Molluscan Research one T Volume 23 ə Number 2 (i 7 cp 5^5nns Ri é əki uUa The Malacological Society of Australasia Molluscan Research Molluscan Research is a publication for authoritative scientific papers dealing with the Phylum Mollusca and related topics. Three numbers are published in each annual volume. General and theoretical papers relating to molluscs are welcome. Papers concerning specific geographical areas or new taxa should normally focus on the Indo-West Pacific region, as well as Australasia and the Southern Ocean. The journal assumes that all authors of a multi-authored paper agree to its submission. The journal uses its best endeavours to ensure that work published is that of the named authors except where acknowledged and, through its reviewing procedures, that any published results and conclusions are consistent with the primary data. It takes no responsibility for fraud or inaccuracy on the part of the contributors. Editors Managing Editor Dr W. F. Ponder, Australian Museum, Sydney, New South Wales Dr M. G. Chapman, University of Sydney, Australia Mr B. A. Marshall, Te Papa Tongarewa, Museum of New Zealand, New Zealand Dr P. Middelfart, Australian Museum, Sydney, New South Wales Dr W. B. Rudman, Australian Museum, Sydney, New South Wales Dr J. Stanisic, Queensland Museum, Brisbane, Queensland C. M. Myers, CSIRO PUBLISHING, Melbourne, Victoria Associate Editors Production Editor Issued 3 times a year The Notice to Authors is published in the first issue of each volume and is available from the journal's Subscription prices for 2003 Sats Institutional rates UE “ə: Publishing enquiries should be sent to: On-line On-line & Print Molluscan Research ' CSIRO PUBLISHING Australia & New $AU100 $AU110 PO Box 1139, Collingwood, Zealand Vic. 3066, Australia. Rest of the world $US75 $US85 Telephone: 61 3 9662 7629 — .—“—— Facsimile: 61 3 9662 7611 Personal subscriptions to the print edition are Email: camilla.myers@csiro.au available as a benefit of membership of the Molluscan Research is abstracted/indexed by: Malacological Society of Australasia. Web edition and on-line licences Prices and further information are available at: Aquatic Science and Fisheries Abstracts; Current Contents/Agriculture, Biology and Environmental Sciences; CAB International; Science Citation Index; Zoological Record. http://www.publish.csiro.au/journals/mr/ E de OPEM TR ISSN 1323-5818 @ Council of The Malacological Society of Australasia Published for The Malacological Society of Australasia Ltd (ACN 067 894 848) by CSIRO PUBLISHING. Printed by Brown Prior Anderson Pty Ltd. Permission to photocopy items from this journal is granted by CSIRO to libraries and other users registered with the Copyright Clearance Centre (CCC), provided that the stated fee for each copy is paid directly to the CCC, 222 Rosewood Drive, Danvers, MA 01923, USA. For copying in Australia, CSIRO PUBLISHING is registered with Copyright Agency Ltd, Level 19, 157 Liverpool Street, Sydney, NSW 2000. Cover design: Jane Guy, Queen Victoria Museum, Launceston, Tasmania and James Kelly, CSIRO PUBLISHING, Melbourne, Victoria. Molluscan Research Contents Volume 23 Number 2 2003 Embryonic and larval development of Pinctada margaritifera (Linnaeus, 1758) M. S. Doroudi and P. C. Southgate 101 Ultrastructure of male germ cells in the testes of abalone, Haliotis ovina Gmelin S. Singhakaew, V. Seehabutr, M. Kruatrachue, P. Sretarugsa and S. Romratanapun “109 Gastropod phylogeny based on six segments from four genes representing coding or non- coding and mitochondrial or nuclear DNA D. J. Colgan, W. F. Ponder, E. Beacham and J. M. Macaranas 123 Reassessment of Australia's oldest freshwater snail, Viviparus (?) albascopularis Etheridge, 1902 (Mollusca : Gastropoda : Viviparidae), from the Lower Cretaceous (Aptian, Wallumbilla Formation) of White Cliffs, New South Wales B. P. Kear, R. J. Hamilton-Bruce, B. J. Smith and K. L. Gowlett-Holmes 149 Relationships of Placostylus from Lord Howe Island: an investigation using the mitochondrial cytochrome c oxidase 1 gene