z s 2sI2_ AMERICAN MALACOLOGICAL BULLETIN J..JHE NATURAL history museum - 6 AUG 2009 L purchased lzooLcy;y l!e;:as v Journal of the American Malacological Society http://www.maIacoIogicaI. org VOLUME 27 July 29, 2009 NUMBER 1/2 From Poe to Ponder... and Lindberg: Introduction to the symposium “Molluscs as models in evolutionary biology”. MATTHIAS GLAUBRECHT and THOMAS VON RINTELEN 1 On “Darwinian Mysteries” or molluscs as models in evolutionary biology: From local speciation to global radiation. MATTHIAS GLAUBRECHT 3 As time goes by: A simple fool’s guide to molecular clock approaches in invertebrates. THOMAS WILKE, ROLAND SCHULTHEI6, and CHRISTIAN ALBRECHT 25 Molluscan models in evolutionary biology: Apple snails (Gastropoda: Ampullariidae) as a system for addressing fundamental questions. KENNETH A. HAYES, ROBERT H. COWIE, ASLAK JORGENSEN, ROLAND SCHULTHEIfi, CHRISTIAN ALBRECHT, and SILVANA C. THIENGO 47 Land snail models in island biogeography: A tale of two snails. BRENDEN S. HOLLAND and ROBERT H. COWIE 59 Molecular phylogeny, taxonomy, and evolution of the land snail genus Pyrenaearia (Gastropoda, Helicoidea). M. ARANTZAZU ELEJALDE, M. JOSE MADEIRA, CARLOS E. PRIETO, THIERRY BACKELJAU, and BENJAMIN J. GOMEZ-MOLINER 69 Documenting molluscan evolution from ancient long-lived lakes: The case of Toxosoma Conrad, 1874 (Gastropoda, Cochliopidae) in Miocene Amazonian Lake Pebas. FRANK P. WESSELINGH and WILLEM RENEMA 83 coiitiiined on bock cover Kenneth M. Brown, Editor-in-Chief Department of Biological Sciences Louisiana State University Baton Rouge, Louisiana 70803, U.S.A. AMERICAN M4J,ACOLOGICAL BULLETIN . • BOARD OF EDITORS Cynthia D. Trowbridge, Managing Editor Oregon State University P.O. Box 1995 Newport, Oregon 97365, U.S.A. Janice Voltzow Department of Biology University of Scranton Scranton, Pennsylvania 18510-4625, U.S.A. Robert H. Cowie Center for Conservation Research and Training University of Hawaii 3050 Maile Way, Gilmore 408 Honolulu, Hawaii 96822-2231, U.S.A. Carole S. Hickman University of California Berkeley Department of Integrative Biology 3060 VLSB #3140 Berkeley, California 94720, U.S.A. Paula M. Mikkelsen Paleontological Research Institution 1259 Trumansburg Road Ithaca, New York 14850-1313, U.S.A. Alan J. Kohn Department of Zoology Box 351800 University of Washington Seattle, Washington 98195, U.S.A. Dianna Padilla Department of Ecology and Evolution State University of New York Stony Brook, New York 1 1749-5245, U.S.A, Roland C. Anderson The Seattle Aquarium 1483 Alaskan Way Seattle, Washington 98101, U.S.A. Timothy A. Pearce Carnegie Museum of Natural History 4400 Forbes Avenue Pittsburgh, Pennsylvania 15213-4007, U.S.A. Janet Voight The Field Museum 1400 S. Lake Shore Dr. Chicago, Illinois 60605-2496, U.S.A. The American Malacological Bulletin is the scientific journal of the American Malacological Society, an international society of professional, student, and amateur malacologists. Complete information about the Society and its publications can be found on the Society's website: h ttp://www. malacological. org AMERICAN MALACOLOGICAL SOCIETY MEMBERSHIP MEMBERSHIP INFORMATION: Individuals are invited to com- plete the membership application available at the end of this issue. SUBSCRIPTION INFORMATION: Institutional subscriptions are available at a cost of $75 plus postage for addresses outside the U.S.A. Further information on dues, postage fees (for members outside the U.S.A.), and payment options can be found on the membership application at the end of this issue. ALL MEMBERSHIP APPLICATIONS, SUBSCRIPTION ORDERS, AND PAYMENTS should be sent to the Society Treasurer: Dawn E. Dittman Tunison Laboratory of Aquatic Science 3075 Gracie Rd. Cortland, New York 13045-9357, U.S.A. CHANGE OF ADDRESS INFORMATION should be sent to the Society Secretary: Paul Callomon Department of Malacology The Academy of Natural Sciences of Philadelphia 1900 Benjamin Franklin Parkway Philadelphia, Pennsylvania 19103-1 195, U.S.A. INFORMATION FOR GONTRIBUTIONS is available on-line and appears at the end of this issue. MANUSCRIPT SUBMISSION, CLAIMS, AND PERMISSIONS TO REPRINT JOURNAL MATERIAL should be sent to the Editor-in-Chief: Kenneth M. Brown, Editor-in-Chief Department of Biological Sciences Louisiana State University Baton Rouge, Louisiana 70803, U.S.A. Voice: 225-578-1740 • Fax: 225-578-2.597 E-mail: kmbrown(<'’lsu.edu AMERICAN MALACOLOGICAL BULLETIN 27(1/2) AMER. MALAC. BULL. ISSN 0740-2783 Copyright © 2009 by the American Malacological Society (xjver photo: The shells and egg cases of, clockwise from top left, Pomacea instdarum, P. guyatiensis, P. diffusa, and P. haustrum. Apple snails are an excellent system to address questions in evolution ami biodiversity, see I layes et al. 47-58. AMERICAN M ALACOLOGICAL BULLETIN THE NATURAL I HISTORY MUSEUM - 6 AUG 2009 PURCHASED ZOOLOGY LIBRARY contents VOLUME 27 I NUMBER l/2 From Poe to Ponder... and Lindberg: Introduction to the symposium “Molluscs as models in evolutionary biology”. MATTHIAS GLAUBRECHT and THOMAS VON RINTELEN On “Darwinian Mysteries” or molluscs as models in evolutionary biology: From local speciation to global radiation. MATTHIAS GLAUBRECHT 3 As time goes by: A simple fool’s guide to molecular clock approaches in invertebrates. THOMAS WILKE, ROLAND SCHULTHEIfi, and CHRISTIAN ALBRECHT 25 Molluscan models in evolutionary biology: Apple snails (Gastropoda: Ampullariidae) as a system for addressing fundamental questions. KENNETH A. HAYES, ROBERT H. COWIE, ASLAK JORGENSEN, ROLAND SCHULTHEIfi, CHRISTIAN ALBRECHT, and SILVANA C. THIENGO 47 Land snail models in island biogeography: A tale of two snails. BRENDEN S. HOLLAND and ROBERT H. COWIE 59 Molecular phylogeny, taxonomy, and evolution of the land snail genus Pyreuaearia (Gastropoda, Helicoidea). M. ARANTZAZU ELEJALDE, M. JOSE MADEIRA, CARLOS E. PRIETO, THIERRY BACKELJAU, and BENJAMIN J. GOMEZ-MOLINER 69 Documenting molluscan evolution from ancient long-lived lakes: The case of Toxosoma Conrad, 1874 (Gastropoda, Cochliopidae) in Miocene Amazonian Lake Pebas. FRANK P. WESSELINGH and WILLEM RENEMA 83 Morphological cladistic analysis as a model for character evaluation in primitive living chitons (Polyplacophora, Lepidopleurina). JULIA D. SIGWART 95 The use of developmental sequences for assessing evolutionary change in gastropods. JENNIFER SMIRTHWAITE, SIMON D. RUNDLE, and JOHN I. SPICER 105 Alien non-marine snails and slugs of priority quarantine importance in the United States: A preliminary risk assessment. ROBERT H. COWIE, ROBERT T. DILLON, JR., DAVID G. ROBINSON, and JAMES W. SMITH 113 New small deep-sea species of Gastropoda from the Campos Basin off Brazil. RICARDO SILVA ABSALAO 133 The genera Myonera, Octoporia, and Protocuspidaria (Pelecypoda, Cuspidariidae) from deep waters of Campos Basin, Rio de Janeiro, Brazil with descriptions of two new species. CLEO DILNEI DE CASTRO OLIVEIRA and RICARDO SILVA ABSALAO 141 X-ray quantitative texture analysis on Helix aspersa aspera (Pulmonata) shells selected or not for increased weight. DANIEL CHATEIGNER, REINIER KAPTEIN, and MATHILDE DUPONT-NIVET 157 Mollusc survey of the lower Bruneau River, Owyhee County, Idaho, U.S.A. STEVEN J. LYSNE and WILLIAM H. CLARK 167 I The shell features of Cornu aspersum (synonym Helix aspersa) and Helix pomatia: Characteristics and comparison. MACIEJ LIGASZEWSKI, KRZYSZTOF SUROWKA, and JULIA STEKLA 173 Rediscovery of the sacoglossan opisthobranch Hermaea wrangeliae (Ichikawa, 1993) in Okinawa, Japan. CYNTHIA D. TROWBRIDGE, YAYOI M. HIRANO, and YOSHIAKI J. HIRANO 183 Index to Vol. 27 190 Membership Form 193 Information for Contributors 195 1 Amer. Maine. Bull. 27: 1-2 (2009) From Poe to Ponder... and Lindberg: Introduction to the symposium “Molluscs as models in evolutionary biology”"^ Matthias Glaubrecht and Thomas von Rintelen Department of Malacozoology, Museum of Natural History, Leibniz University Berlin, Invalidenstrasse 43, D- 101 15 Berlin, Germany Corresponding author: matthias.glaubrecht@mfn-berlin.de Known to students of our profession and concisely sum- marized most recently in Ponder and Lindberg (2008), Mollusca are, with an estimated 200,000 living species, one of the largest animal phyla, second only to the arthropods. The remarkably rich fossil record of molluscs throws light back into the earliest Cambrian revolution 543 million years ago, and ever since then we find them in nearly every ecosystem on Earth. The classes of living and fossil molluscs comprise an array of diverse animals with the most varied body plans, ranging from minute worm-like animals dwelling between sand grains on the beach to giant squids in the deep sea, and from microscopic snails in leaf-litter to giant clams in coral reefs. As objects of fascination, function, and food, molluscs play important roles in many cultures and societies. They include many taxa of immense economic significance, such as oysters, scallops, and squids; some bivalves produce precious pearls, and some snails carry diseases that infect millions of people, especially in the tropics. Yet we feel that it is not only a curious fact in the history of science, but, unfortunately enough, much more a symp- tomatic indication of our discipline that it was not a profes- sional naturalist or scientist with an interest in malacology, but the poet Edgar Allen Poe (1809-1849), who formulated an idea with much future. Poe was among the first to recog- nize and explicitly recommend that the study of molluscs requires a combined analysis, which in his times meant rec- onciling a classification based on hard shells with evidence from soft body anatomy (see details on this in the opening remarks to the symposium by Glaubrecht (2009)). This syn- thetic idea was long ignored by conchologists, who continued to classify molluscs almost exclusively based on features of their shell, while neglecting the soft body and the biological information that it holds. As a consequence, for a long time we knew few hard facts, for example, about tbe evolution and phylogeny of these soft-bodied animals but instead had much speculation by self-proclaimed authorities in the field. Institute for Research in Evolution and Biodiversity at the Humboldt In addition, most contributions in malacology long cen- tered around morphology, anatomy, and in particular phylo- genetic relationships within and among constituent taxa. Only rarely have molluscs been utilized explicitly as models for the study of the general aspects of evolutionary biology. ITowever, molluscs, with their many features and facets, are highly suitable for providing some fundamental insights into the mechanisms of the genesis of biodiversity, its pattern in historical biogeography, and the underlying processes of spe- ciation and radiation. An increasing number of recent studies and publications on molluscs reveal this rich potential. Therefore, it was the aim of this symposium on molluscs as models in evolutionary biology, held during the World Congress of Malacology (WCM) in Antwerp from the 15th to 20th July 2007 (jointly organized by Unitas Malacologia and the American Malacological Society), to bring together experts and their expertise to provide — based on molluscs — some of those fundamental studies, and to show avenues for using data that are of relevance for evolutionary biology. With 43 talks over more than two full days of sessions (plus several posters), this symposium was the largest at the Antwerp WCM. Following the introduction, two invited keynotes or plenary lectures were given, one by Suzanne Williams and David Reid (on global pattern of diversity and speciation) and one by Thomas Wilke and Christian Albrecht (on genesis of biodiversity, focusing on ancient lakes). Other lectures cov- ered a wide array of topics ranging from biogeography, shell morphology and evolution, molecular phylogenetics, radia- tions and extinctions as documented in the fossil record, to mitogenomics, and aspects of development and reproduc- tion. From all these presentations, a selection of eleven con- tributions were made, and we invited tbe authors to work out their main subject as exemplars for their specific area of research, viewed fi'om their individual perspective. Subse- quently, eight of the original speakers have been able to pro- vide manuscripts for the American Malacological Bulletin. *From the symposium “Molluscs as models in evolutionary biology: from local speciation to global radiation” presented at the World Con- gress of Malacology, held from 15 to 20 July 2007 in Antwerp, Belgium. 1 9 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Three other papers of those originally invited have been pub- lished elsewhere in the meantime, viz. Lindberg (2007) on a case study from the limpet Scutellastra flexiiosa (Quoy and Gaimard, 1834) from Moorea, contrasting the respective roles of deep phylogenetic history with recent adaptations in shaping current ecological and life history characteristics, Hershler and Liu (2008) on vicariance and dispersal of hy- drobiid springsnails in the southwestern United States, and Williams and Duda (2008) on biogeography and speciation. We feel that together with these papers, the studies pre- sented here dealing with phenomena from local speciation to global radiations, and including both the paleontological as well as neonotological perspective, underline the potential that molluscs have as models in evolutionary biology. The first contribution by Wilke et al. (“As time goes by...”) pro- vides a concise review of molecular clock methods and is at the same time a hands-on manual for molecular clock analy- ses. This contribution specifically is aimed at researchers in malacology, giving external clock rates for the cytochrome oxidase subunit I which may be applied for most aquatic mol- luscan taxa and even other invertebrates. Molecular clocks have become a standard tool in evolutionary biology and this review provides a sound basis for extending the use of mol- luscs as models. The next two papers by Hayes et al. on “Apple snails as a system for addressing fundamental questions” and by Holland and Cowie on “Land snail models in island bioge- ography” exemplarily highlight the use of taxa like freshwater and terrestrial gastropods for addressing major issues in evo- lutionary research. A common focus of both studies is on exploring patterns of biogeography, and in particular island colonization for the Hawaiian Succineidae and Achatinellinae. In addition, Hayes et al. also discuss the role of ampullariids in advancing knowledge of speciation and adaptation pro- cesses. Subsequently, Elejalde et al. (on the “land snail genus Pyrenaearia") uses these Iberian helicids to investigate the link between molecular data (DNA taxonomy) and classic taxonomy on one hand, and the role of climate in diversifica- tion on the other hand. While all papers so far rely heavily on molecular data, the last three papers in this symposium cover other important aspects of molluscs also relevant in evolu- tionary biology. Wesselingh and Renema (“Molluscan evolu- tion from ancient long-lived lakes”) offer a paleontological perspective on snail diversification in Miocene Amazonian Lake Pebas. I'heir focus is on the tempo and mode of evolu- tionary change in what we like to term a ‘natural laboratory’, as it has model characteristics also for interpreting evidence from modern long-lived lakes, 'fhe paper by Sigwart on “cla- distic analysis in Polyplacophora” again focuses on a recent group, the living chitons, and aims at elucidating the role and value of morphological characters in tracking the phyloge- netic relationships and evolution. In a nutshell, this paper questions the value of morphological characters in a molecular world. The final contribution, by Smirthwaite etal. on “the use of developmental sequences”, offers a glimpse of molluscs in developmental biology, which is a renaissance issue in evolutionary research. Using freshwater pulmonates, the authors discuss the potential of molluscs to offer mecha- nistic explanations of ontogeny, including heterochrony. We hope that these studies and the avenues they suggest will further facilitate the influence of malacology within evo- lutionary biology, even more so in the future than it was the case in the past. It is a pleasure to thank all who participated in this symposium during the Antwerp WCM meeting and, in particular, the speakers who contributed to what we felt to be a stimulating session, as reflected in the many discussions. We are most grateful to the authors that have contributed to this symposium volume and to Ken Brown, who kindly offered to publish these contributions as an outcome of the symposium, finally, we would also like to thank Thierry Backeljau from the Royal Belgian Institute of Natural Sciences in Brussels, and at that time president of Unitas Malacologica, for inviting us (shortly after the Perth meeting in 2004) and then helping to organize this symposium in 2007. LITERATURE CITED Glaiibrccht, M. 2009. On "Darwinian Mysteries" or molluscs as models in evolutionary biology: From local speciation to global radiation. American Malacological Bulletin 27: 3-23. Hershler, R. and H.-P. Liu. 2008. Ancient vicariance and recent dis- persal of springsnails (Hydrobiidae: Pyrgiilopsis) in the Death Valley system, California-Nevada. In: M. C. Reheis, R. Hersh- ler, and D. M. Miller, eds.. Late Cenozoic Drainage History of the Southwestern Great Basin and Lower Colorado River Regio)i: Geologic and Biotic Perspectives. Geological Society of America Special Paper 439. The Geological Society. Pp. 91-101. Lindberg, D. R. 2007. Reproduction, ecology, and evolution of the Indo-Pacific limpet Scutellastra flexuosa. Bulletin of Marine Sci- ence 81: 219-234. Ponder, W. F. and D. R. Lindberg. 2008. Phylogeny and Evolution of the Mollusca. University of Galifornia Press, Berkeley. Williams, S. T. and T. F. Duda. 2008. Did tectonic activity stimulate Oligo-Miocene speciation in the Indo-West Pacific? Evolution 62: 1618-1634. Submitted: 18 April 2009; accepted: 24 April 2009; final revisions received: 28 April 2009 Amer. Maine. Bull. 27: 3-23 (2009) On “Darwinian Mysteries” or molluscs as models in evolutionary biology: From local speciation to global radiation"^ Matthias Glaubrecht Department of Malacozoology, Museum of Natural History, Leibniz Institute for Research in Evolution and Biodiversity at the Humboldt University Berlin, Invalidenstrasse 43, D- 101 15 Berlin, Germany Corresponding author: matthias.glaubrecht@mfn-berlin.de Abstract: Evolutionary biology is not only a biological subdiscipline but also a synthetic theory based on comprehensive scientific achievements. However, to date biodiversity, which is far from being fully documented, and the evolutionary processes leading to it are two of the least understood phenomena in evolutionary biology. Surprisingly, decades after the Modern Synthesis and centuries after the commencement of research in biological systematics, we are still unable to satisfyingly answer apparently simple yet fundamental questions. Here termed “Darwinian mysteries”, these are for example, how many species inhabit Earth today, what are species, where are they distributed, and how did biodiversity originate. While many contributions in malacology center around morphology, anatomy, and phylogenetic relationships within and among constituent taxa, molluscs only rarely have been utilized explicitly as models for the study of general aspects in evolutionary biology. However, this particular group, with its many features and facets, is highly suitable for providing fundamental insights into the mechanisms that generate biodiversity, pattern in historical biogeography, and the underlying processes of speciation and radiation. Here, I discuss some aspects of these fundamental questions that are of relevance for evolutionary biology, hoping that the influence of malacology within evolutionary biology will increase in the future. Keywords: biodiversity, evolution, species, species numbers, species concepts “Science is built of facts as a house is built of bricks; but an accumulation of facts is no more science than a pile of bricks is a house.” Henri Poincare, La Science et I’hypothese ( 1902: 101 ) Molluscs are not only one of the most spectacular animal phyla with great taxonomic diversity and morphological disparity but also have the potential to provide us with some of the most remarkable models in evolutionary biology. Actually, it has escaped many (and not only) malacologists’ attention that molluscs were at the forefront of evolutionary theory in the first place. For example, they were instrumental in Jean-Baptiste de Lamarck’s (1744-1829) hrst evolutionary views. When Lamarck (1801) introduced his then novel classification of invertebrates and included Mollusca as a class on its own, based on his study of the rich molluscan collection in the Paris museum, he also proposed in his introductory “discours preliminaire” a brief exposition of his later, much- debated evolutionary theory (Mayr 1982, Schilling 1989, Burkhardt 1995, Laurent 1997). Although Lamarck suggested an incorrect mechanism for evolution, in comparing fossil molluscs with recent species, he for the first time realized divergent evolution over geological time scales. Molluscs were also the trigger, albeit not the focus, of one of the most famous controversies in the history of science, as it was the debate on cuttlefish anatomy that started the epochal debate in the Paris Academy of Sciences in 1830. Here again a group of molluscs was at center stage when Georges Cuvier’s functionalism opposed Etienne Geoffroy Saint- Hilaire’s “philosophical” morphology (Appel 1987, Le Guyader 2004). Nevertheless, after this crucial epoch, malacology did not remain at the forefront of zoological discoveries and evolu- tionary biology. Although being little more than a marginal note in the history of science, to my view the following episode is not only a curious incidence but also a symptomatic indication for our field. Remarkably, not a professional naturalist or scientist with a keen interest in malacology, but the American poet Edgar Allen Poe (1809-1849) was among the first to recognize and comment that a reliable classification of molluscs requires a combined analysis, which meant in his times reconciling a system based on hard shells (as suggested by Lamarck) with evidence from soft body anatomy (as provided by Cuvier). Illustrated by 215 shells of molluscs, Poe (1839) published a scholarly and most successful textbook with a telling title, viz. The Conchologisds First Book: or, A * From the symposium “Molluscs as models in evolutionary biology: from local speciation to global radiation” presented at the World Con- gress of Malacology, held from 15 to 20 July 2007 in Antwerp, Belgium. 3 4 AMERICAN MALACOLOGICAL BULLETIN 27 • 1 /2 • 2009 system of Testaceous Malacology, arranged expressly for the use of schools, in which the animals, according to Cuvier are given with the shells, a greater number of new species added and the whole brought up as accurately as possible to the present condition of the science. And he explicitly distinguished between conchology as being merely the study of shells versus malacology as being the study of molluscs, i.e., the anatomy of the whole animal including its most important soft parts. After Myra Keen ( 1936) and Joseph Moldenhauer ( 1971) had remarked on this case of “literary curiosity”, the late Stephen Jay Gould (1993, 1995), both a malacologist and evolutionary biologist himself, looked into this episode in more detail. Gould developed the argument that, irrespective of the plagiarism and piracy of Thomas Brown’s earlier book, and Poe’s functioning as a ghost writer and straw man for Thomas Wyatt (as is true for one other case; see Heartman and Canny 1943), it was indeed Poe himself who made the above mentioned important distinction. Although published under Poe’s name, the book was essentially a less-expensive edition of Thomas Wyatt’s (1838) own work Manual of Conchology, which plagiarized most of the text from British naturalist Thomas Brown’s (1833) Conchologist’s Text-book. However, since Poe evidently wrote the preface and introduction, then used the system of classification from Wyatt’s work, but substituted for each genus a paragraph description of the soft parts (maybe taken over from Cuvier), it can be reconstructed that indeed it was Poe’s own novel idea of a combined analysis. Evidently, his early insight into what malacology should be was not borrowed from either Brown and/or Wyatt. Therefore, it remains a curious fact that Poe as a poet with only a marginal interest in science formulated an explicit conceptual reform (i.e., the distinction between conchology and malacology), apparently an idea with much future. Poe’s Conchologist’s First Book, therefore, should be perceived now as more than just a literary curiosity and a collage from others’ texts, as his emphasis of the eminent importance of anatomical features in the distinction and classification of molluscan species reveals innovative potential long unrecognized, paving the way toward malacology as a truly biological discipline. Unfortunately, his insight was ignored for almost another century while conchologists continued to classify molluscs exclusively based on features of their shell while neglecting the soft body and biological information that it holds. Interestingly, following Lamarck’s ( 1792) earliest attempts on classification of invertebrates, where the systematics of molluscs was solely based on shells, Lamarck (1809) later learned from Georges Guvier’s studies on soft-body anatomy, and acknowledged explicitly the importance of these morphological data for his classification (Gorsi 1988: 62). Unfortunately, although Lamarck’s classification of inverte- brates in his original seven-volume llistoirc Nalurelle dcs Animaux sans Vertebres (Lamarck 1815-1822) was widely known and used by many 19''’ century naturalists, his theo- retical insights and reference to Cuvier’s attempt were largely ignored. Continuing in this unhappy tradition, students of molluscs for a long period missed the chance to modernize malacology as a scientific discipline. The problem inherent to the traditional study of only shells, for example, for systematic- taxonomic purposes was that no other sources of information were utilized to evaluate the diagnostic value of shell features. This stands in stark contrast to other disciplines like ornithology, where the so called “/zeiv systematics” provided the foundation for developing the modern theory of evolution (see Glaubrecht 2007, references therein). In the context of malacology developing as a biological science, a study of systematics vs. merely “stamp collecting” (Glaubrecht 2004: 117), I add here as surely more than coincidence that one of Britain’s former leading conchologists, Erancis James Stainforth ( 1797-1866), who was a supplier of many rare shell specimens to Lovell Reeve, also initiated in the early 1860s in London the first stamp collector’s club, thus helping to found philately (Allen 2008). However, molluscs are much more than only collectors’ items. With their many features and facets, they are highly suitable for providing some fundamental insights into the mechanisms of the genesis of biodiversity, its pattern in historical biogeography, and the underlying processes of speciation and radiations. As recently summarized by Ponder and Lindberg (2008), with about 200,000 living species the Mollusca are one of the largest animal phyla, second only to the arthropods. The remarkably rich fossil record of molluscs enlightens the earliest Cambrian revolution some 540 million years ago, and ever since we find them in nearly every ecosystem on Earth. The seven or eight classes of living molluscs plus two extinct class-rank taxa comprise an array of most diverse animals with most varied body plans, ranging from minute worm-like animals dwelling between sand grains on the beach to giant squids in the deep sea, and from microscopic snails in leaf-litter to giant clams in coral reefs. As objects of fascination, function, and food, molluscs play important roles in many cultures and societies. Molluscs include many taxa of immense economic significance, such as oysters, scallops, and squids; some bivalves produce precious pearls, and some snails carry diseases that intect millions of people, especially in the tropics. Yet, for a long time we knew few hard facts, for example, about the evolution and phylogeny ot these soit-bodied animals. For the greater part of the last century, it was lohannes Ihiele’s (1929-1931) epochal llandbuch dcr Systctnalischen Wcichlicrkutule that long provided the stan- dard in molluscan systematics. For his classification, I'hiele evaluated characteristics from the shell but consequently used a synthetic, albeit pre-cladistic manner when he included the MOLLUSCS AS MODELS 5 most important radula features as well as other anatomical characters following a tradition started by Troschel (1856- 1863). Although Thiele sometimes erred, for example when he discarded aplacophoran molluscs as “worms” belonging to the phylum Annelida (see Glaubrecht et al 2005), his systematization was a highly influential masterpiece. Another curious fact was that when Thiele’s handbook was translated into English and re-published by Bieler and Mikkelsen (1992), a substantial amount of new data from both morphology and molecular genetics as well as from paleontology had begun to accumulate, too diverse for a single malacologist to master. New tools — such as fluorescence-coupled antibody stain- ing and confocal laser-scanning microscopy, in concert with new approaches such as computer-assisted cladistics, which allow the recurrent testing of phylogenetic hypotheses under various models and assumptions — have generated a renewed interest in reconstructing evolutionary history with molluscs in a key position (Valentine 2004, Minelli 2009a). Conse- quently, within the last two decades our understanding of molluscan phylogeny has undergone a remarkable, if not even revolutionary, transformation that is about to change fundamentally the classification of phyla. Ponder and Lindberg (2008) provide a collation badly needed when virtually every phylogenetic tree calculated from another partial gene fragment of more or less randomly represented taxa is considered publishable. While many contributions center on morphology, anatomy, and phylogenetic relationships within and among constituent taxa, molluscs have only rarely been utilized as general models for the study of evolutionary biology. Molluscs rarely make it into textbooks on evolutionary theory, and are highly underrepresented when it comes to discussing evolu- tionary concepts and/or phenomena by example (Ridley 1996, Eutuyma 1997, Mayr 2001, Barton et al. 2007). Eor only one more recent example, in a very readable book on evolu- tionary pathways by Avise (2006), only two cases (coiled vs. uncoiled shells and land snail chirality) explicitly referred to recent studies utilizing molluscs. Undoubtedly, however, molluscs have so much more to teach us. As evolution is the vibrant foundation for biology, evolutionary biology in this context has a double function. It is not only a biological subdiscipline but also a synthetic theory and discipline on its own, based on comprehensive scientific achievements. Unfortunately, malacologists in partic- ular have followed one of the most prevailing methodological claims that gathering facts should be the primary role of naturalists (Johnson 2005) which often discourages explicit references to evolutionary theory. In addition, as taxonomy and systematics once marked the beginning of zoology in general, it is still among the primary interests of many malacologists. My point is that malacology has more potential to contribute to evolutionary biology than previously assumed or revealed. Taxonomy, albeit not experimental, does not have to remain merely descriptive, but should be hypothesis- driven as well as drive hypotheses. It forms an integral part of evolutionary biology, as taxonomic facts are a prerequisite to the proper formulation of evolutionary and ecological questions (May 1992, 1999). However, although systematics continues to lay the foundation for many disciplines of the life sciences, its ultimate goals of providing an inventory of biodiversity, and reconstructing a tree of life (Glaubrecht 2007), are far from being achieved in general or in molluscs in particular. The diversity of organisms, and the morphological disparity, as well as the evolutionary processes leading to it, are the least understood phenomena in evolutionary biology. Some facts and reasons why biological diversity is far from being dis- covered, or its origin understood, will be briefly highlighted here. THE SIX “DARWINIAN MYSTERIES” IN BIOSYSTEMATICS AND EVOLUTIONARY BIOLOGY A love of science has much to do with its mysteries that drive basic scientific research; questions are often considered more important than answers in shaping the future of science and its disciplines, fundamental questions can be used as guidelines for future, cutting-edge research and reveal oppor- tunities to be exploited, starting from the supposition that scientists should answer these questions over the next quarter century (Kennedy and Norman 2005). In this context, the apparently simple question “What determines species diver- sity?” was recently ranked among the 25 “big” questions, based on how fundamental they are, how broad-ranging, and whether their solutions will impact other scientific disciplines (Pennisi 2005). Among the others were how Earth’s interior works, the composition of the universe, and whether we are alone in it, or whether the laws of physics can be unified. Highlighting our scientific ignorance, it is surprising that centuries after the commencement of research in biological systematics and decades after the Modern Synthesis of evolutionary biology, we are still unable to answer a series of simple questions linked to this big question on biological diversity. Today, 300 years after Carl Linnaeus (1707-1778) and 200 years after Charles Darwin ( 1809-1882), both biosys- tematics and evolutionary biology are still left with the following six “Darwinian mysteries” (Glaubrecht 2003, 2004, 2005, 2007). One and a half centuries after On the Origin of Species by Darwin (1859), these questions, all relevant to his (r)evolutionary theories, are all largely unanswered, irre- spective of the many fruitful attempts to solve them. ( 1 ) Species numbers: How many species are there? 6 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 (2) Species concepts: What are species and how do we know? (3) Speciation: How do new species evolve? (4) Biogeography: Where are the species and why are they distributed there? (5) Phylogenetics: How are species (groups) related? (6) Genetic causation and molecular processes: How do new forms come into being? Undoubtedly, answering these questions will ultimately help to unravel some of the most important issues not only in biosystematics, but also in evolutionary biology. To accom- plish modern systematists’ tasks, viz. quantifying biodiversity, establishing phylogenies, and understanding the evolutionary process of speciation and radiation, several prerequisites are indispensable, albeit neglected even in many modern systematic approaches. Malacology, with one of the richest animal phyla at hand, both in terms of species numbers and biological phenomena, can have its share in solving these “Darwinian mysteries”. I outline here only the first three of these Darwinian mysteries in more detail. (1) The mystery of species numbers: A Linnean enterprise As taxonomy and systematics provide the reference system for all biology (Wilson 1989), compiling, organizing, and updating taxonomic information has an urgent priority. After decades of de-emphasizing biosystematics (Whitehead 1990, Mikkelsen and Cracraft 2001), and recent taxonomic modernization and renaissance (c.^., Godfray 2002, 2007, Mallet and Willmott 2003, Wheeler 2004, Dayrat 2005, Glaubrecht 2007), the reawakening interest in taxonomy has also led to realization that most animal species, especially invertebrates like molluscs, are far from being discovered yet. In fact, it can be called the “Linnean shortfall” that, while knowing how many atoms are in a molecule, how many craters are on the moon, and how many stars are in the Milky Way or galaxies in the universe, we still have not learned how many species of butterflies live on tropical trees, how many buccinid gastropods are in the sea, freshwater melanopsids around the Mediterranean, or camaenids in Australia, for way too long a period of time, zoologists seemed to perceive “a number of undescribed creatures rather a nuisance”, as Darwin long ago complained (Keynes 2003). Nearly three centuries after Linnaeus’s first attempts to inventory nature, his task remains a daunting challenge for systematists, includ- ing malacologists. Surprisingly, only recently has it become apparent that we lack the most fundamental data on biodiversity, viz. systematic inventories on any organismal, ecological, and geographical level (t’.g., Wilson 1988, 1992, Raven and Williams 1997, for molluscs see Bouchet 1997). Recognizing and describing the living species of plants and animals on Earth, however, is a major task, calling for a large-.scale approach to taxonomy (Raven and Wilson 1992, Wilson 2000, 2003, Lawler 2001, Mikkelsen and Cracraft 2001, Blackmore 2002, Wheeler et al. 2004, Stork 2007), comparable to the Human Genome Project or the NASA next generation space telescope and Sloan Digital Sky Survey. Indeed, it remains an unfinished Linnean challenge to inventory the many species that exist on Earth. Estimates of the number of living species in the world, given by Terry Erwin (based on neotropical insect diversity) as up to 30 million species, have triggered more specific attention, as the question ‘How many species are there?’ has finally been regarded as scientifically important (Erwin 1982, May 1988, 1990, 1992, 1994, 1999, Stork 1988, 1993). Estimates of the total number of species vary now from 5 to over 50 million, using various direct and indirect assessments. Over the last two decades, these global estimates dropped to a total of 5 to 15 million species (Stork 1993, Odegaard 2000). A most comprehensive compilation of species numbers has been provided by Chapman (2005), who settled on between 8 and 9 million species. In this context, the discussion has largely centered around the question of what fraction of the insect species found on a given host-tree is likely to be effectively specialized on it, on species-size relations, or in food web structure (May 1990). for example, after having taken into account the host specificity in particular of herbivorous insects, which is essentially responsible for driving most of these species number estimates, the figure has recently been corrected to 4.8 to 6.6 million species (Novotny et al. 2002, 2007). The latter authors attempted to reconcile an order of magnitude discrepancy between extrapolations based on ecological samples with those based on sampling regional faunas or estimates based on taxonomic collections. However, it is doubtful that neotropical, herbivorous insect diversity is a direct function of plant species number (comprising taxo- nomic and architectural diversity), suggesting that additional factors like the existence of morphologically cryptic species and distributional ranges would again increase global species numbers (Dyer et al. 2007, Condon et al. 2008). Surprisingly, very little solid data has been contributed to the discussion of species richness from other invertebrates, such as molluscs or for aquatic biota, with the notable exception of Bouchet et al. (2002) for marine molluscs, that may account for a large fraction of all marine invertebrate species. I’herelore, not even the magnitude of the world’s diversity is currently known to systematists, let alone exact figures for particular groups of animals. Although it is agreed now that we desperately need a biodiversity assessment comprising a complete inventory ol life, we are still challenged to develop rough estimates on the quantity of species as the units of biodiversity studies and evolution (.see below). Irrespective ot a more exact estimate ot the number of species, we are faced with a tremendous task, given that only MOLLUSCS AS MODELS 7 approx. 1.8 million animal species have been properly described (Stork 1988, May 1990). Amazingly, not even the number of species that have already been named and recorded is known precisely. Thus, not only does Linnaeus lag so far behind Newton (given our knowledge about the universe compared to the living world), but more importantly at the level of individual taxonomic groups we see major differences in the accumulation of recorded numbers of known species, with most invertebrate groups being largely underexplored. May (1990, 1992) has calculated the rate of discovery from thetimesof Linnaeus (1758) up to 1970 and 1990, respectively. In Fig. 1, vertebrates are illustrated by the most representative group, birds, versus an invertebrate group, arthropods. Expressed as a fraction of those species known in 1990, half of all known bird species were already recorded in the century after Linnaeus, by 1 845. In contrast, only half of the arthropods known in 1990 were described within the preceding three decades. The “furries and featheries” are very well known by now, at least at the genus level, with only 134 bird species being added to the total of over 9000 since 1934 (Diamond 1985) and only a few mammal species. With respect to invertebrates (including, of course, molluscs, see below), we have just started to deal with the Linnean challenge of a species inventory. We also have to face the fact that most biological assessments largely use focal groups, i.e., vertebrates and vascular plants, but few invertebrate groups such as butterflies. They all comprise not more than five percent of the known diversity while major environmental players (bacteria, fungi, mites, nematode worms, and beetles) are largely unknown, underrated, and underestimated (May 1988, 1992). The major part of diversity in most invertebrates, the other 99% as it has been called (Ponder and Lunney 1999), is therefore still hidden. In Fig. 2 the most recent figures for individual animal groups used by Chapman (2005) is shown, giving the number of known and of estimated species, respectively, for Mollusca as 70,000 to 120,000 species. Other figures of the number of described species of invertebrates were in the past rightly criticized for not only being a matter of great difficulty but also of little value, and those in most textbooks of zoology were long simply copied from each other. Currently, only a few studies are available for global estimates of specific taxo- nomic groups, such as ( 1 ) arthropods and more specifically Hymenoptera (Odegaard 2000, Dolphin and Quicke 2001), (2) on some crustaceans trying to also quantify the underesti- mation (Adamowicz and Purvis 2005), or (3) on marine fishes ( Mora et al. 2008 ) . However, how much do we as malacologists know about the various subjects of our special interest among the marine, freshwater, and terrestrial taxa? With the world’s oceans and tropical rain forests teeming with life forms, each new survey increases the window through which we see this living world. Our museum collections are already filled with nearly endless numbers of scientific hypotheses, biological tests, and questions. However, can we hope that we will eventually understand the biosphere if we do not even struggle to make comparable inventories of our knowledge on these new and old discoveries? It is estimated that we currently add roughly 10,000 new animal species each year (May 2004), among which are well over 400 new species of Mollusca (Bouchet 1997), discounting for past synonymies (see discussion below). With this plethora of new species and taxa continuously being found, however, we do not even attempt to list how many new species exactly are described annually, or how many await description. Yet we lack, curiously enough, several of the most important parameters or instruments to compile a complete list of species on Earth. To begin with, we still have an incomplete taxonomic database for the known species in the world, as no YEAR YEAR Figure 1. The rate of discovery from the time of Linnaeus ( 1758) up to 1990. Accumulation of species for a vertebrate representative group, birds, versus arthropods (excluding Insecta) expressed as a fraction of those known in 1990 on a logarithmic scale plotted against time. Note that half of all known birds were already described by 1845, in contrast to 1960 in arthropods. Modified from May (1992). 8 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Mollusca: 70,000 - 120,000 □ Hemichordata ■ Echinodermata ■ Insecta □ Arachnida □ Myriapoda □ Crustacea B Onchyophora □ Mollusca B Annelida □ Nematoda o Acanthocephala □ Platyhelminthes ■ Cnidarla ■ Porifera □ Others ■ Vertebrate B. o Cyclostomata ■ Chondrichthyes □ Osteichthyes ■ Amphibia o Reptilia ■ A\«s ■ Mammalia Figure 2. Numbers of known animal species: A, all major taxa; B, vertebrates. Note that for Mollusca the figures is given as 70,000 known species, with the most recent estimate of 120,000 species to- tal. Modified from Chapman (2005). synoptical and coordinated catalog or compilation exists, nor is any effort made to comprehensively list all named species, in particular for invertebrates. This situation basically leaves us with the Zoological Record as the only (albeit with an omission rate of over 20% of new molhiscan names) far from complete compilation of supra-specific names (Bouchet and Rocroi 1992, 1993, Edwards and Thorne 1993). It has been suggested to separate this registration of all new names and nomenclatorial acts from the scientific content of taxonomic papers (Minelli 2003a). However, recently the ICZN has set up a new system, called ZooBank, that requires species descriptions to be registered online (Polaszek 2005). Bouchet and Rocroi ( 1992, 1993) also suggested that names should be registered before they are declared available. InvctUorying the rnalacofaiina Molluscs as the second largest animal phylum are not an exception to this Einnean challenge as outlined above, with exact figures on the number of living and/or fossil species lacking and extrapolations widely variable. Linnaeus in the 1 0th edition of his Systema Naturae in 1758 described among the 4„236 animals listed (there were 549 species in bis first edition of 1735) nearly 700 species of mollu.scs (listed under “4'estacea” that also included brachiopods). Haifa century later, Lamarck assumed in his classification 135,000 species of known invertebrates (Laurent 1997). One of the earliest attempts to more precisely extrapolate the number of living species for all classes of animals from representatives in museum collections was done by the director of the Berlin natural history museum, Karl August Mobius (1825-1908), as detailed in Glaubrecht (2008c). In 1898, using the roughly 30,000 species (with an estimated number of 300.000 specimens) then extant in the Berlin collection (albeit at that time still united with the brachiopods), Mobius arrived at the very conservative figure of 50,000 known species of molluscs (out of or among the well over 400,000 species in total that he at that time estimated). More than half a century later, using instead the available compilations of generic and subge- neric names, Eranz Alfred Schilder (1896-1970) tried to estimate at least for prosobranch gastropods, one of the major classes, the number of fossil and extant families. Eor this subclass alone, he estimated 415 families and about 20,000 genera, a total of 150.000 species (Schilder 1947). He later extrapolated the number of all living molluscs as 250,000 species (among an estimated 3 million animal species), assuming that about half of all species were described at that time (Schilder 1948, 1949). Later, Muller and Campbell (1954) gave the total number of known molluscs species as a much more reduced 73,000, of which over 41,000 (57%) are living species and over 31,000 (43%) fossil species. Later, Boss (1970) doubted the total of 107.000 living species, with 58,000 marine, 14,000 freshwater, and 35,000 terrestrial molluscs estimated at that time (Nicol 1969). Boss (1971) arrived at only about 35,000 species of molluscs known and, guessing the number yet to be described, he suggested a total of nearly 47,000 molluscan species (with 37,500 gastropods and 7,000 bivalves), later corrected to about 60.000 Recent species. As noted earlier (Bouchet 1997) molluscs are a group where the number of described species is problematic, ranging from 45,000 to 150,000. For example. May (1990) gave the estimated number of species in this group recorded scien- tifically to 1970 as 45,000, with 1887 to 1899 being the peak for discovery of new species. According to his estimates, it has taken over 70 years prior to 1970 to record the second hall of this total number of 45,000 species. However, May (1988) listed Mollusca with about 100,000 recorded species, esti- mating the research effort devoted to molluscs (Irom the average number of publications per year in the Zoological Record from 1978 to 1987) as comparable to Lepidoptera and Hymenoptera, but an order of magnitude less than in birds or mammals. Baillie ct al. (2004) gave a total ot 107,7 1 8 described mollu.scan species (75,000 gastropods, 30,000 bivalves, 768 cephalopods, and 1,950 others). ’Lhesc estimates are closer to the one given in Ghapman (2005), who arrived at an estimated number of 120,000 species. However, Ponder and Lindberg (2008) tentatively figured that there might be 200,000 extant MOLLUSCS AS MODELS 9 molluscan species. For marine molluscs alone, Bouchet (2006) estimated 52,500 species worldwide. However, as assessments of species richness are confronted with a plethora of difficulties, reliable figures are still wanting. Only rarely are surveys on mollusc species available to date, even for particular classes or larger sub- groups. Therefore, we can only roughly extrapolate where the species in Mollusca are, i.e., which molluscan taxa hold how many described species. Using various catalogues and check- lists as well as the Zoological Record for taxonomic papers (although the later is not without its deficiencies), Bouchet and Rocroi (1992, 1993) compiled the nomenclaturally available genus-group names for the classes of all Mollusca and Recent Mollusca that were introduced since Linnaeus’ time (1758) until 1989 (Fig. 3). While cephalopods dominate the fossil record with ca. 34% of all named molluscs, bivalves make up ca. 17% among the living taxa. Depending on the method, Bouchet and Rocroi (1992) found between 25,000 and 28,000 supra-specific names, with nearly half of them (all Mollusca) or two thirds (among Recent Mollusca only) for gastropods. They gave the number of genus-group names of Recent molluscs at about 12,000. Before their compilation, Schilder (1947), in his attempt to estimate the number of species in prosobranchs, suggested that there might be 20,000 genus-group names for all Molluscs, with some 5,000 taxa still being extant while Vaught (1989) arrived at 15,000 genus- group names. With an average of 224 new genus-group names per year, the annual increment has remained relatively stable for molluscs since the late century, thus mirroring the same increasing rate of discovery described by May (1990, 1992) for other invertebrates (Fig. 1). Several compilations I found in the literature exhibit this pattern. Thus, with respect to molluscs we surely continue to live in the golden age of discoveries, with a plethora of new species yet to be found and described. As only one of the many cases stated here, we continue to do this for a small group of freshwater gastropods endemic to the island of Sulawesi (Rintelen and Glaubrecht 2003, Rintelen et al. 2007). Bouchet (1997) figured this progressing species inventory in selected groups of Recent molluscs, among them a few groups of marine gastropods, land snails and bivalves (Fig. 4A). The same general trend can also be seen in the data compiled with a much more restricted focus, for example, by Vermeij (1999) for species with shells having a labral tooth although there are two peaks in the number of described species (Fig. 4B). Here, we have charted the progress of knowledge on two randomly picked groups of molluscs, also using Linnaeus and 1758 as starting point, figuring the cumulative number of known species over the subsequent centuries (Fig. 5). As is evident from these data compiled here for Cephalopoda and Conns Linnaeus, 1758, both taxa reveal the one trend already discussed above for invertebrate taxa in general versus vertebrates (Fig. 1). In his compilation, Bouchet (1997) estimated that, on average, 1,395 new species-group molluscs are named each year (69% are fossil species, leaving 430 living species per year), with the number of new marine species described each year increasing by 68%, while the number of new non-marine species decreased by more than 1 5% over the past three decades. Nevertheless, Bouchet (1997: 1 ) also saw no sign of leveling off in the inventory of molluscan diversity, concluding that “for the foreseeable future, micromollusks, the deep-sea, and the marine and non-marine tropics will remain effectively inex- haustible reservoirs of undescribed species”. Actually, there is evidence of hitherto unrecognized taxonomic groups among molluscs from all biotopes. Thus, an inescapable conclusion emerges about a largely uncharted biosphere, and that we are still far from having completed the catalogue of life. However, not only are many biotas incompletely studied, but also our lack of knowledge limits our ability to comprehend and respond urgently to the biodiversity crisis, viz. the loss of many of Earth’s species. Only one of the many reasons is certainly that in- stitutional and financial support for systematics and natural ■ Aplacophora Q Monopiacophora @ Polyplacophora ■ Bivalvia □ Scaphopoda □ Gastropoda M Hyolitha. etc . □ Cephalopoda B. Gastropoda 76.94% Cephalopoda 17.46% Bivalvia ■ Aplacophora B Monopiacophora B Polyplacophora ■ Bivalvia O Scaphopoda □ Gastropoda B Hyolitha. etc. . □ Cephalc^xxja Figure 3. Where are the species in Mollusca? A, genus-group names of all Mollusca, introduced since 1758 (until 1989), partitioned by class, with percentage given for main taxa. B, genus-group names of only Recent Mollusca, introduced from 1758 to 1989, partitioned by class, with percentage given for main taxa. Both figures modified from Bouchet and Rocroi ( 1992). 10 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Number of species described Figure 4. Inventorying the nialacofauna. A, number of species of selected molliiscan groups since I75H, plotted as a percentage ol llie known number of species in (Irom Bouebet I997).'B, number ol gastropod species with a labial tooth describerl (ler 20-year interval Irom 1740 to 1999 (from Vermeij 1999). history studies have continued to dwindle for several decades now, re- sulting in a situation called the “other” taxonomic impediment (Evenhuis 2007). At the same time, there have been various claims that this situation will require a fundamental restructuring of the way we do systematics (Godfray 2007), and calls for accelerated biodiversity assessments, in particular suggesting molecular tax- onomy (and DNA barcoding) as the proper way out (see below). I claim that this barely faces the real and fundamental problems, as there are still other Darwinian mysteries involved that are not tackled or solved by this approach. These problems center on the question if species are real and how many are valid. (2) The mystery of the species concept: ^^Defining the undefinable^' May (2004) noted that detecting new species in the field will remain the rate-limiting step to tomorrow’s taxon- omy. Species are life’s mosaic across space and time. However, there is a fundamental disagreement among sys- tematists of all kinds and expertises about what these species are and what we can regard as the functional unit in evolution on the one hand and in taxonomy on the other hand. Searching for biologically meaningful entities, /.e., species as they are conventionally called, as well as how to define and delimit them has been a century-long challenge, and its full understanding needs to take into account an equally long history which cannot even be outlined here. Discussions of the species problem litter the scientific literature, hampering progress, often obscuring a clear vision of the conceptual issues and practical problems centered around species, with this distinction not being always clear- cut. However, the species problem still remains fundamental, irrespective of the common conjecture ol it being merely a question ol semantics. Although the question “what is meant by a species?” cannot be pursued MOLLUSCS AS MODELS 11 here in more detail (see Bock 2004, Coyne and Orr 2004, Glaubrecht 2004, Reydon 2004, Hey 2006, Stamos 2007, Samadi and Bar- berousse 2006, 2009), a few remarks should be made on the understanding (and often enough misperception) of what earlier, influential systematists and evolutionary theorists have thought about the nature of species and why this is of relevance in particular today. Aston- ishingly, the species problem is not only regarded as being an old one, but it is often erroneously concluded from this fact that the species c]uestion is either irrelevant or prob- lematic (Noor 2002) and, thus, should be avoided. 1 completely disagree and anticipate that this perception is at least as fatal as our lack of knowledge of species numbers. Firmly convinced about the divine origin of nature and the stability of species, for example, Carl Linnaeus and his followers during the subsequent century believed that ‘'‘'species tot sunt quot diversas formas (in principio) in initio produxit infinitum ens'\ i.e., there are so many species as God has created in the beginning (Blunt 2001: 266). Under this assumption, delimiting species was only a matter of describing the most relevant (i.e., morphological) features for identification and classification. However, during Darwin’s time it became a widely realized problem of how to define the “undefinable”, as the species problem is called. It has repeatedly been erroneously referred to in the literature that Darwin, after having struggled long with the species question, applied a somewhat cynical definition of species being merely what “competent naturalists” say they are. Several statements in Darwin’s work seem to prove this view on his confusion over species that many systematists and evolu- tionary biologists commonly hold (Mallet 1995, 2008a, 2008b, Stamos 2007). However, it seems worth noting that Darwin’s species concept changed, for sound reasons, since the time when he started his early notebooks on the transmutation of species in 1837 and publishing his ‘'‘Origin of Species"' in 1859. As Kottler (1978) and Sulloway (1979) have shown in great detail, Darwin clearly sub- scribed to the reality of species and identified the acquisition of reproductive isolation as a mark of completed speciation, irrespective of his later statements on species as arbitrary Figure 5. Increase of described species in Cephalopoda and in Conus, revealing one trend in two randomly chosen taxa. A, cumulative species number in Cephalopoda, 1758-2006, with species currently considered valid (modified from Wood and Day 2006, Cephbase as of July 2007). B, cumulative species number in the neogastropod Conus, 1758-2007, with available species names given (modified from Kohn and Anderson, 2008, Conus Biodiversity Website as of July 2007). 12 AMERICAN MALACO LOGICAL BULLETIN 27 -1 12- 2009 constructions of taxonomists in relation to his new evolu- tionary theory. This discussion and the Darwinian dilemma on the duality of species as a taxon and concept remain today, in particular with the practical procedure of how to delimit species with the availability of new tools. The reality and duality of species In many respects, modern research has returned to some of Darwin’s original interpretations, as Hendry (2009) stated, referring to the most recent perceptions propagated by Mallet (2008a, 2008b). Ernst Mayr (1904-2005) with his 1942 synthe- sis a watershed in the history of the species-problem, triggered a new age of this debate when elevating several different approaches to species identification to the level of concept (Hey 2006, Glaubrecht 2007, Mallet 2008a, 2008b). Mayr’s biological species concept surely remains a highly useful approach to the study of species and speciation. Evolutionary biologists, in particular those who study the pattern and processes of speciation (see below), believe that species of sexually reproducing organisms are real [i.e., they exist with- out us to assign them) and that they exist by virtue of repro- ductive isolation rather than phenotypic distinctness (Mayr 1996, 2001, Coyne and Orr 2004, Glaubrecht 2004, Hendry 2009). However, as is clearly evident from the vast literature still being produced on this subject, the species problem has still not been adequately resolved from its empirical and practical aspects as well as its philosophical perspective. Persisting misconceptions as to the nature and importance of species, in particular the permanent confusion of species taxon with species concept, hampers both biodiversity research and understanding evolution, leading to pertinent problems with merely artificial delineation and naming of species. In light of the many proposals of species concepts, which have become almost a cottage industry, Bock (1995, 2004) and Mayr ( 1996, 2001 ) have explained that there is a funda- mental distinction as to this duality of species as category and concept. On the one hand, there is the practical procedure of how to delineate a described taxon as a species within a taxonomic category of the Linnean hierarchy; on the other hand, there is the theoretical concept of species using objective criteria, as e.g., defining species based on reproductive isola- tion. As species concepts (the meaning of species in nature) and species taxa (as zoological objects to be categorized in an ordering system) are different things, we need to clearly differentiate between describing species and defining species. As pointed out before (Glaubrecht 2004), these two sets of species problems are often ignored, in particular with respect to the description and identification of species, i.e., the question of the delineation ot species as a practical procetlure using operational and empirical criteria. Unfortunately, re- cently it is largely this latter, operational aspect (how to delimit species using new molecular tools, rather than what it actually represents in nature) that has been focused on (Sites and Marshall 2003, 2004, Agapow et al. 2004, Samadi and Barberousse 2009). As much as these procedures are important, I would argue that having a clear idea what species are remains essential. In malacology, for example, naming nearly every conchologically distinguishable specimen or population in the typological manner of the 1 9'^ century has been misleading (Glaubrecht 2000, 2004). As argued there, our task is not to find the least distinguishable unit but to make meaningful inferences on the existence of evolutionary entities in nature. In terms of testing of the null hypothesis of typology {i.e., what looks different is different), we need to strive for identi- fying not only arbitrarily delineated taxa but also specific entities with their own evolutionary trajectories, using all available biological data including the reproductive criterion of the biospecies concept as well as relevant morphological, molecular genetic, geographic, and other data. Irrespective of the enormous use of molecular markers in systematics, phy- logeny, and phylogeography (Avise 2004, 2006), it is worth realizing that as much as morphology can only be used as a proxy to phenotype, molecular sequence differences are not more than a far less reliable proxy to the genetic history of a given taxon. Currently, with about two dozen species concepts around and in view of the ongoing confusion of species concepts and species taxa (see above), the theoretical foundation of any biodiversity approach is weak. At the same time our under- standing of the nature of species has, of course, fundamental implications for evolutionary biology and historical bioge- ography. Thus, it still is a relevant task to establish if and where the concept of biospecies is applicable or if we need to find and define new and improved concepts. In any case it will be necessary to implement a common language for all biodiver- sity assessments, i.e., to specify those units that are used in individual approaches. As much as it is claimed by proponents of the practice of DNA profiling of animal taxa (see below) that species are recognizable by sequence variation as groups of close relatives, the idea of finding and applying statistical generalities for species differentiation is fallacious. Eventually, it needs to be established whether all units currently used, including MOTU (“molecular operational taxonomic units”) hut also, e.g., chronospecies in the fossil record, carry meaningful information in the global task of a bio-inventory in space and time. 'I’herefore, reiterating an earlier plea (Glaubrecht 2004), we should explicitly say what we mean when dealing with the species question, lor example, which species concept we want to apply and for what reason. As crucial as diagnosing and demarcating species are, only if we understand the nature of species and define more properly the various levels of evolu- tionary units, can we arrive at meaningful conclusions for MOLLUSCS AS MODELS 13 evolutionary biology and biological diversity. It will be the challenge of modern biosystematics to review the various approaches and develop unified strategies for the verification of these units of evolution as a major principal theoretical component, not only for biosystematics but also for any biodiversity assessment and evolutionary biology in general. In the following, I will only discuss a few aspects relevant to this task. Taxonomic redundancy: Which of the nominal species are real? The history of taxonomy and of malacology in particular teaches us that there have been many specimens, populations, or taxa named, thus immortalized until later often synon- ymized. Undoubtedly, superfluous or duplicate names for species as biological entities have been introduced repeatedly, at all times and in all taxonomic groups. Thus, in particular the taxonomic literature for invertebrates often deals with names only but not real species or biological entities. This problem of unresolved synonymies might generally be widely known, but its resolution has not been attempted systematically, hampering many scientific hypotheses ranging from biodiver- sity assessments to speciation theory, models for radiation, and historical biogeography. Actually, only rarely has the degree of taxonomic redundancy that inflates species number estimates been derived from any compilation of catalogues and checklists (see, however, for insects Solow et al. 1995). Alroy (2002) studied this bias by comparing historical rates of invalidation and revalidation of named species using taxonomic data of unrivaled completeness for North American fossil mammal species. From analyzing how many named species remained valid, he concluded that diversity estimates are inflated by 32- 44% for the taxa under study (Fig. 6), arguing that this is a conservative figure compared to other methods. Most impor- tantly, Alroy (2002) suggested from several lines of evidence that the same bias probably affects more poorly studied, hyper-diverse living groups, such as insects and, as I anticipate here, also molluscs. As we have to expect substantial inflation of total species counts everywhere in the literature, Alroy suggested that current estimates of total global diversity should be revised downwards to as low as 3.5 to 10.5 million species, if a third of named species are a reflection created by unsettled taxonomy. Others have argued that his optimistic estimation of fossil coverage, while ignoring the vastly different techniques used to delineate species in living and fossil taxa, render such an extrapolation meaningless ( Agapow et al. 2004: 168). Nevertheless, May and Harvey (2009) estimated that with synonyms removed, the total number of known animal species that is currently assumed to be 1.8 to 1.9 million, may be 1.6 million or fewer. For molluscs, though, a level of taxonomic redundancy that is twice as high as this latter estimate and in the range of Alroy ’s figure might be likely. Schilder (1949) estimated that 34% of the names available to classify prosobranchs were synonyms. He found that 49% of the names that were established in the first half of the 19*'’ century were synonyms, with the synonymy ratio decreasing to 37%, 34%, and 21% respectively for the next consecutive quarter-century intervals. As Bouchet and Rocroi (1992: 84) suggested, this is either the result of taxonomists doing better work since the early 19*'’ century or indicates that it takes many decades to assess the value of a given name. That many molluscan species names are synonyms was also assumed by Boss (1970) who estimated that the synonymy ratio {i.e., the number of available nomina for the number of actual species) averages 4: 1 to 5: 1 , suggesting that a single species may have on the average four to five names in molluscs. His guess was certainly driven by extreme cases like the European swan mussel Anodonta cygnea Linne, 1758 with over 500 names. Bouchet (1997) found, based on a small subsample of his data, a synonymy ratio of 1:6. Extra- polating from the assumption that of the average number of 430 molluscan species described as new each year, only 265 are actually valid, we can calculate a synonymy ratio in molluscs to be as high as 38%, or that 62% of the new species described in the last few decades are indeed new species (compare this, for example, to insects with an estimated inflation rate of 28%, Alroy 2002). It would be interesting to thoroughly investigate the influence of revised biological concepts such as that of what a species is and how we define and delimit it (see above) on our diversity estimates, as discussed by Boss (1978) for bivalves and for freshwater cerithioidean gastropods in Glaubrecht (2004). As I argued, such a study would be all the more timely. Figure 6. How many named species are valid? Growth through his- torical time in the number of all named species compared with those considered valid at a time and thought to be valid today. These data from North American fossil mammals suggest diversity estimates are inflated by 32-44% (from Alroy 2002). 14 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 since the increasing application of merely phylogenetic species concepts, that name least distinguishable units instead of biologically meaningful entities, will contribute to the taxo- nomic inflation. The only study explicitly to review in a comparative manner the influence of, for example, the phylogenetic species concept (resting on the idea of diagnosable differences) versus the biological species concept is hopelessly flawed in respect to the only three examples stated for molluscs, given that one of the three studies cited in Agopow et al (2004: 167, table 1 ) is actually a deep-sea shrimp, while the other two deal with only two species under the BSC (biological species concept) versus one under the PSC (phylogenetic species concept). Apart from the doubtful relevance of these cases, this reverses (as the perplexed authors stated) the commonly expected trend they otherwise found in their study, that surveys based on the PSC lead to recognition of a far greater number of much less inclusive units. Accordingly, under the PSC we would end up with 48% more “species”, which has serious consequences not only for conservation and biodiversity (as it will shift identifying taxa and regions of biodiversity) but also for all biology. Again, however, for molluscs what we know is from very few samples, and what we need is a systematic survey on this taxonomic redundancy and inflation. Taxonomic inflation: In search of the “new” species diversity In general, biodiversity is synonymous with the number of species (at least in neonotology, as paleontologists count numbers of genera or even families, which are artificial hier- archical categories, not natural entities). Consequently, if species were used as a proxy for diversity, not only the theoretical understanding of these units becomes essential for all conclusions, but their number itself Recurrent discoveries using more advanced molecular genetic techniques suggest a plethora of cryptic species and thus hidden diversity, implying that we might largely underrate organismal diversity (Bickford et al. 2006, Pfenninger and Schwenk 2007), possibly including molluscs. Clearly, gene .sequence information for many taxa has contributed to the ascendancy of phylogenetic over biological species concepts. However, any kind of taxonomic inflation, c’.^., when known subspecies are raised to species as a result in change of species concept rather than new discoveries, also has important influences on many fields in biology. It will, for example, result in re- and sub-classifications, heat up bio- diversity hot spots, and lead to devaluation if the smallest distinctions are rai.sed up to the level of what makes a species (Agapow et al. 2004, Isaac et al. 2004). Next to the adoption of the PSC over the BS(i as most recent development in this context, the application of routine methods in molecular taxonomy (such as DNA barcoding) have contributed immen.sely to the view ol biodiversity being underestimated. Undoubtedly, for a long time the lack of obvious morphological differences between taxa has impeded the identification of species in many groups of animals including molluscs. Currently, molecular species identifica- tions boost species diversity figures in all parts of the tree of life (Blaxter 2004, Lee 2004, Savolainen et al. 2005, Kohler 2007, Vogler and Monaghan 2007, Meier 2008). Given the increasing armamentarium of molecular genetic techniques for exploring the genetic structure of populations, species, and higher level taxa, not only do we uncover further complex- ities in what we mean by a species but also we are confronted with modifying and/or adopting the species concepts used to evaluate the level of taxonomic and biologically relevant diversity (May 1990). I have discussed this for freshwater certhioidean species diversity elsewhere (Glaubrecht 2000, 2004). Although more thorough comparative studies in concert with systematic revisions are lacking to substantiate this claim, there are cases to support the caveats. Lor example, Meyer and Paulay (2005), using cypraeid gastropods, showed that the much-debated barcoding approach works well only for species that are already well studied but is hardly an instrument for accelerating species inventories, as there is not a certain and universal threshold. Tackling the “barcoding gap”, i.e., using different means in intra- and interspecific sequence variability for congeneric COI sequences to distin- guish between taxa (Meier et al. 2008), Kohler (2007) dis- cussed for pachychilid gastropods the largely overlapping intra- and interspecific genetic variation. Using uncorrected p-distances in COI that vary between 0.0 and 0. 1 1 within, and 0.0 and 0. 16 between species, he also pointed out the variation even among congeners, which argues against the acceptance of threshold values to postulate species identity, as DNA taxonomy promises something that it cannot deliver. A most striking case is reported by Dillon and Robinson (2009) who found extremely high intra- and inter-population sequence divergence in “living fossil” pleurocerid gastropods from the Appalachians, ranging more than 22% within and between populations, respectively. Using any kind of threshold is a license for splitting over lumping but will not provide an.swer to the e.ssential question of the genetic integrity of species in nature. Neverthele.ss, DNA techniques certainly help as an iden- tification tool, e.y., in marine larvae, or where traditional techniques underestimate species diversity, such as in micro- scopic animals, or to identify complex phenomena such as the existence ol cryptic species complexes, particularly in hyper- diver.se taxa, morphologically very similar taxa, or in young radiations. Recent ca.ses Irom nmllu.scs are marine Conus ( Duda et al. 2008), the vermetid Pendropoina Morch, 1861 (( iaivo et al. 2009), or species complexes in calyptraeids (Collin 2000, 2005) as well as freshwater liythinella Moquin- fandon. MOLLUSCS AS MODELS 15 1 856 (Bichain et al. 2007, Haase et nl. 2007, Benke et al. 2009), and the (presumably truly adaptive) radiation of Tylomelania Sarasin and Sarasin, 1897 on Sulawesi (Rintelen et al. 2007, Glaubrecht and Rintelen 2008). Currently, large-scale DNA sequencing provides new opportunities for the study of species and speciation. However, single short mtDNA gene fragments, as in the barcoding approach, should be used for classification and systematics only if placed in a wider phylogenetic context. Since sequence divergence in itself is an extremely crude method for deter- mining species limits, it is the plurality of systematics tools (including morphological, biogeographical, ecological, and ethological) that needs to be employed in concert. Molecular taxonomy such as DNA barcoding should no longer try to bypass traditional taxonomy but be integrated within and done in concert with relevant systematic knowledge and practices. However, as Isaac et al. (2004) predicted, as practi- cally oriented species concepts such as the PSC gain popular- ity, in particular among molecular taxonomists, the resulting taxonomic inflation will affect more taxa, any biodiversity survey, and even evolutionary studies and hypotheses. It will be interesting, if we will eventually witness this current trend from splitting to lumping to reverse again. As we will only be able to understand biodiversity and evolution fully once we know what species are and how many there are, both issues are of eminent importance and not fully reflected yet in our research activities. (3) Speciation: Darwin’s “mystery of the mysteries” As much as of species themselves and their roles in taxonomy and ecology, a better understanding of speciation is not just an academic exercise but crucial for evolutionary and biodiversity studies. Of course, only if species are real, does the process of speciation become a real process and the study of their formation and transformation become a meaningful scientific endeavor (Glaubrecht 2004). Speciation, as the process of how new species evolve, eventually results in the multiplication of species; thus, speciation is the major factor causing biodiversity and is also fundamental to our understanding of evolution. As much as the debate on what species are has created confusion and controversy for a long time, so has the process that brings species about been dis- cussed. Therefore, speciation remains at the forefront of current evolutionary biology (Mallet 2008a, Nosil 2008, Schluter 2009). Unfortunately, too often taxonomists feel that describing any assemblage of species (living as well as fossil taxa) already justifies the term “speciation” in the titles of their publications. The subject of speciation, though, is much more than just an assemblage of species names. With respect to the above- mentioned “big questions” raised for the near future of biological sciences, studying the underlying mechanisms of diversity still remains a most fascinating subject. Toward solving Darwins mystery In this context, it is important to realize the two main focal points of evolutionary biology, namely anagenesis (natural selection acting to transform existing taxa) versus cladogenesis (the actual multiplication of species). While Darwin’s (1859) work is acknowledged as brilliant induction, presenting insightful examples of adaptation and anagenesis, Mayr’s (1942) synthesis provided an explanation for clado- genesis, grounded chiefly in geography, largely ignored before. Thus, our current understanding of how species multiply was influenced by Mayr’s credo that geographic distribution and spatial isolation play a key role in allopatric speciation, as in the course of the Modern Synthesis of evolution it became firmly established that geographic separation is a major factor in speciation (Mayr 1982, 2001, Coyne and Orr 2004). Glaubrecht (2002, 2007) reviewed the historical avenues and intellectual roots, including the contributions of German systematists of the “Berlin School”, built by Erwin Stresemann, Bernhard Rensch, and Ernst Mayr. Their “New Systematics” was regarded as crucial for providing, in the 1920s and 1930s, the foundation for synthetic evolutionary theory. In the last decade, however, it became apparent that although allopatric speciation is important in many cases, it is not the only mechanism and might not even be as important in evolution as Mayr was convinced throughout most of his long life. It became evident recently that an ecological rather than geographical explanation may account for how new species come into being. A key feature of sympatric speciation (Bolnick and Eitzpatrick 2007), ecological speciation by adaptive divergence is an engine of speciation. Accordingly, as assumed implicitly by Darwin, natural selection and adaptation are the means of generating biodiversity, particu- larly in the context of adaptive radiation (Schluter 1996, 2000a, 2000b, 2001, 2009, Orr and Smith 1998, Sudhaus 2004, Rundle and Nosil 2005, Hendry 2009, Nosil et al. 2009). Instead of geography, factors such as phenotypic plasticity, intra- and interspecific competition, with character displace- ment and specialization, lead to divergence and eventually to speciation. Thus, with ecological speciation as a major player in the diversity of life, allopatric versus sympatric speciation remains one of the most conspicuous current battlegrounds in evolutionary biology. This central controversy is far from being settled as it is not resolved yet what genetic chance will result in a new species, and under which conditions, thus what the true influence of genetics, geography, and adaptation on speciation is. With new genetic research and genomic techniques homing in on the molecular and cellular mechanisms that enable diversification to occur, our current understanding of the speciation process has improved considerably, allowing us to answer to the next generation of questions on the frequency and importance of different processes that cause speciation 16 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 (Noor and Feder 2006). In this context, the long debated phenomenon of interspecific hybridization, a major focus for evolutionary biologists since Darwin, has gained attention lately (Seehausen 2004, Mallet 2005, Schwenk et al. 2008). Hybridization and introgression, with various levels of gene flow, in the absence of strict allopatry are antagonistic to the process of speciation but, nevertheless, might play an important role in adaptive radiation (Seehausen 2004, Gavrilets and Losos 2009). Using molluscs as models Molluscs certainly can contribute to these highly inter- esting and controversial subjects of speciation mechanisms, such as hybridization and adaptive radiation. For molluscs, however, we lack textbook examples for sympatric and ecological speciation, adaptive divergence, and hybridization, such as provided by Darwin’s finches, threespine sticklebacks, or hyper-diverse cichlid fishes. Yet, there are certainly suitable potential candidates, for example molluscs in ancient lakes, from oceanic islands, mountains, or other insular environ- ments with endemic species. It is in this context that we regard the freshwater pachychilid snails Tylomelania Sarasin and Sarasin, 1897, endemic to the central lakes on the Indonesian island of Sulawesi (Glaubrecht and Rintelen 2008) and the thalassoid paludomid snails in East Africa’s Lake Tanganyika (Glaubrecht 2008a) as being truly “Darwinian snails”. Unfor- tunately, too few studies have been experimentally designed and/or conducted in such “natural laboratories”. No attempt will be made here to review those but the papers given during the 2007 WCM meeting in Antwerp, of which some are in the present volume, will certainly indicate how much malacology currently contributes to these issues. As was shown by Schwenk et al. (2008) in their introductory overview on hybridization, not only has the number of publications risen enormously during three phases (from up to the 1990s, from 1991 to 2002, and since 2003), with the third phase driven by the availability of multiple nuclear markers, i.e., microsatellite DNA, AFLP, and SNR But, again using the database of the Zoological Record (for 1947-2007), they found that, when corrected for taxon bias and taxonomic practice ii.e., organismal preferences of research teams and number of species per taxonomic group), the rate of interspecific hybridization is relatively homogeneous among animal groups, approx. 1%, fairly low compared to predictions of about 10% on average in animals (compared to 25% in plants. Mallet 2005). Gonsequently, for mollusc species we have a priori the same chance for finding and studying this evolutionary phenomenon as in other groups where of course, birds, butterflies, and mammals again take a lead. I lowever, as in the case of the term “adaptive radiation” (Glaubrecht 2008b), too often malacologists continue to use evolutionary issues for their paper or book section titles in lieu of a soundly designed study. Instead of perpetuating an outdated concept (inferring adaptive radiation from the taxonomic description of any speciose group), we should investigate with adequate and timely methods founded on solid theoretical background the underlying mechanisms of anagenetic and cladogenetic change. As evolutionary biologists working with molluscs, we should test the universality of the known and discussed speciation mechanisms. To date, these are essentially applied to vertebrates (mostly to birds and fishes), but remain largely untested for various groups of invertebrates with the notable exception of butterflies and dipterans among the insects. It is with a clear focus on the above-mentioned mechanisms, however, that we should choose our molluscan models to increase the frequency with which we utilize them as a source of data for studies on speciation, adaptation, hybridization, and radiation, in order to help decipher the underlying mechanisms of biodiversity. (4) Biogeography: Where are the species and why are they distributed there? Since the time of Darwin and Wallace, biogeography was at the forefront and at the same time an integral part of further developments in evolutionary biology. Despite centu- ries of noting the occurrences of animals and plants and the establishment of biogeography as a scientific discipline in its own right (Glaubrecht 2000), we have too few data on distri- butional patterns for most invertebrates, with, unfortunately, the majority of molluscs being no exception. Nevertheless, it would be easy for us to change this situation. Using more molluscs as model cases would immediately result in an increase in contributions toward this discipline, as is indicated by papers published in Journal of Biogeography (Glaubrecht, unpubl. data). For example, in 1980-2000, a total of 345 taxa- related titles yielded the frustrating rate of only 7% being dedicated to Mollusca, with 80% on non-marine taxa (Fig. 7). On the other hand, our contributions to a symposium on the biogeography of SE Asia in 2000, explicitly employing freshwater gastropods as models, resulted in a share of 28% (out of 25) of the taxa-related papers given. Similar results have been found in a recent review of phylogeography by Beheregaray (2008: 3764). However, aside from noting the data for comprehensive and comparative analyses are generally lacking, 1 will not attempt to review the existing contributions to molluscan biogeography. Instead, I will only mention the two most recent seminal analyses by Reid et al. (2006) and Williams and Duda (2008) as stimulating approaches for future studies of this kind. The Ibrmer employed nn)lecular analy.ses ol littorinid snails in the central Indo-West Pacific to test phylo- geograpliic hypotheses and showed that species from ‘continental’ habitats exhibited strong phylogenetic breaks MOLLUSCS AS MODELS 17 Mollusca 7% Aves 22% Reptilia 3% Pisces 6% Amphibia 3% A. others 6% Annelida 2% Crustacea 8% Mammalia 19% Insecta 243 Reptilia 1 2% others 12% Aves 8% Mollusca 28% Mammalia 16% Insecta 24% ■ Mollusca ■ Mammalia ■ Insecta o Reptilia □ Aves □ others Figure 7. Which animal group is employed in biogeographical stud- ies? A, analysis of papers published in Journal of Biogeography in 1980-2000, with a total of 345 taxa yielding a rate of 7% for Mol- lusca. B, increased percentage of contributions on non-marine mol- luscs given at a symposium on the biogeography of SE Asia in Leiden in 2000. between the Indian and Pacific Ocean, in contrast to ‘oceanic’ species. The latter used analyses of molecular phylogenies of three unrelated tropical marine gastropods, in concert with the timing of plate tectonic events (in this case the collision of the Australia and New Guinea plate with the southeast extremity of the Eurasian plate) during the Oligo-Miocene about 25 million years ago, to show that a speciation pulse occurred in the central Indo-West Pacific at that time. (5) Phylogenetics: How are species (groups) related? To Darwin, although helping contemporaries and future generations embrace the common ancestry of organisms on Earth, large parts of the tree of life remained enigmatic. This holds true even today, to the extent that the phylogenetic relationships not only of major animal phyla but also of many terminal branches are still unresolved. Having replaced earlier intuitive ways of finding or subjectively postulating relation- ships by a clear-cut empirical methodology, it was the German entomologist Willi Henning (1913-1976) who in the 1950s and 1960s single-handedly revolutionized biosystematics. With his guidance, phylogenetic studies were transformed from an art to a truly scientific method, i.e., phylogenetic systematics, creating the basis for a completely new sub- discipline, cladistics (Williams and Forey 2004). It was actually a “silent revolution”, as beyond the limits of systematists this new method remained completely unrecognized, with its implications still being underestimated. Another “silent revolution” took place when phylogenetic systematics was wedded recently with modern molecular genetic methods, first with new sequencing techniques, then with genomic approaches. Albeit those latter developments are too young to be fully applied yet, the contributions in Ponder and Lindberg (2008) provide an up-to-date overview of where we stand 150 years after Darwin in respect to the phylogeny of Mollusca. Gurrently, even newer advances in molecular genetic techniques (see below) open further avenues for research in systematics, phylogenetics, phylogeography and, thus, the study of evolution. (6) Genetic causation and molecular processes: How do new forms come into being? The question of how new forms come into being has long fascinated naturalists. An emerging discipline, called “evo- devo”, currently probes how genes involved in development contribute to evolution. It tries to connect developmental and evolutionary biology and to bridge the gap between the genotype and the phenotype, thus linking DNA-focused studies and molecular evolutionary processes with the ques- tion of biological diversity and disparity (Garroll et al. 2001, Minelli 2003b, 2009b, Garroll 2005, Minelli and Fusco 2008). For an overview in Mollusca, see Wanninger et al. (2008). At the same time, genome sequencing is no longer a pricy luxury done only by few, well-funded researchers on even fewer model organisms. With new technological developments and tools, prices are constantly falling. New methods, such as high-throughput SNP custom cDNA, microarrays, whole- genome and next-generation sequencing (e.g., pyro- sequencing) that can also generate large amounts of expressed sequence tags (EST), are developing faster than ever before, making functional and comparative genomics one of the hottest fields and emerging frontiers of research (Clark 2006, Lee and Mitchell-Olds 2006, Service 2006, Schuster 2008, WTieat 2009). For the current situation in Mollusca, see Simison and Boore (2008). As eventually malacology will go genomic, the overview in Ponder and Lindberg (2008) provides available phylogenetic knowledge from which we can proceed into this new genomic era. I anticipate that within the next decade the continued exploration of genomes and the study of the molecular underpinnings of any aspect of the phenotype of organisms will give us a better knowledge of the genetic architecture and those genes responsible for form and function in many animals. In reviewing an example of first new insights from the functional genomic study of a Haliotis Linnaeus, 1758 18 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 species, Medina (2009: 764) noted “that given the, until recently, unimaginable ease with which we can now develop genomic tools for non-model organisms, examining develop- mental processes in multiple species will be a powerful tool”. She concluded that “the transition of comparing a few genes in few species to whole transcriptomes in an even wider range of species will enhance our understanding of the evolution of animal diversity” (2009: 764), as non-model organisms step into the spotlight. Utilizing these new genomic tools and integrating them with evo-devo studies, phylogenetics, ecol- ogy, and historical biogeography will greatly change and enrich biosystematics and evolutionary biology in the near future, for both evolutionary biologists and malacologists it is indeed an exciting time — and what a long way we came since Edgar Allen Poe pondered about the soft bodies of molluscs. CONCLUSION: NEXT GENERATION BIOSYSTEMATICS— A PROGRAM FOR THE FUTURE As I have tried to show, any biosystematic and evolutionary research on the metazoan level is based on (i) inventorying and describing the basic taxonomic components ( /.e., species), (ii) resolving the phylogenetics of the taxa involved {i.e., the genealogical relationships), and (iii) asking questions about the mechanisms of causation, e.g., modes of speciation {i.e., allopatric, sympatric, ecological), including the ecological and developmental constraints, by studying the molecular underpinnings. However, given the lack of knowledge and baseline data, filling these gaps clearly remains the daunting challenge for biological sciences in general and malacology in particular. For quantifying biodiversity, establishing phylog- enies, and understanding the evolutionary process of specia- tion and radiation as well as the taxonomic diversity and morphological disparity, several prerequisites are indispen- sable, albeit neglected in many modern systematic approaches. Understanding how diversity is shaped, what species are and how they come into being, where species are distributed and how they are related, will require a major interdisciplinary effort, involving many different kinds of studies. Along the way toward clarifying the history of life we will have to eventually solve the six Darwinian mysteries, which involve inventoryingandde-scribing the basic taxonomic components, species, resolving the phylogenetics of taxa by uncovering their genealogical relationships, asking questions about mecha- nisms of causation on all levels, from modes of speciation to evolutionary transformations, and including ecological data as well as a study of developmental constraints and other genetic factors. While evolutionary biologists have just started to .sort out the most relevant factors intertwined in answering these questions, molluscs with their multitude of morphological. genetic, and ecological features have much to offer. They are a highly suitable group for providing fundamental insights into mechanisms of the genesis of biological diversity and disparity, into historical zoogeography, and into the underlying pro- cesses of speciation and radiation, as is also shown in the many contributions to the Antwerp symposium on this subject (Glaubrecht and Rintelen 2009). The myriad of mollusc species offer the material basis for conducting a research program along the lines sketched out above. We still need to discover many missing bricks in the sense of Henri Poincare’s statement used as an epitaph here for this article. Surely, molluscs will not only provide many of the yet missing pieces, but we should also strive to integrate new findings based on this extraordinary taxonomic group into a more comprehensive synthesis of forces and factors influ- encing biological diversity. ACKNOWLEDGMENTS Thanks are due to the participants of the above-men- tioned symposium during the WCM Antwerp meeting, their input in discussions, and the authors for contributing to this symposium volume. I thank Thomas von Rintelen for helping in organizing the symposium and in editing the manuscripts for this special volume, for compiling the figures shown here, as well as Kristina von Rintelen for her help with the graph- ics. I am indebted to Ken Brown, who allowed us to put a selection of symposium contributions together, and in par- ticular for his patience in the course of our doing so. My own work referred to herein was substantially supported by several research grants from the Deutsche Forschungsgemeinschaft (DEG GL 297/1-1, 1-2 and 297/7-1, 7-2, and 7-3) that also provided travel funds for the 2007 Antwerp congress. LITERATURE CITED Adamowicz, S. J. and A. lAirvis. 