W. F. Ponder, D. J. Colgan, D. M. Gleeson and G. H. Sherley 159 Short contribution Changes in tissue composition during larval development of the blacklip pearl oyster, Pinctada margaritifera (L.) J. M. Strugnell and P. C. Southgate 179 i M ki. —— 2 — s... an xi E CSIRO PUBLISHING www.publish.csiro.au/journals/mr Molluscan Research, 2003, 23, 101—107 Embryonic and larval development of Pinctada margaritifera (Linnaeus, 1758) Mehdi S. Doroudi^P€ and Paul C. Southgate^ ^School of Marine Biology and Aquaculture, James Cook University, Townsville, Qld 4811, Australia. BNSW Fisheries, c/o Murray Irrigation, Wakool, NSW 2710, Australia. CAuthor to whom all correspondence should be addressed. Email: mehdi.doroudi@fisheries.nsw.gov.au Abstract We describe the developmental stages of the black-lip pearl oyster, Pinctada margaritifera (Linnaeus, 1758), larvae from fertilisation through embryonic development and larval growth in the laboratory at 28 + 1°C. Larvae were anesthetised, fixed, critical-point dried and examined using a scanning electron microscope. We examined embryonic development (fertilisation, polar body, blastomeres, gastrula) and attributes of the larval shell (size, prodissoconch I/II, growth lines, provinculum, shell fracture) and larval velum. The first polar body formed 24 min after fertilisation and fertilised eggs had a mean diameter of 59.9 + 1.4 um. The earliest actively swimming trochophore appeared 8-12 h after fertilisation. The D stage was reached approximately 24 h after fertilisation and measured 79.7 + 2.3 um in shell length. Ten-day-old larvae had umbones that arose opposite each other above the hinge axis and 22-day-old larvae, with a mean shell length of 230.8 + 4.9 um, developed a pigment spot just before entering the pediveliger stage. Additional keywords: hinge, morphology, pearl oyster, scanning electron microscope. Introduction Herdman (1903) studied the early life stages of Pinctada vulgaris (= P. fucata Gould, 1850) larvae up to 3 days after fertilisation. Other studies have investigated larval development and growth of P. fucata, P. martensi Dunker, 1850 and P maxima Jameson, 1901 (Ota 1957; Minaur 1969; Tanaka and Kumeta 1981; Alagarswami et al. 1983). Rose and Baker (1994) described larval development of P. maxima in detail and compared their findings with those reported for other pearl oyster species. Although P margaritifera larvae have been cultured since 1970 (Tanaka et al. 1970; Alagarswami et al. 1989; Southgate and Beer 1997), embryonic development and the morphology of different larval stages have not been described in detail. The present study notes the characteristics of P. margaritifera larvae that have not been reported on previously using scanning electron microscopy (SEM) and provides a basic understanding of larval development during hatchery culture. Materials and methods Larval rearing Pinctada margaritifera broodstock were induced to spawn by thermal stimulation and the addition of sperm in a seawater suspension. Fertilised eggs were stocked at a density of 30 mL”! in aerated fibreglass tanks (500 L) filled with 1 um filtered seawater at 28°C. The salinity of seawater was 33, which was measured using the practical salinity scale. After 24 h, when D-stage larvae (shell becomes D-shaped) had a mean shell length of 79.7 + 2.3 um, they were collected on a 25-um mesh sieve, counted and placed at a density of 2 mL! in 500-L aerated fibreglass tanks containing filtered seawater at the same temperature and salinity. We cultured Tahitian /sochrysis aff. galbana Green (T-ISO) and Pavlova salina Green in 3-L glass flasks and 20-L carboys in autoclaved 0.45-m-filtered and ultraviolet (UV)-treated seawater using f/2 nutrient © Malacological Society of Australasia 2003 10.107 I/MR02015 1323-5818/03/020101 102 Molluscan Research M. S. Doroudi and P. C. Southgate medium (Guillard 1983). Microalgae cultures were provided with illumination from cool white fluorescent lights with a 12-h light: 12-h dark photoperiod. Larvae were fed daily a 1:1 mixture of T-ISO and Pavlova salina at a ration of 