2005. How many branchiopod crus- tacean species are there? Quantifying the components of un- derestimation. CAobnl Ecology aiui Biogcognipliy 14: 455-408. Agapow, P.-M., O. R. P. Bininda-Hmonds, K. A. tirandall, |. 1.. Gittle- man, G. M. Mace, f C. Marshall, and A. I’urvis. 2004. ’Hie im- pact of species concept on biodiversity studies. 'I'lic Qinirtcrly Review oj Biology 79: 161-1 79. Allen, 1). E. 2008. Stamp collecting and natural history. Archive of Niilurol HisloryiS: 172-175. Appel, '!'. 1987. I'he Ciivier-(.k'offroy Pehote.Oxlowl University i’ress, Oxford, UK. Alroy, ). 2002. 1 low many named species are valid? Proceedings of the Nalionol Academy of Sciences 99: 3706-371 1. Avise, |. (',. 2004. Molecular Markers, Nalnnil History, and Evolution. University i’ress, (iamhridge, UK. MOLLUSCS AS MODELS 19 Avise, J. C. 2006. Evolutionary Pathways in Nature. A Phylogenetic Approach. Sinauer Associates, Sunderland, Massachusetts. Baillie, J. E. M., C. Hilton-Taylor, and S. N. Stuart. 2004. lUCN Red List of Threatened Species. A Global Species Assessment. The lUCN Species Survival Commission, Gland, Switzerland, Cam- bridge, UK. Barton, N. H., D. E. G. Briggs, J. A. Eisen, D. B. Goldstein, and N. H. Patel. 2007. Evolution. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. Beheregaray, L. 2008. Twenty years of phylogeography: The state of the field and the challenges for the Southern Hemisphere. Molecular Ecology 17: 3754-3774. Benke, M., M. Brandle, C. Albrecht, and T. Wilke. 2009. Pleistocene phylogeography and phylogenetic concordance in cold-adapted spring snails [Bythinella spp.). Molecular Ecology 18: 890-903. Bichain, J.-M., P. Gaubert, S. Samadi, and M.-C. Boisselier-Dubayle. 2007. A gleam in the dark: Phylogenetic species delimitation in the confusing spring-snail genus Bythinella Moquin-Tandon, 1856 (Gastropoda: Rissoidea: Amnicolidae). Molecular Phylo- genetics and Evolution 45: 927-941. Bickford, D., D. J. Lohman, N. S. Sohi, P. K. L. Ng, R. Meier, K. Wink- er, K. K. Ingram, and I. Das. 2006. Cryptic species as a window on diversity and conservation. Trends in Ecology and Evolution 22: 151-155. Bieler R. and P. M. Mikkelsen. 1992. Handbook of Systematic Mala- cology. Smithsonian Institution Libraries and National Science Eoundation, Washington, D.C. Blackmore, S. 2002. Biodiversity update - progress in taxonomy. Science 298: 365. Blaxter, M. L. 2004. The promise of a DNA taxonomy. Philosophical Transactions of the Royal Society London (B) 359: 669-679. Blunt, W. 2001. Linnaeus. The Complete Naturalist. Frances Lincoln, London. Bock, W. 1995. The species concept versus the species taxon: Their roles in biodiversity analyses and conservation. In: R. Arai, M. Kato, and Y. Doi, eds.. Biodiversity and Evolution. The Natural Science Museum Foundation, Tokyo. Pp. 47-72. Bock, W. J. 2004. Species: The concept, category and taxon. Journal of Zoological Systematics and Evolutionary Research 42: 178-190. Bolnick, D. I. and B. M. Fitzpatrick. 2007. Sympatric speciation: Models and empirical evidence. Annual Reviews in Ecology, Evolution and Systematics 38: 459-487. Boss, K. J. 1970. How many species of mollusks are there? Annual Reports of the American Malacological Union 1970: 41. Boss, K. J. 1971. Critical estimate of the number of Recent Mollusca. Occasional Papers on Molluscs 3: 81-135. Boss, K. J. 1978. Taxonomic concepts and superfluity in bivalve nomenclature. Philosophical Transactions of the Royal Society London (B) 284:417-424. Bouchet, P. 1997. Inventorying the molluscan diversity of the world: What is our rate of progress? The Veliger 40: 1-11. Bouchet, P. 2006. The magnitude of marine biodiversity. In: C. M. Duarte, ed.. The Exploration of Marine Biodiversity. Scientific and Technological Challenges. Fundacion BBVA, Bilbao, Spain. Pp. 31-64. Bouchet, P. and J.-P. Rocroi. 1992. Supraspecific names of molluscs: A quantitative review. Malacologia 34: 75-86. Bouchet, P. and J.-P. Rocroi. 1993. The lottery of bibliographical data- bases: A reply to Edwards & Thorne. Malacologia 35: 407-410. Bouchet, P, P. Lozouet, P. Maestrati, and V. Heros. 2002. Assessing the magnitude of species richness in tropical marine environments: Exceptionally high numbers of molluscs at a New Caledonia site. Biological fournal of the Linnean Society 75: 421-436. Brown, T. 1833. The Conchologist’s Text-book, Embracing the Arrange- ments of Lamarck and Linnaeus. A. Fullarton, Glasgow. Burkhardt, R. W. 1995. The Spirit of Systematics. Lamarck and Evolutionary Biology. Harvard University Press, Cambridge, Massachusetts. Calvo, M., ]. Templado, M. Oliverio, and A. Machordom. 2009. Hid- den Mediterranean biodiversity: Molecular evidence for a cryp- tic species complex within the reef building vermetid gastropod Dendropoma petraeum (Mollusca: Caenogastropoda). Biologi- cal Journal of the Linnean Society 96: 898-912. Carroll, S. B. 2005r Endless Forms Most Beautiful. The New Science of Evo Devo and the Making of the Animal Kingdom. W. W. Norton, New York. Carroll, S. B., J. K. Grenier, and S. D. Weatherbee. 2001. From DNA to Diversity. Molecular Genetics and the Evolution of Animal Design. Blackwell Publications, Oxford. Chapman, A. D. 2005. Numbers of Living Species in Australia and the World. Commonwealth of Australia, Canberra. [Online version with updates available at www.environment.gov.au April 2007]. Clark, A. G. 2006. Genomics of the evolutionary process. Trends in Ecology and Evolution 21: 316- 321. Collin, R. 2000. Phylogeny of the Crepidula plana (Gastropoda: Ca- lyptraeidae) cryptic species complex in North America. Cana- dian Journal of Zoology 78: 1500-1514. Collin, R. 2005. Development, phylogeny, and taxonomy of Bostry- capulus (Caenogastropoda: Calyptraeidae), an ancient cryptic radiation. Zoological Journal of the Linnean Society 144: 75-101. Condon, M. A., S. J. Scheffer, M. L. Lewis, and S. M. Svensen. 2008. Hidden neotropical diversity: Greater than the sum of its parts. Science 320: 928-931. Corsi, P. 1988. The Age of Lamarck. Evolutionary Theories in France 1790-1830. University of California Press, Berkeley. Coyne, J. A. and H. A. Orr. 2004. Speciation. Sinauer Associates, Sun- derland, Massachusetts. Dayrat, B. 2005. Towards integrative taxonomy. Biological Journal of the Linnean Society 85: 407-415. Darwin, C. 1859. On the Origin of Species by Means of Natural Selec- tion, or the Preservation of Favoured Races in the Struggle for Life. John Murray, London. Diamond, J. M. 1985. How many unknown species are yet to be dis- covered? Nature 315: 538-539. Dillon, R. T. and J. D. Robinson. 2009. The snails the dinosaurs saw: Are the pleurocerid populations of the Older Appalachians a relic of the Paleozoic Era? Journal of North American Bcntho- logical Society 28: 1-11. Dolphin, K. and D. L. J. Quicke. 2001. Estimating the global species richness of an incompletely described taxon: An example using parasitoid wasps (Hymenoptera: Braconidae). Biological Jour- nal of the Linnean Society 73: 279-286. Dyer, L. A., M. S. Singer, J. T. Fill, J. O. Stireman, G. L. Gentry, R. J. Marquis, R. E. Ricklefs, H. F. Greeney, D. L. Wagner, H. C. 20 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Morals, I. R. Diniz, T. A. Kursar, and R D. Coley. 2007. Host specificity of Lepidoptera in tropical and temperate forests. Nature 448\ 696-699. Duda, T. R, M. B. Bolin, C. P. Meyer, and A. J. Kohn. 2008. Hidden diversity in a hyperdiverse gastropod genus; Discovery of previ- ously unidentified members of a Conus species complex. Mo- lecular Phylogenetics and Evolution 49: 867-876. Edwards, M. A. and M. J. Thorne. 1993. Reply to ‘Supraspecific names of molluscs: A quantitative review’. Malacologia 35: 153-154. Erwin, T. 1982. Tropical forests: Their richness in Coleoptera and other species. Coleopterist’s Bulletin 36: 74-75. Evenhuis, N. L. 2007. Helping solve the “other” taxonomic impedi- ment: Completing the eight steps to total enlightenment and taxonomic nirvam. Zootaxa 1407: 3-12. Futuyma, D. 1. 1997. Evolutionary Biology, 3rd Edition. Sinauer As- sociates, Sunderland, Massachusetts. Gavrilets, S. and J. B. Losos. 2009. Adaptive radiation: Contrasting theory with data. Science 323: 732-737. Glaubrecht, M. 2000. A look back in time: Toward an historical biogeography as synthesis of systematic and geologic patterns outlined with limnic gastropods. Zoology: Analysis of Complex Systems 102; 127-147. Glaubrecht, M. 2002. The “experience” of nature: From Salomon Muller to Ernst Mayr, or the insights of travelling naturalists toward a zoological geography and evolutionary biology. Ver- handlungen zur Geschichte und Theorie der Biologie 9: 245-282. Glaubrecht, M. 2003. Arten, Artkonzepte und Evolution - Was sind und wie entstehen “biologische Arten”? In: ]. Reichholf, ed., Biologische Vielfalt - Sammeln, Sammlungen, Systematik. Rundgesprache der Komrnission fiir Okologie, Bd. 26. Verlag Dr. Friedrich Pfeil, Miinchen. Pp. 15-42 [In German]. Glaubrecht, M. 2004. Leopold von Buch’s legacy: Treating species as dynamic natural entities, or why geography matters. American Malacological Bulletin 19: 111-134. Glaubrecht, M. 2005. Seitenspriinge der Evolution. Machos und an- dere Mysterien der Biologie. S. Hirzel Verlag, Stuttgart, Leipzig [In German]. Glaubrecht, M. 2007. Die Ordnung des Lebendigen. Zur Geschichte und Zukunft der Systematik in Deutschland. In: J.-W. Wagele, ed., Jubilaumshand zur 100. DZG-Tagung in Koln im September 2007. Basiliken Press, Marburg. Pp. 59-1 10 [In German]. Glaubrecht, M. 2008a. Adaptive radiation of thalassoid gastropods in Lake Tanganyika, East Africa; Morphology and systematiza- tion of a paludomid species flock in an ancient lake. Zoosystem- atics and Evolution 84; 7 1 - 1 22. Glaubrecht, M. 2008b. Hard facts about soft animals [Review of Ponder and Li ndberg, 2008]. Snence 320: 1014-1015. (ilaubrecht, M. 2008c. Homage to Karl Augu.st Mobius { 1825-1908) and his biological contributions: Zoologist, ecologist, and di- rector at the Museum fur Naturkunde in Berlin. Zoosysicmatics and Evolution 84: 7-28. Glaubrecht, M. and T. v. Rintelen. 2008. The species Hocks of lacus- trine gastropods: Tylomelania on Sulawesi as models in specia- tion and adaptive radiation. Hydrobiologia 615: 181-199. Glaubrecht, M. and T. v. Rintelen. 2009. From Poe to Ponder ... and Lindberg: Introduction to the .symposium “Mollu.scs as models in evolutionary biology”. A/nen'cfln Malacological Bulletin 27: 1-2. Glaubrecht, M., M. Maitas, and L. Salvini-Plawen. 2005. Aplacoph- oran Mollusca in the Natural History Museum Berlin. An an- notated catalogue of Thiele’s type specimens, with a brief re- view of “Aplacophora” classification. Mitteilungen Museum fur Naturkunde Berlin, Zoologisches Reihe 81; 145-166. Godfray, H. C. J. 2002. Challenges for taxonomy. Nature 417: 17-19. Godfray, H. C. ]. 2007. Linnaeus in the information age. Nature 446: 259-260. Gould, S. J. 1993. This view of life: Poe’s greatest hit. Natural History 102: 10-19. Gould, S. J. 1995. Dinosaur in a Haystack. Reflections in Natural His- tory. Harmony Books/Crown Publishers, New York. Haase, M., T. Wilke, and P. Mildner. 2007. Identifying species of Bythinella (Caenogastropoda: Rissoidea): A plea for an integra- tive approach. Zootflxu 1563: 1-16. Heartman, C. F. and J. R. Canny. 1943. A Bibliography of Eirst Print- ings of the Writings of Edgar Allan Poe. The Book Farm, Hat- tiesburg, Mississippi. Hendry, A. R. 2009. Speciation. Nature 458: 162-163. Hey, L 2006. On the failure of modern species concepts. Trends in Ecology and Evolution 21: 447-450. Isaac, N. ]. B., L Mattel, and G. M. Mace. 2004. Taxonomic infla- tion: Its influence on macroecology and conservation. Trends in Ecology and Evolution 19: 464-469. Johnson, K. 2005. Ernst Mayr, Karl Jordan, and the history of sys- tematics. History of Science 43: 1-35. Keen, A. M. 1936. Edgar Allan Poe’s conchological text. The Nautilus 50: 42-44. Kennedy, D. and C. Norman. 2005. What don’t we know? Science 309: 75. Keynes, R. D. 2003. Fossils, Finches and Fuegians. Darwin’s Adven- tures and Discoveries on the Beagle. Oxford University Press, Oxford, UK. Kohler, F. 2007. From DNA taxonomy to barcoding - how a vague idea evolved into a biosystematic tool. Mitteilungen aus dent Museum fiir Naturkunde Berlin, Zoologische Reihe 83: 44-5 1 . Kohn, A. J. and T. Anderson. 2008. The Conus Biodiversity website (October 2008). Available at: http://biology.burke. Washington, edu/conus/ 18 October 2008. Kottler, M. J. 1978. Charles Darwin’s biological species concept and theory of geographical speciation: The transmutation note- books. Annals of Science 35: 275-297. Lamarck, J. B. 1 792. Observations sur les coquilles, et sur quelques’uns des genres qu’on a ctablis dans I’ordre des vers testaces. lournal d’histoire naturcllel: 269-280 [In French]. Lamarck, J. B. 1801. Systeme des Animaux sans \'crtcbrcs, ou Tableau Gaicral des Classes, des Ordcs ct des Genres des ees Animaux ... Precede du Diseours d'Ouverture du Cours de Zoologie, Domie Dans le Museum National d’Histoire Naturelle, Pan \TU de la Republique [ 1800]. I’aris ]ln French]. Lamarck, |. B. 1809. Philosophie Zoologiijue, ou Exposition des Con- siderations Relatives a I'llistoire Naturelle lies Animaux. 2 vols. Paris [In French ]. Lamarck, |. B. 1815-1822. Histoire Naturelle des Animaux sans Verte- bres. 7 vols. Paris. MOLLUSCS AS MODELS 21 Laurent, G. 1997. Jean-Baptiste Lamarck, 1744-1829. Comite des travaux historiques et scientifiques, Paris [In French], Lawler, A. 2001. Up for the count? Science 294: 769-770. Lee, M. S. Y. 2004. The molecularisation of taxonomy. Invertebrate Systematics 18: 1-6. Lee, C. E. and T. Mitchell-Olds. 2006. Preface to the special issue: Ecological and evolutionary genomics of populations in na- ture. Mo/ecu/ar Ecology 15: 1193-1196. Le Guyader, H. 2004. Geoffroy Saint-Hilaire: A Visionary Naturalist. University of Chicago Press, Chicago. Linnaeus, C. 1758. Systema Naturae, 10th Edition. Stockholm, fasci- mile vol. 1. London, 1956, vol. 2 Weinheim 1964. Mallet, ]. 1995. A species definition for the Modern Synthesis. Trends in Ecology and Evolution 10: 294-299. Mallet, J. 2005. Hybridization as an invasion of the genome. Trends in Ecology and Evolution 20: 119 -Til . Mallet, J. 2008a. Hybridization, ecological races and the nature of species: Empirical evidence for the ease of speciation. Philo- sophical Transactions of the Royal Society London (B) 363: 2971- 2986. Mallet, J. 2008b. Mayr’s view of Darwin: Was Darwin wrong about speciation? Biological Journal of the Linnean Society 95: 3-16. Mallet, J. and K. Willmott. 2003. Taxonomy: Renaissance or tower of Babel? Trends in Ecology and Evolution 18: 57-59. May, R. M. 1988. How many species are there on Earth? Science 241: 1441-1449. May, R. M. 1990. How many species? Philosophical Transactions of the Royal Society London (B) 330: 293-304. May, R. M. 1992. How many species inhabit the Earth? Scientific American 10: 18-24. May, R. A4. 1994. Conceptual aspects of the quantification of the extent of biological diversity. Philosophical Transactions of the Royal Society London (B) 345: 13-20. May, R. M. 1999. The dimensions of life on earth. In: P. H. Raven and T. Williams, eds.. Nature and Human Society: The Quest for a Sustainable World. National Academy Press, Washington, D.C. Pp. 30-45. May, R. M. 2004. Tomorrow’s taxonomy: Collecting new species in the field will remain the rate-limiting step. Philosophical Trans- actions of the Royal Society London (B) 359: 733-734. May, R. M. and P. H. Harvey. 2009. Species uncertainties. Science 323: 687. Mayr, E. 1942. Systematics and the Origin of Species. Columbia Uni- versity Press, New York. Mayr, E. 1982. The Growth of Biological Thought. The Belknap Press of Harvard University Press, Cambridge, Massachusetts. Mayr, E. 1996. What is a species, and what is not? Philosophy of Sci- ence 63: 262-277. Mayr, E. 2001. What Evolution Is. Basic Books, New York. Medina, M. 2009. Eunctional genomics opens doors to understand- ing metamorphosis in nonmodel invertebrate organisms. Mo- lecular Ecology 18: 763-764. Meier, R. 2008. DNA sequences in taxonomy. Opportunities and challenges. In: Q. D. Wheeler, ed.. The New Taxonomy. The Sys- tematic Association Special Volume Series 76. CRC Press, Boca Raton. Pp. 95-128. Meier, R., G. Tang, and F. Ali. 2008. The use of mean instead of small- est interspecific distances exaggerates the size of the “barcod- ing gap” and leads to misidentification. Systematic Biology 57: 809-813. Meyer, C. and G. Paulay. 2005. DNA barcoding: Error rates based on comprehensive sampling. PlosBiology 3: 2229-2238. Mikkelsen, P. M. and J. Cracraft. 2001. Marine biodiversity and the need for systematic inventories. Bulletin of Marine Science 69: 525-534. Minelli, A. 2003a. The Development of Animal Form. Ontogeny, Morphology, and Evolution. Cambridge University Press, Cam- bridge, UK. Minelli, A. 2003b. The status of taxonomic literature. Trends in Ecol- ogy and Evolution 18: 75-76. Minelli, A. 2009a. Forms of Becoming. The Evolutionary Biology of Development. Princeton University Press, Princeton. Minelli, A. 2009b. Perspectives in Animal Phylogeny and Evolution. Oxford University Press, Oxford, UK. Minelli, A. and G. Fusco. 2008. Evolving Pathways. Key Themes in Evolutionary Developmental Biology. Cambridge University Press, Cambridge, UK. Moldenhauer, J. J. 1971. Beyond the Tamarind Tree: A new Poe letter. American Literature 42: 468-477. Mora, C., D. P. Tittensor, and R. A. Myers. 2008. The completeness of taxonomic inventories for describing the global diversity and distribution of marine fishes. Proceedings of the Royal Society (B) 275: 149-155. Muller, S. W. M. and A. Campbell. 1954. The relative number of living and fossil species of animals. Systematic Zoology 3: 168-170. Nicol, D. 1969. The number of living species of molluscs. Systematic Zoology 18: 251-254. Noor, M. A. F. 2002. Is the biological species concept showing its age? Trends in Ecology and Evolution 17: 153-154. Noor, M. A. F. and J. L. Feder. 2006. Speciation genetics: Evolving ap- proaches. Nature Reviews Genetics 7: 851-861. Nosil, P. 2008. Ernst Mayr and the integration of geographical and ecological factors in speciation. Biological Journal of the Linnean Society 95: 26-46. Nosil, R, L. J. Harmon, and O. Seehausen. 2009. Ecological explana- tions for (incomplete) speciation. Trends in Ecology and Evolu- tion 24: 145-156. Novotny, V., Y. Basset, S. E. Miller, G. D. Weiblen, B. Bremer, L. Cizek, and P. Drozd. 2002. Low host specificity of herbivorous insects in a tropical forest. Nature 416: 841-844. Novotny, V., S. E. Miller, J. Hulcr, R. A. I. Drew, Y. Basset, M. Janda, G. P. Setliff, K. Darrow, A. J. A. Stewart, J. Auga, K. Molem, M. Manumbor, E. Tamtiai, M. Mogia, and G. D. Weiblen. 2007. Low beta diversity of herbivorous insects in tropical forests. Nature 448: 692-695. Odegaard, F. 2000. How many species of arthropods? Erwin’s esti- mate revised. Biological Journal of the Linnean Society 71: 583- 597. Orr, M. R. and T. B. Smith. 1998. Ecology and speciation. Trends in Ecology and Evolution 13: 502-506. Pennisi, E. 2005. What determines species diversity? Science 309: 90. Pfenninger, M. and K. Schwenk. 2007. Cryptic animal species are 22 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 homogeneously distributed among taxa and biogeographical regions. BMC Evolutionary Biology 7: 121. Poe, E. A. 1839. The Conchologist’s First Book: Or, A System of Testa- ceous Malacology, Arranged Expressly for the Use of Schools. Has- well, Barrington and Philadelphia. Poincare, 1. H. 1902. La science et Phypothese. Flammarion, Paris (cited from the English edition: Science and hypothesis, trans- lated by G. B. Elalsted. Scott, London, New York. 1905). Polaszek, A. 2005. A universal register for animal names. Nature 437: 477. Ponder, F. W. and D. R. Lindberg. 2008. Phylogeny and Evolution of the Mollusca. University of California Press, Berkeley. Ponder, F. W. and D. Lunney. 1999. The other 99%. The Conservation and Biodiversity of Invertebrates. Transactions of the Royal Zoo- logical Society of New South Wales, Mosman. Raven, P. H. and T. Williams. 1997. Nature and Human Society. Pro- ceedings of the 1997 Forum on Biodiversity. National Academic Press, Washington, U.C. Raven, P. H. and E. O. Wilson. 1992. A fifty-year plan for biodiversity surveys. Science 258: 1099- 1 100. Reid, D., K. Lai, J. Mackenzie-Dodds, F. Kaligis, D. T. I. Littlewood, and S. T. Williams. 2006. Comparative phylogeography and species boundaries in Echinolittorina snails in the central Indo- West Pacific. Journal of Biogeography 33: 990-1006. Reydon, T. A. C. 2004. Why does the species problem still persist? BioEssays 26: 300-305. Ridley, M. 1996. Evolution, 2"‘' Edition. Blackwell Science, Cam- bridge, Massachusetts. Rintelen, T. v. and M. Glaubrecht. 2003. New discoveries in old lakes: Three new species of Tylomelania Sarasin & Sarasin, 1897 (Gastropoda: Cerithioidea: Pachychilidae) from the Malili lake system on Sulawesi, Indonesia. Journal of Molluscan Studies 69: 3-17. Rintelen, T. v., P. Bouchet, and M. Glaubrecht. 2007. Ancient lakes as hotspots of diversity: A morphological review of an endemic species flock of Tylomelania (Gastropoda: Cerithioidea: Pachy- chilidae) in the Malili lake system on Sulawesi, Indonesia. Hy- drobiologia 592: 1 1 -94. Rundle, LL I), and P. Nosil. 2005. Ecological speciation. Ecology Let- ters 8: 336-352. Samadi, S. and A. Barberousse. 2006. The tree, the network, and the species. Biological Journal of the Linnean Society 89: 509-521. Samadi, S. and A. Barberousse. 2009. Species: Towards new, well- grounded practices. Biological Journal of the Linnean Society 97: 217-222. Savolainen, V., R. S. Cowan, A. P. Vogler, C. K. Roderick, and R. Lane. 2005. Towards writing the encyclopaedia of life: An introduc- tion to DNA barcoding. Philosophical I'ransaclions of the Royal Society London ( B ) 360: 1 805- 1811. Schildcr, E A. 1947. Die Zahl der Pro.sobranchier in Vergangcnheit und (iegenwart. Archiv fi'tr Molluskenkunde 76: 37-44 (In Ger- man |. Schilder, E. A. 1948. Wie vide Tierarten gibt es? L'orschungen und L'ortschritle 24: 42-45 [In German]. Schilder, FA. 1949. Statistical notes on malacology. Proceedings of the Malacological Society oj lAmdon 27: 259-26 1 . Schilling, D. 1989. Biographische und problemgeschichtliche Einlei- tung. In: J. P. Lamarck, Zoologische Philosophie (Teil 1-3). Wis- senschaftlicher Verlag Elarri Deutsch, Frankfurt am Main. Pp. 8-43 [In German] [translated by A. Lang; cited from the 2002 edition]. Schluter, D. 1996. Ecological causes of adaptive radiation. The Amer- ican Naturalist 148: 40-63. Schluter, D. 2000a. Ecological character displacement in adaptive ra- diation. The American Naturalist 156: 4-16. Schluter, D. 2000b. The Ecology of Adaptive Radiation. Oxford Uni- versity Press, Oxford, UK. Schluter, D. 200 1 . Ecology and the origin of species. Trends iti Ecology and Evolution 16: 372-380. Schluter, D. 2009. Evidence for ecological speciation and its alterna- tive. Science 323: 737-741. Schuster, S. C. 2008. Next-generation sequencing transforms today’s biology. Nature Methods 5: 16-18. Schwenk, K., N. Brede, and B. Streit. 2008. Introduction. Extent, pro- cess and evolutionary impact of interspecific hybridization in animals. Philosophical Transactions of the Royal Society London (B) 363: 2805-2810. Seehausen, O. 2004. Hybridization and adaptive radiation. Trends in Ecology and Evolution 19: 198-207. Service, R. E 2006. The race for the $1000 genome. Science 311: 1544-1546. Simison, W. B. and |. L. Boore. 2008. Molluscan evolutionary ge- nomics. In: F. W Ponder and D. R. Lindberg, eds, Phylogeny and Evolution of the Mollusca. University of California Press, Berke- ley. Pp. 447-461. Sites, ). W. and J. C. Marshall. 2003. Delimiting species: A renaissance issue in systematic biology. Trends in Ecology and Evolution 18: 462-470. Sites, J. W. and ). C. Marshall. 2004. Operational criteria for delimit- ing species. Annual Review of Ecology, Evolution, and Systemat- ics 35: 199-227. Solow, A. R., L. A. Mound, and K. |. Gaston. 1 995. Estimating the rate of synonymy. Systematic Biology 44: 93-96. Stamos, D. N. 2007. Darwin atid the Nature of Species. State Univer- sity of New York Press, Albany, New York. Stork, N. E. 1988. Insect diversity: Facts, fiction and speculation. Bio- logical Journal of the Linnean Society 35: 321-337. Stork, N. E. 1993. How many species are there? Biodiversity and Con- servation 2: 2 1 5-232. Stork, N. E. 2007. Biodiversity: World of insects. Nature 448: 657-658. Sudhaus, W. 2004. Radiation within the framework of evolutionary ecology. Organisms, Diversity and Evolution 4: 127-134. Sulloway, E |. 1979. Geographical isolation in Darwin’s thinking: The vicissitudes of a crucial idea. Studies in the History of Biol- ogy 3: 23-65. rhicle, ). 1929-1931. Handbuch der Systematischcn Weichtierkumle. G. I'i.scher, |ena [In German]. Troschel, E i I. 1856-1863. Das Cebiss iler Schnecken zur Begriindung einer natiirlichen Classification. Nicolai.sche Verlag.sbuchhandlung, Berlin [ In German]. Valentine, |. W. 2004. On the Origin of Phyla. University of Chicago Pre.ss, Chicago. MOLLUSCS AS MODELS 23 Vaught, K. C. 1989. A Classificotion of the Living Mollusca. American Malacologists, Melbourne, Florida. Vermeij, G. J. 1999. The accumulation of taxonomic knowledge: The history of species descriptions of some predatory gastropods. Malacologia 41: 147-150. Vogler, A. R and M. T. Monaghan. 2007. Recent advances in DNA taxonomy. Journal of Zoological Systematics and Evolutionary Research 45: 1-10. Wanninger, A., D. Koop, S. Moshel-Lynch, and B. M. Degnan. 2008. Molluscan evolutionary development. In: F. W. Ponder and D. R. Lindberg, eds., Phylogeny and Evolution of the Mollusca. Uni- versity of California Press, Berkeley. Pp. 427-445. Wheat, C. W. 2009. Rapidly developing functional genomics in eco- logical model systems via 454 transcriptome sequencing. Ge- netica (in press). Wheeler, Q. D. 2004. Taxonomic triage and the poverty of phylogeny. Philosophical Transactions of the Royal Society London (B) 359: 571-583. Wheeler, Q. D., P. H. Raven, and E. O. Wilson. 2004. Taxonomy: Im- pediment or expedient? Science 303: 285. Whitehead, P. 1990. Systematics: An endangered species. Systematic Zoology 39: 179-184. Williams, D. M. and P. L. Forey. 2004. Milestones in Systematics. Es- says from a Symposium. The Systematics Association Special Volume series 67. CRC Press, Boca Raton, Florida. Williams, S. T. and T. F. Duda. 2008. Did tectonic activity stimulate Oligo-Miocene speciation in the Indo-West Pacific? Evolution 62: 1618-1634. Wilson, E. O. 1988. Biodiversity. Papers from the National Forum on Biodiversity. National Academic Press, Washington, D.C. Wilson, E. O. 1989. The coming pluralization of biology and the stewardship of systematics. BioScience 39: 242-245. Wilson, E. O. 1992. The Diversity of Life. The Belknap Press of Har- vard University Press, Cambridge, Massachusetts. Wilson, E. O. 2000. A global biodiversity map. Science 289: 2279. Wilson, E. O. 2003. The encyclopedia of life. Trends in Ecology and Evolution 18: 77-80. Wood, J. B. and C. L. Day. 2006. CephBase. A database-driven web site on all living cephalopods (octopus, squid, cuttlefish and nautilus). Available at: www.cephbase.utmb.edu/ 3 June 2009. Wyatt, T. 1838. A Manual of Conchology, According to the System Laid Down by Lamarck. Harper and Brothers, New York. Submitted: 1 1 May 2009; accepted: 2 June 2009; final revisions received: 9 June 2009 » Amer. Make. Bull. 27: 25-45 (2009) As time goes by: A simple fooPs guide to molecular clock approaches in invertebrates'^ Thomas Wilke, Roland Schultheifi, and Christian Albrecht Department of Animal Ecology and Systematics, Justus Liebig University Giessen, Heinrich-Buff-Ring 26-32 IFZ, D-35392 Giessen, Germany Corresponding author: tom.wilke@allzool.bio.uni-giessen.de Abstract: Biologists have used a wide range of organisms to study the origin of taxa and their subsequent evolutionary change in space and time. One commonly used tool is the molecular clock approach, relating substitution rates of nucleotide or amino acid sequences to divergence times. The accuracy of the molecular clock, however, has long been subject to controversy, and numerous papers have addressed problems associated with estimating divergence times. Some workers pointed out a striking imbalance between sophisticated software algorithms used for molecular clock analyses on the one hand, and the poor data on the other hand. Moreover, there is often unease among workers relative to molecular clocks because of the controversy surrounding the approach, the complex mathematical background of many molecular clock tools, the still limited number of available, user-friendly software packages, the often confusing terminology of molecular clock approaches, and the general lack of reliable calibration points and/or external clock rates. The current review therefore briefly provides an overview of analytical strategies, covering approaches based on calibration points and/or bounds, approaches based on external clock rates, and approaches that attempt to estimate relative divergence times in the absence of information that can be used for estimating substitution rates. It also deals with major problems and pitfalls associated with data and analyses, including potential errors of calibration points and bounds, the performance of the gene(s) used, estimation of confidence limits, and misinterpretation of the results of clock analyses due to problems with sampling design. A substantial part of the review addresses the question of “universal” molecular clock rates and summarizes important biological and life history variables that account for deviations from rate constancy both between lineages and at different times within lineages. The usefulness of these factors is discussed within the framework of “trait-specific” molecular clock rates. One such clock rate is introduced here for the cytochrome c oxidase subunit I (COI) gene in small dioecious, tropical and subtropical Protostomia with a generation time of approximately one year. A flow chart is provided as a “simple fool’s guide” to molecular clock analyses, together with a glossary of widely used terms in molecular clock approaches. Finally, step-by-step examples are provided for calculating divergence times in the caenogastropod subfamily Pyrgulinae based on both an internal calibration point and a “trait-specific” molecular clock rate, and it is demonstrated how a relative clock approach can be used for testing evolutionary hypotheses. Our review encourages a judicious use of molecular clock analyses in evolutionary studies of invertebrates by demonstrating their great potential on the one hand and (often-manageable) problems and pitfalls on the other hand. Keywords: molecular clocks, calibrating, Protostomia, Pyrgulinae Over several decades, evolutionary biologists have used a wide range of organisms to study the origin of taxa from a common ancestor and their subsequent change and diversification in space and time. One commonly used tool is the molecular clock approach, relating number of fixed mutations (= substitutions) in nucleotide or amino acid sequences to divergence time of taxa. The introduction of the molecular clock concept is attributed to Zuckerkandl and Pauling (1962), who found amino acid differences in mammalian a and 13 chains of hemoglobin to be roughly proportional to divergence times inferred from paleontological data. In 1965, these workers published a landmark paper (Zuckerkandl and Pauling 1965), naming the molecular clock and describing its stochastic nature as a Poisson process. It was also suggested that, if a molecular clock exists, amino acid changes must be limited almost exclusively to functionally nearly neutral changes — supporting the concept of near neutrality at the molecular level (Takahata 2007). In subsequent years, several workers attributed spontaneous mutations due to replication errors as a driving force of molecular evolution and suggested evolutionarily “neutral” changes in sequences be used to measure divergence times {e.g., Sarich and Wilson 1967, Kimura 1968, Wilson and Sarich 1969). The accuracy of the molecular clock, however, has long been subject to controversy and early on, workers noted that different proteins evolve at different rates. Over the years, numerous papers addressed further problems with molecular clock approaches like rate heterogeneity in different taxonomic groups or body size effects (see Takahata 2007 and references therein, also see below). Unfortunately, the controversy surrounding the molecular clock approach has *From the symposium “Molluscs as models in evolutionary biology: from local speciation to global radiation” presented at the World Congress of Malacology, held from 15 to 20 July 2007 in Antwerp, Belgium. 25 26 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 divided the scientific community. Although an increasing number of workers perform judicious molecular clock analyses, there are still many “believers” who almost uncritically use statistical tools to estimate divergence times, often resulting in unrealistic estimates; “disbelievers” who reject molecular clock approaches as inappropriate; and “ignorant” workers who refrain from using molecular clock approaches due to the critics (right or not) of the method. No matter to what group people belong, there often is a feeling of unease among workers relative to molecular clock approaches. This is due to the controversy of the molecular clock approach (see above), the complex mathematical background of many molecular clock tools, often making it difficult for biologists to understand the statistics involved, the still limited number of user-friendly software packages, the often confusing terminology used in molecular clock approaches, and a general lack of reliable calibration points and/or rates. As pointed out by Takahata (2007: 4) “It is now generally accepted that, although it is uncertain and rejected for a substantial proportion of proteins and genomic regions in comparisons of main taxonomic groups, the molecular clock can put a new timescale on the history of life, thereby allowing exploration of the mechanisms and processes of organismal evolution. Similarly, a molecular clock is an irreplaceable source of information in evolutionary biology and it would be foolish to abandon it altogether”. He also states that the molecular clock needs not be exact and that an approximate clock can still be very useful. However, it often is exactly this approximation of the molecular clock {i.e., estimations of meaningful errors) that is difficult to conduct {e.g., Ayala 1997). Moreover, there frequently is a striking imbalance between sophisticated algorithms used for molecular clock analyses and poor data {e.g., Bandelt 2007). In fact, whereas many studies deal with optimizing the performance of molecular clock tools, there are relatively few publications addressing data and model selection. This is not trivial, as problems with data and/or misinterpretation of the results can account for divergences of molecular clock estimates in one and the same taxon by >1000% (e.g., Wilke 2004, Fulquerio and Nichols 2007). In the present review, we therefore attempt to: ( 1 ) give a brief overview of molecular clock approaches, (2) discuss major problems and pitfalls associated with data and analyses, (3) address the question of universal and trait- specific molecular clocks in invertebrates, (4) provide a conservative “simple fool’s guide” to molecular clock analy.ses, (5) provide a step-by-step example for clock estimations in the caenogastropod subfamily Fyrgulinac based on both an internal calihration point and a trait- specific molecular clock rate, and (6) give definitions of some of the most widely used terms in molecular clock approaches in a glossary. This review is intended neither to provide a full statistical background of the molecular clock hypothesis nor to cover all relevant methods and developments. Instead, we give basic information on principal molecular clock strategies, on approaches for mitigating problems commonly associated with molecular clock analyses, and on major pitfalls in molecular clock estimations. Thus, we specifically target the “ignorant” people men- tioned above, hoping to persuade them to look into the application of molecular dating. At the same time, we hope to make “believers” aware of major pitfalls in molecular clock approaches and to convince “disbelievers” that under a specific set of circumstances, the molecular clock can be a very useful tool for evolutionary analyses. THE MOLECULAR CLOCK APPROACH Molecular-dating methods, the estimation of divergence times of lineages from a common ancestor based on nucleotide or amino-acid sequences, can be broadly classified into population genetic and phylogenetic {i.e., molecular clock) approaches. In population genetic approaches, a coalescent framework is used to estimate the ‘age’ of a most recent common ancestor (MRCA) of a number of alleles. The age of the MRCA is hereby measured in number of generations. This approach works backwards in time and is based on the assumption that a pair of alleles will coalesce, i.e., find their MRCA, at some point in time in the past (see Edwards and Beerli 2000). Several models were developed to describe this process with respect to various parameters such as effective population size, gene flow, and changes in population size over time. These population genetic approaches are, as the name suggests, typically only applicable to estimation of divergence time within a species. for estimating divergence times between species or between groups of species, several phylogenetic approaches have been suggested. Whereas in early studies, genetic distance matrices were used to estimate substitution rates for molecular clock estimations {e.g., Nei 1987, Li and Graiir 1991), today these substitution rates are typically derived from phylogenies or from sequence data in conjunction with tree topologies (Rutschmann 2006). (liven the scope of our review, we focus on these tree- based approaches that now appear to be the most widely accepted molecular-clock methodologies in phylogenetic studies, free-based molecular clock approaches typically use the branching topology ol a phylogeny together with branch length information to estimate the node depth (10 Mya. Problems with saturation may, to a certain extent, be mitigated through the application of sophisticated models of sequence evolution (see Kelchner and Thomas 2007). Moreover, relaxed molecular clocks, which incorporate rate variations across lineages (Fig. 2), may be less prone to saturational effects than strict molecular clock approaches. Nonetheless, tests for saturation, which are implemented in some phylogenetic software packages {e.g., DAMBE, Xia and Xie 2001 ) should be performed for all molecular clock data sets under the respective model of sequence evolution, as data sets with significant levels of saturation are not suitable for molecular clock estimations. This is particularly important for molecular clock approaches utilizing external molecular clock rates as they are typically based on the strict clock model, thus not allowing for rate variation throughout time. Arhogast el al. (2002) pointed out that estimating divergence times between both distantly and closely related taxa are challenging due to the problems discussed above. Whereas many workers are aware of problems in distantly related taxa {i.e., saturation), problems with closely related taxa (/.(’., ancestral polymorphism, power gap), are more free] u e n 1 1 y neglected. MOLECULAR CLOCKS IN INVERTEBRATES 33 Expected divergence time Figure 4. Effect of saturation on estimating divergence times. For genetically divergent taxa, satiirational effects lead to a randomiza- tion of the phylogenetic signal with the number of observed muta- tions (solid line) being lower than the actual number of differences (dotted line). This causes an underestimation of divergence times. Estimation of confidence limits of clock estimations Calculation of confidence limits is a crucial aspect of clock estimations (Elillis et al. 1996, Wilke 2004). This is because confidence limits can be very large (Bromham et al. 1998, Bromham and Penny 2003), often making estimates without considering variability meaningless. Confidence limits are largely affected by two major groups of errors: (a) molecular clock variations and (b) uncertainties of calibration points or external molecular clock rates. Causes of molecular clock variations within and between lineages can have two major sources (reviewed in Bromham and Penny 2003). Eirst, the molecular clock is probabilistic and ticks at irregular intervals. This behavior, commonly described by a Poisson process, potentially causes large confidence intervals. Second, there might be differences in substitution rates within and between lineages (for biological variables that account for deviations from rate constancy, see below). Confidence limits of calibration points and external molecular clock rates include uncertainties in the timing of ancestral DNA, fossils, or geological events, time lags of biogeographical and phylogenetic events as well as the variation of external molecular clock rates. In early molecular clock studies, assessments of clock variations were largely neglected. This has led to numerous, partly conflicting molecular clock estimates, which in turn raised general criticism of the molecular clock approach. Today, estimation of molecular clock confidence limits is standard procedure and many molecular clock software packages incorporate algorithms or approaches for quantification of uncertainties utilizing, for example, bootstrapping, Bayesian posterior distribution, or Poisson distribution (see reviews of Welch and Bromham 2005, Rutschmann 2006). Wliereas most publicly available packages account for variation of the clock itself, not all consider uncertainties of calibration points or external molecular dock rates. In this case, the total error of the clock can be calculated using, for example, a propagation of uncertainty analysis (see “Examples of molecular clock estimations in the caenogastropod subfamily Pyrgulinae” below). Sampling design and interpretation of data Molecular clock approaches allow for a dating of the MRCA of extant lineages. In this regard, the clock approach is unambiguous; split I in Fig. 5, for example, shows the age of the MRCA of taxa 1+2 and taxon 3. The interpretation of the phylogenetic and taxonomic relevance of such events, however, might be affected by sampling design (e.g., missing taxa) and may be subject to misinterpretation. Wilke (2004), for example, compared the results of two molecular clock analyses of similar sets of rissooidean snails that differed in their age estimates by more than an order of a magnitude. He showed that the difference was largely due to missing taxa together with misinterpretations of the results. Thus, these problems, though rarely discussed in the literature, may affect molecular clock analyses more severely than many other problems discussed in the present paper. To demonstrate possible adverse effects of missing taxa, we provide a sample data set with ten species (Fig. 5) with the complete phylogeny (including extinct and unsampled) taxa to the left and a phylogeny with four missing taxa to the right. As mentioned above, molecular clock analyses estimate the age of MRCAs in a given phylogeny. In a complete phylogeny, that is, a phylogeny that contains all species of a given taxon, these age estimates of MRCAs can directly be used for phylogenetic interpretations. The age of node IV, for example, represents the age of genus A, and node II the onset of the intra-generic diversification within genus C. In an incomplete phylogeny such interpretations, however, may be erroneous. Node IV in the right tree is the MRCA of three taxa of genus A and three taxa of genus C. It does, however, not correspond to the age of genus C because that genus is much younger and represents the sister to the extinct genus B (see node III in the complete phylogeny). Also, node II in the sampled phylogeny does not represent the age of the onset of diversification in 34 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 time before present complete phylogeny sampled phylogeny Figure 5. Effects of unsampled (dashed lines) and extinct (dotted lines) spe- cies on inferring the timing of phylogenetic events within three putative genera (A-C). Left, complete phylogeny; right, phylogeny with only six out of ten taxa sampled. Major nodes are marked with Roman numerals. See text for details. genus C because taxon 5 (not sampled) is older than any of the sampled ttixa (see node II in the complete phylogeny). In most cases we do not know whether a phylogeny is complete or not. Thus, molecular clock estimates are best explained within the phylogenetic concept of the MRCA. If taxonomic interpretations were to be made, then this should be done within the context of minimum and maximum ages. In a given phylogeny, the observed onset of the diversification within a supra-specific taxon (genus and beyond) should be expressed as minimum age. Node II in the sampled phylogeny thus is the minimum age of diversification within genus C. Assuming a robust phylogeny, additional taxa previously not sampled cannot render this age younger; they only can render it older (see node II in the complete phylogeny). In contrast, the age of a given supra-specific taxon should be expressed as maximum age. Node IV in the sampled phylogeny thus is the maximum age of genus C. Additional taxa cannot render the age of genus C older; they only can render it younger (see node III in the complete phylogeny). EXTERNAL MOLECULAR CLOCK RATES IN INVERTEBRATES Problems with external molecular clock rates During our survey of molecular clock analyses reported in the literature, we noted an interesting bias. Whereas putative calibration points are often uncritically unitized for clock approaches (often simply as “pers. comm.”), external molecular clock rates (particularly “universal” molecular clock rates) are typically dismissed outright as invalid. In fact, there are several theoretical and practical studies demonstrating that molecular clock rates can largely vary among genes and organisms {e.g., Thomas et al. 2006). While we agree on these findings, we also would like to raise a cautionary note. An external molecular clock rate can only be as good as the individual rates upon which it is based. Given the uncertainties of many calibration points, conflicts between fossil and biogeography-based data as well as general clock problems such as ancestral polymorphism, mi.ssing taxa, and saturation (see above), external molecular clock rates (particularly those applicable to a larger taxon) are hard to establish. Thus, a universal, taxon-specific or trait-specific molecular clock rate may be rejected becau.se there is no such rate, but they may also be rejected because of potential errors in individual rates on which they are based. Nonetheles.s, there is a strong and compelling theoretical background suggesting that there is no clock universal for all genes and taxa, and that mutation rates per se vary among genes and broad taxonomic groups {e.g., Ayala 1997, Drake et al. 1998, Bromham and Penny 2003, Takahata 2007). If we ask whether the concept of a “universal” clock can possibly be saved, we have to understand which biological factors affect mutation rates. Ayala (1999) invoked five biological variables that account for deviations from rate constancy within and between lineages: ( 1 ) generation time (shorter generation time “accelerates the clock” as it shortens the time for fixing new mutations, particularly if DNA replication-dependent errors are the major source of mutations; also see Takahata 2007), (2) population size (larger population sizes will “skwv the clock” because of increased times for fixing new mutations), (3) species-specific differences in properties that affect DNA replication (different species may, for example, have different DNA polymerases with different error rates; Bromham and Penny 2003), (4) changes in the function of a protein as evolutionary time proceeds, and (5) stochasticity of natural selection. Other factors might include: (6) body size (smaller species tend to have faster rates of molecular evolution, Cillooly et al. 2005, Lanfear et al. 2007, but see 'fhomas et al. 2006), (7) body temperature, including ectothermy v.s-. endo- thermy (body temperature affects metabolic rates, which in turn affects production of free radicals causing mutations, Cillooly et al. 2005), and (8) life history, particularly reproductive traits (mutation rates are presumably higher in hermaphrodites compared with gonochorists, Davison 2006, also see Foltz et al. 2004). MOLECULAR CLOCKS IN INVERTEBRATES 35 According to Gillooly et al. (2005), many biological factors can be linked to two major hypotheses explaining rate heterogeneity: the metabolic rate hypothesis (higher metabolic rates are related to higher production of mutation -causing free radicals) and the generation time hypothesis (most mutations are caused by DNA replication errors during division in germ cell lines) (but see Lanfear et al. 2007). It should be noted that most of these effects are largely hypothetical and have only rarely been tested in the context of the molecular clock. Moreover, the relationship between these variables and substitution rates might not be universal, but gene and taxon specific, and the underlying mechanisms often are still poorly understood {e.g., Lanfear et al. 2007). In fact, an extensive study conducted by Thomas et al. (2006) could not find a relationship between body size and mutation rates and most workers simply attribute lineage specific differences to the fickle process of natural selection [e.g., Ayala 1999, Takahata 2007). Gillooly ctn/. (2005), however, introduced a controversial model accounting for body size and temperature effects on metabolic rates, which supposedly could explain rate heterogeneity in different genes, taxa, and environments. Moreover, the authors argue that this model suggests a single molecular clock that ticks according to mass-specific met- abolic energy. Although this model was recently rejected by Lanfear et al. (2007), the effect of these and other biological factors could explain why the existence of an universal molecular clock had to be rejected for larger and biologically diverse groups, such as invertebrates (Thomas et al. 2006), but appears to be valid in some smaller groups with similar biology and life history like birds (Weir and Schluter 2008). In fact, knowledge of the relevant factors affecting the clock could help reduce deviation from rate constancy within larger sets of taxa and lead to the establishment of a series of gene-specific molecular clock rates for groups of species that share similar biological and life history traits. An example for a potential “trait-specific” molecular clock rate in invertebrates is given in the following section. A potential trait-specific molecular clock rate for the Protostomia As outlined above, several biological and life history properties of animals might affect the tick rate of the clock. At the same time, clock rates can vary considerably among genes, and the performance of a given gene might be poor for relatively young (power gap) and relatively old (saturational effects) divergence events. Acknowledging that invertebrates are a paraphyletic and highly diverse group, Wilke (2003) first attempted to establish a specific COI clock for the Protostomia, a clade of bilateral animals including tbe three major groups Ecdysozoa [e.g.. Arthropoda and Nematoda), Lophotrochozoa (e.g., Mollusca and Annelida), and Platyzoa (e.g., Platyhelminthes and Rotifera). Based on published and his own estimates of molecular clock rates for the COI gene in taxa separated by less than 10 Mya (l.e., the presumed time frame in which the COI gene is not saturated), Wilke found relatively coherent rates ranging from 0.7 to 1.2% My“' (uncorrected substitution rates) or 1.4 to 2.4% My“‘ (uncorrected divergence rates). These published individual rates were later reanalyzed and refined within a tree-base approach by Albrecht etal. (2006) utilizing Kimura’s two-parameter model (K2P) model. In this paper, we build upon the studies of Wilke (2003) and Albrecht et al. (2006) in order to establish a preliminary trait-specific molecular clock rate for the COI gene in the Protostomia. The basic idea of this trait-specific clock rate is to find (within a larger taxon) groups of species: ( 1 ) that share biological and life history characteristics supposedly affecting rate heterogeneity (e.g., mode of reproduction, generation time, body size and temperature, population size), (2) which individual clock rates can be calibrated with robust calibration points, and (3) where molecular clock estimations are not affected by the power gap or significant degrees of saturation. These individual rates can be assessed for dispersion and, if applicable, average trait-specific clock rates could be established together with their errors. The trait-specific clock rate for the COI gene in the Protostomia suggested here involves data from a total of 12 pairs of species from several higher taxonomic groups within the Protostomia with the following characteristics: - they are aquatic, - they are dioecious (see point 8 under variables that account for deviations from rate constancy between lineages above), - they have a generation time of approximately one year (see point 1 above), - they are ectothermic and live in tropical or subtropical waters (see point 7 above), and - they are relatively small with body sizes differing by not more than an order of magnitude (see point 6 above). In order to calculate an average trait-specific clock rate for these taxa, we obtained tbe COI sequences used in the original publications (see Table 3), tested the applicability of a strict molecular clock, and analyzed substitution rates under the assumption of the molecular clock and under different models of sequence evolution in MrBayes 3.1.2 (Ronquist and Eluelsenbeck 2003). Note that we were unable to correct for ancestral polymorphism in the clock data sets because of missing 36 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Table 3. Molecular clock (substitution) rates for the COI gene under different models of sequence evolution based on a total of 12 pairs of sister taxa from 5 groups of Protostomia separated by biogeographical events. These taxa share major biological and life history traits. Mean trait-specific molecular clock rates (R,^,,-) together with their standard deviations are given for each model. All values are uncorrected for ancestral polymorphism (JC, lukes-Cantor model; K2P, Kimura’s two-parameter model; F81, Felsenstein 1981 model; HKY, Hasegawa- Kishino-Yano model; GTR, general time reversible model; I, invariable sites; and F, gamma distribution). # Taxon Mean body Divergence Molecular clock rates (in % My ‘)for selected models Taxon Reference pairs size in mm time in My of sequence evolution IC K2P F81 HKY HKY+ GTR GTR-f i+r I+r Salenthydrobia/Peringia Wilke (2003) 1 3-4 5.64-^ 1.33 1.29 1.36 1.32 1.60 1.29 1.96 (Gastropoda) Chlorostoma ( = Tegula) Flellberg and 1 10-20 2.75** 1.37 1.40 1.37 1.37 1.48 1.37 2.06 spp. (Gastropoda) Vacquier (1999) Alpheus spp. Knowlton and 20-50 2.75** 1.21 1.23 1.17 1.18 1.89 1.03 1.94 (Decapoda) Weigt ( 1998) Sesarma spp. Schubart et al. 2 15-30 2.75** 1.01 1.03 1.03 1.10 1.61 1.02 1.59 (Decapoda) (1998) Alvinella/Paralvinella (Annelida) Chevaldonne etai (2002) 1 2-3 2.75** 1.20 1.21 1.20 1.21 1.26 1.19 1.25 Mean {ii - 5) 1.22 1.23 1.23 1.24 1.57 1.18 1.76 95% confidence interval 0.27 0.26 0.28 0.22 0.45 0.31 0.66 (n = 5) * Mediterranean Salinity Crisis, timing for its climax taken from Krijgsman et al. (1999). *’*^Closure of the Isthmus of Panama, timing based on average estimates suggested by Bartoli et al. (2005). ***From 15 pairs of geminate sister species suggested by Knowlton and Weigt ( 1998), the seven most closely related pairs of species were used here (also see Albrecht et al (2006)). population-level data (seesectiononancestralpolymorphism). Thus, the trait-specific clock rate suggested here might be slightly overestimated, but because of the relatively old biogeographical events used for calibration (see Table 3), we would expect the bias to be <10%. Nonetheless, the trait-specific clock rates suggested here for several major models of sequence evolution are surprisingly coherent. The 95% confidence intervals for individual models are typically around 20%. Even among models, the average clock rates are very similar, indicating that they are relatively robust against model misspecifications. The only exceptions are the models with gamma distribution and invariables sites (E+I), which show, as expected, an elevated clock rate and elevated confidence intervals. C)f course, the trait-specific COI clock suggested here would need further refinement involving more taxa and more independent calibration events in order to assess its validity for a large set of taxa. Nonetheless, this example already indicates that deviations from rate constancy can be mitigated with relatively simple means. Moreover, if the validity of this trait-specific clock would be confirmed, it may not only be applicable to many different species; it also could help to establish a general predictive model for substitution rates taking saturation as well as specific life history, biological, and biochemical characteristics into account. A simple fool’s guide to molecular clock approaches The flowchart given (Eig. 6) intends to provide relatively simple and conservative, yet sound, guidance through crucial steps of molecular clock analyses. It is simple because it is based on a series of straightforward tests and tools readily available, and it is conservative because it does not attempt to deal with problems the solution of which would require expert knowledge (e.g., estimation of divergence limes from saturated data). The guide is applicable to molecular dating of data sets for which (a) calibration points or bounds exist, (b) an external clock rate is available, or (c) for which no such information exists. In the latter case, however, only estimations of relative divergence limes (see section “Molecular dating without calibration points or external clock rates” above) would be possible. The initial information required is whether the data set in questions shows significant levels of saturation. The appropriate model of sequence evtdulion (typically the best MOLECULAR CLOCKS IN INVERTEBRATES 37 Model selection^ Calibration point or bounds available? •4- 1 1° 1 4- Test for saturation significant? -► 1 1 -► I I "° I I I I External clock rate available? i i I "° I i I I -► Clock accepted in LRT? Clock accepted in LRT? I Estimation of divergence times and errors with relaxed clock approach’ Estimation of divergence times and errors with strict clock approach’ Clock accepted in LRT? Correction for ancestral polymorphism’ Absolute divergence times (including errors)' Estimation of relative divergence times and errors with relaxed clock approach >■ Estimation of relative divergence times and errors with strict clock approach Relative divergence times (including errors) Figure 6. Simple fool’s guide for molecular clock analyses. See text for details. 'Typically, the best fit model of sequence evolution is used; approaches utilizing external clock rates may, however, require a pre-defined model. ^Error estimations should include both errors of molecular clock variations as well as uncertainties of calibration points or external molecular clock rates, ’if correction for ancestral polymorphism is not possible, divergence times should be treated as maximum divergence times. ‘'For young phylogenetic events, the power gap should be tested. comparing the results for consistence (e.g., Wilke et al. 2007, also see section “Examples of molecular clock estimations in the caenogastropod subfamily Pyrgulinae” below). Calibration point or bounds available If at least one calibration point or two bounds are available, the clock has to be tested {e.g., using LRT) in order to decide whether a strict molecular clock approach can be used for estimating divergence times or whether methods that account for rate heterogeneity (“relaxed clocks”) have to be applied. If tbe clock is accepted, one global rate of substitution can be assumed and several packages are available to estimate divergence times and to provide error estimates (reviewed in Rutschmann 2006). Note that not all of these packages can deal with multiple calibration points. If the clock is rejected, then methods should be applied that either correct for or incorporate rate heterogeneity (reviewed in Rutschmann 2006). However, as with strict clock approaches, not all of these methods can deal with multiple calibration points. Whether strict or relaxed clock approaches are used, it should be checked if error estimations also account for uncertainties in the calibration point(s). If not, this error should be incorporated via, for example, propagation of uncertainty (see section “Examples of molecular clock estimations in the caenogastropod subfamily Pyrgulinae” below). Finally, divergence times should be corrected for ancestral polymorphism, if possible. If the data set does not allow for correction of ancestral polymorphism, divergence times should be treated as maximum times. Moreover, if divergence times are very low, it should be tested whether the data set is sufficient for resolving such young phylogenetic events (see section “Appropriate time frame” above). fit model) has to be selected using, for example, tbe program Modeltest (Posada and Crandall 1998), MrModeltest (Nylander 2004), or similar software tools. Then the data set can be tested under the chosen model for substantial nucleotide saturation {e.g., using the program DAMBE, Xia and Xie 2001). If the data set is saturated, molecular dating is not advisable. If tbe test reveals no substantial saturation, then the question arises whether at least one calibration point or at least one lower and one upper bound for estimating substi- tution rates exist. If so, the tree can be calibrated with those points (see below). If not, tbe guide asks for the existence of an external molecular clock rate. If available, this external rate could be utilized to calibrate tbe clock. Otherwise, only estimations of relative divergence times might be possible. In cases were both calibration point(s) and an external clock rate are available, we suggest using both approaches and External molecular clock rate available If an external clock rate is available, the data set has to be tested for clock like behavior. If the clock is rejected, molecular dating is not advisable, as tbe applicability of an external rate typically requires one global rate of substitution (strict clock). If the clock is accepted, estimation of divergence times is as described above. Some external dock rates are available for a specific model of sequence evolution only. In this case, the same model would need to be applied to the data set in question. Such model misspecifications are, however, controversially discussed (particularly for data sets with distantly related taxa) and should be used with caution. Another consideration is that external clock rates may or may not be corrected for ancestral polymorphism. If uncorrected rates are used, it might be difficult to estimate the bias in tbe data set in question, and dates for young 38 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 divergence events should be treated with particular caution (see section “Examples of molecular clock estimations in the caenogastropod subfamily Pyrgulinae” below). No calibration point or bounds and no external molecular clock rate available If no information is available for calibrating the tree, relative divergence times can be estimated. Depending on whether the clock is accepted, strict or relaxed clock approaches should be used. Relative clock approaches are best implemented by setting the age of the root to 1 by default. Then, after testing for normality of node depth distribution, either paired Mann-Whitney U-tests or Student’s f-tests can be used to study whether two specific MRCA significantly differ in their age utilizing replicates from the phylogenetic search. Although still rarely applied in phylogenetic studies, such relative clock approaches are powerful tools for testing hypotheses in evolutionary biology in the absence of specific divergence times. EXAMPLES OF MOLECULAR CLOCK ESTIMATIONS IN THE CAENOGASTROPOD SUBFAMILY PYRGULINAE To demonstrate different molecular clock approaches in our model taxon, the subfamily Pyrgulinae, we here use the two hydrobiid data sets of Wilke et al. (2007) for three largely independent clock analyses. The full data set A contains combined fragments of the mitochondrial COI gene, the mitochondrial LSU rRNA gene, and the nuclear SSU rRNA gene. The reduced data set B only contains COI sequences. The data set B is used for clock estimations utilizing the trait- specific COI Protostomia clock introduced above. Data set A serves as the basis for estimating divergence times based on an available calibration point as well as for relative clock estimations. The following descriptions are based on the flow chart (Fig. 6) presented above. L Time estimation with calibration point (data set A) Model selection The data set was analysed in MrModeltest 2.3. The models suggested were HKY-i-l+r (Flasegawa-Kishino-Yano model with invariable sites and E distribution), GTR+I+E (general time reversible model with invariable sites and E distribution), and TrNef + 1 (Tamura-Nei model with equal base frequencies and invariable sites) for the COI, LSU rRNA, and SSU rRNA fragments, respectively. 'lest for saturation significant? In order to test whether the individual partitions show significant levels of .saturation, the test of Xia et al. (2003), as implemented in the software package DAMBE 4.2.13 (Xia and Xie 2001), was used with the proportion of invariable sites suggested by MrModeltest [i.e., 0.6082 for COI, 0.5805 for LSU rRNA, and 0.9406 for SSU rRNA). The test did not reveal a significant degree of saturation even under the very conservative assumption of an extremely asymmetrical tree for any of the three data partitions. Therefore, the data set is considered to be suitable for further molecular clock analyses. Calibration point or bounds available? For the Pyrgulinae, the known phylogenetic age of the monotypic genus Salenthydrobia Wilke, 2003 (see Wilke 2003, 2004) could be used as the calibration point for estimating timing of evolutionary events (see node A in Fig. 7). Ecological and biogeographical data strongly indicate that Salenthydrobia originated during the Messinian salinity crisis (MSC), that is, between 5.96 and 5.33 Mya (see Krijgsman et al. 1999 for the dating of the MSC). As Salenthydrobia belongs to the potential sister subfamily of the Pyrgulinae (i.e., the Hydrobiinae), and as their relationships do not show signs of saturation, it is here assumed that the substitution rates in these taxa are similar. Clock accepted in LRT? For the LRT, we first ran two analyses in MrBayes 3.1.2 (clock enforced and clock not enforced) with the partitioned model suggested by MrModeltest (see above) until the chains converged on similar results (i.e., <0.01 after 1.000.000 generations). Then, the best tree from each analysis was used for the LRT. With log L^, = -5493.43, log L, = -5488.52, -2log A = 9.82, and df = 19, the clock hypothesis was not rejected (P< 0.05). Estimation of divergence times and errors with strict clock approach For calculation of the age of the split between the Black Sea/Asia Minor pyrgulinids and the Pyrgulinae from the Balkan (node B in Fig. 7) as well as the split between the Lake Ohrid pyrgulinids from their sister taxon (node C in Fig. 7), we are using all trees generated from the (clock-enforced) Bayesian search, except for those that tall within a predefined burn-in of 10%. For each individual tree, we calculated the age ot nodes B and C using the rule of three with the known variables being the age of node A (i.e., the climax ol the MSC. with 5.64 Mya), the node depth of notle A, and the depth ol nodes B or C. Separately for nodes B and C, we then averaged their ages over all trees and calculated the res[iective 95% confidence intervals (resulting in 7.72 ± 2.37 Mya for node B and 3.00 ± 1 .04 Mya for node C). Finally, the error ol the calibration point has to be added to this error ol the clock. MOLECULAR CLOCKS IN INVERTEBRATES 39 Figure 7. Estimation of divergence times utilizing a calibration point according to approach 1. Shown is the best Bayesian tree under the clock criterion based on three gene fragments for representatives of the nominal subfamilies Pyrgulinae, Pseudamnicolinae, and Hydrobiinae (modified from Wilke etal. 2007). The outgroup taxon was removed a posteriori. The calibration point (i.e., the origin of the genus Salenthy- drobia that is associated with the Messinian salinity crisis [MSC; marked A] ) and the time frame of major phylogenetic events [i.e., the split of the Ponto-Caspian/Asia Minor taxa from the Balkan taxa [B], and the split of Lake Ohrid pyrgulinids from their sister taxon [C]) are shown on the tree. Branch length variations are plotted as black bars at nodes B and C. For estimating the total error of our clock calculations, we also incorporated the error of the calibration point (gray bars, see text for details). Bars right to the node are not corrected for ancestral poly- morphism {AAP); bars to the left constitute corrected values. The timing of the events in question and their confidence intervals are shown as light gray bands. Posterior probabilities are given for all nodes. 40 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Unfortunately, we do not have any information as to when the split between Snlenthydrobia and its sister taxon occurred within the MSC. Thus, each point between the onset of the MSC (I'.e., 5.96 Mya) and its end [i.e., 5.33 Mya) is equally likely. As we used the climax of the MSC for our clock estimation, we here suggest a pragmatic approach for calculating an approx- imate total error of the clock. We simply add the difference from the climax to both the beginning and end of the MSC (i.e., ± 0.315 My) to the conhdence interval of the clock estimations. The values for node B and C thus would be 7.72 ± 2.69 Mya and 3.00 ± 1.35 Mya, respectively (see fig. 7) Correction for ancestral polynwrphism To infer the amount of ancestral polymorphism (Lig. 3) within our data set in the absence of sufficient population genetics data, we here use an approach based on a suggestion of Edwards and Beerli (2000). We assume that the effective population sizes in extant species reflect the population sizes of the ancestral species. Elence, averaging the sequence diversity within the four descendant species of node C (fig. 7) might provide a rough estimate of the ancestral polymorphism that was present when splits C and B occurred. The node depth under the chosen HKY-i-I-i-r model is 0.0039 for Pyr- gula annulata (Linnaeus, 1767), 0.0029 for Ohridopyrgula macedonica (Brusina, 1896), 0.0028 for Chilopyrgula sturanyi Brusina, 1896, and 0.0042 for Xestopyrgula dybowskii Polinski, 1929 (mean node depth; 0.0034). Using node A as calibration point, the amount of ancestral polymorphism corresponds to approx. 0.37 My. Estimation of absolute divergence times including errors Deducing the ancestral polymorphism corresponding to 0.37 My from our uncorrected divergence time estimations results in corrected divergence times and confidence intervals of 7.35 ± 2.69 Mya for node B and 2.63 ± 1.35 Mya ago for node C (see Eig. 7). II. Time estimation with external trait-specific COI clock (data set B) Model selection The COI data set was analysed in MrModeltest 2.3, which suggested the HKY+I-t-E model based on the Akaike information criterion. Calibration point or bounds available? We here ignore the Salenthydrobia-caMhration point used above and continue with the flow chart assuming that there is no calibration point available. Clock accepted in LRT? The COI data set was used to run two analyses in MrBayes 3.1.2 (one with the clock enforced and one without enforced clock) under the HKY-t-I-t-T model suggested by MrModeltest (see above) until the chains converged on similar results (i.e., <0.01 after 1.000.000 generations). Then, the best tree from each analysis was used for the LRT. The clock hypothesis was not rejected (-2log A = 10.66, df= 19, P < 0.05). Estimation of divergence times and errors with strict clock approach for calculating the age of the split between the Black Sea/ Asia Minor pyrgulinids and the Pyrgulinae from the Balkan (node B in Pig. 7) as well as the split between the Lake Ohrid pyrgulinids from its sister taxon (node C in Pig. 7), we use the external trait-specific COI clock rate suggested in the present paper. Based on a rate and 95% confidence interval of 1.57 ± 0.45% My“' under the HKY-hl-t-T model (see Table 3), we used all trees generated from the Bayesian search (under the clock criterion), except for those that fall within a predefined burn-in of 10%. Lor each individual tree (in our case 90,000), we calculated the depths of nodes B and C and their respective standard deviations from the tree files with an R-routine (available upon request), resulting in mean node depths of 11.13 ± 1.90% and 4.46 ± 0.80% for nodes B and C, respectively. Alternatively this calculation can be carried out utilizing the program TreeAnnotatorl.4.8 from the BEAST package (Drummond and Rambaut 2007). Using the rule of three with the known variables being the depth of nodes B (11.13%) and C (4.46%) and the external clock rate of 1 .57% My“ ' , we calculated the mean age resulting in 7.1 1 Mya for node B and 2.85 Mya for node C. finally we calculated the total error of these estimates by combining the error of node depth (standard deviations of 1.90% and 0.80% for nodes B and C, respectively) with the error of the trait-specific clock (the standard deviation ot 0.23% My”' corresponds to a confidence interval of 0.45% My”') by utilizing the method of error propagation (note that this method is based on standard deviations rather than confidence intervals): Test for saturation significant? Utilizing the test of Xia et al. (2003) and the proportion of invariable sites suggested by MrModeltest (i.e., 0.6082), the test did not reveal a significant degree of saturation, even under the very conservative assumption of an extremely asymmetrical tree. I he C(9I data set is therefore considered to be suitable for further molecular clock analyses. AC Ay U' with A(i being the total error, A.v the error of the node defith. Ay the error of the external clock rate, .v the mean node depth, and y the external clock rate. MOLECULAR CLOCKS IN INVERTEBRATES 41 Based on a mean node depth of x = 11.13%, a relative node depth error of Ax = 1.90%, a relative error of the external clock rate of Ay = 0.23% My”', and an external clock rate of y = 1.57% My“', the total error (as standard deviation) for node B in My is: AG + 11.13 (1.57)2 0.23 1.59 Multiplying this standard deviation of 1.59 My with 1.96 results in a 95% confidence interval of 3.10 My. Elence, the age and confidence intervals for node B would be 7.1 1 ± 3.10 Mya (the corresponding values for node C are 2.85 ± 1.29 Mya). Correction for ancestral polymorphism Correction for ancestral polymorphism in the present pyrgulinid data set is not possible. This is because the external clock rate used here is not corrected due to the lack of knowledge of intraspecific diversities within the taxa used for establishing this trait-specific clock rate. However, assuming that the extent of ancestral polymorphism in the latter data sets is similar to the one in our pyrgulinid data set, some approximate information on a potential bias can be given. If, for example, phylogenetic events to be estimated in the pyrgulinid data set have an age similar to the average age of those events used for establishing the trait specific-clock (here approx. 3 Mya, see Table 3), then the bias might be small. If the event to be estimated is older, then we likely will see an underestimation of time. However, if the events to be estimated are younger than those used to establish the trait-specific clock, then we will see an overestimation of divergence times. Estimation of absolute divergence times including errors Without correcting for ancestral polymorphism, the values presented above would have to serve as approximate final values. The time estimate of 2.85 ± 1.29 Mya for node C might represent a relatively unaffected value. The estimate of 7.1 1 ± 3.10 Mya for node B, however, is likely underestimated by an unknown value. section “Time estimation with calibration point” above). Therefore, the data set is considered to be suitable for further molecular clock analyses. Calibration point or bounds available? We here assume that no calibration points and no external rates are available. Clock accepted in LRT? The LRT did not reject the clock hypothesis (-2log A = 9.82, df = 19, P < 0.05, see section “Time estimation with calibration point” above). Estimation of relative divergence times including errors Calculation of relative divergence times of phylogenetic events can be done using all trees generated from the Bayesian search under the clock criterion, except for those that fall within the burn-in. For each individual tree, the relative age of a given node is estimated by either setting the node depth of the root to one or by simply using absolute node depth as relative divergence time. In most cases, relative divergence times, however, are meaningless. Instead there often is an interest in testing whether a specific split occurred simul- taneously with another split in the phylogeny. In Fig. 7, for example, node A (the split of Salenthydrobia from its sister taxon) appears to be younger than node B (the split of the BlackSea/Asia Minor pyrgulinids from the Balkan pyrgulinids). In order to test this assumption, the following statistics can be used. For each individual tree of the Bayesian search (with the trees from the burn-in ignored), the depth of node A is compared to the depth of node B either using a paired Student’s t-test or a paired Mann-Whitney L/-test, depending on whether the data are distributed normally. As a Shapiro-Wilk test did not reject normal distribution of the data (P > 0.05), a paired Student’s t-test was used to test whether depths of the nodes A and B (and therefore their ages) differ significantly. As the test was significant (P < 0.01 ), it can be assumed that node A is significantly younger than node B and that these phylogenetic events do not coincide in time. III. Relative time estimations (data set A) Model selection The best-fit models of sequence evolution suggested by MrModeltest 2.3 were HKY-hl+T, GTR+I-t-T, and TrNef+I for the COI, LSU rRNA, and SSU rRNA fragments, respectively (see section “Time estimation with calibration point” above). Test for saturation significant? The test of Xia et al. (2003) did not reveal a significant degree of saturation for any of the three data partitions (see CONCLUSIONS Over the past few decades, molecular clock approaches have become increasingly popular, and it is now widely accepted that the molecular clock is an important source of information in evolutionary biology. Although not exact, it can provide useful information on divergence times, and a number of approaches are being developed to mitigate problems associated with the clock. This concerns both major clock strategies — the application of calibration points or 42 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 bounds and the application of external molecular clock rates. Lor approaches based on calibration points and bounds, we show that they have the advantage of being locus- (gene-) independent, but that reliable calibration points are rarely available, that the accuracy and the error of calibration points often are difficult to assess, and that some suggested calibration points can only serve as upper or lower bounds, at best. We also show that data typically used for calibrating trees {i.e., ancestral DNA/RNA, fossils, or biogeographical information) all suffer from specific problems that might severely bias molecular clock estimations. Approaches based on external clock rates, however, have the advantage of not requiring such calibration points or bounds. They, however, typically call for a strict molecular clock behavior of the data set in question and are usually locus-specific. Moreover, relatively few external clock rates are available, and many of them are controversial. Whereas universal clocks are often dismissed outright, newer studies suggest that there might be a single substitution rate (including error) for a range of taxa that share biological and life history characteristics supposedly affecting rate heterogeneity, i.e., a trait-specific molecular clock. One such trait-specific clock within the Protostomia is introduced in the present paper. Common to all the approaches above is that they crucially depend on the performance of the gene(s) used, with the lower end of the performance (corresponding to low divergence times) being affected by the “power gap” and the upper end (corresponding to high divergence times) being affected by saturation. In addition, ancestral polymorphism may cause overestimation of divergence times, particularly affecting young phylogenetic events. Another problem in molecular clock approaches is the estimation of confidence limits of the clock. Whereas many available software tools account for the stochastic nature of the clock, not all tools can account for the uncertainties of calibration points or external molecular clock rates. finally, we show that the arguably single most important source of errors in molecular clock estimates is not the underlying statistics, but misinterpretation of the results of clock analyses due to problems with sampling design ( missing and extinct taxa). Nonetheless, the examples presented here for two largely independent molecular clock strategies (calihration point vs. external trait-specific clock) yielded concurrent results, differing by le.ss than 10% (i.e., 7.35 ± 2.69 and 2.63 ± 1.35 Mya vs. 7. 1 I ± 3. 10 and 2.85 ± 1 .29 Mya). Although being an i.solated ca.se, it adds to the increasing evidence that many problems with molecular clocks and associated data are manageable and that the estimation of meaningful confidence intervals is crucial for a judicious interpretation of the results. GLOSSARY Akaike’s information criterion (AIC): A method of model selection developed by Akaike ( 1974). The AIC suggests the best model out of a candidate set of models based on the differences between the individual AIC values of each model. These values are calculated using the likelihood estimator and a penalizing term that increases with the number of model parameters. By doing so, AIC provides a measure of uncertainty of each model rather than a significance value (for further information on model selection see Burnham and Anderson 2002). Ancestral polymorphism: The amount of heterogeneity that is present in an ancestral population prior to the separation of the descending species. As a consequence, genetic divergence predates species divergence by a certain amount of time. This amount corresponds to the coalescent analogue of the polymorphism in the ancestral species. It averages 2N^ generations (with N_ being the effective population size) in a random mating population (see Arbogast et al. 2002, also see Lig. 3). Coalescent theory: A population genetics approach that models the history of gene copies backwards in time. The theory provides a mathematical framework describing the characteristics of coalescent events, i.e., lineages finding their most recent common ancestor (MRCA). The theory was developed by Kingman ( 1982). Divergence rate: Substitution rate x 2. Divergence time: Time since separation of descendent taxa from a most recent common ancestor (MRCA). Effective population size: Wright (1938) defined the term as “the number of breeding individuals in an idealized population that would show the same amount of dispersion of allele frequencies under random genetic drift or the same amount of inbreeding as the population under consideration”. To date, there are several definitions for effective population size (see Ewens 2004). Generation time effect: Assuming that DNA replication- dependent errors constitute a major fraction of the overall number of mutations, taxa with shorter generation times accumulate more mutations per unit time than taxa with longer generation times. Consequently, the substitution rate of the former would be higher (see Takahata 2007). Global clock: A global clock assumes a single substitution rate along all branches of a given phylogeny. It is also termed as “strict clock”. Note that some workers, however, use this term synonymously with “universal clock”. Likelihood ratio test (LRT): A model testing approach ba.sed on the dillerence in likelihood estimators of two nested models. The LR’f approximately follows a chi-squared distribution with - netica 122: 1 15-125. Geyer, )., T Wilke, and E. Petzinger. 2006. The solute carrier family SLGIO: More than a family of bile acid transporters regarding lunclion and phylogenetic relationships. Niumyn-Schmiede- berg’s Archives of Pharmacology 372: 4 1 3-43 1 . Gillooly, j. P., A. P. Allen, G. B. West, and |. 1 1. Brown. 2005. The rate of DNA evolution: Effects of body size and temperature on the molecular clock. Proceedings of the National Academy of Sciences of the U.S.A. 102: 140-145. Govindarajan, A. E, K. M. Halanych, and C. W. Cunningham. 2005. Mitochondrial evolution and phylogeography in the hydrozo- an Obelia geniculata (Cnidaria). Marine Biology 146: 213-222. Hedges, S. B. and S. Kumar. 2004. Precision of molecular time esti- mates. Trends in Genetics 20: 242-247. Hedges, S. B., P. H. Parker, C. G. Sibley, and S. Kumar. 1996. Con- tinental breakup and the ordinal diversification of birds and mammals. Nature 381: 226-229. Hellberg, M. E. and V. D. Vacquier. 1999. Rapid evolution of fertil- ization selectivity and lysin cDNA sequences in teguline gastro- pods. Molecular Biology and Evolution 16: 839-848. Hillis, D. M., B. K. Mable, and C. Moritz. 1996. Applications of molecular systematics: The state of the field and a look to the future. In: D. M. Hillis, C. Moritz, and B. K. Mable, eds.. Molecular Systematics. Sinauer Associates, Sunderland, Mas- sachusetts. Pp. 515-543. Ho, S. Y. W. 2007. Calibrating molecular estimates of substitution rates and divergence times in birds. Journal of Avian Biology 38: 409-414. Kelchner, S. A. and M. A. Thomas. 2007. Model use in phylogenetics: Nine key questions. Trends in Ecology and Evolution 22: 87-94. Kimura, M. 1968. Evolutionary rate at the molecular level. Nature 217: 624-626. Kingman, J. F. C. 1982. The coalescent. Stochastic Processes and Their Applications 13: 235-248. Kishino, H., ]. L. Thorne, and W. J. Bruno. 2001. Performance of a divergence time estimation method under a probabilistic model of rate evolution. Molecular Biology and Evolution 18: 352-361. Knowlton, N. and L. A. Weigt. 1998. New dates and new rates for divergence across the Isthmus of Panama. Proceedings of the Royal Society London (B) 265: 2257-2263. Krijgsman, W., F. J. Hilgen, 1. Raffi, F. J. Sierro, and D. S. Wilson. 1999. Chronology, causes and progression of the Messinian sa- linity crisis. Nature 400: 652-655. Lambert, D. M., P. A. Ritchie, C. D. Millar, B. Holland, A. ). Drum- mond, and C. Baroni. 2002. Rates of evolution in ancient DNA from Adelie penguins. Science 295: 2270-2273. Lanfear, R., J. A. Thomas, J. J. Welch, T. Brey, and L. Bromham. 2007. Metabolic rate does not calibrate the molecular clock. Proceedings of the National Academy of Sciences of the U.S.A. 104: 15388-15393. Li, W.-H. and D. Graur. 1991. Fundanientals of Molecular Evolution. Sinauer Associates, Sunderland, Massachusetts. Luttikhuizen, P. C., |. Drent, W. van Delden, and T. Piersma. 2003. Spatially .structured genetic variation in a broadcast-spawning bivalve: Quantitative versus molecular traits. Journal of Evolu- tionary Biology 16: 260-272. Marshall, C. R. 1990. The fo.ssil record and estimating divergence times between lineages: Maximum divergence times and the importance of reliable phylogenies. Journal of Molecular E,volu- lion 30: 400-408. Nei, M. 1987. Molecular Evolutionary Genetics, P' Eriition. Columbia University Pre.ss, New York. Nylaiidei', I. A. A. 2004. MrModcllcsl v2. Program distributed by the author. Evolutionary Bii>logy Centre, Uppsala finiversity, Sweden. MOLECULAR CLOCKS IN INVERTEBRATES 45 Paxinos, E. E., H. F. James, S. L. Olson, M. D. Sorenson, J. Jackson, and R. C. Fleischer. 