1-18 x 10? cells mL”1 (Southgate and Beer 1997, Doroudi er al. 1999a). We conducted three separate spawnings to collect embryonic and larval samples. Sample preparation We observed embryonic development every 15 min during the first 3 h, then once an hour until the trochophore stage (8-24 h) and D-stage (24 h) using a compound microscope. Larvae were narcotised in a 15% (w/v) solution of MgCl, (Bellolio et al. 1993) and seawater (1:1) at 28°C for 5-10 min. We collected samples from three tanks at 2-day intervals after the D stage (24 h) and fixed them in 2.5% glutaraldehyde in 0.1 M piperazine at pH 7.6. This sampling continued until larvae had developed to the “eyed” stage (i.e. when larvae develop a pigment spot). Subsamples of larvae were post-fixed with osmium tetroxide (OsO,), dehydrated in a graded series of ethanol, critical-point dried in liquid carbon dioxide (CO,), mounted on aluminium stubs with double-sided tape and coated with gold before being examined using an SEM. We collected larvae on a mesh sieve, washed them into a graduated cylinder and removed a subsample, from which the shell length of 40 larvae was measured using a compound microscope. Morphological terminology follows that commonly used for bivalve larvae in similar studies (Waller 1981; Belollio et al. 1993). Results The time series of developmental stages of P margaritifera embryos and larvae is shown in Table 1. Embryo to trochophore Unfertilised P. margaritifera eggs had a mean diameter of 39.7 + 1.3 um (n = 40; Fig. 1). The first polar body formed 24 min after fertilisation and fertilised eggs had a mean diameter of 59.9 + 1.4 um (n = 40). The four blastomeres resulting from the second cleavage formed 2 h after fertilisation. Cell division followed the usual bivalve pattern for bivalves and resulted in the gastrula, 5 h after fertilisation. The change from a ciliated gastrula to the trochophore stage was gradual and the earliest actively swimming trochophore appeared 8—12 h after fertilisation. Morphological changes from trochophore to the D stage included extension along the longitudinal axis and the apical region becoming broader than the posterior region. At this time, cilia on the apical region became longer. With the development of long cilia, larvae began to secrete shell and the resulting larvae swam actively using the velum. Larvae The D stage was reached approximately 24 h after fertilisation and larvae measured 79.7 + 2.3 um (n = 40) in shell length. The D-stage larvae showed preliminary growth rings after 2 days (Fig. 2). The shell showed slight umbonal growth after 6 days development and prodissoconch I and II could be clearly identified (Fig. 3). At a shell length of approximately 100 um, the hinge developed denticulation on either side of a central region. As the larva grew, the hinge developed and formed a series of teeth and sockets on each valve (Fig. 4). Each valve had teeth on either side of a central area (Fig. 5). The central area included a series of tiny teeth and sockets (Fig. 6). Ten-day-old larvae had umbones that arose opposite each other above the hinge axis (Fig. 7) and 22-day-old larvae, with a mean shell length of 230.8 + 4.9 um (n = 40), developed a pigment spot and entered the pediveliger stage shortly after (Fig. 8). Sections of broken shell edges at this stage suggest that calcification of the shell by the mantle had occurred in prodissoconch II (Fig. 9). Fractures through prodissoconch II showed layering in the shell structure and indicated that the shell is thicker in the area of growth lines (Fig. 10). The oval velum was located at the 103 Molluscan Research Development of Pinctada margaritifera 'Ápnjs juosoud “9 :(6861) [D 12 YureAs1eSe[y “ç “(1861) eourmy pue exeuer “p “(6961) MEUN “€ ((p661) səyeq pue osos “Z (£861) 7? zə IUIEASIUGU[ V *[ :SAO[[O] SE IIE LBP IY} JO soo1nos au [ “(p) s&ep 10 (u) sanoq *(urur) sojnurur ur uƏAI8 st (L) our “(uni ur) YASUO] ffəqs 10 19j9urerp [BAIL] 10 339 OY} 1ot[o Jo (S) ƏZIS ULIN P cc 8'0£€c P9I orz = i F = PIC Stc PSI OIT jods-oAq PcI L'OVI PZI al - = PcI STI