2002. mtDNA from fossils reveals a radia- tion of Hawaiian geese recently derived from the Canada goose [Branta canadensis). Proceedings of the National Academy of Sci- ences of the U.S.A. 99: 1399-1404 Posada, D. and T. Buckley. 2004. Model selection and model av- eraging in phylogenetics: Advantages of Akaike information criterion and Bayesian approaches over likelihood ratio tests. Systematic Biology 53: 793-808. Posada, D. and K. A. Crandall. 1998. Modeltest: Testing the model of DNA substitution. Bioinformatics 14: 817-818. Pulquerio, M. J. F. and R. A. Nichols. 2007. Dates from the molecu- lar clock: How wrong can we be? Trends in Ecology and Evolu- tion 22: 180-184. Ronquist, F. and J. P. Huelsenbeck. 2003. MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19: 1572-1574. Rutschmann, F. 2006. Molecular dating of phylogenetic trees: A brief review of current methods that estimate divergence times. Diversity and Distributions 12: 35-48. Sanderson, M. ]. 1997. A nonparametric approach to estimating divergence times in the absence of rate constancy. Molecular Biology and Evolution 14: 1218-1231. Sanderson, M. J. 1998. Estimating rate and time in molecular phy- logenies: Beyond the molecular clock? In: P. Soltis, D. Soltis, and J. Doyle, eds.. Plant Molecular Systematics, II. Kluwer, Bos- ton. Pp. 242-264. Sanderson, M. J. 2002. Estimating absolute rates of molecular evo- lution and divergence times: A penalized likelihood approach. Molecular Biology and Evolution 19: 101-109. Sarich, V. M. and A. C. Wilson. 1967. Immunological time scale for hominid evolution. Science 158: 1200-1203. Schubart, C. D., R. Diesel, and S. B. Hedges. 1998. Rapid evolution to terrestrial life in Jamaican crabs. Nature 393: 363-365. Takahata, N. 2007. Molecular Clock: An Anti-neo-Darwinian Lega- cy. Genetics 176: 1-6. Takezaki, N., A. Rzhetsky, and M. Nei. 1995. Phylogenetic test of the molecular clock and linearized trees. Molecular Biology and Evolution 12: 823-833. Thomas, J. A., J. J. Welch, M. Woolfit, and L. Bromham. 2006. There is no universal molecular clock for invertebrates, but rate varia- tion does not scale with body size. Proceedings of the National Academy of Sciences of the U.S.A. 103: 7366-7371. Thorne, J. L., H. Kishino, and I. S. Painter. 1998. Estimating the rate of evolution of the rate of molecular evolution. Molecular Biol- ogy and Evolution 15: 1647-1657. Walsh, H. E. and V. L. Friesen. 2001. Power and stochasticity in the resolution of soft polytomies: A reply to Braun et al. Evolution 55: 1264-1266. Walsh, H. E., M. G. Kidd, T. Mourn, and V. L. Friesen. 1999. Poly- tomies and the power of phylogenetic inference. Evolution 53: 932-937. Weir, J. T. and D. Schluter. 2008. Calibrating the avian molecular clock. Molecular Ecology 17: 2321-2328. Welch, J. J. and L. Bromham. 2005. Molecular dating when rates vary. Trends in Ecology and Evolution 20: 320-327. Wilke, T. 2003. Salenthydrobia n. gen. (Rissooidea: Hydrobiidae): A potential relict of the Messinian salinity crisis. Zoological Jour- nal of the Linnean Society 137: 319-336. Wilke, T. 2004. How dependable is a non-local molecular clock? A reply to Hausdorf et al. (2003). Molecular Phylogenetics and Evolution 30: 835-840. Wilke, T., C. Albrecht, V. V. Anistratenko, S. K. Sahin, and M. Z. Yildirim. 2007. Testing biogeographical hypotheses in space and time: Faunal relationships of the putative ancient lake Egirdir in Asia Minor. Journal of Biogeography 34: 1807- 1821. Wilson, A. C. and V. M. Sarich. 1969. A molecular time scale for hu- man evolution. Proceedings of the National Academy of Sciences of the U.S.A. 63: 1088-1093. Wright, S. 1938. Size of population and breeding structure in rela- tion to evolution. Science 87: 430-431. Xia, X. and Z. Xie. 2001. DAMBE: Software Package for Data Analy- sis in Molecular Biology and Evolution. The Journal of Heredity 92: 371-373. Xia, X., Z. Xie, M. Salemi, L. Chen, and Y. Wang, Y. 2003. An index of substitution saturation and its application. Molecular Phy- logenetics and Evolution 26: 1-7. Yang, Z. 2004. A heuristic rate smoothing procedure for maximum likelihood estimation of species divergence times. Acta Zoolog- ica Sinica 50: 645-656. Zuckerkandl, E. and L. Pauling. 1962. Molecular disease, evolution, and genic heterogeneity. In: M. Kasha and B. Pullman, eds.. Horizons in Biochemistry. Academic Press, New York. Pp. 189- 225. Zuckerkandl, E. and L. Pauling. 1965. Evolutionary divergence and convergence in proteins. In: V. Bryson and H. J. Vogel, eds.. Evolving Genes and Proteins. Academic Press, New York. Pp. 97-166. Submitted: 9 April 2009; accepted: 15 April 2009; final revisions received: 1 May 2009 »■ * • imaf Amer. Make. Bull. 27: 47-58 (2009) Molluscan models in evolutionary biology: Apple snails (Gastropoda: Ampullariidae) as a system for addressing fundamental questions'^ Kenneth A. Hayes^’^, Robert H. Cowie*, Aslak Jorgensen^, Roland Schultheifi^, Christian Albrecht'*, and Silvana C. Thiengo^ ' Center for Conservation Research and Training, Pacific Biosciences Research Center, University of Hawaii, 3050 Maile Way, Honolulu, Hawaii 96822, U.S.A. ^ Department of Zoology, University of Hawaii, Honolulu, Hawaii 96822, U.S.A. ^ Mandahl-Barth Research Centre for Biodiversity and Health, DBL-Centre for Health Research and Development, Department of Disease Biology, The Faculty of Life Sciences, University of Copenhagen, Thorvaldsensvej 57, DK-1871 Frederiksberg C, Denmark ■' Department of Animal Ecology and Systematics, Justus Liebig University Giessen, Heinrich-Buff- Ring 26-32 IFZ, D-35392 Giessen, Germany ^ Departamento de Malacologia, Instituto Oswaldo Cruz/Fiocruz, Av. Brasil 4365 Manguinhos, 21.040-900, Rio de Janeiro, Brasil Corresponding author: khayes@hawaii.edu Abstract: Molluscs constitute the second largest phylum in terms of the number of described species and possess a wide array of characteristics and adaptations for living in marine, terrestrial, and freshwater habitats. They are morphologically diverse and appear in the fossil record as far back as the early Cambrian (-560 mybp). Despite their high diversity and long evolutionary history, molluscs are often underused as models for the study of general aspects of evolutionary biology. Freshwater snails in the family Ampullariidae have a global tropical and subtropical distribution and high diversity with more than 150 species in nine currently recognized genera, making them an ideal group to address questions of historical biogeography and some of the underlying mechanisms of speciation. They exhibit a wide range of morphological, behavioral, and physiological adaptations that have probably played a role in the processes of diversification. Here we review some of the salient aspects of ampullariid evolution and present some early results from ongoing research in order to illustrate the excellent opportunity that this group provides as a system for addressing numerous questions in evolutionary biology, particularly with regard to the generation of biodiversity and its distribution around the globe. Specifically, we suggest that ampullariids have great potential to inform (1) biogeography, both on a global scale and a smaller intra-continental scale, (2) speciation and the generation of biodiversity, through analysis of trophic relations and habitat partitioning, and addressing issues such as Rapoport’s Rule and the latitudinal biodiversity gradient, and (3) the evolution of physiological and behavioral adaptations. Also, a number of species in the family have become highly successful invasives, providing unintentional experiments that may offer insights into rapid evolutionary changes that often accompany introductions, as well as illuminating invasion biology in general. Key words: biogeography, speciation, freshwater, Pomacea Molluscs are second only to arthropods in number of described species, roughly estimated at about 100,000, with a further 100,000 or so as yet undescribed (Lindberg etal 2004). Although 60-70% of molluscs are marine (van Bruggen 1995), they are also well represented in freshwater and terrestrial habitats. Their adaptations in these environments are displayed through a variety of trophic, ecological, and morphological characteristics (Lindberg et al. 2004). Yet despite their high biodiversity and multifaceted life histories and habits, molluscs remain underused in addressing general aspects of evolution- ary biology. Several features of the group, including its long history, global distribution, ecological and morphological diversity, and high biodiversity, make it amenable to providing fundamental insights into many evolutionary issues, including patterns of historical biogeography, mechanisms generating biodiversity, and the underlying processes of adaptation and speciation. Freshwater snails offer many opportunities for such studies {e.g., Dejong et al. 2001, Mavarez et al. 2002, Facon et al. 2003, Albrecht et al. 2007, Strong et al. 2008), and among them the operculate family Ampullariidae seems particularly valuable in this regard. The Ampullariidae have a primarily circumtropical distribution, reaching their highest diversity in South America. There are records of ampullariids from the Lower Cretaceous, ~145 million years before present (mybp), and the Upper Jurassic, -160 mybp, in Africa and Asia respectively (Wang 1984, Tracey et al. 1993, Van Damme and Pickford 1995), and their fossil record dates back at least 50 mybp in the Neotropics (Melchor et al. 2002). More than 150 nominal species are recognized in nine extant genera: Afropomns Pilsbry and * From the symposium “Molluscs as models in evolutionary biology: from local speciation to global radiation” presented at the World Congress of Malacology, held from 15 to 20 July 2007 in Antwerp, Belgium. 47 48 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Bequaert, 1927, Sau/ea Gray, I867,andLan2sfesMontfort, 1810 are African; Pila Roding, 1798 is African and Asian; Asolene d’Orbigny, 1838, Felipponea Dali, 1919, Marisa Gray, 1824, and Pomella Gray, 1847 are South American; Pomacea Perry, 1810 ranges from Argentina to the southeastern U.S.A. and the Caribbean (Berthold 1991, Cowie and Thiengo 2003). While the overall family-level morphology of ampullariids is relatively constrained, many species exhibit wide ontogenetic and ecophenotypic conchological variation, making identi- fication and delimiting of species based on conchology alone very difficult. Internal anatomy offers some resolution (Thiengo 1989, Thiengo et al. 1993, 2007), but molecular analyses have begun to make it possible to identify well- demarcated lineages (species) (Cowie etal 2006), and provide a phylogenetic framework to resolve fundamental taxonomic and systematic problems and address major evolutionary questions. Their long evolutionary history, wide geographic distribution, and high biodiversity make them especially well suited for studying biogeography, biodiversification, and novel adaptations to provide further insights into evolutionary biology in general. This paper makes no attempt to review ampullariid evolution, phylogenetics, or systematics comprehensively. Rather, we summarize the basic knowledge of the evolution and biogeography of the group and discuss some of the opportunities mentioned above, especially in the light of our ongoing research on ampullariids, to illustrate their potential as vehicles for addressing questions in evolutionary biology. BIOGEOGRAPHY The family Ampullariidae is thought to have originated in the part of Gondwana that is now Africa. The origin of the family more than 150 mybp was followed by spread and diversification across Africa, Asia, and the Neotropics. Its absence in Australia is thought to be a result of the early separation ofthat continent (>160 mybp) prior to ampullariids reaching it (Berthold 1991 ). Studies of the wide distributions of the diverse species in Africa and the Neotropics should allow insights into both the higher level (generic) origins and patterns of diversification within the family. This will, in turn, provide additional insight into the biological, phylogenetic, and evolutionary consequences of the break up of Gondwana, by corroborating or contradicting patterns revealed by other groups of plants and animals. Insights gained from those ampullariid taxa with narrow distributions {Afropomiis, Saulcn, Felipponen, and perhaps Asoleiic) may also provide more detailed information about the precise geographic relationships and pathways of dispersal between particular sub-regions of the main parts of Gondwana, in particular certain parts of eastern South America and western Africa. Glarification of the exact order and timing of ampullariid diversification and associated biogeographic patterns will not only provide a much better understanding of evolution within the group, but also contribute significantly to our knowledge of the mode and tempo of evolution, adaptive radiation, and distribution of other freshwater fauna. Also, because the distributions of ampullariids and speciation within the family are probably influenced by both vicariant events like the splitting of Gondwana and passive, long distance dispersal with flow in major river systems, studying ampullarids may help clarify the relative roles of each, an ongoing debate in biogeography (Gowie and Holland 2006, Holland and Gowie 2006, Nelson 2006). Global The global, historical biogeography of the Ampullariidae is not fully understood. Resolving remaining questions about the origins and diversification of the genera will require additional fossil and molecular data and a more complete phylogenetic analysis. However, published hypotheses, based until now only on anatomical and morphological data (Berthold 1991, Simone 2004), have provided an important starting point in answering these questions, and new molecular information is also refining our understanding. According to the morphological analyses (Berthold 1991, Bieler 1993), the Ampullariidae originated in Gondwana and, 140 million years ago, were restricted to parts of Gondwana as follows: Afropomiis and Saulea in southern Africa, Lanistes in southern Africa and Madagascar, and the most recent common ancestor (MRGA) of Pila and the Recent Neotropical genera in southern Africa, Madagascar, southwestern India, and eastern South America (Fig. lA). The subsequent break up of Gondwana led to diversification of this MRGA on the different land masses. It gave rise in South America to five genera: Pomella, Felipponea, Asolene, Marisa, and Pomacea with the first three diversifying early and Pomacea and Marisa being more derived (Berthold 1991, Bieler 1993). In Africa it gave rise to Pila, which spread also into Asia. Afropomiis, Saulea, and Lanistes remained on the African continent and, in the case of Lanistes, in Madagascar (Fig. lA). However, recent studies using DNA sequence data challenge this scenario (Schultheifs et al. 2007, lorgensen et al. 2008). In contrast to Berthold’s (1991) phylogeny, Pila and Lanistes appear as sister taxa in most molecular analyses, suggesting a different Gondwanan distribution of the Ampullariidae from that Berthold had suggested. Acct>rding to the molecular scenario, the ampullariid ancestor gave rise to two lineages: that giving rise to modern Afropomiis and that giving rise to all other extant ampullariids. This second lineage then split into two lineages. One of these gave rise to the sister taxa Lanistes and Pila and diversified within Africa, colonizing Madagascar and, in the case of Pila, Asia; the other, the MRGA o( Saulea and the New World taxa, colonized South America APPLE SNAILS IN EVOLUTIONARY BIOLOGY 49 Afropomus (A) Saulea (S) Lanistes (L) Pila (Pi) Pomella (Pm) Felipponea (F) Asolene (As) Mahsa (M) Pomacea (Pc) : — rt Afropomus (A) Pila (Pi) Lanistes (L) Saulea (S) Asolene (As) Felipponea (F) Marisa (M) Pomella (Pm) Pomacea (Pc) Figure 1. Two hypotheses of Ampullariidae biogeography and diversification. A, the morphological based hypothesis of Berthold (1991) as- sumes an ampullariid ancestor (AA) giving rise to Afropomus, Saulea, and Lanistes in southern Africa, with Lanistes spreading to Madagascar, and the MRCA of Pila and recent Neotropical genera splitting in Africa with Pila diversifying in southern Africa, Madagascar and Asia, and the five Neotropical genera diversifying throughout South and Central America. B, the DNA based scenario (Hayes 2007, Schultheifi et al. 2007, lorgensen et al. 2008) showing the initial divergence of two main lineages in Africa, one giving rise to Afropomus and the other diversifying again and giving rise to Pila and Lanistes, which diversified within Africa but also colonized Madagascar with Pila spreading to Asia. The final lineage, probably sharing a MRCA with Saulea, colonized South America, diversifying into the five currently recognized New World genera. Shading highlights represent the major differences between the two hypotheses. Late jurassic Gondwana maps (ca. 150 mybp) are redrawn from Scotese (2002). and gave rise to the modern Saulea in Africa and the five New World genera (Fig. IB). However, ongoing research (Hayes et al. 2009) shows that this relationship is sensitive to the inclusion of Cyclophoridae as an outgroup (cf. McArthur and Harasewych 2003), which results in Saulea being basal in a monophyletic African clade {Saulea, Pila, Lanistes, Afropomus). Old World Of the four African ampullariid genera, Lanistes, Pila, Afropomus, and Saulea, the latter two are monotypic and restricted to Liberia, Sierra Leone, and the Ivory Coast (Brown 1994). This area is known for its high proportion of endemic freshwater fauna and is considered a distinct freshwater bioregion. Upper Guinea (Thieme et al. 2005). The ampul- lariid fauna in this small region has been considered old and relictual (Van Damme 1984), and given the region’s close geological ties with northeastern South America, makes it a likely candidate for the location of the sister taxa of the New World ampullariids. This contrasts somewhat with Berthold’s (1991) scenario in which he placed the original distribution of ampullariids in South America along the southern coast of what is now Brazil. Pila and Lanistes are generally more widespread in Africa though both genera have widespread as well as locally restricted species. Fossils of both genera are known from the late Cenozoic of the Albertine Rift Valley (Van Damme and Pickford 1995). The five African species of Pila currently recognized (Brown 1994) are distributed across Africa with no clearly discernable biogeographic pattern. Pila ovata (Olivier, 1804) is the only widespread species, although with four of its historically named forms being relatively distinct (Mandahl- Barth 1954). Lanistes is more speciose than African Pila with 19 currently recognized species, mostly with very limited (known) distributions (Brown 1994). Lanistes occurs both in Madagascar and Africa from the lower Nile south to KwaZulu- Natal and the Okavango Delta (Brown 1994). Presuming an age of Lanistes of tens of millions of years, and with the difficulty of delimiting species morphologically, intensive geographical sampling and molecular analysis, as in other ampullariid groups, will probably reveal additional cryptic species. For instance, Lanistes ovum Peters, 1845 is the only widespread species of Lanistes, and although many nominal taxa have been reduced to synonymy with it, some at least may represent cryptic species, with L. ovum in fact being a species complex rather than a single species. Molecular analyses of this widely distributed African taxon will be necessary to resolve this issue. The biogeography of Lanistes especially is a field wide open for investigation. In Asia, Pila is the only native genus. Prashad (1925) recognized ten species of Pila in India, including two species of Turbinicola Annandale and Prashad, 1921, which is a junior synonym (Berthold 1991). Otherwise, Asian Pila, of which 50 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 there may be about 25 species based on Berthold’s (1991) estimate of about 30 species in the genus as a whole, have not been revised comprehensively and their distributions, systematics, and biogeography are therefore also ripe for study. Neotropics In general. Neotropical freshwater biogeography may be better understood than that of Africa, especially with regard to ampullariids. Tectonic events and climatic fluctuations have probably influenced the diversity and distribution of New World ampullariids. The emergence of the Antillean archipelago (~49 mybp; Graham 2003) may have facilitated diversification by both vicariance and dispersal. Connection of the West Indies to northwestern South America, ending 32 mybp (Iturralde-Vinent and MacPhee 1999), may have been important. Other ‘land-based’ scenarios are also possible. There are two main groups of hypotheses to explain Neotropical freshwater biodiversity. Refuge hypotheses (Haffer 1982) posit that diversification resulted primarily from multiple habitat fragmentation and coalescence events driven primarily by Pleistocene climate changes (1.8 million - 11,000 ybp). Hydrogeological hypotheses (Lundberg 1998) suggest that current diversity was reached much earlier, resulting from the changing relationships among South American river systems and their drainages 90-10 mybp. Hydrogeological changes related to tectonic events drove diversification by fracturing and reuniting aquatic habitats multiple times, leading to allopatric speciation. These hypotheses place divergences among drainage biotas much earlier than refuge hypotheses and offer multiple time points that may be correlated to cladogenic events (Sivasundar et al. 2001, Montoya-Burgos 2003). Finally, the rise and completion of the Isthmus of Panama ~3 mybp (Goates and Obando 1996) provided non- marine connections between South and Central American drainages. Phylogeographic patterns in several freshwater fish genera suggest multiple waves of dispersal through Central America from South America (Bermingham and Martin 1998, Perdices et al. 2002). A combination of hypotheses may thus explain New World ampullariid diversification. But historical biogeographic inferences rely on the fossil record and knowledge of phylogenetic relationships of extant taxa; in both regards New World ampullariids are poorly known. Limited fossil evidence places Potiiacea in South America ~50 mybp but an earlier, possibly Gondwanan, origin has been suggested (Berthold 1991, Melchor et al. 2002), with origin and diversification of contemporary New World taxa occurring in South America soon after breakup of the supercontinent (~I80 mybp). An ancient origin of South American ampullariids supports the hydrogeological hypothesis but remains conjecture without knowledge of the temporal pattern of diversification. Thus, New World ampullariids have the potential to illuminate and discriminate among these various general hypotheses of the diversification of the freshwater biota. In South and Gentral America, based on molecular analysis (Fig. IB), the lineage that probably gave rise to the genus Saulea in Africa diversified into five currently recognized genera, Pomella, Asolene, Marisa, Felipponea, and Pomacea. Reconstruction of the relationships within the family, based on morphology, placed Asolene as the most basal of the New World ampullariids with dose ties to both Felipponea and Pomella (Berthold 1991, Bieler 1993). Pomacea and Marisa were placed in more derived positions and as sister taxa (Fig. lA). Pomella has a rather disjunct distribution, with Pomella americanista (Ihering, 1919) and Pomella megastoma (Sowerby, 1825) (subgenus Pomella sensu stricto) occurring in the south (Argentina, Uruguay, Paraguay, southern Brazil) and Pomella sinamarina (Bruguide, 1792) (subgenus Surinamia) in the north of the continent (Guyana, Suriname, French Guiana). Similarly, the distribution of Asolene is non-contiguous, with some species occurring in the south and others restricted to the north. The three species recognized by Cowie and Thiengo (2003) in Felipponea are restricted to the south (Argentina, Uruguay, Paraguay, southern Brazil). Taxa in the more derived Marisa-Pomacea clade have much wider and somewhat more contiguous distributions. The two species of Marisa are distributed from southern Brazil through northern South America and Trinidad and Tobago. Marisa planogyra Pilsbry, 1933 occurs primarily in the south and Marisa cornuarietis (Linnaeus, 1758) in the north, with some possible overlap in northern Brazil. Pomacea is the largest and most diverse genus and has the widest distribution, occurring from Argentina through Central America, the Caribbean and into southeastern North America. Multi-gene phylogenetic results from recent work on New World ampullariids are largely in agreement with previous hypotheses, placing Pomacea as the most derived group of New World ampullariids (Hayes 2007, Hayes et al. 2009). However, Pomacea as currently recognized (Cowie and Thiengo 2003) is not monophyletic, as also suggested by Simone (2004) in his morphological study. Similarly, Hayes (2007) found no support for monophyletic Aso/cne, Felipponea, or Marisa. Instead, both species of Marisa, two species of Felipponea, and several species of Asolene were recovered in a single well -supported basal clade, sister to a clade consisting of other species of Asolene and Pomacea (Fig. 2). These two clades were in turn sister to the larger well-supported group containing the remaining Pomacea species and the only Pomella species included in the analysis, Pomella megastoma. In the past, assignment of species to these genera, especially to Asolene, Felipponea, and Pomella, based on morphological criteria, has been inconsistent (C(owie and 'Lhiengo 2003); these molecular results will help to circumscribe these poorly understood genera. Also, these preliminary data begin to clarify APPLE SNAILS IN EVOLUTIONARY BIOLOGY 51 _Pomacea canaliculata group includes Pomella megastoma 1* , -'.1 -3 Round Calcareous -Pomaceabridgesii group ^ •Pomacea glauca group -Pomacea ftageltata group Honeycomb Calcareous Pomella megastoma eggs Above the water oviposition —Asolene, Pomacea spp. _Marisa. Asolene Felipponea spp. Round Gelatinous Irregular /Round Gelatinous / Calcareous Below or at the water oviposition Marisa planogyra and Pita conica eggs Figure 2. Ampullariidae phylogeny based on analysis of one mitochondrial and three nuclear genes showing the relationships among the ma- jor clades of the monophyletic New World genera (Hayes 2007). Egg morphology and oviposition location are mapped onto the major groups illustrating the evolutionary shift from laying gelatinous eggs below the water to laying calcareous eggs on emergent vegetation. Representative egg clutches from species within each of the major lineages illustrate a high level of morphological conservation within groups. From left to right egg clutches are: Top row, Pomacea insularum (d’Orbigny, 1835), Pomacea canaliculata, Pomacea paludosa; second row, Pomacea diffusa Blume, 1957, Pomacea bridgesii (Reeve, 1856); third row, Pomacea sp., Pomacea glauca, Pomacea guyanensis (Lamarck, 1822); fourth row, Pomacea catemacensis (Baker, 1922); fifth row, Asolene spixii (d’Orbigny, 1838), Felipponea sp.; bottom row, Lanistes ovum, Pila conica (Wood, 1828). Photo credits: J. F. R. Amato {Felipponea sp.), K. C. M. Heiler {Lanistes ovum), J.-P. Pointier {Pomacea glauca), R. C. Joshi (Pila conica), K. Gallagher {Asolene spixii), S. C. Thiengo {Marisa planogyra), and K. A. Hayes (all others). the biogeographic patterns of diversification in New World ampullariids and reveal insights into factors like desiccation resistance, oviposition, and predation pressure that may have played a role in this diversification (Hayes 2009). These issues are discussed below. BIODIVERSITY AND SPECIATION The family Ampullariidae contains more than 150 species, and reaches its highest diversity in the Neotropics with the genus Pomacea, which contains 117 nominally valid species (Cowie and Thiengo 2003). The question of why there are so many Neotropical species has intrigued scientists since Alfred Russel Wallace (1852) first proposed that rivers act as barriers, driving allopatric speciation in Amazonian monkeys. Since then several explanations have been proposed, including Haffer’s (1969) explanation of bird diversity in the context of the so-called “refuge hypothesis” and explanations of freshwater ichthyofaunal diversity based on the “hydrogeological hypothesis” (Lundberg et al. 1998, Montoya-Burgos 2003). Most explanations of high levels of Neotropical diversity have 52 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 been based on vertebrates, invoking vicariance as the primary isolating mechanism (Hall and Harvey 2002, Costa 2003, Ribas et al. 2005). Conspicuously lacking are studies of the huge diversity of invertebrates, including Neotropical freshwater molluscs and studies looking specifically at the complex inter- and intraspecific interactions that drive speciation. Ampullariids have the potential to provide insight into these issues, as well as to reveal other less studied mechanisms generating diversity (Vermeij and Covich 1978, Endler 1982). Since Darwin ( 1859), evolutionary biologists have suggested that ecology plays a vital part in the origin of species (Schluter 2001, Via 2002). The evolutionary ecology of apple snails has great potential for shedding light on the various processes involved in population divergence and ultimately speciation. Here, we address three of these, which may overlap; there are probably others equally amenable to investigation using these snails. Lirst, the interaction between apple snails and their predators, their trophic relations, may influence natural selection regimes acting on both predator and prey. Studies focusing on these interactions should provide key insights into the evolution of these relationships and predator prey co-evolution in general, which in turn will provide information about how such processes help shape biodiversity. Second, habitat partitioning among ampullariids within an ecosystem may provide the necessary isolation for divergent selection to reinforce adaptations leading to reproductive isolation. These interactions may represent major drivers of evolutionary change. Third, because of their wide latitudinal distribution and high species level diversity, ampullariids offer an exciting opportunity to investigate large-scale patterns, or rules, of biodiversity, including Rapoport’s Rule and the latitudinal diversity gradient. Trophic relations Throughout their range, ampullariids are major constituents of tropical/subtropical freshwater diversity and are key taxa in important aquatic ecosystems such as the Elorida Everglades, the Llanos of Venezuela, and the Pantanal of central South America (l)onnay and Beissinger 1993, Fellerhoff 2002, Brown et al. 2006). They may even serve as important indicators of ecosystem health (Ogden et al. 2005). In these systems, they are the main food of snail kites, which include the endangered Everglades Snail Kite, Rostrhamus sociabilis pluniheus, and closely related congeners [i.e., Rostrhanins hamatiis). Because of the abundance of apple snails in these ecosystems and their role in the diet of a variety of animals (birds, fishes, turtles, crocodilians), they could be considered keystone prey species and an important link between aquatic and terrestrial food chains (Donnay and Beissinger 1993, Ebenman and Jonsson 2005). Studios of trophic relationships indicate that these links may have a large influence on species diversity (Paine 1966, Kondoh 2003). Populations of species coupled in predator-prey relationships may often be ecologically linked through both conspicuous predator-prey interactions (Connell 1961, Vermeij 1982) and more cryptic evolutionary dynamics (Yoshida et al. 2007). For instance, variation in the distribution and abundance of apple snail species has been shown to influence the distribution (Angehr 1999), abundance (Darby et al. 2006), and behavior (Tanaka et al. 2006) of snail kites, which have evolved both morphologically and behaviorally for extreme specialization on apple snails (Beissinger et al. 1994). At the same time, predation pressure, by kites and other predators, has probably shaped the morphological and behavioral adaptations of apple snails. For example, Reed and Janzen (1999) determined that the foraging behavior of limpkins {Aramus guarauna) resulted in disruptive selection on shell size in Pomacea flagellata (Say, 1829) in Costa Rica. They also observed directional selection against larger, light-colored snails by snail kites. Dieckmann and Doebeli ( 1 999) modeled just such a predator-prey system and found that predator-prey interactions, when coupled with demographic stochasticity and the resulting genetic drift and assortative mating, often leads to evolutionary branching, which could result in sympatric speciation. Providing further evidence of possibly important evolutionary interactions, Snyder and Snyder (1971) found that Pomacea paludosa (Say, 1829), Pomacea glauca (Linnaeus, 1758), and Pomacea dolioides (Reeve, 1856) exhibit alarm responses to chemical cues from turtle predators, injured or dead conspecifics, and mechanical disturbance. Similar processes may also have shaped the diversity of Old World ampullariids. Van Damme and Pickford (1995) suggested that rapid radiations of Lanistes spp. and changes in Pila spp. in the Rift Valley lakes of East Africa were probably the result of selection pressure from specialized predators, particularly fishes. Rapid and successive morphological changes in these two genera were inferred to have occurred ca. 8-2.5 mybp. A series of impressive Laidstes radiations involving rapid, major changes in shell morphology provides a good model for understanding speciation processes (Van Damme and Pickford 1995). Specifically, two successive radiations occurred, first in Paleolake Obweruka and later in Lake Malawi, both demonstrating convergence on anti- predatory behaviors and morphologies characteristic of a number of Rift Valley lake mollusc species. Some of the patterns seen, particularly lhalassoidism (i.e., shell form resembling marine gastropod species), were attributed to predator-prey interactions that may have triggered speciation. I'he repetitive ampullariid radiatitins were considered as conforming to a punctuated equilibrium model of evolutionary change (Van Damme and Pickford 1995) although such interpretations ol Alrican Cireat Lake fossil gastropod faunas APPLE SNAILS IN EVOLUTIONARY BIOLOGY 53 have long been criticized {e.g., Jones 1981). The fossil radiations may be useful in understanding patterns of more recent radiations of African ampullariids, especially if change in shell morphology can be linked to genetic change. Almost all theory on the tempo and mode of speciation in Lake Malawi, while providing major insight into evolutionary processes, rests on the study of cichlid fishes (Kocher 2004). The recent ampullariid radiation in Lake Malawi was studied morphologically by Berthold (1990), but a detailed molecular study of these snails would potentially shed further light on these questions using a non-fish model system. Recent studies of the Lake Malawi radiation ( Jorgensen etal. 2008, Schultheifi, Van Bocxlaer, Albrecht, and Wilke, pers. comm.) revealed relatively low genetic variation within this clade. This might indicate a young evolutionary age of the radiation, a suggestion previously made by Berthold (1990). In combination with modern morphometric analyses, molecular methods will help to identify general patterns of diversification in ancient lakes, which will help clarify any differences in mode and tempo of speciation between vertebrate taxa like cichlid fish and invertebrates like the ampullariid genus Lanistes. Habitat partitioning In addition to predator-prey interactions, other aspects of ampullariid ecology have probably influenced their current diversity. African Lanistes have both lacustrine (e.g., Lake Malawi, see above) and riverine (e.g., Congo River basin) radiations. In Lake Malawi Lanistes nyassanids Dohrn, 1865 and Lanistes solidus Smith, 1877 differ in their use of microhabitats along a depth gradient, probably related to food availability and differential response to cichlid predators (Louda et al. 1984). Further, using the Lanistes spp. of Lake Malawi as an example, Berthold (1991) explored aspects of speciation and evolution of shell sculpture within the framework of a multidimensional niche concept. He proposed that speciation of Lake Malawi Lanistes was driven by differential adaptations to wave action, food resources, and predators (particularly habitat and behavioral shifts for predator avoidance). Such a scenario has the classic elements that would be anticipated in a case of ecological speciation, whereby reproductive isolation builds between two populations that accumulate adaptations to unique aspects of their environment (Schluter 2001). The Congo River basin radiation of Lanistes consists of Lanistes bicarinatus Germain, 1907, Lanistes congicus Boettger, 1891, Lanistes intortus Martens, 1877, and Lanistes nsend- weensis (Dupuis and Putzeys, 1901). The high levels of concho- logical variation among these species makes inferring their monophyly difficult based on morphological analysis. None- theless, this great variation suggests a role of ecology in species diversification. Investigation of this radiation should focus initially on ascertaining its age, documenting genetic variation. determining monophyly, and identifying common ancestors. However, while it is an example of neither true riverine nor true lacustrine speciation, it offers the possibility of investigating speciation in a habitat type (river basin) that is more permanent on both ecological and evolutionary time- scales than most lake habitats (Giller and Malmquist 1998). If future molecular investigations reject the monophyly of the Congo species, then the basin might be interpreted as a refuge that has been stable over the long term rather than a place of species radiation. Large-scale biodiversity rules A number of patterns of species diversity have been documented across a wide range of taxonomic groups. Most notably, these patterns include Rapoport’s Rule - a positive correlation between species ranges and latitude (Stevens 1989), Bergmann’s rule - increasing body size with increasing latitude in mammals and birds (Bergmann 1847), and the latitudinal biodiversity gradient - decreasing species richness from tropical to polar latitudes (Dobzhansky 1950). The latitudinal biodiversity gradient is one of the longest recognized and most universally accepted patterns in nature (Darwin 1859, Wallace 1878, Hutchinson 1959, Wright et al. 2006), yet there remains little agreement regarding the underlying mechanisms responsible for it (Mittelbach et al. 2007). Three broad categories of explanations have been proposed to explain the gradient, involving ecological, evolutionary, and historical hypotheses (Mittelbach et al. 2007). Ecological hypotheses focus on processes of species coexistence and the maintenance of species diversity through species interactions, and apple snails have been mentioned above as a system with which to investigate such processes. Evolutionary hypotheses focus on rates of diversification. And historical explanations are based on the persistence and extent of tropical environments. Understanding the relationship between latitude and speciation has been hindered by a lack of comparative analyses across a single clade that inhabits both tropical and temperate regions. Here again, apple snails may serve as a good system for investigating the underlying processes, and they can be used to test explicitly several of the proposed hypotheses. The oft-cited “diversification rate hypothesis” suggests that high tropical diversity results from high rates of speciation (Fischer 1 960) caused by one or more of the following: ( 1 ) greater opportunities for reproductive isolation because lower latitudes contain larger area (Terborgh 1973), (2) increased rates of molecular evolution due to higher metabolic rates in warmer regions (Rohde 1992, Wright et al. 2006), (3) enhanced biotic interactions because of increased specialization and reduced dispersal (Dobzhansky 1950, Janzen 1967), and (4) lower extinction rates due to increased climatic stability (DanUn 1859, Fischer 1960) or larger population sizes (Terborgh 1973). 54 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Using ampullariids to investigate the various mechanisms responsible for higher levels of tropical than of temperate diversity w'ill require data from paleontology, biogeography, ecology, and phylogenetics. However, using preliminary data (Hayes 2009) we can begin addressing at least one of these hypotheses. The explanation of Rohde (1992), supported by Wright et al. (2006), posits that higher tropical diversity results from an increased rate of molecular evolution in the tropics relative to higher latitudes. Species of Pomacea are an ideal group to test this hypothesis, as they range from temperate Argentina to the southeastern U.S.A. Using the approach of Wright et al. (2006), rate heterogeneity in molecular evolution can be tested using sister taxa, one of which occurs in the tropics and the other in a temperate region (e.g., Pomacea canaliculata (Lamarck, 1822) and Pomacea dolioides). Hayes (2009) found that ampullariid diversity indeed decreases with increasing latitude. If future research finds a difference in rate of evolution, these taxa could be used to investigate further the possible mechanisms driving the differences. The “historical time and area hypothesis” contends that areas with tropical climates are historically larger and older, which has allowed more opportunity for diversification (Lischer 1960, Wiens et al. 2006). If this were the primary driver of greater tropical versus temperate diversity, we should expect tropical species to be older and temperate species to be nested within clades of tropical taxa. Also we should expect diversity to be correlated with the age of geographical regions. Data emerging from ongoing work on ampuUariids (Schultheifi et al. 2007, Jorgensen et al. 2008, Hayes et al. 2009) are beginning to provide the phylogenetic and biogeographic framework to address such hypotheses. PHYSIOLOGY AND BEHAVIOR In addition to addressing broad c^uestions of biogeography and speciation, apple snails provide an excellent system for studying the evolution of physiological and behavioral adaptations, aspects of which may have profound implications for the generation of diversity, and for addressing important questions in behavioral ecology and evolution. Mapping apple snail oviposition location onto a preliminary phylogeny, Hayes (2007) found that laying eggs on emergent vegetation or other above-water hard surfaces is a synapomorphy that unites the most derived clade consisting predominantly of snails currently referred to Pomacea. Other ampullariids, including Old World and basal New World taxa, oviposit either on vegetation below or at the water line or in mud close to it (Gowie 2002) (Fig. 2). 'Fhis observation, combined with the fact that this derived group is also the most speciose and covers the widest geographical range, leads to speculation that this shift to above water oviposition may have been a key innovation that accompanied the diversification and spread of the group. Other characteristics in the above-water egg- laying group seem to include longer siphons (for aerial respiration), increased lung size, and increased desiccation resistance (Gowie 2002). All these factors may be correlated with the success of the group. Unique egg morphologies are associated with each of the clades in this above-water oviposition group, with the most derived group having spherical eggs that cluster relatively loosely in the egg mass. The more basal taxa in this group lay eggs that are honeycombed or polygonal in shape and abut tightly against one another within the egg mass (Fig. 2). It is possible that the derived condition of spherical, loosely clustered eggs may also have contributed to the success of these taxa through increased hatching rate resulting from more efficient respiration through the egg shell although respiration rates in clutches with different morphologies have yet to be measured. Nuptial feeding is any form of nutrient transfer from the male to female during or directly after courtship or copulation. Burela and Martin (2007) reported nuptial feeding in Pomacea canaliculata, the first time it has been reported in a gastropod. Such behavior has implications for sexual selection and fitness. Burela and Martin (2007) discussed several possible advantages, including enhanced male fitness through benefits conferred to the offspring via additional nutrients, and mate attraction, mate acceptance, or increasing the length of copulation to maximize sperm transfer. Either way, this is a fascinating behavior that has interesting evolutionary implications for apple snails and mating behavior in general. Burela and Martin (2007) suggested that given the high level of similarity in the general body plan across the Ampullariidae, this behavior is probably not exclusive to this species, and may be found more widely. BIOLOGICAL INVASIONS Invasive species are now recognized throughout tlie world as a major economic and environmental threat (Pimentel et al. 2005, Puth and Post 2005). While these alien invasions cause tremendous agricultural, conservation, and human health problems, the rapid evolutionary changes that often accompany such unplanned invasion experiments may permit a greater understanding of the natural world, and at the same time provide insights into a variety of ecological and evolutionary processes (Sax et al. 2007). 'I'hat rapid evolutionary changes occur alter the introduction of alien species has become increasingly well documented (Gox 2004, Garroll et al. 2005, I liiey et al. 2005), and more studies are taking advantage ot these “accidental experiments” to investigate contemporary evolution. Such changes may olten APPLE SNAILS IN EVOLUTIONARY BIOLOGY 55 take place in tens to hundreds of generations instead of the millions that most evolutionary processes are normally thought to occur over, and they take place in both the alien and the native species that interact during invasions (Sax etal. 2007). A number of ampullariid species have become invasive outside their native ranges, particularly species of Pomacea and Marisa (Joshi and Sebastian 2006, Rawlings et al. 2007, Hayes et al. 2008). It is possible that the adaptive genetic changes necessary to be successful invasives are occurring rapidly in these species, and these processes may be understood better by integrating ecological and evolutionary perspectives. Wada and Matsukura (2007) have shown that Pomacea canaliculata has adopted at least two strategies for dealing with overwintering in its introduced range in Japan; burial in mud or seeking refuge under rice straw before the onset of winter. They found seasonal differences in cold hardiness of snails, suggesting that cold winters may impose strong natural selection on such populations. It is still uncertain whether this is an adaptation acquired after introduction or one possessed by source populations in their native ranges. However, in either case it demonstrates that ampullariids may be an illuminating system for studying adaptive strategies of introduced species. Because multiple ampullariid species have been introduced, comparative studies among them may reveal key differences in such strategies. The results of such studies will not only strengthen our understanding of invasion biology but may also allow us to investigate the patterns and tempo of adaptive genetic changes along with the influence of founder events on these processes. CONCLUSIONS More than 150 years after Darwin’s revolutionary idea of descent with modification, there remain a number of unanswered questions fundamental to our understanding of evolution and biodiversity. How many species are there? How are these species distributed? What are the processes that generate this biodiversity? Many of these questions remain unanswered simply because of the complexity of the evolutionary process. Eor example, the myriad mechanisms that might lead to the evolution of reproductive isolation { i.e., speciation) are often difficult to disentangle. Yet addressing these issues in a range of groups, particularly those with the highest diversity, may reveal additional insights that will go a long way to answering these big evolutionary questions. Ampullariids offer an excellent system for addressing many of these questions, particularly regarding the generation of biodiversity and how it spread and diversified around the globe. Lessons learned from this group may be generalized not only to other freshwater taxa but also to more profound and over-arching themes in evolutionary biology. ACKNOWLEDGMENTS We thank Matthias Glaubrecht and Thomas von Rintelen for inviting us to participate in the symposium they organized and for the invitation to write this contribution. We owe a debt of gratitude to numerous collaborators for their help obtaining ampullariid material from around the world, particularly those in SE Asia and Brazil. Funding was provided to RHC by the U.S. Department of Agriculture and to KAH by the Ecology, Evolution, and Conservation Biology program of the University of Hawaii (NSF grants DGE0232016 and DGE0538550 to K. Y. Kaneshiro) and an American Malacological Society Student Research Award. KAH’s attendance at the symposium was partially funded by a grant from Unitas Malacologica. The Villum Kami Rasmussen Foundation provided funding to AJ in support of his research and attendance at the symposium. LITERATURE CITED Albrecht, C., K. Kuhn, and B. Streit. 2007. A molecular phylogeny of Planorboidea (Gastropoda, Pulmonata): Insights from en- hanced taxon sampling. Zoologica Scripta 36: 27-39. Angehr, G. R. 1999. Rapid long-distance colonization of Lake Gatun, Panama by snail kites. Wilson Bulletin 111: 265-268. Beissinger, S. R., T. J. Donnay, and R. Walton. 1994. Experimental analysis of diet specialization in the snail kite: The role of beha- vioral conservatism. Oecologia 100: 540-565. Bergmann, C. 1847. Uber die Verhaltnisse der Warmeokonomie der Thiere zu ihrer Grosse. Gottinger Studien, Gottingen 3: 595-708 [In German]. Bermingham, E. and A. P. Martin. 1998. Comparative mtDNA phy- logeography of neotropical freshwater fishes; Testing shared history to infer the evolutionary landscape of lower Central America. Molecular Ecology 7: 499-517. Berthold, T. 1990. Intralacustrine speciation and the evolution of shell sculpture in gastropods of ancient lakes - application of Gunther’s niche concept. Abhandlungen des Naturwissenschoft- lichen Vereins in Hamburg (NF) 31132: 85-1 18. Berthold, T. 1991. Vergleichende Anatomie, Phylogenie und Histori- sche Biogeographie der Ampullariidae (Mollusca, Gastropoda). Abhandlungen des Naturwissenschaftlichen Vereins in Hamburg (NF) 29: 1-256 [In German]. Bieler, R. 1993. Ampullariid phylogeny — Book review and cladistic re-analysis. The Veliger 36: 291-299. Brown, D. S. 1994. Freshwater Snails of Africa and Their Medical Im- portance, 2"^* Edition. Taylor 8< Francis, London. Brown, M. T., M. J. Cohen, E. Bardi, and W. W. Ingwersen. 2006. Species diversity in the Florida Everglades, USA: A systems approach to calculating biodiversity. Aquatic Sciences 68: 254- 277. Burela, S. and P. B. Martin. 2007. Nuptial feeding in the freshwater snail Pomacea canaliculata (Gastropoda: Ampullariidae). Mala- cologia 49: 465-470. 56 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Carroll, S. R, I. E. Loye, H. Dingle, M. Mathieson, T. R. Famula, and M. R Zalucki. 2005. And the beak shall inherit — evolution in response to invasion. Ecology Letters 8: 944-951. Coates, A. G. and I. A. Obando. 1996. The geologic evolution of the Central American isthmus. In: ]. Jackson, A. F. Budd, and A. G. Coates, eds.. Environment and Evolution in Tropical America. Chicago University Press, Chicago. Pp. 21-56. Connell, I. H. 1961. The influence of interspecific competition and other factors on the distribution of the barnacle Chthamalus stellatus. Ecology 42: 710-723. Costa, L. P. 2003. The historical bridge between the Amazon and the Atlantic forest of Brazil: A study of molecular phylogeography with small mammals. Journal of Biogeography 30: 71-86. Cowie, R. H. 2002. Apple snails (Ampullariidae) as agricultural pests: Their biology, impacts and management. In: G. M. Barker, ed.. Molluscs as Crop Pests. CABI Publishing, Wallingford, U.K. Pp. 145-192. Cowie, R. FI. and S. C. Thiengo. 2003. The apple snails of the Ameri- cas (Mollusca: Gastropoda: Ampullariidae: Asolene, Felipponea, Marisa, Pomacea, Pomella): A nomenclatural and type catalog. Malacologia 45: 41-100. Cowie, R. H. and B. S. Holland. 2006. Dispersal is fundamental to biogeography and the evolution of biodiversity on oceanic is- lands. Journal of Biogeography 33: 1 93- 1 98. Cowie, R. H., K. A. Hayes, and S. C. Thiengo. 2006. What are apple snails? Confused taxonomy and some preliminary resolution. In: R. C. loshi and L. C. Sebastian, eds.. Global Advances in Ecology and Management of Golden Apple Snails. Philippine Rice Research Institute, Munoz, Nueva Ecija, Philippine. Pp. 3-23. Cox, G. W. 2004. Alien Species and Evolution: The Evolutionary Ecol- ogy of Exotic Plants, Animals, Microbes, and Interacting Native Species. Island Press, Washington, D.C. Darby, P. C., R. E. Bennetts, and L. B. Karunaratne. 2006. Apple snail densities in habitats used by foraging snail kites. Florida Field Naturalist 34: 37-68. Darwin, C. 1859. On the Origin of Species by Means of Natural Selec- tion. John Murray, London. Dejong, R. J., J. A. T. Morgan, W. Lobato Paraense, J. P. Pointier, M. Amarista, P. E K. Ayeh-Kumi, A. Babiker, C. S. Barbosa, P. Bremond, A. P. Canese, C. Pereira de Souza, C. Dominguez, S. File, A. Gutierrez, R. N. Incani, T. Kawano, E Kazibwe, J. Kpikpi, N. J. S. Lwambo, R. Mimpfoundi, F. Njiokou, J. N. Poda, M. Sene, L. E. Velasc]uez, M. Yong, C. M. Adema, B. V. Hofkin, G. M. Mkoji, and E. S. I.oker. 2001. Evolutionary relationships and biogeography of Biomphalaria (Gastropoda: Jdanorbidae) with implications regarding its role as host of the human blood- Iluke, Schistosotna mattsoni. Molecular Biology and Evolution 18: 2225-2239. Dieckmann, U. and M. Doebeli. 1999. On the origin of species by sympatric speciation. Nature 4{)i): 354-357. Dobzhansky, T. 1950. Evolution in the tropics. American Scientist 38: 208-221. Donnay, 'J'. J. and S. R. Beissinger. 1993. Apple snail (Pomacea doliodes (s/c|) and freshwater crab (Dilocarcinus dcntatus) population fluctuations in the I.lanos of Venezuela. Biotropica 25: 206-214. Ebenman, B. and T. Jonsson. 2005. Using community viability analy- sis to identify fragile systems and keystone species. Trends in Ecology and Evolution 20: 568-575. Endler, J. 1982. Convergent and divergent effects of natural selection on color patterns in two fish faunas. Evolution 36: 178-188. Facon, B., J.-P. Pointier, M. Glaubrecht, C. Poux, P. Jarne, and P. Da- vid. 2003. A molecular phylogeography approach to biological invasions of the New World by parthenogenetic thiarid snails. Molecular Ecology 12: 3027-3039. Fellerhoff, C. 2002. Feeding and growth of apple snail Pomacea lin- eata in the Pantanal wetland, Brazil — a stable isotope approach. Isotopes in Environmental Health Studies 38: 227-243. Fischer, A. G. 1960. Latitudinal variations in organic diversity. Evolu- tion 14: 64-81. Giller P. S. and B. Malmqvist. 1998. The Biology of Streams and Rivers. Oxford University Press, Oxford. Graham, A. 2003. Historical phytogeography of the Greater Antilles. Brittonia 55: 357-383. Haffer, J. 1969. Speciation in Amazonian forest birds. Science 165: 131-137. Haffer, J. 1982. General aspects of the refuge theory. In: G. T. Prance, ed.. Biological Diversification of the Tropics. Columbia Univer- sity Press, New York. Pp. 6-24. Hall, J. P. and D. J. Harvey. 2002. The phylogeography of Amazonia revisited: New evidence from Riodinid butterflies. Evolution 56: 1489-1497. Hayes, K. A. 2007. Molecular systematics and evolutionary patterns of diversification in New World Ampullariidae. In: K. Jordaens, N. Van Houtte, J. Van Goethem, and T. Backeljau, eds.. Abstracts of the World Congress of Malacology 2007, Antwerp, Belgium. Unitas Malacologica, Antwerp. P. 93. Hayes, K. A. 2009. Evolution, molecular systematics and invasion biology of Ampullariidae. Ph.D. Dissertation, Department of Zoology, University of Hawaii, Honolulu. Hayes, K. A., R. C. Joshi, S. C. Thiengo, and R. H. Cowie. 2008. Out of South America: Multiple origins of non-native apple snails in Asia. Diversity and Distributions 14: 701-712. Hayes, K. A., S. C. Thiengo, and R. H. Cowie. 2009. A global phylog- eny of apple snails: Gondwanan origin, generic relationships and the influence of outgroup choice (Caenogastropoda: Amp- ullariidae). Biological Journal of the Linncan Society (in press) Holland, B. S. and R. H. Cowie. 2006. Dispersal and vicariance in Hawaii: Submarine slumping does not create deep inter-island channels. Journal of Biogeography 33: 2 1 55-2 1 57. Huey, R. B., G. W. Gilchrist, and A. P. Hendry. 2005. Using inva- sive species to study evolution: Case studies with Drosophila and salmon. In: D. F. Sax, J. ). Stachowicz, and S. D. Ciaines, eds.. Species Invasio)is: Insights into Ecology, Evolution and Bio- geography. Sinauer Associates, Sunderland, Ma.ssachusetts. Pp. 139-164. Hutchinson, G. E. 1959. Homage to Santa Rosalia or why are there so many kinds ol animals? American Naturalist 93: 145-159. Iturralde-Vinent, M. A. aiul R. D. E. MaePhee. 1999. Paleogeography ol the Caribbean region: Iniplications for tienozoic biogeogra- phy. Bulletin oj the American Museum of Natural History 238: I -95. APPLE SNAILS IN EVOLUTIONARY BIOLOGY 57 lanzen, D. H. 1967. Why mountain passes are higher in the tropics. American Naturalist 101: 233-249. lones, I. S. 1981. An uncensored page of fossil history. Nature 293: 427-428. Jorgensen, A., T. K. Kristensen, and H. Madsen. 2008. A molecu- lar phylogeny of apple snails (Gastropoda, Caenogastropoda, Ampullariidae) with an emphasis on African species. Zoologica Scripta 37: 245-252. Joshi, R. C. and L. S. Sebastian. 2006. Global Advances in Ecology and Management of Golden Apple Snails. Philippine Rice Research Institute, Nueva Ecija, Philippine. Kocher, T. D. 2004. Explosive speciation: The cichlid fish model. Nature 5: 288-298. Kondoh, M. 2003. Foraging adaptation and the relationship between food-web complexity and stability. Science 299: 1388-1391. Lindberg, D. R., W. F. Ponder, and G. H. Haszprunar. 2004. The Mollusca: Relationships and patterns from their first half-bil- lion years. In: J. Cracraft, ed.. Assembling the Tree of Life. Ox- ford University Press, Cary, North Carolina. Pp. 252-278. Louda, S. M., K. R. McKaye, T. D. Kocher, and C. J. Stackhouse. 1984. Activity, dispersion, and size of Lanistes nyassanus and L. solidus (Gastropoda, Ampullariidae) over the depth gradient at Cape Maclear, Lake Malawi, Africa. The Veliger 26: 145-152. Lundberg, J. G. 1998. The temporal context of the diversification of Neotropical fishes. In: L. R. Malabarba, R. E. Reis, R. P. Vari, Z. M. S. Lucena, and C. A. S. Lucena, eds., Phylogeny and Clas- sification of Neotropical Fishes. Edipucrs, Porto Alegre, Brazil. Pp. 49-68. McArthur, A. G. and M. G. Harasewych. 2003. Molecular systematics of the major lineages of the Gastropoda. In: C. Lydeard and D. R. Lindberg, eds.. Molecular Systematics and Phylogeography of Mollusks. Smithsonian Institution, Washington, D.C. Pp. 140- 160. Mandahl-Barth, G. 1954. The freshwater mollusks of Uganda and adjacent territories. Annales du Musee Royal du Congo Beige, Tervuren, 8°, Sciences Zoologiques 52: 1-206. Mavarez, ]., C. Steiner, J.-P. Pointier, and P. Jarne. 2002. Evolutionary history and phylogeography of the schistosome-vector fresh- water snail Biomphalaria glabrata based on nuclear and mito- chondrial DNA sequences. Heredity 89: 266-271. Melchor, R. N., J. F. Genise, and S. E. Miquel. 2002. Ichnology, sedi- mentology and paleontology of Eocene calcareous paleosols from a palustrine sequence, Argentina. Palaios 17: 16-35. Mittelbach, G. G., D. W. Schemske, H. V. Cornell, A. P. Allen, J. M. Brown, M. B. Bush, S. P. Harrison, A. H. Hurlbert, N. Knowl- ton, H. A. Lessios, C. M. McCain, A. R. McCune, L. A. Mc- Dade, M. A. McPeek, T. J. Near, T. D. Price, R. E. Ricklefs, K. Roy, D. F. Sax, D. Schluter, J. M. Sobel, and M. Turelli. 2007. Evolution and latitudinal diversity gradient: Speciation, extinc- tion and biogeography. Ecology Letters 10: 315-331. Montoya-Burgos, J. I. 2003. Historical biogeography of the catfish genus Hypostomus (Siluriformes: Loricariidae), with implica- tions on the diversification of Neotropical ichthyofauna. Mo- lecular Ecology 12: 1855-1867. Nelson, G. 2006. Hawaiian vicariance. Journal of Biogeography 33: 2154-2155. Ogden, J. C., S. M. Davis, T. K. Barnes, K. J. Jacobs, and J. H. Gentile. 2005. Total system conceptual ecological model. Wetlands 25: 955-979. Paine, R. T. 1966. Food web complexity and species diversity. Ameri- can Naturalist 100: 65-75. Perdices, A., E. Bermingham, A. Montilla, and I. Doadrio. 2002. Evo- lutionary history of the genus Rhamdia (Teleostei: Pimelodi- dae) in Central America. Molecular Phylogenetics and Evolution 25: 172-189. Pimentel, D., R. Zuniga, and D. Morrison. 2005. Update on the en- vironmental and economic costs associated with alien-invasive species in the United States. Ecological Economics 52: 273-288. Prashad, B. 1925. Revision of the Indian Ampullariidae. Memoirs of the Indian Museum 8: 69-89. Puth, L. M. and D. M. Post. 2005. Studying invasion: Have we missed the boat? Ecology Letters 8: 715-721. Rawlings, T. A., K. A. Hayes, R. H. Cowie, and T. M. Collins. 2007. The identity, distribution and impacts of non-native apple snails in the continental United States. BMC Evolutionary Biology 7: 97. Reed, W. L. and F. J. Janzen. 1999. Natural selection by avian preda- tors on shell size and color of a freshwater snail. Biological Jour- nal of the Linnean Society 67: 331-342. Ribas, C. C., R. Gaban-Lima, C. Y. Miyaki, and J. Cracraft. 2005. His- torical biogeography and diversification within the Neotropical parrot genus Pionopsitta (Aves: Psittacidae). Jounal of Biogeog- raphy 52: 1409-1427. Rohde, K. 1992. Latitudinal gradients in species diversity: The search for the primary cause. Oikos 65: 514-527. Sax, D. E, J. J. Stachowicz, J. H. Brown, J. F. Bruno, M. N. Dawson, S. D. Gaines, R. K. Grosberg, A. Hastings, R. D. Holt, M. M. Mayfield, M. I. O’Connor, and W. R. Rice. 2007. Ecological and evolutionary insights from species invasions. Trends in Ecology and Evolution 22: 465-471. Schluter, D. 2001. Ecology and origin of species. Trends in Ecology and Evolution 16: 372-380. Schultheifi, R., T. Geertz, K. Heiler, and C. Albrecht. 2007. Challenging the biogeographical scenario of diversification of the Ampullariidae (Caenogastropoda) using molecular meth- ods. In: K. Jordaens, N. Van Houtte, J. Van Goethem, and T. Backeljau, eds.. Abstracts of the World Congress of Malacology 2007, Antwerp, Belgium. Unitas Malacologica, Antwerp. P. 199. Scotese, C. R. 2002. PALEOMAP project. Available at http://www. scotese.com 1 July 2007. Simone L. R. L. 2004. Comparative morphology and phylogeny of representatives of the superfamilies of architaenioglossans and the Annulariidae (Mollusca, Caenogastropoda). Arquivos do Museu Nacional (Rio de Janeiro) 62: 387-504. Sivasundar, A., E. Bermingham, and G. Orti. 2001. Population structure and biogeography of migratory freshwater fishes (Prochilodus:Characiformes) in major South American rivers. Molecttlar Ecology 10: 407-417. Snyder, N. F. R. and H. A. Snyder. 1971. Defenses of the Florida apple snail Pomacea paludosa. Behaviour 40: 175-215. Stevens, G. C. 1989. The latitudinal gradient in geographical range: How so many species coexist in the tropics. American Natural- ist 133: 240-256. 58 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Strong, E. E., O. Gargominy, W. E. Ponder, and P. Bouchet. 2008. Global diversity of gastropods (Gastropoda; Mollusca) in fresh- water. Hydrobiologia 595: 149-166. Tanaka, M. O., A. L. T. Souza, and E .S. Modena. 2006. Habitat struc- ture effect on size selection of snail kites {Rostrhamus sociabilis) and limpkins [Aramus giiarauna) when feeding on apple snails [Pomacea spp.). Acta Oecologica 30: 88-96. Terborgh, ]. 1973. On the notion of favorableness in plant ecology. American Naturalist 107: 481-501. Thieme, M. L., R. Abell, M. L. 1. Stiassny, P. Skelton, B. Lehner, G. G. Teugels, E. Dinerstein, A. K. Toham, N. Burgess, and D. Olson. 2005. Freshwater Ecoregions of Africa and Madagascar. A Con- servation Assessment. Island Press, Washington, D.G., Govelo, London. Thiengo, S. C. 1989. On Pomacea sordida (Swainson, 1823) (Proso- branchia, Amptillariidae). Memorias do Instituto Oswaldo Cruz 84: 351-355. Thiengo, S. C., C. E. Borda, and J. L. Barros Araujo. 1993. On Poma- cea canaliculata (Lamarck, 1822) (Mollusca, Pilidae: Ampulla- riidae). Memorias do Instituto Oswaldo Cruz 88: 67-71. Thiengo, S. G., K. A. Hayes, A. C. Mattos, M. A. Fernandez, and R. H. Gowie. 2007. South American Ampullariidae: Morphology and taxonomy of the genus Pomacea. Ill Taller de Biologta de Am- pullariidae, Tupungato, Argentina, 29 November - 2 December 2007. Abstracts, P. 8. Tracey, S., L A. Todd, and D. H. Erwin. 1993. Mollusca: Gastropoda. In: M. J. Benton, ed.. The Fossil Record 2. Chapman & Hall, New York. Pp. 131-167. van Bruggen, A. C. 1995. Biodiversity of the Mollusca: Time for a new approach. In: A. C. van Bruggen, S. M. Wells, and T. C. M. Kemperman, eds.. Biodiversity and Conservation of the Mollus- ca. Backhuys Publishers, Oegstgeest-Leiden, The Netherlands. Pp. 1-19. Van Damme, D. 1984. The Freshwater Mollusca of Northern Africa. Dr. W. funk Publishers, Dordrecht, The Netherlands. Van Damme, D. and M. Pickford. 1995. The late Cenozoic Ampul- lariidae (Mollusca, Gastropoda) of the Albertine Rift Valley (Ugandda-Zaire). Hydrobiologia 316: 1-32. Vermeij, G. ). 1982. Unsuccessful predation and evolution. The American Naturalist 120: 701-720. Vermeij, G. ). and A. P. Covich 1978. Coevolution of freshwater gas- tropods and their predators. The American Naturalist 112: 833- 843. Via. S. 2002. The ecological genetics of speciation. American Natu- ralist 159: SI-S7. Wada, f. and K. Matsukura. 2007. Seasonal changes in cold hardi- ness of the invasive freshwater apple snail, Pomacea canalicu- lata (Lamarck) (Gastropoda: Ampullariidae). Malacologia 49: 383-392. Wallace, A. R. 1852. On the monkeys of the Amazon. Proceedings of the Zoological Society of London 20: 1 07- 1 1 0. Wallace, A. R. 1878. Iropiccd Nature and Other Essays. Macmillian, London. Wang, ll.-l. 1984. 'I'wo Upper jurassic gastropod opercula in China. Acta Palaeontologica Sinica 23: 369-372. Wiens, J. )., C;. II. (iraham, 1). S. Moen, S. A. Smith, and T. W. Re- eder. 2006. Evolutionary and ecological causes of the latitu- dinal diversity gradient in hylid frogs: Treefrog trees unearth the roots of high tropical diversity. American Naturalist 168: 579-596. Wright, S., L Keeling, and L. Gillman. 2006. The road from Santa Ro- salia: A faster tempo of evolution in tropical climates. Proceed- ings of the National Academy of Sciences 103: 7718-7722. Yoshida, T., S. P. Ellner, L. E. Jones, B. J. M. Bohannan, R. E. Lenski, and N. G. Hairston, Jr. 2007. Cryptic population dynamics: Rapid evolution masks trophic interactions. PLOS Biology 5: e235 doi:l 0.1 371 /journal. pbio.0050235. Submitted; 9 November 2008; accepted: 4 February 2009; final revisions received: 16 March 2009 Amer. Maine. Bull. 27: 59-68 (2009) Land snail models in island biogeography: A tale of two snails'^ Brenden S. Holland and Robert H. Cowie Center for Conservation Research and Training, Pacific Biosciences Research Center, University of Hawaii, 3050 Maile Way, Gilmore 408, Honolulu, Hawaii 96822, U.S.A. Corresponding author: bholland@hawaii.edu Abstract: Oceanic islands have long been important in evolutionary biology. Land snails are a major component of oceanic island biotas and have much to offer as systems for addressing major questions in evolution and biogeography. We review patterns of within-archipelago biogeography and diversification in two large Hawaiian land snail groups, the Succineidae and the Achatinellinae. Molecular studies suggest that long-distance oceanic dispersal and colonization of the Hawaiian Islands has been rare but between-island dispersal has been far more common. Long-distance oceanic dispersal is the most important driver for deep phylogenetic divergence. Dispersal is also important within the archipelago, while among-island vicariant processes result in only a portion of tip clade diversity. The Achatinellinae are monophyletic but there is evidence of a deep phylogenetic split between the two Hawaiian succineid clades, a result of two independent colonizations reflecting two oceanic dispersal events. Hawaiian succineids have also dispersed to Samoa and Tahiti. Dispersal is an important biogeographical phenomenon, and its role in shaping distributions of island lineages should not be underestimated. Because of their relatively sedentary nature, yet a proclivity for long-distance passive dispersal, island snails can facilitate insights into mechanisms of evolutionary diversification. Important phylogenetic lessons are emerging from studies of island snails and such studies will eventually allow estimation of ages of species groups, speciation rates, timing of the processes involved in community assembly, and other dynamics, all of which are important contributions to the overall understanding of evolution. Key words: Succineidae, Achatinellinae, phylogeography, vicariance, Hawaii Islands have long been considered important model systems for the study of evolution and biogeography. In the Pacific, the Hawaiian and Galapagos Islands in particular have provided riumerous novel scientific insights (e.g., Cain 1984, Grant 1986). Land snails are an important component of biodiversity in many oceanic island groups (e.g., Cowie 2004, Cameron et al. 2007) and have great potential as informative models with which to address major evolutionary, especially biogeographic, questions. Specific biological attributes of island snails that render them informative as models for biogeographic analyses include their high levels of species endemism and often broad distributions of families. This allows phylogenetic studies to use a hierarchical systematic approach to illuminate nested evolutionary patterns from the levels of populations to families. A number of snail families have remarkable passive dispersal and colonization abilities, given geological time, in seeming contrast with their low active mobility and often relatively small spatial ranges at the species level (e.^.. Barker and Mayhill 1999, Fontaine et al. 2007). Certain land snails are adept at dispersing across vast geographic distances, including ocean basins. This has been widely acknowledged and documented {e.g., Darwin 1859, Anonymous 1936, Rees 1965, Dundee et al. 1967, Vagvolgyi 1975) and is evidenced by the worldwide distributions of a number of lineages. It is now supported by molecular evidence (Gittenberger et al. 2006). In this paper, we review the phylogenetics, biogeography, and phylogeography of two major groups of land snails, the Succineidae and Achatinellinae, that have diversified across the Hawaiian archipelago in contrasting ways, illustrating the value of studies of island snails in evolution and biogeography. Within each group, we first address interspecific phylogenetic relationships and colonization patterns and then focus on intraspecific phylogeography. Since the Hawaiian land snail fauna is highly threatened (Solem 1990, Lydeard et al. 2004), understanding diversification patterns and evolutionary history can also provide valuable insights for conservation {e.g., Holland and Hadfield 2002). THE HAWAIIAN ISLANDS AND THEIR BIOTA The Hawaiian archipelago consists of a sequence of oceanic islands formed as the Pacific plate moves northwest- ward over a stationary “hot spot” in the earth’s mantle. The hot spot sends magma up through the plate, creating a chain From the symposium “Molluscs as models in evolutionary biology: from local speciation to global radiation” presented at the World Con- gress of Malacology, held from 15 to 20 july 2007 in Antwerp, Belgium. 59 60 AMERICAN MALACOLOGICAL BULLETIN 27 • 1 /2 • 2009 of volcanoes, each sequentially younger than the one that preceded it, northwestward away from the hot spot. High islands eventually subside and erode to become low atolls, then submerged seamounts, and are ultimately subducted as the Pacific plate slides under the adjacent tectonic plate (Price and Clague 2002). The current islands are divided into the younger “high” islands and the older northwestern islands, which have become low atolls or small, eroded pinnacles or islets. Currently, the oldest, northwestern-most island is Kure Atoll (29 Ma) and the oldest high island is Kauai (5.1 Ma), with the youngest island, Hawaii itself, being less than 0.5 Ma and still forming. The islands harbor unique forms of plants and animals that are the products of evolution in isolation over tens of millions of years, as older islands vanished and new islands formed. Most of the current diversity occurs on the high islands, from the island of Kauai in the northwest to the island of Hawaii in the southeast (Pig. 1). The islands’ extreme isolation, diversity of microhabitats, and dynamic geology have led to spectacular biological endemism as well as prominence as a global biodiversity hotspot (Simon 1987, Ziegler 2002). The unique, terrestrial evolutionary radiations have attracted scientific attention for over a century (Gulick 1905, Carson 1987, Wagner and Punk 1995, Hormiga et al. 2003). In recent decades, however, the Hawaiian biota has become increasingly threatened. Anthropogenic habitat destruction and the devastating impacts of non-native species (Staples and Cowie 2001, Eldredge and Evenhuis 2003) have led to high levels of extinction (Vitousek 1988, Pimm et al. 1994, Wagner et al. 1999). There are over 750 nomenclaturally valid, native, land snail species in 1 1 families in the Hawaiian Islands; over 99% of these species are endemic (Cowie 1995, Cowie et al. 1995). It has long been assumed that each endemic group (genus, subtamily, family) of species within this huge diversity has resulted from in situ speciation follow- ing a single colonization by a single ancestral lineage (Zimmerman 1948) although until recently this had not been rigorously tested. Ascertaining biogeographic origins of land snail radiations is important for under- standing their natural history, and the fact that multiple origins have now been demonstrated in some Hawaiian terrestrial invertebrate groups (Gillespie et al. 1994, Robinson and Sattler 2001, Rundell etal. 2004) suggests that other taxa may have similarly complex evolutionary histories. Hawaiian land snails have suffered higher levels of extinction than perhaps any other group of Hawaiian organ- isms, with estimates as high as 90% of species now extinct (Cowie 2001 ) and most of the remainder seriously threatened. As a consequence of their present conservation status, there is an urgent need for data regarding the systematics and evolutionary history of the fauna, especially as legislative decisions regarding conservation require an unambiguous understanding of taxonomic status (e.g., Avise 2000, Holland and Hadfield 2004, 2007). COMPARATIVE BIOGEOGRAPHY Integration of molecular phylogenetics with geological history is a powerful approach that permits estimation of lineage ages and polarity. Knowing absolute island ages, questions regarding temporal aspects of radiations such as rates of evolutionary diversification can be addressed; DNA sequences can provide high resolution, quantitative data relevant, for instance, to colonization patterns of many taxa including endemic terrestrial snails (Chiba 1999, Goodacre 2002, Holland and Hadfield 2002, Rundell etal. 2004, Holland and Cowie 2007). F-'iglire 1. Map of the main Hawaiian I.slaiuis showing subniarinc topography and selected channel depths. Dotted lines show now submerged, maximum historical shorelines for each volcano as it formed. The islands of Maui, Lanai, Molokai, and Kahoolawe are known as Maui Nui. These islands and Oahu formed a single super island as recently as I8,()()() years ago dur- ing the last glacial maximum. LAND SNAIL ISLAND BIOGEOGRAPHY 61 Molecular phylogeography addresses spatial patterns of evolutionary diversity within species or species complexes and has recently been expanded to include comparison of phylogeographic patterns of multiple co-distributed lineages, comparative phylogeography (Arbogast and Kenagy 2001). Such studies have revealed pervasive and unanticipated biogeographic patterns and suggest that biotic assemblages contain much greater cryptic biological diversity than tradi- tional systematics has recognized (e.g., Bird et al. 2007, Sha et al. 2007, Victoriano et al. 2008). In a comparative frame- work, phylogeography can be used to evaluate both historical patterns and evolutionary processes, providing a basis for new avenues of research into the regional historical, ecological, and coevolutionary factors generating and maintaining biodiversity. The well-documented geological history and known ages of the Hawaiian Islands allow examination of the timing of lineage splitting and species formation. The phylogenetic pattern characteristic of Hawaii’s more diverse endemic clades, especially nonvolant groups with limited active dispersal ability, is for the older, basal members of a clade to occur on the oldest island and for successively more recently derived members of the clade to occur on successively younger islands (Wagner and Funk 1995, Fleischer et al. 1998, Roderick and Gillespie 1998, Price and Clague 2002, Cowie and Holland 2006), a pattern termed the progression rule of island biogeography (Wagner and Funk 1995). Since land snails are generally considered poor active long-distance dispersers; they are predicted to adhere to the progression rule pattern on oceanic archipelagos. Native invertebrates, among them the endemic land snails, comprise the most species-rich faunal radiations in the Hawaiian Islands and offer valuable opportunities for the study of historical biogeography and evolutionary diversification of insular radiations in a comparative phylogenetic framework. HAWAIIAN LAND SNAILS AS BIOGEOGRAPHIC MODELS Hawaiian land snail phylogenies provide opportunities to investigate a number of general issues in evolutionary biology, including the relative roles of dispersal versus vicariance in the generation and maintenance of endemic lineages and the predictions of the “taxon cycle” (sensii Wilson 1961 ). Since the early 1980s, a debate has periodically flared up in the primary biogeographic literature regarding the importance of vicariance versus dispersal in shaping distribu- tions of plants and animals. Since most species have patchy distributions at some scale, and patchiness or allopatry has long been considered to play an important role in diversifi- cation (Mayr 1942), the mechanisms contributing to natural distribution patterns are fundamental to evolutionary biology and our understanding of biodiversity (see de Queiroz 2005, Cowie and Holland 2006). According to the predictions of Wilson’s (1961) taxon cycle, island lineages initially colonize marginal habitat, eventually dispersing and diversifying into higher elevation cloud forests, and ultimately go extinct as habitat changes and islands erode and subside. One of the key features of Wilson’s taxon cycle is the notion that oceanic islands serve as evolutionary blind alleys (Whittaker 1998) and, therefore, do not give rise to further long-distance colonization. Also, the diverse Hawaiian land snails provide natural systems with which to use molecular techniques to investigate many such patterns of island biogeography and lineage splitting (Holland and Hadfield 2002, 2004, Rundell et al. 2004, Cowie and Holland 2006, Holland and Cowie 2007) as well as to address more basic systematics issues (Holland and Hadfield 2007). HAWAIIAN SUCCINEIDAE Succineids occur worldwide (Pilsbry 1948, Patterson 1971), reaching their highest diversity in Pacific islands, India, and the Americas (Barker 2001). Though often associated with riparian areas {e.g., Kerney and Cameron 1979), succineids also occur in a range of different habitats (Barker 2001, Rundell et al. 2004). There are 42 recognized Hawaiian succineid species, all endemic, with 35 single-island endemics (Cowie et al. 1995), suggesting that dispersal between islands occurs, but is sufficiently rare to allow speciation. The Hawaiian Succineidae have radiated into a diverse array of habitats, from montane rainforests to xeric coastal dunes (Cowie 1995, Cowie et al. 1995). Their shell types appear to reflect these ecological differences, making them candidates for studies of adaptive radiation, convergence, and morphological evolution. We have been unable to find approximately half of the described species despite intensive field efforts focused on type localities and other suitable habitat, e.g., Catinella riibida Pease, 1870 from Kauai, the type species of the genus Catinella Pease, 1870. These species are either extremely rare or possibly extinct. Rundell et al. (2004) estimated that as many as one third of the Hawaiian succineid species may be extinct, but this may well be an underestimate. Phylogenetics and biogeography Although the family in the Hawaiian Islands was historically assumed to be monophyletic and the result of a single colonization (Zimmerman 1948), Rundell et al. (2004) showed, based on mtDNA evidence, that this is not the case. Here we present a global, multi-locus data set that permits this surprising biogeographic result to be investigated further. 62 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Hawaiian Islands 'T!22_r Hniiu Hawaii (3) Tahiti Hawaii (2) -T~ Oahu Nursery Thailand Ml Aorai Tahiti (2) Orofero Tahiti Pacific Rim and Islands Hawaiian Islands I Island of Hawaii Invasive Clade B French Polynesia Costa Rica, Oahu invasive Saipan (3) I , ' Tampa FL Invasive Northern Marianas, Melanesia • Tampa -^^^-[^Sarigan Island (3) I lOOp Papua New Guinea I Solomon Islands I C New Zealand (2) Santa Cruz Island (2) SW Australia I L Santiago Island (2) • San Cristobal Eastern/ Central Pacific 100 New Zealand, Australia, Galapagos I South Africa -ISendai (3) I Japan H^L(4) I Brazil lOoH United Kingdom (3) Michigan (2) Netherlands i00C-(2) ^(2) England, Netherlands, North America 100 loof lOgasawara, Japan South China 94 -[ (2) I South Africa ^t!^Hawai'i'(2) — Molokai I North America -C:'^ Maui (3) 0 ahu Samoa 100-^ -Kauai 100 100 H Oahu (10) Kauai Kauai Oahu Molokai Maui Lanai Hawaii Clade A _New Caledonia (2) — Oahu — 0.005 substitutions/site I Lymnaeidae outgroups Rimatara Australs (2) Ua Huka. Marquesas (2) Rarotonga Cooks (2) South Pacific Islands Figure 2. Phylogenetic tree for the Succineidae, based on three genes, two mitochondrial and one nuclear (COI, 16S, 18S; 1740 base pairs), for 40 ingroup species and two outgroups. This area phylogram was generated with PAUP (Swofford 2002) using a maximum-likelihood ap- proach. Numbers near nodes represent bootstrap support based on 1000 replicates. Numbers in parentheses show numbers of specimens seciuenced per lineage. The shaded boxes highlight the two Hawaiian lineages. A combined multi-locus phylogeny of 40 ingroup species indicates that the Hawaiian succineids fall into two monophyletic groups (Fig. 2). Clade A (Fig. 2) includes succineids from Kauai, Oahu, Molokai, Maui, Lanai, and Hawaii, with a species from Samoa nested within this clade, sister to a species from Kauai. The general patterns of relationships within this clade provide support, though weakly, for the progression rule of successive colonization from older islands (Kauai, 5.1 Ma, Oahu, 3.7 Ma) to younger islands (West Maui, 1.3 Ma, Hawaii, <0.5 Ma), with species from Kauai and Oahu associated with the basal nodes of the clade. Basal to this clade is a species from the islands of the South Pacific ( Samoan, Marc]uesas, Cook, and Austral Islands), indicating an ancient South Pacific origin. Clade B includes species from the island of Hawaii plus Succinea caduca Mighels, 1845, which is the one Hawaiian succineid that occurs on all six main islands (Holland and Cowie 2007), with a species from Tahiti nested within the clade, sister to a species from the island of Hawaii. Basal to Clade B is a species from Thailand (that is also found as a modern invasive species in Hawaii), suggesting a southeast Asian origin. Clade A is the older of the Hawaiian succineid lineages. The en- demic Hawaiian succineids are not monophyletic and the two main Hawaiian lineages differ in their geo- graphic origin and inferred relative time of colonization. An Asian or Australasian origin has been thought likely for a number of other Pacific island land snail groups (Cowie 1996, Pokryszko 1997). The phylogenetic analyses presented here, with I'hai species basal to Clade B ( Fig. 2), are consistent with this suggestion. The positions of a Samoan species in Clade A and a Tahitian species in Clade B suggest that Samoa was colonized from Kauai and Tahiti was colonized from Hawaii. The.se may be the first recognized ca.ses of natural colonizations of other locations by endemic Hawaiian animal species, contrasting, since they are both rainforest not marginal habitat species, with predictions of Wilson’s ( 1961 ) “taxon cycle”, which considers endemic island lineages to be es.sentially evolutionary blind alleys. The absolute ages of the Hawaiian succineid lineages are uncertain but theoretically an ancestral succineid could have arrived in the Hawaiian archipelago as early as about 29 Ma ago. This is the age of Kure, the adjacent older islands in the chain having already vanished below the ocean surface, and islands have been consistently present above sea level since then (Carson and Claguc 1995,Clague 1996). 1 lowever, many groups of Hawaiian plants and animals exist exclusively at high elevation. TTierefore, since there were probably periods of hundreds of thousands of years when, as older islands subsided and eroded and newer islands were as yet low, only low elevation habitats wereavailable. Absence of high elevation habitat may thus preclude an ancient origin {i.c., 29 Ma) (Price and Clague 2002). In particular, such a scenario may LAND SNAIL ISLAND BIOGEOGRAPHY 63 have happened when Kauai, the oldest, high island, emerged because Nihoa (the adjacent older island) had already declined greatly in elevation (Price and Clague 2002). Nevertheless, some modern Hawaiian succineids inhabit coastal dune lands and low-elevation forests, indicating that their ancestors could also have thrived in similar environments; land of sufficient, if low, suitable elevation was indeed consistently available from 29 Ma ago to the present. Archaeological and paleontological evidence suggests that succineids, amastrids, and other endemic Hawaiian land snail species probably once had ranges that extended into low elevation dry forest and arid habitats ( Christensen and Kirch 1986). Phylogenetic studies of other Hawaiian invertebrate radiations suggest a few origins far exceeding the age of Kauai (Russo etal. 1995, Jordan etal. 2003). The island of Hawaii is less than 0.5 Ma old, and 22 (19 endemic) of the 42 recognized Hawaiian succineid species are from this island (Cowie et al. 1995). Additional work on the evolution of succineid species on the island of Hawaii, such as application of molecular clock theory to robust phylogenies consisting of additional nuclear genes and the possible inclusion of extinct species from museum collections, may provide more insights into the complexity and rate of evolutionary change. Single species phylogeography: Succinea caduca Although the Hawaiian succineids are primarily single- island endemics, Succinea caduca Mighels, 1845 occurs on all six major high islands (Holland and Cowie 2006). Holland and Cowie (2007) used mtDNA COI sequences to evaluate geographic patterns of variation in S. caduca, sampling 24 populations on the six islands. Six clades based on 276 COI sequences were revealed, indicating substantial geographic genetic structuring. Low nucleotide diversity and low pairwise molecular-divergence values within populations coupled with higher between-population values suggested multiple founder events. High overall haplotype diversity suggested diversification involving rare dispersal, fragmentation by historical lava flows, and variation in habitat structure. Within- island rather than between-island population comparisons accounted for the majority of molecular variance. Population partitioning patterns suggested that genetic fragmentation has been driven by punctuated, passive dispersal of groups of closely related haplotypes that subsequently expanded and persisted in isolation for long periods (average >2 Ma), and that inter- island dispersal has led to population fragmentation. However, Mantel tests for isolation by distance, statistical correlation of geographic and genetic distance, were not significant. Thus, rather than a widespread panmictic distribution that was subsequently divided into its present day distribution, we see evidence that dispersal has led to a chaotic pattern of genetic partitioning. Historical availability of mesic coastal habitat, together with effective dispersal, may explain the long-term persistence and unusual multi-island distribution of this species, which contrasts with the single island endemism of most Hawaiian succineids as well as much of the Hawaiian biota (Holland and Cowie 2007). Haplotype networks from the northwestern portion of the archipelago, representing the older islands of Kauai and Oahu that are separated by the Kauai Channel (Fig. 1) and have never been connected, are broken into 5 clades (Holland and Cowie 2007). One of these clades contains haplotypes from Kauai and Oahu, demonstrating unambiguously that dispersal across the Kauai Channel has occurred. Other haplotypes from Oahu and all those from Maui Nui (Maui, Molokai, Lanai) group in a single clade and provide support for vicariant separations and evidence that Pleistocene island connections may have been important in enhancing gene flow. Vicariant separation in this clade has not led to deep divergence, a reflection of the geologically recent separation of the Maui Nui component islands (Price and Elliott-Fisk 2004) , which conforms with the estimated most recent sea level minimum at the last glacial maximum 18,000 years ago, about 130 m below the current level. Haplotypes from the island of Hawaii cluster in a separate, single island clade. Thus, Succinea caduca provides evidence for both within (Kauai, Oahu) and between (Kauai-Oahu, Maui-Hawaii) island diversification, suggesting that a combination of vicariance and dispersal has led to the allopatric diversification of this species. HAWAIIAN ACHATINELLINAE With 99 recognized species, all single-island endemics, in four genera (Cowie et al. 1995, Holland and Hadfield 2004), the endemic Hawaiian tree snails (Achatinellinae) are a species-rich radiation. The richly colored, highly varied banding patterns of their shells captured the attention of early naturalists and shell collectors, many of whom collected thousands of snails during the late 1800s and early 1900s (Hadfield 1986). Historically, appreciation for Hawaiian tree snails helped to inspire and promote general awareness of the diverse and unique Hawaiian biota. In recent years, however, the Hawaiian tree snails have also gained scientific and regulatory attention because of their dire conservation status. Of the 41 species of Achatinella Swainson, 1828 recognized by the U.S. Fish and Wildlife Service, only 10 are now thought to survive and are listed as endangered, with range reductions approaching 90% (U.S. Fish and Wildlife Service 1993). All other achatinelline species are seriously threatened. Phylogenetics and biogeography The molecular phylogenetic pattern for the Hawaiian tree snails (Fig. 3; see also Thacker and Hadfield 2000, Holland and Hadfield 2004) contrasts markedly with the topology for Hawaiian succineids (Fig. 2). The extant species analyzed 64 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 form a monophyletic group of three main clades: the Oahu Clade (I) consisting only of five Achatinella species from Oahu; the Maui Nui Clade (2) with 13 species from three genera and four islands (Oahu, Molokai, Lanai, Maui); and the Mixed Clade (3) with five species from three genera and four islands (Oahu, Molokai, Maui, Hawaii). The outgroup {Auriculella sp., Achatinellidae: Auriculellinae) was selected based on Holland and Hadfield (2004). Achatinella mustelina Mighels, 1845, endemic to the oldest mountain range on the island, was basal to the other four members of the clade. Therefore, in reconstructing the biogeographic pattern and colonization sequence for the tree snails (Fig. 4), we begin with A. mustelina. As few as two of the twelve colonization events depicted in Lig. 4 may have involved dispersal over an ocean channel; the remainder could have taken place via past island connections (forested land bridges) and relatively recent vicariant separation. However, this is the minimum possible number of ocean channel crossings based on this analysis. The following genetic divergence values were generated using mtDNA COI sequence data, for which published divergence rates for invertebrates are typically in the range of 1 .4 to 2.2% per million years (Kn owl ton and Weigt 1998). The mean divergence among the 12 species of Parfn/mn Pfeiffer, 1854 (5.4%) was lower than that among the Achatinella species, all from a single island (6.1%), despite the Partulina species coming from what are presently four islands (Maui, Molokai, Lanai, Hawaii) (Lig. 1), suggesting that the Partulina and Perdicella species in the Maui Nui Clade are younger than Achatinella species. The mean genetic divergence among species of all three genera from the Maui Nui Clade was 5.4%, slightly lower than that for all members of the genus Partulina at 5.5%. Interestingly, the divergence between Partulina physa (Newcomb, 1854) from the island of Hawaii and the next closest relative, P tappaniana (Adams, 1851), was low (3.3%) in spite of substantial differences in shell morphology and separation of the islands by the Alenuihaha Channel. This relatively close relationship, spanning a deep marine feature, reflects historical dispersal across a marine barrier. Newcombia cumingi Partulina redfieldi Partulina tappaniana Partulina physa Acbatlnella fulgens Partulina perdix Partulina crocea Partulina splendida Partulina porcellana Perdicella Helena Partulina proxima Partulina sp. Partulina mighelsiana Partulina variabilis Partulina semicarinata Achaffnelfadectpwns Achatinefla IHa AchPpnbttafu&cobPais Achattnpllamueteilna Achatfpeilpabwerbyaba Achatinella iivida Achatinella apextuiva Achatinefla PPhcavospira Auriculella sp. Maui Maui Nui Hawaii Oahu Maui Nui Oahu main clades = O seeding’ lineages = ^ Figure 3. Maxinuim likelihood cladogram generated using I’AUP (SwofTord 2002) for the achatinelline tree snails reconstructed using COI gene sequences and 1000 bootstrap replicates. Some of the main features include monophyly of the subfamily and family (see i lolland and I ladfield 2004), progression rule pattern of hiogeography, and presence of seeding lineages from the oldest island in the group in each of the three main clades. LAND SNAIL ISLAND BIOGEOGRAPHY 65 Figure 4. Island colonization sequence inferred from molecular evi- dence for achatinelline tree snails, based on the phylogeny of Holland and Hadfield (2004). Map includes submarine historical shorelines. Solid arrows follow the progression rule and represent forward colo- nization events from older to younger geological features, numbered according the order of the nodes in the tree (Fig. 3) and starting from Achntinella mustelina. Dotted arrows show back colonizations, from newer to older geological features or islands. In a phylogenetic reconstruction that includes species representing all five achatinellid subfamilies, the Achatinellinae are monophyletic and well supported (R. H. Cowie et al, unpubl. data). Achatinellid sequences formed a monophyletic group with high bootstrap support (91-100%). Single species phylogeography: Achatinella mustelina Holland and Hadfield (2002) used mitochondrial DNA ( mtDNA) GOI sequences to evaluate phylogeographic structure within and among 21 populations of Achatinella mustelina {N = 78). In contrast to the multi-island distribution of Succinea caduca,A. mustelina has a relatively small distribution spanning a single mountain range on Oahu. Pairwise intraspe- cific mtDNA sequence divergence between haplotypes ranged from 0 to 5.3%, and population genetic partitioning and mountain topography were strongly correlated. Maximum genetic distances were observed across deep, largely deforested valleys and sheer mountain ridges, independent of geographic distance. However, in certain areas where forest cover is presently fragmented, there was little sequence divergence despite large geographic distances. Genetic data were used to define five evolutionarily significant units (ESUs) that will guide conservation decisions regarding, for instance, placement of predator-exclusion fences, captive propagation, and eventual re-introduction and translocation. To account for the observed phylogeographic pattern, an evolutionary scenario was proposed (Holland and Hadfield 2002) beginning with an initial panmictic phase, followed by long-term, large-scale habitat fragmentation, and finally recent fine-scale fragmentation resulting in the current patchy distribution. In the early geological history of the island, western Oahu consisted of a single massive shield volcano, similar in shape to the less than half-million year old volcanoes on the island of Hawaii. The modern topography of western Oahu is deeply eroded and rugged, with complex features resulting from action of wind and rain over its 3.7 Ma history. It is possible that large populations of tree snails gradually became separated from one another by vicariant forces resulting from the formation of valleys and ridges, and isolated by distance into the present pattern of “islands” of genetically cohesive populations. In support of this idea, a Mantel test (Holland and Hadfield 2002) demonstrated a correlation between genetic and geographic distance (in contrast to Succinea caduca). Achatinella mustelina has a far more restricted and genetically structured distribution than S. caduca, indicating relatively limited dispersal ability, possibly related to the former s much larger and heavier shell. CONCLUSIONS The molecular studies of Hawaiian land snails reviewed here strongly support the current shift in perception in historical biogeography from vicariance as the dominant process to more of a balance between dispersal and vicariance driving allopatric diversification. In the Hawaiian Succineidae, molecular data reveal two independent colonization events from two different, distant geographic origins. Once the tvs^o lineages became established and began to radiate, each acted as a colonization source for long-distance dispersal to two different South Pacific island groups. Within one Hawaiian succineid lineage, dispersal across marine barriers occurred multiple times, resulting in one species {Succinea caduca) that occurs on all six major high islands. Closely related haplotypes that span a deep permanent feature such as the Kauai Channel attest to the dispersal ability of this species. Eor the Hawaiian tree snails dispersal has also helped to shape their natural history in important ways. Sister species that span the Alenuihaha Channel between the relatively young islands of Maui and Hawaii attest to recent across- ocean dispersal. However, the fact that all 99 species are single- island endemics indicates that dispersal of these sedentary, heavy-shelled tree snails, while it certainly has occurred, is rare. Patterns of molecular diversification in the achatinellines indicate an Oahu origin, progression rule pattern, isolation by distance, and at least two channel-crossing dispersal events. 66 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 In contrast to the deep separation of two distinct lineages in the succineids, the achatinellines form a monophyletic clade (Holland and Hadfield 2004). However, preliminary molecular data indicate that the family Achatinellidae, which consists of five subfamilies (Cowie et al. 1995), two of which are endemic to Hawaii, colonized the Hawaiian Islands via four independent events (Holland and Cowie, unpubl. data). Based on the well understood geological history of the Hawaiian Islands and the use of molecular phylogenetics, it has been possible to reconstruct the evolutionary and biogeographical history of two major groups of land snails, including the timing of lineage splitting and direction of colonization, while at the same time determining their geographic origins and whether presumptive radiations comprise monophyletic groups. Because of the relatively sedentary nature of land snails, yet the proclivity of some for long-distance passive dispersal, these groups have facilitated a major insight into the mechanisms via which evolutionary diversification takes place. Important phylogenetic lessons emerging from these studies of island snails include: ( 1 ) assumptions of monophyly are not always met; (2) in multi- island lineages composed of single island endemic species, monophyletic clades may include species from different islands; and (3) although some lineages conform to the isola- tion by distance and progression rule patterns of biogeography, others do not. Studies of island land snails will eventually lead to estimations of ages of species groups, speciation rates, timing of the processes involved in community assembly, and other dynamics, all of which are important contributions to the overall understanding of mechanisms of evolution. ACKNOWLEDGMENTS We thank Matthias Glaubrecht and Thomas von Rintelen for inviting us to participate in the symposium they organized and for the invitation to write this contribution. Our work on Succineidae was supported by NSF grant DEB 0316308. Molecular phylogeographic analysis of Hawaiian tree snails was carried out in the laboratory of Michael G. Hadfield, to whom we are grateful. LITERATURE CITED Anonymous. 1936. Succiiica carried by a bird. Nauliliis 50: 3 1 . Arbogast, B. S. and G. ). Kcnagy. 2001. ('.omparative pbylogeograpby as an integrative approach to historical biogeography, louniul of liiogco^riiphy 28: 8 1 9-825. Avise, |. (7 2000. I’hyloyeoyrophy: The lihtory and Ivniuilioii of Spe- cies. Harvard University Press, ( iambritlge, Massacluisells. Barker, G. M. 2001. Gastropods on land. In: G. M. Barker, ed.. The Biology of Terrestrial Molluscs. CABI Publishing, Wallingford, U.K. Pp. 1-146. Barker, G. M. and P. C. Mayhill. 1999. Patterns of diversity and habitat relationships in terrestrial mollusc communities of the Pukeamaru Ecological District, northeastern New Zealand. Journal of Biogeography 26: 215-238. Bird, C. E., B. S. Holland, B. Bowen, and R. Toonen. 2007. Contrast- ing phylogeography in three endemic Hawaiian limpets (Cel- lana spp.) with similar life histories. Molecular Ecology 16: 3173-3186. Cain, A. ]. 1984. Islands and evolution: Theory and opinion in Darwin’s earlier years. Biological Journal of the Linnean Society 21: 5-27. Cameron, R. A. D., R. M. T. Da Cunha, and A. M. Erias Martins. 2007. Chance and necessity: Land-snail faunas of Sao Miguel, Azores, compared with those of Madeira. Journal of Molluscan Studies 73: 11-21. Carson, H. L. 1987. Colonization and speciation. In: A. I. Gray, M. I. Crawley, and P. J. Edwards, eds.. Colonization, Succession and Stability. Blackwell, Oxford. Pp. 187-205. Carson, H. L. and D. A. Clague 1995. Geology and biogeography of the Hawaiian Islands. In: W. L. Wagner and V. L. Punk, eds., Ha- waiian Biogeography. Smithsonian Institution Press, Washing- ton, D.C. Pp. 14-29. Chiba, S. 1 999. Accelerated evolution of land snails Mandarina in the oceanic Bonin islands: Evidence from mitochondrial sequenc- es. Evolution 53: 460-471. Christensen, C. C. and P. V. Kirch, 1986. Nonmarine mollusks and ecological change at Barber’s Point, Oahu, Hawaii. Bishop Mu- seuni Occasional Papers 26: 52-80. Clague, D. A. 1996. The growth and subsidence of the Hawaiian- Emperor volcanic chain. In: A. Keast and S. E. Miller, eds.. The Origin and Evolution of Pacific Islands Biotas, New Guinea to Eastern Polynesia: Patterns and Processes. SPB Academic Pub- lishing, Amsterdam. Pp. 35-50. Cowie, R. H. 1995. Variation in species diversity and shell shape in Hawaiian land snails: In situ speciation and ecological relation- ships. Evolution 49: 1191-1 202. Cowie, R. H. 1996. Pacific island land snails: Relationships, origins, and determinants of diversity. In: A. Keast and S. E. Miller, eds., The Origin and Evolution of Pacific Island Biotas, New Guinea to Eastern Polynesia: Patterns and Processes. SPB Academic Pub- lishing, Amsterdam. Pp. 347-372. Cowie, R. H. 2001. Invertebrate invasions on Pacific islands and the replacement of unique native launas: A synthesis of the land and freshwater snails. Biological hivasions i: 1 19-136. Cowie, R. 1 1. 2004. Disappearing snails and alien invasions: 'I'he bio- diversity/conservation interlace in the Pacific. Journal of Gon- chology Special Publications 3: 23-37. Cowie, R. 1 1. and B. S. I lolland. 2006. Dispersal is fundamental to evo- lution on oceanic islands. lournal of Biogeography 193-200. Cowie, R. 11., N. L. Evenhuis, and C. Ci. Christensen. 1995. Catalog of the Native Land and L'reshwater Molluscs of the Ilawiuian Is- lanils. Backhuys Publishers, Leiden, The Netherlands. I )arwin, C. 1 859. ( )n the ( trigin of Species by Means of Natural Selec- tion. Garamond Pre.ss, Aurora, Ontario. LAND SNAIL ISLAND BIOGEOGRAPHY 67 de Queiroz, A. 2005. The resurrection of oceanic dispersal. Trends in Ecology and Evolution 20: 68-73. Dundee, D. S., P. H. Phillips, and ]. D. Newsom. 1967. Snails on mi- gratory birds. The Nautilus 80: 89-9 1 . Eldredge, L. G. and N. L. Evenhuis. 2003 Hawaii’s biodiversity: A de- tailed assessment of the numbers of species in the Hawaiian Islands. Bishop Museum Occasional Papers 76: 1-28. Fleischer, R. C., C. E. McIntosh, and C. L Tarr. 1998. Evolution on a volcanic conveyor belt: Using phylogeographic reconstructions and K-Ar-based ages of the Hawaiian Islands to estimate mo- lecular evolutionary rates. Molecular Ecology 7: 533-545. Fontaine, B., O. Gargominy, and E. Neubert. 2007. Priority sites for conservation of land snails in Gabon: Testing the umbrella spe- cies concept. Diversity and Distributions 13: 725-734. Gillespie, R. G., H. B. Groom, and S. R. Palumbi. 1994. Multiple ori- gins of a spider radiation in Hawaii. Proceedings of the National Academy of Sciences of the U.S.A. 91: 2290-2294. Gittenberger E., D. S. J. Groenberg, B. Kokshoom, and R. G. Preece. 2006. Molecular trails from hitch-hiking snails. Nature 439: 409. Goodacre, S. L. 2002. Population structure, history and gene flow in a group of closely related land snails: Genetic variation in Partula from the Society Islands of the Pacific. Molecular Ecol- ogy I 55-68. Grant, P. R. 1986. Ecology and Evolution of Darwin’s Finches. Prince- ton University Press, Princeton, New Jersey. Gulick, J. T. 1905. Evolution, racial and habitudinal. Carnegie Institu- tion of Washington Publication 25: 1-269. Hadfield, M. G. 1986. Extinction in Hawaiian achatinelline snails. Malacologia 27: 67-81. Holland, B. S. and R. H. Gowie. 2006. New island records of an en- demic Hawaiian land snail species, Succinea caduca Mighels, 1845 (Gastropoda: Pulmonata: Succineidae). Bishop Museum Occasional Papers 88: 58-60. Holland, B. S. and R. H. Gowie. 2007. A geographic mosaic of pas- sive dispersal: Population structure in the endemic Hawaiian amber snail Succinea caduca (Mighels, 1845). Molecular Ecology 16: 2422-2435. Holland, B. S. and M. G. Hadfield. 2002. Islands within an island: Phylogeography and conservation genetics of the endangered Hawaiian tree snail Achatinella mustelina. Molecular Ecology 11: 365-375. Holland, B. S. and M. G. Hadfield. 2004. Origin and diversification of the endemic Hawaiian tree snails (Achatinellinae: Achatinel- lidae) based on molecular evidence. Molecular Phylogenetics and Evolution 32: 588-600. Holland, B. S. and M. G. Hadfield. 2007. Molecular systematics of the endangered Oahu tree snail Achatinella mustelina: Synony- mization of subspecies and estimation of gene flow between chiral morphs. Pacific Science 61: 53-66. Hormiga, G., M. Arnedo, and R. G. Gillespie. 2003. Speciation on a conveyor belt: Sequential colonization of the Hawaiian Islands by Orsonwelles spiders (Araneae, Linyphiidae). Systematic Biol- ogy 52: 70-88. Jordan, S., C. Simon, and D. Polhemus. 2003. Molecular systematics and adaptive radiation of Hawaii’s endemic damselfly genus Megalagrion (Odonata: Coenagrionidae). Systematic Biology 52: 89-109. Kerney, M. P. and R. A. D. Gameron. 1979. A Field Guide to the Land Snails of Britain and North-west Europe. Gollins, London. Knowlton, N. and L. A. Weigt. 1998. New dates and new rates for divergence across the Isthmus of Panama. Proceedings of the Royal Society of London (B) 265: 2257-2263. Lydeard, C., R. H. Gowie, W. F. Ponder, A. E. Bogan, P. Bouchet, S. Clark, K. S. Cummings, T. J. Frest, O. Gargominy, D. G. Herbert, R. Hershler, K. Perez, B. Roth, M. Seddon, E. E. Strong, and F. G. Thompson. 2004. The global decline of nonmarine mollusks. BioScience 54: 321-330. Mayr, E. 1942. Systematics and the Origin of Species. Columbia Uni- versity Press, New York. Patterson, C. M. 1971. Taxonomic studies of the land snail family Succineidae. Malacological Review 4: 131-202. Pilsbry, H. A. 1948. Land Mollusca of North America (north of Mexico), Vol. II, Part 2. Academy of Natural Sciences of Phila- delphia Monographs 3: 771-847. Pilsbry, H. A. and C. M. Cooke, Jr. 1912-1914. Manual of Conchol- ogy. Structural and Systematic. With illustrations of the Species. Second Series: Pulmonata. Vol. XXII. Achatinellidae. Academy of Natural Sciences, Philadelphia. Pimm, S. L., M. P. Moulton, and L. J. Justice. 1994. Bird extinctions in the central Pacific. Philosophical Transactions of the Royal So- ciety of London (B) 344: 27-33. Pokryszko, B. M. 1997. Lyropupa Pilsbry, 1900. Systematics, evolu- tion and dispersal (Gastropoda: Pulmonata: Pupilloidea). Ge- nus S: 377-487. Price, J. P. and D. A. Clague. 2002. How old is the Hawaiian biota? Geology and phylogeny suggest recent divergence. Proceedings of the Royal Society London (B) 269: 2429-2435. Price, J. P. and D. Elliott-Fisk. 2004. Topographic history of the Maui Nui complex, Hawaii, and its implications for biogeography. Pacific Science 58: 27-45. Rees, W. J. 1965. The aerial dispersal of Mollusca Proceedings of the Malacological Society of London 36: 269-282. Robinson, G. S. and K. Sattler. 2001. Plutella in the Hawaiian Islands: Relatives and host-races of the diamondback moth (Lepidoptera: Plutellidae). Bishop Musetim Occasional Papers 67: 1-27. Roderick, G. K. and Gillespie, R. G. 1998. Speciation and phylogeog- raphy of Hawaiian terrestrial arthropods. Molecular Ecology 7: 519-531. Rundell, R. J., B. S. Holland, and R. H. Gowie. 2004. Molecular phy- logeny and biogeography of the endemic Hawaiian Succineidae (Gastropoda: Pulmonata). Molecular Phylogenetics and Evolu- tion 31: 246-255. Russo, C. A. M., N. Takezaki, and M. Nei. 1995. Molecular phylogeny and divergence times of drosophilid species. Molecular Biology and Evolution 12: 391-404. Sha, Z.-L., G.-D. Zhu, R. W. Murphy, and D.-W. Huang. 2007. Dig- lyphus isaea (Hymenoptera: Eulophidae): A probable complex of cryptic species that forms an important biological control agent of agromyzid leaf miners. Journal of Zoological Systemat- ics and Evolutionary Research 45: 128-135. 68 AMERICAN MALACOLOGICAL BULLETIN 27 -1/2 -2009 Simon, C. 1987. Hawaiian evolutionary biology: An introduction. Trends in Ecology and Evolution 2: 175-178. Solem, A. 1990. How many Hawaiian land snail species are left? and what we can do for them. Bishop Museum Occasional Papers 30: 27-40. Staples, G. W. and R. H. Cowie. 2001. Hawaii’s Invasive Species. Mu- tual Publishing and Bishop Museum Press, Honolulu. Swofford, D. L. 2002. PAUP*. Phylogenetic Analysis Using Parsimony i*and Other Methods). Version 4.0bl0. Sinauer Associates, Sun- derland, Massachusetts. Thacker, R. W. and M. G. Hadfield. 2000. Mitochondrial phylogeny of extant Hawaiian tree snails ( Achatinellinae). Molecular Phy- logenetics and Evolution 16: 263-270. U. S. Fish and Wildlife Service. 1 993. Recovery Plan. 0‘ahu Tree Snails of the Genus Achatinella. U. S. Department of the Interior, U. S. Fish and Wildlife Service, Portland, Oregon. Vagvolgyi, 1. 1975. Body size, aerial dispersal and origin of the Pacific land snail fauna. Systematic Zoology 24: 465-488. Victoriano, P. F., ). G. Ortiz, E. Benavides, B. I. Adams, and I. W. Sites, Jr. 2008. Comparative phylogeography of codistributed species of Chilean Liolaemus (Squamata: Tropiduridae) from the central-southern Andean Range. Molecular Ecology 17: 2397-2416. Vitousek, P. M. 1988. Diversity and Biological Invasions of Oceanic Islands. National Academy Press, Washington, D.C. Wagner, W. L. and V. A. Funk. 1995. Hawaiian Biogeography. Smith- sonian Institution Press, Washington, D.C. Wagner, W. L., D. R. Herbst, and S. H. Sohmer. 1999 Manual of the Elowering Plants of Hawai'i. University of Hawaii Press, Bishop Museum Press, Honolulu. Whittaker, R. J. 1998. Island Biogeography. Oxford University Press, Oxford. Wilson, E. O. 1961. The nature of the taxon cycle in the Melanesian ant Emrux. American Naturalist 95: 169-193. Ziegler, A. C. 2002. Hawaiian Natural History, Ecology, and Evolution. University of Hawaii Press, Honolulu. Zimmerman, E. C. 1948. Introduction. Insects of Hawaii 1: 1-206. Submitted: 9 November 2008; accepted: 23 January 2009; final revisions received: 16 March 2009 Amer. Malac. Bull. 27: 69-81 (2009) Molecular phylogeny, taxonomy, and evolution of the land snail genus Pyrenaearia (Gastropoda, Helicoidea)"^ M. Arantzazu Elejalde\ M. Jose Madeira\ Carlos E. Prieto^, Thierry Backeljau^ '*, and Benjamin J. Gomez-Moliner* ' Departamento Zoologia, Facultad de Farmacia, Universidad del Pais Vasco, c/ Paseo de la Universidad 7, 01006 Vitoria (Alava), Spain ^ Departamento Zoologia, Facultad Ciencia y Tecnologia, Universidad Pais Vasco, Barrio Sarriena s/n 48940 Leioa, Spain ^ Royal Belgian Institute of Natural Sciences, Vautierstraat 29, B-1000 Brussels, Belgium Evolutionary Biology Group, Department of Biology, University of Antwerp, Groenenborgerlaan 171, B-2020, Antwerp, Belgium Corresponding author: benjamin.gomez@ehu.es Abstract: We reconstructed the molecular phylogeny of 78 specimens, including nearly all nominal species of the genus Pyrenaearia, and discussed the implications of the molecular phylogeny for species delimitation. The four basal clades obtained by mitochondrial COI and 16S sequences were highly congruent with nuclear ITS markers, shell morphology, and geographic distribution patterns, and were considered different species under the phylogenetic species concept: Pyrenaearia carascalopsis (Bourguignat in Fagot, 1884), P. carascalensis (Ferussac, 1821), P. parva Ortiz de Zarate, 1956, and P cantabrica (Hidalgo, 1873). Evidence of reproductive isolation between coexisting species of these basal clades has also been obtained. Because of genetic divergence, together with peculiarities of habitat use, distributional range, and shell morphology, P. organiaca (Fagot, 1905) and P. navasi (Fagot, 1907) were regarded as valid species. Further subdivisions within Pyrenaearia were also considered. Some incongruencies were found between shell morphology and DNA-based taxonomy, including the presence of hairs in adult shells and shell shape adaptations to different altitudes. The deepest split in Pyrenaearia involves four different lineages and represents an ancient event, occurring during the Pleistocene or even the Pliocene. Subsequent Pyrenaearia speciation within the four main groups has been a very recent process, and it is hypothesized to have occurred during Pleistocene cycles of climatic cooling and warming. Chronologically, a third speciation process is occurring from the Wurm de-glaciation to the present time. All these processes included allopatric speciation of the high-altitude taxa during warmer interglacial periods, but parapatric or peripatric speciation events could also be involved. Based on the present day distribution of the species and their different altitudinal preferences, the whole genus could constitute a good model to investigate the evolutionary processes that created this diversity. It could also be an excellent group of closely related taxa for the study of the effect of Plio- Pleistocene climate changes on species distribution, population structure, speciation processes, secondary contacts, and passive dispersal. Key words: DNA, systematics, distribution, land snails The genus Pyrenaearia Hesse, 1921 is endemic in the north Iberian Peninsula and southern France, where it occurs in the Cantabrian Mountains and the Pyrenees (Fig. 1). Species of this genus generally live at high altitudes in open limestone habitats and occupy very restricted geographical areas (Altonaga et al. 1994, Puente 1994). They usually occur on rock walls (Ortiz de Zarate 1956), but above 2000 m they live predominantly under stones (Gomez-Moliner, pers. obs.). Based on distributional patterns, the 16 currently recognized nominal species in the genus can be divided in six species groups. (1) Six morphospecies are endemic in the Cantabrian Mountains, where Pyrenaearia cantabrica (Hidalgo, 1873) has a relatively wide distribution, while P covadongae Ortiz de Zarate, 1956, P. daanidentata Raven, 1988, P. oberthueri (Ancey, 1884), P. poncebensis Ortiz de Zarate, 1956, and P. schaufiissi (Kobelt in Rossmassler, 1876) are confined to the Picos de Europa Massif. (2) Pyrenaearia velascoi (Hidalgo, 1867) is restricted to the Basque Mountains. (3) Another five morphospecies are endemic in the central and western Pyrenees with P. carascalensis (Ferussac, 1821) showing the widest distribution, whereas P. carascalopsis (Bourguignat in Fagot, 1884), P. cotiellae (Fagot, 1906), and P. esserana (Bourguignat in Fagot, 1888) are limited to small areas in the central Pyrenees. Pyrenaearia transfuga (Fagot 1885) is even restricted to a single locality (Escot) in the Aspe valley at the northern slope of the western Pyrenees. (4) Two species are confined to the eastern pre-Pyrenees: P. organiaca (Pagot, 1905) and P. parva Ortiz de Zarate, 1956. (5) Pyrenaearia molae Haas, 1924 lives in the mountains of Tarragona (Catalonia). (6) Finally, P. navasi (Fagot, 1907) is From the symposium “Molluscs as models in evolutionary biology: from local speciation to global radiation” presented at the World Congress of Malacology, held from 15 to 20 July 2007 in Antwerp, Belgium. 69 70 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 6^ ofSP' OH TM W Ot- TM UM XP YP BJ P XM VM bG DH _ '(JH rq. DG- EG Acladei Ac LADE 2 O CL ADE 3a OCLADE3b ■1CLADE3C □ CLADE3d □ CLADE 3e ■ CL ADE 3f O CLADE 4 Figure 1. Map showing collection sites of the specimens used in this study. The small box at the lower left corner indicates the distribution of the genus Pyrenaearia in black. the only species living south of the Ebro river valley, where it is confined to Mount Moncayo. Nevertheless, Prieto (1991) recognized two subspecies within P. navasv. P. n. navasi and P n. sylvatica. Most Pyrenaearia spp. are considered as sub-alpine organisms (Ortiz de Zarate 1956, Prieto 1986, Puente 1994), including Pyrenaearia daanidentata, P oberthneri, P. velascoi, P. carascalensis,P carascalopsis, P cotielIae,P esserana, P parva, and P. navasi navasi. These cold-adapted species have allopatric distributions, with often strongly isolated populations inhabiting the highest mountains, while other taxa are con- fined to the valleys (P. covadongae, P. poncebensis, P schaufusi, P. transfuga, P. organiaca and P. navasi sylvatica). Only P. can- tabrica shows a wide altitudinal range from 200 m to 2600 m above sea level. The taxonomic validity of the 16 nominal Pyrenaearia species and subspecies is not clear. Most species show no diagnostic differences in their reproductive apparatus although characters such as body color and relative sizes of structures in the reproductive system have been used to distinguish a few taxa (Ortiz de Zarate 1956, Puente 1994). Thus, taxon identification in Pyrenaearia is based mostly on shell characters, including shell size, shape and color, peris- tome morphology, and the presence of hairs on the shell surface (Ortiz de Zarate 1956, Puente 1994). However, shell morphology can be protoimdly influenced by environmental cues, including altitude (Raven 1988), and may be subject to parallel or convergent evolution, resulting sometimes in arbitrary classifications and incorrect species determination (e.g., Giusti and Manganelli 1992, Uit de Weerd el al. 2004). Delimiting species and reconstructing their phyloge- netic relationships are the two major goals of systematics (Wiens and Penkrot 2002). In this context, recent developments in DNA- based taxonomy aimed at defining species using the evolutionary and phylogenetic species concepts (Simpson 1961, Wiley 1978, Cracraft 1983, Coyne and Orr 2004) are extremely useful in validating morphospecies in several animal groups (Vogler and Monaghan 2007). Molecular genetic tools have also become very popular for reconstructing phylogenetic relationships in animal species (Avise 2000) and for identifying speciation processes, analyzing popu- lation structure, and inferring phylo- geographic patterns. The present work is the first molec- ular phylogenetic study of the genus Pyrenaearia. We used DNA sequence data to (1) assess the validity of the 16 nominal Pyrenaearia species under the morphological, phylo- genetic, and biological species concepts, (2) reconstruct their phylogenetic relationships, and (3) evaluate the geographic limits of the different lineages and taxa. Based on this infor- mation we develop some tentative speciation scenarios in relation to the Plio-Pleistocene climatic changes. MATERIALS AND METHODS Seventy-eight specimens of Pyrenaearia were included in this study with Hygroinia limbata (Draparnaud, 1805) and Iberus gualtieranus (Linnaeus, 1758) as outgroups. Locality and sample data are provided (Fig. 1, Appendix 1 ). Whenever possible, topotypes were included in the analysis. Freshly collected animals were kept frozen at -20 °C until analysis. Collection specimens were preserved in 70% ethanol. Individual DNA was extracted from foot muscle (15 to 20 mg), using the DNAeasy Tissue kit (QIAGEN). In some cases tissue was removed, squashed, and homogenized in sucrose buffer in order to eliminate excess mucopolysac- charides (Moorsel et al. 2000). Polymerase chain reaction (PGR) was used to amplify two mitochondrial gene fragments, cytochrome oxidase subunit 1 (COD and I6S rRNA (16.8), and one complete nuclear gene, the ribosomal internal transcribed spacer 1 (ITS-I)." 4’he COl fragment was initially amplified using the primers described by Folmer et al. (1994) ll.GC) 1490 (5'-GGTGAAGAAATGATAAAGATAfTGG-3'), 1 ICO 2198 (5'-TAAACn'GAGGGTGAGt:AAAAATGA-3')|. The ampli- cons were sequenced for various Pyrenaearia species and a MOLECULAR PHYLOGENY OF PYRENAEARIA 71 new specific forward primer was designed for this study to amplify a shorter fragment of about 500 base pair (bp) [COIPIF (5'-TAATGTTGTAGTTACTGC-3')]. The COIPIF primer was used together with the HCO 2198 reverse primer of Folmer et al. (1994). The 16S fragment was amplified by the primers of Palumbi et al. (1991) [16SarL (5'-CGCCTGTTTATCAAAAACAT-3') and 16SbrH (5'-CC- GGTCTGAACTCAGATCACGT-3')] . The ITS-1 marker was amplified using the primers ITSIL (5'-TCCGTAGG- TGAACCTGCGGAAGGAT-3') and 58C (5'-TGCGTTCAA- GATATGGATGTTCAA-3') (Idillis and Dixon 1991). Amphcons were sequenced using the dRhodamine Terminator Cycle Sequencing Ready Reaction Kit (Applied Biosystems) run on an ABI PRISM Model 3100 Avant Genetic Analyzer. The sequences were deposited in GenBank (accession numbers in Appendix 1). All sequences were aligned with Clustal X version 1.8 (Thompson et al 1997) with default settings and they were refined manually where necessary. Phylogenetic analyses using PAUP^^ 4.0b lO(PPC) (Swofford 2002) were performed for the three DNA fragments independently as well as for the combined data set. Gaps were treated as missing data. The best model of sequence evolution was selected using the Akaike information criterion (AIC) implemented in Modeltest v3.06 (Posada and Crandall 1998). A heuristic search was performed for the maximum parsimony (MP) analyses, with ten random-addition replicates, using the tree bisection reconnection (TBR) option generating multiples trees to determine the most parsimonious solution. Weights of transversions (Tv) and transitions (Ts) varied depending on the fragment and were estimated by maximum likelihood. The weighting scheme was 4:1 for COI and 1:1 for 16S and ITS-1. Neighbor-joining (Nj) (Saitou and Nei 1987) trees were constructed using the GTR-l-I-t-G model (Rodriguez et al. 1990) for all data matrices, according to Modeltest results. Uncorrected pairwise “p” distances were also calculated for the four data sets (see Table 1). For both the MP and NI analyses, bootstrap (BS) confidence estimates were based on 1000 replicates (Felsenstein 1985). Any BS values higher that 70% were considered as providing strong support (Hillis and Bull 1993). Bayesian analyses (BA) were performed using the MrBayes v3.0 package (Huelsenbeck and Ronquist 2001 ). The GTR model was used for the four data matrices, and rate variation across sites was modeled using a gamma distribu- tion for COI data set and equal distribution for 16S, ITS-1, and combined data sets, with a proportion of the sites being variants. The Markov Chain Monte Carlo (MCMC) search was run with four chains for 2 million generations, with trees being sampled every 100 generations (the first 2000 trees were discarded as “burn-in”). In the combined analyses, variation was partitioned among genes. Table 1. Molecular diversity indices for the different clades of Pyre- naearia for COI (first line), 16SrRNA (second line), and ITS-1 (third line). Values of number of haplotypes (H), haplotype diversity (h), nucleotide diversity (k), and mean number of nucleotide differences (k) are reported. CLADE 1 CLADE 2 CLADE 3 CLADE 4 H 4 3 20 27 3 4 23 24 3 1 8 5 h 0.9 0.833 0.965 0.986 0.7 1.0 0.953 0.975 0.7 0 0.713 0.321 K 0.021 0.004 0.03 0.027 0.008 0.006 0.027 0.016 0.005 0 0.002 0.002 k 9.4 2.0 13.361 12.068 3.1 2.5 10.729 6.383 3.4 0 1.112 1.444 RESULTS Sequence characteristics and genetic divergences The number of mtDNA haplotypes (H), haplotype diver- sity (h), nucleotide diversity (Jt), and number of nucleotide differences (k) were calculated for each gene fragment and for the combined data set (Table 1). The aligned COI fragment comprised 443 base pairs (bp). Sequences were A:T rich (74.17%) with nucleotide composition of: T (43.72%), C (11.2%), A (30.45%), and G (14.63%). One hundred and thirteen (25.51%) positions were parsimony informative. Using the Drosophila mitochon- drial genetic code, 100 (68.03%) of the 147 amino acids had synonymous nucleotide substitutions and 47 (31.97%) had no substitutions. The aligned 16S fragment consisted of 408 bp. The se- quences were A:T rich (69.84%), with nucleotide compositions of: T (33.78%), C (12.54%), A (36.06%), and G (17.62%). Eighty-nine (21.81%) positions were parsimony informative. The aligned ITS-1 gene comprised 692 characters. The proportion of A:T (40.95%) was lower, with nucleotide compositions of: T (22.18%), C (27.12%), A (18.77%), and G (31.93%). Seventeen (2.46%) positions were parsimony informative. The combined data set of the COI, 1 6S, and ITS- 1 analyses consisted of 1543 bp, 1069 of which (66.69%) were invariable, while 284 (18.4%) were variable but parsimony uninformative and 230 (14.91%) were parsimony informative. Phylogenetic analyses The four NJ trees had nearly identical topologies (Figs. 2-4) and were also almost identical to the MP and BA 72 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 topologies (not shown). The p-distances based on COI were larger than those based on 16S. All haplotypes were grouped within four basal clades. Average p-distances between these four clades were higher than 8.5% and 4.5% for COI and I6S, respectively (Table 2). The mean values of the average pairwise p-distances between a basal clade and the other three (Table 2) were very similar for each mtDNA gene fragment (8.6- 9.8% for COI; 5.2-6.9% for I6S). ITS- 1 was nearly lOx more conservative than COI and I6S (Table 2) and also recovered four major groups, the p-distances between which amounted to a maximum of 1.1%. The four ITS-1 groups corresponded to mtDNA clades, with clade 3a joined together with the other clade 3 subgroups in a common polytomy. The combined data set of the three genes, COI, 16S, and ITS-1, yielded a well-supported topology (fig. 4). Basal nodes were not resolved but the monophyly of each of the four basal clades was strongly supported by bootstrap values (>95%), except for clades 3 and 4 in the MP analysis. The four basal clades joined the same populations in the COI, ITS-1, and the concatenated phylogenetic analyses (Ligs. 2-4). Clade 3a was recovered as an independent lineage in the 16S tree. These results suggest confidence in the phylogeny obtained for Pyreriaearia, since the separate and combined analyses of COI, 16S, and ITS-1 did not yield conflicting results. There- fore, we will use the combined phylogeny for further discussion. Clade 1 comprised the haplotypes of Pyrenaearia carascalopsis and P. esserana, with the former constituting a paraphyletic group. Clade 2 comprised the haplotypes of P. parva from two localities, including topotypes. Clade 3 grouped P carascalensis together with P. cotiellae, P molae, P navasi, P. organiaca, P transfuga and P. velascoi. Pyrenaearia organiaca was the basal group of clade 3 and showed the greatest divergence within it (mean values of the average p-distances of 7.7% and 6.9% for COI and 16S, respectively) (Table 3). Pyrenaearia navasi was the sister group of clades 3c-f. The mean values of the average p-distances among clade 3 subgroups (3b-f) were below 5.0%, 3.6%, and 0.3% for COI, 16S and ITS-1, respectively (Table 3). The analysis fully resolved the phylogenetic position of P. organiaca and P. navasi. Nevertheless, the phylogenetic relationships of P. carascalensis, P cotiellae, and P. molae could not be resolved. The monophyly of P. velascoi was well supported (BS >85%) and this nominal species was grouped in subclade 3f with P. transfuga and with the populations of P. carascalensis from the West-Pyrenees. Pyrenaearia transfuga and P. carasca- lensis appeared in an unresolved poly- tomy. Excluding P. velascoi, all the haplotypes of the Cantabrian Pyrenae- ar/fi populations were grouped in clade4, which thus comprised P. cantabrica, P. oberthueri, P. daanidentata, P. ponce- bensis, and P. schaufussi. DISCUSSION Species delineation and taxonomy The present study is the first attempt ever to determine species delineation o( Pyrenaearia using DNA- based systematics. All haplotypes were unambiguously grouped within four main clades. fhe phylogeny of Pyrenaearia was highly consistent with the current geographical distribution of the clades. Accortling to the explicit tree-based species delimitation protocol recommended by Wiens and Penkrot (2002) and the geographic distribution, we conclude that the four clades should CLADE 1 P. oberthueri -0 1 P. oberthueri -02 P. oberthueri -03 P. oberthueri -04 P. oberthueri -05 P. oberthueri -06 P. poncebensis -0 1 P. poncebensis -02 P. schaufussi -01 P. schaufussi -02 P. cantabrica -01 P. cantabrica -02 P. cantabrica -03 P. cantabrica -04 P. cantabrica -05 P. cantabrica -06 P cantabrica -07 P. cantabrica -08 P. cantabrica -09 P cantabrica - 1 0 P cantabrica - 1 1 P. cantabrica - 1 2 P. cantabrica - 1 3 P cantabrica - 1 4 P cantabrica - 1 5 P cantabrica - 1 6 P cantabrica - 1 7 P cantabrica - ! 8 P cantabrica - 1 9 P. (laaniilentata -01 ! cantabrica -2 1 3/3/2 P cantabrica -22 Jaanidentata -02 P cantabrica -20 5/5/4, P. esserana -0 1 ! esserana -02 'esserana -03 P. carascalopsis -0 1 P carascalopsis -0^ 5/5/; P. parva -OT P, parva -02 P. parva -03 sP. parva -0^ CLADE 2 -/-/I 3/5/— \ 4/5/5 i^Pfor^aniaca -0 1 P. organiaca -Oi P. n, navasi -01 P n. navasi -02 P. n. sylvatica -03 P. n. svlvatica -04 — P. molae -Ol 5/5/2 — P molae -02 ‘ — P. cotiellae -0 1 P. cotiellae -02 I P cotiellae -03 ' P. cotiellae -04 — P. carascalensis -0 1 • P. carascalensis -02 — P carascalensis -03 — P carascalensis -04 P. carascalensis -05 P. carascalensis -06 P carascalensis -08 ' P carascalensis -09 P. carascalensis -07 P velascoi -0 1 P. transfuga -01 P. transfuga -02 P. velascoi -02 P velascoi -03 P. velascoi -04 P velascoi -05 P. velascoi -06 P. velascoi -07 P velascoi -08 P. velascoi -09 P. velascoi - 1 0 P. velascoi - 1 ! P. velascoi - 1 7 CLADE 3 o.noi substimiion/sitc Figure 2. Unrooted N| tree of nuclear ITS-1 from Pyrenaearia .specimens. Bootstrap values under NJ, Ml^, and posterior probabilities under MB are shown at each node (N)/M1VMB) when >70'M). Number codes indicate bootstrap values aiul posterior probabilities: 1, 100%; 2, 99-95%; 3, 94-90%; 4, 89-80'M.; and 3, 79-65%. MOLECULAR PHYLOGENY OE PYRENAEARIA 73 0.005 substitution/site 1/1/3 \ OLTCROCPS 1 CL.ADE 1 CLADE 2 3a 3a 3b 3b 3c 3e 3d 3d 3c 3e CLADE 3 3f 3f CLADE4 0,005 substitulion/site Figure 3. Molecular phylogenies of COI (left) and 16S rRNA (right) by NJ analyses. Bootstrap values under NJ, MP, and posterior probabilities under MB, are given at each node (NJ/MP/MB) when >70%. Bootstrap values are indicated by number codes (see legend of Fig. 2). be considered different species under the phylogenetic species concept. With the exception of clades 3f and 4, all the clades are allopatric and, consequently, their reproductive isolation could not be tested. The species inhabiting the Pyrenees, the eastern pre-Pyrenees, the Tarragona, and the Moncayo mountains involved three basal clades. Pyrenaearia carascalopsis and P. esserana formed a strongly supported phylogroup (clade 1). Both nominal species were not reciprocally monophyletic suggesting that they constitute a single taxon instead of two different species as proposed by Fagot (1906). In addition, some authors placed both nominal species in the synonym P. carascalensis (Germain 1930, Puente 1994). Yet our analyses showed that this opinion cannot be supported, and we instead agree with Prieto (1986) who joined P carascalopsis and P. esserana in a single taxon because both taxa have neither well-defined morphological differences, nor particular geographical ranges. Pyrenaearia parva (clade 2) is a valid morphospecies based on differences in shell morphology and body color (Ortiz de Zarate 1956, Prieto 1986, Puente 1994). It constitutes a well-defined clade that will not be discussed further. In this case, morphological and molecular evidence concerning the species status of P. parva were concordant. Clade 3 joined seven nominal species. Pyrenaearia organiaca and P. navasi were recovered as two well-supported clades. Both morphospecies are very different from other Pyrenaearia taxa in terms of shell shape and geographic distribution. Pyrenaearia organiaca lives at low altitudes in the eastern pre-Pyrenees and its shell is very different in size, color, and sculpture (Ortiz de Zarate 1956, Bech 1990, Puente 1994). Pyrenaearia navasi is the only Pyrenaearia taxon occurring south of the Ebro valley and it also has a unique shell shape, shell color, and fragility (Prieto 1991). Therefore, we suggest that P. organiaca and P. navasi should be treated as valid species, based on phylogenetic and morphological criteria. The four P navasi haplotypes form an unresolved intraspecific polytomy, thus eventual subspecific categories cannot be recognized. Pyrenaearia carascalensis, P. molae, P cotiellae, P velascoi, and P. transfuga formed a species complex (referred to as P. carascalensis s.l.) whose taxonomy can only be resolved after more intensive sampling and/or by screening more-sensitive markers. Pyrenaearia cotiellae. 74 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 H. Umbata l/l/l ,'I;I 2/5/1 4/3/1 4/4/ 1/1/1 suhsuiiilion/siii; P. carascalopsis-0\ I — P. carascalop.s'is-02 H I P. esserana-Q\ H P esserana-02 ' P. essenimi-03 P. parva-()\ P. parva-02 L| Ppan’a-03 - P parva-QA /l/l j- P. urganiaca-Q\ P. organi(ica-Q2 P. n. navasi-0\ P. n. navasi-Q2 P. n. sylvatica-03 P. n. sylvatica-iSA P. P. iiu)lae-(i2 P. a)tiellae-70%. Topotype specimens are indicated in bold. Bootstrap values are indicated by number codes (see legend of I'ig. 2). MOLECULAR PHYLOGENY OE PYRENAEARIA 75 Table 2. Uncorrected average (± SD) p-distances of COI (first line), 16S (second line), and lTS-1 (third line) within and among clades. Mean values of the average p-distances between a clade and the other three are also indicated. CLADE 1 CLADE 2 CLADE 3 CLADE 4 2.12 % CLADE 1 0.77 % 0.52 % 11.36% ±0.0238 0.45 % CLADE 2 6.26 % ± 0.003 0.63 % 0.72 % ± 0.0035 0% 10.48% ±0.0067 9.50 % ± 0.0066 3.01 % CLADE 3 7.23 % ± 0.0048 5.76% ±0.0041 2.82 % 1.07% ± 0.0030 0.70 % ± 0.0008 0.16% 8.94 % ± 0.0053 9.06 % ± 0.0053 8.50 % ± 0.0090 2.72 % CLADE 4 6.72 % ± 0.0038 4.56 % ± 0.0044 6.05% ±0.0061 1.6% 1.24% ±0.0038 0.81 %± 0.0016 1.08% ±0.0018 0.21 % 9.81 %± 0.01 16 9.42 % ± 0.0095 8.79% ±0.011 8.56 % ± 0.0087 Mean values 6.94 % ± 0.0052 5.24 % ± 0.0077 6.17% ±0.007 5.99 % ± 0.0076 1.13% ± 0.0036 0.76% ±0.0016 1.05% ±0.0022 1.08% ±0.0023 Table 3. Uncorrected average (± SD) p-distances of COI (first line), 16S rRNA (second line), and ITSl (third line) within and among clade 3 subgroups. Mean values of the average p-distances between clade 3 subgroups and the other five are also indicated. CLADE 3a CLADE 3b CLADE 3c CLADE 3d CLADE 3e CLADE 3f % % CLADE 3a 1.58 % 0.25 % 0.15 % 8.92 % ± 0.0040 0.3 CLADE 3b 5.98% ±0.0017 0.13 0.37 % ± 0.0007 0 7.44 % ± 0.0026 5.64 CLADE 3c 6.16% ± 0.0014 2.83 0.22 % ± 0.0008 0.15 7.90 % ± 0.0034 5.42 CLADE 3d 6.66 % ± 0.0023 3.21 0.29% ±0.0011 0.22 8.80 % ± 0.000 7.45 CLADE 3e 6.18% ±0.0032 3.85 0.22 % ± 0.0009 0.15 7.36 % ± 0.0050 4.51 CLADE 3f 7.31 %± 0.0057 3.63 0.37% ±0.0010 0.3 7.71 %± 0.0073 4.92 Mean values 6.93 % ± 0.0071 3.53 0.34% ±0.0011 0.27 % ± 0.0000 % ± 0.0024 0% %± 0.0012 0% % ± 0.000 0% % ± 0.0033 3.5% ±0.0012 1.05% %± 0.0031 2.39 % ± 0.003 0.92 % % ± 0.0008 0.07 % ± 0.0008 0.1 % % ± 0.0024 5.19% ±0.000 4.4 % ± %± 0.0012 1.76 % ± 0.000 1.95 % ± % ± 0.000 0 % ± 0.000 0.07 % ± % ± 0.0048 2.44 % ± 0.0045 1.97 %± % ± 0.0032 2.63 % ± 0.0043 3.21 %± % ± 0.0006 0.15 % ± 0.0006 0.22 % ± % ± 0.009 3.17% ±0.0126 2.72 % ± % ± 0.0038 2.57 % ± 0.0043 3.06 % ± % ± 0.0008 0.13 % ± 0.0007 0.2 % ± 0.0012 0% 0.0026 0.76 % 0.0008 0% 0.0036 4.01 % ± 0.0047 0.88 % 0.004 2.95 % ± 0.0033 1.24% 0.0006 0.15% ±0.0006 0.06 % 0.0132 4.58% ±0.0122 3.24% ±0.0121 0.0052 2.86 % ± 0.0063 3.21 %± 0.0051 0.001 0.13% ±0.0007 0.22 % ± 0.0009 P. rnolae, and P velascoi are monophyletic lineages and constitute three phylogeographic units in the sense of Avise (1989). Therefore, we suggest that they should maintain their nomenclatorial identity within the P. cnrascalensis s.l. group. Definition of other phylogeographic units within P caras- calensis s.l. would require a wider geographic sampling. Einally, clade 4 comprised five nominal species from the Cantabrian mountains; Pyrcnaearia cantabrica, P. 76 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 daanidentata, P. oberthueri, P. poncebensis, and P. schaiifussi. Haplotypes of P poncebensis and P schaiifussi topotypes were assigned to the P cantabrica haplogroup. Thus, we include both names in the synonymy of P. cantabrica. Moreover, the morphological distinction between P schaiifussi and P cantabrica has never been well documented (Azpeitia 1925, Ortiz de Zarate 1956, Prieto 1986, Puente 1994) and both taxa were usually defined by their geographical distribution (Altonaga et al. 1994), with P cantabrica living in the West- Cantabrian mountains and P. schaiifussi in the East. The morphospecies P poncebensis was defined by the presence of periostracal hairs on adult shells (Ortiz de Zarate 1956). Haired Pyrenaearia specimens have been observed in several localities, interspersed between the localities of P cantabrica (Puente 1994) and recently, we have observed haired adult shells even in the locus typicus of P. cantabrica. Clearly, our molecular results suggest that the presence of hairs in adult shells is not diagnostic in Pyreanaearia but rather represents a microhabitat adaptation that probably allows snails to retain water from fog in dry places. This hypothesis is based on the observation that haired P. poncebensis live under overhanging rocks in the locus typicus while smooth, hairless shells occur on neighboring vertical walls. Pyrenaearia oberthueri did not represent a monophyletic group within clade 4. This taxon possesses an altitudinal dine in shell size and shape (Ortiz de Zarate 1956), with small, conical forms at higher altitudes {oberthueri morph) and larger, flattened shells in the valleys {cantabrica morph). Both morphs possess a bluish shell color with a brownish-red peristome (Ortiz de Zarate 1956). As such, P. oberthueri seems to be merely a high-altitude form of P. cantabrica, living above 1500 m in the Central Massif of Picos de Europa. The bluish color of the valley morph shells is probably a local variation of P. cantabrica. More studies are needed to survey and understand this clinal variation and to uncover additional geographical structure in populations {e.g., haplotypes from Santa Ana mountains). Pyrenaearia daanidentata is the only monophyletic morphospecies in clade 4. It is restricted to the highest altitudes (>2000 m) of the West Massif of Picos de Europa (Pena Santa) and its shell differs strongly from that of P. cantabrica by growing slower, being smaller and possessing two strong teeth in the peristome (Raven 1988). Based on the phylogenetic results, together with morphology and geographic distribution, P. daanidentata should retain its taxonomic identity. Whether this taxon has a specific or subspecific status requires further research. Interestingly, Pyrenaearia cantabrica and P. vclascoi (clades 4 and 3f, respectively) coexist above 1300 m in three mountains of the Ba.sque country: Corbea [Aldamin (30IWNI865), Pena Eekanda (30rWNI668)l and Beriain (301WN8349) (Prieto 1986; present work). Yet, without exception, the haplotypes of P. velascoi and P. cantabrica from these areas were assigned to clades 3f and 4, respectively. These results, together with the absence of intermediate shell forms in these localities (Gomez-Moliner, pers. obs.), suggest that both taxa do not hybridize, so that P. cantabrica and P. carascalensis s.l. (including the nominal species P. transfuga and P. velascoi) should be considered different species under the morphological, biological, and phylogenetic species concepts. Biogeographical considerations and divergence times The phylogenetic relationships of the four basal clades remain unclear since they were joined in an unresolved polytomy. It is not excluded that this could be a hard poly- tomy (Walsh et al. 1999) if the radiation of the four clades happened in a short period of time. This seems to be supported by the very similar mean p-distances between them. However, from the beginning of the Pyrenaearia diversification to the present time, several lineages could have become extinct, although fossil evidence has not been reported. Missing evolutionary intermediates can indeed lead to phylogenetic artifacts, including unresolved nodes (Emerson 2002, Holland and Hadfield 2004). A molecular clock is required for estimating divergence times. Unfortunately, the evolutionary rate of COI, 16S, or ITS-1 sequences of Pyrenaearia is unknown. Evolutionary rates of mtDNA are often estimated at 2% pairwise sequence divergence per million years (My) for invertebrates (DeSalle et al. 1987). Yet, faster clocks have been postulated in some terrestrial snails (Thomaz et al. 1996), so that mtDNA evolutionary rates of terrestrial gastropods tentatively vary between 1% and 10% per My (Thomaz et al. 1996, Douris et al. 1998, Chiba 1999, Hayashi and Chiba 2000). Sometimes, 2-5% sequence divergence per My has been tentatively used for several helicoids (Thomaz et al. 1996, Pfenninger et al. 2003). Hayasi and Chiba (2000) estimated a divergence rate of 10% for 16S in a bradybaenid using more reliable data than previous speculations. The mean values of the uncorrected average p-distances among basal clades were 8. 5-9. 8% for COI and 5. 2-7. 2% for 16S. Using a fast rate of 10% per My, the four clades would have separated during the Pleisto- cene (0.5- 1.0 My). However, applying a more conventional divergence rate of 2% would imply that separation of the four clades could be dated between 2.6 and 4.9 My ago, in the Pliocene. Consequently, the split of the four major lineages in Pyrenaearia probably predated the Pleistocene glaciations as also postulated lor other animal groups (Hewitt 1996, Taberlet et al. 1998). The genetic distances of mtDNA genes between clade 3a and the other subgroups of clade 3 (b-f) were similar to those indicated for the four basal clades (7.7%, 6.9‘M) for COI, 16S) and probably predated the Pleistocene glaciations as well. The mean values of the average pairwise MOLECULAR PHYLOGENY OF PYRENAEARIA 77 genetic distances among the clades 3b-f ranged between 2.8% and 5.7% for COI and 2.2%-3.3% for 16S. Hence, the separation of clades 3b-f could be dated during Pleistocene glaciations [from Gunz on, 600-500 thousand years (ky)], when considering the fast molecular rate of 10% per My. Gonsequently, Pyrenaearia speciation within the four main groups is supposed to have occurred during cooling and warming cycles of the Pleistocene, which were particularly intense during the last 600 ky. During Pleistocene cooling periods, the high-altitude adapted populations had to migrate from their warm period refuges on the top of the mountains to lower altitudes in the valleys of the Cantabro-Pyrenaean region and the Ebro valley as a consequence of the permanent ice coverage of the mountains. In some cases, this downward migration was probably followed by expansions of the distributional ranges of cold-adapted species, resulting in possible secondary contacts. The presence of Pyrenaearia velascoi in the Cantabrian mountains seems to be the result of one of such expansion events that occurred probably during the last glaciation (Wurm, 120-18 ky). This expansion of P. carascalensis s.l. to the west would have resulted in the secondary contact between the pyrenaean P. carascalensis s.l and the Cantabrian P cantabrica lineages, but the previous allopatric speciation of these two lineages must have early enough to build up reproductive isolation. The most recent speciation process began with the last deglaciation phase, starting ca. 14,000 ky BP, resulting in the shrinkage of the distribution range of cold-adapted species. This may have led to, for example, the isolation and subsequent genetic differentiation between Pyrenaearia carascalensis s.s. and P velascoi. The small population of P. carascalensis living in Aspe valley {-P. transfuga morphospecies) could be a relict population surviving at low altitudes or may have been established by passive transport (heavy wind, rainstorms, birds) as was described in some other terrestrial molluscs (Gittenberger et al. 2006). The separation of two populations of P. navasi, one restricted to high altitudes (above 2000 m), the other sheltered within deciduous forests in the same mountainous system seems to also postdate the Last Glacial Maximum. The resolution of all these and some other questions needs more field and laboratory research. Several workers have demonstrated that terrestrial molluscs are valuable organisms to investigate evolutionary processes in species with poor dispersal abilities (Davison and Clarke 2000, Hausdorf and Henning 2004, Schilthuizen et al. 2004, 2006, Uit de Weerd 2004, Ketmaier et al. 2006). The land snail genus Pyrenaearia is a polytypic group of closely related taxa that could constitute a good model organism for molecular phylogenetic studies. The existence of several taxa, which are confined to high-altitudes and, consequently, are intensely affected by insularity phenomena, makes these organisms suitable for the study of allopatric and/or peripat- ric speciation processes of isolated lineages. Moreover, the presence of some species living at high altitudes in close vicinity to other species restricted to the valleys within the same mountainous system also makes them a good model to study parapatric speciation. The effects of geo-historical climate changes on high-altitude species is poorly understood (Haubrich and Schmitt 2007). The genus Pyrenaearia could also represent an ideal group for the study of the effects of cyclic climate changes on geographic distribution of gene lineages and population structure in a group containing several taxa with subalpine affinities. ACKNOWLEDGMENTS This work has been funded by the Basque Country University. Project numbers: 00076. 125-EA-7876/2000 and 00076.3 10-E- 15256/2003. A. Elejalde holds a Ph.D. fellowship awarded by the Basque Country University. Authors wish to thank A. Bertrand, J. Corbella, and G. Renobales for supplying Pyrenaearia specimens for this study. We are also greatly indebted to K. Breugelman’s Ph.D. students and technicians of The Royal Belgian Institute of Natural Sciences for their valuable help. LITERATURE CITED Altonaga, K., B. Gomez, R. Martin, C. E. Prieto, A. I. Puente, and A. Rallo. 1994. Estudio faunistico y biogeogrdfico de los moluscos terrestres del norte de la Peninsula Iberica. Eusko Legebiltzarra- Parlamento Vasco, Vitoria-Gasteiz, Spain [In Spanish]. Avise, L C. 1989. A role for molecular genetics in the recognition and conservation of endangered species. Trends in Ecology and Evolution 4: 279-281. Avise, J. C. 2000. Phylogeography: The History and Pormation of Spe- cies. Harvard University Press, Gambridge, Massachusetts. Azpeitia, F. 1925. Estudio de las formas de moluscos espanoles, mas afines a las Helix cantabrica y Helix oreina. Boletin de la Sociedad Iberica de Ciencias Naturales 23: 138-177 [In Spanish]. Bech, M. 1990. Fauna malacolbgica de Catalunya. Mollusc terrestres I d'aigua dol<;a. Treballs de la Institucio Catalana dHistoria Nat- ural 12: 1-229 [In Catalan]. Chiba, S. 1999. Character displacement, frequency-dependent se- lection and divergence of shell colour in land snails Manda- rina (Pulmonata). Biological Journal of the Linnean Society 66: 465-479. Coyne, I. A. and H. A. Orr. 2004. Speciation. Sinauer Associates, Sun- derland, Massachusetts. Cracraft, ]. 1983. The significance of phylogenetic classifications for systematic and evolutionary biology. In: ). Felstenstein, ed.. Numerical Taxonomy. Springer-Verlag, Berlin. Pp. 1-17. Davison, A. and B. Clarke. 2000. History or current selection? A molecular analysis of ‘area effects’ in the land snail Cepaea 78 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 nernoralis. Proceedings of the Royal Society of London (B) 267: 1399-1405. DeSalle, R., A. Templeton, I. Mori, S. Pletscher, and I. S. Johnston. 1987. Temporal and spatial heterogeneity of mtDNA polymor- phism in natural populations of Drosophila mercatoriun. Ge- netics 116: 215-223. Douris, V., R. A. D. Cameron, G. C. Rodakis, and R. Lecanidou. 1998. Mitochondrial phylogeography of the land snail Albinaria in Crete: Long-term geological and short-term vicariance effects. Evolution 52: 1 16-125. Emerson, B. C. 2002. Evolution on oceanic islands: Molecular phy- logenetic approaches to understanding pattern and process. Molecular Ecology 11:951 -966. Fagot, R 1906. Mollusca Nova. Provinciae Aragoniae. Boletln de la So- ciedad Aragonesa de Ciencias Naturales 5: 171-173 [In French]. Felsenstein, J. 1985. Confidence limits on phylogenies: An approach using the bootstrap. Evolution 39: 783-791. Folmer, O., M. Black, W. Floeh, R. Lutz, and R. Vrijenhoek. 1994. DNA primers for amplification of mitochondrial cytochrome c oxidase subunit 1 from diverse metazoan invertebrates. Molecu- lar Marine Biology and Biotechnology 3: 294-299. Germain, L. 1930. Mollusques terrestres et fluviatiles. Tomo 21. Iiv. P. Lechevalier, ed., Eaune de la Erance. Federation Franpaise des Societes de Sciences Naturelles, Paris. Pp. 477 [In French]. Gittenberger, E. D., S. L Groenenberg, B. Kokshoorn, and R. G. Preece. 2006. Biogeography: Molecular trails from hitch-hiking snails. Nature 439: 409. Giusti, F. and G. Manganelli. 1992. The problem of the species in malacology after clear evidence of the limits of morphological systematics. In: E. Gittenberger and J. Goud, eds.. Proceedings in the Ninth International Malacological Congress, Edinburgh, 31 August-6 September, 1986. Leiden. Pp. 153-172. Haubrich, K. and T. Schmitt. 2007. Cryptic differentiation in alpine- endemic, high-altitude butterflies reveals down-slope glacial refugia. Molecular Ecology 16: 3643-3658. Hausdorf, B. and C. Hennig. 2004. Does vicariance shape biotas? biogeographical tests of the vicariance model in the north-west European land snail fauna. Journal of Biogeography 31: 1751- 1757. Hayashi, M. and S. Chiba. 2000. Intraspecific diversity of mitochon- drial DNA in the land snail Euhadra peliomphala (Bradybaeni- dae). Biological Journal of the Linnean SocietylO: 391-401. Llewitt, G. M. 1996. Some genetic consequences of ice ages, and their role, in divergence and speciation. Biological Journal of the Lin- nean Society 58: 247-276. I lillis, D. M. and M. T. Dixon. 1991 . Ribosomal DNA-molecular evo- lution and phylogenetic inference. Quarterly Review of Biology 66: 410-453. I lillis, D. M. and |. I. Bull. 1993. An empirical test of bootstrapping as a method lor as.sessing confidence in phylogenetic analysis. Systematic Biology 42: 1 82- 1 92. Holland, B. S. and M. G. Iladficld. 2004. Origin and diversification of the endemic I lawaiian tree snails ( Achatinellidae: Achatinel- linae) based on molecular evidence. Molecular Phylogenetics and Evolution 32: 588-600. Huelsenbeck, J. P. and F. R. Ronquist. 2001. MRBAYES: Bayesian in- ference of phylogeny. Bioinformatics 17: 754-755. Ketmaier, V., F. Giusti, and A. Caccone. 2006. Molecular phylogeny and historical biogeography of the land snail genus Solatopupa (Pulmonata) in the peri-Tyrrhenian area. Molecular Phyloge- netics and Evolution 39: 439-451. Moorsel, C. H. M., W. J. Nes, and H. J. Megens. 2000. A quick, simple, and inexpensive mollusc DNA extraction protocol for PGR-based techniques. Malacologia 42: 203-206. Ortiz de Zarate, A. 1956. Observaciones anatomicas y posicion sistematica de varios Heliddos espanoles. IV. Genero Pyrenae- aria Hesse, 1921. Boletln de la Real Sociedad Espanola de Histo- ria Natural 54: 35-61 [In Spanish]. Palumbi, S. R., A. P. Martin, S. Romano, W. O. McMillan, L. Stice, and G. Grabowski. 1991. The Simple Pool’s Guide to PGR. Spe- cial Publication Department of Zoology, University of Hawaii, Honolulu. Pfenninger, M., D. Posada, and F. Magnin. 2003. Evidence for sur- vival of Pleistocene climatic changes in Northern refugia by the land snail Trochoidea geyeri (Sods 1926) (Helicellinae, Stylo- matophora). BMC Evolutionary Biology 3: 8. Posada, D. and K. A. Crandall. 1998. Modeltest: Testing the model of DNA substitution. Bioinformatics 14: 817-818. Prieto, C. E. 1986. Estudio sistematico y biogeografico de los He- licidae sensu Zilch, 1959-60 (Gastropoda: Pulmonada: Sty- lommatophora) del Pais Vasco y regiones adyacentes. Ph.D. Dissertation, University of the Basque Country, Spain [In Spanish]. Prieto, C. E. 1991. Pyrenaearia navasi sylvatica ssp. nov., una subespe- cie originada por una regresion glaciar cuaternaria en el macizo del Moncayo. Kobie 20: 69-75 [In Spanish]. Puente, A. I. 1994. Estudio taxonomico y biogeografico de la super- familia Helicoidea Rafinesque, 1815 (Gastropoda: Pulmonata: Stylommatophora) de la Peninsula Iberica e Islas Baleares. Ph.D. Dissertation, University of the Basque Country, Spain [In Spanish]. Raven, J. G. M. 1988. Pyrenaearia daanidentata spec. nov. (Helici- dae), a toothed species from the Cantabrian mountains, Spain. Basteria 52: 121-123. Rodriguez, E, J. L. Oliver, A. Marin, and ). R. Medina. 1990. Ehe general stochastic model of nucleotide substitution, lournal of Theoretical Biology 142: 485-501. Saitou, N. and M. Nei. 1987. 4’he neighbour-joining method: A new method for reconstructing phylogenetic trees. Molecular Biol- ogy and Evolution 4: 406-425. Schilthuizen, M., R. E. I loekstra, and E. Gittenberger. 2004. 1 lybrid- ization, rare alleles and adaptive radiation. Trends in Ecology and Evolution 19: 404-405. Schilthuizen, M., A. Van Til, M. Salverda, T. S. Liew, S. lames, B. Bin Elahan, and |. I. Vermeulen. 2006. Microge(^graphic evolution of snail shell shape ami predator behaviour. Evolution 60: 1851- 1858. Simpson, ). ). 1961. Principles of Animal Ja.xonomy. Columbia Uni- versity Pre.ss, New York. MOLECULAR PHYLOGENY OF PYRENAEARIA 79 Swofford, D. L. 2002. PAUP* Beta Version Phylogenetic Analysis Using Parsimony (* and Other Methods). Sinauer Associates, Sunder- land, Massachusetts. Taberlet, P., L. Fumagalli, A. G. Wust-Saucy, and ]. F. 1998. Com- parative phylogeography and postglacial colonization routes in Europe. Molecular Ecology 7: 453-464. Thomaz, D., A. Guiller, and B. Clarke. 1996. Extreme divergence of mitochondrial DNA within species of Pulmonate land snails. Proceedings of the Royal Society (B, Biological Sciences) 263: 363-368. Thompson, J. D., T. D. Gibson, F. Plewniak, F. Jeanmougin, and D. G. Fliggins. 1997. The CLUSTAL_X windows interface: Flex- ible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 25: 4876-4882. Uit de Weerd, D. R. 2004. Molecular phylogenetic history of eastern Mediterranean Alopiinae, a group of morphologically indeter- minate land snails. Ph.D. Dissertation, Leiden University, The Netherlands. Uit de Weerd, D. R., W. H. Piel, and E. Gittenberger. 2004. Wide- spread polyphyly among Alopiinae snail genera: When phy- logeny mirrors biogeography more closely than morphology. Molecular Phylogenetics and Evolution 33: 533-548. Vogler, A. P. and M. T. Monaghan. 2007. Recent advances in DNA taxonomy. Journal of Zoological Systematics and Evolution Re- search 45: 1-10. Walsh, FI. E., M. G. Kidd, T. Mourn, and V. L. Friesen. 1999. Poly- tomies and the power of phylogenetic inference. Evolution 53: 932-937. Wiens, ]. J. and T. L. Penkrot. 2002. Delimiting species based on DNA and morphological variation and discordant species limits in spiny lizards {Sceloporus). Systematic Biology 51: 69-91. Wiley, E. O. 1978. The evolutionary species concept reconsidered. Systematic Zoology 27: 17-26. Submitted: 9 November 2008; accepted: 30 January 2009; final revisions received: 12 May 2009 80 AMERICAN MALACOLOGICAL BULLETIN 11 'Ml- 2009 Appendix 1. Species abbreviations, geographical coordinates (using Spanish grid references), and GenBank accession sequences of Pyrenaearia. Asterisks {*) and bold indicate topotypes. GenBank access Abbreviation Localities-U.T.M. GenBank access number of COI number of 16S rRNA P. carascalopsis-Ol* Port d'Salau (Lleida) 31TCH4733 EU3 10004 EU3 10083 P. carascalopsis-Q2 Alto Aneu (Lleida) 31TCEI4133 EU3 10005 EU310084 P. esserana-Ol* La Renclusa (Huesca) 31TCH0727 EU3 10006 EU310085 P esserana-02 Port de Viella (Lleida) 31TCH1624 EU3 10007 EU3 10086 P. esserana-03 Port de Viella (Lleida) 31TCH1624 EU3 10008 EU310087 P. parya-01* Pedraforca (Barcelona) 31TGG9377 EU3 10009 EU3 10088 P. parva-02* Pedraforca (Barcelona) 31TCG9377 EU310010 EU3 10089 P. parva-Q3 Sierra de Cadi (Barcelona) 31TCG9582 EU310011 EU3 10090 P. parva-04 Sierra de Cadi (Barcelona) 31TCG9582 EU310012 EU3 10091 P. organiaca-Ol* Congost d'Organya (Lleida) 31TCG6377 EU310013 EU3 10092 P. organiaca-02* Congost d'Organya (Lleida) 31TCG6377 EU310014 EU3 10093 P. navasi navasi-Ol* Moncayo (Zaragoza) 30TWM9726 EU310015 EU3 10094 P. navasi navasi-02* Moncayo (Zaragoza) 30TWM9726 EU310016 EU310095 P. navasi sylvatica-03* Moncayo (Zaragoza) forest 30TWM9829 EU310017 EU310096 P. navasi sylvatica-04* Moncayo (Zaragoza) forest 30TWM9829 EU310018 EU310097 P. molae-OM Mola de Colldejou (Tarragona) 31TCF2153 EU310019 EU3 10098 P. molae-02* Mola de Colldejou (Tarragona) 31TCE2153 EU3 10020 EU3 10099 P. cotiellae-0\* Circo de Armena (Huesca) 31TBH8010 EU3 10021 EU310100 P. cotiellae-02* Circo de Armena (Huesca) 31TBH8010 EU3 10022 EU310101 P. cotiellae-03 Collado de Sahiin (Huesca) 31TBH8616 EU3 10023 EU310102 P. cotiellae-Q4 Collado de Sahun (Huesca) 31TBH8616 EU3 10024 EU310103 P. carascalensis-Ol Muntaya de Casamanya (Andorra) 31TCH8215 EU310025 EU310104 P carascalensis-02 Muntaya de Casamanya (Andorra) 31TCH8215 EU310026 EU310105 P. carascalensis-03 Port Boucharo. Gavarnie (Francia) 30T YN4032 EU3 10027 EU310106 P. carascalensis-04 Port Boucharo. Gavarnie (Francia) 30T YN4032 EU3 10028 EU310107 P carascalensis-05 Pico Anie (Navarra) 30TXN8555 EU3 10029 EU310108 P carascalensis-06 Pico Anie (Navarra) 30TXN8555 EU3 10030 EU310109 P carascalensis -07 Pico Anie (Navarra) 30TXN8555 EU3 10031 EU310110 P. carascalensis-08 Portalet (Huesca) 30TYN1042 EU3 10032 EU310111 P. carascalensis-09 Pico Anie (Navarra) 30TXN8555 EU3 10033 EU310112 P. transfuga-Ol* Pont d'Esquit (Francia) 30TXN9559 EU3 10034 EU310113 P. transfuga-02* Pont d'Esquit (Francia) 30TXN9559 EU3 10035 EU310114 P. velascoi-01* Aldamin (Bizkaia) 30TWN1865 EU3 10036 EU310115 P. velascoi-02* Aldamin (Bizkaia) 30TWN1865 EU310037 EU310116 P. velascoi-03 Gorbea frente Aldamin (Bizkaia) 30TWN1865 EU3 10038 EU310117 P. velascoi-04 Txindoki (Gipuzkoa) 30TWN7463 EU3 10039 EU3101 18 P. velascoi-05 Txindoki (Gipuzkoa) 30TWN7463 EU3 10040 EU310119 P. velascoi-06 Aitzgorri (Gipuzkoa) 30TWN5456 EU3 10041 EU310120 P. vela scot -07 Monte Altxueta (Navarra) 30TWN8456 EU3 10042 EU310I21 P. velascoi-OS Aitzgorri (Gipuzkoa) 30TWN5456 EU3 10043 FU310I22 P. velascoi-09 Pena Lekanda (Bizkaia) 30TWN1668 FU3 10044 FU310I23 P. vclascai- 1 0 Pena Lekanda (Bizkaia) 301’WN1668 FU3 10045 FU310124 P. velascoi- 1 1 Beriain (Navarra) 3(n’WN8349 EU3 10046 FU310I25 P. velascoi-\2 Beriain (Navarra) 30TWN8349 FU3 10047 FU310126 P. cantahrica-0\* Galdas de Oviedo (Asturias) 30TTP630() FU3 10048 FU310127 P. cantabrica-02* Caldas de (Aiedo (Asturias) 3()TrP630() FU3 10049 FU3I0128 P. cantnbrica-03 Pena Lekanda (Bizkaia) 30rWNl668 FU3 1 0050 f:U3 101 29 P. caritabrica-l)4 Txarla/.o (Bizkaia) 3()'rVN9659 FU3 10051 FU3 10130 P. cmttabrica-()5 Txarlazo (Bizkaia) 3()’I'VN9659 FU3 10052 FLI3I013I numbers for all the GenBank access number of ITS 1 EU310162 EU310163 EU310164 EU310165 EU310166 EU310167 EU310168 EU310169 EU310170 EU310171 EU310172 EU310173 EU310174 EU310175 EU310176 EU310177 EU310178 EU310179 EU310180 EU310181 EU310182 EU3 10183 EU310184 EU3 10185 EU310186 EU310187 EU310188 EU310189 EU310190 EU310191 EU310192 EU3 10193 EU3 10194 EU3 10195 EU3 10196 EU310197 EU310198 EU310199 EU3 1 0200 EU3 10201 EU3I0202 EU3I0203 EU3I0204 EU3 10205 EU3 10206 EU3 10207 EU3 10208 EU3 10209 EU310210 MOLECULAR PHYLOGENY OF PYRENAEARIA 81 Appendix 1. (continued) Abbreviation Localities-U.T.M. GenBank access number of COl GenBank access number of 16S rRNA GenBank access number of ITS 1 P. cantabrica-06 Beriain (Navarra) 30TWN8349 EU3 10053 EU310132 EU310211 P. cnntabrica-07 Beriain (Navarra) 30TWN8349 EU3 10054 EU310133 EU310212 P. cantabrica-08 Beriain (Navarra) 30TWN8349 EU3 10055 EU310134 EU310213 P cantabrica-09 Proaza (Asturias) 29TQH4192 EU3 10056 EU310135 EU310214 P. cantabrica-lO Langro-Pto Tarna (Asturias) 30TUN1281 EU310057 EU310136 EU310215 P. cantabrica-\\ Langro-Pto Tarna (Asturias) 30TUN1281 EU310058 EU310137 EU310216 P. cautabrica-\2 Aldamin (Bizkaia) 30TWN1865 EU3 10059 EU310138 EU310217 P. cantabrica-l3 Poncebos (Asturias) 30TUN5191 EU310060 EU310139 EU310218 P. cantabrica-l4 Poncebos (Asturias) 30TUN5191 EU3 10061 EU310140 EU310219 P. cantabrica-l5 Caranga-Quiros (Asturias) 29TQH4089 EU310062 EU310141 EU310220 P. cantabrica- 16 Caranga-Quiros (Asturias) 29TQH4089 EU3 10063 EU310142 EU3 10221 P. cantabrica-\7 Cordinanes-Cain (Asturias) 30TUN4582 EU3 10064 EU310143 EU3 10222 P. cantnbrica-\8 Cordinanes-Cain (Asturias) 30TUN4582 EU3 10065 EU310144 EU3 10223 P. cantnbrica-\9 Desfiladero de los Beyos (Asturias) 30TUN3088 EU3 10066 EU310145 EU3 10224 P. cantabrica-20 Desfiladero de los Beyos (Asturias) 30TUN3088 EU3 10067 EU310146 EU310225 P. cantabrica-2\ Anboto (Bizkaia) 30TWN3270 EU3 10068 EU310147 EU3 10226 P cantabrica-22 Anboto (Bizkaia) 30TWN3270 EU310069 EU310148 EU3 10227 P. daanidentata-Ol* Hoyo del Burro (Leon) 30TUN4082 EU310070 EU310149 EU310228 P daanidentata-02 Canal del Perro (Leon) 30TUN4082 EU3 10071 EU310150 EU310229 P. oberthiieri-Ol Vega Urriello (Asturias) 30TUN5285 EU3 10072 EU310151 EU310230 P. oberthueri-02 Vega Urriello (Asturias) 30TUN5285 EU3 10073 EU310152 EU310231 P. oberthueri-03 Vega Urriello (Asturias) 30TUN5285 EU3 10074 EU310153 EU3 10232 P. oberthiieri-04 Vega Urriello (Asturias) 30TUN5285 EU3 10075 EU310154 EU310233 P oberthueri-05 Pena Santa Ana (Leon) 30TUN5282 EU3 10076 EU310155 EU310234 P. oberthueri-06 Pena Santa Ana (Leon) 30TUN5282 EU3 10077 EU310156 EU310235 P. poncebens'is-Ol* Poncebos (Asturias) 30TUN5191 EU3 10078 EU310157 EU310236 P. poncebensis-02* Poncebos (Asturias) 30TUN5191 EU3 10079 EU310158 EU310237 P. schaufussi-Ol* Urdon (Asturias) 30TUN6791 EU3 10080 EU310159 EU310238 P. schaufussi-02* Urdon (Asturias) 30TUN6791 EU3 10081 EU310160 EU3 10239 Hygromia limbata Pagasarri (Bizkaia) 30TWN0485 EU310082 EU310161 EU3 10240 Iberus gualtieranus LIuercal de Almeria (Almeria) 30SWF4880 AY928568 AY928596 EU446026 r I- ! ■» ■k V? ■"i> . ■ V art* ■ I »r lia ' . :. *' m r- .« wr- » ■■ ■ ■ ■ ■ ■< -r - ■ • ' '■ Hli. ♦■■ - ‘VK*.; '■♦■'■■!*. ,fc. »r> ' '♦ = t •*. ♦.-•■jT,*.- :jf ' ■ »■ :f ■•S Arner. Malac. Bull. 27; 83-93 (2009) Documenting molluscan evolution from ancient long-lived lakes: The case of Toxosoma Conrad, 1874 (Gastropoda, Cochliopidae) in Miocene Amazonian Lake Pebas"^ Frank P. Wesselingh and Willem Renema National Museum of Natural History Naturalis, P.O. Box 9517, 2300 RA Leiden, The Netherlands Corresponding author: wesselingh@naturalis.nnm.nl Abstract: Long-lived lakes with their often in situ evolved faunas form an excellent model system to study evolution. The paleontological record of long-lived lakes provides possibilities to investigate the tempo and mode of evolutionary change. However, in order to demonstrate such processes, a combined rigorous temporal and spatial framework is needed in order to make sure ( 1 ) alleged evolutionarily successive morphs are not misidentified immigrant species and (2) morphological variation is not due to ecophenotypy. Other factors that may cause the development of discrete morphs, such as sexual dimorphism, also must be considered. In this paper, evolution in the Miocene snail genus Toxosoma Conrad, 1874 from long-lived Lake Pebas in western Amazonia is studied within such a temporal and paleo-environmental framework. Keywords: anagenesis, cladogenesis, temporal/spatial framework, gastropod evolution Long-lived lakes are considered as laboratories and archives of evolution (Martens 1997). Faunas of these lakes, such as the well-known African cichlid fish, have served in many groundbreaking studies on adaptation, niche partition- ing, sexual selection, and other mechanisms underlying the process of speciation. Molluscs (and ostracods) are abundant in these lakes and contain hard shells that have a good fossilization potential. Their fossils have frequently been used in exploring these archives of evolution (e.g., Williamson 1981a, 1981b, Mensink 1984, Geary 1990, Geary et al 2002, Harzhauser and Mandic 2004, Wesselingh 2007). However, evolutionary insights based on the fossil record of long-lived lakes have been diverse and partially contradictory. A well- known example is the debate on alleged punctuated evolu- tionary change in Kenyan Lake Turkana mollusc faunas as proposed by Williamson (1981a, 1981b) and hotly contested afterwards (see references in Bocxlaer et al. 2007). In several studies, morphological change in successive stratigraphic intervals has been taken as evidence for evolution, underesti- mating the potential role of ecosystem variability and intrinsic biological processes involved in the generation of morpho- logical variation in molluscs. In this paper we aim to demonstrate that the fossil record of long-lived lakes is very well suited for the study of evolu- tionary patterns as long as adequate account is taken of the spatial and temporal context. The study is based on strati- graphic successive morphological change in snails of the genus Toxosoma Conrad, 1874 (Rissooidea: Cochliopidae), in Amazonian Lake Pebas during the Middle and early Late Miocene. The methodological context by which evolution might be demonstrable is outlined. LONG-LIVED LAKES Long-lived lakes are lakes that combine geological longevity and the development of endemic faunas. Geological longevity is not strictly delimited but usually taken as being longer than an interglacial time interval. The current Lake Victoria in East Africa, with its more than thousand endemic cichlid fish species, was completely desiccated only about 13 ka ago (Johnson et al. 1996). However, Quaternary Lake loanina in northern Greece existed for at least several hundreds of thousands of years, yet almost completely lacked endemic species (Frogley and Preece 2007). In order to develop endemic faunas, a lake must have been habitable for some time. Thirteen thousand years is the shortest known long-lived lake. Although lakes may persist over long time periods, it is the temporal continuity of ecological conditions under which lacustrine biota can live that is the most important feature of long-lived lakes (Wesselingh 2007). Typically, ecological continuity is on the order of 100 ka to several Ma. The faunas in long-lived lakes contain a significant proportion of endemic species. These lakes often show high * From the symposium “Molluscs as models in evolutionary biology: from local speciation to global radiation” presented at the World Con- gress of Malacology, held from 15 to 20 July 2007 in Antwerp, Belgium. 83 84 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 within-habitat diversity. The endemic species can be anage- netic derivatives of more widespread sister species (references in Wesselingh 2007), but in many of the long-lived lakes in situ radiations occur (Michel 1994, Muller et al. 1999, Nevesskaja et al. 2001, Geary et al. 2002). Such radiations have led to species flocks, i.e., monophyletic clades of mostly or entirely endemic species (West and Michel 2000). Long- lived lake faunas have a broad range of morphologies, often containing marine-like features such as unusually thick and strongly ornamented shells and apertural modifications including siphonal canals. A general trait shared by endemic taxa in long-lived lakes is the tendency to evolve K- reproductive strategies (Michel 1994). An exception has been documented from Late Miocene lake Pannon of central-eastern Europe, where secondary opportunist, endemic bivalve species evolved as r-strategists (Harzhauser and Mandic 2004). Long-lived lakes can be found in rift valleys (such as Lake Baikal in Russia and Lake Tanganyika in East Africa), in karst depressions (Lake Pozo in Indonesia and Lake Ohrid on the Balkans), in tectonic depressions (Lake Titicaca in South America and Lake Biwa in Japan), and even in meteorite craters (Miocene Lake Steinheim in southern Germany). They may be located at or below sea-level and contain brackish water such as the Gaspian Sea and Neogene Paratethyan lakes of southeastern Europe. The oldest, long- lived lacustrine mollusc fauna known is that of the Permian South American Parana Lake (Wesselingh 2007). Lake Pebas, occupying western Amazonia during the Early to early Late Miocene (ca. 23-11 Ma), was located in the Andean foreland basins and adjacent pericratonic basins. At its maximum, the lake measured over 1 million km^ in area (Wesselingh et al. 2002). EVOLUTION AND THE FOSSIL RECORD by certain, unique combinations of ranges of morphological characters. However, (sexual) dimorphism is an inherent biological process known to produce distinct morphs within species. The possibility that different morphs may represent males and females must therefore be considered in evolu- tionary studies as well. Freshwater gastropod species are also well known to show ecophenotypic variation, i.e., morpho- logical variation related to habitat differences lacking a hered- itary basis. Several mollusc groups exist whose species, as identified by DNA or ecological preferences, cannot be Toxosoma contorium Wesselingh, 2006 Intermediate sized elongate to ovate-conical shell Aperture flaring Plane of aperture tilted backwards Aperture contorted by well developed columellar plicae Well developed teeth behind aperture within outerlip Toxosoma grande Wesselingh, 2006 Large bulbous, comparatively thin shell Convex whorl profile Lacking teeth/spiral ridges inside outerlip Columellar plicae absent or very low Large and wide aperture Toxosoma denticulatum Wesselingh, 2006 Intermediate-large ovate to ovate-conical shell Thick shell Suture very shallow Whorl profile comparatively flat Apertural margins reinforced Columellar plicae very robuste Well developed teeth behind aperture within outerlip High aperture Toxosoma eboreum Conrad, 1874 Intermediate sized ovate-conical shell Thick shell Somewhat deepened suture Semiconvex whorl profile Base bodywhorl slightly bulbous Columellar plicae small but well developed Teeth behind aperture within outerlip of variable strength When working with fo.s.sils, species concepts are almost entirely morpho- logically based. Species can be defined Figure I. Toxosoiiui species from the Middle to early Late Miocene Pebas Formation of west- ern Amazonia. Specimen and locality information in Wesselingh (2()()6). Scale bar = I mm. GASTROPOD EVOLUTION IN MIOCENE LAKE PEBAS 85 distinguished on shell morphology, either by the lack of sufficient characters or by an abundance of apparently widely variable and overlapping characters. Within long-lived lakes taxa that are extremely difficult to separate based on shell characters are present, among others. Lake Tanganika snail species of the genus Lavigeria Bourguignat, 1888 (Michels 2004). Elowever, detailed study of distribution patterns, ecology, and DNA may lead to grouping of specimens that turn out to have more or less subtle but distinctive morpho- logical traits after all (Papadopoulos etal. 2004). Nevertheless, it can be expected that shell morphology cannot always be informative as to species discrimination. This may especially be the case in evolutionary transitional forms. The importance of sound systematics is not a trivial issue. Bocxlaer et al. (2007) showed that the erroneous generic identification of gastropod taxa in Lake Turkana has severely corrupted the evolutionary claims ofWilliamson (1981a, 1981b). Evolution in fossil biotas can be defined as morphological change in lineages (anagenesis) and the origination of novel morphologies within clades (cladogenesis) through time in as far these cannot be ascribed to ecophenotypy or polymor- phism. In order to demonstrate evolution, both control over stratigraphic time (temporal control) as well as the architecture of depositional environments is needed to address ecopheno- typy (spatial control). MATERIALS AND METHODS Toxosoma eboreum Conrad, 1874 is the single Toxosoma species (Gastropoda, Cochliopidae) known from Middle Miocene intervals of the Pebas Eormation (Wesselingh 2006). In successive late Middle to early Late Miocene intervals, three species are found {Toxosoma denticulatum Wesselingh, 2006; Toxosoma contortum Wesselingh, 2006; Toxosoma grande Wesselingh, 2006). The species have a range of shell characters that allow for their discrimination (Eig. 1). Since Toxosoma is endemic to the Pebas system, it is possible that the latter three species originated from a single Middle Miocene ancestor (see Discussion section below). Together with a further three species found in earlier strati- graphic intervals in the Pebas Eormation, Toxosoma may constitute a classical species flock. In the Pebas system, several putative species flocks occur, including a flock of the cochliopid genus Sioliella Haas, 1949. In order to assess whether the three species are not mere ecophenotypes or sexual dimorphs we have measured specimens from different stratigraphic and depositional environments. A total of 278 specimens belonging to the four species were retrieved from 17 Miocene Pebas Eormation localities in Colombian and Peruvian Amazonia (Table 1, Pig. 2, Appendix 1). A single unidentified specimen was also included in the analyses. Table 1. Localities; detailed information (including coordinates, collection dates, and maps) can be found in Wesselingh (2006) and Wesselingh et al. (2002, 2006b). MZ, mollusc zonation from Wessel- ingh et al. (2006b). ENZ, inferred environment from facies analyses (Rasanen et al. 1998, Wesselingh etal. 2006a) and from multivariate analysis of mollusc samples (Wesselingh et al. 2002). F, fluvial; ML, marginal lacustrine (prodelta, interdistributary bay, shoreface); L, lake shelf and bottom. Sample Locality MZ ENV F4 Mocagua (Amazonas, Colombia) 10 ML F6 Mocagua (Amazonas, Colombia) 10 ML F16 Los Chorros I (Amazonas, Colombia) 10 F F21 Los Chorros III (Amazonas, Colombia) 10 ML F22 Los Chorros III (Amazonas, Colombia) 10 L F29 Puerto Narino I (Amazonas, Colombia) 11 ML-F F32 Macedonia (Amazonas, Colombia) 11 ML F34 Zaragoza (Amazonas, Colombia) 11 ML F363 Iquitos Puerto GansoAzul (Loreto, Peru) 6 ML-L F367a Nuevo Horizonte III (Loreto, Peru) 9 L F417 Pebas XIII (Loreto, Peru) 7 L F489 Santa Elena I (Loreto, Peru) 8 ML F498 Santa Elena II (Loreto, Peru) 8 L F535 Santa Rosa de Pichana (Loreto, Peru) 7 L F685 Tamshiyacu (Loreto, Peru) 7 ML F702 Porvenir II (Loreto, Peru) 9 F F707 Porvenir IV (Loreto, Peru) 9 L F830 Momon III (Loreto, Peru) 7 L-ML F836 Nuevo Horizonte II (Loreto, Peru) 9 L The specimens were first identified using apertural outline and inclination, whorl profile, nature and number of columellar plicae, and grooves/denticles within the outer lip (Pig. 1). Three characters [height (H), height of aperture (Hap), and width of body whorl half a whorl before termination (WBW)] were measured with a resolution of 20 pm (Pig. 3). In Toxosoma specimens, the last quarter of the body whorl has a tendency to flare, which can be considered an adult apertural modification (Papadopoulos et al. 2004). The extent of flaring can vary considerably within and between populations. The maximum shell width (W) that is linked to the extent of flaring is too variable a measure and has been excluded from study. In many of the specimens, the aperture is damaged, precluding width measurements. In this study, the width of the shell half a whorl before the aperture is used in order to circumvent the above-mentioned problems. In three of the four species, the aperture tends to become tilted, making the height of the aperture (Hap) another potentially problematic character for study. How- ever, we observed a remarkable correlation between W and 86 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Age(Ma) c 10 Santa Rosa de Pichana Tamshiyacu Santa Elena Momon Q Iquitos Pebas Nuevo Horizonte O rietici Porvenir Yav9'' 100 km Puerto Narin^ Los Chorros Mocagua Macedonia Zaragoz Figure 2. Localities and stratigraphic context of studied material. A, fieldwork area. B, localities: white areas denote alluvial plains, grey areas indicate uplands (so called “terra firme”) that are underlain mostly by Neogene deposits. C, stratigraphic framework for the studied samples. Tor, Tortonian; Ser, Serravallian; Lan, Langhian; PZ, pollen zones (Hoorn 1993); G, Grimsdalea zone; C, Crassoretitrilctes zone; numbers refer to mollusc zones from Wesselingh et al. (2006b). Hap, strongly suggesting that the Hap is a usable character in spite of variations in the tilt of the apertural plane. Finally, the number of whorls was counted (Fig. 3). A Principal Components Analysis (PCA) was performed on the variables H, Hap, WBW, and Wh {N = 276) with Primer 6.1 (Clarke and Warwick 2001 ). Width (W) was ex- cluded becau.se of the large number of missing measurements. The first two principal components were plotted in morpho- logical space for each stratigraphical interval. The depositional environments of the Toxosoma species are inferred from sedimentary facies analyses (Riisanen cl al. 1998, We.sselingh et al. 2006a) and from multivariate analysis of mollu.se .samples (Wesselingh cl al. 2002). A variety of habitats, ranging from fluvial to lake bottoms, have been discerned. RESULTS In stratigraphic zones MZ6-MZ8, only the species Toxosoma cborcum is present. In samples from zones MZ9- MZl 1, T. coiitortiim, T. grande, and T. dcnticulatnm co-occur. Within species, size differences were observed in different samples. All identifications apart from one (see Discussion below) were unambiguous. Shell measurements are pre.sented (Appendix 1). The first two principal components (PC',1 and PC2) plotted here account for 97% of the variance in the data (per-axis inertia: PCI = 0.83; PC.2 = 0.12). PCI represents mostly size parameters (11, Hap, WBW-1). The .second PCA axis is strongly determined by the number of whorls. PCI and PC2 GASTROPOD EVOLUTION IN MIOCENE LAKE PEBAS 87 Figure 3. Characters measured. H, height; WBW-1, width of body whorl between half a whorl and one whorl before the termination of the shell; Hap, height of aperture. Wliorls are counted as in upper right. The width per whorl is counted as in lower right. scores are provided for the different species per stratigraphic interval (Eig. 4). Toxosoma eboreum occurs in the lower three stratigraphic zones (MZ6-MZ8). The populations in the MZ6 samples have low PCI scores and variability. PCI scores and variability are larger in the overlying two stratigraphic zones. The popula- tions in all three zones have intermediate PC2 scores. In stratigraphic zones MZ9-MZ11, Toxosoma contortum is characterized by low PCI scores and intermediate PC2 scores. In MZ9, PCI scores are most variable and partially overlap those of T. eboreum in the underlying MZ8. In the overlying MZIO and MZll, the average PCI score slightly increases. In MZ9, Toxosoma grande and T. denticulatum have very similar PCI scores, but are well distinguished by PC2 scores (low for T denticulatum and high for T grande). In the overlying MZIO and MZll, the PC2 values of both species largely overlap. A notable increase in PCI values and variability is found for T grande in successive stratigraphic zones. The unidentified Toxosoma species from interval MZIO scores between PC values of T contortum and T grande in that interval. In three of the four species {Toxosoma eboreum, T contorum, T grande) an increase in PCI values in successive time intervals is observed and corresponds to an anagenetic increase in size. DISCUSSION The data indicate anagenetic change in Toxosoma eboreum in stratigraphic zones MZ6-MZ8. The apparent existence of two morphs within MZ7 is discussed below. It is possible that T eboreum evolved from T. ovatum Wesselingh, 2006 that is common in MZ5. The apparent (because of low number of samples) sudden occurrence in MZ9 of Toxosoma contortum, T dentic- ulatum, and T grande is interpreted to result from cladogenesis (see below). More detailed sampling is required to establish the exact order of branching. The three species are well delim- ited through morphological characters as well as PC scores in MZ9 samples. Elowever, the PC scores of the latter two species largely overlap in the upper two stratigraphic zones (MZIO and MZll). The PCI values of both T contortum and T grande increase in successive intervals, again indicating anage- netic change towards larger sizes. On the scale of the mollusc biozones, both apparent “punctuated” (the occurrence of three species in MZ9) as well as gradual change {e.g., directional change within T contortum, T eboreum, and T denticulatum lineages) occurs. The exact time associated with the biozones is unknown but presumably on the order of several hundred thousand years per zone (Wesselingh et al. 2006a). The current sampling density and time control does not allow assessing whether the cladogenesis between MZ8 and MZ9 was sudden. The unidentified specimen in MZIO may be an aberrant specimen of one of the three species occurring in that zone. It also may represent a fifth species. However, we consider the last option unlikely as no similar specimens have been found in the large collection of Pebas material available. Einally, the unidentified morph may also represent a hybrid between Toxosoma contortum and T grande or T denticulatum. The limits imposed by the fossil record do not allow for an inter- pretation of the unidentified species. Toxosoma eboreum lived in lacustrine and marginal lacustrine habitats (see Table 1; Wesselingh et al. 2002). Toxosoma grande is a species from marginal lake habitats (inter- distributary bay, delta, and pro-delta) and fluvial habitats (see Table 1; Wesselingh et al. 2002). The lack of morphological intermediates with the other two co-occurring Toxosoma species in marginal lacustrine habitats indicates that they are not ecophenotypic morphs. Toxosoma denticulatum inhabited lake margins and lake shelves although low numbers occur in samples from lake bottom settings that were not sampled in this study. Toxosoma contortum has a similar distribution although more common in lake bottom samples. Since both species occur in the same habitats (Table 1) and lack morphological intermediates, they too are unlikely to be ecophenotypic variants. The three Toxosoma species occurring in MZ9-MZ11 have partially overlapping environmental optima. Toxosoma grande has its optimum in marginal lacustrine habitats, T. denticulatum on the lake shelf, and T. contortum on the lake shell and lake bottom. A number of large samples containing 88 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Figure 4. Character development in Pebasian Toxosoma species. MZ numbers refer to mollusc zones. PCI and PC2 refer to the PCA axes. .solely T. grande suggest that the form is not simply a male or female sexual dimorph, as at least some of the specimens corresponding to the other sex would be expected. Various samples containing only T. contortuin were found in strati- graphic zones and deposilional environments where T. dcntic- iilalutn also may have lived. 