PI Ori PII SEI oqunN ps 6601 P6 OII POI SII PS 96 Ps OTT F 001 oquin A[req ur L'6L Upc SL V 07 SL = s ut $8 Voc L9 edeus q u cI-8 VOL = a i x - = us SL ` " e1oqdougoo1[ uç 9°69 > = m = ” > qs SL *5 ^ e[nnser) qc 09 = = = F DM x ul 09 = ^ 9IoUIOJSE[q NOY utu pc 66S m = = zi = = = = = = Apoq 1v[oqd = Lor = Sy = > F 6S = = = Ly 534 1 S At S m S L S m S il S Ə5u1S 67-LT 0-97 8c-9c It-Lc S'6c-0 Sc 8'6c-t vc (Os) o11je1odurop 9 S v € T I xSIotpiny puəfulupöapul ppp)oulq DUIXDUL nppj2ulq pipəonf ppploulq sərəəds 19js&o [1eod juej10dur ouros yo juauido[oAop [EAB]; pug o&1qurq ` AQEL 104 Molluscan Research M. S. Doroudi and P. C. Southgate Figs 1-6. 7, Unfertilised egg of Pinctada margaritifera with sperm (s) on the surface. 2, External view of the right valve of Pinctada margaritifera D-stage larvae; I, prodissoconch I; II, prodissoconch II. 3, Prodissoconch I (I) and II (II) in the early umbo stage of Pinctada margaritifera larvae. 4, Development of the provinculum of Pinctada margaritifera larvae (umbo stage); t, tooth; c, central area. 5, Main teeth on either side ofa central area of Pinctada margaritifera larvae (umbo stage); t, tooth; s, socket. 6, Central area of the Pinctada margaritifera larval hinge (umbo stage) showing the series of teeth (t) and sockets (s). Development of Pinctada margaritifera i Molluscan Research 105 Fig. 7-12. 7, Hinge length (h) and umbones (u) arise over the hinge axis of Pinctada margaritifera larvae (umbo stage). 8, An eyed larva of Pinctada margaritifera showing details of umbonal features (u); r, right valve. 9, The mantle (m) viewed from a cross-section of the prodissoconch II (IT) of Pinctada margaritifera larvae (umbo stage). 10, Fracture of the prodissoconch in Pinctada margaritifera umbo larva; p, outer prismatic layer; gh, granular homogeneous layer. /1, Lateral view of the entire Pinctada margaritifera D-stage larva with velum (v) extended at the anterior dorsal side of the shell; r, right valve. 12, Enlargement of cilia (c) on the velum of Pinctada margaritifera D-stage larvae. posterior dorsal side of the larva (Fig. 11) and was well developed with a peripheral ring of cilia (Fig. 12). Discussion Pinctada margaritifera larvae need a period of 8 days to reach the early umbo stage (shell length 110 um) and exhibit an average daily growth rate of 3.7 um. Elsewhere, a daily 106 Molluscan Research M. S. Doroudi and P. C. Southgate growth rate of 5 um has been reported for P. margaritifera during the first 7 days of the larval rearing period (Tanaka ez al. 1970). Growth rates of bivalve larvae are likely to be influenced by genetic factors, as well as endogenous and exdogenous nutrition and culture conditions. Because P. margaritifera larvae have exponential growth (Doroudi et al. 19992), the mean daily growth rate increased up to 7.2 um over the period of 22 days required for larvae to reach the eye spot stage (230 um). Eye spots in P. margaritifera generally occur in larvae that are 230 um or greater in shell length. In a previous study, P. margaritifera larvae developed a pigment spot at 210 um shell length (Alagarswami et al. 1989). In P. fucata (Alagarswami et al. 1983) and P. maxima (Rose and Baker 1994), eye spots form in individuals that are approximately 210 and 230 um in shell length, respectively. Despite variations in rearing conditions (e.g. environmental factors, type of food and genetic differences), P. margaritifera, P. fucata and P. maxima settle at approximately the same size (230-266 um) and age (20—23 days) from fertilisation (Rose and Baker 1994). The overall development of P. margaritifera larvae described in the present study is similar to the more general descriptions of pearl oyster larvae reported in previous studies (Table 1). The larvae of bivalves are similar in exterior appearance and are difficult to differentiate without detailed anatomical study. The present SEM study of P. margaritifera larvae revealed some anatomical features of the larval shell that have not been observed previously using other techniques. For instance, the punctate region on