'I'his also indicates that it is unlikely that both forms are male and female morphs of a single species. Furthermore, in such a case a similar develop- ment in early ontogeny would he expected with divergence in adult stages (Shaver 1953). In some .samples of T. chorciini and T. denticulatum, apparent slender and broad morphs co-occurred. Pro- nounced sexual dimorphism has been demonstrated to occur in some coch- liopid gastropod species (Taylor 1987). Width/whorl measurements on both afore-mentioned Toxosoma species refute the presence of two well-delimited forms that might have represented sexual di- morphs (Pig. 5). Documenting morphological change through time does not equate with documenting evolution. Other possible causes for morphological variability, such as ecophenotypy and sexual dimorphism, need to be accounted for. Purthermore, one needs to ascertain that the successive morphs are members of a single lineage. Endemic lineages in fossil long-lived lakes, such as the gastropod genus Toxosoma in Miocene Lake Pebas, comply with the latter assumption. The inferred cladogenetic event giving rise to three Toxosoma species in interval MZ9 may also have resulted from the invasion of species that developed elsewhere in the large (>1 million km^) Pebas system. How- ever, the morphological variability of closely related species is very low within stratigraphic intervals, even in samples from locations far apart (Wesselingh etal. 2006b). The geological record also provides opportunities to consider the potential role of environmental variation in generating morphological variation, and the shells themselves may provide clues about the possibility of dimorphism, making them a very suitable model to study speciation. The three species whose distribu- tional optima are in lacustrine habitats {Toxosoma eboreum, T denticulatum, and T contortuin) developed thick shells with pronounced denticulation. The single marginal lacustrine-to-fluvial species {T grande) has a comparatively thin shell, large apeiiure, and lacks well- developed denticles. The thick shells of the lacustrine species are possibly related to very abundant shell-cracking fish populations in the lake habitats. I'eeth of sciaenid fish, a shell-cracking group, are olten lound in lacustrine samples from the Eebas Formation. A causal relationship is speculative at the moment. Scars ot tailed predatory attacks are common GASTROPOD EVOLUTION IN MIOCENE LAKE PEBAS 89 D (mm) D (mm) Figure 5. Width per half revolution (see Fig. 3 for explanation). Toxosoma eboreum and T. denticulatum have almost identical growth curves, the latter grows only to larger sizes and appears to develop two morphs that may represent sexual dimorphs in later stages of ontogeny. D, diameter. F707 and F685 refer to sample numbers (see Table 1). on Toxosoma shells, however. In the lacustrine species such scars appear to match attacks from crushing predators (exemplified by the damage illustrated in Wesselingh 2007 on Pebasian Pachydon obliquus Gabb, 1869). The outer lip of T. grande is often damaged in a way that suggests the action of predatory crabs. Geary (1990) studied morphological change within species of Melanopsis Ferussac, 1807 in the Late Miocene long-lived Lake Pannon in Austria and Hungary and found anagenetic change followed by a gradual cladogenetic split. The split probably developed over a time interval of approx, one million years and was accompanied by the shift of the two daughter species into different depositional environments. Geary’s (1990) study is one of the very few documentations of gradual evolutionary change in long-lived lakes and shows that fossilized, long-lived lake shells provide great potential for such studies, as is corroborated with the Toxosoma study presented here. CONCLUSIONS In the case of the gastropod genus Toxosoma, an apparent cladogenetic event is documented in the Middle to early Late Miocene intervals of the Pebas Formation. The morphs correspond to species, and sexual dimorphism is found implausible as a possible cause for the observed morphologies. ACKNOWLEDGMENTS Matthias Glaubrecht and Thomas von Rintelen are thanked for organizing the symposium on “Molluscs as evolutionary models” at the World Malacological Congress in Antwerp (Belgium) in 2007 and for their invitation to contribute to this volume. We thank two anonymous review- ers for their comments. In particular, the thorough comments of one of the reviewers were of much help when considering data quality and interpretation as well as the language of an earlier version of this manuscript. LITERATURE CITED Bocxlaer, B. van, D. van Damme, and C. S. Feibel. 2007. Gradual versus punctuated equilibrium evolution in the Turkana Basin molluscs: Evolutionary events or biological invasions? Evolu- tion 62: 511-520. Clarke K. R. and R. M. Warwick. 2001. Change in Marine Communi- ties: An Approach to Statistical Analysis and Interpretation. 2"*^ Edition. PRIMER -E, Plymouth, UK. Frogley, M. R. and R. C. Preece. 2007. A review of the aquatic Mol- lusca from Lake Pamvotis, loannina, an ancient lake in NW Greece. Journal of Conchology 39: 271-295. Geary, D. FI. 1990. Patterns of evolutionary tempo and mode in the radiation of Melanopsis (Gastropoda; Melanopsidae). Paleobi- ology 16: 492-511. Geary, D. H., A. W. Staley, P. Muller, and I. Magyar. 2002. Iterative changes in Lake Pannon Melanopsis reflect a recurrent theme in gastropod morphological evolution. Paleobiology 28: 208-221. Harzhauser, M. and O. Mandic. 2004. The muddy bottom of Lake Pannon - a challenge for dreissenid settlement (Late Miocene; Bivalvia). Palaeogeography, Palaeoclimatology, Palaeoecology 204:331-352. Hoorn, M. C. 1993. Marine incursions and the influence of Andean tectonics on the Miocene depositional history of northwestern Amazonia: Results of a palynostratigraphic study. Palaeogcog- raphy, Palaeoclimatology, Palaeoecology 109; 1-55. 90 AMERICAN M ALACOLOGICAL BULLETIN 27 • I /2 • 2009 Johnson, T. C., C. A. Scholz, M. A. Talbot, K. Kelts, R. D. Ricketts, G. Ngobi, K. Beuning, I. Ssemmanda, and ). W. McGill. 1996. Late Pleistocene desiccation ot Lake Victoria and rapid evolution ot Cichlid fishes. Science 273\ 1091-1093. Martens, K. 1997. Speciation in ancient lakes. Trends in Ecology and Evolution 12; 177-182. Mensink, H. 1984. Die Entwicklung der Gastropoden im miozanen See des Steinheimer Beckens (Siiddeutschland). Palaeonto- graphica A-183: 1-63 [In German]. Michel, E. 1994. Why snails radiate: A review of gastropod evolution in long-lived lakes, both recent and fossil. Archiv fiir Hydrobi- ologie, Beiheft Ergebnisse Limnologie 44: 285-317. Michel, E. 2004. Vinundu, a new genus of gastropod (Gerithioidea; “Thiaridae”) with two species from Lake Tanganyika, East Africa, and its molecular phylogenetic relationships. Journal of Molluscan Studies 70: 1-19. Muller, R, D. G. Geary, and I. Magyar. 1999. The endemic molluscs of the Late Miocene Lake Pannon; Their origin, evolution, and family level taxonomy. Lethaia 32: 47-60. Nevesskaja, L. A., N. P. Paramonova, and S. V. Popov. 2001. History of the Lymnocardiinae (Bivalvia, Cardiidae). Paleontological Journal 35: sl47-s217. Papadopoulos, L. N., J. A. Todd, and E. Michel. 2004. Adulthood and phylogenetic analysis in gastropods: Character recognition and coding in shells of Lavigeria (Gerithioidea, Thiaridae) from Lake Tanganyika. Zoological Journal of the Linnean Society 140: 233-240. Rasanen, M., A. Linna, G. Irion, L. Rebata Hernani, R. Vargas Hua- man and F. Wesselingh. 1998. Geologla y geoformas de la zona de Iquitos. In: R. Kalliola and S. Flores Paitan, eds., Geoecologia y desarollo Aniazdnico: estudio integrado en la zona de Iquito, Peru. Annales universitatis Turkuensis (A, II) 114: 59-137. [In Spanish]. Shaver, R. H. 1953. Ontogeny and sexual dimorphism in Cytherella bidlata. Journal of Paleontology 27: 471-480. Taylor, D. W. 1987. Freshwater molluscs from New Mexico and vicin- ity. New Mexico Bureau of Mines Research Bulletin 116: 1-50. Wesselingh, F. P. 2006. Molluscs from the Miocene Pebas Formation of Peruvian and Colombian Amazonia. Scripta Geologica 133: 19-290. Wesselingh, F. P. 2007. Long-lived lake molluscs as island taunas: A bivalve perspective. In: W. Renema, ed.. Biogeography, Time and Place: Distributions, Barriers and Islands. Springer, Dordrecht, The Netherlands. Pp. 275-314. We.s.selingh, F. P, R. |. G. Kaandorp, H. B. Vonhof, M. E. Ra.siinen, and W. Renema. 2006a. The nature of aquatic landscapes in the Miocene of western Amazonia: An integrated palaeontological and geochemical approach. Scripta Geologica 133: 363-393. Wes.selingh, F. IT M. C. Hoorn, J. Guerrero, M. E. Rasanen, L. Rome- ro Pittmann, and j. Salo. 2006b. The stratigraphy and regional structure of Miocene deposits in western Amazonia (Peru, Co- lombia and Brazil), with implications for l.ate Neogene land- .scape evolution. Scripta Geologica 133: 291-322. Wes.selingh, F. P, M. F. Rasanen, G. Irion, 1 1. B. Vonhof, R. Kaandorp, W. Renema, L. Romero Pittman, and M. Gingras. 2002. Lake Pebas: A palaeoecological reconstruction of a Miocene, long- lived lake complex in western Amazonia. Cainozoic Research 1: 35-81. West, K. and E. Michel. 2000. The dynamics of endemic diversifica- tion: Molecular phylogeny suggests an explosive origin of the thiarid gastropods of Lake Tanganyika. Advances in Ecological Research 31: 33\-373. Williamson, P. G. 1981a. Palaeontological documentation of specia- tion in Cenozoic Mollusks from Turkana Basin. Nature 293; 437-443. Williamson, P. G. 1981b. Morphological stasis and developmental constraint: Real problems for neo-Darwinism. Nature 294: 214-215. Submitted: 9 April 2009; accepted: 13 April 2009; final revisions received: 9 June 2009 Appendix 1. Measurements of Toxosoma. H, height; Hap, height of aperture; WH, number of whorls *100; WBW, width of shell half a whorl before termination (see Fig. 3). Sample Species H Hap WH WBW F685 T. eboreurn 510 260 610 285 F685 T. eboreurn 460 220 580 265 F685 T. eboreurn 465 260 550 275 F685 T. eboreurn 470 240 590 275 F685 T. eboreurn 470 210 605 265 F685 T. eboreurn 440 220 550 255 F685 T. eboreurn 445 210 580 255 F685 T. eboreurn 465 230 605 265 F685 T. eboreurn 485 230 570 285 F685 T. eboreurn 480 230 605 260 F685 T. eboreurn 450 215 560 260 F685 T. eboreurn 495 240 610 280 F685 T. eboreurn 445 220 570 250 F685 T. eboreurn 525 275 595 295 F685 T. eboreurn 475 240 555 290 F685 T. eboreuin 440 215 545 250 F685 T. eboreurn 475 230 580 275 F685 T. eboreuin 430 210 530 245 F685 T. eboreurn 465 240 560 265 F685 T. eboreuin 480 220 620 270 F685 T. eboreuin 475 230 615 275 F685 T. eboreuin 450 225 600 255 F685 T. eboreuin 435 230 560 255 F685 T. eboreuin 500 250 600 275 F685 T. eboreuin 455 230 605 260 F685 T. eboreuin 475 220 600 270 F685 T eboreuin 485 250 595 280 F685 T eboreuin 470 230 595 265 F707 T contortuin 285 135 505 140 F707 T contortuin 335 150 600 160 1707 P. contortuin 285 140 550 140 1707 T coiitortuni 340 135 600 155 1707 T contortuin 310 175 530 170 GASTROPOD EVOLUTION IN MIOCENE LAKE PEBAS 91 Appendix 1. Sample , (Continued) Species H Hap WH WBW Appendix 1. Sample . (Continued) Species H Hap WH WBW F707 T. contortum 290 130 575 140 F535 T. eboreum 370 185 570 205 F707 T. contortum 355 150 580 180 F535 T eboreum 445 240 570 250 F707 T contortum 325 150 605 165 F535 T eboreum 420 205 595 235 F707 T. contortum 380 165 620 185 F535 T eboreum 370 180 550 210 F707 T contortum 300 130 565 155 F535 T eboreum 340 175 520 195 F707 T contortum 305 140 550 155 F535 T eboreum 435 230 560 240 F707 T. contortum 345 160 590 170 F535 T. eboreum 435 215 575 245 F707 T contortum 320 155 560 165 F535 T eboreum 390 190 550 225 F707 T. contortum 305 140 530 155 F535 T eboreum 350 185 510 200 F707 T contortum 315 145 550 160 F535 T eboreum 390 180 595 220 F029 T grande 555 295 520 290 F836 T denticidatum 520 290 655 250 F029 T. grande 535 335 515 295 F836 T. denticidatum 400 240 560 215 F029 T. grande 520 265 495 275 F836 T denticidatum 445 245 650 225 F029 T. grande 595 295 510 340 F836 T denticidatum 435 245 545 245 F029 T grande 610 310 515 340 F836 T. denticulatum 410 210 560 205 F029 T grande 575 310 550 320 F836 T. denticulatum 430 230 560 235 F029 T. grande 550 320 510 325 F836 T. denticulatum 400 235 560 210 F029 T. grande 515 275 500 325 F836 T denticulatum 430 235 640 220 F029 T. grande 505 275 520 295 F836 T denticulatum 470 205 580 240 F029 T grande 515 290 505 290 F367 T. denticulatum 485 260 605 240 F032 T grande 485 250 515 265 F367 T denticulatum 505 280 615 255 F032 T. grande 580 335 590 300 F367 T. denticulatum 455 250 610 220 F032 T grande 560 290 580 305 F367 T denticulatum 415 255 590 215 F032 T. grande 500 290 575 265 F367 T denticulatum 455 255 600 250 F032 T. grande 590 325 595 320 F367 T. denticulatum 430 240 560 215 F032 T grande 560 315 565 295 F367 T. contortum 330 170 545 165 F032 T. grande 550 310 525 295 F367 T. contortum 340 160 560 175 F032 T grande 520 295 525 275 F367 T contortum 340 160 600 170 F032 T grande 550 310 550 300 F367 T. contortum 310 150 545 150 F032 T grande 575 320 585 300 F367 T contortum 325 145 550 160 F032 T grande 600 350 555 330 F367 T contortum 330 140 585 165 F032 T. grande 495 265 495 275 F367 T contortum 410 180 645 190 F032 T. grande 550 310 525 310 F367 T contortum 325 160 580 160 F032 T grande 495 245 560 260 F367 T. contortum 340 145 580 165 F032 T. contortum 375 170 575 190 F367 T contortum 310 145 615 155 F032 T. contortum 350 160 580 185 F367 T contortum 300 150 565 150 F032 T contortum 370 125 590 175 F367 T contortum 325 150 560 160 F032 T contortum 380 140 590 180 F367 T. contortum 350 175 560 175 F032 T contortum 380 120 610 175 F367 T contortum 315 145 545 165 F032 T contortum 350 125 585 165 F367 T contortum 335 160 560 175 F032 T. contortum 340 115 575 175 F367 T contortum 335 150 565 165 F032 T contortum 375 115 575 175 F367 T. contortum 350 155 595 175 F032 T contortum 350 140 570 175 F367 T contortum 335 155 590 160 F032 T. contortum 340 135 570 165 F367 T contortum 315 145 555 160 F4I7 T. eboreum 455 230 570 255 F367 T contortum 300 140 540 155 F417 T eboreum 435 215 555 255 F034 T denticulatum 460 255 555 235 F535 T. eboreum 425 240 560 250 F034 T denticulatum 520 295 575 270 F535 T eboreum 360 185 550 200 F034 T denticulatum 415 220 505 245 F535 T eboreum 420 215 585 235 F498 T eboreum 415 205 605 230 F535 T eboreum 390 195 525 230 F498 T eboreum 405 200 570 220 F535 T. eboreum 355 175 545 195 F498 T. eboreum 355 180 530 205 92 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Appendix 1. (Continued) Sample Species H Hap WH WBW Appendix 1. Sample , (Continued) Species H Hap WH WBW F498 T. eboreum 390 185 555 205 F489 T. eboreum 460 200 570 250 F498 T eboreum 405 195 575 215 F489 T eboreum 485 205 605 250 F498 T. eboreum 385 170 570 185 F022 T contortum 345 140 545 160 F498 T eboreum 410 200 595 215 F022 T. contortum 350 150 565 175 F498 T eboreum 435 220 570 240 F022 T. contortum 335 155 570 165 F498 T eboreum 460 200 620 245 F022 T. contortum 340 150 550 180 F498 T. eboreum 395 170 590 180 F022 T contortum 345 150 575 165 F498 T. eboreum 350 165 510 180 F022 T contortum 350 150 575 155 F498 T eboreum 400 180 555 195 F022 T contortum 350 165 575 170 F498 T. eboreum 420 230 585 245 F022 T contortum 335 135 575 160 F498 T. eboreum 430 220 610 230 F022 T contortum 335 150 550 165 F498 T. eboreum 370 175 540 200 F022 T. contortum 320 135 555 165 F498 T. eboreum 380 195 590 220 F022 T. contortum 330 135 560 170 F498 T. eboreum 380 160 545 190 F022 T contortum 315 145 545 150 F498 T. eboreum 370 170 545 205 F022 T contortum 300 145 545 155 F498 T eboreum 375 190 560 200 F022 T contortum 300 155 525 155 F498 T eboreum 405 200 560 240 F022 T contortum 330 155 555 160 F498 T eboreum 410 195 590 215 F004 T grande 460 255 495 270 F498 T eboreum 340 150 545 180 F004 T grande 515 280 545 270 F498 T. eboreum 400 190 570 190 F004 T grande 420 245 500 240 F498 T. eboreum 360 170 535 180 F004 T grande 450 250 505 255 F498 T. eboreum 385 165 550 190 F004 T grande 445 240 500 250 F498 T. eboreum 350 170 520 180 F702 T. grande 420 210 465 220 F489 T eboreum 490 235 615 250 F702 T grande 475 220 480 250 F489 T. eboreum 395 185 545 210 F702 T grande 430 180 465 220 F489 T eboreum 455 200 540 255 F702 T. grande 435 195 475 235 F489 T eboreum 445 210 550 225 F702 T grande 460 240 480 235 F489 T. eboreum 470 210 565 250 F702 T grande 400 195 460 215 F489 T. eboreum 420 195 555 220 F702 T. grande 415 180 465 220 F489 T. eboreum 495 215 575 255 F702 T grande 405 190 455 215 F489 T. eboreum 500 220 550 270 F702 T grande 425 205 475 225 F489 T eboreum 460 215 515 255 F702 T. grande 415 200 475 225 F489 T eboreum 500 220 510 275 F702 T grande 380 185 480 215 F489 T. eboreum 470 205 535 255 F702 T. grande 450 225 465 235 F489 T eboreum 535 230 640 265 F702 T grande 435 205 485 230 F489 T. eboreum 475 230 550 260 F702 T. grande 405 185 460 215 F489 T. eboreum 505 245 545 290 F702 T. grande 405 220 480 225 F489 T eboreum 485 225 545 260 F702 T grande 470 215 505 250 F489 T. eboreum 495 210 600 250 F702 T grande 420 185 475 210 F489 T. eboreum 400 180 530 210 F702 T. grande 470 220 460 245 F489 T. eboreum 500 220 555 265 F702 T grande 420 195 445 240 F489 T. eboreum 480 220 545 270 F363 T eboreum 360 180 560 220 F489 T eboreum 495 210 555 270 F363 T. eboreum 370 160 560 215 F489 T. eboreum 525 240 550 290 F363 T. eboreum 335 175 550 210 F489 T. eboreum 405 190 515 200 F363 T. eboreum 340 170 540 205 R89 T. eboreum 435 180 560 215 F363 'll eboreum 360 180 575 200 F489 T. eboreum 460 190 590 235 F363 T. eboreum 330 165 550 195 F489 I', eboreum 510 250 560 280 F363 'll eboreum 325 150 525 190 F'489 T. eboremu 505 220 595 255 F363 'll eboreum 315 170 515 185 F489 T eboreum 450 205 545 235 F'363 'll eboreuni 360 180 560 180 J-489 1. eboreum 510 225 560 280 l'363 'll eboreum 355 165 560 175 GASTROPOD EVOLUTION IN MIOCENE LAKE PEBAS 93 Appendix 1. (Continued) Sample Species H Hap WH WBW F363 T. eboreum 385 190 565 170 F363 T. eboreum 320 145 530 165 F363 T eboreum 370 160 570 160 F363 T eboreum 330 155 510 155 F363 T. eboreum 340 160 520 150 F006 T. denticulatum 550 290 580 285 F006 T. denticulatum 520 280 560 290 F006 T. denticulatum 490 260 540 260 F006 T denticulatum 525 280 550 285 F006 T denticulatum 485 270 545 265 F016 T grande 430 220 510 250 F016 T. grande 495 245 505 270 F016 T spec. 390 205 550 200 F016 T. grande 490 265 525 270 F016 T grande 505 260 505 290 F021 T grande 420 220 480 240 F830 T eboreum 345 160 550 180 F830 T eboreum 355 160 555 205 F830 T. eboreum 335 160 555 190 F830 T eboreum 345 160 560 180 F830 T eboreum 320 165 510 190 F830 T. eboreum 320 160 545 180 F830 T. eboreum 360 165 555 200 F830 T eboreum 365 170 560 195 F830 T eboreum 340 160 555 190 F830 T eboreum 350 155 550 180 F830 T eboreum 340 160 555 195 F830 T eboreum 355 160 550 195 F830 T. eboreum 365 160 565 205 F830 T eboreum 390 170 560 200 F830 T. eboreum 350 170 555 200 F830 T. eboreum 330 155 555 180 F830 T. eboreum 335 155 555 180 F830 T. eboreum 365 170 575 195 F830 T. eboreum 375 175 575 195 F830 T eboreum 380 170 605 205 F830 T. eboreum 340 160 565 185 F830 T. eboreum 305 145 510 185 F830 T eboreum 350 175 540 195 Amer. Maine. Bull. 27: 95-104 (2009) Morphological cladistic analysis as a model for character evaluation in primitive living chitons (Polyplacophora, Lepidopleurina)^ Julia D. Sigwart National Museum of Ireland, Natural History Division, Merrion Street, Dublin 2, Ireland and Queen’s University Belfast, School of Biological Sciences, University Road, Belfast BT7 INN, Northern Ireland, U.K. Corresponding author: julia.sigwart@ucd.ie Abstract: Chitons are often referred to as “living fossils” in part because they are proposed as one of the earliest-diverging groups of living molluscs, but also because the gross morphology of the polyplacophoran shell has been conserved for hundreds of millions of years. As such, the analysis of evolution and radiation within polyplacophorans is of considerable interest not only for resolving the shape of pan-molluscan phylogeny but also as model organisms for the study of character evolution. This study presents a new, rigorous cladistic analysis of the morphological characters used in taxonomic descriptions for chitons in the living suborder Lepidopleurina Thiele, 1910 (the earliest-derived living group of chitons). Shell-based characters alone entirely fail to recover any recognized subdivisions within the group, which may raise serious questions about the application of fossil data (from isolated shell valves). New analysis including characters from girdle armature and gill arrangements recovers some genera within the group but also points to the lack of monophyly within the main genus Leptochiton Gray, 1847. Additional characters from molecular data and soft anatomy, used in combination, are clearly needed to resolve questions of chiton relationships. However, the data sets currently available already provide interesting insights into the analytical power of traditional morphology as well as some knowledge about the early evolution and radiation of this group. Key words: morphology, molluscan evolution, cladistics, Leptochiton Phylogenetic studies have shown that chitons (Polypla- cophora) retain many features that are plesiomorphic within molluscs and indeed appear to have close morphological similarity with the hypothesized common ancestor of the Mollusca {e.g., Haszprunar 1996, Sigwart and Sutton 2007; Fig. 1 A). Nevertheless, given the radical disparity encompassed by molluscan morphology, understanding the patterns of evolution within the Polyplacophora is of particular im- portance to Lindertand how a bauplan that has remained so conserved over hundreds of millions of years in the direct living descendants of the ancestral chiton may have also given rise to forms as different as bivalves and cephalopods. The living chitons are contained in the subclass Neoloricata, which has a fossil record extending from the Carboniferous (ca. 350 Mya) to Recent. However, in spite of this deep fossil record, the majority of fossil species are known only from isolated, disarticulated plates, which makes it very difficult to infer the morphology of the whole animal (Cherns 2004, Vendrasco et al. 2004, Sigwart and Sutton 2007). The three major clades within the Neoloricata (Lepidopleurida, Chitonina, and Acanthochitonina) are separated stratigraphically in the fossil record as well as morphologically, which has resulted in a broad acceptance of their monophyly and phylogenetic separation at the ordinal level (Buckland-Nicks 1995, Okusu etal. 2003). The suborder Lepidopleurina is uncontroversially accepted to be the earliest-derived group of living chitons (Fig. IB). In spite of this broadly accepted separation, to date there is limited understanding of the relationships between taxa within the major clades of chitons and not a robust testing of the monophyly of these groups across the broad range of taxa they each cover. Living members of the order Lepidopleurida (suborder Lepidopleurina) include nine genera comprising five families {sensu Sirenko 2006); but the genus Leptochiton Gray, 1847 contains about 80 of the 120 valid extant species in the order. Complex shell and girdle features that are used to define non-lepidopleuran chitons are often lacking in these species, which are small, plain, and superficially homogenous in their external morphology. However, species are described using the same conventional features across all chitons — shell sculpture, girdle armature microstructure, gills, and gross radular morphology. In addition, several morphological features, particularly from soft anatomy, have been proposed to assess the intra-clade relationships of chitons, particularly sperm acrosome and egg hull structures (Hodgson et al. 1988, Sirenko 1993, Buckland-Nicks 1995, 2006), relative positions of gonopores Mrom the symposium “Molluscs as models in evolutionary biology: from local speciation to global radiation” presented at the World Con- gress of Malacology, held from 15 to 20 July 2007 in Antwerp, Belgium. 95 96 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Solenogastres Caudofoveata Polyplacophora Monoplacophora Bivalvia Scaphopoda Gastropoda Cephalopoda > n c B Solenogastres Caudofoveata Paleoloricata Multiplacophora Lepidopleurida Chitonina Acanthochitonina (D o n QJ CD Figure 1. A, generalized topology of the aculiferan model of mol- luscan evolution, showing relationships between living classes of molluscs. B, generalized topology of major clades within the Aculif- era (redrawn from Sigwart and Sutton 2007). All living chitons are classified in the subclass Neoloricata, in the two orders Chitonida (containing two suborders, Chitonina and Acanthochitonina) and Lepidopleurida (the focus of the present study). and nephridiopores within the gill row (Sirenko 1993, Sigwart 2008b), and patterns of aesthete canals within the valves (Fernandez et al. 2007, Vendrasco et al. 2008). The aim of this paper is to empirically assess the usefulness of the morphological features that are currently used to define species, by testing their ability to build phylogenetic hypotheses within and between the taxa that comprise this clade. This is accomplished here by using a subset of exemplar lepidopleuran taxa. The hypothesis tested is that if the characters used to diagnose and to describe species were defined expressly to separate species, genera, and other taxonomic subsets, then the expected results of a c]uantitative phylogenetic analysis of those characters would be a tree replicating the established Linnaean hierarchy. In tact, many non-diagnostic characters are regularly described for chitons, and the accepted diagnoses of genera in particular are fraught with inconsistency. Consequently, this study afso examines the information content of morphological characters. MATERIALS AND METHODS Characters and taxa This study draws on morphological examination by the author of preserved specimen material, particularly from the collections of the Museum national d’Histoire naturelle (MNHN, Paris); Zoologische Staatssammlung Miinchen (ZSM, Munich), Naturhistorisches Museum in Wien (NHMW, Vienna); and the National Museum of Ireland, Natural History Division (NMINH, Dublin). Sixty-nine morphological characters were described based on features used in every modern species description, derived from a standard implicitly set by the work of Kaas and Van Belle (1985). Character states are defined based on features identified in the literature and from the author’s personal observation of specimen morphology. Codings for the present morphological matrix are based on original anatomical observation by the author. In principle, these character states are defined based on variability that is morphologically constant for all individuals within a species but may vary, encompassing the different states defined, in different species. The characters include 32 shell features, 27 girdle features, two aspects of gross body shape, two radula characters, and six characters of gill arrangement (Table 1). Part of this data set was based on the matrix of shell features published by Sigwart et al. (2007), incorporating new refinements to the characters used in that analysis. Characters were described with two (binary), three, or four potential character states. All characters were considered to be unordered (that is, evolutionary change could hypothetically transform freely between any of the described states) and without polarity (no ancestral state is assumed). All characters were equally weighted for the analysis. The complete character matrix is presented in Table 2. The characters were coded for 39 ingroup taxa selected to cover the morphological diversity of the living Lepidopleurina. Two species of Callochiton and the monotypic Clioriplax grayi (H. Adams and Angas, 1864) were coded as outgroup taxa. I'he genus Callochiton resolved as sister to the family Leptochitonidae in a molecular phylogenetic analysis (Okusu et al. 2003). This is a controversial result which has not been supported by other studies (Buckland-Nicks 2006, Sirenko 2006, B. Lieb, pers. comm.). However, as there is so little information about the internal topology of any of the major polyplacophoran clades, this is the best evidence for choosing a most proximal outgroup from living taxa. Choriplax has traditionally been classified within the Lepidopleurina but has recently been considered by Sirenko (2006) to be an advanced form with .secondarily derived leatures in common with lepidopleurans. Phylogenetic analysis, con.sensus, and support The complete matrix was initially subjected to a permu- tation tail probability test (PTP) to a.s.sess data quality, using MORPHOLOGICAL CLADISTICS WITH PRIMITIVE LIVING CHITONS 97 Table 1. Characters formulated from external morphology and used to code chiton taxa in the phylogenetic analysis. Cl (confidence index) from main analysis of complete matrix. Character description Cl 1 Ratio: apophyses outside diameter / valve width: <0.8 (0); >0.8 (1). 0.125 2 Ratio: combined diameter of apophyses / valve width: <0.4 (0); >0.4 (1). 0.125 3 Thickened on terminal margins: no (0); yes ( 1 ). 0.250 4 General character of arch (intermediate plates): straight sides (0); rounded ( 1 ); concave (2). 0.333 5 Dorsal elevation (height / width) of intermediate plates: <0.4 (0); >0.4 (1). 0.200 6 Valves beaked: no (0); yes (1). 0.111 7 Lateral area elevated on intermediate plates: no (0); yes ( 1 ). 0.091 8 Intermediate plates with distinct diagonal separating lateral areas: no (0); yes ( 1 ). 0.077 9 Apophyses’ jugal margin: straight (0); concave (1). 0.091 10 Head valve shape: semicircular (0); shape < semicircle (1); shape > semicircle (2). 0.222 1 1 Tail valve shape: semicircular (0); shape < semicircle ( 1 ); shape > semicircle (2). 0.286 12 Mucro prominent: no (0); yes (1). 0.100 13 Mucro position: posterior (0); median (1); anterior (2). 0.095 14 Post-mucronal slope: straight (0); concave (1); convex (2). 0.125 15 Articulamentum character: weak, transparent (0); moderate, opaque at least in major central areas (1); 0.182 strong, forming thickened calluses and/or extending into insertions (2). 16 Insertion plates present: no (0); yes (1). 0.333 17 Tail valve apophyses shape same as on intermediate valves: yes (0); no, difference(s) ( 1 ). 0.077 18 Intermediate valve shape: trapezoidal (0); rectangular (1); ovate or circular (2). 0.400 19 Intermediate valves: anterior margin: straight (0); concave (1); convex (2). 0.167 20 Intermediate valves: posterior margin: straight (0); convex (1); concave around apex (2). 0.182 21 Tegmentum — general sculpture: smooth (0); granulose (1); pustulose or joined (2). 0.250 22 Tegmentum — gradation of sculpture: regular (0); larger toward margin (1); faded posteriorly (2). 0.333 23 Tegmentum — dominant granule shape: no granules (0); roundish ( 1 ); square or irregular (2). 0.250 24 Central areas of intermediate valves — distinct jugal sculpture: not distinct from pleural area (0); longitudinally 0. 167 granulate in jugal area; pleural areas coarser (1); jugal sculpture in quincunx, grading to radiating longitudinal or diagonal series (2). 25 Intermediate valves — sculpture with longitudinal rows: no pattern (0); longitudinal rows (1); quincunx or 0.214 diagonal series (2); irregular, wavy, or zigzag (3). 26 Intermediate valves — sculpture interstices: close / narrow or sandy (0); evenly distributed, series of granules 0.1 18 may be coalescing / beading ( 1 ); widespread or punctured (2). 27 Areas of intermediate plates with corresponding sculpture as terminal plates: no (0); yes ( 1 ). 0.200 28 Ratio: apophyses inside separation / valve width: <0.4 (0); >0.4 ( 1 ). 0. 100 29 Thick periostracum forming pustules: absent (0); present (1). 1.000 30 Valve carinated or keeled: no (0); semi-carinated or keeled at posterior (1); yes (2). 0.133 31 Lateral area sculpture in radiating rows: no pattern or quincunx (0); strong radiating rows (1); central sculpture 0.300 continuing in longitudinal rows (2); wavy or zigzag (3). 32 Tail valve more than twice the length of intermediate valves: no (0); yes ( 1 ). 0.500 33 Dorsal girdle scales — length / width ratio of typical scale (total length / width at base): absent (0); square (1); 0.200 length >1.5 X width (2); width >1.5 X length (3). 34 Dorsal girdle scales: absent (0); straight (1); distally curved (2). 0.333 35 Dorsal girdle scales or spicules — generalized density: “dense” (1); unremarkable (2); “sparse” (3). 0.133 36 Dorsal girdle scales — distribution: absent (0); regular (1); irregular (2). 0.200 37 Dorsal girdle scales or spicules — lateral size gradient: no, equal at all points (0); smaller toward margin (1); 0.167 larger toward margin (2). 38 Dorsal girdle scales — shape: absent (0); square top ( 1 ); round top (2); pointed (3). 0.250 39 Dorsal girdle scales — texture: absent (0); smooth to weakly ribbed or striated ( 1 ); distinct ribs, more than 10 (2). 0.273 98 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Table 1. (Continued) Character description Cl 40 Ventral girdle scales — length relative to typical dorsal scale: absent (0); ec]ual length (1); shorter, half or less length (2); 0.250 longer, 1.5 times length or more (3). 41 Ventral girdle scales — width relative to typical dorsal scale: absent (0); equal width ( 1 ); narrower half or less length (2); 0.250 wider 1.5 times length or more (3). 42 Ventral girdle scales — texture: absent (0); smooth to weakly ribbed or striated (1); distinct ribs, 10 or less ( 1 ); distinct ribs, 0.286 more than 10(2). 43 Ventral girdle scales or spicules: absent (0); straight ( 1 ); distally curved (2). 0.667 44 Ventral girdle scales or spicules — shape: absent (0); elongate oval, round top or pointed scales (1); rectangular scales (2); 0.375 spiculose (3). 45 Ventral girdle scales or spicules — generalized density: “dense” (1); unremarkable (2); “sparse” (3). 0.167 46 Ventral girdle scales or spicules — distribution: absent (0); regular ( 1 ); irregular (2). 0.286 47 Ventral girdle scales or spicules — lateral size gradient: no, equal at all points (0); smaller toward foot ( 1 ); larger toward 0.500 foot (2). 48 Dorsal spicules, including intersegmental armature — length relative to major dorsal scales: absent (0); spicules form major 0.429 dorsal coverage, scales absent ( 1 ); dorsal or intersegmental spicules not substantially longer than major dorsal scales (2); dorsal or intersegmental spicules substantially longer, three times length of dorsal scales or longer (3). 49 Dorsal spicules, including intersegmental armature — shape: absent (0); straight ( 1); distally curved (2). 0.500 50 Dorsal spicules, including intersegmental armature — terminal shape: absent (0); sharp, pointed ( 1 ); blunt (2). 0.182 51 Intersegmental spicules — complex base: absent or simple (0); spicules in chitinous cupules (1); spicules in complex 0.167 rigschaftnadel (2). 52 Dorsal spicules, including intersegmental armature — texture: absent (0); smooth, strongly tapered ( 1 ); striated or 0. 154 grooved (2); smooth, cylindrical (3). 53 Marginal scales / spicules — length relative to typical dorsal armature: absent or not differentiated (0); same length ( 1 ); 0.200 shorter (2); substantially longer, three times length or longer (3). 54 Marginal scales / spicules — width relative to typical dorsal armature: absent or not differentiated (0); same width ( 1 ); 0.182 narrower (2). 55 Marginal scales/ spicules — texture: absent (0); smooth, strongly tapered (1); striated or grooved (2); smooth, cylindrical (3). 0.176 56 Marginal spicules — shape: absent (0); straight (1); distally curved (2). 0.222 57 Marginal spicules — terminal shape: absent (0); pointed, sharp (1); blunt (2). 0.154 58 Dorsal sutural spines differentiated from major armature: no (0); yes ( 1 ). 0.077 59 Marginal spicules — complex base: absent or simple (0); spicules in chitinous cupules ( 1 ); spicules in complex 0. 1 82 rigschaftnadel (2). 60 Ratio: animal body length / width: <2 (0); >2 (1). 0.063 61 Maximum adult body length (mm): <10 mm (0); >10 mm (1). 0.125 62 Radula — number of cusps on major lateral teeth: monocuspidate (0); bicuspidate ( 1 ); tricuspidate (2). 0. 1 54 63 Radula — smallest cusp on major lateral teeth: monocuspidate or equal (0); anterior / interior ( 1 ); posterior / exterior (2); 0.2 14 central, tricuspid with two outer denticles equal (3). 64 Gills — number per side in adult animal: <4 (0); 5-10 ( 1 ); 1 1-16 (2); > 1 6 (3). 0.167 65 Gills — size distribution: largest at posterior, or all equal (0); smallest anterior and posterior ( 1 ). 0. 100 66 Gills interrupted by anal interspace: yes (0); no, gills on anal stem ( 1 ). 0. 125 67 Gills “merohranchial”, posterior: yes (0); no, other arrangement ( 1). 1.000 68 Ciill coverage, as percentage of foot length: <40% (0); >40% ( 1 ). 0.200 69 Gills abanal (single gill posterior of nephridiopore in adult animal): no (0); yes ( I ) 0.500 MORPHOLOGICAL CLADISTICS WITH PRIMITIVE LIVING CHITONS 99 Table 2. Character-taxon data set utilized for the phylogenetic analysis. Asterisks (*) denote taxa designated as outgroups. Ingroup taxa are listed in alphabetical order by species epithet. For some characters, transitional states are coded r = {0,1 ); t = {1,2}. 1 2 3 4 5 6 123456789012345678901234567890123456789012345678901234567890123456789 *Callochiton Fouveh Thiele, 1906 *Callochitori septemvalvis (Montagu, 1803) *Choriplax grayi (H. Adams and Angas, 1864) Parachiton acuminiatus (Theile, 1909) Leptochiton aeqiiispinus (Bergenhayn, 1933) Leptochiton algesirensis (Capellini, 1859) Leptochiton alveolus (Sars MS, Loven, 1846) Leptochiton arnericanus Kaas and Van Belle, 1985 Nierstraszella andamanica (Smith, 1906) Leptochiton asellus (Gmelin, 1791) Leptochiton binghami (Boone, 1928) Leptochiton boucheti Sirenko, 2001 Lepidopleurus cajetanus (Poll, 1791) . Leptochiton cancellatus (Jeffreys, 1839) Parachiton communis Saito, 1996 Deshayesiella curvata (Capenter MS, Pilsbry, 1892) Leptochiton deforgesi Sirenko, 2001 Leptochiton denhartogi Strack, 2003 Leptochiton foresti (Leloup, 1981) Leptochiton hirasei (Taki and Taki, 1929) Leptochiton inquinatus (Reeve, 1847) Leptoch i to n i n termed i us (Salvini-Plawen, 1968) Leptochiton japonicus (Theile, 1909) Leptochiton juvenis (Leloup, 1981) Leptochiton kerguelensis Haddon, 1886 Nierstraszella lineata (Nierstrasz, 1905) 110001111010201101201010101002000010100331112??1110111122010122300111 110001111000001101221010201002000010100221112721110132121010122300111 111100010220222111212020111001102132?31????????????????????1120?????1 000100110010010001001 02 011110?rl2 122732 111112? ?2 11 02 31121001120???0?0 01 02 0000100020200110101 02 Oil 02 0021777221 12 11777000001131200010 02 11070 1 1 01 00 01 010020 701 1001 0701 71 10?r01222 112221 1L2222 121 1112 1200011020 1010 1001100000001170171010100771010021211721 12 11271211 02 77777101100271070 110000017100012012001011171001102117731111117172110177777070110111070 1110010010000r2 Oil 02 000022 1012000012 0001111312 71 12020000010111 0211010 110 70101101101 ?0??2220101?0101r022121222211221??????121110001011?1000 777071107771217777021011111? 02 1021171313111 32 723110100000070170777070 000100000000210011101010221100000012100311121221122231311120121000000 0001 701 1 10 7122 70 7 70 1 00 1012 110020 11120 7232111 7 122111tl22 11 00 11202 710 00 110100111111210011001010101107101222112121122122111177777101007111000 010100110020001011001020101000012121733111112177777700000001123311000 01010111 02 012 02 007 02 17711000002021271 32211112172110112111071110111000 100100000001110011001020221100001121122312112122122111312121021000000 110100111111210011001010111100101221112111112122110111211000000111000 1110010010001020010210101210 02 10001200011113717112 02 11221010120117010 000 1 000 1 10 7 12 1000200101 71 ?0000r021 17 711 121 1121 72 11 Oil 12 1101 10201 17070 01 01 001 1 100000001 1 0010 701 71 000r01211 1231211 12 72212 12 112 Itl 71 1001 7 1000 10010001000001100100101012110000???????????????????????????1??????0?0 010200001000202001101010201102002177722112117770000011312 0001002 11000 000001010000201011001112101102002111112121112122120012311110123211000 00011011000011001700001 12 710000021171 12 mill 122 11 12 77777100010117070 llOrOOOlOlOOO 12011110 72 03110113000 120001111312 71 120200000101110211010 100 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Table 2. (Continued) Leptochiton /nedinne ( Plate, 1899) Hanleya nagelfar (Loven, 1846) Leptochiton n. sp. 4 Sigwart, 2008a Leptochiton n. sp. 5 Sigwart, 2008a Oldroydia percrassa (Dali, 1894) Leptochiton pergranatus Dali, 1889 Ferreimella plana (Nierstrasz, 1905) Parachiton politus Saito, 1996 Leptochiton rugatus (Capenter MS, Pilsbry, 1892) Leptochiton saitoi Sirenko, 2001 Leptochiton scabridus (Jeffreys, 1880) Leptochiton thandari Sirenko, 2001 Leptochiton vanbellei Sirenko, 2001 Leptochiton vaubani Kaas, 1991 Leptochiton vietnamensis Sirenko, 1998 Ferreimella xylophaga Gowlett- Holmes and Jones, 1992 1 2 3 4 5 6 123456789012345678901234567890123456789012345678901234567890123456789 000100010?0100001100101110100010112?132112111 122220211211000110111000 11001001000010000020000000?000000020000121132??1110100000001120311010 000100010001200011201011111000002121122331112123122100000100021000000 000000000000210011001112221101002111122111111123122131112120121000000 1100010011000 12 01222?0111?1001100010?001311121211101?????00111021100? ??00000110?021100? 02 1010111? 02r0212?l 32 3311312231101000000?01?02??000 11 020100102 10 12 012 11101 0207101000012 000000002001 12 02 112 120?01203??0?0 01 010010002 000 1001001020 101 0000 12 121 722211 1121 72120200000001 12331 1000 77010011 7101 17001 10020701 71 ?00r012??lt311211?l?22202112221??1001?1000 001000010000201011001011111002000022000000032221211111111110023111000 000100011007110001201110170001101127113122112722111177777170020171000 100000001000000011001011110101002121121111112122121111111120021000000 100110000001100001001010221102000022000111212221122111311121121000000 110011000000110011101012211002202122031111112222110100000100111300010 010001010000201001001010101000000022100000032221111111111110011211000 100000000200102112110000021102000020000000002001110111111111123301010 50 permutation replicates (Laith and Cranston J991). Having determined that the data differ significantly from random, they were then analyzed in the standard software package PAUL* version 4.0b 10 (Swofford 2002), using a heuristic search algorithm. To improve the efficiency of analysis, an initial heuristic search was performed with ten random addition sequence replicates to determine the size {n steps) of the shortest common most parsimonious trees (MPTs). The analysis was then repeated with 1 00 random addition sequence replicates, limiting the search to trees of length n or less. All trees of minimum length were retained as the primary set of MPTs to a maximum of 3000 MPTs. To assess the relative degree of clade support, bootstrap values were calculated from the analysis, using 50 bootstrap replicates of 10 random addition sequence replicates each. Strict consensus cladograms can show dramatic reduction of resolution when the positions of a few ingroups are highly mobile ie.g., Wilkinson 1999). Reduced consensus methods can recover additional ingroup relationships by removing the.se unresolved taxa. It is important to note that taxa are pruned only after the analysis is performed using the complete data matrix so no data are selectively “ignored”. This study employed the “Strict” program in the software package RedCon version 3.0 (Wilkinson 2001) to produce a set of strict reduced consensus trees. Two methods were employed to examine the influence of the two major data partitions (shell characters and girdle characters). Pirst, consensus index (Cl) values for the two data sets were compared from within the main analysis (the value for each character of the minimum number of potential state changes, divided by the minimum number actually observed on the tree). Two secondary analyses were run on restricted sections of the matrix (one with shell characters only, one with girdle characters alone) with the .same parameters and protocol as the main analysis (initial PTP lest and heuristic search protocol in PAUL'* outlined above). RESULIS The permutation tail probability calculated for the matrix was 0.02, which is significant (P < 0.05), indicating that the matrix should contain a resolvable phylogenetic signal. The phylogenetic analysis resulted in 68 MPTs, each of length 517 MORPHOLOGICAL CLADISTICS WITH PRIMITIVE LIVING CHITONS 101 Leptochiton algesirensis Leptochiton asellus Leptochiton cancellatus Leptochiton denhartogi Lepidopleurus cajetanus Leptochiton hirasei Leptochiton scabridus Leptochiton aequispinus Leptochiton japoniciis Leptochiton alveolus Parachiton communis Parachiton politus Parachiton acuminiatus Leptochiton boucheti Deshayesiella curvata Leptochiton deforgesi Leptochiton inquinatus Leptochiton intermedius Leptochiton juvenis Leptochiton kerguelensis Leptochiton medinae Leptochiton n. sp. 4 Leptochiton n. sp. 5 Leptochiton rugatus Leptochiton thandari Leptochiton vanbellei * Choriplax grayi Leptochiton americanus Leptochiton binghami Leptochiton pergranatus Leptochiton vaubani Nierstraszella andamanica Nierstraszella I i neat a Ferreiraella plana Ferreiraella xylophaga Leptochiton saitoi Leptochiton vietnamensis Leptochiton foresti Hanleya nagelfar Oldroydia percrassa * Callochiton houveti * Callochiton septemvalvis B C d d d Leptochiton boucheti Leptochiton vanbellei Leptochiton deforgesi Leptochiton n. sp. 5 Leptochiton n. sp. 4 Leptochiton thandari Leptochiton aequispinus Leptochiton japonicus Leptochiton alveolus Leptochiton juvenis Leptochiton algesirensis Leptochiton asellus Leptochiton cancellatus Leptochiton denhartogi Lepidopleurus cajetanus Leptochiton hirasei Leptochiton scabridus Leptochiton inquinatus Leptochiton kerguelensis Leptochiton medinae Leptochiton rugatus Parachiton communis Parachiton politus Parachiton acuminiatus * Choriplax grayi Leptochiton americanus Leptochiton binghami Leptochiton pergranatus Leptochiton vaubani Nierstraszella andamanica Nierstraszella llneata Ferreiraella plana Ferreiraella xylophaga Leptochiton saitoi Leptochiton vietnamensis Leptochiton foresti Hanleya nagelfar Oldroydia percrassa * Callochiton houveti * Callochiton septemvalvis Figure 2. Phylogenetic trees resulting from analysis of morphological characters for Lepidopleurida. Asterisks (*) denote taxa designated as outgroups. A, strict consensus of 68 MPTs; bootstrap values are marked on internal nodes (where the bootstrap value is over 50). B, preferred tree, strict reduced consensus tree with Leptochiton intermedius and Deshayesiella curvata pruned from trees contributing to Fig. 1 A. Major in- ternal clades discussed in the text are marked with vertical lines to the right: predominantly Pacific Leptochiton spp. (top, Leptochiton boucheti- Leptochiton juvenis); predominantly Atlantic Leptochiton spp. (lower, Leptochiton algesirensis-Leptochiton scabridus); major Leptochiton clade including Parachiton {Leptochiton boucheti-Parachiton acuminatus). steps (Fig. 2A). The strict consensus of these trees recovers a monophyletic Lepidopleurida including Choriplax, relative to the Callochiton outgroup (Fig. 2A). Whether Choriplax is designated an outgroup taxon or an ingroup taxon does not change the topology of the resulting trees. Un-rooted trees also resolve the same topology in all cases. Within the ingroup, a single major clade containing most Leptochiton species (21 of 28 taxa included in the analysis), including the type species Leptochiton asellus (Gmelin, 1791) as well as Lepidopleurus Risso, 1826, Parachiton Thiele, 1909, and Deshayesiella Carpenter in Dall, 1879. The remaining lepidopleurans are arranged as a grade sister to this main clade. Strong sister- taxa relationships resolve Nierstraszella Sirenko, 1992 and Ferreiraella Sirenko, 1988 as monophyletic (each genus represents its own family). The monophyly of Lepidopleurina is supported by three synapomorphies: merobranchial gill arrangement (unique to Lepidopleurina), adanal gill place- ment (although this is reversed in Choriplax), and lack of shell insertion plates (although this is reversed in Ferreiraella as well as in Choriplax). 102 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Reduced consensus analysis resulted in a set of seven additional strict reduced consensus trees; four excluded a single taxon from the trees, two trees excluded two taxa, and one excluded nine taxa. Pruning Leptochiton intermedins (Salvini- Plawen, 1968) produced better resolution within the main Leptochiton clade. However, other permutations recovered in these seven SRC trees did not change the topology of the cladogram. The two SRC trees that pruned two taxa both excluded L. intermedins and one of Parachiton acnrninatns (Theile, 1909) or Deshayesiella cnrvata (Capenter MS, Pilsbry, 1892). Parachiton acnrninatns is the type species of the genus Parachiton so its exclusion is not preferred. The preferred final tree representing the phylogeny, based on this character matrix, is the SRC tree excluding L. intermedins and D. cnrvata (Pig. 2B). Bootstrap values for the nodes recovered were generally low, with support (>50) for local sister-group relationships but low support (<50) for the major clade of Leptochiton and allies (Pig. 2A). Within this major Leptochiton clade, there is one consistently resolved group of primarily Atlantic species although one Japanese species, Leptochiton hirasei (Taki and Taki, 1929), is included. This is the only cluster of species that resolves consistently in all MPTs although bootstrap support for this clade is low. Reduced consensus methods further resolve a separate group of Pacific species of Leptocinton although one Atlantic species, Leptochiton alveolus (Sars MS, Loven, 1846), is included (Pig. 2B). Consistency index (Cl) values for most characters were quite low (Table 1). The average Cl values for the shell characters (0.22) were lower but not substantially different from the average Cl for girdle characters (0.26). The analysis of 32 shell characters resulted in 348 MPTs of length 206 steps; analysis of 27 girdle characters resulted in a large number of MPTs (>3000), length 180 steps. Consensus trees from each of these analyses resulted in a totally unresolved polytomy that failed to separate the outgroup and ingroup taxa. DISCUSSION Lepidopleuran chitons are recognized by three major synapomorphies; the absence (or very minimal development) of shell insertion plates; latero-ventral shell eaves that anchor the valves in the fleshy girdle; and the posterior (i.e., “merobranchial”) and adanal gill arrangement. The adanal condition is identified by the placement of multiple gills posterior to the nephridiopore; gills are added to both the anterior and posterior ends of the gill row during ontogeny. These features (absent in.sertion plates and gill arrangement) are traditionally considered plesiomorphic within chitons. Fossil polyplacophoran valves with morphology very similar to those in living lepidopleiirans are known in many taxa from the Lower Carboniferous (ca. 350 Mya), whereas shell ■ insertion plates first appear in the fossil record in the Permian ,i (ca. 290-250 Mya) and become more common in Cretaceous i taxa and later. Merobranchial adanal gills, although not preserved in the fossil record, are a feature unique to I lepidopleiirans, and this is assumed to be the ancestral i condition on that basis. Sirenko (2006) revised the classification of Lepidopleurida t and assigned the genera Choriplax, Hemiarthrnm, and i Weedingia to the family Hemiarthridae in the Chitonida. fl These three genera had previously been classified in the ti Lepidopleurida because, although they have shell insertion $ plates, the insertion plates are weak and unslitted. Their | removal from the Lepidopleurida was based on their abanal ij gill arrangement (with only one gill located posterior to the | nephridiopore in adult animals). Sirenko (2006) interpreted i| the gill arrangement in these three genera to be indicative that i they are more derived and that the insertion plates in their i shells are secondarily simplified. In the present analyis, I Choriplax resolves within Lepidopleurina. I The nature of nephridiopore arrangement in lepi- (i dopleurans and all chitons clearly requires further study. Other I genera that are uncontroversially considered to be within I Chitonida may have adanal (plesiomorphic) gill arrangements; I Chaetoplenra, Stenoplax, Callistochiton, and Onithochitoti I species all have three or more gills posterior to the nephridiopore p in adult animals. This classification of gill arrangements may ft not provide the straightforward synapomorphy that has been ft proposed. In the present analysis, the large and dominant genus ft Leptochiton is clearly non-monophyletic. The major clade & (Pig. 2B) could be considered to represent the family li Leptochitonidae including Parachiton. Although the shell ft morphology that defines Parachiton (having a substantially It enlarged tail valve) is distinctive, it may be interesting to | consider in future studies whether that feature has evolved I more than once. Outside of the major “Leptochitonidae” clade, the two other lepidopleuran families represented by more than one species are resolved in this tree: Nierstraszellidae (Nierstraszella spp.) and Ferreiraellidae (Ferreiraella spp.) as ! part of a basal polytomy including Hanleya (Hanleyidae) and I Oldroydia ( Protochitonidae). The consistent recovery of i Mediterranean and North Atlantic taxa in close topological i|| proximity (although not supported by bootstrap values) | indicates that there may be a single regional radiation that I links these taxa, which is worthy of further study. The two I genera included, Lepidoplcnrns and Leptochiton, have been ' the subject of taxonomic controversy; the type species of both genera are included in this putative North Atlantic clade. The finst cladistic analy.sis published to assess the relation- ships within a clade of chitons was presented by Sigwart ct al. (2007); as that analysis was intended to assess the phylogenetic MORPHOLOGICAL CLADISTICS WITH PRIMITIVE LIVING CHITONS 103 position of a new fossil species, it used only shell morphology. The resulting phylogeny did not recover any of the family or genus-level groups within Lepidopleurina although there were some relationships between individual sister taxa that were well supported (bootstrap >50), as was also found in this expanded study. Understanding the potential usefulness of the major data sets available is an important step in building our understanding of the evolution of this group. The analysis of the two largest data partitions in this study, shell morphology characters and girdle morphology characters, unambiguously demonstrates that employing more characters produces more resolution. Analysis of either data set alone results in large numbers of trees with effectively random distribution of taxa. This was also demonstrated by the large number of initial MPTs recovered in the study of Sigwart et al. (2007) using shell character data. In studies of molecular phylogenetic methods, it is well known that resolution and support recovered by phylogenetic analyses improves with increased character sampling (i.e., number of genes and/or length of sequences). This argument suggests that adding more data of any type produces better resolved and presumably more accurate trees {e.g., Poe and Swofford 1999, Simmons and Miya 2004). Questions remain un- resolved about the relative benehts of adding more characters or more taxa in phylogenetic reconstruction, particularly using morphological data (Graybeal 1998, Gobbett et al. 2007). Some workers have even argued that large-scale cladistic data sets of morphological data are less informative than those concentrating on few well-described characters (Scotland et al. 2003). The basis for this position is that morphological data alone cannot provide a sufficient character base to derive robust phylogenetic hypotheses, so it may be more accurate to rely on molecular data and use a few broadly defined morphological features as a reassuring appendum. This has been informally called the“Christmas tree” approach to the use of morphological attributes: using morphological ornaments to make an adequate tree more aesthetic. In fact, one argument for the dominance of molecular phylogenetic methods to the exclusion of morphological data is that “much of the useful morphological ^ diversity has already been scrutinized” (Scotland et al. 2003: ' 545). This statement is at odds with the state of anatomical I knowledge for many biological groups, of which chitons are a 1 quite typical example. Other studies have concluded that “much more [mor- phological] variation could be included in phylogenetic analyses than is used presently” (Poe and Wiens 2000: 33-34). This is true not only of computational phylogenetic analyses but of the descriptive taxonomic studies that form a basis for phylogenetic investigation. The sources of data to describe species of chitons rely almost exclusively on external anatomy (gills, girdle, and shell morphology), with use of radular morphology (Saito 2004) and aesthete arrangement (Sirenko 2001, Fernandez et al. 2007) gaining popularity and regular usage. Further sources of data, such as gamete morphology (Buckland-Nicks 2006) and nephridiopore arrangement (Sirenko 1993, Sigwart 2008b) and others will become standard sources of morphological character data as anatomical descriptions are published for larger numbers of closely related taxa. The present study, based on “classical” morphological characters used in traditional species descriptions of chitons, functions as a model to demonstrate that cladistic analyses of morphological data sets can successfully generate phylo- genetic hypotheses. Wliether or not the hypotheses proposed by these cladograms are “true” will be determined by the progres- sive accumulation of a body of work examining the internal relationships of taxa within well-established clades, based on multiple data sets including genetic and novel morphological information. Morphological data that are available from currently routine description and analysis of specimens can be used to produce phylogenetic hypotheses and provide new insights into the relationships within known clades. The anecdotal superficial homogeneity of chiton morphology is disputed by this finding. Maximizing the number of morphological characters in analysis improves the quality of phylogenetic signal recovered (making the transition from a complete polytomy to resolved clades). Application of morphological data is critical not only because it is sensible to use all available data for living taxa, but also because it creates a link to fossil taxa. Encompassing all forms of data for fossil and living species is critical to understanding the ongoing evolution of chitons and other groups of living fossils. ACKNOWLEDGMENTS The author thanks M. Glaubrecht and T. von Rintelen for their work in organizing the symposium Molluscs as Models in Evolutionary Biology (2007) and the opportunity to contribute to this collection of papers. Michael Schrodl and Bernhard Lieb provided improving reviews and discussion. This work was supported by the SYNTHESYS Project, funded by the European Gommunity FP6 programme, for research visits to MNHN (award FR-TAF-1157) and NHMW (award AT-TAF-405). V. Heros (MNHN), E. Schwabe (ZSM), and A. Eschner (NHMW) are thanked for their help with access to and loan of specimen material. LITERATURE CITED Buckland-Nicks, ). 1995. Ultrastructure of sperm and sperm-egg in- teraction in Aculifera: Implications for molluscan phylogeny. 104 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Memoirs du Museum national d'Histoire naturelle, Paris 166; 129-153. Buckland-Nicks, J. 2006. Fertilization in chitons; Morphological clues to phylogeny. Venus 65: 51-70. Cherns, L. 2004. Early Palaeozoic diversification of chitons (Poly- placophora, Mollusca) based on new data from the Silurian of Gotland, Sweden. Lethaia 37: 445-456. Cobbett, A., IVl. Wilkinson, and M. A. Wills. 2007. Fossils impact as hard as living taxa in parsimony analyses of morphology. Sys- tematic Biology 56: 753-766. Faith, D. P. and P. S. Cranston. 1991. Could a cladogram have arisen by chance alone? On permutation tests for cladistic structure. CladisticsJ: 1-28. Fernandez, C. Z., M. 1. Vendrasco, and B. Runnegar. 2007. Aesthete canal morphology in twelve species of chiton (Polyplacopho- ra). The Veliger49: 51-69. Graybeal, A. 1998. Is it better to add taxa or characters to a difficult phylogenetic problem? Systematic Biology 47: 9-17. Haszprunar, G. 1996. The Mollusca: Coelomate turbellarians or mesenchymate annelids? In: ]. D. Taylor, eds.. Origin and Evo- lutionary Radiation of the Mollusca. Oxford University Press, Oxford. Pp. 3-28. Hodgson, A. N., ]. M. Baxter, M. G. Sturrock, and R. T. F. Bernard. 1988. Comparative spermatology of 1 1 species of Polyplacoph- ora (Mollusca) from the suborders Lepidopleurina, Chitonina, and Acanthochitonina. Proceedings of the Royal Society, London (B) 235: 161-177. Kaas, P. and R. A. Van Belle. 1985. Monograph of Living Chitons (Mol- lusca: Poly placophora), Volume 1. Order Neoloricata, Lepidopleu- rina. Backhuys, Leiden. The Netherlands. OkiisLi, A., E. Schwabe, D. J. Eernisse, and G. Giribet. 2003. Towards a phylogeny of chitons (Mollusca, Poly placophora) based on combined analysis of five molecular loci. Organisms, Diversity and Evolution 3: 281-302. Poe, S. and D. L. Swofford. 1999. Taxon sampling revisited. Nature 398: 299-300. Poe, S. and 1. J. Wiens. 2000. Character selection and the method- ology of morphological data. Iti: J. J. Wiens, eds.. Phylogenetic Analysis of Morphological Data, Smithsonian Institution Press, Washington, D.C. Pp. 20-36. Saito, H. 2004. Phylogenetic significance of the radulae in chitons, with special reference to the Cryptoplacoidea (Mollusca: Poly- placophora). Bollettino Malacologico 39 (Supplement 5): 83-104. Scotland, R. W., R. G. Olmstead, and ). R. Bennetl. 2003. Phylogeny reconstruction: The role of morphology. Systematic Biology 52: 539-548. Sigwarl, ). 1). 2008a. Phylogeny and Evolution of Basal Living Chitons (Mollusca: Polyplacophora: Lepidopleurida). Ph.D. Dissertation, (Queen’s University, Belfast, UK. Sigwart, |. 1). 2008b. Gross anatomy and positional homology of gills, gonopores, and nephridiopores in “basal” living chitons ( Polyplacophora; Lepidopleurina). American Malacological Bulletin 25: 43-49. Sigwart, ). D. and M. I). Sutton. 2007. Deep mollu.scan phylogeny: Synthesis of palaeontological and neontological data. Proceed- ings of the Royal Society, London ( B ) 274: 24 1 3-24 1 9. Sigwart, |. D., S. B. Andersen, and K. I. Schnetler. 2007. Eirst record of a chiton from the Palaeocene of Denmark (Polyplacophora: Leptochitonidae) and its phylogenetic affinities. Journal of Sys- tematic Palaeontology 5: 123-1 32. Simmons, M. P. and M. Miya. 2004. Efficiently resolving the basal clades of a phylogenetic tree using Bayesian and parsimony | approaches: A case study using mitogenomic data from 100 | higher teleost fishes. Molecular Phylogenetics and Evolution 31: j 351-362. Sirenko, B. I. 1993. Revision of the system of the order Chitonida (Mollusca: Polyplacophora) on the basis of correlation between the type of gills arrangement and the shape of the chorion pro- cesses. Riithenica 3: 93-1 17. Sirenko, B. I. 2001. Deep-sea chitons (Mollusca, Polyplacophora) from sunken wood off New Caledonia and Vanuatu. Memoirs du Museum national d'Histoire naturelle, Paris 185: 39-71. Sirenko, B. 1. 2006. New outlook on the system of chitons (Mollusca; Polyplacophora). Venus 65: 27-49. j" Swofford, D. L. 2002. PAUP*. Phylogenetic Analysis Using Parsimony (*and Other Methods). Version 4. Sinauer Associates, Sunder- land, Massachusetts. Vendrasco, M. J., T. E. Wood, and B. N. Runnegar. 2004. Articu- lated Palaeozoic fossil with 1 7 plates greatly expands disparity of early chitons. Nature 429: 288-29 1 . Vendrasco, M. J., C. Z. Eernandez, D. J. Eernisse, and B. Runnegar. 2008. Aesthete canal morphology in the Mopaliidae (Polypla- cophora). American Malacological Bulletin 25: 51-69. Wilkinson, M. 1999. Choosing and interpreting a tree for the oldest mammal. /onr/ifl/ of Vertebrate Paleontology 19: 187-190. Wilkinson, M. 2001 . REDCON 3.0: Software and Documentation. De- partment of Zoology, The Natural History Museum, London. Submitted: 9 October 2008; accepted: 25 February 2009; final revisions received: 16 March 2009 ; Anier. Malac. Bull. 27: 105-11 1 (2009) The use of developmental sequences for assessing evolutionary change in gastropods'^ Jennifer Smirthwaite, Simon D. Rundle, and John I. Spicer School of Biological Sciences, University of Plymouth, Plymouth PL4 8AA, U.K. Corresponding author: srundle@plymouth.ac.uk Abstract: First introduced by Ernst Haeckel in the nineteenth century, the use of developmental sequences has recently seen a renaissance as part of the study of the evolutionary biology of embryos; here we review briefly the literature describing gastropod developmental sequences, appraising the extent to which it has contributed to this renaissance. Gastropods have figured extensively in studies of early development with cell lineage analysis available for numerous taxa. Phylogenetic comparisons of these data reveal strong evolutionary signals, particularly in relation to early cell divisions. In contrast, although the description of post cell division developmental stages, including functional elements of development, in gastropods is extensive, interspecific comparisons are rare and tend to focus instead on developmental mode. However, a recent comparison of the sequence of functional and morphological events in a clade of basommatophoran snails demonstrated several alterations in the timing of developmental events {i.e., heterochronies) across the phylogeny. Many gastropod groups may offer the potential to carry out similar investigations of the evolutionary importance of sequence heterochrony and to try to unravel the mechanistic basis for such patterns in developmental sequences. Key words: Pulmonata, Basommatophora, embryo, development, heterochrony "'....Haeckel can be seen as the father of a sequence- based phylogenetic embryology" (Richardson and Keuck 2002; 495)." DEVELOPMENTAL SEQUENCES IN BIOLOGY Haeckel is perhaps best known for his Biogenetic Law, which, in its simplest interpretation, proposed that ontogeny is a brief and rapid re-run of phylogeny, evolution occurring by terminal addition (Haeckel 1866). Much has been made of the rescinding of this law, and Haeckel’s reputation and work has to some extent been damaged by these discussions (Garstang 1922, deBeer 1958, Gould 1977, Richardson 1995, 1998, Richardson and Jeffery 2002, Richardson and Keuck 2002). However, it is frequently overlooked that, in the process of formulating his now discredited law, Haeckel made other valuable contributions to the field of evolutionary biology (Richardson and Keuck 2002). First, he flagged the importance of variation in embryonic development between species, the same variation that underpins much of the current emphasis in “Evo-Devo” research (Raff 2000, Arthur 2002). Second, he was first to champion the use of developmental sequences in cross-species comparisons (Richardson and Keuck 2002): his approach of using letters of the alphabet to represent the succession of developmental stages that could be lost or replaced through evolutionary time provided a significant advance on other approaches that focused on the relative rates of development of different traits (e.g., as emphasized by workers such as Wilhelm His) and is very similar to approaches currently being advocated for analyzing changes in developmental sequence between descendent and ancestral species (i.e., heterochronies — see below). Related to the use of developmental sequences, another of Haeckel’s major contributions to biology was the coining of the term heterochrony, which he used to describe “anomalies” to his biogenetic law (Richardson and Keuck 2002). Such anomalies were alterations to developmental sequences across species, whereby developmental events appeared either later or earlier in ontogeny (Haeckel also flagged heterotopes, which were alterations in the position of developmental events within the embryo). While Haeckel viewed heterochronies as exceptions to his biogenic law, they appear to have assumed major importance in the investigation of evolution and have even been suggested by some to be one of the key drivers of evolutionary change (e.g., Gould 1977, but see also Raff 1996). Much of this emphasis of heterochrony as a major evolutionary shaping force has centered around shifts in the developmental timing of morphological traits and, in particular, body size (as a proxy for developmental time) in relation to reproductive maturation. This so-called global heterochrony shifted the emphasis from sequences to one concerned mostly with size and shape (i.e., towards an allometric approach). From the symposium “Molluscs as models in evolutionary biology: from local speciation to global radiation” presented at the World Congress of Malacology, held from 15 to 20 July 2007 in Antwerp, Belgium. 105 106 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 It is only comparatively recently that those interested in heterochrony have re-established the Haeckelian focus on using developmental sequences and the potential importance of comparing these sequences across taxa in order to elucidate patterns in the evolution of development [e.g.^ Smith 2001, 2003, Bininda-Emonds et al. 2002). This “new” emphasis has seen the development of sophisticated analytical procedures that allow the formal comparisons of changes in develop- mental sequence across species within an explicit phylogenetic context (sequence heterochronies) (Mabee and Trendler 1996, Smith 1996, 1997, 2001, Richardson etal. 2001, Bininda-Emonds et al. 2002, Jeffery et al. 2002a, 2002b, 2005, Schulmeister and Wheeler 2004). Given the great wealth of information that exists on gastropod development, stretching back over one hundred years, coupled with a voluminous recent literature marrying molecular techniques with classical embryology, the question can be posed as to what extent has the study of gastropod development figured in this renaissance? Here we address this question by reviewing briefly work that has aimed to compare developmental sequences in gastropods. We de- monstrate that work on gastropods has been at the forefront of investigations assessing the evolutionary importance of very early developmental stages but that, paradoxically, despite a voluminous literature documenting patterns for individual species, cross-species comparisons of events later in the devel- opmental sequence appear to be rare. Einally, we present the findings of a recent study that begins, in a small way, to redress this imbalance. Developmental sequences in gastropods Cell lineage analysis There is a long and distinguished history of studying cell lineages in gastropods, initiated by workers such as Blochmann (1882), Gonklin (1897), and Delsman (1914) [see Raven (1958) and Eretter and Graham ( 1962) for summaries of this early work and Lindberg and Guralnick (2003) for a list of papers]. However, it was not until Hyman (1951) that the information from these pioneering studies alongside that on cell lineages for other spiralian taxa were used to propose a link between development and evolution; this reluctance to link ontogeny with phylogeny was almost certainly a reaction to the controversy surrounding recapitulation (Lindberg and Guralnick 2003). In the past couple of decades, with the advent of sophis- ticated approaches for exploring phylogenetic relationships, gastropods have again been at the center of research linking early cell-cleavage patterns with evolution although it should be noted that many of these analyses draw on data from the studies at the turn of the nineteenth century (freeman and Lundelius 1992, van den Biggelar 1993, van den Biggelar and Haszprunar 1996, Bonder and Lindberg 1997). In particular. there has been a focus on the timing of the formation of the 4d mesentoblast (the precursor of the mesoderm), with the relation of this timing to the cleavage of other cells being linked to the evolution of the major gastropod groups (van den Biggelar and Haszpruner 1996, Lindberg and Guralnick 2003). The main observation is that the onset of the 3d macromere division, which leads to the formation of the 4d mesentoblast, is accelerated through evolutionary time. Hence, in more derived gastropod groups such as the caeno- gastropods and heterobranchs, it occurs at the 24-cell stage compared with at the 63-cell stage in the stem gastropod taxa (e.g., Patellogastropoda and Vestigastropoda). In effect, these studies demonstrate a heterochrony in the se- quence of very early development, with a shift in the timing of developmental events between ancestral and descendent taxa. A more recent analysis on more extensive cell-lineage data for more taxa (Lindberg and Guralinick 2003) confirmed that there was congruence between phylogenetic trees derived using cell lineages and those using morphological and molecular approaches. This study also identified a long branch within the cell-lineage phylogeny, indicating a large number of developmental event changes, between the Patellogas- i tropoda/Vestigastropoda and Neritopsina/Apogastropoda ■ clades. This evolutionary change again indicated an accelera- tion in development, a shortening of the trochophore stage, and an accompanying lengthening of the veliger stage. It was proposed that this shift towards the earlier development of > a longer planktotrophic stage may have been a response to increased levels of primary production in the oceans during i the Silurian period. In effect, this is a heterochronous change i in the timing of developmental stage, and it is clear that the ! investigation of cell lineages in relation to gastropod evolution > has led us full circle in terms of the important link between . ontogeny and phylogeny. Sequences in later developmental events The evidence for an evolutionary role for development from extensive phylogenetic analysis of very early gastropod developmental events has not been extended to any great degree to the later stages of development. This is somewhat ; surprising, given the extent to which the developmental : events of many gastropod species have been described in detail [for example see Raven ( 1958) and Lretter and Graham i ( 1962)]. Indeed, the only substantial phylogenetic analysis on ' gastropod development has focused on an a.s.sessment of the : evolution ol developmental mode rather than developmental i' sequences per sc. Ciollin (2004) mapped the developmental ' mode (ol 72 calyptraeid gastropods) onto a phylogeny and i found that there was no evidence that phylogenetic effects ^ had constrained the evolution of this trait; species with i planktotrophic, lecithotrophic, or tlirect development with i DEVELOPMENTAL SEQUENCES IN GASTROPODS 107 nurse eggs all had the potential to evolve a different devel- opmental mode. Other studies that focus more on developmental se- quences are either qualitative or restricted in their comparative element. Page (1994), for example, provided an interesting qualitative comparison of the occurrence of eight develop- mental structures in opisthobranchs and prosobranchs. She concluded that young planktonic opisthobranch larvae represented a good approximation of an ancestral gastropod larva by not expressing many of the structures of the defini- tive body found early in prosobranch development {i.e., at the veliger stage) until late in the larval phase. Gibson (2003), in contrast, found that one group of opisthobranchs, the Notospidea, possessed adult characters (notum differen- tiation, adult shell growth, lack of operculum) during the early larval stage. Collin and Wise (1997) included observa- tions of early cell development in their investigation of development in the pyramidellid Odostomia columbiana Dali and Bartsch, 1907 and concluded that larvae with un- equal cleavage and early development of eyes and tentacles might represent the common ancestors of pyramellids and opisthobranchs. At the same time, gastropods have been used by workers taking a functional approach to embryonic development, including studies of embryonic ionic balance [e.g., Taylor 1977), calcification (e.^., Bielefeld and Becker 1991), respiration {c.g., Baldwin 1935), and muscle and nerve development {e.g., Croll and Voronezhskaya 1996, Page 1998, Yamanaka et al. 2000). Clearly, there is a lot of information on the later developmental stages of gastropod species both in terms of morphological and functional traits, including some evidence for differences in developmental sequences of these traits between species. More extensive, formal analysis of these data could shed much light on the role that developmental sequences have played in gastropod evolution; indeed, gastropods might provide particularly good models for integrated approaches that focus on functional as well as morphological aspects of development. In the next section, we describe a study that takes such a formal and integrative approach. Case study: A quantitative comparison of developmental sequences in basonvnatophoran gastropods So far, we have seen that there is an imbalance between cell lineage and later developmental characters when it comes to comparative approaches with developmental se- quences in gastropods, with a lack of studies focusing on later development. In fact, the bulk of studies explicitly testing for differences in developmental sequences across phylogenies have been for mammals. In these examples, there is clear evidence for heterochrony in terms of altered sequences of developmental events between ancestors and descendents. There are, for example, differences in the timing ot the central nervous system and the craniofacial apparatus between eutherians and marsupials (Smith 1997, Nunn and Smith 1998). Jeffery et al. (2002a) also demonstrated that, within the amniotes, mammals were characterized by delayed development of the eyes and that there were several hetero- chronies involving shifts in cardiac events relative to non- cardiac events. However, a more recent study suggested that the role of sequence heterochrony in mammals may not be that extensive; Bininda-Emonds et al. (2004) hypothesized that a lack of heterochronies in mammals might be expected due to the time during which the organo-genetic period occurs being spent in the protective environment of the amniotic egg. In many invertebrate groups, however, including many gastropods, embryonic development is external and, potentially, more open to evolutionary processes. Hence, we might predict that alterations to the developmental sequence might be more pronounced in such instances. A recent study by Smirthwaite et al. (2007) took the first step towards exploring heterochrony in invertebrates, by comparing developmental sequences in three clades of basommatophoran snails. This work involved observing the precise timings of “traditional” morphological stages and functional/physiological traits in 12 species, mapping these events onto developmental time lines and then making a formal comparison of the relative timing of these events among species. The morphological stages used were those that have been described previously by workers such as Raven (1958) and Cummin (1972) and included the trochophore, veliger, and hippo stages (Eigs. lA-C). It is important to note that it was the onset of these stages that were used as events. The physiological and functional events typically used occurred after these morphological stages (but see below) and included the formation of eyespots, heart, and radula, the onset of body flexing and contraction of mantle muscle, the time when the animal migrated from the egg capsule and, finally, the egg mass (Figs. 1-2). Once event timings had been determined, timelines were constructed for each species, with that for Lymnaea stagnalis (Linnaeus, 1758) used as the “standard” (Fig. 2) and event- paircracking (PARSIMOV approach — Jeffery efu/. 2005) was used to test formally for heterochronies. In essence, this technique involves comparing the timing of each pair of events by classifying each event depending on whether it occurred earlier, simultaneously, or later than each other event in all other species. These scores are then mapped onto the phylogeny for the group. This analysis provided formal support for several heterochronies, some of which are illustrated in Fig. 2. The lines on the first panel of this figure [i.e., Fig. 2 A) illustrate three traits (hippo, eyespot formation, and mantle muscle control) that do not change their timing relative to one another 108 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 A B 1 # o D f X : - X ^ m e E t F Figure 1. Developmental stages of Lymnaea stagnalis embryos; A, trochophore; B, veliger; C, early hippo; D, free-swimming hippo; E, crawling hippo; F, juvenile snail migrating from egg mass. Arrows and letter for free-swimming hippo indicate the location of the eye (e), body flexing (f), and mantle muscle contraction (m). across species; hence, these traits are non-heterochronous. In contrast, the second panel (Lig. 2B) illustrates three traits that are heterochronous. One of these heterochronies involves the early occurrence of the embryo attaching to the egg capsule wall in relation to eyespot formation and body flexing in the two species of Physa Draparnaud, 1801 compared with all other species. Eyespot formation and body flexing were also heterochronous; within the Planorbidae and the Physidae, body flexing occurred before the eye was formed, whereas only two [Radix aiirkularia (Linnaeus, 1758) and Omphiscola glabra (O. E Muller, 1774)] of the six species in the family Lymnaeidae showed this timing pattern. Hence, it appears that, within this clade of gastropods, heterochronies are associated with speciation events and larger taxonomic divergences at the family level, fhe number of events changing their position was also similar in relative terms to those identified for mammals ( Bininda-Emonds cl al. 2004). Where can the investigation of developmental sequences in gastropods take us? The study by Smirthwaite el al. (2007) provides evidence that sec|uence heterochrony occurs in a clade of gastropods. Figure 2. Sequences of events mapped on a time line for twelve spe- cies of basommatophoran snail; the three main clades represent (from top to bottom) the Lymnaeidae, Planorbidae, and Physidae. Event labeling (note: event timing was measured as the starting point for that event): 1, laying; 2, trochophore; 3, veliger; 4, hippo; 5, eyespot formed; 6, heart beat; 7, free swimming; 8, body tlexing; 9, mantle muscle contraction; 10, attachment to egg; 1 1, crawling; 12, radula; 13, emergence from egg capsule; 14, migration from the egg mass. A, illustrates three non-heterochronous events, i.c., which do not change their relative sequence across species: hippo, dashed line; eyespot formation, solid line; mantle muscle control, diUted line. B, illustrates sequence heterochronies among three events: in the two Physa species, attachment to the egg capsule (solid line) occurs ear- lier in development than body Hexing (dotted line) and eyespot for- mation (dashed line) compared with in other species: body Hexing (doited line) occurs earlier in the Physidae, Planorbidae, and two species of Lymnaeidae [Radix ballhica Linnaeus, 1758 and ('hiiplds- cola gl(d’ra) than in the other species within Lymnaeidae. which suggests that, potetitially, heterochrony make have playetl a part in the evolution of this groitp. 1 lowever, we must be wary of a.ssuming that, because we have demonstrated DEVELOPMENTAL SEQUENCES IN GASTROPODS 109 heterochrony as essentially a pattern of evolutionary change, that the evolutionary mechanism must also involve hetero- chrony; there is no necessity for pattern and mechanism of evolutionary change both to be heterochronous (Spicer and Rundle 2007). Indeed, we would suggest that one of the big challenges for research on developmental sequences will be to establish whether there is a link between heterochronic process and heterochrony as a pattern (Spicer and Rundle 2007) . Clearly, this approach will need to be guided by com- parative studies such as that by Smirthwaite et al. (2007) that flag functional heterochronies between extant species, al- lowing investigations such as whether alterations to devel- opmental sequences between species can be replicated within species. Such intraspecific changes in developmental sequence have been termed heterokairies (Spicer and Burggren 2003, Spicer and Rundle 2006, 2007) and their investigation should promote research on developmental sequences. So might gastropods be a good model for future studies on the role of developmental sequences in evolution? We feel that the answer to this question is yes, for several reasons, first, the embryonic development of many gastropods is external and visible, which means that observations of early development and manipulations of the developmental environment are tractable. Second, phylogenies have been constructed for several groups and the relatedness of the major gastropod clades has also been formulated, allowing the explicit test of patterns in sequences within an evo- lutionary framework, finally, the extensive information on cell lineages within the gastropods has meant that they have been one of the groups at the forefront of the develop- ment of techniques for generating cell-fate maps such as fluorescent stains used with confocal microscopy. Such tech- niques have already allowed some workers to trace devel- opmental pathways from cell cleavage through to structures such as nerves, muscles, the mantle, and cilia within gastropod species (Render 1997, Hejnol et al. 2007, Wanninger et al. 2008) . The emphasis so far has been in elucidating how major taxonomic divergences may be linked with different devel- opmental pathways, yet it might be possible that species- level divergence might also be driven by small-scale variation in induction patterns during early cell divisions. Such variation might also be linked to later differences in the developmental sequence. Clearly, this is a highly speculative hypothesis but perhaps one that deserves consideration. At the same time, the ability to measure gene expression and up-regulation in gastropod embryos (Lartillot et al. 2002, Hinman et al. 2003) will also enhance our understanding of the genetic basis of developmental sequence change. Cou- pled with studies that assess fitness implications, there is real potential to take a truly integrated approach to the study of developmental sequences using gastropods (Spicer and Rundle 2006). LITERATURE CITED Arthur, W. 2002. The emerging conceptual framework of evolution- ary developmental biology. Nature 415: 757-764. Baldwin, E. 1935. The energy sources in ontogenesis. VIII The respi- ratory quotient of developing gastropod eggs. Journal of Experi- mental Biology 12: 27-35. Bielefeld U. and W. Becker. 1991. Embryonic development of the shell in Biomphalaria glabrata (Say). International Journal of Developmental Biology 35: 121-131. Bininda-Emonds, O. R. R, J. E. Jeffery, M. I. Coates, and M. K. Richardson. 2002. From Haeckel to event-pairing: The evolu- tion of developmental sequences. Theory in Biosciences 121: 297-320. Bininda-Emonds, O. R. R, J. Jeffrey, and M. K. Richardson. 2004. Is sequence heterochrony an important evolutionary mech- anism in mammals? Journal of Mammalian Evolution 10: 335-359. Blochmann, F. 1882. Uber die Entwicklung der Neritina fluviatilis Mill. Zeitschrift fiir wissenschlaftliche Zoologie 36: 125-174 [In German]. Collin, R. 2004. Phylogenetic effects, the loss of complex characters, and the evolution of development in calyptraeid gastropods. Evolution 58: 1488-1502. Collin, R. and J. B. Wise. 1997. Morphology and development of Odostomia Columbiana Dali & Bartsch (Pyramidellidae): Impli- cations for the evolution of gastropod development. Biological Bulletin 192: 243-252. Conklin, E. G. 1897. The embryology of Crepidula, a contribution to the cell lineage and early development of some gastropods. Journal of Morphology 13: 1-266. Croll, R. P. and E. E. Voronezhskaya. 1996. Early elements in gastro- pod neurogenesis. Developmental Biology 173: 344-347. Cummin, R. 1972. Normentafel zur Organogenese von Lymnaea stagnalis (Gastropoda Pulmonata) mit besonderer Beriicksich- tigung der Mitteldarmdrtise. Revue Suisse Zoologie 79: 709-774 [In German]. deBeer, G. 1958. Embryos and Ancestors, 3^^ Edition. Clarendon Press, Oxford, U.K. Delsman, H. C. 1914. Entwicklungsgeschichte von Littorina obtusa- ta. Tijdschrift der Nederlandsche Dierkundige Vereeniging 14: 383-498 [In German]. Freeman, G. and J. W. Lundelius. 1992. Evolutionary implications of the mode of D quadrant specification in coelomates with spiral cleavage. Journal of Evolutionary Biology 5: 205-247. Fretter, V. and A. Graham. 1962. British Prosobranch Molluscs: Their Eunctional Anatomy and Ecology. Ray Society, London. Garstang, W. 1922. The theory of recapitulation. A critical restate- ment of the biogenetic law. Journal Linnean Society, London (Zoology) 35: 81-101. Gibson, G. D. 2003. Larval development and metamorphosis in Pleurobraitchaea maculata, with a review of development in the Notaspidea (Opisthobranchia). Biological Bulletin 205: 121-132. Gould, S. J. 1977. Ontogeny and Phylogeny. Harvard University Press, Cambridge, Massachusetts. 110 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Haeckel, E. H. R A. 1866. Generelle Morphologic der Organismen allgemeine Griindziige der organischen Formen-Wisseiischnft: rnechanisch begriindet durch die von Charles Darwin re- forniirte Descendenz-Theorie. Vol. 1-2. G. Reimer, Berlin [In German]. Hejnol, A., M. Q. Martindale, and J. Q. Henry. 2007. High- resolution late map of the snail Crepidula fornicata: The origins of ciliary bands, nervous system, and muscular elements. De- velopmental Biology 305: 63-76. Hinman, V. R, E. K. O’Brian, G. S. Richards, and B. M. Degnan. 2003. Expression of anterior Hox genes during larval develop- ment of the gastropod Haliotis asinina. Evolution and Develop- ment 5: 508-521. Hyman, L. H. 1951. The Invertebrates: Platyhelminthes and Rhyn- chocoela, the Acoelomate Bilateria, Vol. 2. McGraw-Hill, New York. leffery, J. E., M. K. Richardson, M. I. Goates, and O. R. P. Bininda- Emonds. 2002a. Analyzing developmental sequences within a phylogenetic framework. Systematic Biology 51: 478-491. Jeffery, J. E., O. R. R Bininda-Emonds, M. I. Goates, and M. K. Richardson. 2002b. Analyzing evolutionary patterns in ver- tebrate embryonic development. Evolution and Development 4: 292-302. Jeffery, J. E., O. R. R Bininda-Emonds, M. I. Coates, and M. K. Richardson. 2005. A new technique for identifying sequence heterochrony. Systematic Biology 54: 230-240. Lartillot, N., O. Lespinet, M. Vervoort, and A. Adouette. 2002. Ex- pression pattern of Brachyury in the mollusc Patella vulgata suggests a conserved role in the establishment of the AP axis in Bilateralia. Development 129: 141 1-1421. Lindberg, D. R. and R. P. Guralnick. 2003. Phyletic patterns of early development in gastropod molluscs. Evolution and Develop- ment 5: 494-507. Mabee, P. M. and T. A. Trendler. 1996. Development of the cranium and paired fins in Betta splendens (Teleosti: Percomorpha): In- traspecific variation and interspecific comparisons. Journal of Morphology 227: 249-287. Nunn, C. L. and K. K. Smith. 1998. Statistical analysis of develop- mental sequences: The craniofacial region in marsupial and placental mammals. T/ncn'cnn Naturalist 152: 82-101. Page, L. R. 1994. The ancestral gastropod larval form is best approxi- mated by hatching-stage opisthobranch larvae: Evidence from comparative developmental studies. In: W. H. Wilson Jr., S. A. Strieker, and G. L. Shinn, eds.. Reproduction and Development of Marine Invertebrates. 4'he Johns Hopkins University Press, Baltimore and London. Pp. 206-223. Page, L. R. 1998. Sequential developmental programmes for retrac- tor muscles of a cacnogastropod: Reappraisal of evolutionary homologues. Proceedings of the Royal Society (B) 265: 2243- 2250. Ponder, W. P. and 1). R. Lindberg. 1997. 4’owards a phylogeny of gastropod molluscs: An analysis using morphological char- acters. Zoological Journal of the l.innean Society, London I 19: 83-265. Raff, R. A. 1996. The Shape oj Life. University of Ghicago Press, Ghicago. Raff, R. A. 2000. Evo-devo: The evolution of a new discipline. Nature \ Reviews Genetics 1: 74-79. Raven, G. P. 1958. Morphogenesis: The Analysis oj Molluscan Develop- ment. Pergamon Press, London. Render, J. 1997. Gell fate maps in the Ilyanassa obsoleta embryo ' beyond the third division. Developmental Biology 189: 301- 310. Richardson, M. K. 1995. Heterochrony and the phylotypic period. { Developmental Biology 172: 412-421. Richardson, M. K. 1998. Haeckels’s embryos continued. Science 281: 1 1289. Richardson, M. K. and J. E. Jeffery. 2002. Haeckel and modern biol- ogy. Theory in Biosciences 121: 247-251. Richardson, M. K. and G. Keuck. 2002. Haeckel’s ABC of evolution i and development. Biological Reviews 77: 495-528. Richardson, M. K., J. E. Jeffery, M. I. Coates, and O. R. P. Bininda- I Emonds. 2001. Comparative methods in developmental biol- ogy. Zoology 104: 278-283. Schulmeister, S. and W. C. Wheeler. 2004. Comparative and phylo- genetic analysis of developmental sequences. Evolution and i Development 6: 50-57. Smirthwaite, J. J., S. D. Rundle, O. R. P. Bininda-Emonds, and J. J. Spicer. 2007. An integrative approach identifies developmen- tal sequence heterochronies in freshwater basommatophoran , snails. Evolution and Development 9: 122-130. Smith, K. K. 1996. Integration of craniofacial structures during de- 1 velopment in mammals. Amena?;; Zoologist 56: 70-79. Smith, K. K. 1997. Comparative patterns of craniofacial develop- |' ment in eutherian and metatherian mammals. Evolution 51: < 1663-1678. Smith, K. K. 2001. Heterochrony revisited: The evolution of devel- > opmental sequences. Biological Journal Linncan Society, London I 73: 169-186. Smith, K. K. 2003. Time’s arrow: Heterochrony and the evolution i of development. Inter?iational Journal of Developmental Biology r 47: 613-621. Spicer, J. I. and W. W. Burggren. 2003. Development of physiological ^ regulatory systems: Altering the timing of crucial events. ZooT r ogy 106: 91-99. Spicer, J. I. and S. D. Rundle. 2006. Out of place and out of time: n Towards a more integrative approach to heterochri)ny. Animal r Biology 57: 487-502. Spicer, J. I. and S. D. Rundle. 2007. Plasticity in the timing ol physi- r: ological development: Physiological heterokairy — what is it, how frequent is it, and does it matter? Comparative Biochemis- n try and Physiology (A) 148: 712-719. Taylor, 11. 11. 1977. The ionic and water relations of embryos of ' Lynmaea stagnalis, a freshwater pulmonate mollusc, lournal of ' Experimental Biology 69: 1 43- 1 72. van den Biggelar, |. A. M. 1993. Cleavage pattern in embryos o\ I tali- L Otis tiibercidata (Archaeogastropoda) and gastropod phylogeny. n journal of Morphology 216: 121-1 39. van den Biggelar, |. A. M. and G. Haszpnmar. 1996. Cleavage pat- k terns aiul mesentoblast formation in the Gastropoda: .An evo- >' hitionary perspective. Evolution 50: 1520-1540. DEVELOPMENTAL SEQUENCES IN GASTROPODS Wanninger, A., D. Koop, S. Moshel-Lynch, and B. M. Degnan. 2008. Molluscan evolutionary development. In: W. R Ponder and D. R. Lindberg, eds., Phylogeny and Evolution of the Mollusca. Univer- sity of California Press, Berkeley and Los Angeles. Pp. 427-446. Yamanaka, M., D. Hatakeyama, H. Sadamoto, T. Kimura, and E. Ito. 2000. Development of key neurons for learning stimulates learning ability in Lymnaea stagnalis. Neuroscience Letters 278: 113-116. Submitted: 9 November 2008; accepted: 9 February 2009; final revisions received: 16 April 2009 f Amer. Maine. Bull. 27: 1 13-132 (2009) Alien non-marine snails and slugs of priority quarantine importance in the United States: A preliminary risk assessment Robert H. Cowie,* Robert T. Dillon, Jr.^, David G. Robinson^, and James W. Smith'* ' Center for Conservation Research and Training, Pacific Biosciences Research Center, University of Hawaii, 3050 Maile Way, Gilmore 408, Honolulu, Hawaii 96822, U.S.A. - Department of Biology, College of Charleston, Charleston, South Carolina 29424, U.S.A. ^ USDA-APHIS-PPQ / Department of Malacology, Academy of Natural Sciences, 1900 Benjamin Franklin Parkway, Philadelphia, Pennsylvania 19103, U.S.A. ■' USDA-APHIS-PPQ-CPHST (Center for Plant Health Science and Technology), Raleigh, North Carolina 27606, U.S.A. Corresponding author: cowie@hawaii.edu Abstract: In 2002, the U.S. Department of Agriculture requested assistance from the American Malacological Society in the development of a list of non-native snails and slugs of top national quarantine significance. From a review of the major pest snail and slug literature, together with our own experience, we developed a preliminary list of gastropod species displaying significant potential to damage natural ecosystems or agriculture, or human health or commerce, and either entirely absent from the United States to our knowledge or restricted to narrow areas of introduction. Comments on the list from the worldwide malacological community were then solicited and led us to modify the original list. We then evaluated the taxa on this list by ranking them according to 12 attributes — seven biological variables and five aspects of human interaction — based on thorough review of the detailed literature. The ranked list that emerged from this risk assessment process included 46 taxa (species or species-groups) in 18 families. The highest ranked taxa were in the Ampullariidae, Hygromiidae, Cochlicellidae, Helicidae, Veronicellidae, Succineidae, Achatinidae, and Planorbidae. We validated the risk assessment model by scoring a suite of non-native snail and slug species already present in the United States. The list is not definitive but rather is offered as a framework for additional research. There remain important gaps in biological knowledge of many of the taxa evaluated, and rigorous reporting of economic impacts is extremely limited. We expect the prioritizing and listing of taxa to be dynamic, not only as these knowledge gaps are filled but also as environmental, agricultural, international trade, and societal factors change. Key words: Gastropoda, invasive species, life-history, natural ecosystems, pests Alien species are being moved around the world at unprecedented rates as a result of the globalization of trade and the increased ability of people to travel widely. These alien species have serious impacts on agriculture, the natural environment, commerce, and human health and well-being (Bright 1998, Cox 1999, Mack et al. 2000, Staples and Cowie 2001 ), and these effects maybe complex (Didham et al. 2007). In the United States, annual costs associated with damage to the environment and to agriculture caused by alien species have been most recently estimated as US$ 1 20 billion ( Pimentel etal. 2005). Combined costs for the United States (Pimentel et al. 2000), the United Kingdom, Australia, South Africa, India, and Brazil have been estimated as US$314 billion per year (Pimentel et al. 2001). Although the level of uncertainty is high, these estimates indicate that the problem is severe. While much attention is paid to invasive plants (e.g., Gordon etal. 2008), insects (Simberloff 1986), and pathogens (Palm 2001), with some notable exceptions (e.g., zebra mussels (Dreissena polymorpha (Pallas, 1771)): Britton and McMahon 2005; apple snails (Pomacea spp.): Hayes et al. 2008; New Zealand mud snails (Potamopyrgus antipodanim (Gray, 1853)): Kerans et al. 2005, Hall et al. 2006), molluscs receive relatively little attention (Keller et al. 2007). Nonetheless, invasive molluscs can have important impacts on agriculture (Godan 1983, Henderson 1989, 1996, Barker 2002a), bio- diversity (Coote and LoWe 2003, Lydeard et al. 2004), and human health (Madsen and Frandsen 1989, Pointier et al. 2005, Hollingsworth and Cowie 2006, Boaventura et al. 2007, Hollingsworth et al. 2007) and can become major public nuisances (Civeyrel and Simberloff 1996). Quarantine measures to limit the spread of invasive species include pre-introduction screening of species to assess their potential for invasiveness (Ruesink et al. 1995). Formal systems of weed risk assessment have been put into regulatory use widely for plants (Gordon et al. 2008), driven in part by the continuing demands of the global horticulture trade to move many species to new localities, with the horticultural industry playing probably by far the most important role in the introduction of invasive plants (Dehnen-Schmutz etal. 2007). Similar science-based risk assessment protocols based on the guidelines of the International Plant Protection Convention (IPPC) have been developed by Australia, New Zealand, and 113 114 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 other countries for other major groups of organisms. There have been many assessments of individual species of concern {e.g., Ruesink et al. 1995) and many jurisdictions have lists of prohibited species, but for the most part these have not been developed by applying objective, science-based, standardized protocols. Some countries have nascent protocols but have yet to implement them widely {e.g., Mito and Uesugi 2004, Gederaas etal. 2007). Many studies of various animal and plant groups, re- viewed by Kolar and Lodge (2001) and Hayes and Barry (2008), have attempted to develop formal screening proto- cols by assessing potential risk based on suites of characters thought a priori to correlate with invasiveness, e.g., in fish (Kolar and Lodge 2002), birds (Veltman et al. 1996, Duncan et al. 2001), and reptiles and amphibians (Bomford et al. 2008). The goals of such screening systems are primarily to provide an objective means of analyzing the legal, deliberate import of alien species. But they could also be used to allocate special attention to the interception of species trans- ported inadvertently that are potentially invasive. However, increasingly it is being suggested that any species-level characteristics that might identify successful invaders are both taxon and location specific (Sakai et al. 2001, Hayes and Barry 2008), and general approaches to risk analysis of potential invasive species remain challenging (Stohlgren and Schnase 2006). With some notable exceptions, most alien snails and slugs are transported inadvertently (Cowie and Robinson 2003). Quarantine agencies around the world routinely intercept numerous species of snails and slugs. Robinson ( 1999) listed those that were intercepted by U. S. quarantine officials between 1993 and 1 998. The purpose ofthe present study was, on behalf of the American Malacological Society (AMS) and at the request of the U. S. Department of Agriculture, Animal and Plant Health Inspection Service, Plant Protection and Quarantine (USDA-APHIS-PPQ), to develop a much shorter list of the snail and slug species considered as top priority for prevention of their introduction and establishment in the United States. This list would then be used by USDA-APHIS- PPQ officials as a list of species of quarantine importance to the United States and upon which to focus their attention. A preliminary version ofthe list (Cowie 2()()2a) was submitted to the USDA; the present paper is a revised version based on further analysis and more extensive review of the literature. MATERIALS AND METHODS Scope Species to be considered were species not present in the United States or, if present, only distributed highly locally and with the possibility of eradication or at least containment. A number of species found only in Hawaii, although wide- I spread there, were considered containable with respect to invasion of the remainder of the United States and were • therefore included. Only species falling under the jurisdiction , of USDA-APHIS-PPQ were included, that is, pest species with i the potential to cause damage to either agriculture or natural i ecosystems. Marine species (the responsibility of the National ' Marine Lisheries Service) were excluded, as were species only f affecting endangered species (the responsibility of the U. S. Pish and Wildlife Service). However, we treated these con- straints fairly broadly because they are often inter-related and : considered pest problems in four areas: agriculture (including i livestock health), environment, human health, and commerce. . Initially, the charge from the USDA was to generate a list of ■ 15 species, selected and prioritized using an explicit protocol. ) It soon became clear that a simple list of 15 species would not serve the interests of USDA-APHIS-PPQ adequately, for the I following reasons. ( 1 ) Most snail and slug species are generalist i herbivores. They do not in general exhibit the kind of precise i host-specificity exhibited, for instance, by many of the insect t pests upon which PPQ focuses greater attention. Congeners (and even less closely related species) are therefore likely to i have similar feeding habits, and listing just one species would b exclude other, related species that may not differ markedly in i pest potential. (2) Detailed information regarding species- ", level differences in feeding preferences among related species »i is available for few taxa. Therefore, listing one and not others i; of a number of species in a group {e.g., a genus) might again i divert attention away from potential pests. (3) Distinguishing n closely related species is difficult even for experts in the group i and would be impossible for PPQ field personnel without / extensive training, except in certain clear cases. (4) Limiting c the list to just 15 species could result in a focus on only a few r taxonomic groups that include multiple species considered ! potential pests while omitting species in other groups that i might be equally problematic but for which information was B limited. Conversely, selecting 15 well-known species from 1 a range of larger groups might also have meant omitting i other species in those groups that were potential pests. Lor c these reasons, we decided to create a prioritized list of larger 'J taxonomic groups (families) with a number of known or ( potential pests considered within each. Development of an initial iinranked list Focusing primarily on species intercepted by LISDA- i APHIS-PPQ (Robin.son 1999; D. G. Robin.son, unpubl. data), , we developed a preliminary list by scanning the literature on i mollusc pests worldwide, including primarily Godan ( 1983), ! Henderson ( 1989, 1996), Barker (2002a), augmented by our own knowledge. Some well-known pests were immediately excluded from the list because they were already widely distributed in the United States, e.g., Pcroecra^ rctieulatuni GASTROPODS OF QUARANTINE IMPORTANCE IN U.S.A. 115 (Muller, 1774) (Barker 2002a), Cornu aspersiim (Muller, 1774) (Dundee 1974, Roth and Sadeghian 2003). Other less well- known taxa were evaluated provisionally but omitted from the list, including, notably, the following. Brndybaena similar is (Rang, 1831) (Bradybaenidae). This species is probably already too widespread in the United States, occurring in much of the southeast (Dundee 1974). Otala lactea (Muller, 1774) (Helicidae). This species is a minor plant pest but is probably already too widespread in the United States, as it is known from southeastern states, Arizona, and a number of counties in California (Roth and Sadeghian 2003). Theba Risso, 1826 (Helicidae). Theba pisana (Muller, 1774) is a serious pest (Baker 1989, 1991, 2002, Coupland 1996) , currently confined to a small number of localities in southern California (Roth and Sadeghian 2003), and is included in the list. However, no other species in the genus appears to have pest potential as none is referred to in the pest snail literature. Trochulus Chemnitz, 1786 (Hygromiidae). There is no clear evidence that these species have pest potential (D. G. Robinson, unpubl. data) and they are not mentioned widely in the pest snail literature. Xerotricha Monterosato, 1892 (Hygromiidae). Xerotricha conspurcata (Draparnaud, 1801) is established in four or five counties in the San Erancisco Bay area, and although USDA-APHIS-PPQ still takes action on it when intercepted, the agency decided some years ago not to address these infestations. We therefore excluded it and other Xerotricha spp. from our analyses. Milax gagates (Draparnaud, 1801) (Milacidae). This species is a major pest in Europe and elsewhere (Barker 2002a) but is already probably too widespread in the United States, occurring in much of eastern North America, the Pacific Northwest, and California (Pilsbry 1948, Roth and Sadeghian 2003). Gonaxis Taylor, 1877 (Streptaxidae). At least two species of Gonaxis have been introduced to Hawaii as putative biocontrol agents for Achatina fulica Bowdich, 1822 (Cowie 1997) . However, although they have been implicated in the decline of native snail species, there is no evidence that they are a serious problem on the scale of that caused by the better known predator Euglandina rosea (Eerussac, 1821) (Cowie 2001a). They are not listed by Robinson (1999) as having been intercepted and there is no intention of introducing them deliberately to the mainland United States. Subulinidae. Too little is known of the pest potential of subulinids; they are rarely mentioned in the pest literature; and a number of species are already widespread in the United States (Robinson and Slapcinsky 2005). Belocaulus angustipes (Heynemann, 1885) (Veroni- cellidae). This slug may not be important as a major plant pest but is known as a disease vector (Rueda et al. 2002), although it is probably already too widespread in the United States (D. G. Robinson, unpubl. data). Aegopinella nitidula (Draparnaud, 1805) (Zonitidae). This small European land snail has been reported in British Columbia, with the suggestion that it could affect the native land snail fauna through predation (Forsyth et al. 2001). However, there is no evidence of this and it is not listed by Robinson (1999) as having been intercepted. Pomacea diffusa Blume, 1957 (Ampullariidae). We include all other species of Pornacea Perry, 1810, but this species, which is often referred to incorrectly as Pornacea bridgesii (Reeve, 1856) (Rawlings et al. 2007), has been considered a microherbivore (feeding on algae) (Howells 2002) and therefore not a potential pest, although its food preferences may be wider (Aditya and Raut 2001 ). It is also widely used as a domestic aquarium snail. Regulatory changes have banned live Pornacea spp., with the exception of P. bridgesii {i.e., P. diffusa), from any United States trade. Potamopyrgus antipodarum (Gray, 1853) (Hydrobiidae). This freshwater species may outcompete native species and change stream ecology but is probably already too widespread in the United States to be eradicated or contained, having been found in ten western states, as well as in the Great Lakes (Kerans et al. 2005, Hall et al. 2006, Bersine et al. 2008). Thiaridae. Within this freshwater family, the two most invasive species, Melanoides tubercidata (Muller, 1774) and Tarebia granifera (Lamarck, 1816), are already too widespread in the United States, the former having been reported from at least 15 states, the latter from seven (Dundee and Paine 1977, Burch and Tottenham 1980, Mitchell et al. 2007, NatureServe 2008). Triculinae (Pomatiopsidae). Some of these freshwater taxa transmit Schistosoma and most triculines can transmit Paragonimus, helminth parasites infecting people (Davis et al. 1999). However, none of them is a threat, as their ecological requirements probably cannot be met in the United States (G. M. Davis, pers. comm.). Consultation with the malacological community Having developed a preliminary version of this list we disseminated it widely over the Internet, primarily through the MOLLUSCA listserver, with an explanation of the purpose of the project and a request for comments and suggestions of additional or alternative species to include on it. The MOLLUSCA listserver has approximately 1,000 members throughout the world. The message was also sent to the AMS membership of about 340 malacologists although many of these are also subscribers to MOLLUSCA. Responses were received from over 20 people. The first author also presented a talk at the 2002 annual meeting of the AMS, outlining the progress of the project and again requesting 116 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 input. A number ot helpful comments were made by various conference attendees. All these comments were considered when developing the final prioritized list. Scoring taxa and prioritizing the list Eollowing this consultation phase we evaluated each of the species or species-groups in the list according to 12 non- exclusive attributes that are generally thought to correlate with a species’ invasiveness and that seemed particularly pertinent to non-marine molluscs (e.^., Veltman etal. 1996, Goodwin et al. 1 999, Lockwood 1 999, Duncan et al. 200 1 , Kolar and Lodge 2001, Sakai et al. 2001, Daehler et al. 2004, Leung et al. 2004, Marchetti et al. 2004, Theoharides and Dukes 2007, Alonso and Castro-Diez 2008, Bomford et al. 2008, Hayes and Barry 2008). Our evaluations were based on information obtained via a thorough search of the literature. Species and species groups were scored by giving them a ‘1’ if the data suggested that an attribute would enhance their pest potential and a ‘0’ if the data suggested it would not do so. If an attribute was mixed or would enhance pest potential only somewhat, we scored it as ‘0.5’, and if the data were insufficient, we did not assign a score. We were conservative in using 0.5 or not assigning a score if there was any question about giving 1 or 0. for each species or group we summed the scores to obtain S, a simple measure of the pest potential of each species or group. This measure, however, downplays a species’ pest potential when fewer attributes can be scored {i.e., when we had less knowledge). We therefore also divided each value of S by the total number of attributes scored, to obtain P, a proportional measure of pest potential not influenced by the number of scores, and ranging from 0 to 1, least to greatest concern. The species/groups were then ranked from highest to lowest based on the values of S and P. The attributes scored included both biological attributes of the species and attributes related to their interaction with people. The biological attributes evaluated were as follows. Range. If a species has a wide natural climatic range, it coulci invade a larger area within the United States. For example, among the Ampullariidae, one or more species of Pomacea occur from temperate Argentina to the Amazon basin and have the potential to spread widely in the United States (.scored as 1 ), contrasting with the two species of Marisa (scored as 0), which arc more restricted in South America and thus probably less likely to become widespread in the United States (Rawlings et al. 2007, Hayes et al. 2008). Similarly, among Helicidae, Otala punctata is confined to the western Mediterranean, primarily close to the coast, a limited climatic range (scored as 0), whereas riieha pisana occurs from the southwest of the British Isles to the eastern Mediterranean, a much wider geographic span, but nonetheless almost exclusively close to the coast (Cowie 1990), and therefore T. pisana was scored as 0.5 rather than 1. The extent ol the natural ranges of some species has been confounded by human- mediated spread, e.g. Archachatina marginata and Achatina fulica (Raut and Barker 2002), or by misidentification, e.g., , Achatina achatina (Bequaert 1950), and are probably smaller ; than sometimes supposed. Nevertheless, A. fulica may have . the potential to spread widely within the United States (Smith ) 2005). Ranges were determined by scanning the literature, web sites, and from our personal knowledge. Detailed data for many species are unavailable, and while those with very i wide or very narrow ranges are easy to assess, others are more i difficult. Our scoring of range size was thus in some cases . somewhat subjective. Phylogenetic relationships. If a species is closely related ■ to known pests (pest status assessed below), the likelihood of f it becoming a pest is greater (Hayes et al. 2008, examples in n Barker 2002a). We scored taxa as 1 if in the same or a very i closely related genus as a known serious pest, 0.5 if in a less / closely related genus or in the same or a very closely related >*. genus as a less serious pest, and 0 if more distantly related to iJ any known pest. Species known themselves to be serious pests were scored as 1. Adult size. Larger species are favored for deliberate 1 introductions (Mead 1979, Smith 2005, Thiengo et al. 2007) but for inadvertent introductions smaller species have a i greater chance of evading quarantine (Cowie and Robinson 2003). For species we knew to be introduced predominantly ;l deliberately, we scored large size (maximum shell dimension of snails and maximum extended length of slugs roughly >2 F cm) as increasing invasive potential (1), whereas for species e introduced primarily accidentally we scored small size »; ( roughly < 1 cm) as increasing invasive potential. Deliberately H introduced taxa <1 cm and accidentally introduced taxa t: >2 cm were scored as 0. Intermediate-sized snails (1-2 cm), ( regardless of mode of introduction, were scored as 0.5. ? Assessments were based on information from basic field ) guides and the taxonomic literature, augmented by our ii knowledge of probable modes of introduction {e.g., Cowie k 1998a, Cowie and Robinson 2003). Egg/juvenile size. Production of smaller and therefore i more readily dispersed offspring could lead to a species’ more > rapid and wider dispersal once introduced (cf. Vagvolgyi 1 975, < Paulay and Meyer 2002). Egg size is rellected by hatchling size > and is broadly correlated with adult size ( I leller 2001 ). 1 Idler i (2001) tabulated known egg sizes Idr terrestrial species and ’ we augmented those data with information for addititmal species from other published sources: Barrientos (1998) i (Ovachlaniys fulgens)-, Staikou and Lazaridou-Dimitriadou i (1991) (AVro/nV/u); I’hompson (1957) (Euglandina)', 'Wwncr ' and McCabe ( 1990) and Bai nes c/ 7 mm as large (0), and those between these sizes as intermediate (0.5). Heller (2001) gave ranges of sizes for some species and we have combined some species into groups {e.g., Pornacea, Helix) for our analyses. Thus, for the few taxa in which egg size data straddled these categories, we were conservative and scored them as 0.5. Reproductive potential. In general, larger snails produce more eggs over their lifetime (Heller 2001) although there is great variation in both longevity and productivity among species. However, if a species produces large numbers of young in a short period of time, e.g., an annual reproductive season, the chances of it being more invasive may be greater (Keller et al. 2007). Annual productivity data were obtained from: Hodasi (1979) [Achatina achatina)-, Raut and Barker (2002) {Achatina fulica); Plummer (1975) {Archachatina marginata)-, Barrientos (1998) {Ovachlamys fulgens); Cowie (1984) and Baker (1991) {Theba pisana, Cernuella virgata (da Costa, 1778)); Baur and Raboud (1988) [Arianta arbustorurn)-, Lazaridou and Chatziioannou (2005) {Xerolenta obvia); Baker and Hawke (1991) {Cochlicella acuta); Rueda et al. (2002) {Sarasinula plebeia, Leidyula moreleti); Cowie (2002b) {Pornacea); Keller et al. (2007) {Marisa cornuarietis, Biomphalaria glabrata); Dillon (2000; annualized from data in his table 4.1) {Biomphalaria, Bulinus). Eor Limicolaria aurora we used data from a conge- neric (Ergonmwan 2007). We scored mean per snail annual production of >1,000 eggs as 1, of 500-1,000 eggs as 0.5, and of <500 eggs as 0. In some cases productivity appears highly variable among regions, straddling categories {e.g., Achatina fulica; Raut and Barker 2002); we scored these as 0.5. Semelparous or iteroparous. Semelparous species put all their reproductive effort into a single reproductive event (or season), a life-history trade-off that results in a shortened life- cycle. Semelparity is probably correlated with high reproductive potential so semelparous species may be more invasive than iteroparous species (Dillon 2000, Heller 2001, Barker 2002b). We treated species with an annual (or shorter) life-cycle as semelparous, scoring them as 1. Other species were scored as semelparous if they breed only during one season before dying, regardless of their overall life-cycle, which may be biennial or longer (Heller 2001). Iteroparous species, including some that reproduce more or less continuously over multiple years (Dillon 2000), were scored as 0. We based our scores on the following: Raut and Barker (2002) (Achatinidae); Txurruka et al. (1996) {Arion ater); South (1992) (Arionidae, Tandonia budapestensis); Barrientos (1998) {Ovachlamys fulgens); Baur and Raboud (1988) {Arianta arbustorurn); Cowie (1984), Baker (1989, 1991, 2002), and Baker et al. (1991) {Theba pisana, Cernuella virgata, Cochlicella spp.); Heller (2001), Staikou et al. (1988), and Staikou and Lazaridou-Dimitriadou (1991) {Helix, Xeropicta); Lazaridou and Chatziioannou (2005) {Xerolenta obvia); Barker (2002b) {Tandonia sowerbii); Cowie (2002b) (Ampullariidae); Dazo et al. ( 1966), Sturrock (1973), and Loreau and Baluku (1987) {Biomphalaria, Bulinus); Yapi et al. (1994) {Indoplanorbis exustus); Remais et al. (2007) {Oncomelania). Eobania vermiculata is “marginally iteroparous”, with most individuals reproducing only once but a significant number reproducing for at least one additional season (Lazaridou-Dimitriadou and Kattoulas 1991); we scored it as 0.5. In some cases we generalized from information for one or a few species, e.g., Xeropicta in our list: information ior Xeropicta vestalis (Pfeiffer, 1841) (Heller 2001) and Xeropicta derbentina (Krynicki, 1836) (Staikou and Lazaridou-Dimitriadou 1991, Kiss et al. 2005). Breeding system. Selling or parthenogenetic rather than outcrossing species may be better invaders (Eoltz et al. 1984, Baur and Bengtsson 1987, Dybdahl and Kane 2005). All ampullariids and pomatiopsids were scored as outcrossing (0) as they have separate sexes and no records of parthenogenesis (Dillon 2000, Cowie 2002b). All other species on the list are hermaphrodites. None exhibits parthenogenesis (Jordaens et al. 2007), but selling may occur to a greater or lesser degree in most species, along something of a continuum of strategies. Many normally outcrossing species may self under rare circumstances, especially if kept in isolation (Duncan 1975), though usually producing eggs/young at a very much reduced rate. Eor example, achatinids, helicids, and hygromiids are generally considered obligate outcrossers {e.g., Duncan 1975, Barker 1999, Raut and Barker 2002) although limited selling may be possible {e.g., Arianta arbustorurn; Heller 2001); all were scored as outcrossing. Arion lusitanicus is predominantly, if not exclusively, outcrossing (Eoltz et al. 1982). Some species adopt either strategy although in some cases selling only in isolation, e.g., Arion ater (Eoltz et al. 1982), Sarasinula plebeia (Rueda et al. 2002) and Laevicaulis alte (Duncan 1975); they were scored as 0.5. Most planorbids are capable of outcrossing and selling although preference for one mode or the other differs among species (Jarne et al. 1993, Dillon 2000, Jordaens et al. 2007). Even in a preferential outcrosser, Biomphalaria glabrata (Say 1818), there is little loss in productivity when forced to self (Paraense 1959). However, planorbids were scored as 0.5, since although the potential in some species to self without loss of fecundity is equivalent, from the current perspective, to being selfers, it is not known how widely this applies in the taxa considered. The capacity to self is widespread in Succineidae, but whether important in natural situations and in our listed taxa is not known (Barker 2001); they were not assigned a score. Ovachlamys fulgens seifs readily with no loss of fecundity and this may be the predominant mode (Barrientos 1998), as it is 118 AMERICAN MALACOLOGICAL BULLETIN 27 • 1 /2 • 2009 in Tnndonia budapestensis and Tandonia sowerbii (Eoltz et al. 1984); these were scored as 1. The human-interaction attributes evaluated were as follows. Introduction pressure. Frequent interception implies higher introduction pressure and hence greater likelihood of establishment (Cowie and Robinson 2003). Species listed by Robinson (1999: table 3) were the species most commonly intercepted by USDA-APHIS-PPQ during 1993-1998; those on our list we scored as 1. Robinson (1999) also mentioned Helix pomatia, Cantareus apertus, Achatina spp., and Archachatina marginata as being frequently intercepted; they also were scored as 1 . Others scored as 1 include Xeropicta spp., based on Kiss et al. (2005, reporting on Xeropicta derbentina), Succinea tenella, based on Cowie et al. (2008), Pornacea spp. and Marisa spp. because of their worldwide popularity in the aquarium trade ( Rawlings et al. 2007, Hayes et al. 2008), and a number of taxa based on data (D. G. Robinson, unpubl. data) accumulated since 1998 (Robinson 1999). We scored other species as 0.5 if they were listed by Godan ( 1983) or Robinson ( 1999) as having been intercepted entering the United States or Canada. Others were scored as 0, and no score was assigned if we were unsure of their introduction pressure. Invasion history. Invasiveness elsewhere in the world suggests a greater likelihood of becoming invasive in the United States. Species known to be invasive (as opposed to simply recorded as present, e.g., Macrochlamys indica: Robinson 1999; Barker and Efford 2004) elsewhere in the world (including Hawaii, as being distinct from the continental United States) were scored as 1 based on the literature, including the following: Mead (1979), Raut and Barker (2002), Smith (2005), and Thiengo et al. (2007) {Achatina fulica)-, Grimm (2001) and Shoaib and Cagan (2004) {Arion lusitanicus, Xerolenta obvia); Hollingsworth et al. (2007) {Pannarion niartensi)-, Robinson and Fields (2004) {Zachrysia provisoria); Robinson (1999) and Gowie et al. (2008) (Ovachlaniys fulgens)-, Baker (1989, 2002, 2008) (Theba pisana, Cernuella virgata, Cochlicella spp.); Kiss et al. (2005) (Xeropicta)-, Barker (1999, 2002a) (Tandonia budapestensis, T. sowerbii)-, Gowie et al. (2008) (Succinea tenella)-, Cowie (1998b) and Cowie et al. (2008) (Laevicaulis alte, Sarasinula plebeia, Veronicella ciibensis)-, Cowie (2002b), Rawlings et al. (2007), and Hayes et al. (2008) (Poniacea spp.); Coelho da Silva et al. (1997), Pointier et al. (2005), Majoros et al. (2008) (Indoplanorbis exustus, Bioinphalaria spp., with Hulinus spp. explicitly not considered invasive). Pila was .scored as 0.5 on the basis of its localized but .serious invasive status on one of the Hawaiian Islands (Tran et al. 2008), as was Liniicolaria aurora becau.se of its invasive status in Martinique (Raut and Barker 2002). Cantareus apertus, a Mediterranean species, is invasive in southern Germany (Godan 1983); I-iobania verniiculata, another Mediterranean specie.s, is locally established in California (Roth and Sadeghian 2003) and Japan (Ueshima et al. 2004), and mtiy be invasive; Tandonia rustica, a central European species is arguably invasive in Western Europe, where it is widespread (e.g., Philp 1987); all were scored as 0.5. Species that appeared not to have become i invasive anywhere or that were explicitly stated to be only i minimally invasive, were scored as 0. Species for which we were ■ unsure were not scored. Major pest elsewhere. If a species is a major pest elsewhere ; of a crop grown in the United States, or causes other major ) problems elsewhere (e.g., environmental damage, human I disease), there is a greater likelihood that it will cause serious u problems in the United States. Species scored as having a history of invasion (above) are often considered invasive on the basis of being major pests where introduced. The two attributes are closely linked. Some species, however, lack an extensive history of invasion but are pests (perhaps relatively minor pests) within their native ranges (e.g., Arion ater, Mariaella dussumieri). Many, if not most, snails and slugs can act as intermediate hosts of human and livestock parasites (Godan 1983, Grewal et al. 2003). Assessment of whether a species causes sufficient problems to be catego- rized as a major pest is somewhat subjective. We have been conservative in scoring as such only those taxa that are explicitly referred to in the literature as causing substantial problems. Many species have been reported as pests although many of them may cause little loss. Numerous crops have been listed as susceptible to damage by certain species but with no indication of the severity of the problem (e.g., Raut and Barker 2002: table 3.1). And some species have been reported as pests but only on the basis of occasionally being found in association with a particular crop, as we suspect is the case for many of the instances listed by Godan (1983). Our assessments were based primarily on the following: Raut and Barker (2002) (Achatinidae); Frank (1996) and Grimm (2001) (Arion lusitanicus)-, Godan (1983), South (1992), and Barker (2002b) (Arionidae, Milacidae); Kumar and Ahmed (2000) (Macrochlamys indica)-, Godan (1983) (Mariaella dussumieri, Pannarion martensi)-, Hollingsworth et al. (2007) (Pannarion martensi)-, Rohinson and Fields (2004) (Zachrysia spp.); Sanderson and Sirgel (2002) (Theba pisana)-, Godan (1983) (Helix, Arianta [-,\$ 'Helicigona'] arbustoruni, Cantareus apertus, Eobania verniiculata, Otala punctata, Xerolenta obvia)-. Baker (1989, 2002, 2008) and Coupland (1996) (Theba pisana, Cernuella virgata, Cochlicella spp.); Kiss et al. (2005) (Xeropicta); Cowie et al. (2008) (Succinea tenella); de lager and Daneel (2002) (Elisolimax flavesccns); Godan ( 1983), Raul ( 1996), 1 lala et al. ( 1997), Rueda et al. (2002), Fields and Robin.son (2004), USHA-APHIS-EEQ (2006), Hollingsworth et al. (2007), Naranjo-Garcia et al. (2007), and (iowie et al. (2008) ( Veronicellidae); Stange (2006) (Zachrysia provisoria, Ovachlaniys fulgens, Veronicella sloanii); (iowie (2002h), loshi and Sebastian (2006), and GASTROPODS OF QUARANTINE IMPORTANCE IN U.S.A. 119 Rawlings et al. (2007) (Ampullariidae); Stevens (2002) and Pointier et al. (2005) (Planorbidae); Davis et al. (1999) [Oncomelania) . A ‘'miilti-pest’l The severity of the problems caused has been scored above, according to whether a species is a major pest. Here we score species as 1 if they cause problems in more than one of agriculture (including livestock health), environment, human health, and commerce, regardless of degree. Thus, for example, Achatina fulica is not only a serious plant pest but also an important vector of parasitic diseases, as well as a major public nuisance (Mead 1979, Civeyrel and Simberloff 1996, Raut and Barker 2002, Smith 2005, Thiengo et al. 2007); Vewnicella cubensis is an important parasitic disease vector (Hollingsworth et al. 2007) as well as an agricultural and garden pest; Pomacea spp. are major crop pests (Cowie 2002b, Joshi and Sebastian 2006) and important parasite vectors (Hollingsworth and Cowie 2006) . Other species may cause serious problems in one area but only minor problems in another. Por instance, Parmarion martensi is a plant pest in Malaysia (Godan 1983) and a possibly serious human disease vector in Hawaii (Hollingsworth et al. 2007) . Other taxa cause problems in more than one area but they are not severe in either. For example, Arion ater is a minor crop pest and also causes environmental damage by feeding on young tree seedlings (South 1992); Pila spp. are local or minor crop pests (Cowie 2002b, Levin et al. 2006) and recognized parasite vectors (Hollingsworth and Cowie 2006), as are Indoplatwrbis exustus (Stevens 2002), Thelidornus aspera (Lindo et al. 2002), ^nd. Diplosolenodes occidentalis (Rueda et al. 2002); Laevicaulis alte is a disease vector, although not as important as Vewnicella cubensis or Parmarion martensi (Hollingsworth etal. (2007), and a relatively minor plant pest (Raut 1996). All were scored as 1 . Economic potential. We evaluated whether the problems a species could cause would be likely to result in major economic loss in the United States, including costs of control or eradication. This attribute overlaps with the attribute of being a major pest elsewhere, but is explicitly focused on economic cost. Our evaluation was based on the likelihood of the taxon becoming widespread in the United States and on either quantified assessments of costs in other regions, e.g., Baker (1989) {Cernuella virgata, Cochlicella spp., Theba pisana), Andrews (1989) [Sarasinula plebeia), Cheng (1989), Naylor (1996), and Levin et al. (2006) (Pomacea), or unquantified statements of the pest’s economic importance, e.g.. Mead (1979) and Raut and Barker (2002) (Achatina fulica). Prank (1996) and Grimm (2001) (Arion lusitanicus). South (1992) (Tandonia budapestensis), de lager and Daneel (2002) (Elisolimax flavescens). If we found no report in the literature explicitly indicating major economic costs or only highly localized costs, or found explicit statements that a species/ group was not a major economic problem we scored it as 0, e.g., Archachatina marginata and Lirnicolaria aurora (Raut and Barker 2002). Others, for which the economic literature was limited or equivocal, or for which we considered the potential economic costs unlikely to be widespread were scored as 0.5, e.g., Achatina achatina (Raut and Barker 2002), Zachrysia provisoria (Robinson and Pields 2004), Ovachlamys fulgens (Stange 2006, Cowie et al. 2008), Tandonia spp. (South 1992), Veronicella cubensis (USDA-APHIS-PPQ 2006), Veronicella sloanii (Stange 2006), Pila spp. (Cowie 2002b, Levin et al. 2006). Validating the model We assessed the appropriateness of the model by scoring the following representative suite of species that have already been introduced to the United States and determining whether it would accurately predict their invasion status. We ex- cluded information from the United States when scoring the species’ attributes, to avoid circularity. We selected a non- random sample of taxa that ( 1 ) have been subject to relatively substantial amounts of research in the United States, so that there is an appropriate level of knowledge of their distributions and impacts, (2) are already widespread in the United States, and (3) represent a range of impacts. Scores of attributes were obtained as follows. Deroceras reticidaturn (Muller, 1774) (terrestrial slug, Agriolimacidae): native range (Kerney and Cameron 1979, Barker 1999), adult size (Kerney and Cameron 1979), egg size, reproductive potential, and semelparity/iteroparity (Heller 2001), breeding system (Loltz et al. 1984), introduction pressure (Robinson 1999), invasion history (Barker 1999), pest status, and economic damage (Barker 2002a). Cepaea nernoralis (Linnaeus, 1758) (terrestrial snail, Helicidae): native range, adult size, and invasion history (Kerney and Cameron 1979), phylogenetic relationships (scored as 0.5 since Cepaea is somewhat closely related to Cornu), egg size (Heller 2001), reproductive potential and semelparity/ iteroparity (Cowie 1984), breeding system (helicids in gen- eral are outcrossers; Duncan 1975), introduction pressure (Robinson 1999), pest status, and economic damage (Godan 1983, Henderson 1989, 1996, Barker 2002a). Cornu aspersum (Muller, 1774) (terrestrial snail, Helicidae): native range and adult size (scored as 0.5 because it is prob- ably introduced both deliberately and accidentally: Barker 1999) (Kerney and Cameron 1979), egg size (Heller 2001), reproductive potential (Desbuquois et al. 2000), semelparity/ iteroparity (R. H. Cowie, pers. obs.), breeding system (Selander and Hudson 1976), introduction pressure (Robinson 1999), invasion history (Barker 1999), pest status, and economic damage (Godan 1983, Sanderson and Sirgel 2002). Potamopyrgus antipodarum (Gray, 1853) (freshwater snail, Hydrobiidae): native range, adult size, reproductive potential, breeding system, invasion history (Alonso and Castro-Diez 2008, Radea etal. 2008), juvenile size (Radea etal. 120 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 2008), semelparity/iteroparity (Winterbourn 1970), invasion pressure (Robinson 1999, Alonso and Castro-Diez 2008), and pest status (Alonso and Castro-Diez 2008, Holomuzi and Biggs 1999). Milax gagates (Draparnaud, 1801) (terrestrial slug, Milacidae): native range and adult size (Kerney and Cameron 1979), egg size (Heller 2001), semelparity/iteroparity (South 1992), breeding system (Loltz etal. 1984), introduction pressure (Robinson 1999), invasion history (Barker 1999), pest status, and economic damage (Godan 1983, Henderson 1989, 1996, South 1992, Barker 2002a). Rumina decollata (Linnaeus, 1758) (terrestrial snail, Subulinidae): native range (Batts 1957), phylogenetic relation- ships (scored as 0 as it is not known as a pest nor closely related to a known pest), adult size (scored as 0.5 because it is probably introduced both deliberately and accidentally: Cowie 2001a), invasion history (De Lrancesco and Lagiglia 2007), egg size (Heller 2001), reproductive potential (extrapolated from Batts 1957, Selander and Hudson 1976, Lisher and Orth 1985), semelparity/iteroparity (Dundee 1986), breeding system (Batts 1957, Selander and Hudson 1976, Lisher and Orth 1985), pest status, status as a “multi-pest”, and economic damage (Cowie 2001a). Melanoides tuberculata (Muller, 1774) (freshwater snail, Thiaridae): native range (not assigned a score because it is now so widespread that its true region of origin and its extent is not known), phylogenetic relationships (scored as 1 as it is itself a minor pest), adult size (Dudgeon 1986, Pointier et al. 1994) (scored as 0.5 because it is probably introduced both deliberately and accidentally: Cowie and Robinson 2003), juvenile size (Dudgeon 1986, Pointier et al. 1992), reproductive potential (Berry and Kadri 1974), semelparity/ iteroparity (Berry and Kadri 1974, Dudgeon 1986, Pointier et al. 1992), breeding system (Berry and Kadri 1974, Dudgeon 1986, Ben-Ami and Heller 2007), introduction pressure (Robinson 1999), invasion history (Berry and Kadri 1974, Dudgeon 1986, Pointier et al. 1994, Pointier 1999, Cowie 2001b), pest potential, status as a ‘multi-pest’, and economic damage (Berry and Kadri 1974, Dudgeon 1986, Pointier 1999, Ben-Ami and Heller 2001 ). RESULTS AND DISCUSSION The prioritized list We created a ranked list of 46 species or groups of species repre.senting 18 families ('fable 1). Ranks based on simple (S) and proportional (P) values for each taxon were generally similar. However, some species exhibited relatively large disparities between the two scores although none rellected grossly different placement of the.se species in the overall rankings, for instance Irom the top to the bottom third. Nevertheless, we argue that the rank based on P values probably captures the true pest potential better, as it is less biased by the number of attributes it was possible to score for a particular species. The S rank will inevitably increase as more attributes are scored (unless they are all scored as 0), which is not the case for the P rank. The data for the individual attributes on which these scores and ranks are based are provided (Appendix 1). The evaluated species/groups belong to 18 families (Table 2). The top-ranked 12 species or groups fell in eight families, and these eight families included 28 of the 46 taxa evaluated (Table 2). The top-ranked potential pest groups were the Ampul- lariidae and Hygromiidae (Table 2). The former ranked highly because of Pomacea spp. These freshwater snails have become major pests of rice and other crops in southeast Asia and Hawaii (Joshi and Sebastian 2006). Pour species of Pomacea have been introduced to the continental United States, where they threaten rice crops and natural ecosystems (Rawlings et al. 2007). Cowie and Thiengo (2003) recognized 1 1 7 nomenclaturally valid species, many of which may have a similar pest potential to those already introduced. Hygromiids ranked highly because of Cermiella spp. and Xeropicta spp. Cernuella virgata has become a major cereal and pasture pest in Australia (Baker 2002). These and many other hygromiids are especially prone to being introduced in association with domestic tiles imported to the United States from southern Europe (Robinson 1999). Some of them also occur in temperate localities in their native Europe (Kerney and Cameron 1979) and collectively they thus have the potential to invade large parts of the United States. Helicidae and the closely related Cochlicellidae ranked immediately below the ampullariids and hygromiids (Table 2). Helicids ranked highly essentially because of the value for Theba pisana (Table 1, Appendix 1). The value for cochlicellids (Appendix 1) was based on information for Cochlicella acuta (Muller, 1774) and C. barbara (Linnaeus, 1758). Both T. pisana and these cochlicellids have become pests in various parts of the world where they have been introduced, notably in Australia where they are major cereal and pasture pests (Baker 2002). Theba pisana is also a pest of grape vines in South Africa (Sanderson and Sirgel 2002) and was formerly an important citrus pest in California but was thought to have been eradicated (Armitage 1949). It has now reappeared but is not widespread (Id)th and Sagedhian 2003). Veronicellid slugs ranked next highest ('fable 2). Veroni- cellids are large slugs. Sarasinula plcbeia and Vewnicella cubensis especially can become extremely abundant and are important pests in numerous crops, horticultural facilities, and gaixlens, and can become a public nuisance in urban/ suburban areas (Riieda et al. 2002, Naranjo-Carcia el al. 2007, R. 1 1. Cowie, pers. t)bs.). l.iicvicaiilis alle is less well recognized GASTROPODS OF QUARANTINE IMPORTANCE IN U.S.A. 121 Table 1. List of mollusc species and species-groups of potential major pest significance to the United States, ranked according to their pest potential from greatest ( 1) to least (46). S and P denote Simple and Proportional values and the ranks based on them (see methods). Species/species-group Cernuella Schliiter, 1838 Pomacea Perry, 1810^ Cochiicella Ferussac, 1821 Theba pisana (Muller, 1774) Sarasimila plebeia (Fischer, 1868) Xeropicta Monterosato, 1892 Laevicaulis alte (Ferussac, 1822) Succinea tenella Morelet, 1865‘^ Veronicella cubensis (Pfeiffer, 1840) Achatina fulica Bowdich, 1822 Indoplanorbis exustus (Deshayes, 1834) Biomphalaria Preston, 1910‘* Bi(/im/s Muller, 1781 Ovachlamys fidgens (Gude, 1900) Zachrysia provisoria (Pfeiffer, 1858) Tandonia budapestensis (Flazay, 1881) Xerolenta obvia (Menke, 1828) Arion lusitaniciis Aucl., non Mabille, 1868'’’ Elisolitnax flavescens (Keferstein, 1866) Marisa Gray, 1824 Parrnarion martensi Simroth, 1893 Pila Roding, 1798 Tandonia sowerbii (Ferussac, 1823) Cantareus apertus (Born, 1778) Eobania verrnicidata (Muller, 1774) Veronicella sloanei (Cuvier, 1817) Diplosolenodes occidentalis (Guilding, 1825) Macrochlamys indica Godwin-Austen, 1888 Succinea s.g. Calcisuccinea Pilsbry, 1948* Arion ater (Linnaeus, 1758) Oncomelania Gredler, 1881 Enidae Woodward, 1903 Achatina achatina (Linnaeus, 1758) Thelidomiis aspera (Ferussac, 1821) Zachrysia auricoma (Ferussac, 1821) Eitglandina Crosse and Fischer, 1870® Tandonia rustica (Millet, 1843) Helix Linnaeus, 1758 Limicolaria aurora (lay, 1839) Otala punctata (Muller, 1774) Archachatina marginata (Swainson, 1821) Mariaella dussumieri Gray, 1855 Arianta arbustorum (Linnaeus, 1758) Acusta touranensis (Souleyet, 1842) Family^ S score Hygromiidae 9.5 Ampullariidae 9.5 Cochlicellidae 9.0 Flelicidae 9.0 Veronicellidae 6.5 Hygromiidae 6.5 Veronicellidae 5.5 Succineidae 5.5 Veronicellidae 5.5 Achatinidae 7.5 Planorbidae 7.5 Planorbidae 7.0 Planorbidae 6.5 Chronidae 7.0 Pleurodontidae 4.5 Milacidae 5.5 Hygromiidae 5.5 Arionidae 5.5 Urocyclidae 4.0 Ampullariidae 5.0 Ariophantidae 4.0 Ampullariidae 5.0 Milacidae 5.0 Helicidae 4.5 Helicidae 5.0 Veronicellidae 3.0 Veronicellidae 2.5 Ariophantidae 2.5 Succineidae 2.5 Arionidae 4.0 Pomatiopsidae 4.0 Enidae 3.5 Achatinidae 4.0 Pleurodontidae 2.5 Pleurodontidae 2.5 Spiraxidae 2.5 Milacidae 2.5 Helicidae 3.0 Achatinidae 3.0 Helicidae 3.0 Achatinidae 3.0 Ariophantidae 2.0 Helicidae 2.5 Bradybaenidae 1.5 P score S rank P r 0.79 1 ] 0.79 1 1 0.75 3 3 0.75 3 3 0.72 9 5 0.72 9 5 0.69 12 7 0.69 12 7 0.69 12 7 0.68 5 10 0.68 5 10 0.64 7 12 0.59 9 13 0.58 7 14 0.56 22 15 0.55 12 16 0.55 12 16 0.50 12 18 0.50 24 18 0.50 18 18 0.50 24 18 0.50 18 18 0.50 18 18 0.45 22 24 0.45 18 24 0.43 30 26 0.42 35 27 0.42 35 27 0.42 35 27 0.40 24 30 0.40 24 30 0.39 29 32 0.33 24 33 0.31 35 34 0.31 35 34 0.28 35 36 0.28 35 36 0.27 30 38 0.27 30 38 0.27 30 38 0.25 30 41 0.25 43 41 0.21 35 43 0.19 44 44 (continued) 122 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Table 1. (continued) Species/species-group Famil/ S score P score S rank P rank Leidyula moreleti (Crosse and Fischer, 1872) Veronicellidae 1.5 0.19 44 44 Zachrysia trinitaria (Pfeiffer, 1858) Pleurodontidae 1.0 0.13 46 46 All assignments to family from Robinson (1999), except for Wilke et al. (2001 ) for Pomatiopsidae and Vaught (1989) for Ariophantidae, while accepting that some are in flux {e.g., Wade et al. 2007). *’ All species of Pomacea except P. diffusa Blume, 1957, which is often referred to, incorrectly (Rawlings et al. 2007, Hayes et al. 2008), as P. bridgesii (Reeve, 1856), and the native P. paludosa (Say, 1829). ‘ May also include the similar Siiccinea horticola Reinhardt, 1877. All species of Bioinplialaria except the native B. obstructa (Morelet, 1849). ' Arion liisitankus Auct., non Mabille is now referred to as Arion vulgaris Moquin-Tandon, 1855 by many workers. Arioii lusitaukus Mabille, 1868 is increas- ingly acknowledged as a species of Mesarion Hesse, 1926, restricted to Spain and Portugal. The issue is not satishictorily resolved. '^Only species of Sucdiiea {Cakisuccinea) not native to the United States. ® Only species of Euglandma not native to the United States. as a major pest, but some of its other attributes resulted in a high ranking (Table 1, Appendix 1). This may be an instance in which differential knowledge of the attributes scored among these veronicellids resulted in a higher ranking of a species (L. alte) than its potential may warrant, relative to other species (S. plebeia and U ciibensis), and reflects the need for caution when interpreting the results of analyses of this kind when based on limited knowledge. The succineids, achatinids, and planorbids included the remaining taxa in the top ranked 12 (Table 1). In general, succineids have not been considered significant pests until recently as a number of species, notably Siiccinea tenella, are increasingly transported around the world in the horticultural trade (Cowie et al. 2008). What their impacts will be is not entirely clear. Achatina fiilica has often been thought of as one of the world’s worst land snail pests (Mead 1979, Raut and Barker 2002) and was the driver of the high ranking of the achatinids (Ttible 2). Like most snails and slugs, it can act as a vector of human and animal diseases and, with its large size and potential for explosive population growth following introduction, can become a major public nuisance (Foucher 1975, Civeyrel and Simberloff 1996). Other achatinids ranked much lower. 1 lowever, complacency about them would be misplaced, as little is known about the biology of most of them and many are difficult for untrained specialists to distinguish. Quarantine officials should be vigilant of any achatinids. The planorbids’ biological attributes make them potentially highly invasive (Appendix 1). The planorbids role as potential pests is primarily in the arena of human disease, as they are major parasite vectors. However, in this regard their potential is more difficult to evaluate than the more straightforward agricultural potential of most of the other taxa evaluated and it may be that sanitary conditions and people’s behavior may minimize the chance of the parasites cycling in the United States (D. S. Woodruff, pers. comm.). The potential of planorbids may be overestimated by our model. Table 2. Families ranked according to the highest rank achieved by a species or group of species in each family, with the number of species or groups ranked in the top 12 (based on P rank) for each family, and the total number of species or groups that we assessed in each family. Family Highest species or group rank (P/S) Number of species or groups in top 12 Total number of species or groups assessed Ampullariidae 1/1 1 3 Hygromiidae 1/1 2 3 Helicidae 3/3 1 6 Cochlicellidae 3/3 1 1 Veronicellidae 5/9 3 6 Succineidae 7/12 1 2 Achat inidae 10/5 1 4 Planorbidae 10/5 2 3 Chronidae 14/7 0 1 Pleurodontidae 15/22 0 4 Milacidae 16/12 0 3 Arionidae 18/12 0 2 Ariophantidae 1 8/24 0 3 Urocyclidae 18/24 0 1 Pomatiopsidae 29/24 0 1 Fnidae 31/29 0 1 Spiraxidae 36/35 0 1 Bradybaenidae 44/43 0 1 GASTROPODS OF QUARANTINE IMPORTANCE IN U.S.A. 123 Validation of the model To test the validity of our model, we scored a number of additional species and compared the outcome with their known status in the United States. Their attribute scores are available (Appendix 2). Deroceras reticulatum (Agriolimacidae) scored 7.5 (S value) and 0.68 (P value), ranking it 5 and 10, respectively, among the more serious ‘potential’ invaders, and appropriately pre- dicting its wide distribution and major pest status in the United States (chapters in Barker 2002a). Cepaea nemoralis (Helicidae) scored 3.5 (S) and 0.29 (P), ranking it 29 and 36, respectively, toward the bottom of the list. While it is widespread in the eastern United States (Brussard 1975, Wliitson 2005), it appears not to be a pest and although a role as a competitor of native snail species has been suggested (Whitson 2005), it has not been demonstrated. The prediction of the model, especially the P rank, which we deem more appropriate, concurs with the essentially non- pest status of this species in the United States. Cornu aspersmn (Elelicidae) scored 7.0 (S) and 0.58 (P), ranking it 7 and 14, respectively, among the top-ranked one third. While it is widely distributed in the United States (Roth and Sadeghian 2003), it is only a major agricultural pest, notably of citrus, in California (e.g., Sakovich 2002). Elsewhere it may be more of a garden nuisance. Nevertheless, its status as invasive in the United States is unquestionable and its ranking may reflect the relatively lesser relevance of its biological attributes (which included relatively few 1 scores) as opposed to its human interaction attributes (see discussion below). Thus, the model, at least regarding the P rank, may have underestimated its potential. Potamopyrgus antipodarum (Hydrobiidae) scored 7.5 (S) and 0.63 (P), ranking it 5 and 13, respectively, among the top third. This ranking is reflected appropriately in its increasing spread through much of the western United States and increasing but as yet somewhat limited documentation of its ecological impacts (Kerans et al. 2005, Hall et at. 2006). Milax gagates (Milacidae) scored 7.0 (S) and 0.58 (P), ranking it 7 and 14, respectively, also in the top third. Although widely distributed in the United States (Pilsbry 1948, Roth and Sadeghian 2003), the relative lack of literature {e.g., Godan (1983) reports it as damaging Brussels sprouts; it is not mentioned in the chapters of Barker (2002a) dealing with the United States) suggests that it has not yet become a major widespread pest. In this case the model (at least the S value) may have overestimated this species’ potential, although given its pest status in Europe, it would be unwise to assume this. However, simply changing the multi-pest score from 0 to 1 on the basis of its damaging endemic plants in Hawaii (Cowie 1997), changes its scores to 8.0 (S) and 0.73 (P), thereby ranking it 5 (both S and P ranks) and illustrating both the sensitivity of the ranking system to minor changes in the scores and perhaps the serious potential of this species as both an agricultural and environmental pest. Rumina decollata (Subulinidae) scored 5.0 (S) and 0.42 (P), ranking it 18 and 27, respectively. Having been initially introduced accidentally, it has now been spread deliberately as a putative control agent for Cornu aspersum, and is now found widely in southern states from the east coast to California (Cowie 2001a). It has not been considered a serious agricultural pest although it may occasionally become sufficiently abundant in domestic gardens to be considered a nuisance (Eisher and Orth 1985, Cowie 2001a). As a facultative snail predator, it has been suggested that it could affect native, including endangered, snail species, but any such impacts have not been documented (Cowie 2001a). Thus, its wide distribution but low, though not negligible, effects are reflected appropriately in its ranking in the middle third. Melanoides tuberculata (Thiaridae) scored 6.5 (S) and 0.59 (P), ranking it 9 (S) and 13 (P), among the top third. This ranking of its invasive potential is reflected in its presence in 15 states (Mitchell et al. 2007). Almost no studies have attempted to demonstrate any negative effects. However, it can reach high densities and acts as a vector of various trematode parasites. Thus it may have serious ecological impacts as a result of both competition with other freshwater organisms (including native snails and mussels) and transmission of parasites to fish (including endangered species) and indirectly to birds; it potentially may also have a human health impact as a result of the indirect transmission of trematodes to people (Mitchell et al. 2007). Broadly, the model appropriately predicted the invasive pest status of this range of species, suggesting that it works at a gross level. Nevertheless, it is clearly sensitive to minor scoring changes and to the scoring algorithm used, and because some of the scores, especially the human attribute ones, are somewhat subjective, the model can only provide a rather general categorization. Alternative models In addition to the uncertainty in an analysis of this kind resulting from a lack of adequate basic knowledge of the attributes scored, subjectivity in scoring some of them, and choice of ranking algorithm, one could arguably include other attributes or weight the attributes differentially, as certain ones may be more important than others in determining potential invasiveness. Notably, climate/habitat match, introduction pressure, and being invasive elsewhere seem to be especially important {e.g., Kolar and Lodge 2001, Theoharides and Dukes 2007, Bomford et al. 2008, Hayes and Barry 2008). However, weighting some attributes more than others would involve even more subjectivity than is already inherent in our model 124 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 and we preferred to take the more objective approach of not weighting. Nevertheless, some of the categories are strongly related to each other (e.g., invasion history, major pest elsewhere, economic potential) and by including scores for each of them we are in a sense positively weighting the more fundamental underlying attribute. Also, many of the biological attributes scored do indeed seem to be generally correlated with the human interaction attributes. Eurthermore, by scanning Appendix 1, it is possible to identify those species that, tor instance, are frequently intercepted, that are invasive/pests elsewhere, and so on, and to emphasize certain attributes in order to hne-tune or re-evaluate the ranking of a particular species or group of species. By doing so, it may be possible to tailor quarantine interventions to the threats from individual species or groups. In simplest terms, however, if a species is an invasive pest elsewhere and occurs in habitats/ climates represented in the United States, in the absence of any more sophisticated risk assessment, the simplest approach is to assume that it also has that pest potential in the United States. CONCLUSIONS Our extensive review of the pest snail and slug literature and consultation with the malacological community, combined with our testing of the model against known alien pests in the United States, makes us conhdent that our prioritized list does indeed include those taxa most likely to become pests in the United States if they breach quarantine and/or if they cannot be contained locally. The ampullariid genus Pomacea, hygromiids, Cochlicella spp., helicids (notably Theba pisaiui), veronicellids, succineids, achatinids (primarily Achatina fiilica), and planorbids topped the list. However, while the ranks, particularly the P ranks, assigned to these species/ groups may be reasonable approximations of the relative seriousness of their threats, they should not be adhered to rigidly. Similarly, paying strict attention to the relative rankings of the other taxa that constitute the remainder of the list is also probably not warranted, especially as these species rank as potential pests for a variety of rea.sons in addition to their potential specifically as agricultural pests. Other snail and slug species not listed may well have pest potential of which we are currently unaware or may develop pest potential as a result of future environmental changes, changes in agricultural practice, and changes in commercial activities including import/export routes and societal preferences. Notable among these are the numerous hygromiid species from around the Mediterranean, where the group exhibits immense diversity, exemplified by the long list of hygromiids given hy Robinson ( 1999: 438) and of ‘Hclicclla species given by Codan ( 1983: 272). A key need, however, is better knowledge of the basic biology of many of these potential pests, and rigorous documentation of the levels of damage they cause (including economic data) rather than statements such as ‘is a pest of legumes’ or ‘causes damage to fruit trees’, which do not permit assessment of the severity of damage caused. Also, the relative lack of study of their environmental as opposed to agricultural impacts means that the potential of some species to cause serious environmental harm may be underestimated in studies such as this, since with little knowledge, it may not be possible to assign a score for their environmental pest status and potential economic impact on the environment. Nevertheless, we consider this prioritized list of potential pest snails and slugs of quarantine importance to the United States to be a good approximation that we hope will be used as a basis for further development and more detailed evaluation of the pest potential of the taxa included. ACKNOWLEDGMENTS We thank Art Bogan, Carl Christensen, Jay Cordeiro, Mike Cortie, James Coupland, Chuck Lydeard, Cameron Nickerson, Tim Pearce, Emilio Power, John Slapcinsky, I9avid Woodruff, and the other people who offered comments on the species to include in the list. George Davis provided information on Pomatiopsidae and Ken Brown made valuable editorial suggestions. Eunding was provided by USDA-APHIS-PPQ to the American Malacological Society (AMS) in the form of a grant to the University of Hawaii. LITERATURE CITED Aditya, G. and S. K. Raut. 2001. Food of the snail Powacca bridgesi, introduced in India. Current Science 80: 919-921 . Alonso, A. and P. Castro-Dicz. 2008. What explains the invading success of the aquatic mud snail Potamopyrgus mitipodarum (Hydrohiidae, Mollusca)? Hydrobiologia 614: 107-1 16. Andrews, K. L. 1989. Slug pests of dry beans in Central America. British Crop Protection Council Monogrnph 41: 85-89. Armitage, I I. M. 1949. Bureau of Entomology. Thirtieth Annual Re- port. Colifornia Deportment ofAgrienItnre Bulletin 38: 157-216. Baker, G. Tl. 1989. Damage, population dynamics, movement and control of pest helicid snails in southern Australia. British Crop Protection Council Monograph 41: 175-185. Baker, G. II. 1991. Prmluction of eggs and young snails by adult Theba pisana (Muller) and Cernnella virgata (da Gosta) (Mol- lusca: llelicidae) in laboratory cullures and held populations. Australian journal of Zoology 39: 673-679. Baker, G. II. 2002. llelicidae and I lygromiidae as pests in cereal crops aiul pastures in southern Australia. In: G. M. Barker, ed.. Molluscs as Crop Pests, G.ABI Publishing, Wallingford, U.K. Pp. 193-215. GASTROPODS OF QUARANTINE IMPORTANCE IN U.S.A. 125 Bcikcr, G. H. 2008. The popuEition dyricUTiics of the Mediterrtinecin snails Cernuella virgata, Cochlicella acuta (Hygromiidae) and Theba pisana (Helicidae) in pasture-cereal rotations in South Australia: A 20-year study. Australian Journal of Experimental Agriculture 48: 1514-1522. Baker, G. H. and B. G. Hawke. 1991. Fecundity of Cochlicella acuta (Muller) (Mollusca: Helicidae) in laboratory cultures. Inverte- brate Reproduction and Development 20: 243-247. Baker, G. H., B. G. Hawke, and B. K. Vogelzang. 1991. Life history and population dynamics of Cochlicella acuta (Muller) (Gastropoda: Helicidae) in a pasture-cereal rotation. Journal of Molluscan Studies 57: 259-266. Barker, G. M. 1999. Naturalised Terrestrial Stylommatophora (Mol- lusca: Gastropoda). Fauna of New Zealand 38. Manaaki When- ua Press, Lincoln, Canterbury, New Zealand. Barker, G. M. 2001. Gastropods on land: Phylogeny, diversity and adaptive morphology. In: G. M. Barker, ed.. The Biology of Terrestrial Molluscs. CABI Publishing, Wallingford, U.K. Pp. 1-146. Barker, G. M. 2002a. Molluscs as Crop Pests. CABI Publishing, Wall- ingford, U.K. Barker, G. M. 2002b. Gastropods as pests in New Zealand pastoral agriculture, with emphasis on Agriolimacidae, Arionidae and Milacidae. In: G. M. Barker, ed.. Molluscs as Crop Pests. CABI Publishing, Wallingford, U.K. Pp. 361-423. Barker, G. M. and M. G. Efford. 2004. Predatory gastropods as natu- ral enemies of terrestrial gastropods and other invertebrates. In: G. M. Barker, ed.. Natural Enemies of Terrestrial Molluscs. CABI Publishing, Wallingford, U.K. Pp. 279-403. Barnes, M. A., R. K. Fordham, R. L. Burks, and J. ]. Hand. 2008. Fe- cundity of the exotic applesnail, Pomacea insularum. Journal of the North American Benthological Society 27: 738-745. Barrientos, Z. 1998. Life history of the terrestrial snail Ovachlamys fulgens (Stylommatophora: Helicarionidae) under laboratory conditions. Revista de Biologia Tropical 46: 285-296. Batts, L H. 1957. Anatomy and life cycle of the snail Rumina decollata (Pulmonata: Achatinidae). The Southwestern Naturalist 2: 74-82. Baur, B. and J. Bengtsson. 1987. Colonizing ability in land snails on Baltic uplift archipelagos. Journal of Biogeography 14: 329-341. Baur, B. and C. Raboud. 1988. Life history of the land snail Arianta arbustorurn along an altitudinal gradient. Journal of Animal Ecology 57: 71-87. Ben- Ami, F. and L Heller. 2001. Biological control of aquatic pest snails by the black carp Mylopharyngodon piceus. Biological Control 22: 131-138. Ben-Ami, F. and L Heller. 2007. Temporal patterns of geographic parthenogenesis in a freshwater snail. Biological Journal of the Linnean Society 91: 711-718. Bequaert, ]. C. 1950. Studies in the Achatininae, a group of African snails. Bidletin of the Museum of Comparative Zoology at Har- vard College 105: 1-216, pis. 1-81. Berry, A. ]. and A. bin H. Kadri. 1974. Reproduction in the Malayan freshwater cerithiacean gastropod Melanoides tuberculata. Jour- nal of Zoology, London 172: 369-381. Bersine, K., V. E. F. Brenneis, R. C. Draheim, A. M. W. Rub, 1. E. Za- mon, R. K. Litton, S. A. Hinton, M. D. Sytsma, ). R. Cordell, and J. W. Chapman. 2008. Distribution of the invasive New Zealand mudsnail [Potarnopyrgus antipodarum) in the Columbia River Estuary and its first recorded occurrence in the diet of juvenile Chinook salmon (Oncorhynchus tshawytscha). Biological Inva- sions 10: 1381-1388. Boaventura, M. R, S. C. Thiengo, and M. A. Eernandez. 2007. Gastropodes llmnicos hospedeiros intermediarios de trem- atodeos digeneticos no Brasil. In: S. B. dos Santos, A. D. Pimenta, S. C. Thiengo, M. A. Eernandez, and R. S. Absalao, eds., Topicos em Malacologia. Ecos do XVIII Encontro Bra- sileiro de Malacologia, Rio de Janeiro, 21-25 de julho de 2003. Sociedade Brasileira de Malacologia, Rio de laneiro. Pp. 327- 337 [in Portuguese]. Bomford, M., F. Kraus, S. C. Barry, and E. Lawrence. 2008. Predicting establishment success for alien reptiles and amphibians: A role for climate matching,. Biological Invasions DOI 10. 1007/s 10530- 008-9285-3. Bright, C. 1998. Life Out of Bounds. W. W. Norton and Company, New York and London. Britton, D. K. and R. F. McMahon. 2005. Analysis of trailered boat traffic and the potential westward spread of zebra mussels across the lOO''^ meridian. American Malacological Bulletin 20: 147-159. Brussard, P. F. 1975. Geographic variation in North American colo- nies of Cepaea nemoralis. Evolution 29: 402-410. Burch, J. B. and ]. L. Tottenham. 1980. North American freshwater snails. Species lists, ranges and illustrations. Walkerana 1: 81-215. Cheng, E. Y. 1989. Control strategy for the introduced snail, Pomacea lineata, in rice paddy. British Crop Protection Council Mono- graph 41: 69-75. Chi, L. W. and E. D. Wagner. 1962. Oviposition observations in On- comelania formosana. Transactions of the American Microscopi- cal Society 81: 244-246. Civeyrel, L. and D. Simberloff. 1996. A tale of two snails: Is the cure worse than the disease? Biodiversity and Conservation 5: 1231- 1252. Coelho da Silva C. L. P. A., M. S. Soares, and M. G. M. Barreto. 1997. Occurrence of Biomphalaria tenagophila and disappearance of Biomphalaria straminea in Paracambi, R], Brazil. Memorias do Instituto Oswaldo Cruz 92: 37-38. Coote, T. and E. Loeve. 2003. From 61 species to five: Endemic tree snails of the Society Islands fall prey to an ill-judged biological control programme. Oryx 37: 91-96. Coupland, L B. 1996. The biological control of helicid snail pests in Australia: Surveys, screening and potential agents. British Crop Protection Coimcil Symposium Proceedings 66: 255-261. Cox, G. W. 1999. Alien Species in North America and Hawaii. Island Press, Washington, D. C. Cowie, R. H. 1984. The life-cycle and productivity of the land snail Theba pisana (Mollusca: Helicidae). /or/mn/ of Animal Ecology 53:311-325. Cowie, R. H. 1990. Climatic selection on body colour in the land snail Theba pisana (Pulmonata: Helicidae). Hcrcd/f)' 65: 123-126. Cowie, R. H. 1997. Catalog and bibliography of the nonindigenous nonmarine snails and slugs of the Hawaiian Islands. Bishop Museum Occasional Papers 50: 1-66. 126 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Cowie, R. H. 1998a. Patterns of introduction of non-indigenous non-marine snails and slugs in the Hawaiian Islands. Biodiver- sity and Conservation 7: 349-368. Cowie, R. H. 1998b. Catalog of the nonmarine snails and slugs of the Samoan Islands. Bishop Museum Bulletin in Zoology 3: i-viii, 1-122. Cowie, R. H. 2001a. Can snails ever be effective and safe biocontrol agents? International Journal of Pest Management 47: 23-40. Cowie, R. H. 2001b. Invertebrate invasions on Pacific islands and the replacement of unique native faunas: A synthesis of the land and freshwater snails. Biological Invasions 3: 1 19-136. Cowie, R. H. 2002a. List of potential pest mollusks in the USA. In- terim Report. Unpublished report submitted to the USDA- APHIS-PPQ. University of Hawaii, Honolulu. Cowie, R. H. 2002b. Apple snails (Ampullariidae) as agricultural pests: Their biology, impacts and management. In: G. M. Barker, ed.. Molluscs as Crop Pests. CABI Publishing, Wallingford, U.K. Pp. 145-192. Cowie, R. H. and D. G. Robinson. 2003. Pathways of introduction of nonindigenous land and freshwater snails and slugs. In: G. Ruiz and 1. T. Carlton, eds.. Invasive species: Vectors and Management Strategies. Island Press, Washington, D.C. Pp. 93-122. Cowie, R. H. and S. C. Thiengo. 2003. The apple snails of the Ameri- cas (Mollusca: Gastropoda: Ampullariidae: Asolene, Felipponea, Marisa, Pomacea, Pornella): A nomenclatural and type catalog. Malacologia 45: 41-100. Cowie, R. H., K. A. Hayes, C. T. Tran, and W. M. Meyer, III. 2008. The horticultural industry as a vector of alien snails and slugs: Widespread invasions in Hawaii. Internatiotial Journal of Pest Management 54: 267-276. Daehler, C. C., 1. S. Denslow, S. Ansari, and H.-C. Kuo. 2004. A risk-as- sessment system for screening out invasive pest plants from Ha- waii and other Pacific islands. Conservation Biology 18: 360-368. Davis, G. M., T. Wilke, Y. Zhang, X.-L Xu, C.-P. Qiu, C. Spolsky, D.-C. Qiu, Y Li, M.-Y Xia, and Z. Feng. 1999. SnaW-Schistosoma, Paragonimus interactions in China: Population ecology, genetic diversity, coevolution and emerging diseases. Malacologia 41: 355-377. Dazo, B., N. Hairston, and 1. Dawood. 1966. The ecology of Bidintis truncatus and Biomphalaria alexandrina and its implication for the control of bilharziasis in the Egypt 49 project area. Bulletin of the World Health Organization 35: 339-356. De Francesco, C. G. and H. Lagiglia. 2007. A predatory land snail invades central-western Argentina. Biological Invasions 9: 795- 798. de lager, K. and M. Daneel. 2002. Urocyclus Jlavcscens Kerferstein ( Urocyclidae) as a pest of banana in South Africa. In: G. M. Barker, ed.. Molluscs as Crop Pests. CABI Publishing, Walling- ford, U.K. Pp. 235-239. 1 )ehnen-Schmutz, K., ). Ibuza, C. Perrings, and M. Williamson. 2007. A century of the ornamental plant trade and its impact on invasion success. Diversity and Distributions 13: 527-534. 1 )esbiK|uois, L. Chevalier, and L. Madec. 2000. Variability of egg cannibalism in the land snail Helix aspersii in relation to the number of eggs available and the pre.sence of .soil, lounud of Molluscan Studies 66: 273-28 1 . Didham, R. K., J. M. Tylianakis, N. I. Gemmell, T. A. Rand, and R. M. Ewers. 2007. Interactive effects of habitat modification and species invasion on native species decline. Trends in Ecology and Evolution 22: 489-496. Dillon, R. T. 2000. The Ecology of Ereshwater Molluscs. Cambridge University Press, Cambridge, U.K. Dudgeon, D. 1986. The life cycle, population dynamics and produc- tivity of Melanoides tuberculata (Muller, 1774) (Gastropoda: Prosobranchia: Thiaridae) in Hong Kong. Journal of Zoology, London (A) 208: 37-53. Duncan, C. j. 1975. Reproduction. In: V. Eretter and j. Peake, eds., Pulmonates, Vol. 1. Eunctional Anatomy and Physiology. Aca- demic Press, London. Pp. 309-365. Duncan, R. R, M. Bomford, D. M. Eorsyth, and L. Conibear. 2001. High predictability in introduction outcomes and the geo- graphical range size of introduced Australian birds: A role tor climate. Journal of Animal Ecology 70: 621-632. Dundee, D. S. 1 974. Catalog of introduced molluscs of eastern North America (north of Mexico). Sterkiana 55: 1-37. Dundee, D. S. 1986. Notes on the habits and anatomy of the intro- duced land snails, Rumina and Lamellaxis (Subulinidae). The Nautilus 100: 32-37. Dundee, D. S. and A. Paine. 1977. Ecology of the snail Melanoides tuberculata (Muller), intermediate host of the human liver fluke (Opisthorchis sinensis) in New Orleans, Louisiana. The Nautilus 9l:\7-20. Dybdahl, M. E. and Kane S. L. 2005. Adaptation vs. phenotypic plas- ticity in the success of a clonal invader. Ecology 86: 1592-1601. Egonmwan, R. 1. 2007. Light and electron microscopy study of late embryonic development in the land snail Limicolaria jlammca (Miiller) (Pulmonata, Achatinidae). Revista Brasileira de Zoo- logia 24: 436-441 . Eields, A. and D. G. Robinson. 2004. The slug Veronicella sloanci (Cuvier, 1817) - an important pest in the Caribbean. In: I. H. Leal, E. Grimm, and C. Yorgey, eds.. Program and Abstracts. 70''' Annual Meeting, American Malacological Society, Sanibel Ishmd, Plorida, 30 Inly - 4 August 2004, Bailey-Matthews Shell Muse- um, Sanibel, Florida. P. 26. Fisher, T. W. and Orth, R. E. 1985. Biological control of snails. Occasional Papers, Department of Entoniology, University of California, Riverside 1: i-viii, 1-111. Foltz, D. W., H. Ochmann, and R. K. Selander. 1984. Genetic di- versity and breeding systems in terrestrial slugs of the families Limacidae and Arionidae. Malacologia 25: 593-605. Foltz, 1). W., 11. Ochmann, |. S. lones, S. M. Evangelisti, and R. K. Selander. 1982. Genetic population structure and breed- ing systems in arionid slugs (Mollusca: Pulmonata). Biological Jounud of the l.innean Society 17: 225-241. Forsyth, R. (i., |. M. CL 1 lutchinson, and 1 1. Rei.se. 2001. Acgopinclla nitidula (Draparnaud, 1805) (Ciastropotia: Zonitidae) in Brit- ish Columbia - first confirmed North American record. Ameri- can Midacologic(d Bulletin 16: 65-69. Frank, T. 1996. Sown wildllower strips in arable land in rela- tion to slug density aiul slug damage in rape and wheat. British Crop Protection Council Symposium Proceedings 66: 289-296. GASTROPODS OF QUARANTINE IMPORTANCE IN U.S.A. 127 Gederaas, L„ I. Salvesen, and A. Viken. 2007. Norsk svarteliste 2007- 0kologiske risikoviirdeririger avfremrnede arter. 2007 Norwegian Black List — Ecological Risk Analysis of Alien Species. Artsdata- banken, Trondheim. Godan, D. 1983. Pest Slugs and Snails. Springer- Verlag, Berlin, Heidelberg, New York. Goodwin, B. A. I. McAllister, and L. Fahrig. 1999. Predicting inva- siveness of plant species based on biological information. Con- servation Biology 13: 422-426. Gordon, D. R., D. A. Onderdonk, A. M. Fox, and R. K. Stocker. 2008. Consistent accuracy of the Australian weed risk assess- ment system across varied geographies. Diversity and Distribu- tions 14: 234-242. Grewal, P. S., S. K. Grewal, L. Tan, and B. ]. Adams. 2003. Parasitism of molluscs by nematodes: Types of associations and evolution- ary trends. Journal of Nernatology 35: 246-156. Grimm, B. 2001. Life cycle and population density of the pest slug Arion lusitanicus Mabille (Mollusca: Pulmonata) on grassland. Malacologia 43: 25-32. Hall, R. O, Jr., M. F. Dybdahl, and M. C. VanderLoop. 2006. Ex- tremely high secondary production of introduced snails in riv- ers. Ecological Applications 16: 1121-1131. Hata, T. Y., A. H. Hara, and B. K.-S. Hu. 1997. Molluscicides and me- chanical barriers against slugs, Vaginula plebeia Fischer and Ve- ronicella cubensis (Pfeiffer) (Stylommatophora: Veronicellidae). Crop Protection 16: 501-506. Hayes, K. A., R. C. Joshi, S. C. Thiengo, and R. H. Cowie. 2008. Out of South America: Multiple origins of non-native apple snails in Asia. Diversity and Distributions 14: 701-712. Hayes, K. R. and S. C. Barry. 2008. Are there any consistent predictors of invasion success? Biological Invasions 10: 483-506. Heller, ]. 2001. Life history strategies. In: G. M. Barker, ed.. The Biol- ogy of Terrestrial Molluscs. CABI Publishing, Wallingford, U.K. Pp. 413-445. Henderson, I. F. 1989. Slugs and Snails in World Agriculture. British Crop Protection Council, Thornton Heath, U.K. Henderson, 1. F. 1996. Slug and Snail Pests in Agriculture. British Crop Protection Council, Farnham, U.K. Hodasi, L K. M. 1979. Life-history studies of Achatina [Achatina) achatina (Linne). Journal of Molluscan Studies 45: 328-339. Hollingsworth, R. G. and R. H. Cowie. 2006. Apple snails as disease vectors. In: R. C. loshi and L. C. Sebastian, eds.. Global Advanc- es in Ecology and Management of Golden Apple Snails. Philip- pine Rice Research Institute, Munoz, Nueva Ecija, Philippines. Pp. 121-132. Hollingsworth, R. G., R. Kaneta, J. ]. Sullivan, H. S. Bishop, Y. Qvarnstrom, A. I. da Silva, and D. G. Robinson. 2007. Distribu- tion of Parmariott cf. rnartensi (Pulmonata: Helicarionidae), a new semi-slug pest on Hawai'i Island, and its potential as a vec- tor for human angiostrongyliasis. Pacific Science 61: 457-467. Holomuzi, J. R. and B. ). F. Biggs. 1999. Distributional responses to flow disturbance by a stream-dwelling snail. Oikos 87: 36-47. Howells, R. G. 2002. Comparative feeding of two species of apple snails (Poinacea). Ellipsaria 4: 14-16. Jarne, P, M. Vianey-Liaud, and B. Delay. 1993. Selfing and outcross- ing in hermaphrodite freshwater gastropods (Basommato- phora): Where, when and why. Biological Journal of the Linnean Society 49: 99-125. lordaens, K., Dillen, L., and Backeljau, T. 2007. Effects of mating, breeding system and parasites on reproduction in hermaph- rodites: Pulmonate gastropods (Mollusca). Animal Biology 57: 137-195. loshi, R. C. and L. C. Sebastian. 2006. Global Advances in Ecology and Management of Golden Apple Snails. Philippine Rice Research Institute, Munoz, Nueva Ecija, Philippines. Keller, R. P, j. M. Drake, and D. M. Lodge. 2007. Pecundity as a ba- sis for risk assessment of nonindigenous freshwater molluscs. Gonservation Biology 21: 191-200. Kerans, B. L., M. F. Dybdahl, M. M. Gangloff, and ). E. jannot. 2005. Potamopyrgus antipodarum: Distribution, density, and effects on native macroinvertebrate assemblages in the Greater Yellow- stone Ecosystem Journal of the North American Benthological Society 24: 123-138. Kerney, M. P. and R. A. D. Cameron. 1979. Pield Guide to the Land Snails of Britain and North-west Europe. Collins, London. Kiss, L., C. Labaune, F. Magnin, and S. Aubry. 2005. Plasticity of the life cycle of Xeropicta derbentina (Krynicki, 1836), a recently introduced snail in Mediterranean France. Journal of Molluscan Studies 71: 221-231. Kolar, C. S. and D. M. Lodge. 2001. Progress in invasion biology: Predicting invaders. Trends in Ecology and Evolution 16: 199- 204. Kolar, C. S. and D. M. Lodge. 2002. Ecological predictions and risk assessment for alien fishes in North America. Science 298: 1233- 1236. Kumar, S. and S. I. Ahmed. 2000. New records of pestiferous land molluscs from Rajasthan, India. Records of the Zoological Survey of India 98: 67-70. Lazaridou-Dimitriadou, M. and M. E. Kattoulas. 1991. Energy flux in a natural population of the land snail Eobania vermicula- ta (Muller) (Gastropoda: Pulmonata: Stylommatophora) in Greece. Ganadian Journal of Zoology 69: 881-891. Lazaridou, M. and M. Chatziioannou. 2005. Differences in the life histories of Xerolenta obvia (Menke, 1828) (Hygromiidae) in a coastal and a mountainous area of northern Greece. Journal of Molluscati Studies 71: 247-252. Leung, B., ). M. Drake, and D. M. Lodge. 2004. Predicting invasions: Propagule pressure and the gravity of Allee effects. Ecology 85: 1651-1660. Levin, P, R. H. Cowie, J. Taylor, K. Burnett, K. A. Hayes, and C. Perguson. 2006. Apple snail invasions and the slow road to control: Ecological, economic, agricultural and cultural per- spectives in Hawaii. In: R. C. loshi and L. C. Sebastian, eds.. Global Advances in Ecology and Management of Golden Apple Snails. Philippine Rice Research Institute, Munoz, Nueva Ecija, Philippines. Pp. 325-335. Liang, Y.-S. 1974. Cultivation of Bulinus globosus and Biomphalaria pfeifjeri, snail hosts of schistosomiasis. Sterkiana 54: 1-75. Liang, Y.-S. and H. van der Schalie. 1975. Cultivating Lithoglyphop- sis aperta, a new snail host for Schistosoma japonicum, Mekong strain. Joiirnal of Parasitology 61: 915-919. 128 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Lindo, J. R, C. Waugh, J. Hall, C. Cunningham-Myrie, D. Ashley, M. L. Eberhard, J. J. Sullivan, H. S. Bishop, D. G. Robinson, T. Holtz, and R. D. Robinson. 2002. Enzootic Angiostrongylus cantonensis in rats and snails after an outbreak of human eosinophilic meningitis, Jamaica. Emerging Infectious Diseases 8: 324-326. Lockwood, J. L. 1999. Using taxonomy to predict success among in- troduced avifauna: Relative importance of transport and estab- lishment. Conservation Biology 13: 560-567. Loreau, M. and B. Baluku. 1987. Population dynamics of the fresh- water snail Bioniphalaria pfeifferi in eastern Zaire. Journal of Molliiscan Studies 53: 249-265. Lydeard, C., R. H. Cowie, W. F. Ponder, A. E. Bogan, P. Bouchet, S. Clark, K. S. Cummings, T. J. Frest, O. Gargominy, D. G. Herbert, R. Hershler, K. Perez, B. Roth, M. Seddon, E. E. Strong, and F. G. Thompson. 2004. The global decline of nonmarine mollusks. BioScience 54: 321-330. Mack, R. N., D. Simberloff, W. M. Lonsdale, H. Evans, M. Clout, and F. A. Bazzaz. 2000. Biotic invasions: Causes, epidemiology, global consequences, and control. Ecological Applications 10: 689-710. Madsen, H. and F. Frandsen. 1989. The spread of freshwater snails including those of medical and veterinary importance. Acta Tropica 46: 139-146. Majoros, G., Z. Feher, T. Deli, and G. Foldvari. 2008. Establishment of Biomphalaria tenagophila snails in Europe. Emerging Infec- tious Diseases 14: 1812-1813. Marchetti, M. R, P. B. Moyle, and R. Levine. 2004. Invasive species profiling? Exploring the characteristics of non-native fishes across invasion stages in California. Ereshwater Biology 49: 646-661. Mead, A. R. 1979. Economic Malacology with Particular Reference to Achatina fulica. Pulmonates Vol. 2B. Academic Press, London. Mitchell, A.J., M. S. Hobbs, and T. M. Brandt. 2007. The effect of chemical treatments on red-rim Melania Melanoides tubercu- lata, an exotic aquatic snail that serves as a vector of trematodes to fish and other species in the USA. North American Journal of Eisheries Management 27: 1287-1293. Mito, T. and T. Uesugi. 2004. Invasive alien species in Japan: The sta- tus quo and the new regulation for prevention of their adverse effects. Global Environmental Research 8: 171-191. Naranjo-Garcia, E., J. W. Thome, and J. Castillejo. 2007. A review of the Veronicellidae from Mexico (Gastropoda: Soleolifera). Revista Mexicana de Biodiversidad 78: 41-50. NatureServe. 2008. NatureServe Explorer: An online encyclopedia of life [web application]. Version 7.0. NatureServe, Arlington, Virginia. Available at: http://www.nature.serve.org/explorer 22 August 2008. Naylor, R. 1996. Invasions in agriculture: A.s.sessing the cost of the golden apple snail in Asia. Ambio 25: 443-448. O’Keeffe, |. H. 1985. Population biology of the freshwater snail Bu- liniis globosiis on the Kenya Coast. I. Population fluctuations in relation to climate. Jourtud of Applied Ecology 22: 73-84. Palm, M. E. 2001. Systematics and the impact of invasive fungi on agriculture in the United States. BioScicnce 5\: 141-147. Paraen.se, W. L. 1959. One-sided reproductive i.solation between geographically remote populations of a planorbid snail. Ameri- can Naturalist 93:93-101. Parashar, B. D., A. Kumar, and K. M. Rao. 1986. Role of food in mass cultivation of the freshwater snail Indoplanorbis exustus, vector of animal schistosomiasis. Journal oj Molluscan Studies 52: 120-124. Paulay, G. and C. Meyer. 2002. Diversification in the tropical Pacific: Comparisons between marine and terrestrial systems and the importance of founder speciation. Jntegrative and Comparative Biology 42: 922-934. Philp, E. G. 1987. Tandonia rustica (Millet), a slug new to the British Isles. Journal of Conchology 32: 302. Pilsbry, H. A. 1948. Land Mollusca of North America (north of Mex- ico). Vol. II part 2. Academy of Natural Sciences of Philadelphia Monograph 3: i-xlvii, 521-1113. Pimentel, D., L. Lach, R. Zuniga, and D. Morrison. 2000. Environ- mental and economic costs of nonindigenous species in the United States. BioScience 50: 53-65. Pimentel, D., S. McNair, J. Janecka, J. Wightman, C. Simmonds, C. O’Connell, E. Wong, L. Russel, J. Zern, T. Aquino, and T. Tsomondo. 2001. Economic and environmental threats of al- ien plant, animal, and microbe invasions. Agriculture, Ecosys- tems and Environment 84: 1-20. Pimentel, D., R. Zuniga, and D. Morrison. 2005. Update on the en- vironmental and economic costs associated with alien-invasive species in the United States. Ecological Economics 52: 273-288. Plummer, J. M. 1975. Observations on the reproduction, growth and longevity of a laboratory colony of Archachatina [Calachatina) marginata (Swainson) subspecies ovum. Proceedings of the Mal- acological Society of London 41: 395-412. Pointier, J. P. 1999. Invading freshwater gastropods: Some conflicting aspects for public health. Malacologia 41: 403-41 1. Pointier, J. R, P. David, and P. Jarne. 2005. Biological invasions: The case of planorbid snails. Journal of Helminthology 79: 249-256. Pointier, J. R, B. Delay, J. L. Toffart, M. Lefevre, and R. Romero- Alvarez. 1992. Life history traits of three morphs of Melanoides tuberculata (Gastropoda: Thiaridae), an invading snail in the French West Indies. Journal of Molluscan Studies 58: 415-423. Pointier, J. R, R. N. Incani, C. Balzan, R. Chrosciechowski, and S. Prypehan. 1994. Invasion of the rivers of the littoral central re- gion of Venezuela by Thiara granifera and Melanoides tubcrcu- lata (Mollusca: Prosobranchia: Thiaridae) and the absence of Biomphalaria glabrata, snail host of Schistosoma tfumsoni. The Nautilus 107: 124-128. Poucher, C. 1975. Eradication of the giant African snail in Florida. Proceedings of the Elorida State Horticultural Society 88: 523- 524. Radea, C., I. Louvrou, and A. Economou-Amilli. 2008. First record of the New Zealand mud snail Potamopyrgus antipodarum 1. E. Gray 1843 (Molhi.sca: I lydrobiidae) in Greece - notes on its population structure and a.ssociated microalgae. A(}uatic Inva- sions 3: 34 1 -344. Rant, S. K. 1996. Factors determining the effectiveness of the mites Euscuropoda marginata in the control of the slug pests Lacvicau- lis altc. British Crop Protection Council Symposium Proceedings 66: 247-254. Raut, S. K. and G. M. Barker. 2002. Achatina fulica Bowdich and other Achatinidae as pests in tropical agriculture. In: G. M. GASTROPODS OF QUARANTINE IMPORTANCE IN U.S.A. 129 Barker, ed., Molluscs as Crop Pests. CABI Publishing, Walling- ford, U.K. Pp. 55-114. Rant, S.K., M. S. Rahman, and S. K. Samanta. 1992. Influence of temperature on survival, growth and fecundity of the freshwa- ter snail Indoplanorbis exustus. Memorias do Instituto Oswaldo Cniz%l\ 15-19. Rawlings, T. A., K. A. Hayes, R. H. Cowie, and T. M. Collins. 2007. The identity, distribution, and impacts of non-native apple snails in the continental United States. BMC Evolutionary Biol- ogy?: 97 [14 pp.] Remais, ]., A. Hubbard, W. Zisong, and R.C. Spear. 2007. Weather- driven dynamics of an intermediate host: mechanistic and sta- tistical population modelling of Oncomelania hupensis. Journal of Applied Ecology 44: 781-79 1 . Robinson, D. G. 1999. Alien invasions: The effects of the global economy on non-marine gastropod introductions into the United States. Malacologia 41: 413-438. Robinson, D. G. and A. Fields. 