the exterior surface of prodissoconch I of P. margaritifera is also observed in Crassostrea virginica Gmelin, 1791 (Carriker and Palmer 1979) and Ostrea edulis Linnaeus, 1750 (Waller 1981). Hinge structure can be a primary character in identification of bivalve larvae (Le Pennec 1980). Lutz et al. (1982) reported that hinge structure differed among 12 genera of bivalves, whereas the basic hinge morphology of P. margaritifera seems to be similar to that of other pearl oysters; that is, a tooth and socket at each end with a thin central area. The present study has shown that 8-day-old larvae of P. margaritifera, with a shell length of 110 um, have five teeth in each valve, with three at the anterior end of the hinge line and two at the posterior. This is the same as reported for P. maxima larvae with a shell length of 90 um (Rose and Baker 1994). Our observations in the present study on the larval rearing of P. margaritifera provide a basic understanding of larval development during hatchery culture of this species. Acknowledgments This study was partially funded by the Australian Centre for International Agricultural Research (ACIAR) as part of Project No. FIS/97/31 *Pearl Oyster Resource Development in the Pacific Islands'. References Alagarswami, K., Dharmaraj, S., Velayudhan, T. S., Chellam, A., Victor, A. C. C., and Gandhi, A. D. (1983). Larval rearing and production of spat of pearl oyster Pinctada fucata (Gould). Aquaculture 34, 287-301. Alagarswami, K., Dharmaraj, S., Chellam, A., and Velayudhan, T. S. (1989). Larval and juvenile rearing of black-lip pearl oyster Pinctada margaritifera (Linnaeus). Aquaculture 76, 43—56. Bellolio, G., Lohrmann, K., and Dupre, E. (1993). Larval morphology of the scallop Argopecten purpuratus as revealed by scanning electron microscopy. The Veliger 36, 332-342. Carriker, M. R., and Palmer, R. E. (1979). Ultrastructural morphogenesis of prodissoconch and early dissoconch valves of the oyster Crassostrea virginica. Proceedings of the National Shellfisheries Association 69, 103—128. Development of Pinctada margaritifera Molluscan Research 107 Doroudi, M. S., Southgate, P. C., and Mayer, R. (1999a). The combined effects of temperature and salinity on the embryo and larvae of black-lip pearl oyster, Pinctada margaritifera (L). Aquaculture Research 30, 271—277. Doroudi, M. S., Southgate, P. C., and Mayer, R. (19992). Growth and survival of the black-lip pearl oyster (Pinctada margaritifera, L.) larvae fed at different algal density. Aquaculture International 7, 179—187. Guillard, R. L. (1983). Culture of phytoplankton for feeding marine invertebrates. In *Culture of Marine Invertebrates’. (Ed. C. L. Berg.) pp. 108-132. (Hutchinson Ross: Stroudberg.) Herdman, W. A. (1903). Observations and experiments on the life-history and habits of the pearl oyster. In *Report Pearl Oyster Fisheries, Gulf of Mannar'. (Ed. W. A. Herdman.) pp. 125-146. (Royal Society: London.) Le Pennec, M. (1980). The larval and post-larval hinge of some families of bivalve mollusks. Journal of Marine Biology Association UK 60, 601—617. Lutz, R., Goodsell, M., Castagna, M., Chapman, S., Newell, C., Hidu, H., Mann, R., Jablonski, D., Kennedy, V, Siddall, S. et al. (1982). Preliminary observations on the usefulness of hinge structures for identification of bivalve larvae. Journal of Shellfish Research 2, 65—70. Minaur, J. (1969). Experiments on the artificial rearing of the larvae of Pinctada maxima (Jameson) (Lamellibranchia). Australian Journal of Marine and Freshwater Research 20, 175—187. Ota, S. (1957). Notes on the identification of free swimming larvae of pearl oyster (Pinctada martensii). Bulletin of the National Pearl Research Laboratory, Japan 2, 128-132. Rose, R. A., and Baker, S. B. (1994). Larval and spat culture of the Western Australian silver or gold lip pearl oyster, Pinctada maxima (Jameson) (Mollusca: Pteriidae). Aquaculture 126, 35-50. Southgate, P. C., and Beer, A. C. (1997). Hatchery and early nursery culture of the blacklip pearl oyster (Pinctada margaritifera, L.). Journal of Shellfish Research 