2004. The Cuban land snail Zachry- sia: The emerging awareness of an important snail pest in the Caribbean basin, hr. J. H. Leal, E. Grimm, and C. Yorgey, eds.. Program and Abstracts of the 70''' Annual Meeting, American Malacological Society, Sanibel Island, Elorida, 30 July - 4 August 2004. Bailey-Matthews Shell Museum, Sanibel, Florida. P. 73. Robinson, D. G. and J. Slapcinsky. 2005. Recent introductions of alien land snails into North America. American Malacological Bulletin 20: 89-93. Roth, B. and P. S. Sadeghian, 2003. Ghecklist of the land snails and slugs of California. Santa Barbara Museum of Natural History Contributions in Science 3: 1-81. Rueda, A., R. Caballero, R. Kamnsky, and K. L. Andrews. 2002. Vagi- nulidae in Central America, with emphasis on the bean slug Sarasinula plebeia (Fischer). In: G. M. Barker, ed.. Molluscs as Crop Pests. CABI Publishing, Wallingford, U.K. Pp. 1 15-144. Ruesink, J. L., I. M. Parker, M. J. Groom, and P. M. Kareiva. 1995. Reducing the risks of nonindigenous species introductions. Guilty until proven innocent. BioScience 45: 465-477. Saha, T. C. 1993. Effect of crowding on growth, fecundity and life cycle of two freshwater snails, Lymnaea [Radix) luteola and In- doplanorbis exustus. Journal of Ereshwater Biology 5: 49-58. Sakai, A. K., E. W. Allendorf, J. S. Holt, D. M. Lodge, J. Molofsky, K. A. With, S. Baughman, R. J. Cabin, J. E. Cohen, N. C. Ellstrand, D. E. Mccauley, P. O’Neil, I. M. Parker, J. N. Thompson, and S. G. Weller. 2001. The population biology of invasive species. Annual Review of Ecology and Systematics 32: 305-332. Sakovich, N. J. 2002. Integrated management of Cantareus aspersus (Muller) (Helicidae) as a pest of citrus in California. In: G. M. Barker, ed.. Molluscs as Crop Pests. CABI Publishing, Walling- ford, U.K. Pp. 353-360. Sanderson, G. and W. Sirgel. 2002. Helicidae as pests in Australian and South African grapevines. In: G. M. Barker, ed.. Molluscs as Crop Pests. CABI Publishing, Wallingford, U.K. Pp. 255-270. Selander, R. K. and R. O. Hudson. 1976. Animal population struc- ture under close inbreeding: the land snail Rumina in southern Prance. American Naturalist 1 10: 695-718. Shoaib, M. and L. Cagah. 2004. Natural enemies of slugs and snails re- corded in Slovakia. Acfn Eytotecluuca et Zooiechnica 7: 275-278. Simberloff, D. 1986. Introduced insects: A biogeographic and sys- tematic perspective. In: H. A. Mooney and J. A. Drake, eds.. Ecology of Biological Invasions of North America and Hawaii. Springer- Verlag, New York, Berlin, Heidelberg, London, Paris, Tokyo. Pp. 3-26. Smith, J. W. 2005. Recently recognized risks of importing the giant African snail, Achatina fulica Bowdich, 1822, and its relatives into the United States and the efforts of the U.S. Department of Agriculture to mitigate the risk. American Malacological Bul- letin 20: 133-141. South, A. 1992. Terrestrial Slugs. Biology, Ecology and Control. Chap- man & Hall, London. Staikou, A. and M. Lazaridou-Dimitriadou. 1991. The life cycle, population dynamics, growth and secondary production of the snail Xeropicta arenosa Ziegler (Gastropoda: Pulmonata) in northern Greece. Zoological Journal of the Linnean Society 101: 179-188. Staikou, A., M. Lazaridou-Dimitriadou, and N. Farmakis. 1988. As- pects of the life cycle, population dynamics, growth and sec- ondary production of the edible snail Helix lucorum Linnaeus, 1758 (Gastropoda, Pulmonata) in Greece. Journal of Molluscan Studies 54: 139-155. Stange, L. A. 2006. Pest alert. Snails and Slugs of Regulatory Sig- nificance to Elorida. Florida Department of Agriculture and Consumer Services, Division of Plant Industry. Available at: http://www.doacs.state.fl.us/pi/enpp/ento/snail_slugs-pa.html. Accessed 9 September 2008. Staples, G. W. and R. H. Gowie. 2001. Hawaii’s Invasive Species. Mutual Publishing, Bishop Museum Press, Honolulu. Stevens, M. M. 2002. Planorbidae and Lymnaeidae as pests of rice, with particular reference to Isidorella newcombi (Adams & Angus). In: G. M. Barker, ed.. Molluscs as Crop Pests. CABI Publishing, Wallingford, U.K. Pp. 217-233. Stohlgren, T. J. and J. L. Schnase. 2006. Risk analysis for biological hazards: What we need to know about invasive species. Risk Analysis 26: 163-173. Sturrock, R. F. 1973. Field studies on the population dynamics of Biornphalaria glabrata, intermediate host of Schistosoma man- soni on the West Indian island of St. Lucia. International Jour- nal of Parasitology 3: 165-174. Theoharides, K. A. and J. S. Dukes. 2007. Plant invasion across space and time: Factors affecting nonindigenous species success during four stages of invasion. New Phytologist 176: 256-273. Thiengo, S. C., F. A. Faraco, N. C. Salgado, R. H. Cowie, and M. A. Fernandez. 2007. Rapid spread of an invasive snail in South America: The giant African snail, Achatina fulica, in Brasil. Bio- logical Invasions 9: 693-702. Thompson, F. G. 1957. A collection of mollusks from northern \^en- ezuela. Occasional Papers of the Museum of Zoology University of Michigan 591: 1-10, pis. I-II. Tran, G. T., K. A. Hayes, and R. H. Gowie. 2008. Lack of mitochon- drial DNA diversity in invasive apple snails (Ampullariidae) in Hawaii. Malacologia 50: 351-357. Turner, R. L. and C. M. McCabe. 1990. Calcium source for proto- conch formation in the Florida apple snail, Pomacea paludosa 130 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 -2009 (Prosobranchia: Pilidae): More evidence for physiologic plas- ticity in the evolution of terrestrial eggs. The Veliger 33: 185- 189. Txurruka, J. M., O. Altzua, and M. M. Ortega. 1996. Organic matter partitioning in Arion ater: Allometric growth of somatic and reproductive tissues throughout its lifespan. British Crop Pro- tection Council Symposium Proceedings 66: 141-148. Ueshima, R., M. Okamoto, and Y. Saito. 2004. Eobania vermiculata, a land snail newly introduced into Japan. Chiribotan 35: 71-74. USDA-APHIS-PPQ. 2006. Pest alert. Stop the spread of the Cuban slug! USDA Program Aid 1834: 1-2. Available at www.aphis. usda.gov/publications/plant_health/content/printable_version/ pa_cubanslug.pdf. Accessed 8 September 2008. Vagvolgyi, J. 1975. Body size, aerial dispersal and origin of the Pa- cific land snail fauna. Systematic Zoology 24: 465-488. Vaught, K. C. 1989. A Classification of the Living Mollusca. American Malacologists, Inc., Melbourne, Florida. Veltman, C. J., S. Nee, and M. J. Crawley. 1996. Correlates of intro- duction success in exotic New Zealand birds. American Natu- ralist 147: 542-557. Wade, C. M., C. Hudelot, A. Davison, F. Naggs, and P. B. Mordan. 2007. Molecular phylogeny of the helicoid land snails (Pulmo- nata: Stylommatophora: Helicoidea), with special emphasis on the Camaenidae. Journal of Molluscan Studies 73:411-415. Whitson, M. 2005. Cepaea nemoralis (Gastropoda, Helicidae): The invited invader. Journal of the Kentucky Academy of Science 66: 82-88. Wilke, T., G. M. Davis, A. Falniowski, F. Giusti, M. Bodon, and M. Szarowska. 2001. Molecular systematics of Hydrobiidae (Mol- lusca: Gastropoda: Rissooidea): Testing monophyly and phy- logenetic relationships. Proceedings of the Academy of Natural Sciences of Philadelphia 151: 1-21. Winterbourn, M. ). 1970. Population studies on the New Zealand freshwater gastropod, Potamopyrgus antipodarum (Gray). Pro- ceedings of the Malacological Society of London 39: 139-149. Yapi, Y., K. E. N’Goran, D. Saha, P. Cunin, and C. Bellec. 1994. Pop- ulation dynamics of Indoplanorbis exustus (Deshayes, 1834) (Gastropoda: Planorbidae), an exotic freshwater snail recently discovered at Yamoussoukro (Ivory Goast). Journal of Mollus- can Studies 60: 83-87. Submitted: 18 January 2008; accepted: 24 October 2008; final revisions received: J7 March 2009 GASTROPODS OF QUARANTINE IMPORTANCE IN U.S.A. 131 Appendix 1. Scores of each of the 46 taxa evaluated against the 12 attributes related to potential invasiveness (see text for explanation). Egg/ Repro- Intro- Inva- Present Native Phylogenetic Adult juvenile ductive Semelparous/ Breeding duction sion Major Multi- Economic Taxon in USA“ range relationships size size potential iteroparous system pressure history pest pest damage Land snails and slugs Achatinidae Achatina achatina N 0 1 1 0 0 0 0 1 0 0.5 0 0.5 Achatina fulica R - 1 1 0 0.5 0 0 1 1 1 1 1 Archachat'uia N 0 1 1 0 0 0 0 1 0 0 0 0 rnarginata Limicolaria aurora N 0 0 1 0.5 0 0 0.5 0.5 0.5 0 0 Arionidae Arion ater N 1 0 0.5 1 0.5 0.5 0 0 0.5 0 Arion iusitanicus N 0 1 0 0.5 - 1 0 0 1 1 0 1 Ariophantidae Macrochlamys indica N 1 0.5 0 0.5 0.5 0 Ma riaella d ussu in ieri N 0 1 0.5 - - - - 0 0 0.5 0 0 Parmariori rnartensi R 0 1 0.5 - - - - 0 1 0.5 1 0 Bradybaenidae Acusta touranensis N 0 0.5 0.5 0.5 0 0 0 0 Cochlicellidae Cochlicella N 1 1 1 1 0 1 0 1 1 1 0 1 Chronidae Ovachlamys fidgens R 0 1 0 1 0 1 1 1 1 0.5 0 0.5 Enidae Enidae N 0.5 1 0.5 1 0.5 0 0 0 0 Helicidae Arianta arbustorum N 0.5 0 0.5 0.5 0 0 0 0.5 0 0.5 0 0 Cantareus apertus R 0 1 1 - - 0 0 1 1 0.5 0 0 Eobania vermiculata R 0.5 1 1 0.5 - 0.5 0 1 0.5 0 0 0 Elelix R 0 0.5 1 0.5 - 0 0 1 0 0 0 0 Otala punctata R 0 1 1 0.5 - 0 0 0.5 0 0 0 0 Theba pisana R 0.5 1 0.5 1 1 1 0 1 1 1 0 1 Hygromiidae Cernuella R 1 1 0.5 1 1 1 0 1 1 1 0 1 Xerolenta obvia R 1 - 0.5 1 0 1 0 0.5 1 0.5 0 - Xeropicta N 1 - 0.5 1 - 1 0 1 1 1 0 - Milacidae Tandonia R 0 1 0.5 _ _ 1 0 0.5 1 1 0 0.5 budapestensis Tandonia rustica N 0 1 0.5 0.5 . _ 0 0.5 0 0 0 Tandonia sowerbii N 0 1 0.5 0.5 - 1 0 - 1 0.5 0 0.5 Pleurodontidae Thelidornus aspera N 0 1 0 _ _ _ - 0.5 0 0 1 0 Zachrysia auricoma N 0 1 0 - - - - 0.5 0.5 0.5 0 0 Zachrysia provisoria R 0 1 0 - - - - 0.5 1 1 0 1 Zachrysia trinitaria R 0 1 0 - - - 0 0 0 0 0 Spiraxidae Euglandina N 0 1 0 1 - - - 0.5 0 0 0 0 Succineidae Succinea N 0 1 0.5 1 0 0 (Calcisuccinea) Succinea tenella R 0 1 1 1 - - - 1 1 0.5 0 - 132 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Appendix 1. (continued) Egg/ Repro- Intro- Inva- Present Native Phylogenetic Adult juvenile ductive Semelparous/ Breeding duction sion Major Multi- Economic Taxon in USA" range relationships size size potential iteroparous system pressure history pest pest damage Urocyclidae Elisolimax flavesceiis Veronicellidae N 0 1 0.5 - - - 0.5 0 1 0 1 Diplosolenodes occidentalis N - 1 0 - - - 0.5 - 0 1 0 Laevicaidis alte R 1 1 0 - - 0.5 0.5 1 0.5 1 - Leidyida moreleti N 0 1 0 - - - 0.5 0 0 0 0 Samsinida plebeia R - 1 0 0 - 0.5 1 1 1 1 1 Veronicella ciibensis R 0 1 0 - - - 1 1 1 1 0.5 Veronicella sloonii Freshwater snails N 0 1 0 - - - 0.5 - 1 0 0.5 Ampullariidae Marisa R 0 0.5 1 1 0 0 1 0.5 1 0 Pila R 1 0.5 1 - 0 0 0.5 0.5 0.5 1 0 Pomacea ' Planorbidae R 1 1 1 0.5 1 0 0 1 1 1 1 1 Biomphalaria ^ N 1 1 1 1 0.5 0 0.5 0 1 1 0 - Bidiniis N 1 1 1 1 1 0 0.5 0 0 1 0 - Itidoplanorbis exustus Pomatiopsidae R 0 1 0.5 1 1 0 0.5 0.5 1 1 1 - Oncomelania N 0 1 1 1 0 0 0 0 1 0 - '' Not present (N) or locally restricted (R). '' Only species of Eiiglamiina not native to the United States. ‘ Only species of Sucdnea (Calcisiicdnea) not native to th' e United States. May also include the similar Sucdnea horticola. ' All species of Pomacea except P. dijfiisa (often referred to, incorrectly, as P. bridgesii) and the native P. paludosa. ^ All species of Biomphalaria except the native B. ohstructa. Appendix 2. Scores of each of the seven species already present in the United States that were used to validate the model fo r assessing invasive potential. Egg/ Repro- Native Phylogenetic Adult juvenile ductive Semelparous/ Breeding Introduction Invasion Major Multi- Economic Taxon range relationships size size potential iteroparous system pressure history pest pest damage Agriolimacidae Deroccras 1 0 1 0.5 1 0 1 1 1 0 1 reticidatwn Helicidae Cepaca ncmoralis 1 0.5 0.5 0.5 0 0 0 0.5 0.5 0 0 0 Cornu aspcrsum 1 1 0.5 0.5 0 0 0 1 1 1 0 1 Hydrobiidae Potamopyrgiis antipodarum Milacidae 0 1 1 1 0 1 1 0.5 1 1 0 0 Milax goyntcs Suhulinidac 0.5 1 0 1 1 0 0.5 1 1 0 1 Riimina decollnin Thiaridac 0.5 0 0.5 1 0.5 1 0.5 0 1 0 0 0 Melnnoidcs - 1 0.5 1 0 0 1 0.5 1 0.5 1 0 tubercidala Amer. Malac. Bull. 27: 133-140 (2009) New small deep-sea species of Gastropoda from the Campos Basin off Brazil Ricardo Silva Absalao Departamento de Zoologia, Institute de Biologia, Universidade do Estado do Rio de Janeiro, Avenida Sao Francisco Xavier 524, Maracana, Rio de Janeiro, RJ, Brazil CEP 20550-900 and Departamento de Zoologia, Institute de Biologia, Centro de Ciencias da Saude, Universidade Federal do Rio de Janeiro, Ilha do Fundao, Rio de Janeiro, RJ, Brazil CEP 21941-570 Corresponding autlior: absalao@liotmail.com Abstract: During environmental characterization of the Campos Basin, Rio de Janeiro State (22°S), about 120 samples from depths between 400 and 2000 m were dredged with a 0.25-m^ box-core, and a high biodiversity of micro-molluscs was found. Although only dead animals were collected, the shells were often in a good state of preservation. Six new species, belonging to gastropods families Aclididae {Adis kanela n. sp.), Trochidae {Mirachelus urueuauau n. sp. and Calliotropis pataxo n. sp.), Skeneidae {Palazzia pankakare n. sp. and Adeuomphalus xerente n. sp.), and Tornidae {Ponderinella xacriaba n. sp.) are described. Key words: biodiversity, deep-water, continental slope, new species This paper consists of observations on several groups of southern Atlantic gastropods, accumulated during the past 10 years. It is based on material from the continental slope of Brazil obtained during environmental characterization of the Campos Basin, the main Brazilian oil-production area, sponsored by the Brazilian petroleum company Petrobras. Finds of very small and well-preserved shells are still rare worldwide, and there are no published records from the Brazilian coast. This is the most extensive collection of deep- water molluscs from South America. Of the 1575 species of marine molluscs reported for Brazil by Rios (1994), 1112 belong to the Gastropoda. Only 126 (1 1.33%) of these are reported to occur on the continental slope (below 200 m depth); many also occur on the continental shelf. These numbers are not indicative of an impoverished malacological fauna but rather reflect quite limited collecting effort at such depths. The first gastropod species from Brazilian deep water was reported by Watson (1879), who described Margarites dnopherus (Watson, 1879). Many of the 126 above-mentioned deep-water species were also described or reported by him. After the great 19'*’ century exploratory expeditions (the “Challenger,” “Albatross,” and others), no additional reports on Brazilian deep-water molluscs appeared. This situation began to change after 1980, when Leal and co-workers described some new macro-mollusc species (Leal and Bouchet 1989, 1991, Leal and Rios 1990, Leal and Simone 1998, 2000). Other reports appeared by Harasewych (1983) and more recently by Absalao et al. (2001), Absalao and Pimenta (2003, 2005), Absalao and Santos (2004), Zelaya etal. (2006), Simone (1999, 2002, 2003, 2005), Simone and Birman (2006), Simone and Cunha (2006), Barros et al. (2007), Lima and Barros (2007), and Lima et al. (2007). Despite the efforts of these investigators, our knowledge about molluscs and especially about deep-water micro-molluscs is incomplete. Our limited knowledge about biodiversity of these deep waters is even worse when we consider our lack of knowledge about the anatomy of the soft parts and the radulae of most deep-water species, making it difficult to determine their generic and/or sub-generic placement. Therefore, despite taxonomic difficulties, the goal of this study was to describe several newly discovered deep-water micro-gastropod species. MATERIALS AND METHODS About 120 samples were dredged with a 0.25-m^ box- core from the continental slope off Rio de Janeiro state, at depths ranging from 700 to 1950 m. Dredging was performed by the R/V Astro-Garoupa between 2001 and 2003 as part of the program “Environmental Characterization of Campos Basin, RJ, Brazil” under the auspices of PETROBRAS S.A. Each sample was washed through a 0.5-mm mesh and preserved in 70% ethanol. In the laboratory, these residual materials were sorted under magnification and the molluscs picked out. Although no live specimens were collected, the shells were often in a good state of preservation. Shells were mounted on specimen stubs and photographed under a Scanning Electron Microscope (ZEISS EVO 40) at “Gerencia de Bioestatigrafia e Paleoecologia Aplicada (BPA)” belonging to the Centro de Pesquisas da Petrobras (CENPES). Most of the material is deposited in the mollusc collection of Departamento de Zoologia, Instituto de Biologia da Universidade Federal do Rio de Janeiro (JBUFRJ) unless otherwise stated. Other abbreviations and terms used in this paper are: MNRJ ( Museu Nacional do Rio de Janeiro), MNHN 133 134 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 (Museum national d’Histoire naturelle Paris), BC (Bacia de Campos = Campos Basin), PETROBRAS (Brazilian Petroleum Company), n. sp. (new species), Norte (north), and Sul (south). RESULTS Family Trochidae Rafinesque, 1915 Genus M/rac/;e/us Woodring, 1928 Mirachelus unieuauaii new species (Figs. lA-B) Description Shell small, conical, and solid, of about 4 whorls. Protoconch relatively large and smooth, with about IVi whorls. Teleoconch with 1 2 well-separated, thin axial cordlets. There is a spiral cord on the lower ‘A of the whorl, forming nodules where it crosses the axial cordlets. No microsculpture. Shell profile pagoda-like, with a deep suture. Base almost straight, with a smooth cord bordering the umbilicus. Aperture subcircular-polyhedric because the outer lip is Figures I. A-B, Mirachelus uruciiauau n. sp. I lolotypc IBUI RI 18031. A, whole shell, frontal view; B, protoeoneh. (LI), Calliolropis pataxo n. sp. I lolotype IBUFR) 18035. (i, whole shell, frontal view; I), protoconch. L-l I, Palazzia pankakarc n. sp. I lolotype IBUl'RI 18041. L, dorsal view; F-, ventral view; G, frontal view; 1 1, protoconch. Scale bar for A-(i: 500 pin; D: 250 pm; lL(i: 400 pm; 1 1: 200 pm. NEW DEEP-SEA GASTROPODA FROM BRAZIL 135 marked to both the spiral cord and basal cord. Columella short, slightly arched. Radula and operculum unknown. Etymology This species is named in honor of the Uru-eu-au-au Indians, one of the indigenous peoples of Brazil. The name is employed as a noun in apposition. Holotype IBUFRJ 18031; 2.65 mm length, 2.26 mm width, Campos Basin, BC Sul I #75 (22°31'28"S, 40°03'50"W), 19.XI.2002, 1050 m. Paratypes MNRJ 12847, BC Sul II #80 (22°24'30"S, 39°57'28"W), 20.VI.2003, 1044 m; IBUFRJ 18032, BC Sul II #86 (22°3 1 '37"S, 39°55'14"W), 16.VI.2003, 1630 m; MNHN, BC Sul II #81 (22°26'28"S, 39°54'08"W), 21.VI.2003, 1345 m; IBUFRJ 18033 BC Norte I #61 (21°52'51"S, 39°48' 11" W), 12.XII.2002, 1350 m. Additional material IBUFRJ 18034BCNorteII#61(21°52'51"S,39°48'12"W), 26.VI.2003, 1372 m. Remarks The only species of Mirachelus previously reported for Brazil is Mirachelus clinocnemus Quinn, 1979 (Quinn 1979: 18-19, figs. 33-34). Rios (1994, fig. 83) illustrated the species with an SEM photograph. Mirachelus urueuauau n. sp. can be distinguished from M. clinocnemus by its more delicate axial riblets, compared to the thicker axial ribs in M. clinocnemus. The spaces between the axial riblets in M. urueuauau n. sp. is about 4 times the riblet width whereas the spaces are almost equal to the width of the ribs in M. clinocnemus. Finally, the base is almost straight and smooth in M. urueuauau n. sp., while in M. clinocnemus it is convex and has three strong spiral cords. The umbilicus is wider in M. urueuauau n. sp. than in M. clinocnemus. Although the holotype of Mirachelus urueuauau n. sp. is only a fourth whorl shell, these differences cannot be attributed to size differences. Mirachelus corbis (Dali, 1889) (Quinn 1979: 18, figs. 35-36) is a very distinctive species, with a more turreted shell profile, stronger and numerous axial ribs, an additional spiral cord on the whorls, an excavated suture, and an ornamented base rib. Despite its small size, Mirachelus urueuauau n. sp. cannot be confused with any other species of the genus. Genus Calliotropis Seguenza, 1903 Calliotropis pataxo new species (Figs. IC-D) Description Shell small, broadly conical, of about 6 whorls, iridescent, carinate, and with broad umbilicus. Shell profile straight or slightly concave, strongly stepped. Protoconch small, smooth, with slightly more than one whorl, ornamented with very small irregular stellar nodules connected by protuberances concentrated on the initial part. Proto-teleoconch boundary well marked. First and second whorls of teleoconch with a narrow keel on the shoulder, then becoming obsolete and arising again on the body whorl as a more nodulose keel. Body whorl with three keels: one nodulose near the suture; the second in the middle of the whorl, less evident than the others; and the third with very light nodules, delimiting the body whorl from the base. Opisthocline axial growth scars present on the entire shell, but more visible on the first two teleoconch whorls. There are about 22 low but somewhat pointed nodules on the spiral keel on the body whorl, suggesting the presence of very retractive axial ribs. Base convex, smooth or with hardly discernible axial cords. Umbilicus broad and deep, bordered by somewhat evanescent cord; two additional umbilical cords may also be present inside. Aperture subquadrate, narrow; inner lip slightly reflected. Columella slightly oblique and arched. Etymology This species is named in honor of the Pataxo Indians, one of the indigenous peoples of Brazil. The name is employed as a noun in apposition. Holotype IBUFRJ 18035, 1.22 mm length, 1.40 mm width, Campos Basin, BC Sul I #73 (22°41'35"S, 40°00'45"W), 22.XI.2002, 1950 m. Paratypes MNRJ 12848, BC Sul II #77 (22°36'12"S, 39°58'22"W), 13.VI.2003, 1670 m; IBUFRJ 18036, BC Sul II #83 (22°30'34"S, 39°51'44"W), 16.VI.2003, 1970 m; MNHN, BC Norte I #46 (22°10'55"S, 39°49'00"W), 10.XII.2002, 1350 m; IBUFRJ 18037,BCNorteI#62(21°52'4T'S,39°46'17"W),ll.XII.2002, 1650 m. Additional material IBUFRJ 18038,BCNorteI#63(21°52'44"S,39°40'45"W), 11.XII.2002, 1950 m; IBUFRJ 18039, BC Norte II #57 (2r57'15"S,39°47'41"W),28.VI.2003, 1587 m. Remarks There are only three species previously reported for Brazil: Calliotropis actinophora (Dali, 1890), Calliotropis aeglees (Watson, 1879), and Calliotropis calatha (Dali, 1927) (Rios 1994). The first two are deep-water species reported from off 136 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 northern and northeastern Brazil; the third is reported from the edge of the continental shelf off southern Brazil. Calliotropis pataxo n. sp. can be distinguished from C. actinophora (type illustrated by Quinn 1979, hgs. 21-22) by its smaller spire angle, less-rounded profile of the body whorl, absence of thin radial riblets, absence ol nodulous middle spiral keel, and absence of two strong spiral cords on the base. It can be distinguished from C. aeglees (type illustrated by Quinn 1979, figs. 11-12) by its stepped shell; the less-nodulose spiral cords, especially the supra-sutural cord; absence of axial riblets on the first whorl; absence of a third keel forming the whorl periphery; and absence of three spiral cords on the base. It can be distinguished from C. calatha (type illustrated by Quinn 1979, figs. 15-16) by the less-expanded body whorl, absence of axial riblets, sutures not being channeled, less-prominent middle keel, and absence of four spiral cords on the base. Calliotropis rhina (Watson, 1886) (type illustrated by Quinn 1979, figs. 27-28) shows a much higher spire than Calliotropis pataxo n. sp., in addition to more nodulous spiral cords and spiral cords on the base. Calliotropis lissocona (Dali, 1881) (type illustrated by Quinn, 1979, figs. 13-14) has a smaller spire angle than C. putnxo n. sp., shows no trace of a middle keel, has two very strong spiral cords on the base, and a nodulose spiral cord bordering the umbilicus; none of these cords is present in C pataxo n. sp. In summary, the stepped shell, the slight nodulose ornamentation, and smooth base preclude any chance of confusing Calliotropis pataxo n. sp. with other species of the Atlantic basin. Lamily Skeneidae Thiele, 1929 Genus Palazzia Waren, 1991 Palazzia pankakare new species (figs. lE-H) Description Shell small, diameter about 1 .5 mm, white, solid, planispiral; spire not projecting beyond the last whorl profile. Protoconch with about 1 .5 whorls, with clear cut between proto-teleoconch boundary. It is entirely covered by irregular corrugations. Teleoconch with 15 annular ribs (triangular in section) and broad smooth interspaces between them, these spaces about 3-4 times the width of the annular ribs. No spiral sculpture nor pits present. Lips strong but not thickened. Aperture holostomate, circular. Operculum unknown. Etymology This species is named in honor of the Pankakare Indians, one of the indigenous peoples of Brazil. 4'he name is employed as a noun in apposition. Holotype I BULK) 1 804 1, shell diameter 1.40 mm, height 0.68 mm, Gampos Basin, BG Sul I #77 (22°36'03"S, 39°57'54"W), 16.X1.2002, 1650 m. Paratypes MNRl 12849, BC Sul I #85 (22°29'33"S, 39°56'17"W), 19.XI.2002, 1350 m; IBULRJ 18042, BC Sul II #85 (22°30'21"S, 39°56'53"W), 21.VI.2003, 1353 m; MNHN, BC Sul II #86 (22°31'37"S, 39°55'14"W), 16.VI.2003, 1630 m; IBULRJ 18043, BC Norte I #61 (21°52'51"S, 39°48' 1 1" W), 12.XII.2002, 1350 m. Additional material IBULRJ 18044, BC Norte II #61 (21°52'51"S, 39°48'12"W), 26.VI.2003, 1372 m. Remarks There is no record of this genus in the South Atlantic, and Palazzia phmorbis (Dali, 1927) and Palazzia ausonia (Palazzi, 1988) are the only known species reported for the North Atlantic (Waren 1991a: 77, figs. 16A-D; 78-79, figs. 17A-G, 18A-C). Palazzia pankakare n. sp. can be distinguished from the other congeners by its incomplete and much more numerous axial rings. The micro-ornamentation of the protoconch of P. pankakare n. sp. is identical those showed by P planorhis and P. ausonia (see Waren 1991a: 79, figs. 18A- C). The genus Ammonicera Vayssiere, 1893 has conchological similarities with Palazzia pankakare n. sp. but is distinguished from Palazzia by its spirally ornamented protoconch (Rolan 1992). Despite the good fit of P. pankakare n. sp. to genus Palazzia, Waren (1991a: 75) states that “the teleoconch surface is finely and irregularly pitted” and we were unable to find any sign of pits on the teleoconch. So, attribution to the generic level should be viewed with caution. Genus Adeuoniphalus Seguenza, 1876 Adeuomphalus xerente new species (Pigs. 2A-D) Description Shell planospiral, small, white, with axial rings and spiral cords. Protoconch size about 1.75 whorls, with well-marked boundary with teleoconch. Protoconch with irregular stellar nodules connected by protuberances that spread over ■% of the protoconch length. Teleoconch with about 60 narrow, equally spaced rings. Width of interspaces about 2-4 times the width of ring. Peripherally there is a spiral cord on both the dorsal and ventral side of the shell, and sometimes it is crossed by rings forming low and rounded nodules. Lips not thickened, aperture irregularly ovoid with anterior whorl projecting slightly over it. Operculum and radula unkin)wn. Etymology This species is named in honor ol the Xerente Indians, one ol the indigenous peoples t)l Brazil. The name is employed as a noun in apposition. NEW DEEP-SEA GASTROPODA PROM BRAZIL 137 Figures 2. A-D, Adeuomphalus xerente n. sp. Holotype IBUFRJ 18045. A, dorsal view; B, ventral view; C, frontal view; D, protoconch. E-H, Poderinella xacriaba n. sp. Holotype IBUFRJ 18048. E, whole shell, frontal view; F, dorsal view; G, base; H, protoconch. I-K, Adis kanela n. sp. Holotype IBUFRJ 18056. 1, whole shell, frontal view; 1, protoconch and K, detail of suture. Scale bar for A-C, I: 250 pm; D, H, J-K: 100 pm; E, G: 500 pm Holotype IBUFRJ 18045, shell diameter 0.74 mm, height 0.55 mm, Campos Basin, BC Sul II #85 (22°30'21"S, 39°56'53"W), 21.VI.2003, 1353 m. Paratypes MNRJ 12850, BC Sul II #86 (22°31'37"S, 39°55'14"W), 16.VI.2003, 1630 m; IBUFRJ18046, BC Norte I #46 (22°10'55"S, 39°49'00"W), 10.XII.2002, 1350 m; MNHN, BC Norte I #61 (21°52'51"S, 39°48'11"W), 12.XII.2002, 1350 m; IBUFRJ 18047, BC Norte II #45 (22°10'53"S, 39°52'18"W), 01.VII.2003, 1039 m. Remarks There is no record of this genus in the South Atlantic, and Adeuomphalus ammoniformis Seguenza, 1 876 is the only known species reported for the North Atlantic (Waren 1991a: 74-76, hgs. 14F, 15A-B). It can be distinguished from Adeuomphalus xerente n. sp. by its much more numerous axial rings (about 85); A. xerente n. sp. has only about 60. These rings in A. ammoniformis 138 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 also disappear abruptly towards the periphery of the shell, and frequently have a tubercle % from the periphery; in Adeiiomphalus xerente n. sp. the axial rings are constant over all the shell, not disappearing in any part. The tubercles are more discrete than those in A. ammoniformis. The protoconch ornamentation of Adeiiomphalus xerente n. sp. is identical to that described for A. ammoniformis. But despite that, the two species can be clearly distinguished. family Tornidae Sacco, 1896 Genus Ponderinella Marshall, 1988 Ponderinella xacriaba new species (figs. 2E-H) Description Shell ovoid-conical, almost globular, small, smooth, solid, and white. Resembing a small naticid. Protoconch with about 1.5 whorls, ornamented with delicate irregular lines that ramify and anastomose (very small irregular stellar nodules connected by protuberances), and a well-marked boundary with the teleoconch. Teleoconch with convex whorls, smooth except for irregular growth scars. Suture well marked, bordered at body whorl by a slight depression. Base smooth, umbilicus moderately narrow. Lips strong, aperture rounded, prosocline, slightly pointed posteriorly. Operculum and radula unknown. Etymology This species is named in honor of the Xacriaba Indians, one of the indigenous peoples of Brazil. The name is employed as a noun in apposition. Holotype IBULRJ 18048, 1.23 mm length, 1.23 mm width, Campos Basin, BC Sul I, #73 (22°41'35"S, 40°00'45"W), 22.XI.2002, 1950 m. Paratypes MNRI 12851, BC Sul 1 #77 (22°36'03"S, 39°57'54"W), 16.X1.2002, 1650 m; IBULRJ 18049, BC Sul 1 #82 (22°28'49"S, 39°53'24"W), 17.XI.2002, 1650 m; MNHN, BC Sul I #85 (22°29'33"S, 39°56'17"W), 19.X1.2002, 1350 m; IBULRJ 18050, BC Sul II #81 (22°26'28"S, 39°54'08"W), 21. VI. 2003, 1345 m. Additional material IBUFRJ 18051, BC Sul II #85 (22°30'2 l"S, 39°56'53"W), 21. VI. 2003, 1353 m; IBUFRJ 18052, BC Norte I #46 (22°10'55"S, 39°49'00"W), I0.X11.2002, 1350 m; IBUF'RI 18053, BC Norte I #61 (2I°52'5I"S, 39°48'II"W), 12. XI 1.2002, 1350 m; IBUFRJ 18054, BC Norte II #50 (22°04'33"S, 39°52'05"W), 30.VI.2003, 1030 m; IBUFR) 18055, BC Norte II #61 (21°52'51"S,39°48'12"W),26.VI.2003, 1372 m. Remarks At first glance, Ponderinella xacriaba n. sp. resembles Cirsonella in the general shell shape, but the protoconch ornamentation and aperture are very distinctive. While Cirsonella has a protoconch with very fine irregular spiral lines (Waren, 1991b: 210, fig. E), Ponderinella xacriaba n. sp. has tiny irregular stellar nodules connected by protuberances (Lig. 2H). The aperture of Cirsonella is regularly rounded (Waren, 1991b: 212, figs. 1 lA-D), whereas in P xacriaba n. sp. it is posteriorly constricted (fig. 2E). These differences preclude assignment of this new taxon to the genus Cirsonella. The genus Ponderinella, on other hand, was originally created to accommodate a characteristic group from the southeastern Pacific. Recently, Rolan and Rubio (2002) assigned some eastern Atlantic species to this genus, and it seems to be an appropriate systematic placement for Ponderinella xacriaba n. sp. as well. The genus Ponderinella was, until now, restricted to the eastern Atlantic. The record of Ponderinella xacriaba n. sp. expands the genus distribution to the southwestern Atlantic. Among Atlantic Ponderinella, P. xacriaba n. sp. is distinguished from Ponderinella tornatica (Moolenbeek and Hoenselaar, 1995) (Rolan and Rubio 2002: 39, figs. 101-106) which has a strong peripheral keel on the body whorl and another keel bordering the umbilicus, whereas P. xacriaba n. sp. is devoid of such characters. Ponderinella minutissima Rolan and Rubio, 2002 (pp. 42-43, figs. 118-126) has a thickened umbilical cord, which is absent in P. xacriaba n. sp. Ponderinella skeneoides Rolan and Rubio, 2002 (pp. 40-41, figs. 107-117) is the most similar species to P. xacriaba n. sp. Although P. skeneoides has a very variable shell profile, some of them (Rolan and Rubio2002: 41, fig. 115) resemble those present in P. xacriaba n. sp., and in addition both species share the same kind of protoconch ornamentation (Rolan and Rubio 2002: 41, fig. 1 16, and figs. 2E-L herein). Despite these similarities, the two species can be distinguished because P. skeneoides usually has a much broader umbilicus (Rolan and Rubio 2002: 41, figs. 107-1 10), an umbilical cord (sometimes nodulose) (Rolan and Rubio 2002: 41, figs. 108, 113-1 14), and a less-rounded aperture with the ventral part somewhat retracted (same illustrations mentioned above); P. xacriaba n. sp. has a narrower umhilicus, no umbilical cord, and a more-rounded aperture with no retraction on the ventral side (Figs. 2L, 2C). Family Aclididae Sars, 1 878 Genus Ac//s Loven, 1846 Adis kaneta new species (Figs. 2I-K) Description Shell small, white, thin, conical-elongated. Protoconch globose, smooth, with about 1 'A whorls. Proto-teleoconch NEW DEEP-SEA GASTROPODA FROM BRAZIL 139 transition not discernible. Teleoconch with convex profile; in the third and fourth whorls, the main whorl diameter is in the anterior part of the whorl. Whorls increasing moderately in diameter. Smooth, except for irregular axial growing scars. Suture well impressed, with posterior border clearly extending over the anterior one, forming a projecting border (Fig. 2K). Supra-sutural microscopic spiral striation, visible only under very strong magnification, disappears towards the middle of the whorl. Base conical, convex. Aperture ovoid, peristome reflected on anterior side, with no thickening. Umbilicus narrow, partially covered by parietal wall (fig. 21). Etymology This species is named in honor of the Kanela Indians, one of the indigenous peoples of Brazil. The name is employed as a noun in apposition. Holotype IBUFRJ 18056, 1.90 mm length, 0.75 mm width, Campos Basin, BC Sul I #75 (22°31'28"S, 40°03'50"W), 19.XI.2002, 1050 m. Paratypes MNRJ 12852, BC Sul I #80 (22°24'31"S, 39°57'28"W), 20.XI.2002, 1050 m; IBUFRJ 18057, BC Sul II #75 (22°3 1 '28"S, 40°03'49"W), 18.Vi.2003, 1043 m; MNHN, BC Sul II #84 (22°26'28"S, 39°58'53.3"W), 20.VI.2003, 1046 m; IBUFRJ 18058, BC Norte I #45 (22°10'54"S, 39°52'19"W), 10.XII.2002, 1050 m. Additional material IBUFRJ 1 8059, BC Norte I #60 (2 1°52'50"S, 39°5 1 '04" W), 12.XII.2002, 1050 m; IBUFRJ 18060, BC Norte II #63 (21°52'43"S, 39°40'4T'W), 26.VI.2003, 1941 m. Remarks Five species of Adis have been previously reported for Brazil (Rios 1994). Two of them are shallow- water species {Adis bermudensis Dali and Bartsch, 1911 and Adis underwoodae (Bartsch, 1947)) and show distinctive shell profiles. Adis hyalina Watson, 1881 was reallocated to the genus Costadis Bartsch, 1947 by Bouchet and Waren (1986). The other two, Ac/issfln'ssfl Watson, 1881 and Adis macrostoma Barros, Lima, and Francisco, 2007 are deep-water species from northern Brazil and will be discussed below. Adis sarissa shows the typical elongated-turreted Adis shell profile, with the first whorls markedly smaller than the last ones; whereas in Adis kanela n. sp. the increase in shell width is more regular. The apex is also blunter than in the two former species. Finally, A. sarissa shows very fine axial lines, which are lacking in A. kanela n. sp. Adis macrostoma and A. kanela n. sp. share the same type of supra-sutural microscopic spiral striae. Adis macrostoma has a more pointed protoconch than A. kanela n. sp., a much more obtuse spire angle, and an opisthocline aperture with reflected lips posteriorly and anteriorly; A. kanela n. sp. has the aperture almost orthocline and reflected lips restricted on the anterior side. A similar protoconch is present in Adis attenuans Jeffreys, 1883 (Bouchet and Waren 1986: 305, fig. 730), but the whorls are regularly convex, whereas in Adis kanela n. sp. the greatest whorl diameter is on the anterior part of the whorl; moreover, the lip in A. attenuans is more extensively reflected, whereas this reflection is more restricted in A. kanela n. sp. Finally, A. attenuans is a Mediterranean species, while A. kanela n. sp. is a South American one. ACKNOWLEDGMENTS My sincere thanks are extended to Dr. Emilio Rolan for his critique and suggestions about the generic placement of some of taxa herein described. I also thank Petrobras (Brazilian Petroleum Co.) for the possibility to get this material, and the SEM supporting the descriptions. This research was partially supported by CNPq from Brazil. LITERATURE CITED Absalao, R. S. and A. D. Pimenta. 2003. A new subgenus and three new species of Brazilian deep waters Olivella (Mollusca, Gas- tropoda, Olivellidae) collected by the RV Marion Dufresne in \9%7 . Zoosystema 25: 177-185. Absalao, R. S. and A. D. Pimenta. 2005. New records and new species of Vetulonia Dali, 1913 and Brookula Iredale, 1912 from Brazil (Gastropoda, Trochidae). The Veliger 47: 193-201. Absalao, R. S. and R N. Santos. 2004. Recent deep-sea species of Benthonellania Lozouet, 1990 (Gastropoda, Rissoidea) from the south-western Atlantic with description of two new species utilizing a shell morphometric multivariate. Journal of Con- chology 38: 329-340. Absalao, R. S., G. Miyaji, and A. D. Pimenta. 2001. The genus Brookula Iredale, 1912 (Gastropoda: Trochidae) from Brazil; Description of a new species, with notes on other South American species. Zoosystema 23: 675-687. Barros, J. C. N, S. R B. Lima, and J. A. Francisco. 2007. Two new spe- cies of Adis (Mollusca: Gastropoda: Aclididae) from the conti- nental slope of northeast Brazil. Zoofnxfl 1614: 61-68. Bouchet, P. and A. Waren. 1986. Revision of the Northeast Atlantic bathyal and abyssal Aclididae, Rulimidae, Rpitoniidae (Mol- lusca, Gastropoda). Bollettino Malacologico (Supplement 2): 299-576. Harasewych, M. G. 1 983. A review of the Golumbariinae (Gastropoda; Turbinellidae) of the western Atlantic with notes on the anat- omy and systematic relationships of the subfamily. Ncnwuria 27: 1-42. 140 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Leal, J. H. and P. Bouchet. 1989. New deep-water Volutidae from off southeastern Brazil (Mollusca; Gastropoda). The Nautilus 103: 1-12. Leal, J. H. and P. Bouchet. 1991. Distribution patterns and disper- sal of prosobranch gastropods along a seamount chain in the Atlantic Ocean. Journal of the Marine Biological Association of the United Kingdom 71: 11-25. Leal, 1. H. and E. C. Rios. 1990. Nanomelon vossi, a new deep-water Zidoninae from off southern Brazil (Gastropoda: Volutidae). The Veliger 33: 317-320. Leal, ). H. and L. R. L. Simone. 1998. Propilidium curumim, a new species of Lepetidae (Gastropoda, Patellogastropoda) from off southern and southeastern Brazil. Bulletin of Marine Science 63: 1 57-165. Leal, J. H. and L. R. L. Simone. 2000. Copulabyssia riosi, a new deep- sea limpet (Gastropoda: Pseudococculinidae) from the conti- nental slope off Brazil with comments on the systematics of the genus. The Nautilus 114: 59-68. Lima, S. E B. and J. C. N. Barros. 2007. Two new species of Cerithiella ( Apogastropoda: Cerithiopsidae) for the continental slope of Pernambuco (northeast Brazil). Zoofaxu 1441: 63-68. Lima, S. F. B, J. C. N. Barros, and R. E. Petit. 2007. A new species of Gerdiella (Gastropoda; Cancellariidae) from the South Atlantic Ocean off Brazil with discussion of an undescribed species. The Nautilus 121: 99-103. Quinn, ]. F. 1979. Biological results of the University of Miami deep- sea expeditions. 130. The systematic and zoogeography of the gastropod family Trochidae collected in the Straits of Florida and its approaches. Malacologia 19: 1-62. Rios, E. C. 1994. Seashells of Brazil. Funda(;:ao Universidade do Rio Grande, Rio Grande. Rolan, E. 1992. The family Omalogyridae G. O. Sars, 1878 (Mollusca, Gastropoda) in Cuba with description of eight new species. Apex 7: 35-46. Rolan, E. and F. Rubio. 2002. The family Tornidae (Gastropoda, Rissoidea) in the East Atlantic. Resenhas Malacologicas 13 (Suplemento): 1-98. Simone, L. R. L. 1999. The anatomy of Cochlespira Conrad (Gastropoda, Conoidea, Turridae) with a description of a new species from the Southeastern coast of Brazil. Revista Brasileira dc Zoologia 16: 103-115. Simone, L. R. L. 2002. Thee new deepwater species of Eulimidae (Caenogastropoda) from Brazil. Novapex 3: 55-60. Simone, L. FL L. 2003. Revision of the genus Bcnthobia (Caenogastropo- da, Pseudolividae). Journal of Molluscan Studies 69: 245-262. Simone, L. R. L. 2005. A new species of Gemmula (Caenogastropoda Turridae) from Brazilian deep waters. Stromlms 12: 7-10. Simone, L. R. L. and A. Birman. 2006. A new species of Iphinopsis (Caenogastropoda, Cancellariidae) from Brazil. Journal of (Jonchology 39: 141-144. Simone, L. R. I,, and C. M. Cimha. 2006. Revision of genera Gaza and Gallogaza (Vetigastropoda, Trochidae), with description of a new Brazilian spcc'ws. Zoolaxa 1318: 1-40. Watson, R. B. 1879. Mollu.sca of II.M.S. ‘Challenger’ Expedition. Part IV. Zoological Journal of the Litnican Society 14: 692-716. Waren, A. 1991a. New and little known Mollusca from Iceland and Scandinavia. Sarsia 76: 53- 1 24. Waren, A. 1991b. New and little known “skeneimorph” gastropods from the Mediterranean Sea and the adjacent Atlantic Ocean. Bollettino Malacologico 27: 149-248. Zelaya, D. G., R. S. Absalao, and A. D. Pimenta. 2006. A revision of Benthobrookula Clarke, 1961 (Gastropoda, Trochoidea) in the Southwestern Atlantic Ocean. Journal of Molluscan Studies 72: 77-87. Submitted: 18 June 2008; accepted: 4 December 2008; final revisions received: 8 May 2009 Amer. Make. Bull. 27: 141-156 (2009) The genera Myonerfl, Octoporia, and Protocuspidaria (Pelecypoda, Cuspidariidae) from deep waters of Campos Basin, Rio de Janeiro, Brazil with descriptions of two new species Cleo Dilnei de Castro Oliveira and Ricardo Silva Absalao Departamento de Zoologia, Instituto de Biologia, Centro de Ciencias da Saude, (Jniversidade Federal do Rio de Janeiro, Illia do Fundao 21941-590 Rio de Janeiro, Rio de Janeiro, Brazil Corresponding author; cleo.oliveira@gmail.com Abstract: Eight species of Cuspidariidae belonging to the genera Myonera Dali, 1886, Octoporia Scarlato and Starobogatov, 1983, and Protocuspidaria Allen and Morgan, 1981 were obtained in samples from the continental slope (700-2000 m depth) at Campos Basin, Rio de Janeiro state (22 S), Brazil. Myonera paucistriata Dali, 1886 and Protocuspidaria verityi Allen and Morgan, 1981 are now documented from the Campos Basin. Though previously reported for Brazil, the known range of Octoporia octaporosa (Allen and Morgan, 1981) is enlarged southward. Myonera limatula (Dali, 1881) and Protocuspidaria atlantica Allen and Morgan, 1981 are reported from the South Atlantic Ocean for the first time. Myonera kaiwa sp. nov. and Protocuspidaria jarauara sp. nov. are described herein. Myonera sp., an eighth taxon, is present as one unique specimen. The presence of micro-pits on the shell surface is reported for the first time in the Septibranchia. Key words: Mollusca, Anomalodesmata, biodiversity, continental slope, micro-pits The septibranchs comprise some of the most intriguing pelecypods, with anatomical adaptations for a carnivorous diet. Reflecting the importance of the group in deep waters, in the last 30 years the number of described species has increased considerably, with many faunal surveys and taxonomic reviews (Kuroda 1952, Bernard 1974, Knudsen 1982, Poutiers 1984, Poutiers and Bernard 1995). In spite of this activity, there are still only a few studies regarding exclusively the septibranchs in deep waters off Brazil (e.g., Marini 1974). The classification of septibranchs has a long taxonomic history (Pelseneer 1911, Thiele 1935, Newell 1969, Allen and Morgan 1981, Morton 1981, Scarlato and Starobogatov 1983, Poutiers and Bernard 1995, Harper et al. 2006). Scarlato and Starobogatov (1983) proposed a new classification based on the branchial apparatus-septum structure that increased the number of higher-rank taxa, many of them new. Poutiers and Bernard (1995) completed a revision of species from the Pacific Ocean, and their classification disagreed with that of Scarlato and Starobogatov (1983). Harper et al. (2000) and Morton (2003: 378, table 3), both based on morphological data, did not recognize the higher-rank taxa classification proposed by Scarlato and Starobogatov (1983). Additionally, Dreyer et al. (2003) and Harper etal. (2006) addressed the same issue, but worked with all anomalodesmatans, and with molecular tools as well, also did not support Scarlato and Starobogatov (1983). In our opinion, there is insufficient justification to adopt the higher-rank taxa proposed by Scarlato and Starobogatov (1983) and we follow the classification of Poutiers and Bernard (1995), regarding Myonera Dali, 1886, Octoporia Scarlato and Starobogatov, 1983, and Protocuspidaria Allen and Morgan, 1981 as full genera of Cuspidariidae. MATERIALS AND METHODS All material was dredged with a 0.25-m^ box-corer, from the continental slope off Rio de Janeiro state, at depths ranging from 700 to 1950 m. The dredging was carried out by the R/V Astro-Garoupa between 2001 and 2003 as part of the program “Environmental Characterization of Campos Basin, RJ, Brazil” under the auspices of PETROBRAS S.A. Each sample was washed through a mesh of 0.05 mm and preserved in 70% ethanol. In the laboratory, the residues were examined under magnification and the molluscs picked out. Although no live specimens were collected, many of the shells were in a good state of preservation. Shells were mounted on specimen stubs and photographed under a Scanning Electron Microscope (ZEISS EVO 40), at the Gerencia de Bioestatigrafia e Paleoecologia Aplicada (BPA), belonging to the Petrobras Research Center (Centro de Pesquisas da Petrobras-CENPES). The diameters of the micro-pits were measured on the SEM photos. Though other authors have given the distribution of species studied herein, only those who expanded the kno\vn geographic range of these species were included in the tables of distribution. The material is deposited in the Mollusca collection of the following institutions: Departamento de Zoologia, Instituto 141 142 AMERICAN MALACOLOGICAL BULLETIN 11 'Ml' 2009 de Biologia, Universidade Lederal do Rio de Janeiro (IBULRJ); Museu de Zoologia, Universidade de Sao Paulo (MZUSP); Museu Nacional, Universidade Lederal do Rio de Janeiro (MNRJ); and Museum national d’Histoire naturelle, Paris (MNHN). RESULTS Cuspidariidae Dali, 1886 Genus Myonera Dali, 1886 Type species: Myonera paucistriata Dali, 1886 by original designation in Dali, 1886b Genus characterization Shell small, outline variable, inequilateral, rostrate, usually inflated. Externally with concentric and/or radial orna- mentation. Hinge edentulous. Ligament internal, deflected and posteriorly pointed. Gills with four or five pairs of pores. (Adapted from Allen and Morgan 1981, Poutiers and Bernard 1995). Discussion The name Myonera was introduced as a subgenus of Neaera by Dali (1886a). Later, Dali (1886b) established Myonera as a genus and included Neaera siilcifera Jeffreys, 1881, Neaera angularis Jeffreys, 1876, Neaera lamellifera Dali, 1881, Neaera lirnatula Dali, 1881 [plus its synonym Neaera contracta Jeffreys, 1881], Myonera laticella Dali, 1886, Neaera undata Verrill, 1884, and Neaera fragilissima Smith, 1885, and designated Myonera paucistriata as the type species. Currently, some of these species are allocated to other genera ie.g., Cuspidaria sulcifera, Cuspidaria contracta, Cuspidaria undata, Cardiornya fragilissima). Some classifications still rank Myonera as a subgenus of Cuspidaria (Thiele 1935, Allen and Morgan 1981 ), as a section of the genus Cuspidaria (Eischer 1887) or, most frequently, as a genus of the family Cuspidariidae (Grasse 1960, Yokes 1967, Bernard 1974, Rios 1994, Poutiers and Bernard 1995,Absalao et al. 2003). Scarlato and Starobogatov (1983) proposed the family Myoneridae, comprising 1 1 genera, a classification not followed by many authors (Rios 1994, Poutiers and Bernard 1995, Harper et al. 2000, 2006, Morton 2003, Dreyer ct al. 2003). We agree with Poutiers and Bernard (1995: 139) that “although the usage of several ordinal taxa [...] may prove to be useful, many of them seem presently unwarranted.” Myonera paucistriata Dali, 1886 (Pigs. lA-D) Neaera paucistriata Bush, 1885: 473 [nomen nudum] Myonera paucistriata Dali, 1886: 302; Abbott 1974: 568; Allen and Morgan 1981: 473; Rios 1994: 303 Cuspidaria paucistriata: Pelseneer, 1911: 80 Characterization Shell white, small (max. length 8 mm), elongated, inequi- lateral, rostrate, inflated. Umbo large, centralized, rostrum short, ventrally pointed, two keels between umbo and postero-ventral margin, anterior margin rounded. Anteriorly with an undulating appearance. External surface smooth. Micro-pits absent. Hinge edentulous. Resilifer posteriorly pointed. Material examined IBUFRJ 17855 (21°58'36"S, 39°46'30"W, 1700 m), 08.X.01, [1 valve], IBUFRJ 17860 (22°07'17"S, 39°50'02"W, 1230 m), 13.V.02, ]2 valves], IBUFRJ 17865 (22°04'52"S, 39°49'04"W, 1330 m), 09.V.02, ]2 valves], IBUFRJ 17866 (22°02'36"S,39°49'36"W, 1330 m), 08.V.02, ] 1 valve], IBUFRJ 17867 (22°03'27"S, 39°45'07"W, 1730 m), 08.V.02, ]2 valves], IBUFRJ 17871 (22°05'45"S, 39°45'55"W, 1730 m), 09.V.02, ]1 valve], IBUFRJ 17880 (22°09'10"S, 39°44'50"W, 1930 m), 08.V.02, ]1 valve], IBUFRJ 17911 (22°04'44"S, 39°46'31"W, 1650 m), 24.XI.02, ]2 valves], IBUFRJ 17912 (21°57'15"S, 39°47'43"W, 1650 m), 14.X1I.02, ]1 valve], IBUFRJ 17924 (21°52'41"S, 39°46'17"W, 1650 m), 11.X11.02, ]1 valve], IBUFRJ 17925 (22°41 ' 10''S, 40°02'20"W, 1650 m), 13.V1.03, ]3 valves], IBUFRJ 17935 (21°52'41"S, 39°46'17"W, 1650 m), 26.VI.03, ]1 valve], IBUFRJ 17936 (2 1°57' 1 5"S, 39°47'4 T'W, 1650 m), 28.V1.03, ]3 valves], IBUFR) 17954 (22°04'45"S, 39°46'31"W, 1650 m), 27. VI. 03, ]4 valves], IBUFR] 17979 (22°11'04"S, 39°47'04"W, 1650 m), 22.V1.03, ]7 valves], IBUFRJ 18025 (21°57'I5"S, 39°49'37"W, 1350 m), 25.V1.03, ]6 valves], IBUFRJ 18026 (22°26'27"S, 39T58'51''W, 1050 m), 20.XI.02, ] I valve], IBUFR) 18027 (21°57'15"S, 39°49'37"W, 1350 m), 14.XII.02, ]2 valves], IBUFRI 18030 (21°52'41"S, 39°46'17"W, 1650 m), 1 I.XI1.02, ] 6 valves]. Distribution of Myonera paucistriata Dali, 1886 References Locality Type locality: U. S. Coast Survey Steamer ‘Blake’ sta. Depth (m) Dali (1886b) 226 and 230 ]St. Vicent]; sta. 43 [near Tortugas, Florida]. 620-849 Abbott (1974) North Carolina to West Indies. 353-1609 Allen and Morgan (1981 ) Pacific Ocean: Hawaiian Islands. Atlantic Ocean: northwest Atlantic, west coast of Malabar, Bay of Biscay. 678-3806 Rios ( 1994) North Carolina to West Indies. Brazil. 600-760 Rosenberg (2005) USA: Florida: Florida Keys. 35“’N to 34.42°S; 80°W to 4.25°W. 166-1609 Present study Campos Basin, Rio de Janeiro, Brazil. 1050-1930 PELECYPODS PROM DEEP WATERS OF CAMPOS BASIN 143 Figure 1. A-D, Myonera paucistriata Dali, 1886. IBUFRJ 17924. A, external view; B, umbo detail; C, anterior margin detail; D, rostrum detail. E-I, Myonera limatula (Dali, 1881). IBUFRJ 14798. E, external view; F, internal view; G, umbo detail; H, rostrum detail; I, anterior margin detail. Scale bar A: 2 mm; E-F: 1 mm; B-D, G-I: 100 (am. 144 AMERICAN MALACOLOGICAL BULLETIN 27 • 1 /2 • 2009 Discussion The name Neaera paucistriata was first cited in Bush (1885), but in a manner that does not comply with articles 12 and 16 of the Code (ICZN 1999), characterizing it as a nomen nudum. A proper description of this species was made by Dali (1886b) as Myonera paucistriata. This is probably the most common of all septibranchs (Allen and Morgan 1981). In Campos Basin, it was present in 18 of 1 17 stations where pelecypods occurred. Regarding micro-pits, this species is different from all other Myonera in this study, because, for Myonera paucistriata, no micro-pits were observed on the shell surface (Ligs. IB-D). Myonera limatula (Dali, 1881) (Ligs. lE-I) Neaera limatula Dali, 1881: 112 Neaera contractu: leffreys, 1881: 941, pi. LXXI, fig. 4 Myonera limatula: Dali, 1886: 304, pi. Ill, fig. 5 Characterization Shell white, small ( max. length 6mm), elongate, inequilateral, rostrate, somewhat inflated. Umbo large, closer to anterior end, rostrum oblique, pointing downwards, postero-ventral sinuation slightly pronounced, anterior margin rounded. Externally with about 20 equidistant concentric lamellae with fine concentric growth lines between them. Micro-pits present on umbo (size: X = 3 pm ± 0.9 SD) and restricted to distal surface of concentric lamellae on the rostrum (size: x = 2 pm ± 0.5 SD). These pits are unevenly distributed over the shell, being concentrated on the posterior part. Hinge edentulous. Resilifer somewhat oblique. (1886b: 304) argued that the posterior margin of the right valve is beveled off and is not a tooth, and we agree with this interpretation. Unlike Myonera paucistriata, micro-pits are present in Myonera limatula. These micro-pits are restricted to the umbo region (fig. IG) and on concentric lamellae, which shows a gradient of distribution of the micro-pits, being very abundant on the rostrum region (fig. IH) and vanishing towards the anterior and ventral margins (fig. II). In Gampos Basin it was present at 1 of 117 stations where pelecypods occurred. Myonera sp. (figs. 2A-E) Description Shell small (max. length 3 mm), fragile, inequilateral. Umbo slender, small, closer to anterior end, dorsal margin straight, rostrum long, postero-ventral sinuation absent, ante- rior margin rounded. Ornamented from anterior to posterior ends with seven equally spaced concentric continuous ridges, which are present from the middle to the ventral margin of the shell. Hardly visible concentric growth striae. Micro-pits (size: X = 3 pm ± 0.5 SD) are abundant at umbo and less numerous at rostrum, but vanish toward antero-ventral margin and are absent from the ridges of shell. Hinge edentulous. Resilifer elongated, posteriorly pointed. Distribution Restricted to Campos Basin, Rio de Janeiro state, Brazil. Distribution of Myonera limatula (Dali, 1881) References Locality Depth (m) Dali (1881) Type locality: U. S. Coast Survey Steamer ‘Blake’ sta. 44 [near Tortugas, Florida]. 985 Abbott (1974) off Nantucket, Massachusetts, Florida Strait 985-1000 Foutiers and Bernard ( 1995) Northwest and West Central Atlantic. 230-1000 Rosenberg (2005) USA: Florida: West Florida. 42‘’N to 25‘’N. 985-1000 Present study Campos Basin, Rio de Janeiro, Brazil. 1700 Material examined IBULRJ 14798 (21°58'36"S, 39°46'30"W, 1700 m), 08.X.2001 [ 1 valve]. Discussion The hinge plate of Myonera limatida has a lamellar process on the postero-dorsal margin of the right valve (Fig. If). Ihis was not mentioned in the original description (Dali 1881: 112-113) but was noted for Neaera contracta (Jeffreys, 1881), a junior synonym of M. limatula. Jeffreys ( 1 88 1 : 94 1 ) noted one laminar tooth on the posterior side of the right valve, extending parallel to the hinge plate. Dali Material examined IBULRJ 14795 (21°58'36"S, 39°46'30"W, 1700 m),08.X.2001 [1 spec.]. Discussion A very rare taxon, only one specimen was found. Although we strongly suspect that it is new to science, a formal epithet will be delayed until additional material is collected. Myonera sp. can be distinguished from all other Atlantic species by a unique set of characters: the straight dorsal margin and the narrow length of the dorso-ventral axis, which is the smallest among the known Atlantic species. The most similar species in the South Atlantic is Myonera allcni Foiitiers, 1995 (figured at Allen and Morgan 1981: 472, fig. 35 as Myonera atlantica). Myonera sp. can be distinguished from M. alleni by, in the former, the ab.sence of postero-ventral sinuation and rostral ridge, antero-dorsal margin straight (Fig. 2A). Olten M. alleni shows incomplete concentric ridges, whereas in Myonera sp. these are always complete from the PELECYPODS PROM DEEP WATERS OP CAMPOS BASIN 145 Figure 2. A-E, Myonera sp. IBUFRJ 14795. A, external view; B, internal view; C, umbo detail; D, rostrum detail; E, anterior margin detail. Scale bar A-B: 1 mm; C-E: 100 pm. anterior to the posterior end, including the rostral region (Pig. 2A). The micro-pits of Myonera limatula show an opposite distribution to those of Myonera sp. In the former species the micro-pits are present on the umbo but are restricted to the distal margins of the concentric lamellae on the remainder of the shell. In Myonera sp. they occur, except on the concentric lamellae, over the umbo region (Pig. 2C) and over the rostrum region (Pig. 2D) vanishing towards the antero-ventral margin (Pig. 2E). Since Myonera sp. is a bit smaller than M. limatula, such differences in pit distribution might be explained by the intrinsic differences among growth stages, but when one compares the same shell regions of both species it is clear that such differences are not related to growth. In addition, the material of M. limatula and Myonera sp. is not worn, so such differences cannot be attributed to preservation stage of the shell either. Myonera kaiwa sp. nov. (Pigs. 3A-I) Description Shell white, small (max. length 5 mm), elongate, inequi- lateral, rostrate, inflated. Umbo large, closer to anterior end, rostrum slender, elongate, gently curved dorsally, postero- ventral margin slightly sinuate, anterior margin well rounded. Ornamented with about five concentric foliaceous lamellae, and countless growth lines between them. Rostral ridge present, with a secondary ridge parallel to the dorsal margin usually visible, with growth scars and about seven faint longitudinal striae formed by the periostracum, more conspic- uous at the rostrum end. Micro-pits absent on lamellae, but present over entire shell. The micro-pits are more abundant and larger (size: x = 10 pm ± 2.8 SD) on the rostrum, decreasing in number and in size (size: x = 5 pm ± 0.6 SD) towards the anterior margin. Hinge edentulous. Resilifer elongate, deflected, posteriorly pointed. 146 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Figure 3. A-l, Myonera kaiwa sp. nov. A, external view, left valve. Holotype IBUl RI 17923; B, external view, right valve. I’aratype IBUFRI 17934; C, internal view, right valve. Raratype IBUFRJ 17933; 1), internal view, left valve. Raratype IBURRl 17932; R, dorsal view. Raratype IBUFRI 17886. F-l, Raratype IBUFR) 14799. F, anterior margin; G, rostrum detail; II, anterior margin detail; I, rostrum detail. Scale bar A-H; I mm; F-G: 100 pm; II-l: 40 pm. PELECYPODS FROM DEEP WATERS OF CAMPOS BASIN 147 Figure 4. A-E, Octoporia octaporosa (Allen and Morgan, 1981). A-B, IBUFRJ 14805. A, external view; B, internal view; C, rostrum detail. IBUFRJ 14802; D, anterior margin detail. IBUFRJ 14800; E, umbo detail. IBUFRJ 14804. F-I, Protocuspidaria verityi Allen and Morgan, 1981. F-G, I, IBUFRJ 14998. F, external view; G, internal view, right valve; H, internal view, left valve. IBUFRJ 17896; I, anterior margin detail. Scale bar A-B: 1 mm; C, E: 25 |im; D, I: 50 |am; F-H: 500 |im. 148 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Figure 5. A-1-, Protocuspidaria atlnntiai Allen and Morgan, 1981. A-B, L-I', IBUI'K] 14997. C-D, IBUl'R) I8()0(i. A, external view; B-C, internal view from left and right valve.s, re.spectively; D-F., hinge detail; F, anterior margin detail. Cl-M, I’lvtocuspiddrid jdiiiudvd .sp. nov. Ilolotype IBUFR) I4996. (I, ), external view from right and left valve.s, respectively; IM, internal view from right and left valves, respec- tively; K-L, hinge detail; M, anterior margin detail. Scale bar CM), (I-l: 500 pm; A-B, F, K-F: 200 pm; I', M: 30 pm. PELECYPODS PROM DEEP WATERS OP CAMPOS BASIN 149 Etymology This species is named in honor of the Kaiwa Indians, one of the indigenous peoples of Brazil. The name is employed as a noun in apposition. Distribution Restricted to Campos Basin, Rio de Janeiro state, Brazil. Holotype IBUPRJ 17923 (21°52'41"S, 39°46'17"W, 1650 m), 26. VI.03, [left valve]. Paratypes IBUFRJ 17934 (21°52'41"S, 39°46'17"W, 1650 m), 26. VI.03, [1 valve], IBUFRJ 17933 (21°52'41"S 39°46'17"W, 1650 m), 26.VI.03, [1 valve], IBUFRJ 17932 (21°52'41"S, 39°46'17"W, 1650 m), 26.VI.03, [1 valve], IBUFRJ 17886 (22°05'11"S, 39°42'40"W, 1930 m), 08.V.02, [1 valve], IBUFRJ 14799 (21°58'36"S, 39°46'30"W, 1700 m), 08.X.2001 [1 valve], MNRJ 12859 (22°36'03"S, 39°57'54"W, 1650 m), 16.XI.02, [2 valves], MZUSP 40595 (22°04'45"S, 39°46'31"W, 1650 m), 27. VI.03, [2 valves], MNHN (22°37'02"S, 39°56'20"W, 1950 m), 13.VI.03, [2 valves], MNHN (22°03'27"S, 39°45'07"W, 1730 m), 08.V.02, [3 valves]. Other material examined IBUFRJ 17879 ' (22°09'10"S, 39°44'50"W, 1930 m), 08.V.02, [15 valves], IBUFRJ 17885 (22°06'52"S, 39°44'13"W, 1930 m), 08.V.02, [6 valves], IBUFRJ 17887 (22°05'11"S, 39°42'40"W, 1930 m), 08.V.02, [7 valves], IBUFRJ 17904 (22°01'16"S, 39°43'44"W, 1950 m), 25.XI.02, [4 valves], IBUFRJ 17909 (22°04'44"S, 39°46'31"W, 1650 m), 24.XI.02, [5 valves], IBUFRJ 17916 (21°57'26"S, 39°40'33"W, 1950 m), 11.XII.02, [6 valves], IBUFRJ 17917 (21°52'44"S, 39°40'45"W, 1950 m), 11.XII.02, [4 valves], IBUFRJ 17940 (21°52'43"S, 39°40'4T'W, 1950 m), 26.VI.03, [1 valve], IBUFRJ 17950 (22°41'35"S, 40°00'45"W, 1950 m), 22.XI.02, [5 valves], IBUFRJ 17988 (21°57'26"S, 39°40'34"W, 1950 m), 27.VI.03, [5 valves], IBUFRJ 18000 (22°33'08"S, 39°54'21"W, 1950 m), 15.VI.03, [3 valves], IBUFRJ 18011 (22°41'31"S, 40°00'47"W, 1950 m), 06.XII.03, [7 valves], IBUFRJ 18014 (22°30'34"S, 9°51'44"W, 1950 m), 16.VI.03, [7 valves]. Discussion The most similar species in the South Atlantic is Myonera alleni. Myonera kaiwa sp. nov. can be distinguished from M. alleni, and also from Myonera sp. (this study) by less-marked sinuation on the postero-ventral margin, a more concave postero-dorsal margin, a longer and more slender rostrum, and the concentric foliaceous lamellae extending to the umbo and complete to the rostral ridge (Figs. 3A-B). Regarding the micro-pits, Myonera kaiwa sp. nov. is the only taxon that shows this set of features: pits restricted to the shell (not occurring on the concentric ridges) (Figs. 3F-G), the largest pits and a size gradient (Figs. 3H-I). In the Campos Basin, it was present at 20 of 1 17 stations where pelecypods occurred. Genus Octoporia Scarlato and Starobogatov, 1983 Type species: Cuspidaria (Myonera) octaporosa Allen and Morgan, 1981 original designation by Scarlato and Starobogatov (1983) Genus characterization Shell small, elongate, inequilateral, rostrum long. Sculptured with growth lines and concentric ribs, these more conspicuous on anterior part. Hinge edentulous. Septum with 8-20 pairs of pores. (Adapted from Allen and Morgan 1981, Scarlato and Starobogatov 1983, Poutiers and Bernard 1995). Discussion The genus Octoporia includes species that resemble the shells of Myonera but shows anatomical similarities with Halonympha Dali, 1886 (Poutiers and Bernard 1995). The name Octoporia was introduced by Scarlato and Starobogatov (1983) as a genus with only one known species, Cuspidaria (Myonera) octaporosa (Allen and Morgan, 1981), in their new family Halonymphidae, which accommodates species with 8-20 pairs of septal pores. Subsequently, Krylova (1994) revised Octoporia, describing new species. Currently, the family Hanonymphidae is not generally recognized [except by Scarlato and Starobogatov (1983) and Krylova (1994)] but the name Octoporia was accepted as a genus by Poutiers and Bernard (1995) and as a subgenus of Halonympha by Harper et al. (2006), in both cases, as a taxonomic category in Cuspidariidae. Octoporia octaporosa (Allen and Morgan, 1981) (Figs. 4A-E) Cuspidaria (Myonera) octaporosa Allen and Morgan, 1981:476-479, figs. 40-41 Octoporia octaporosa: Scarlato and Starobogatov 1983 translated in Poutiers and Bernard 1995: 176; Krylova 1994: 40 Characterization Shell white, small (max. length 5 mm), elongate, inequi- lateral, rostrate. Umbo small, triangular, blunt, centralized, postero-dorsal margin concave, rostrum slender, faintly postero-ventral sinuation, anterior margin rounded. Ornamen- tation varying from almost smooth to covered by live or more slender concentric ribs, usually present from ventral margin to middle of shell. Concentric growth lines may be present over entire shell but are much more conspicuous on rostrum. Rostral ridge slight. Additional irregular radial lines barely 150 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 visible on rostrum. Micro-pits equally distributed over entire shell and lamellae, but smaller (size: x = 0.5 pm ±0.1 SD) on the umbo and on rostrum and larger on the anterior margin (size: x = 1 pm ± 0.6 SD). Hinge edentulous. Resilifer not visible in present material. IBULRJ 17857 (22°03'03"S, 39°50'32"W, 1230 m), 13.V.02, [5 valves], IBULRl 17863 (22°06'58"S, 39°48'41"W, 1330 m), 09.V.02, [5 valves], IBULRJ 17869 (22°03'27"S, 39°45'07"W, 1730 m), 08.V.02, [19 valves], IBULRJ 17870 (22°05'45"S, 39°45'55"W, 1730 m), 09.V.02, [12 valves], IBULRJ 17875 (22°08'23"S, 39°46'23"W, 1730 m), 09.V.02, [12 valves], IBULRJ 17881 (22°09' 10"S, 39°44'50"W, 1930 m), 08.V.02, [37 valves], IBULRJ 17882 (22°06'52"S, 39°44'13"W, 1930 m), 08.V.02, [15 valves], IBULRJ 17889 (22°05'H"S, 39°42'40"W, 1930 m), 08.V.02, [29 valves], IBULRJ 17890 (22°38'01"S, 40°17'26"W, 900 m), 18.V.02, [1 valve], IBULRJ 17895 (22°41'18"S, 40°14'05"W, 1100 m), 15.V.02, [4 valves], IBULRJ 17900 (22°11'04"S, 39°47'04"W, 1650 m), 25.XI.02, [11 valves], IBULRJ 17901 (22°1 1 ' 16"S, 39°43'44"W, 1950 m), 25.XI.02, [ 14 valves], IBULRJ 17910 (22°04'44"S, 39°46'31"W, 1650 m), 24.XI.02, [10 valves], IBULRJ 17907 (22°04'46"S, 39°43'02"W, 1950 m), 24.XI.02, [8 valves], IBULRJ 17915 (21°57'26"S, 39°40'33"W, 1950 m), 11.XII.02, [13 valves], IBULRJ 17922 (21°52'41"S, 39°46'17"W, 1650 m), 11.XII.02, [34 valves], IBULRJ17958 (22°46'59"S, 40°07'49"W, 1650 m), 22.XI.02, [9 valves], IBULRJ 17981 (22°38'53"S, 40°04'14"W, 1350 m), 23.XJ.02, [1 valve], IBULRJ 18003 (22°41'03"S, 40°02'29"W, 1650 m), 23.XI.02, [23 valves], IBULRJ 17947 (22°41'35"S, 40°00'45"W, 1950 m), 22.XI.02, [75 valves], IBULRJ 17985 (22°3 1 '28"S, 40°03'50"W, 1050 m), 19.XI.02, [ 1 6 val ves ] , 1 BU L R J 1 7966 ( 22°36 ' 03"S, 39°57 ' 54"W, 1650m), 16.XI.02, [7 valves], IBULRJ 17964 (22°37'02"S, 39°56'20"W, 1950 m), 23.XI.02, [22 valves], IBULRJ 18021 (22°24'31"S, 39°57'28"W, 1050 m), 20.X1.02, [6 valves], IBULRJ 17995 (22°27'18"S, 39°54'50"W, 1350 m), 17.XI.02, [2 valves], IBULRJ 17930 (22°30'35"S, 39°51'45"W, 1950 m), 23.XI.02, [32 valves], IBULRJ 14800 (2 1°58'36"S, 39°46'30"W, 1700 m), 08.X.2001 [1 valve], IBULRJ 14801 (2 1°58'36"S, 39°46'30"W, 1700 m), 08.X.2001 ]4 valves], IBULRJ 14802 (21°58'36"S, 39°46'30"W, 1700 m), 08.X.2001 [4 valves], IBULRJ 14803 (21°58'36"S, 39°46'30"W, 1700 m), 08.X.2001 ]ll valves]. IBULRJ 14804 (21°58'36"S, 39°46'30"W, 1700 m), 08.X.2001 [5 valves], IBULRJ 14805 (21°58'36"S, 39°46'30"W, 1700 m), 08.X.2001 [4 valves], IBULRJ 14806 (21°57'05"S, 39°49'58"W, 1200 m), 24.IX.2001 [1 valve], IBULRJ 18022 (22°33'10"S, 39°54'22"W, 1950 m), 23.XI.02, [72 valves], IBULRJ 17959 (22°10'53"S, 39°52'18"W, 1050 m),01.VII.03, [4 valves], IBULRJ 1 7978 (22°1 1 '04"S, 39°47'04"W, 1650), 22.VJ.03, [10 valves], IBULRJ 17990 (22°11'16"S, 39°43'44"W, 1950 m), 22.VI.03, [18 valves], IBULRJ 18024 (22°04'33"S, 39°52'05"W, 1050 m), 30. VI. 03, [2 valves], IBULRJ 18023 (22°04'43"S, 39°49'09"W, 1350 m), 25.VI.03, [1 valve], IBULRJ 17953 (22°04'45"S, 39°46'31"W, 1650 m), 27.VI.03, [2 valves], IBULRJ 17996 (22°04'45"S, 39°41'58"W, 1950 m), 27.VI.03, [15 valves], IBULRJ 17938 (21°57'15"S, 39°47'41"W, 1650 m), 28.VI.03, [7 valves], IBULRJ 17989 (21°57'26"S, 39°40'34"W, 1950 m), 27.VI.03, [29 valves], IBULRJ 18012 (21°52'51"S, 39°48' 12"W, 1350 m), 26.VI.03, [2 valves], IBULRJ 17931 (21°52'41"S, 39°46'17"W, 1650 m), 26.VI.03, [27 valves], IBULRJ 17939 (21°52'43"S, 39°40'4T'W, 1950 m), 26.VI.03, [56 valves], IBULRJ 17974 (22°48'05"S, 40°06'38"W, 1950 m), 06.XII.03, [25 valves] , IBULRJ 1 7926 (22°4 1 ' 1 0"S, 40°02 ' 20"W, 1 650 m), 13.VI.03, [6 valves], IBULRJ 18004 (22°34'05"S, 40°00'12"W, 1350 m), 15.VI.03, [2 valves], IBULRJ 17946 (22°36'12"S, 39°58'22"W, 1650 m), 13.VI.03, [16 valves], IBULRJ 18007 (22°28'46"S, 39°53'27"W, 1650 m), 17.VI.03, [8 valves], IBULRJ 17977 (22°31'37"S, 39°55'14"W, 1650 m), 16.VI.03, [2 valves], MNHN (22°31'28"S, 40°03'49"W, 1050 m), 18. VI. 03, [17 valves]. Discussion This species may show variation in rostrum shape and features of ornamentation. The elongated rostrum is normally concave, pointing dorsally (Ligs. 4A-B) but some specimens show the rostrum almost straight. The shell ornamentation grades from almost smooth to about 9 concentric ribs. We suspect that the shell illustrated by Routiers (1984: 291, Hg. 4) as Cuspidnria sp. is in fact Octoporia octnpowsa, but only direct examination of this shell could resolve this matter. This species exhibits a simple pattern of distribution of micro-pits. Though smaller on rostrum (Lig. 4C) and on the lamellae, specifically on the distal ends (Lig. 4D), the micro-pits are dispersed over the entire shell (Ligs. 4C-L). Since there is no information about micro-pits on Octoporia spp. and because Octoporia octaporosa is the only species of the genus Octoporia studied herein, it is not possible to compare this pattern of distribution of micro-pits with other congeneric species. This species was quite abundant Distribution of Octoporia octaporosa (Allen and Morgan, 1981) References Locality Depth (m) Allen and Morgan (1981) Type locality: Atlantis II, sta. 92, 36°20.0'N, 67°56.0'W. 4800 Other material: 37°24.0'N to 0.0°46.0'S; 69°26.2'W to 29°28.0'W. 3459-5000 Present study Campos Basin, Rio de Janeiro, Brazil. 900-1950 Material examined PELECYPODS PROM DEEP WATERS OF CAMPOS BASIN 151 in our samples (743 valves), and in the Campos Basin it was present at 49 of 117 stations where pelecypods occurred. Genus Protociispidaria Allen and Morgan, 1981 Type species; Protocuspidaria verityi Allen and Morgan, 1981 by original designation in Allen and Morgan (1981) Genus characterization Shell small, rounded, equivalve, inequilateral, rostrate, laterally compressed. Umbo small, hemispherical, closer to anterior end. Rostrum very short, with variable width. Postero-ventral sinuation with individual variation, anterior margin rounded, and postero-dorsal margin nearly straight. Shell surface covered by countless striae, more conspicuous toward ventral margin and rostrum. Hinge edentulous, or with anterior lateral tooth on one or both valves. Resilifer small, central. Septum thin, membranous, with no muscle attachments to the shell (Adapted from Allen and Morgan 1981, Krylova 1995, Poutiers and Bernard 1995). Discussion This genus poses a great challenge, in part because, as pointed out by Allen and Morgan (1981; 495), “All species have a similar external appearance, in that they are small, rounded, laterally flattened with a very short rostrum”, and still according to Allen and Morgan (1981; 500), “there is sufficient variation between individuals to make identification other than by reference to dentition extremely difficult.” The genus Protocuspidaria was established by Allen and Morgan (1981), with species showing quite variable outlines. Three subgenera are characterized by the presence or absence of hinge teeth, or by which valve bears the teeth. Accordingly, the subgenus Protocuspidaria is characterized by an anterior tooth only on the right valve, the subgenus Edentaria Allen and Morgan, 1981 is characterized by a hinge devoid of teeth on both valves, and the subgenus Bidentaria Allen and Morgan, 1981 is characterized by an anterior tooth on both valves (Allen and Morgan 1981; 495, 497, and 499, respectively). Poutiers and Bernard (1995) stated that this subgeneric division does not represent this group in a realistic way. According to them, this scheme of separation based on the hinge structure fails to distinguish each group, but and Morgan (1981) raised to genus level. Subsequently, Poutiers (1984: 295, fig. 6a-b) described Protocuspidaria [Edentaria) thomassini and Krylova (1995) added 10 new species but recognized the family Protocuspidariidae with only two genera: Protocuspidaria (with three subgenera: Protocuspidaria, Bidentaria, and Edentaria) diagnosed by the presence of 7 tentacles on the siphon border and Multitentaculata Krylova, 1995 that shows between 7 and 33 tentacles on the siphon border. Besides that, Krylova (1995) subdivided Multitentaculata according to the hinge. The subgenus Multitentaculata s.s. lacks teeth, whereas the subgenus Dentaria would be characterized by the presence of an anterior lateral tooth on the right valve. According to Krylova (1995: 34) “the tentacles number at siphon border is, at time, the unique distinguishing morphological character between the subgenus Protocuspidaria and Multitentaculatad At the same time, Poutiers and Bernard (1995) and Morton (2003: 378, table 3) did not recognizes the family Protocus- pidariidae. Because there are intense discussions about the phylogeny of Septihranchia (Dreyer et al. 2003, Harper et al. 2006) and because of the lack of taxonomic evidence to maintain the higher-rank level of Protocuspidaria, we prefer to follow the authors that retain Protocuspidaria at the genus level [e.g., Allen and Morgan 1981, Poutiers and Bernard 1995) although some divisions may prove to be useful in the future. Micro- pits were absent in all species examined herein. Protocuspidaria [Protocuspidaria) verityi AWen and Morgan, 1981 (Figs. 4F-I) Protocuspidaria [P) verityi Allen and Morgan, 1981: 496- 497, figs. 61-62; Krylova, 1995: 31. Characterization Shell white, small (max. recorded length 5 mm), rounded, inequilateral, rostrate. Umbo small, closer to anterior end. Anterior margin rounded, usually giving rise to a plateau immediately next to the umbo. Rostrum truncate, usually short, variable in height. Postero-ventral sinuation with individual variation. Dorsal margin straight. Shell surface covered by countless striae. Micro-pits absent. Hinge with anterior lateral tooth only on right valve. Resilifer small, central. subgeneric division proposed by Allen and Morgan (1981). Scarlato and Starobogatov (1983) proposed the super- family Protocuspidarioidea and the family Protocuspidariidae, with each subgenus of Allen Distribution of Protocuspidark 1 (Protocuspidaria) verityi Allen and Morgan, 1981 References Locality Depth (m) Allen and Morgan (1981) Type locality: Atlantis 11, sta. 167, 7°58.0'S, 34°17.0'W to 7°50.0'S, 34°17.0'W. 943-1007 Other material: 47°35.5'N to 36°05.2'S; 11°35.0'E to 68°31.0'W. 943-4706 Present study Campos Basin, Rio de Janeiro, Brazil. 750-1950 152 AMERICAN MALACO LOGICAL BULLETIN 27 • 1/2 • 2009 Material examined IBULRI 17854 (21°58'36"S,39°46'30"W, 1700 m),08.X.01, [1 valve], IBULRI 17859 (22°07'17"S, 39°50'02"W, 1230 m), 13.V.02, [5 valves], IBULRJ 17861 (22°06'58"S, 39°48'41"W, 1330 m), 09.V.02, [3 valves], IBULRJ 17873 (22°05'45"S, 39°45'55"W, 1730 m), 09.V.02, [4 valves], IBULRJ 17878 (22°09' 10"S, 39°44'50"W, 1930 m), 08.V.02, [2 valves], IBULRJ 17893 (22°33'31"S, 40°12'05"W, 900 m), 18.V.02, [2 valves], IBULRJ 17896 (22°39'34"S, 40°08'22"W, 1200 m), 15.V.02, [1 valve], IBULRJ 17897 (22°10'54"S, 39°52'19"W, 1050 m), 10.XII.02, [7 valves], IBULRJ 17902 (22°11'16"S, 39°43'44"W, 1950 m), 25.XI.02, [3 valves], IBULRJ 17905 (22°04'43"S, 39°49'08"W, 1350 m), 24.XI.02, [1 valve], IBULRJ 17906 (22°04'46"S, 39°43'02"W, 1950 m), 24.XI.02, [2 valves], IBULRJ 17918 (21°52'44"S, 39°40'45"W, 1950 m), 11.XII.02, [3 valves], IBULRJ 17941 (21°52'43"S, 39°40'41"W, 1950 m), 26.VI.03, [2 valves], IBULRJ 18019 (22°27'31"S, 40°09'23"W, 750 m), 18.VI.03, [3 valves], IBULRJ 17944 (22°36'12"S, 39°58'22"W, 1650 m), 13.VI.03, [1 valve], IBULRJ 17956 (22°46'59"S, 40°07'49"W, 1650 m), 22.XI.02, [4 valves], IBULRJ 17962 (22°10'53"S,39°52'18"W, 1050 m), 01.VII.03, [ 1 valve], IBULRJ 17980 (22°11'04"S, 39°47'04"W, we cannot find micro-pits at any other Protocuspidaria species studied herein, we suppose that the absence of micro-pits might be a character of the generic level. In the Campos Basin it was present at 36 of 117 stations where pelecypods occurred. Protocuspidaria (Bidentaria) atlantica Allen and Morgan, 1981 (Ligs. 5A-L) Protocuspidaria [B.) atlantica Allen and Morgan, 1981: 499, figs. 64-67; Krylova, 1995: 33 Characterization Shell white, small (max. recorded length 5 mm), rounded, inequilateral, rostrate. Umbo small, hemispherical, closer to anterior end. Anterior margin rounded, usually giving rise to a plateau immediately next to the umbo. Rostrum truncate, variable in height. Postero-ventral sinuation with individual variation, from almost inconspicuous to quite accentuated. Dorsal margin straight. Shell surface covered by countless striae. Micro-pits absent. Hinge with anterior lateral tooth on both valves. Resilifer small, central. 1650 m), 22. VI. 03, [1 valve], IBULRJ 17963 (22°37'02"S, 39°56'20"W, 1950 m), 23.XI.02, [2 valves], IBULRJ 17969 (22°28'49"S, 39°53'24"W, 1650 m), 17.XI.02, [1 valve], IBULRJ 17968 (22°36'03"S, 39°57'54"W, 1650 m), 16.XI.02, [1 valve], IBULRJ 17982 (22°38'53"S, 40°04'14"W, 1350 m), 23. XI. 02, [3 valves], IBULRJ 17991 (22°11'16"S, 39°43'44"W, 1950 m), 22. VI. 03, [1 valve], IBULRJ 17992 (22°29'33"S, 39°56'17"W, 1350 m), 19.XI.02, [1 valve], IBULRJ 17997 (22°04'45"S, 39°41'58"W, 1950 m), 27.VI.03, [2 valves], IBULRJ 17998 (22°37'02"S, 39°56'20"W, 1950 m), 13.VI.03, [2 valves], IBULRJ 18001 (22°33'08"S, 39°54'21"W, 1950 m), 15.VI.03, [2 valves], IBULRJ 18016 (22°26'28"S, 39°54'08"VV, 1350 m), 21.VI.03, [1 valve], IBULRJ 18017 (22°24'30"S, 39°57'28"W, 1050 m), 20.VI.03, [2 valves], IBUFRJ 18018 (22°35'04"S, 40°08'53"W, 1050 m), 21.XI.02, [2 valves], IBUFRJ 18020 (21°52'59"S, 39°55'32"W, 750 m), 29.VI.03, [1 valve], IBUFRJ 17949 (22°41'35"S, 40°00'45"W, 1950 m), 22. XI. 02, [2 valves], IBUFRJ 17874 (22°08'23"S, 39°46'23"W, 1730 m), 09.V.02, ]1 valve], IBUFRJ 17884 (22°06'52"S, 39°44'13"W, 1930 m), 08.V.02, ]3 valves], IBUFRJ 17986 (22°31'28"S, 40°03'50"W, 1050 m), 19.X1.02, ]4 valves], MNHN (22°03'03"S,39°50'32"W, 1230 m), I3.V.02, |2 valves]. Distribution of Protocuspidaria (Bidentaria) atlantica Allen and Morgan, 1981 References Locality Depth (m) Allen and Morgan (1981) Type locality: Discovery, sta. 6696, 28°6.0'N, 13°28.0'W. 1780 Other material: 46°31.2'N to 28°06.0'N; 66°47.0'W to 10°19.5'W. 1150-4706 Present study Campos Basin, Rio de Janeiro, Brazil. 900-1950 Material examined IBUFRJ 14997 (21°58'36"S, 39°46'30"W, 1700 m), 08.X.2001 [5 valves], IBUFRJ 14998 (21 °57'05"S, 39°49'58"W, 1200 m), 24.IX.2001 [4 valves], IBUFRJ 17858 (22°05'04"S, 39°50'01"W, 1230 m), 09.V.02, [4 valves], IBUFRJ 17864 (22°04'52"S,39°49'04"W, 1330 m),09.V.02, [2 valves], IBUFRJ 17891 (22°38'01"S, 40°17'26"W, 900 m), 18.V.02, [2 valves], IBUFRJ 17894 (22°37'54"S, 40°13'36"W, 1000 m), 19.V.02, [3 valves], IBUFRJ 17921 (21°52'41"S, 39°46'17"W, 1650 m), 11.XII.02, ]3 valves], IBUFRJ 17928 (22°41 ' 10"S, 40°02'20"W, 1650 m), 13.V1.03, [2 valves], IBUFRJ 17929 (22°30'33"S, 39°51'45"W, 1950 m), 23. X1.02, [1 valve], IBUFRJ 17937 (21°57'15"S, 39°47'41"W, 1650 m), 28.V1.03, ]2 valves], IBUFRJ 17945 (22°36'I2"S, 39°58'22"W, 1650 m), 13.VI.03, [1 valve], IBUFRJ 17961 (22°I0'53"S, 39°52' I8"W, 1050 m), 01.VII.03, ]7 valves], IBUFRI 17965 (22°37'02"S, 39°5(V20"W, 1950 m), 23.XI.02, |l valve], IBUFRI 17970 (22°28'49"S, Discussion Despite the many individuals examined, no specimen showed micro-pits on any part of the shell (Fig. 41). Since 39°53'24"W, 1650 m), I7.XI.02, ]1 valve], IBUFRJ 17973 (22°48'05"S, 40°06'38"W, 1950 m), 06.XI1.03, [3 valves], IBUFRI 17975 (22°3 I '37"S, 39°53' 1 4"W, 1630 m), 16.V1.03, |3 valves], IBUFRJ 17983 (22°38'33"S, 40°04' I4"W, 1330 m), PELECYPODS FROM DEEP WATERS OF CAMPOS BASIN 153 23.XL02, [3 valves], IBUFRJ 17984 (22°31'28"S, 40°03'50"W, 1050 m), 19.XI.02, [2 valves], IBUFRJ 17987 (2r57'26"S, 39°40'34"W, 1950 m), 27.VI.03, [2 valves], IBUFRJ 17993 (22°29'33"S, 39°56'17"W, 1350 m), 19.XI.02, [3 valves], IBUFRJ 17994 (22°27'18"S, 39°54'50"W, 1350 m), 17.XI.02, [3 valves], IBUFRJ 18002 (22°41'03"S, 40°02'29"W, 1650 m), 23.XI.02, [3 valves], IBUFRJ 18005 (22°34'05"S, 40°00'12"W, 1350 m), 15.VI.03, [1 valve], IBUFRJ 18006 (22°28'46"S, 39°53'27"W, 1650 m), 17.VI.03, [1 valve], IBUFRJ 18009 (22°3U28"S, 40°03'49"W, 1050 m), 18.VI.03, [8 valves], IBUFRJ 18013 (21°52'5T'S, 39°48'12"W, 1350 m), 26.VI.03, [4 valves], IBUFRJ 18015 (21°52'51"S, 39°48'11"W, 1350 m), 12.XII.02, [5 valves], MNHN (22°41'31"S,40°00'47"W, 1950 m), 06. XII. 03, [2 valves]. Discussion This species shows the most variable outline and teeth variation in shape, but the presence of the anterior lateral teeth in both valves is diagnostic (Figs. 5D-E). These teeth can vary in their degree of development according to the size of the specimen (e.g., Allen and Morgan 1981, Poutiers and Bernard 1995), but this kind of variation, or expression, lacks taxonomic significance. Tike all other species of this genus, no micro-pits were observed, even at high magnification (Fig. 5F). In the Campos Basin, it was present at 28 of 117 stations where pelecypods occurred. Pwtocuspidaria {Bidentaria) jarauara sp. nov. (Figs. 5G-M) Myonera aff. ruginosa auct. non Jeffreys, 1881: Absalao etai, 2003: 327, figs. 10-11 Description Shell white, small (max. recorded length 5 mm), rounded, inequilateral, rostrate, laterally compressed. Umbo small, closer to anterior end. Anterior margin rounded, an anterior plateau present immediately next to the umbo. Rostrum large, truncate. Ventral and dorsal margins of the rostrum sub- parallel. Shell surface covered by countless striae. Micro-pits absent. Hinge with bifid anterior lateral tooth on both valves. Resilifer small, central. Etymology This species is named in honor ot the Jarauara Indians, one of the indigenous peoples of Brazil. The name is employed as a noun in apposition. Distribution Restricted to Campos Basin, Rio de Janeiro state, Brazil. Holotype IBUFRJ 14996 (21°58'36"S, 39°46'30''W, 1700 m), 08.X.2001 ] 1 spec.]. Paratype IBUFRJ 17888 (22°05'H"S, 39°42'40"W, 1930 m), 08. V.02, ]2 valves], MNRJ 12860 (22°10'54"S, 39°48'59"W, 1350 m), 25.VI.03, [2 valves], MZUSP 40957 (22°04'45"S, 39°46'31"W, 1650 m), 27.VI.03, [2 valves], MNHN (22°10'55"S, 39°49'00"W, 1350 m), 10.XJI.02, [2 valves], MNHN (21°52'44"S, 39°40'45"W, 1950 m), 11.XII.02, [2 valves]. Other material examined IBUFRJ 17862 (22°06'58"S, 39°48'41"W, 1330 m), 09. V.02, [1 valve], IBUFRJ 17872 (22°05'45"S, 39°45'55"W, 1730 m), 09.V.02, [4 valves], IBUFRJ 17876 (22°08'23"S, 39°46'23"W, 1730 m), 09.V.02, [1 valve], IBUFRJ 17877 (22°09'10"S,39°44'50"W, 1930 m), 08.V.02, [ 1 valve], IBUFRJ 17883 (22°06'52"S, 39°44'13"W, 1930 m), 08.V.02, ]5 valves], IBUFRJ 17892 (22°38'01"S, 40°17'26"W, 900 m), 18.V.02, [1 valve], IBUFRJ 17898 (22°10'55"S, 39°49'00"W, 1350 m), 10. XJI.02, [1 valve], IBUFRJ 17903 (22°11'16"S, 39°43'44'W, 1950 m), 25.XI.02, [1 valve], IBUFRJ 17908 (22°04'46"S, 39°43'02"W, 1950 m), 24.XI.02, [1 valve], IBUFRJ 17913 (21°57'15"S, 39°47'43"W, 1650 m), 14.XII.02, [1 valve], IBUFRJ 17914 (21°57'26"S, 39°40'33"W, 1950 m), 11.XII.02, [1 valve], IBUFRJ 17920 (21°52'41"S, 39°46'17''W, 1650 m), 11. XII.02, [1 valve], IBUFRJ 17927 (22°41'10"S, 40°02'20"W, 1650 m), 13.VI.03, [1 valve], IBUFRJ 17942 (2r52'43"S, 39°40'41"W, 1950 m), 26.VI.03, [2 valves], IBUFRJ 17943 (22°36'12"S, 39°58'22"W, 1650 m), 13.VI.03, [1 valve], IBUFRJ 17948 (22°41'35"S, 40°00'45"W, 1950 m), 22.XI.02, [1 valve], IBUFRJ 17955 (22°46'59"S, 40°07'49"W, 1650 m), 22.XI.02, [1 valve], IBUFRJ 17957 (22°46'59"S, 40°07'49"W, 1650 m), 22.XI.02, [3 valves], IBUFRJ 17960 (22°10'53"S, 39°52'18"W, 1050 m), 01.VII.03, [2 valves], IBUFRJ 17971 (22°28'49"S, 39°53'24"W, 1650 m), 17.XI.02, [3 valves], IBUFRJ 17972 (22°48'05"S, 40°06'38"W, 1950 m), 06.XII.03, [1 valve], IBUFRJ 17976 (22°31'37"S, 39°55'14"W, 1650 m), 16.VI.03, [4 valves], IBUFRJ 18028 (22°31'36"S, 39°55'15"W, 1650 m), 16.XI.02, [1 valve], IBUFRJ 18061 (21°58'36"S, 39°46'30"W, 1700 m), 08.X.2001 [1 valve]. Discussion The diagnostic character of this species is a bifid anterior lateral tooth on both valves (Figs. 5K-F), since this bifid tooth is absent in all other species previously reported in the genus. The presence of this tooth could suggest that a fourth subgenus is present — and still unnamed — if one used exclusively hinge characters to determine the subgenera of Protociispidaria. But, because we do not have any other information about soft parts or any other kind of data beyond the conchological one, we prefer to keep the new species in Bidentaria. Exteriorly, Protocuspidavia {Bidentaria) jarauara sp. nov. could be initially confused with Protociispidaria {Bidentaria) 154 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 atlantica and Proto cusp idaria [Protocuspidaria] verityi, but the hinge differences clearly distinguish the three species. Absalao et al. (2003: 327) has previously assigned Protocuspidaria (Bidentaria) jarauara sp. nov. to the Brazilian coast under the name Myonera aff. ruginosa Jeffreys, 1881. In fact, the description of the genus Protocuspidaria is very similar to the description given for Myonera ruginosa Jeffreys (1881: 942, pi. LXXI, fig. 7), which distinguishes M. ruginosa by the external surface of the shell covered by narrow concentric striae, a short and truncated rostrum, anterior border rounded, a small prominent umbo, and an anterior tooth on the left valve. So, the identification of M. ruginosa to Brazil must be disregarded. The transfer of M. ruginosa to the genus Protocuspidaria was first suggested by Allen and Morgan (1981: 995) and followed by Krylova ( 1995: 29). No micro-pits were observed, even at high magnification (Eig. 5M). In the Campos Basin it was present at 29 of 117 stations where pelecypods occurred. GENERAL DISCUSSION Reflecting the difficulties involved in collecting material from deep waters, and despite the efforts of several investigators over the past 30 years (Allen and Turner 1974, Allen and Morgan 1981, Leal and Simone 2000, Absalao et al. 2001, 2003, 2005, Simone 2002, 2003, Absalao and Pimenta 2003, 2005, Absalao and Santos 2004, Caetano et al. 2006, Simone and Cunha 2006, Zelaya et al. 2006, Barros et al. 2007, Lima and Barros 2007), the Brazilian deep-water species are essentially unknown. Most (five of eight) of the species reported here were not previously recorded in Brazilian waters. Two species are new to science [Myonera kaiwa sp. nov. and Protocuspidaria jarauara sp. nov.) and for one taxon, Myonera sp., a formal epithet will be delayed until additional material is available. Except for Myonera paucistriata, which is probably the most common of all septibranchs at Campos Basin, all others species studied herein have their known range expanded geographically and/or bathymetrically. Protocuspidaria verityi though well represented at North Atlantic Ocean, has been scarcely represented at South Atlantic Ocean, with an occurrence gap between the latitudinal coordinates 09° and 36°S. Myonera liinatula and Protocuspidaria atlantica are for the first time recorded in the South Atlantic Ocean and Brazil. Octoporia octaporosa had been previously recorded for the South Atlantic Ocean, but we have enlarged its range to the .south. Bathymetrically, O. octaporosa (900 m), P. verityi (750 m), and P. atlantica (900 m) show their shallowest record, while M. liinatula { 1700 m) shows its deepest record. The.se new data show that our understanding of the taxonomic composition and distribution of deep-water pelecypod species inhabiting Brazilian coast is still unsatisfactory. Under high resolution of a Scanning Electron Microscope (SEM), a character not yet reported for septibranchs was observed: the presence of micro-pits on the shell surface. The distribution of micro-pits found on the species studied here is not random and seems to be a taxonomically useful pattern. Only the type species of Octoporia, O. octaporosa, is represented in our samples. This is numerically the most abundant species sampled and, in spite of the variation in the shell ornamentation, which grades from almost smooth to concentrically ribbed, all specimens show the same pattern of distribution of the micro-pits. On Protocuspidaria, despite the many individuals and the three species examined in this paper, no micro-pits were observed on any part of the shell. These findings suggest the importance of the micro-pits for taxonomic proposals. Eor the genus Myonera, only the genus type species, M. paucistriata, has no micro- pits. The other species studied herein (four taxa) exhibit, each one, a different pattern of distribution of the micro- pits on the shell surface and additional research is needed to establish the potential use of such micro-pits in this taxonomic category. The micro-pits resembles the “pores” described for polypla- cophorans that house the aesthetes and have been observed in other molluscs [e.g, according to Reindl and Haszprunar (1996), in Polyplacophora, Leptochiton cancellatus (Sowerby, 1839); Gastropoda, Diodora graeca (Linnaeus, 1758); and Pelecypoda, Area noae Linnaeus, 1758)]. The homology of these pits is not established and their probable function is a matter for speculation, with widely differing interpretations. Some authors have suggested that their function is sensory (Baxter et al. 1990), or for excretion (Waller 1980), or for maintenance of the periostracum (Baxter et al. 1987). The function of such micro-pits for septibranchs is thus still unclear and open to future research. ACKNOWLEDGMENTS Our greatest thanks are to Dr. John Allen (University Marine Biological Station) for his encouragement, exchange of ideas, bibliography, and critical information on all aspects of this work; Dr. Alexandre Pimenta (Museu Nacional/UFRI) for his help on nomenclatural matters; Dr. Paula MikkeLsen (Paleontological Research Institution) and one anonymous reviewer for their criticisms and suggestions that improved this manuscript; Ms. Raquel Pigueira (IIPRI) 1(U' the first English revision of this ms and Dr. lanet Reid for the final English revision ol this paper; and PE'fROBRAS (Brazilian Petroleum Go.) for making this material available and for SEM support. This research was partially supported by fellowships from PAPER) (P'uIuk^'ao de Amparo a Pesquisa PELECYPODS PROM DEEP WATERS OF CAMPOS BASIN 155 do Estado do Rio de Janeiro) to the senior author and from CNPq (Conselho Nacional de Desenvolvimento Cientifico e Tecnologico) to the second author. LITERATURE CITED Abbott, R. T. 1974. American Seashells, 2"^* Edition, van Nostrand Re- inhold Co., New York. Absalao, R. S. and A. D. Pimenta. 2003. A new subgenus and three new species of Brazilian deep waters Olivella (Mollusca, Gas- tropoda, Olivellidae) collected by the RV Marion Dufresne in 19S7 . Zoosystema 25: 177-185. Absalao, R. S. and F. N. Santos, 2004. Recent deep-sea species of Benthonellania Lozouet, 1990 (Gastropoda, Rissoidea) from the south-western Atlantic with description of two new species utilizing a shell morphometric multivariate. Journal of Con- chology 38: 329-340. Absalao, R. S. and A. D. Pimenta. 2005. New records and new spe- cies of Vetulonia Dali, 1913 and Brookula Iredale, 1912 from Brazil (Gastropoda, Trochidae). The Veliger47: 193-201. Absalao, R. S., C. Miyaji, and A. D. Pimenta. 2001. The genus Brookula Iredale, 1912 (Gastropoda: Trochidae) from Brazil: Description of a new species, with notes on other South Ameri- can species. Zoosystema 23: 675-687. Absalao, R. S., C. H. Gaetano, and A. D. Pimenta. 2003. Novas Ocor- rencias de Gastropodes e Bivalves Marinhos no BrasU (Mollusca). Revista Brasileira de Zoologia 20: 323-328 [In Portuguese]. Absalao, R. S., A. D. Pimenta, and C. H. Gaetano. 2005. Turridae (Mollusca, Gastropoda, Conoidea) coletados durante as cam- panhas do programa REVIZEE/Score Central (1996-2002). Biociencias 13: 19-47 [In Portuguese]. Allen, J. A. and J. F. Turner. 1974. On the functional morphology of the family Verticordiidae (Bivalvia) with descriptions of new species from the abyssal Atlantic. Philosophical Transactions of the Royal Society of London (B) 268: 401-536. Allen, J. A. and R. E. Morgan. 1981. The functional morphology of Atlantic deep-water species of the families Cuspidariidae and Poromyidae (Bivalvia) - an analysis of the evolution of the Sep- tibranch condition. Philosophical Transactions of the Royal Soci- ety of London (B) 294: 413-546. Barros, J. C. N, S. F. B. Lima, and J. A. Francisco. 2007. Two new spe- cies of Adis (Mollusca: Gastropoda: Aclididae) from tbe conti- nental slope of northeast Brazil. Zootaxa 1614: 61-68. Baxter, J. M., A. M. Jones, and M. G. Sturrock. 1987. The ultra- structure of aesthetes in Tonicella marmorea (Polyplacophora: Ischnochitonina) and a new functional hypothesis. Journal of Zoology 2U: 589-604. Baxter, J. M., M. G. Sturrock, and A. M. Jones. 1990. The structure of the intrapigmented aesthetes and the periostracum layer in Callochiton achatinus (Polyplacophora). /oHn?n/ of Zoology 220: 447-468. Bernard, F. R. 1974. Septibranchs of the Eastern Pacific (Bivalvia Anomalodesmata) . Allan Hancock Monographs in Marine Biology, no. 8: 1-279. Bush, K. J. 1885. Additions to the shallow- water Mollusca of Cape Hatteras, N.C., dredged by the U.S. Fish Commission Steamer ‘Albatross,’ in 1883 and 1884. Transactions of the Connecticut Academy of Arts and Sciences 6: 453-480. Gaetano, C. H., V. Scarabino, and R. S. Absalao. 