16, 561—568. Tanaka, Y., and Kumeta, M. (1981). Successful artificial breeding of the silver-lip pearl oyster, Pinctada maxima (Jameson). Bulletin of the National Research Institute of Aquaculture, Japan 2, 21-28. Tanaka, Y., Inoha, S., and Kakazu, K. (1970). Studies on seed production of black-lip pearl oyster Pinctada margaritifera, in Okinawa. V. Rearing of the larvae. Bulletin of Tokai Region Fisheries Research Laboratory, Japan 63, 97—106. Waller, T. R. (1981). Functional morphology and development of veliger larvae of the European oyster, Ostrea edulis Linne. Smithsonian Contributions to Zoology 328, 1—70. http://www.publish.csiro.au/journals/mr egeat ^ j RGC allea 2 Eo CSIRO PUBLISHING www.publish.csiro.au/journals/mr Molluscan Research, 2003, 23, 109—121 Ultrastructure of male germ cells in the testes of abalone, Haliotis ovina Gmelin Sombat Singhakaew^, Viyada Seehabutr®, Maleeya Kruatrachue^, Prapee Sretarugsa© and Suppaluk Romratanapun® ‘Department of Biology, Faculty of Science, Mahidol University, Bangkok 10400, Thailand. Department of Zoology, Faculty of Science, Kasetsart University, Bangkok 10400, Thailand. “Department of Anatomy, Faculty of Science, Mahidol University, Bangkok 10400, Thailand. To whom correspondence should be addressed. Email: scmkt@mahindol.ac.th Abstract An ultrastructural study of male germ cells in the testes of Haliotis ovina revealed that spermatogenesis could be classified into 13 stages, based on the pattern of chromatin condensation and distribution of organelles, as follows: the spermatogonium; five stages of the primary spermatocyte; the secondary spermatocyte; five stages of the spermatid; and the spermatozoa. Each spermatogonium was round or oval, with a euchromatic nucleus and prominent nucleolus. The primary spermatocytes were divided into five stages: leptotene (LSc); zygotene (ZSc); pachytene (PSc); diplotene (DSc); and metaphase (MSc). The nucleus of the LSc contained scattered small heterochromatin blocks that were increasingly thickened in the ZSc. The PSc was characterised by a bouquet pattern of heterochromatin fibres. The DSc decreased in size, resulting in close clumping of chromatin blocks, whereas in the MSc, long and large blocks of chromosomes were formed and then moved to be aligned along the equatorial region. Secondary spermatocyte showed thick chromatin blocks that appeared reticulate. The spermatid could be divided into five stages (St,_;). The St, was a large round cell and its nucleus contained homogeneous chromatin granules. In St), the nuclear chromatin started to condense into patches. The St; was smaller with a round nucleus containing dark blocks of heterochromatin. The St, became smaller still, with a round opaque nucleus. The St; was the smallest round cell, with almost completely condensed chromatin. The spermatozoon had a round to barrel-shaped head that contained completely condensed chromatin covered by a conical acrosome. The posterior border of the nucleus was flanked by five large spherical mitochondria and the tail consisted of axonemal microtubules surrounded by the plasma membrane. Introduction There are almost 100 species of abalone all belonging to the genus Haliotis (Fallu, 1991). These snails are widely distributed in tropical and temperate seas and inhabit submerged rock (Crofts 1929). Since ancient times, the abalone has been an economically important snail because of its decorative shell and food value. Abalone are commercially important molluscs in many countries, such as Japan, America, Mexico and Australia, where the culture of commercial abalone is well established. In Thailand, there are three species of abalone, namely Haliotis ovina Gmelin, 1791, H. asinina Linnaeus, 1758 and H. varia Linnaeus, 1929 (Nateewathana and Hylleberge 1986; Tookvinart ef al. 1986; Bussarawit et al. 1990). Of these the three species, H. ovina is the one with economic potential (Nateewathana and Hylleberge 1986; Bussarawit et al. 1990). Species of Haliotis are dioecious, with clear sexual dimorphism. The single testis envelops the digestive