2006. Scaphopoda (Mollusca) from the Brazilian continental shelf and upper slope (13° to 21°S) with descriptions of two new species of the genus Cadulus Philippi, Zootaxa 1267: 1-47. Dali, W. H. 1881. Reports on the results of dredging, under the su- pervision of Alexander Agassiz, in the Gulf of Mexico, and in the Caribbean Sea, 1877-1879, by the U.S. coast survey steamer ‘Blake’ XV. Preliminary report on the Mollusca. Bulletin of the Museum of Comparative Zoology 2: 33-144. Dali, W. H. 18862.. Neaera. Nature 34: 122. Dali, W. H. 1886b. Report on the Mollusca Part I - Brachiopoda and Pelecypoda - Reports on the results of dredging under the supervision of Alexander Agassiz, in the Gulf of Mexico (1877- 79), and in the Caribbean Sea (1879-80), by the U. S. Coast Survey Steamer ‘Blake’. Bulletin of the Museum of Comparative Zoology 12: 171-318. Dreyer, H., G. Steiner, and E. M. Harper. 2003. Molecular phylog- eny of Anomalodesmata (Mollusca: Bivalvia) inferred from 18S rRNA sequences. Zoological Journal of the Linnean Society 139: 229-246. Fischer, P. 1887. Manuel de Conchyliologie et de Paleontologie Con- chyliologique on LJistoire Naturelle des Mollusques Vivants et Fossiles. Librairie F. Savy, Paris [In French]. Grasse, P. 1960. Traite de Zoologie. Tome V. Masson et Cie Editeurs, Paris. Pp. 1053-2219 [In French]. Harper, E. M., E. A. Hide, and B. Morton. 2000. Relationships between the extant Anomalodesmata: A cladist test. In: E. M. Harper, J. D. Taylor, and J. A. Crame, eds., The Evolutionary Biology of the Bivalvia.Vol. 177. Geological Society, London. Pp. 129-143. Harper, E. M., H. Dreyer, and G. Steiner. 2006. Reconstructing the Anomalodesmata (Mollusca: Bivalvia): Morphology and mol- ecules. Zoological Journal of the Linnean Society 148: 395-420. ICZN. 1999. International Code of Zoological Nomenclature, 4''’ Edi- tion. International Commission on Zoological Nomenclature, London. Jeffreys, G. 1881. On the Mollusca procured during the ‘Lightening’ and ‘Porcupine’ expeditions 1868-70, Part V. Proceedings of the Zoological Society of London. Pp. 656-687. Knudsen, J. 1982. Anomalodesmata (Mollusca, Bivalvia) from Saba-Bank, the Caribbean region. Proceedings of the Konin- klijke Nederlandse Akademie Van Wetenschappen (C) 85: 121-146. Krylova, E. M. 1994. Clams of the genus Octoporia (Septibranchia, Halonymphidae) in the world oceans. Zoologicheskyi Zhurnal 73: 38-45 [In Russian]. Krylova, E. M. 1995. Clams of the family Protocuspidariidae (Sep- tibranchia, Cuspidarioidea): Taxonomy and distribution. Zoo- logicheskyi Zhurnal 74: 20-38 [In Russian]. Kuroda, T. 1952. On the Verticordiidae from Japan. Venus 17: 6-16. Leal, L H. and L. R. L. Simone. 2000. Copulahyssia riosi, a new deep- sea limpet (Gastropoda: Pseudococculinidae) from the conti- 156 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 nental slope off Brazil with comments on the systematics of the genus. The Nautilus 114: 59-68. Lima, S. F. B. and J. C. N. Barros. 2007. Two new species of CerithieT la (Apogastropoda: Cerithiopsidae) for the continental slope of Pernambuco (northeast Brazil). Zootaxa 1441: 63-68. Marini, A. C. 1974. O genero Verticorciio Wood, 1844 (Bivalvia: Ver- ticordiidae) na plataforma continental brasileira. Papeis do Departainento de Zoologia de Sdo Paulo 28: 241-244 [In Por- tuguese]. Morton, B. 1981. The Anomalodesmata. Malacologia 21: 35-60. Morton, B. 2003. The functional morphology of Bentholyonsia tera- machii (Bivalvia: Lyonsiellidae): Clues to the origin of preda- tion in the deep water Anomalodesmata Journal of Zoology 261: 363-380. Newell, N. D. 1969. Classification of Bivalvia. In: R. C. Moore, ed.. Treatise on Invertebrate Paleontology. Part N. Mollusca 6. Vol. 1. Bivalvia. Geological Society of America and University of Kansas, Boulder, Colorado and Lawrence, Kansas. Pp. N205-N224. Pelseneer, P. 1911. Les Lamellibranches de I’expedition du Siboga, par- tie anatomique. Siboga Expedition (A) 53: 1-125 [In French [. Poutiers, J. M. 1984. Septibranches abyssaux de I’ocean indien occi- dental (mollusques bivalves Anomalodesmata). Journal of Con- chology 31: 281-306 [In French]. Poutiers, j. M. and F. R. Bernard. 1995. Carnivorous bivalve mol- lusks (Anomalodesmata) from the tropical western Pacific Ocean, with a proposed classification and a catalogue of Recent species. In: P. Bouchet, ed., Resultats des Campagnes MUSORS- TOM, Vol. 14. Memoirs du Museum national d’Histoire natureT le 167: 107-187. Reindl, S. and G. Haszprunar. 1996. Shell pores (caeca, aesthetes) of Mollusca: A case of polyphyly. In: J. Taylor, ed.. Origin and Evo- lutionary Radiation of the Mollusca. Oxford University Press, The Malacological Society of London, Oxford. Pp. 115-118. Rios, E. C. 1994. Seashells of Brazil, 2"‘* Edition. Fundac^ao Universi- dade Federal do Rio Grande, Rio Grande, Brazil. Rosenberg, G. 2005. Malacolog 4.1.0: A Database of Western Atlan- tic Marine Mollusca. Available at: http://www.malacolog.org/ 17 September 2008. Scarabino, F. 2003. Lista sistematica de los Bivalvia marinos y es- tuarinos vivientes de Uruguay. Comunicaciones de la Sociedad Malacoldgica del Uruguay B: 227-258 [In Spanish]. Scarlato, O. A and Y. I. Starobogatov. 1983. System of the bivalve mollusks of the superorder Septibranchia. In: 1. M. Likharev, ed.. Molluscs. Their Systematics, Ecology and Distribution. Nauka, Leningrad. Pp. 7-13. Simone, L. R. L. 2002. Three new deepwater species of Eulimidae (Caenogastropoda) from Brazil. Novapex3: 55-60. Simone, L. R. L. 2003. Revision of the genus Benthobia (Caenogas- tropoda, Pseudolividae). Journal oj Molluscan Studies 69: 245- 262. Simone, L. R. L. and C. M. Cunha. 2006. Revision of genera Claza and Callogaza ( Vetigastropoda, Trochidae), with description of a new Brazilian species. ZooRuv) 1318: 1-40. Thiele, |. 1935. Ilandbuch dcr Systematischen Weichticrkundc. Gustav ITscher, Jena. Vol. 2. Pp. 1023-1 134 [In German [. Vokes, H. E. 1967. Genera of the Bivalvia, a systematic and biblio- graphic catalogue. Bulletins of American Paleontology 51: 111-394. Waller, T. R. 1980. Scanning electron microscopy of shell and man- tle in the order Arcoida (Mollusca, Bivalvia). Smithsonian Con- tributions of Zoology 313: 1-58. Zelaya, D. G., R. S. Absalao, and A. D. Pimenta. 2006. A revision of Benthobrookula Glarke, 1961 (Gastropoda, Trochoidea) in the Southwestern Atlantic Ocean. Journal of Molluscan Studies 72: 77-87. Submitted: 27 June 2008; accepted: 3 December 2008; final revisions received: 20 March 2009 Amer. Malac. Bull. 27: 157-165 (2009) X-ray quantitative texture analysis on Helix aspersa aspera (Pulmonata) shells selected or not for increased weight Daniel Chateigner*, Rainier Kaptein^ and Mathilda Dupont-NiveP 'Laboratoire CRISMAT-ENSICAEN, UMR CNRS n°6508, and lUT-Caen, Universite de Caen - Basse Normandie, 6 Boulevard Marechal Juin 14050 Caen, Erance "INRA, UR544 Unite de Genetique des Poissons, F-78350 Jouy-en-Josas, Erance Corresponding author: daniel.chateigner@ensicaen.fr Abstract: X-ray Quantitative Texture Analysis (QTA) results are examined for the outer aragonitic shell layers of Helix aspersa aspersa (Muller, 1774) to probe the relevance of the approach to non-flat surfaces. Two sets of H. aspersa aspersa were studied, for a total of 29 samples. Quantitative texture analysis showed that although the nature of the texture present was roughly constant, the textural strength varied significantly among specimens because of biologically inherited surface irregularities. A statistical analysis showed that textural strength exhibited larger standard deviations for snails selected for greater shell weight than for control snails. The H. aspersa aspersa aragonite texture is the same as observed in previous studies, with <110> shell growth directions. This texture causes elastic behavior of the mineral part of the shell, which accommodates moderate shear and compression. We furthermore determine that the colored bands at the shell surface were aligned with the <020> crystal directions. Key words: Helix texture, aragonite, shell growth Quantitative Texture Analysis (QTA) is frequently used to characterize the macroscopic organization of layered crystals in mollusc shells. A high degree of order (or textural strength) has been reported (Hedegaard and Wenk 1998, Chateigner et al. 1999), which varies among taxa, with qualitatively identical textures in closely related species (Chateigner et al. 2000). In a single specimen, textures can vary with location in the shell, either between different layers as in Cypraea testudinaria (Linnaeus, 1758) (Chateigner et al. 1996) or in the same layer, e.g., in Pterioida (Zolotoyabko and Quintana 2002, Checa and Rodriguez-Navarro 2005). The organic matrices control the inorganic crystal orienta- tion (Falini et al. 1996) and the crystal shapes themselves (Aizenberg et al. 1996), but these two traits can be seen as non-redundant characters in terms of phylogeny. For instance, it has been demonstrated that scanning electronic microscopy images can be misleadingly interpreted in terms of orientation (Chateigner et al. 2000). QTA has also been proposed to link living species to extinct fossils in the Bivalvia (Chateigner et al. 2002). Two types of QTA have been applied to the Mollusca in the literature, which differ in the radiation used, thereby probing different material scales. While X-ray diffraction was formerly used (Chateigner et al. 1996, 1999, 2000, 2002, Hedegaard and Wenk 1998), Electron Backscatter Diffrac- tion (EBSD) has more recently provided a way for local characterization of texture variation in molluscs (Checa et al. 2005, Rousseau et al. 2005). Dealing with X-ray analysis using whole X-ray diffraction profiles accounts for all the crystallites. even the smallest ones. However, the X-ray beam extends on the specimens’ surface for several mm^ during the measurements. This simple fact alters the results because of the irregular surface of specimens and prevents quantitative results. Only once in the literature have QTA measurements been made on two specimens of the same species. Helix aspersa, and these indicated variability in the quantification of the texture although qualitatively, orientations were the same (Chateigner et al. 2000). Quantitative variation in the results {e.g., variation in textural strength) could also come from real variation in specimen textures, for instance due to a growth anomaly, rather than from an artifact from the irregular irradiation of the surface. These two latter effects could also be explained by the natural texture variation inherent to a species reared in different conditions. In Europe, snail production has increased considerably in the last two decades. Snail farming could give rise to modifications of shell growth that could be undesirable from an economical point of view. In par- ticular, rearing larger snails for an increased tissue weight could modify the shell texture because of the faster growth (Dupont-Nivet et al. 2000b). In turn, the texture modifica- tions could affect other shell characteristics (mechanical properties, colors, etc.). Here we statistically analyzed X-ray QTA results of outer aragonitic layers of Helix aspersa aspersa shells, obtained under controlled conditions to optimize growth. Our first aim was to examine a potential effect of the selection method used to increase weight on the degree of preferred orientation. 157 158 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 We then determined the distribution of QTA results on equally shaped samples to show how this distribution changed in two lines of animals, one selected for growth and the other not. Both lines were reared in the same environmental con- ditions. Our second aim was to exemplify how QTA results can vary between individuals when using X-ray investiga- tions of texture, using the usual measurement characteristics. The certification of the methodology allowed further identihcation of alignment of given crystal axes with specihc shell directions, such as in between shell layers and with colored bands. MATERIALS AND METHODS Figure 1. Shell samples used for X-ray measurements. S-specimens come from a line selected for increased weight while C-specimens were control samples. The arrows indicate the growth direction G that guides sample positioning with respect to the X-ray beam. The specimens of Helix aspersa aspersa used in these experiments were from two lines: one selected for increased weight (S) and a control line (C). In the S line, at each generation, the largest snails were chosen for reproduction while in the C line, breeders were chosen at random. The selection procedure is further described in Dupont-Nivet et al. (2000b). Animals were all reared in the same room with environmental conditions optimal for growth (density, food, temperature, relative humidity, and photoperiod) as detailed elsewhere (Dupont-Nivet et al. 1998, 2000a). Selected snails used in this experiment were from the 7'^ generation. Their mean weight was 16.98 g versus 10.61 g for the control snails. We collected ‘adult’ snails, i.e., snails for which the peristome was reflected and, thus, shell growth was completed. Eourteen and fifteen samples were analyzed for the C and S lines, respectively. They were chosen from the whole population available, with weight and age close to the population means, i.e., at a similar growth stage. Shell specimens were all prepared the same day according to the following procedure. Animals were frozen at -18 °C, thawed after one day, and the body manually separated from the shell. Shells were washed with water and air-dried. Eor the X-ray QTA experiments, a sample of approx. 1 X 1 cm^ was cut out the mollusc shells about 0.5 cm from an omnipresent growth irregularity near the macroscopic margin of the shell (Eig. 1). The position of the sample on the diffractometer was as in Chateigner et al. (1999). The pole figure plotting was with the projection normal as the N direction of the shell, with the G and M directions respec- tively vertical and horizontal in the pole figure projection planes (Fig. 2). The Helix aspersa aspersa shell is composed typically of 95% crystallized biogenic aragonite, and of approx. 5% in volume of residual materials, mainly intercrystalline and intracrystalline biomolecules. I lowever, these two latter com- ponents are not visible and do not perturb the X-ray diffrac- tion diagrams because of their weak pre.sence and scattering factors and their poor crystallization. Aragonite is one of the three CaCO^ polymorphs and crystallizes in the orthorhombic Pmcn space group with the following cell parameters: a = 4.961 1 A, b = 7.9672 A, and c = 5.7407 A for the reference non-biogenic mineral (Pilarti et al. 1998). The X-ray QTA measurements were carried out using a 4-circle diffractometer and a monochromatized Cu-Ka averaged radiation (1.5418 A) in point focused tube mode (Ricote and Chateigner 2004), with a beam cross-section of 1 X 1 mm^. The sample was mounted in the center of an Eulerian Cradle (Huber) and rotated in all necessary space directions (at a fixed X-ray incident angle (O = 1 6.64°, scanning for 0 < X < 60° and 0 < d < 355° with 5° steps). Each diagram was acquired for 60 seconds, using the Curved Position Sensitive detector (CPS 120, INEL) which spans all the Bragg diffracted intensities in a 120° 20-range at once for all given sample orientations (Eig. 3). After acquisition, data treatment and QTA involved the so-called “combined analysis” (Chateigner 2004) which used as a first step Rietveld’s (1969) refinement of all the 936 resulting diagrams. After this step, integral intensities were G B I'igiirc 2. A, Sample reference frame with margin (M), growth (G), and normal (N) directions; H, eorrespoiuling pole figure frame. TEXTURE ANALYSIS OE HELIX ASPERSA ASPERSA 159 Mono- chromator Eulerian Cradle X-ray tube Figure 3. A, Schematic and, B, picture of the X-ray instrumental set-up. Scale: 1: 20. extracted by the Le Bail extraction procedure (Le Bail et al. 1988) and used for quantitative texture analysis with the E-WIMV algorithm (Lutterotti et al. 2004). During this latter step, the Ori- entation Distribution Eunction (ODE) (Matthies etal. 1987) was refined. These steps were iterated 4 times to find the best solution at the convergence of the program, after which pole figures were reconstructed. Instrumental aberrations were calibrated on a standard LaB^ powder from NIST (SRM660b) and de-convoluted for all the acquired data. Defocusing aberrations with the tilt angle were refined on each sample since they depend on the sample curvature, using a polynomial approach (Chateigner 2004). Pole figures were normalized into orientation density values, expressed in “multiples of random distribution” units (or m.r.d.). In these units, samples without any texture (powders) exhibit homogeneous pole figures at the 1 m.r.d. level, while textured samples show maxima and minima in the pole figures, respectively above and below 1 m.r.d., the former corresponding to the texture components. The E-WIMV approach provided the maximum and minimum values of the ODE, which were quantitative appreciations of the texture strength for specific points of the orientation space. An overall texture strength value was the texture index (Bunge 1982). During the Rietveld and E-WIMV cycles, the phase cell parameters were also refined, together with other effects that could be detected (crystallite sizes, d-spacing microstrains, stresses, etc). Quality of the results were assessed by the reliability factors for the Rietveld (R^, R^,, R^^^) and ODE (R Rg.j.) refinements, respectively, as defined for this combined analysis (Chateigner 2004) and implemented in the MAUD package (Lutterotti et al. 1999). Scanning Electron Microscope images were obtained from secondary electrons using a Philips XL 30 EEC instru- ment at an operating voltage of 20 kV. Energy Dispersive Spectrometry could reveal only a single composition with respect to the CaCO^ stoichiometry. Four to five locations were measured for each specimen. RESULTS AND DISCUSSION Individual X-ray diagrams, measured on the same sample for different orientations, clearly showed the unique presence of textured aragonite (Fig. 4). Diagrams measured for different orientations of the sample exhibited different peak intensity ratios, indicating the texture was probably strong. None of these diagrams corresponded to randomly oriented powder. Peak positions corresponded only to the aragonite unit-cell. For this phase and our X-ray energy, the linear absorption coefficient was p, = 208 cm \ which corresponded to 99% of the diffracted intensity coming from typically the first 46 pm of the shell, i.e., approx, two thirds of the total shell thickness, then probing the outer, crossed-lamellar layers. For this probe depth the diffraction peaks were very narrow, a signature of well-developed crystallites, to sizes larger than our instrumental limit of typically 1 pm for their mean smallest dimension. No micro-distortion of the d-spacings could be detected, indicating that crystallization occurred in a smooth manner in these layers. The refinements of both textures (E-WIMV) and structures (Rietveld) were in good agreement with the experimental values as indicated by the reliability factors Figure 4. Plot of three spectra measured for three different couples of tilt and azimuth angles, respectively X and (j). The intensity ratio changes as a function of these two rotations, indicating preferred orientation. 160 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 (Table 1). On the 29 measured individuals, the Rietveld reliability factors ranged from roughly 14 to 30%, which for 936 diagrams (3 x 10^ measured points per sample) was considered as very good, and was also indicated by the low standard deviations ranging between 1 and 2 r.m.s. Lor the texture refinement (taking out extremes Si- 13 and S2-7), reliability factors ranged from approx. 12% to 32% (with 5 to 6 r.m.s. standard deviation) which again were satisfactory for the level of textures shown (Chateigner 2005). Refined cell parameters corresponded to the values of synthetic non- biogenic aragonite, and no difference was observed between the two sample sets within the standard deviations. We conclude selection does not affect aragonite structure. Orientation distribution function minima were all close to 0 m.r.d., indicating that 100% of the total shell volume was textured. Orientation distribution function maxima were very large and fluctuated strongly, with a tendency for larger ODE max in the control samples, even if both sets overlapped within their standard deviations. Variation of the ODE max values reached 35% of the mean value in C samples versus 50% in S. The same maximum variations were visible for the texture index values. Interestingly, standard deviations for texture strengths (ODE max and E‘) were lower for C samples than for S samples. This is particularly significant if we bear in Table 1. Refinement results from the MAUD package. Grayish specimens represent extreme results (larger texture reliability factors then worst refinements) and have been removed for mean and standard deviation calculations. The standard deviation resulting from the refinements on the parameters is typically 2 units on the last shown digit. Specimen R„ (%) R„ (%) R (%) Re, (%) Kr (%) a (A) b(A) c (A) ODF min (m.r.d.) ODF max (m.r.d.) F- (m.r.d.) Sl-1 21.79 27.70 19.18 18.94 17.33 4.9503 7.9540 5.7410 0.0005 142.00 12.60 Sl-3 19.95 25.46 17.39 15.13 13.74 4.9604 7.9760 5.7493 0.0027 160.33 14.45 Sl-5 19.88 25.46 17.88 13.39 12.89 4.9680 7.9803 5.7500 0.0006 223.39 20.4 Sl-8 20.09 25.97 17.16 15.21 15.35 4.9579 7.9658 5.7438 0.0001 278.05 33.98 Sl-9 22.41 28.73 18.45 16.61 15.89 4.9764 7.9956 5.7520 0.0001 252.27 21.97 Sl-10 19.52 24.85 19.52 13.79 13.92 4.9672 7.9809 5.7528 0.0344 94.99 7.60 Sl-13 17.01 21.44 15.34 13.23 11.74 4.9615 7.9676 5.7408 0.0057 183.23 18.12 Sl-14 23.27 30.09 20.90 15.31 13.88 4.9687 7.9813 5.1444 0.0001 191.49 19.62 Sl-21 19.83 25.51 17.64 12.77 11.85 4.9579 7.9662 5.7408 0.0002 177.16 24.13 S2-2 20.2 25.77 17.82 13.82 13.26 4.9674 7.9852 5.7453 0.0026 180.38 17.48 S2-3 19.62 25.21 17.60 13.08 12.96 4.9580 7.9740 5.7426 0.0003 206.53 19.58 S2-5 22.72 29.16 17.59 20.62 20.15 4.9559 7.9830 5.7465 0.0154 97.94 9.42 S2-7 20.07 25.12 14.24 36.67 35.8 4.9636 7.9888 5.7409 0.0390 85.40 5.30 S2-11 20.07 28.16 16.79 14.40 14.53 4.9545 7.9627 5.7448 0.0004 142.40 18.61 S211b 22.56 28.85 17.85 19.00 18.35 4.9539 7.9589 5.7433 0.0014 161.01 16.85 Mean 20.60 26.50 17.69 16.80 16.11 4.9614 7.9747 5.7452 0.0069 171.77 17.34 a(r.m.s.) 1.64 2.24 1.58 6.00 5.94 0.0070 0.0118 0.0041 0.0128 47.04 7.07 Cl-2 20.73 26.65 17.87 15.08 13.65 4.9561 7.9568 5.7377 0.0081 190.80 15.18 Cl-4 19.97 25.53 15.74 22.24 19.32 4.9546 7.9681 5.7420 0.0130 182.90 20.00 Cl-6 21.01 26.90 18.17 14.26 13.73 4.9715 7.9833 5.7390 0.0023 173.77 19.32 Cl-7 19.59 25.27 17.69 12.63 1 1.25 4.9553 7.9613 5.7405 0.0016 202.65 28.24 Cl-9 22.46 28.72 18.23 16.09 14.73 4.9649 7.9762 5.7395 0.0029 193.19 24.06 Cl-1 1 23.42 29.51 16.46 31.44 28.00 4.9756 7.9829 5.7434 0.0006 166.57 20.27 CI-12 22.83 29.66 18.23 14.84 13.14 4.9664 7.9795 5.7417 0.0008 189.94 25.92 CI-17 19.31 24.8 16.74 15.96 14.52 4.9614 7.9741 5.7397 0.0053 1 50.38 16.3 CI-23 19.43 24.68 15.82 19.40 22. 1 5 4.9691 7.9867 5.7542 0.0082 266.34 24.23 C2-5 22.23 28.48 17.30 18.67 16.91 4.9579 7.9576 5.7400 0.0003 2 1 7.9 1 27.95 C2-6 20.21 26.05 17.75 13.52 13.68 4.9524 7.9619 5.7382 0.0003 272.31 24.79 C2-7 20.09 25.81 1 7.67 14.29 12.57 4.9547 7.9518 5.7.364 0.0034 163.58 15.07 C2-I4 21.57 27.37 1 6.93 20.25 17.17 4.9560 7.9677 5.74.56 0.0009 184.00 22.74 C2- 1 9 20.66 26.48 15.91 18.02 17.88 4.9.547 7.9685 5.7406 0.0019 168.25 18.72 Mean 20.97 26.85 17.18 17.62 16.54 4.9608 7.9697 5.7412 0.0033 194.47 21.65 a(r.m.s.) 1.33 1.67 0.91 4.86 4.58 0.0074 0.0 no 0.0043 0.0038 36.07 4.45 TEXTURE ANALYSIS OF HELIX ASPERSA ASPERSA 161 mind that S samples were flatter than C’s, and consequently should have provided less fluctuating X-ray results. Histo- grams of the F^ fluctuations are presented for S (Fig. 5A) and C (Fig. 5B) samples, respectively. The error bars on these diagrams are standard deviations. One can clearly notice the reduced variability of F^ on C samples. The significantly stronger textures observed for control shells underlined the higher degree of orientation in control samples. Smaller standard deviation of F^ and ODF max in C compared to S indicated a larger textural resemblance among the control specimens than in the ones selected for increased weight. This could be attributed to the larger shells obtained by selection, which implies a faster, less acute, crystallographic growth. The overlap of the results (including standard deviations) indicated that the overall preferred crystallite orientation was only mildly dependent on the selection carried out although Figure 5. A, histogram for S (S2-7 omitted) specimens with an average (Av.) of 18.2 m.r.d.^, including an error bar indicating the standard deviation for the entire class: 6.7 r.m.s. and B, F" histogram for C (Cl-1 1 omitted) specimens with an average of 21.7 m.r.d. in- cluding an error bar indicating the standard deviation for the entire class: 4.5 r.m.s. the degree of this orientation was slightly influenced. To better visualize the influence of selection, one can calculate (Fig. 6) the {020} and {002} pole figures for the extremes as well as for an average sample (C2-14, the closest to the F‘ average value) including their minimum and maximum pole distribution values. This figure shows that the same mean preferred - orientation components were present in the samples, includ- ing the extreme samples: QTA provided reliable results although the maximum densities were subjected to deviations due to surface effects like roughness, flatness, and growth anomalies. The textures stabilized corresponded to crystallite c-axes (revealed by the {002} pole figures) aligned with the macro- scopic normal of the shell, while the b-axes (revealed by the {020} pole figures) of aragonite were mainly aligned at approx. 10° from G. Slight variations between samples included the error created by positioning the samples on the diffractometer. However, the b-axes were always found at this 10° angle from G, certifying our positioning. The full width at half maximum of the c-axes distribution was around 20° in the direction of the b-axes and around 10° perpendicularly. The fact that all samples exhibited pole figures which had the same shape but different maximum ODF values (and F^ values) indicated that the direction of the preferred orientation of the crystallites did not change with selection, but the number of crystallites that oriented within a given width of distribution did. Indeed, the non-oriented part of the irradiated samples was in practice zero for all specimens in both C and S sets (Table 1 ), but the maxima for S were more variable than for control specimens, indicating that selection for larger size induced order-disorder fluctuations of the aragonite layers throughout the population. Adult age, i.e., time to complete growth (until the peristome was reflected), was not significantly different between lines (Dupont-Nivet et al. 2000b). This means that larger size in S snails was achieved only through faster growth, with 60% more weight in S animals in the same growing time. The fact that the preferred orientation of crystallites changed only slightly with selection showed that this was a phenomena which was determined more by the species than by growth conditions. However, the faster growth seemed to have potentially unfavorable effects through order-disorder fluctuations of the aragonite layers. Other experiments showed that neither mortality nor shell shape or shell proportion (ratio between shell and animal weight) differed significantly between both lines (Dupont-Nivet, pers. comm.). Preferred orientations also condition mechanical properties of aggregates, in particular when the constituting crystals possess strong anisotropy of their elastic stiffness constants, as for aragonite. The elastic mechanical behavior of the mineral aragonite fraction can be simulated from the ODF-weighted average of the single crystal stiffness tensor. 162 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2-2009 020 ■C-0.22 020 r- — * o 'iS' o V >; • > ', ; ^ V7i V \ ^ I r''X<\<:>^ ^ • , ■ • '0 ^3e- '■ V'^ a'^ '' W -0.76 >00.-'' -29.78 0.01 A -16.59 ta ■^0.17 B 25.36 tl 0.03 C Figure 6. {020} and {002} pole figures for two extreme and one average samples. A, Sl-8; B, SI -10; and C, C2- 14. Equal area projections, logarithmic distribution density scale. Linder tlie hypothesis of regular grain boundary behaviors, using the geometric mean approach (Ouhenia et al. 2008). In such an approach, we ignore the biocomposite nature of the shell and organic-inorganic interactions, and illustrate how the mineral part of the shell affects elastic behavior of the Helix aspersa aspersn shell. For an aragonite single crystal there are 9 independent values for the c stiffness values (in GLa): c,, = 159.58, c^^ = 86.97, c„ = 85.(')3, c,, = 41.32, Cj^ = 25.64, c^,_ = 42.74, c,, 36.63, c,, = 1 .97, and c^, = 1 5.9 1 . In'the frame of this calculation, axes 1 , 2, and 3 for i and j indices are the M, G, and N directions, respectively. Using the ODE geometric mean, we obtained mean macro.scopic stiffness values of (in GPa): c“J‘J = 1 17, c^> 107, c;; = 86, c“ = 36, cJJ = 34, c“ = 41, c"^^ = 31, C|3 = 14, and c“ = 15, within 6% of the standard deviation, and no significant difference occurred between mean values for the S and C sample sets. Several orientation effects on the macroscopic constants of the shell, com- pared to the single crystal values, kept the C33 constant practically unchanged around 85 GPa. This is because of the strong c-axis orientation with N. How- ever, the texture imposes an average value for c^[ and c^^, intermediate values compared to the single crystal. This causes a homogeneous mechanical re- sponse of the mineral part for com- pression in the (G, M) plane of the shell. All the off-diagonal c,^ coefficients are homogenized, being much less aniso- tropic than in the single crystal, giving rise to moderate anisotropic transverse strains in the shell. Because of this, c^| is 7 times larger than in a single crystal, at only a small expense of c^. The shear coefficients c”^', c^^, and c“ were balanced compared to the single crystal values, and in particular c^^and c^^. This allows the shell to accommodate relatively large shear coefficients along all direc- tions. Hence, from an elastic anisotropic theory point of view, the strong texture exhibited by tbe H. aspersa aspersa shell behaves, at least for the mineral part, in an optimal manner relative to moderate compressive and shear forces. This is caused by balanced weak and strong elastic coefficients that are not optimal in all macroscopic directions of the shell as observed for instance in the marine gastropod Cluironia lampaslampas{L\nnAcus, 1758) (Ouhenia etal. 2008). One of our objectives was to determine if growth selection had important detrimental effects on shell characteristics. Indeed, even if it is not a selected trait, shell strength is a key point in snail farming. Animals are often manipulated (structure changes, sorting, collecting) which creates multiple chances for shell fracture. These broken shells are problematic for snail survival (dehydration) and growth, and also for the commercial value of animals. The results of this study clearly show that .selection did not compromise shell structure, at least in our experimental comlitions. 1 lowever, any side effects TEXTURE ANALYSIS OE HELIX ASPERSA ASPERSA 163 of faster growth should be checked on a regular basis. Moreover, we should check changes in shell thickness and I measured shell strength and correlate them with crystallo- graphic results to test if order-disorder fluctuations of the aragonite layers have unfavorable effects. Looking at a local scale, the crystal organizations on SEM images, the different orientations of crystallites can be made visible. Erom the shell top ( ( G, M) plane) Helix aspersa aspersa has successively alternating lamellae (Eig. 7). The crystals are elongated alternatively along a direction at 62° (arrow) from the vertical direction (G) and a direction of about 53° I from it. This results in a counterclockwise angle between the ' respective elongations of -115° in the two succeeding ! elongation directions. The two main visible directions of crystal elongation showed a striking angular correspondence between the (110) and (-110) crystallographic planes within aragonite (or between the [110]’^ and [-llO]”^ reciprocal crystallographic directions, respectively normal to the (110) and (-110) planes). Calculating the angles between (100) and (-110) planes using the cell-parameters (a = 4.96 A and b = 7.97 A), one finds an angle of -116°. When looking at the {110} pole figure (Fig. 8A), the two directions of crystal elongation in the two Helix aspersa aspersa layers clearly could be identified to the [110]’^ and [-110]’^ crystalline directions (Fig. 8B). However, this does not mean that crystallographic alignment occurred with two components of orientation. Indeed, looking at Fig. 6C, only one, previously described orientation component was present throughout the shell, which corresponded to the {110} four- fold multiplicity of Fig. 8A. We conclude that elongation of the crystals is operated along the {110} plane of stacking directions. From one layer to the next, this biologically driven growth of crystals occurred without loss of crystallographic orientation, but with change in for individuals selected for growth from one of the {110} planes (for instance (110)) to the other {e.g., (-110)). This is a Figure 7. SEM photograph of a fractured Helix aspersa aspersa shell parallel to the (G, M) plane. Magnification is 742 x. The bottom of the image corresponds to the outermost CCL (Comarginal Crossed- Lamellar ) layer while the top of the image is the next inner RCL (Ra- dial Crossed-Lamellar) layer. Arrows illustrate the main crystal di- rections in the two layers. The shell frame is indicated on the right. different growth scheme than observed in the gastropod Cypraea testudinaria (Linnaeus, 1758), in which the crossed lamellae of either the radial or co-marginal layers were obtained by a twinning relationship (Chateigner et al. 1996). In Helix aspersa aspersa, we always found the same single- components of orientation whatever the size of the shell, i.e., whatever the thickness probed, indicating all the layers do keep the crystal orientation. This appears a common pattern in land snails, the same textures having been observed in Helix pomatia (Linnaeus, 1758), Helmintoglypta (Binney, 1897), and Euglandina (Ferussac, 1818) (Chateigner et al. 2000) while all the marine gastropods analyzed to date showed orientation modifications from layer to layer (Chateigner et al. 1996, 2000, Checa and Rodriguez-Navarro 2005). Finally, looking at brownish bands on the surface of Helix aspersa aspersa shells (Figs. 1-2), these typically made an angle around 10° with the growth direction G. Such angular values were retrieved between the <020> crystallographic directions and G ({020} pole figure (Fig. 6C and Fig. 8B)). A close direc- tional relationship may exist between the alignment of the colored bands at the shell surface and the <020> crystallo- graphic directions, using for instance pending bonds from the carbonate groups as they coincide with the observed orien- tations (Fig. 8C). Such bands were recently associated in Helix aspersa aspersa to the presence of un-substituted, methyl terminated chains of 8-13 conjugated double-bond polyenes, either isolated or bound to other molecules, the density of which rendered the color intensity (Hedegaard et al. 2006). Since crystallographic texture in molluscs is associated with inter- crystalline macromolecule interaction, the fact that the colored bands of H. aspersa aspersa were linked to the specific [020] di- rection indicates bound polyenes or pigments in this species. In conclusion, comparison of QTA results of 29 speci- mens of Helix aspersa aspersa, using a statistical analysis, indicated quantitative agreement within standard deviations of -5 m.r.d.“ for the texture index and 40-50 m.r.d. for the maximum value of the orientation distribution function, sug- gesting such standard deviations vary from about 20 to 35%. We observed a difference between control specimens and the ones selected for larger size: growth stimulation affects the preferred orientations. The standard samples have a higher average texture index with less variability (lower standard deviation). These results indicated the degree to which one can see texture variation between individuals, at least for H. aspersa aspersa having a quite regular but curved shell shape. A clear identification of the elongation direction of the individual crystals in the radial and co-marginal crossed lamellar layers indicated <110> crystallographic directions whatever the layer, while the crystals in different layers all had the same orientation. The colored bands at the surface of the shells were linked to the b-axis of the aragonite structure. The elastic behavior of the mineral part of the shell is averaged by 164 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 [110] b=[020] [-110] a=[200] 19.0 0.01 A B C Figure 8. A, { 1 1 0} pole figure of Hexlix aspcrsa aspersa C2- 1 4; B, sche- matic of the corresponding main crystalline directions; and C, one theoretical aragonite unit-cell in the (G, N, M) shell reference frame. the crystallite dispersions to accommodate moderate shear and compression stresses. ACKNOWI.EIKJMENTS The authors would like to thank the I’rench Region Basse-Normandie for its partial funding of the X-ray texture instrument and Helene Rousseliere for the SEM analysis. The two anonymous referees are greatly acknowledged for their constructive comments which contributed to greatly improve the outline of the paper. LITERATURE CITED Aizenberg, J., M. Ilan, S. Weiner, and L. Addadi. 1996. Intracrystalline macromolecules are involved in the morphogenesis of calcitic sponge spicules. Connective Tissue Research 34: 255-261. Bunge, H. J. 1982. Texture Analysis in Materials Science: Mathematical Methods (translated by R R. Morris). Butterworths, London. Chateigner, D. 2004. Combined analysis: Structure-texture-micro- structure-phase-stresses-reflectivity determination by x-ray and neutron scattering. Available at: http://www.ecole.ensicaen. fr/~chateign/texture/combined.pdf 28 April 2009. Chateigner, D. 2005. Reliability criteria in Quantitative Texture Analysis with experimental and simulated orientation distribu- tions. jourttal of Applied Crystallography 38: 603-61 1. Chateigner, D., C. Hedegaard, and H. R. Wenk. 1996. Texture analysis of a gastropod shell: Cypraea testudinaria. In: Z. Liang, L. Zuo, and Y. Chu, eds.. Textures of Materials ICOTOM-1 1: Proceedings of the 1 1th International Conference on Textures of Materials, Xi’an, China, 1996. Academic Publishers. Pp. 1221- 1226. Chateigner, D., C. Hedegaard, and H. R. Wenk. 1999. Quantitative characterisation of mollusc shell textures. In: ). A. Szpunar, ed.. Textures of Materials, Vol. 2. NRC Research Press, Ottawa. Pp. 1495-1500. Chateigner, D., C. Hedegaard, and H. R. Wenk. 2000. Mollusc shell microstructures and crystallographic textures. Jourtial of Struc- tural Geology 22: 1723-1735. Chateigner, D., M. Morales, and E. M. Harper. 2002. QTA of pris- matic calcite layers of some bivalves, a link to trichite ances- trals. Materials Science Torum 408-412: 1687-1692. Checa, A. and A. Rodriguez-Navarro. 2005. Self-organisation of na- cre in the shells of Pterioida (Bivalvia: Mollusca). Biomaterials 26: 1071-1079. Dupont-Nivet, M., ). Mallard, J. C. Bonnet, and J. M. Blanc. 1998. Quantitative genetics of reproductive traits in the edible snail Helix aspersa Muller. Journal of Experimental Zoology 281: 220- 227. Dupont-Nivet, M., |. Mallard, I. C. Bonnet, and M. ). Blanc. 2000a. Direct and correlated responses to individual selection for large adult weight in the edible snail Helix aspersa Miiller. The Jour- nal of Experimental Zoology 287: 80-85. Dupont-Nivet, M., V. Cioste, P. Cioinon, |. C. Bonnet, and M. |. Blanc. 2000b. Rearing density effect on the production perfor- mance of the edible snail Helix aspersa Muller in indoor rear- ing. Annales Zooteclmiques 49: 447-456. Falini, C., S. Albeck, S. Weiner, and L. Addadi. 1996. Control of ara- gonite or calcite polymorphism by mollusk shell macromol- ecules. Science 271: 67-69. I ledegaard, C. aiul 1 1. R. Wenk. 1998. Microstructure and texture pattern ol mollusc shells. Journal of Molluscan Studies 64: 133-1.16. TEXTURE ANALYSIS OE HELIX ASPERSA ASPERSA 165 Hedegaard, C., J. F. Bardeau, and D. Chateigner. 2006. Molluscan shell pigments: An in-situ resonance Raman study. Journal of Molluscan Studies 72; 157-162. Le Bail, A., H. Duroy, and J. L. Fourquet. 1988. Ab-initio structure determination of LiSbWO^ by X-ray powder diffraction. Mate- rial Research Bulletin 23: 447-452. Lutterotti, L., S. Matthies, and H. R. Wenk. 1999. National Research Council of Canada, Ottawa. Available at: http://www.ing.unitn. it/maud/ 28 April 2009. Lutterotti, L., D. Chateigner, S. Ferrari, and J. Ricote. 2004. Texture, residual stress and structural analysis of thin films using a com- bined X-ray analysis. Thin Solid Films 450: 34-41. Matthies, S., G. W. Vinel, and K. Helming. 1987. Standard Distribu- tions in Texture Analysis,Vo\. 1. Akademie-Verlag, Berlin. Ouhenia, S., D. Chateigner, M. Belkhir, and E. Guilmeau. 2008. Mi- crostructure and crystallographic texture of Charonia lampas larnpas shell. Journal of Structural Biology 163: 175-184. Pilarti, T., F. DeMartin, and C. M. Gramaccioli. 1998. Lattice- dynamical estimation of atomic displacement parameters in carbonates: Calcite and aragonite Ca CO^, dolomite Ca Mg (COj)^, and magnesite Mg CO^. Acta Crystallographica (B) 54: 515-523. Ricote, L and D. Chateigner. 2004. Quantitative microstructural and texture characterisation by X-ray diffraction of polycrystalline ferroelectric thin films. Journal of Applied Crystallography 37; 91-95. Rietveld, H. M. 1969. A profile refinement method for nuclear and magnetic structures. Journal of Applied Crystallography 2: 65- 71. Rousseau, M., E. Lopez, P. Stempfle, M. Brendle, L. Eranke, A. Guette, R. Naslain, and X. Bourrat. 2005. Multiscale structure of sheet nacre. Biomaterials 26: 6254-6262. Zolotoyabko, E. and L P. Quintana. 2002. Non-destructive micro- structural analysis with depth resolution: Application to sea- shells. Journal of Applied Crystallography 35: 594-599. Submitted: 8 April 2008; accepted: 26 March 2009; final revisions received: 29 April 2009 N . i. Ainer. Maine. Bull. 27: 167-172 (2009) Mollusc survey of the lower Bruneau River, Owyhee County, Idaho, U.S.A. Steven J. Lysne* and William H. Clark^’^ Orniii ]. Smith Museum of Natural History, The College of Idaho, 21 12 Cleveland Boulevard, Caldwell, Idaho 83605, U.S.A. ^ Idaho Power Company, 1221 West Idaho Street, Boise, Idaho 83702, U.S.A Corresponding author: stevelysne@cwidaho.ee Abstract: The Bruneau River in southwestern Idaho is a largely free-flowing desert stream characterized by a snowmelt-driven hydrograph, flash floods, and geothermal inputs. We surveyed the lower Bruneau River from its confluence with the Snake River upstream to Hot Creek, a distance of approx. 41 river kilometers, between 1997 and 2008. We supplemented our work with a review of collections held at the Orma J. Smith Museum of Natural History, College of Idaho and with collections available online from national and international museums. The known mollusc fauna consists of 18 species ( 1 1 gastropods and 7 bivalves) from eight families. Molluscan richness is greatest in free-flowing stream reaches and is dominated by hydrobiid and unionid taxa. Key words: gastropods, bivalves, invasive species The freshwater molluscan fauna of Idaho has interested malacologists for many years (Hendersen 1924) yet its diverse taxa remain poorly understood (Frest and Johannes 2000, Frest et al. 2001). Further, the taxonomic status of some species in Idaho is under revision (Taylor 2003, Hershler and Liu 2004, Rogers and Wethington 2007, Wethington and Lydeard 2007) with considerable work yet to be completed. Of the 117 putative species in Idaho (Frest and Johannes 2000), few published accounts exist on their occurrence in specific stream segments. Similarly, the ecology of most gastropods, including species in Idaho, has received very little attention (Brown et al. 2008) resulting in a vague understanding of interactions between species and their environment. Conservation of freshwater molluscs requires: (1) an understanding of the relationships between species in a geographic area (f.c., stabilized taxonomy), (2) an understanding of the distribution of species [i.e., species inventory), and (3) an understanding of the ecology of species within the context of long-term species persistence (Brown and Johnson 2004, Stewart and Dillon 2004, Lysne et al. 2008). In reality, natural resource managers must frequently prioritize conservation goals with incomplete biological knowledge (Regan et al. 2003, 2005). Within this context, we join other efforts (Dillon 2008) to document molluscan biodiversity and to provide information for natural resource managers to set conservation priorities. MATERIALS AND METHODS Study area The Bruneau River, in southwestern Idaho, is a largely free-flowing, high-desert stream originating in the mountains of northern Nevada and stretching approx. 185 km north to join the Snake River (Fig. 1). Only two known structures divert Bruneau River water between the Bruneau Arm and its headwaters: Buckaroo Diversion at Bruneau River kilometer (RKM) 34 and Harris Diversion at RKM 34.4. Mean in- stantaneous flow of the Bruneau River, measured at RKM 35.4 between 1987 and 2007 was 8.47 cubic meters second ' (ems) and ranged from 2.86 ems in December to 30.58 ems in May (USGS 2008). Mean monthly water temperatures ranged from 14 °C in May to 21 °C in September, with a high of 26 °C in August (USGS 2008). The Bruneau River has few perennial tributaries but has numerous ephemeral creeks that can significantly increase flow during intense summer storms. A defining characteristic of the Bruneau River system is the geothermal aquifer that underlies much of southern Idaho (Berenbrock 1993) and contributes a base flow to the river year round. Water temperature of geothermal spring flows range between 11 °G and 40 °G (Myler et al. 2007), and an abrupt thermocline has been observed in the main channel as a result of geothermal, hyporheic additions to the river. Field sampling We surveyed sections of the Bruneau River from the Bruneau Arm of C. J. Strike Reservoir {i.e., the confluence of the Bruneau and Snake Rivers; 42.92375°N, 1 15.90353°W) to Hot Greek (42.76226°N, 1 15.73084°W), a distance of approx. 41 km (Fig. 1). We used a Venturi dredge apparatus and SGUBA (Stephenson et al. 2004) to collect 432 samples from the Bruneau Arm of C. J. Strike Reservoir. A 0.25- m^ quadrat was placed on the reservoir bottom and the substrate suctioned to a depth of approx. 5 cm. Samples were sorted by hand to identify, and return to the river, endangered gastropod species, if present. Remaining sample material was 167 168 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Figure 1. Figure of the study area in southwestern Idaho. Samples were observed or collected in the Bruneau Arm as well as the Bruneau River between Hot Creek and Highway 31. preserved in 95% ETOH and shipped to EcoAnalysts Inc., Moscow, Idaho, for identification to the lowest possible taxonomic level. We estimated densities of taxa for species importance curves based on all collections between 1997 and 2008 where density information was available. In the free-flowing reaches of the Bruneau River, we conducted visual inspections (via a benthic viewer) of all available habitat types {e.g., springs, river, pools, riffles, submerged and emergent aquatic vegetation, fines, cobbles, and bedrock) below Hot Creek. We spent more than 100 person-hours surveying the stream and associated habitats for molluscs. We used hand collections only in the river below Hot Creek and made no attempt to quantify densities or total numbers of individuals of any taxa. Relative abundance of taxa for species importance curves was estimated qualitatively by professional judgment. We collected and preserved individuals in 95% ETOH only for purposes of voucher specimens. These individuals were identified to the lowest possible taxonomic level. Collections in the Bruneau Arm occurred throughout the year but our observations in the free-flowing Bruneau River were conducted, for safety reasons, between July and March when flows were <5.7 cms. Available information including approximate sample locations, physico- chemical data, and lists of non- molluscan taxa are available from the authors. Museum collections We supplemented our work with a review of collections held at the Orma J. Smith Museum of Natural History, The College of Idaho (http://www.collegeofidaho.edu) between August 2007 and September 2008. The Smith Museum houses the largest collection of freshwater molluscs in Idaho and has staff dedicated to the curation and study of freshwater molluscs. We conducted a query of the Smith Museum’s curation database resulting in approx. 1,100 lots from the Bruneau River. We also inspected approx. 150 Lin-cataloged lots of freshwater molluscs, looking for collections from the Bruneau River. In addition, we surveyed online databases and/or requested searches from The University of Michigan’s Museum of Zoology, the California Academy of Sciences, and I'he Canadian Museum of Nature. From these searches, we obtained 418 lots containing molluscs fix>m the Bruneau River. Detailed intormation on all collections is available from the authors. Voucher specimens from our work have been deposited at the Orma |. Smith Museum of Natural History. Identifications followed Burch (1989); nomenclature follows I'urgeon ct al. (1998) for most taxa and Mackie (2007) lor the corbiciilid and sphaeriid clams. MOLLUSCS OF THE LOWER BRUNEAU RIVER 169 RESULTS We documented eleven species of gastropods from five families, including the endangered, thermophillic springsnail Pyrgulopsis bruneauensis (Hershler, 1990) (Myler et al. 2007) and two non-native gastropods: Potamopyrgus antipodariim (Gray, 1853) and Radix auricularia (Linnaeus, 1758). We also documented seven species of bivalve molluscs from three families, including the invasive Corbicula fluminea (Muller, 1774). Table 1 lists species collected or observed, the location of collection, and information on the conservation status of each taxon globally and in Idaho (NatureServe 2008). Thirteen species of molluscs were collected or observed in the free-flowing reaches of the Bruneau River below Hot Creek. Of these molluscs, one hydrobiid gastropod, Fluminicola fuscus (Haldeman, 1847), and one unionid bivalve, Gonidea angulata (Lea, 1838), dominated numerically. Nine species of molluscs were collected from the impounded Bruneau Arm of C. J. Strike Reservoir. Of these, Gyraulus parvus (Say, 1817) and Vorticifex effusa (Lea, 1856) dominated numerically. Four species were found in both the free-flowing sections of the river and the reservoir and all are considered habitat generalists. The three non-native species (C. fluminea, P. antipodariim, and R. auricularia) occurred in the river and reservoir reaches. We observed a change in species importance from reservoir to river habitats (Figs. 2A-B). Based on our collections, the Bruneau Arm molluscan community is dominated by Vorticifex effusa and Gyraulus parvus. Estimated densities of both species exceeded 1,500 individuals/m^ whereas estimated densities for all other taxa fell below 400 individuals/m^. By contrast, in the river below Hot Creek the molluscan community is dominated by Fluminicola fuscus and Gonidea angulata. Three additional species, Gyraulus parvus, Physa gyrina, and Pyrgulopsis bruneauensis, represented an inter- mediate relative abundance with the remaining taxa best described as uncommon. DISCUSSION The molluscan fauna of the Bruneau River below Hot Creek is rich (containing about 15% of the known Idaho fauna) relative to other tributaries of the Snake River in southern Idaho. The reasons for this diversity are not entirely known, but the relatively long length of the stream at the landscape scale, the diversity of habitats at multiple spatial scales, water chemistry, geothermal influences, and the relatively unimpaired conditions of the Bruneau River corridor are probably important influences. Table 1. Molluscs present in the lower Bruneau River (below Hot Creek), Idaho. Data show presence related to general habitat as well as con- servation status. Voucher specimens of each taxon are located in the Orma J. Smith Museum of Natural History, The College of Idaho. Habitat descriptors: RIV, river; RES, reservoir. Rankings are from NatureServe (2008): G1 /SI, critically imperiled; G2/S2, imperiled; G3/S3, vulnerable; G4/S4, apparently secure; G5/S5, secure; SNR, state not ranked; EXO, exotic/introduced; NA, no ranking available. Family Genus Species Authority Habitat G Rank S Rank Gastropods Ancylidae Ferrissia rivularis Say, 1817 RIV, RES G5 SNR Hydrobiidae Fluminicola fuscus Haldeman, 1847 RIV G2 SI Hydrobiidae Potamopyrgus antipodariim Gray, 1853 RIV, RES G5 EXO Hydrobiidae Pyrgulopsis bruneauensis Hershler, 1990 RIV G1 SI Hydrobiidae Pyrgulopsis robusta Walker, 1908 RES G4 SI Lymnaeidae Fossaria sp. Westerlund, 1885 RES NA NA Lymnaeidae Radix auricularia Linnaeus, 1758 RIV G5 EXO Physidae Physa gyrina Say, 1821 RIV, RES G5 SNR Planorbidae Gyraulus parvus Say, 1817 RIV, RES G5 SNR Planorbidae Planorbella subcrenata Carpenter, 1857 RIV G5 SNR Planorbidae Vorticifex effusa Lea, 1856 RES G3 SNR Bivalves Corbiculidae Corbicula fluminea Muller, 1774 RES G5 EXO Sphaeriidae Pisidium casertanum Poli, 1791 RIV G5 SNR Sphaeriidae Pisidium variabile Prime, 1852 RIV NA SNR Sphaeriidae Pisidium compressum Prime, 1852 RES G5 SNR Sphaeriidae Sphaerium simile Say, 1817 RIV G5 SNR Unionidae Anodonta californiensis Lea, 1852 RIV G5 SNR Unionidae Gonidea angulata Lea, 1838 RIV G3 SNR 170 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 A FlumGon Gyr PhyPyrgAno Pot Stag Plan Per Fos RadP.caf'.varSph B Figure 2. Importance curves for molluscs in the Bruneau River from Hot Creek to the Bruneau Arm of C. J. Strike Reservoir (A), and in the Bruneau Arm of C. ]. Strike Reservoir (B). Abbreviations are: Flumitiicolu fuscus, Gonidca atigiilata, Gyraulus parvus, Physn gyritin, Pyrgulopsis bnineaiiensis (A),Anodouta californiensis, Potamopyrgus antipodanim, Stagm'co/n hitikleyi, Planorbella subcrenatum, Perrissia rividaris, Possaria sp., Radix auricularia, Pisidiiim casertanum, Pisidium variabile, Sphaerium simile (A), Vorticifex effusa, Pyrgulopsis robusta (B), Pisidium compressum, Sphaerium sp. (B), and Coricula jlumitiea. The Bruneau River is a long (ca. 185 km) desert stream. It is generally larger, in terms of volume, than most tributaries in southern Idaho hut is considerably smaller than the Boise, Layette, or Salmon Rivers further to the north. 'I'he underlying basalt geology (Ross and Savage 1967) permits the Bruneau to run relatively clean without the excessive sedimentation that is observed in many desert streams. Further, much of the Bruneau River runs through remote canyon-lands with steeply incised walls that limit grazing and other activities that tend to result in increased sedimentation (Allan 1995, Belsky et al. 1999). Similarly, the Bruneau River above our study area lacks irrigation diversions, intensive agriculture, or timber harvest practices that are typical of many landscapes surrounding Idaho streams to the east and north. The result is a desert stream system unique in Idaho with a natural flow regime and a relatively undisturbed stream corridor. The Bruneau River has a steep gradient as it leaves the Owyhee Uplands which begin to attenuate below Hot Creek. As the river approaches the town of Bruneau, Idaho, the topography flattens out and the river begins to meander, which reduces stream velocity and sediment load and alters benthic habitat. As the river continues toward the Bruneau Arm of C. J. Strike Reservoir, it bisects a large wetland created by the impoundment of the Snake and Bruneau Rivers. The Bruneau Arm is essentially lentic and different ecologically from the river upstream with a very different species assemblage. These varied landscape scale habitat characteristics may explain the high molluscan diversity in the Bruneau River (Newton et al. 2008). Perhaps more importantly, the Bruneau River below Hot Creek has increased habitat heterogeneity owing to the geothermal aquifer that underlies much of the Owyhee Uplands in southern Idaho. Water temperature is perhaps one of the most important determinants of where species are presently found (Allan 1995), it has important implications with regard to a species biology (Brown et al. 1998, Lysne and Koetsier 2006a), and can be used as a predictor of species presence (Malcom and Radke 2005). Hundreds of geothermal springs enter the Bruneau along its length (Myler etal. 2007), both above and below the water’s surface, creating a patchwork of different habitats, thermal characteristics, and floral and faunal assemblages. The geothermal aquifer also moderates cold winter temperatures and provides thermal refugia for aquatic invertebrates eliminating the need for hibernation- like, overwintering behaviors observed in other Idaho molluscs (Lysne and Koetsier 2006b). for example, water temperature in the Bruneau River measured at 2/3 depth in February of 2008 was 4.8 °C, below the known thermal range of Pyrgulopsis bruneauensis (Mladenka and Minshall 2001). However, at the boundary layer, water temperature was 12.7 °C, within the thermal tolerance of the species, and the river supports P. bruneauensis activity year-round given the appropriate substrates (Myler el al. 2007). Other species may respond similarly to the unusual thermal regime: Richards (2004) found populations of I'ayloreoucha serpenticola to fluctuate seasonally but reproduction occurred in winter with the thermally constant water temperatures. Similarly, Mladenka and Minshall (2001) found that Pyrgulopsis bruneauensis reproduced year-round in super-heated springs. Thus, the varied geothermal habitat characteristics, at a smaller spatial scale than the landscape, may also contribute to the high molluscan diversity of the Bruneau River. MOLLUSCS OF THE LOWER BRUNEAU RIVER 171 Downstream of Hot Creek the malacofauna is dominated numerically by two species: Fhiminicola fuscus and Gonidea angulata. Conversely, the Bruneau Arm is dominated by Gyranlus parvus and Vorticifex effusa and no unionid bivalve molluscs were present (Stephenson et al. 2004). The non- native Radix auricularia was found only in the free-flowing river and the non-native Potamopyrgus antipodarum was found in both lotic and lentic habitats. The relative importance of molluscs in the river versus the reservoir follows habitat requirements of unionid mussels (Vaughn and Taylor 1999, Dillon 2000) and the comparative ecology of pulmonate and “prosobranch” snails (Brown et al. 1998, Brown and lohnson 2004). Pulmonate snails such as Vorticifex effusa and Gyraulus parvus are considered more tolerant of lentic systems with relatively eutrophic water quality whereas “prosobranchs” such as Fhiminicola fuscus and unionids such as Gonidea angulata are more characteristic of lotic systems and relatively mesotrophic water quality. Of note in the Bruneau River below Hot Creek are three species of intermediate relative abundance: Gyraulus parvus, Physa gyrina, and Pyrgulopsis bruneauensis. Physa gyrina occupies various habitat types and and is common throughout the area we studied. By contrast, based on our direct observations, both G. parvus and Pyrgulopsis bruneauensis are patchily distributed depending on the presence of suitable habitats. The presence of Potamopyrgus antipodarum is of concern. This species is now the most abundant mollusc in parts of the middle Snake River (Richards et al. 2001) and can reach extremely high densities in some locations (Richards et al. 2004). The invasion of Bruneau River may be recent. Clark (1979) sampled two sites within the present study area in 1975, approx. 10 years prior to the invasion of the Snake River basin by this species, and failed to detect P. antipodarum. Similarly, surveys between Hot Creek and the Bruneau Arm completed between 1990 and 2005 also failed to detect and/or report the presence of P. antipodarum (Myler et al. 2007). Today the species is abundant in C. J. Strike Reservoir and appears to be moving upstream. The relatively recent establishment of P. antipodarum thus provides an opportunity to monitor its invasion and potential community level effects. ACKNOWLEDGMENTS We thank M. Stephenson and B. Bean, Idaho Power Company, for collections on the lower Bruneau River (the Bruneau Arm). The Orma J. Smith Museum ol Natural History provided much information on historic mollusc occurrences in the Bruneau River and supplied us with curation materials. Information from collections was supplied by J. M. Gagnon at the Canadian Museum of Nature, D. M. O E oighil at the University of Michigan Museum of Zoology, and R. Van Syoc at the California Academy of Sciences. J. Burch, R. Hershler, and J. Keebaugh verified species determinations. Special thanks go to J. Keebaugh at the Smith Museum, M. Radko at the Idaho Power Company, and D. Ames and J. Whelihan at the College of Western Idaho. LITERATURE CITED Allan, J. D. 1995. Stream Ecology: Structure and Function of Running Waters. Chapman and Hall, London. Belsky, A. J., A. Matzke, and S. Uselman. 1999. Survey of livestock influences on stream and riparian ecosystems in the west- ern United States. Journal of Soil and Water Conservation 54: 419-431. Berenbrock, C. 1993. Effects of well discharge on hydraulic heads in and spring discharges from the geothermal aquifer system in the Bruneau area, Owyhee County, southwestern Idaho. U.S. Geological Survey Water Resources Investigations Report 93-4001. Brown, K. M. and P. D. lohnson. 2004. Comparative conservation ecology of pleurocerid and pulmonate gastropods of the United States. American Malacological Bulletin 19: 57-62. Brown, K. M., J. E. Alexander, and 1. H. Thorp. 1998. Differences in the ecology and distribution of lotic pulmonate and prosobranch gastropods. American Malacological Bulletin 14; 91-101. Brown, K. M., B. Lang, and K. E. Perez. 2008. The conservation ecology of North American pleurocerid and hydrobiid gas- tropods. Journal of the North American Benthological Society Tl\ 484-495. Burch, 1. B. 1989. Freshwater Snails of North America. Malacological Publications, Hamburg, Michigan. Clark, W. H. 1979. Water quality status report Bruneau River, Owyhee County, Idaho 1975. Water Quality Series 36, Idaho Department of Health and Welfare, Division of Environment, Boise. Dillon, R. T. 2000. The Ecology of Freshwater Molluscs. Cambridge University Press, Cambridge, UK. Dillon, R. T. 2008. Ereshwater gastropods of North America proj- ect. Available at: http;//www.cofc.edu/~fwgna/fwgnahome.htm 1 May 2009. Frest, T. J. and E. J. Johannes. 2000. An annotated checklist of Idaho land and freshwater mollusks. Journal of the Idaho Academy of Science 36; 1-51. Frest, T. J., E. J. Johannes, W. H. Clark, G. Stephens, and M. G. Plew. 2001. A bibliography of Idaho freshwater and terrestrial mol- lusks. Journal of the Idaho Academy of Science 37: 9-120. Hendersen, J. 1924. Mollusca of Golorado, Utah, Montana, Idaho, and Wyoming. University of Colorado Studies 13; 65-223. Hershler, R. and H.-P. Liu. 2004. Taxonomic reappraisal of species assigned to the North American freshwater gastropod subgenus Natricola (Rissooidea: Hydrobiidae). The Veligcr 47: 66-81. Lysne, S. and P. Koetsier. 2006a. Growth rate and thermal tolerance of two endangered Snake River snails. Western North American Naturalist 66: 230-238. Lysne, S. and P. Koetsier. 2006b. The life history of the Utah (desert) valvata, Valvata utahensis, in the Snake River, Idaho. Journal of Freshwater Ecology 21: 285-291. 172 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Lysne, S. J„ K. E. Perez, K. M. Brown, R. L. Minton, and J. D. Sides. 2008. A review of freshwater gastropod conservation: Challenges and opportunities. Journal of the North American Benthological Society 27: 463-470. Mackie, G. L. 2007. Biology of freshwater corbiculid and sphaereiid clams of North America. Ohio Biological Survey Bulletin New Series 15: 1-436. Malcom, J. and W. R. Radke. 2005. Habitat associations of the San Bernardino springsnail, Pyrgulopsis bernardina (Hydrobiidae). Journal of Freshwater Ecology 20: 1\-71 . Mladenka, G. C. and G. W. Minshall. 2001. Variation in the life history and abundance of three populations of Bruneau hot springsnails (Pyrgulopsis bruneauensis) . Western North Ameri- can Naturalist 61: 204-212. Myler, C. D., G. C. Mladenka, and G. W. Minshall. 2007. Trend analysis shows decline of an endangered thermophillic spring- snail (Pyrgulopsis bruneauensis) in southwestern Idaho. West- ern North American Naturalist 67: 199-205. NatureServe. 2008. NatureServe Explorer: An online encyclopedia of life. Version 7.0. NatureServe, Arlington, Virginia. Available at http://www.natureserve.org/explorer. 1 December 2008. Newton, T. J., D. A. Woolnough, and D. L. Strayer. 2008. Using landscape ecology to understand and manage freshwater mus- sel populations. Journal of the North American Benthological Society 27: 424-439. Regan, H. M., H. R. Ak^akaya, S. Eerson, K. V. Root, S. Carroll, and L. R. Ginzburg. 2003. Treatments of uncertainty and variability in ecological risk assessment of single-species populations. Hu- man and Ecological Risk Assessment 9: 889-906. Regan, H. M., Y. Ben-Haim, B. Langford, W. G. Wilson, P. Lund- berg, S. J. Andelman, and M. A. Burgman. 2005. Robust deci- sion-making under severe uncertainty for conservation man- agement. Ecological Applications 15: 1471-1477. Richards, D. G. 2004. Competition between the threatened Bliss Rapids snail, Taylorconcha serpenticola (Hershler et al.j, and the inva- sive, aquatic snail Potamopyrgus antipodarum (Gray). Ph.D. Dissertation, Montana State University, Bozeman, Montana. Richards, D. C., L. D. Cazier, and G. T. Lester. 2001. Spatial distribu- tion of three snail species, including the invader Potamopyrgus antipodarum, in a freshwater spring. Western North American Naturalist 6\: 375-380. Richards, D. C., P. O’Connell, and D. Cazier-Shinn. 2004. Simple control method to limit the spread of the New Zealand mud- snail Potamopyrgus antipodarum. North Anterican Journal oj Fisheries Management 24: I 14-1 17. Rogers, D. C. and A. R. Wethington. 2007. Physa natricina Tliylor 1988, junior synonym of Physa acuta Draparnaud, 1805 (Pul- monata: Physidae). Zootuxu 1662:45-51. Ross, S. 1 1. and C. N. Savage. 1967. Idaho Earth Sciences. Idaho Bureau of Mines and Geology, Earth Science Series 1 , Moscow, Idaho. Stephenson, M. A., B. M. Bean, A. |. foster, and W. 11. Clark. 2004. Sfiakc River Aquatic Macroinverlebrate and ESA Snail Survey, 'fcchnical report for U.S. fish and Wildlife Service, Boi.se, Idaho. Stewart, T. W. and R. T. Dillon. 2004. Species composition and geo- graphic distribution of Virginia’s freshwater gastropotl fauna: A review using historical records. American Malacological Bul- letin 19: 79-91. Taylor, D. W. 2003. Introduction to the Physidae (Gastropoda: Hygrophila) biogeography, classification, morphology. Revista de Biologta Tropical International Journal of Tropical Biology and Conservation 51: 1-299. Turgeon, D. D., J. F. Quinn, In, A. E. Bogan, E. V. Coan, F. G. Hochberg, W. G. Lyons, P. M. Mikkelsen, R. ]. Neves, C. F. E. Roper, G. Rosenberg, B. Roth, A. Scheltema, F. G. Thompson, M. Vecchione, and J. D. Williams. 1998. Common and Scientific Names of Aquatic Invertebrates from the United States and Canada: Mollusks, 2"‘* Edition. American Fisheries Society, Special Publication 26, Bethesda, Maryland. U. S. Geological Survey. 2008. National streamflow data. Available at: http://waterdata.usgs.gov/nwis/nwisman/?site_no = 13168500&agency_cd=USGS 1 December 2008. Vaughn, C. G. and C. M. Taylor. 1999. Impoundments and the de- cline of freshwater mussels: A case study of an extinction gradi- ent. Conservation Biology 13: 912-920. Wethington, A. R. and C. Lydeard. 2007. A molecular phylogeny of Physidae (Gastropoda: Basommatophora) based on mitochon- drial DNA sequences. Journal of Molluscan Studies 73: 241-257. Submitted: 21 October 2008; accepted: 7 April 2009; final revisions received: 1 May 2009 Amer. Make. Bull. 27: 173-181 (2009) The shell features of Cornu aspersum (synonym Helix aspersa) and Helix pomatia: Characteristics and comparison Maciej Ligaszewski\ Krzysztof Sur6wka^ and Julia Stekla^ National Research Institute of Animal Production-NRI, Department of Technology, Ecology and Economy of Animal Production, 1 Krakowska Street, 32-083 Balice, Poland ^Agricultural University of Krakow, Department of Refrigeration and Food Concentrates, 122 Balicka Street, 10-149 Krakow, Poland ^ Experimental station of the National Research Institute of Animal Production at Grodziec Sl^ski, 43-386 Swi^toszowka, Poland Corresponding author: mligasze@izoo.krakow.pl Abstract: We examine the morphometric, chemical, and physical properties of adult shells from breeding populations of Cornu aspersum maxima (Taylor, 1883), Cornu aspersum aspersum (Muller, 1774), and Helix pomatia (Linnaeus, 1758). The higher thermal requirements of the African subspecies C. aspersum maxima were confirmed by the fact that normal shell maturation, indicated by a decreasing calcium content as the snail ages, was related to an increased mean air temperature of over 22.9 °C during the breeding season. In contrast, normal shell maturation of the European subspecies C. aspersum aspersum occurred with a temperature in the range of 20.6-23.6 °C. Based on the results of texturometric analysis, shell puncture force increased with an increase in temperature during breeding. In contrast, shell puncture force decreased and collapsing force increased with increasing relative humidity. The mechanical strength of C. aspersum and H. pomatia shells was related to their chemical composition and the level of their structural maturity. Shells containing a higher percentage of calcium were characterized by lower mechanical strength than those containing a lower amount. Key words: shell features, measurement methods, shell maturation The physical, chemical, and morphometric features of snail shells are closely related to their internal structure. The internal structure of the shell and the mechanisms of its shaping differ in various mollusc species (Saleuddin and Hare 1970, Weiner and Traub 1980, Wilbur and Saleuddin 1983, Bowen and Hieng Tang 1996, Chateigner et al. 1996, 2000, Hedegaard and Wenk 1998, Kaplan 1998, Dauphin and Denis 2000). Shell formation relies on the production of an organic “matrix” on the outer surface of the mantle. The matrix is composed of specific glycoproteins and amino acids, creating an orderly environment for the crystallization of calcium carbonate, which is the main structural material of a shell. Calcium carbonate is synthesized in the form of a biomineral characteristic of molluscs, aragonite, with a higher specific gravity and different crystalline structure compared to calcite. In a mature shell, properly organized crystals of aragonite form several layers of microstructure specific to individual species of molluscs. There are insufficient data on the shaping of selected physical, chemical, and morphometric features of shell structure which are of primary importance from the point of view of breeding the edible species of Helicidae. Information exists on the effect of microclimate and genotype on color diversification and the pattern ot the shell in Helicidae (Albuquerque de Matos 1984, Lecompte et al. 1998). Moreover, methods for measuring the mechanical strength of the shell have been described for species with shell symmetry and a habitat different from those for Helicidae (Kent 1981, Garden 1998). This study provides data regarding the shaping of selected physical, chemical, and morphometric features of shells in bred populations of Cornu aspersum and Helix pomatia. Biometric and physical-chemical analyses of mature shells from C. aspersum maxima and C. aspersum aspersum bred in variable microclimates were carried out. Shells from mature H. pomatia were used as a reference. MATERIALS AND METHODS Breeding populations of Helicidae The following populations in an experimental snail farm in the National Research Institute of Animal Production - NRI, near Krakow (Poland) were included in this study: (1) “Balice” and “Albino” populations of Cornu aspersum maxima; (2) “Balice” and “French” populations of Cornu aspersum aspersum; and (3) the natural “Balice” population of Roman snail (Helix pomatia) occupying a park surrounding the Radziwill palace in Balice, near Krakow. Shells of mature individuals were collected. Cornu aspersum specimens were aged O-h (first season) and l-h (second season) years and H. pomatia specimens were aged 2+ to 3-t- years. Snails were collected over two consecutive seasons in 2002 and 2003 from microclimatically variable breeding sites, such as field 173 174 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 enclosures, greenhouse enclosures, plastic basins, and a natural park. Snail breeding for research Cornu aspersum and Helix poniatia were bred in non- heated ground enclosures located either in a greenhouse or in a field from May to the end of September. Agricultural lime was applied annually at a rate of 0.5 kg/m^ and the field sown with a mixture ot pasture plants. The spring hatching density was 300 individuals per 1 m^. The animals were provided with wooden pallets onto which feed was sprinkled. Animals could find shelter under the feeders. Breeding enclosures were equipped with a water spray. Throughout the entire study, snails were fed the same standard dry feed. In autumn, 7-month old C. aspersum specimens were somatically and commercially mature, in contrast to Roman snails that maturated 1 -2 years later. Cornu aspersum and Helix pomatia shell sampling Sixty mature snails were randomly selected from each population. Samples of mature Cornu aspersum (age 0-h) were collected in autumn when the snails reached physiological and commercial maturity, on the basis of visual inspection of the shell. Shells were prepared and dried at room temperature for morphometric measurements and further physical and chemical analyses. Mechanically damaged shells and shells of irregular structure were discarded prior to the tests. Samples from reproductive adults (age 1+) of Cornu aspersum were collected in Lebruary, after the first hibernation, and in May, after reproduction in breeding enclosures. Damaged or irregular shells were discarded. Cornu aspersum aged O-H and Helix pomatia from the natural populations were collected within a 750-m^zone, at a distance of 100 m from the breeding farm. Microclimatic and breeding condition Relative air humidity (%) was measured twice daily in June, July, and August in all the breeding areas of the snail farm at the NRl in Balice and in the park surrounding the Radziwill palace. Measurements were made in the morningand afternoon, using a portable electronic thermo-hygrometer, and a mean value for the whole sea.son was calculated. The air temperature was also recorded. Methods of measuring shell features used in the study The following parameters were examined: seven morpho- metric parameters and indices; two physical parameters connected with shell mechanical strength; and five chemical parameters and indices related to the calcium and phosphorus content in the shell. 'I’he hibernating snail was boiled for 30 seconds in hot water, the carcass was removed from the shell, and the shell was dried at room temperature. Morphometric parameters were measured with an electronic scale and electronic micrometer .screw according to the method adopted by the Nature Protection Laboratory of the Polish Academy of Science (Pigs. lA-C). The method to measure shell thickness involved crushing it and taking measurements ol the thickness of ten fragments chosen at random to calculate average thickness. The shape index, or quotient of shell width and height, was used to classify Cornu aspersum (Chevallier 1977). The solidity index allows comparison of the solidity of shells of various sizes and weights (Cooke 1973, Ireland 1991 ). Another index designated “the weight ratio of calcium in the solidity index” allows comparison of the mineral fractions ratio in shell on solidity of shells of various sizes and weights too. Puncture strength is related to the hardness of the most resistant internal layer of a shell, while collapse strength is related to external damage to a shell. A computerized TA.XT2 Figure I. Mcllioil ol mcasiiroiiieiit of (A) height, (B) width, and (C) diameter of shell. PULMONATE SHELL CHARACTERISTICS EROM DIEEERENT POPULATIONS 175 Texture Analyser with a 25-kg load cell was used for measuring these parameters. The XTRA Dimension (ver. 3.7) computer program by Stable Micro Systems, Haselemere, Surrey, UK was used for data collection. The shells were punctured at a rate of 0.1 mm s ‘ by a needle plunger (SMS-P/2N) from the inside at the point where the widest coil was at its most convex (Eig. 2). The puncture force was defined as the force corresponding to the highest, usually first maximal peak on the force-deformation curve. In measuring collapse strength, pressure was exerted with a flat probe (SMS-P/4) 4 mm in diameter moving at a rate of 2 mm s‘‘ (Eig. 3). Shells were crushed from the outside at the middle part of the last coil. Because of the damage occurring, it was possible to make only one measurement of each shell, and it was impossible to use the same shell for both puncture and collapse strength. The values for these parameters were measured separately in two sub-samples, and the relations between the two forces were calculated using intra-group analysis of a single-factor regression. A list of measured features, corresponding methods, and terms of measurement is given (Table 1). Statistical analysis of the results Statistical relationships between pairs of measured shell features were calculated according to a single-factor regression analysis. The same method was used to analyze relationships between the features of the shells, relative humidity, and air temperature. A single-factor analysis of variance based on the least significant differences (LSD) test was performed in order Figure 2. Method of measurement of puncture force of shell. Figure 3. Method of measurement of collapse force of shell. to compare shell feature values between individual breeding populations in two subspecies of Cornu aspersum. RESULTS Comparison of Cornu aspersum and Helix pomatia No statistically significant differences in shell weight, thickness, and diameter were found for age groups 0-t- and l-i- of the two breeding populations of Cornu aspersum maxima and the two populations of Cornu aspersum aspersum (Table 2), indicating the uniformity of breeding conditions for all experimental groups. Statistically significant differences (P < 0.05 or P < 0.01) between populations of both subspecies were found for shell shape, measured using shape index value; mechanical strength, measured using collapsing force; and, in snails aged 1+, the calcium content in the shell solidity index. Eor C. aspersum maxima, collapsing force increased with age, whereas for C. aspersum aspersum it decreased. Cornu aspersum maxima shells, compared to C. aspersum aspersum shells, were always characterized by a higher weight, solidity index, diameter, thickness, and mechanical strength for puncture and collapse. Despite the close relationship between both subspecies and identical experimental conditions, their shells differed significantly in chemical composition. Cornu aspersum maxima shells, for corresponding age groups, contained less calcium and phosphorus and more raw ash than C. aspersum aspersum shells. The shells of Helix pomatia. 176 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Table 1. List of measured features, corresponding methods, and terms of measurement. Parameter Unit Shell weight (g) Shell diameter (mm) Shell height (mm) Shell width (mm) Shell thickness (mm) Shell volume (ml) Shape coefficient - Solidity index (gcmT 100 Weight ratio of calcium in solidity index (gem"^ ) 100 Shell puncture force (N) Shell collapse force (N) Calcium (g. %) Phosphorus (mg, %) Raw ash (g,%) Measurement method Measurement made after drying the shell at room temperature Measurement method (Nature Protection Laboratory of the NAS) Measurement method according to Chevallier (1977) Measurement method according to Chevallier (1977) Mean thickness of 10 measurement points Volume of water contained in a shell Quotient of shell width and height, Chevallier (1977) S.i = [shell weight (height width) ' ] 100 Interpretation: the lower the coefficient value, the less solid the shell Measurement method according to Cooke (1973) S.i. Ca = [Ca weight in shell (height width)"' ] 100 Interpretation: the lower the coefficient value, the less effect of the mineral fractions ratio in shell on the shell solidity. TA.XT2 texture analyzer, SMS-P/2N needle 0.1 mm s ' TA.XT analyser, SMS-P/4 plumper, 2 mm s ' Complexometric titration with sodium versenate (Hermanowicz et al. 1999) Colorimetric method with ammonium molybdate and menthol as a reducer in a mineralized sample (Hermanowicz et al. 1999). Ash determined at 550 °C (Hermanowicz et al. 1999) maturating at the age of 2+-3+, were different from the shells of both subspecies of C. aspersum, containing more calcium in the solidity index, higher mechanical strength for collapse, and a lower value for shape index. Effect of microclimatic conditions On average, relative humidity in 2002 was higher for the whole breeding season (May- September) compared to 2003. The differences ranged between 7.2% (field enclosure) and 8.9% (park), and extreme season-related differences between the various locations ranged from 0.4% to 23.7% (Fig. 4). The mean air temperature in the greenhouse in 2002 was exactly the same as in 2003, and for the other breeding areas the differences ranged from 0.2 to 0.3 °C. However, the differences in the same season between locations were much more pronounced, ranging from 0.9 °C to 2.8 °C (Fig. 5). For Cornu aspersum aspersum and Cornu aspersum maxima, in a 6-month breeding cycle, puncture force for mature snails decrea.sed and the collapse force increased with increasing mean .sea.sonal relative air humidity in the breeding areas used for the study. Correspondingly, the calcium content increased in C. aspersum aspersum shells, and the calcium and phosphorus content decrea.sed in C aspersum maxima shells ('lable 3). Similarly, the thickness of the shells of both Cornu aspersum subspecies increased with increasing mean .sea.sonal temperature. In contrast to the effect of relative air humidity, the shell puncture force increased for the shells of both C aspersum subspecies. 4'he calcium content in the shells of both subspecies decrea.sed with increasing temperature. Interrelation between shell features The regression analyses between the pairs of shell features indicated that the highest number of statistically significant (P < 0.05) and highly significant (P < 0.01 ) relationships were found in Cornu aspersum maxima shells, followed by Cornu aspersum aspersum, and Helix pomatia (Table 4). Mechanical strength of Cornu aspersum maxima shells The puncture force for Helix aspersa maxima shells in- creased (P < 0.05 or P < 0.01) with the increased weight, height, and thickness of the shell as well as with an increasing solidity index and shell collapsing force. Puncture force also increased with decreasing calcium content. Because mean calcium content in shells decreased and shell puncture force increased with snail age, declining calcium content should be considered an index of shell .structural maturation, which is accompanied by increasing mechanical strength. No relationship was found between collap.se force and the width, volume, and .shape of a shell or its pho.sphorus content (Table 4). However, relationships between the collapsing force and other shell features differed from the relationships described for puncture force. ( 1 ) Shell collapsing force decrea.sed (P < 0.05) with increasing diameter, width, and \'olumc of a shell whereas puncture force was not related to these features. (2) Shell collapsing force increa.sed (P < 0.01) with decreasing phosphorus content, lust as calcium content was a measure ofshell maturation in relation to puncture force, phosphorous played a similar role in relation to collapsing force, consti- tuting an index of the organic structure of a shell. PULMONATE SHELL CHARACTERISTICS FROM DIFFERENT POPULATIONS 177 Table 2. Mean values of features of mature shells of Cornu aspersum and Helix pomatia from all samples collected in 2002-2003. differences of statistical significance [P < 0.05); differences of high statistical significance (P < 0.01 ). Cornu i aspersum maxima populations Cornu aspersum aspersum populations Helix “Bailee “Albino” “Bailee” “Albino” “Bailee “French” “Bailee “French” pomatia Shell parameters age 0-t age 1 + age 0-1- age 1-t- age 2+ - 3-1- Weight (g) 3.5 3.6 4.4 4.5 1.7 1.8 2.1 2.0 3.9 Solidity index [(g cm ") 100] 21.2 21.0 28.3* 26.2* 16.9* 17.8* 20.6 21.3 26.4 Diameter 33.8 35.1 33.0 32.8 27.1 27.2 27.0 27.2 32.4 Shell thickness (mm) 0.33 0.33 0.44 0.42 0.25 0.27 0.31 0.31 0.37 Shape index 1.05'^* 1.09** 1.05** 1.09** 1.08* 1.07* 1.08** 1.05** 0.97 Calcium content (g, %) 34.8 34.6 32.4 32.3 41.5 41.5 38.1 39.8 36.8 Calcium content in 179 182 212* 00 00 107 113 117 117 212 150 shells (g) Calcium content in solidity 7.2 7.1 9.1** 7.3** 7.1 7.4 7.7** 8.3** 9.7 index (g cm'") Phosphorus content (mg, %) 0.0024 0.0026 0.0011* 0.0016* 0.0087*” 0.0054** 0.0044** 0.0055** 0.0015 Raw ash content (g, %) 81.9 79.3 83.2 84.5 62.9 70.2 65.3 64.7 66.6 Shell puncture force (N) 20.0 20.0 26.7 26.0 9.5* 10.7* 14.6** 17.0** 23.2 Shell collapse force (N) 77.6** 62.0** 105.3** 79.8** 51.7** 64.9** 39.7** 50.0** 122.6 100 60 50 ' ' ' ' ' ' Park Basin Glass-house Field enclosure Figure 4. Mean relative humidity in 2002 (open boxes) and 2003 (hatched boxes). Box represents ± S£; whiskers indicate maximum and minimum values. 32 30 28 o o 18 16 Park Basin Glass-house Field enclosure Figure 5. Mean temperature in 2002 (open boxes) and 2003 (hatched boxes). Box represents ± S£; whiskers indicate maximum and minimum values. Mechanical strength o/ Cornu aspersum aspersum shells The shell puncture force increased [P < 0.05 or P < 0.01) with increasing weight and thickness of a shell, tind with an increase in the shell solidity index. As in the case of Cornu aspersum maxima, the puncture force increased (P < 0.01 ) with decreasing calcium content, providing further proof that decreasing calcium content constitutes an index of shell structural maturation. No relation was found between punctuie force and the diameter, width, and volume of a shell. The increase in force with decreasing raw ash content proved to be statistically not significant. No correlations were found between the shell collapsing force and other shell features or even with the puncture force, this being an important difference between the shells of this subspecies and those of Cornu aspersum maxima. Mechanical strength of Helix pomatia shells In this study there was no statistically significant rela- tionship between the mechanical strength of Helix pomatia 178 AMERICAN MALACOLOGICAL BULLETIN 27 • 1/2 • 2009 Table 3. Correlation coefficients (r) for statistically significant {P < 0.05 or 0.01 ) relationships between mean relative humidity or temperature in the snail breeding period and mean values of selected features of mature Cornu aspersiitn shells. Climatic factor Cornu aspersum subspecies Weight Diameter Calcium content Thickness (%) Relative humidity Cornu aspersum aspersum -0.79 -0.93 0.85’ Cornu aspersum maxima -0.75 -0.95’ Phosphorus Raw ash content content Puncture Collapse (%) (%) force force -0.84 0.52- -0.71 -0.74 -0.93 0.49 Temperature Cornu aspersum aspersum Cornu aspersum maxima 0.73 ‘ for humidity over 80%, relationship between the features disappeared ■for humidity over 72%, relationship between the features disappeared ^relationship found for temperatures over 22.9 °C shells and other shell features. Particulary, relatively high correlation coefficients between the shell puncture force and its shape and phosphorus content were not statistically significant. The same, statistically not significant were high correlation coefficients between the shell collapsing force and the shape index, calcium content, and phosphorus content. Differences between Helix pomatia and Cornu aspersum regarding the relationships between the mechanical strength of a mature shell and its chemical composition were the result of differences in the ages of snails in separate age groups and differences related to sampling sites. Samples of the Roman snail originated mainly from the natural environment, with different microclimatic conditions than in the breeding enclosures of C. aspersum. Relationship between Cornu aspersum and Helix pomatia shell solidity and calcium content No relationships were found between the shell solidity index and the total content by weight of accumulated calcium, nor between the calcium content in the solidity index and the percentage of calcium content in shells. However, for Cornu aspersum maxima, Cornu aspersum aspersum, and Helix pomatia, significant relationships were found between the shell solidity index and the percentage of calcium content in the shell (Table 5) and calcium content in the solidity index. DISCUSSION Shell calcium, phosphorus content, and shell mechanical strength The results of regression analysis indicated that increasing puncture and collapsing force in the ca.se ot Cornu aspersum maxima shells and increasing puncture force in Cornu aspersum aspersum were accompanied by a decrea.se in calcium 0.58 -0.74 0.88 0.53 -0.62^ 0.98 content. Calcium content decreased with the age of the snail whereas shell mechanical strength increased with age, which confirms the relationship between the observed quantitative chemical changes and the structural maturation of a shell. Shell mechanical strength was also largely dependent on phospho- rus content, despite the fact that this element constituted only a small fraction of the weight of the shell. Phosphorus was, however, an indicator of the organic matter contained in a mature shell, or was a residue after transformations important for shell maturation. In C. aspersum maxima, collapsing force decreased with an increase in phosphorus content, and in Helix pomatia, the force increased. The puncture force for C. aspersum aspersum and H. pomatia shells decreased with an increase in the phosphorus content. The role of calcium and phosphorus as indicators of shell structure differentiation is confirmed by the fact that the per- centage content of phosphorus was more closely correlated to shell collapsing force than to shell puncture force. However, puncture force, which increased with the increasing unitary weight of a shell and its solidity index, was negatively correlated with the percentage of calcium content for all the studied species and subspecies of snails. It was assumed that both puncture and collapsing forces for both species of snails would increa.se with increasing shell weight, solidity index, and thickne.ss. Llowever, in both subspecies of Cornu aspersum and in the Roman snail [Helix pomatia) this a.ssumption was true only for shell puncture force. In Cornu aspersum aspersum, no relationship was found for the collapsing force, and in the Roman snail, the shell collapsing force actually decrca.sed with increases in weight and shell .solidity index, fhe puncture force value does indeed depend on the above-mentioned physical features, but collapsing force was more dependent on the variable micro- structural features of a shell in individual species and populations of snails. Table 4. Correlation coefficients (r) for relationship between mean values of features of mature Cornu aspersum and Helix pornatia shells. statistically significant result [P < 0.05); highly statistically significant result (P < 0.01). PULMONATE SHELL CHARACTERISTICS FROM DIEFERENT POPULATIONS 179 o -C Cu ★ ★ ★ ★ ★ ★ * ★ ★ ★ ★ ★ 9r ie cr> ON o o 00 CO in Cs) O NO m NO LO CO NO Oh d d d 1 d d d d d 1 d c u c/D Cu ★ ★ ★ * ★ ■k * ★ ★ ★ (N r^i o fO o ON 00 00 00 ON 00 00 On u -S d d d 1 d 1 d d 1 d d (U Vh D tj C D Dh ★ ★ k k ★ ★ * ■* ★ ★ k ★ ★ k ★ k k ★ k ★ k ■“rt ^ ON LT) (N f— ( o in ON 00 CO ON GO LO NO NO T'v ON ON NO ON ON ON oo o d d d t d d 1 d d d d d d d d O- oj ? ^ OO -S OJ ★ ★ ★ ★ ★ 9r * * ★ * E k ★ ★ k ★ k ★ ★ ★ k ★ ★ k ★ m 00 ON NO ON 00 00 NO ON o _> NO ON ON ON 00 ON 00 ON ON ON NO NO ON 00 1 d d d d d d d d d d d 1 d t d d -o (U C u IS H OJ E Zj :> ^ X QJ ]_i a> >'- 12