gland and, together, these structures form a large cone-shaped appendage called the conical organ, which wraps around the right and posterior margins of the shell muscle (Crofts 1929). The colour of the gonad indicates the sex of the abalone. The testis is cream to ivory, whereas the ovary is green in colour (Singhagraiwan 1989). © Malacological Society of Australasia 2003 10.1071/MR02016 1323-5818/03/020109 110 Molluscan Research S. Singhakaew et al. There is no genital duct in Haliotis (Crofts 1929). The sperm from the testis are expelled into the cavity of the right renal organ, which is seen on the dorsal surface of the body and is overlapped by the digestive gland, and finally pass through shell perforations, which are located on the left side (Fallu 1991), into the water. There have been extensive ultrastructural studies of male germ cells, especially the spermatozoa, in several haliotid species, such as Haliotis rufescens Swainson, 1822 (Lewis et al. 1980), H. aquatilis Reeve, 1846 (Shiroya and Sakai 1992), H. diversicolor supertexta Lischke, 1870 (Gwo et al. 1997), H. discus Reeve, 1846 (Sakai et al. 1982; Usui 1987), H. midae Linnaeus, 1758 (Hodgson and Foster 1992), H. laevigata Donovan, 1808 (Healy et al. 1998), and H. asinina (Apisawetakan et al. 2000; Sobhon et al. 2001). In general, spermatogenesis in these haliotid species can be classified into the following stages: spermatogonia; primary spermatocytes; secondary spermatocytes; spermatids; and spermatozoa. Sobhon et al. (2001) differentiated male germ cells in H. asinina into 14 stages based on the ultrastructure and pattern of chromatin condensation. These stages were the spermatogonium, six stages of the primary spermatocyte, the secondary spermatocyte, four stages of the spermatid and the spermatozoa (immature and mature). The general ultrastructure of haliotid spermatozoa is typical of the primitive type described for molluscs that reproduce by external fertilisation. The spermatozoa have a very simplified midpiece that is composed of a cluster of spherical mitochondria surrounding a pair of orthogonally arranged centrioles (Lewis et al. 1980; Sakai er al. 1982; Hodgson and Bernard 1986; Hodgson and Foster 1992; Shiroya and Sakai 1992; Gwo et al. 1997; Healy et al. 1998; Apisawetakan et al. 2000). In the present study, we examined the ultrastructure of male germ cells in H. ovina, an abalone of potential economic importance, which is common along the coast of Thailand. The results are compared with those of H. asinina and other abalone species. Materials and methods Adult H. ovina (approximately 7 cm shell length, 170 g weight) were collected during June and July 1999 from Samed Island, Rayong Province, Thailand. A total of 20 male abalone was used in the present study. For the ultrastructural study, very small pieces of testes were fixed in 306 glutaraldehyde in 0.1 M sodium cacodylate buffer (pH 7.8) at 4°C overnight, washed in 0.1 M sodium cacodylate buffer and post-fixed in 1% osmium tetroxide in 0.1 M sodium cacodylate buffer for 1 h at 4°C. The pieces were then dehydrated in a graded series of ethanol, cleared using propylene oxide and infiltrated and embedded in Araldite 520 epoxy resin. Sections were cut with glass knives on a Sorvall MT-2 ultramicrotome. Semithin sections were stained with toluidine blue and observed with a light microscope. Ultrathin sections were stained with uranyl acetate and lead citrate and were viewed with a Hitachi H300 transmission electron microscope (TEM) operated at 75 kV. Results The male germ cells of H. ovina can be classified into 13 stages based on cell shape and size, nuclear shape and size, and pattern of chromatin condensation (Fig. 1). These stages comprise the spermatogonium, five stages of the spermatocyte, the secondary spermatocyte, five stages of the spermatid and the spermatozoa. Spermatogonium These cells lie on the outer edges of the lobe of the testis. The spermatogonium is spherical or oval, with a diameter of approximately 8-10 um (Fig. 14,3). The nucleus is round or slightly oval and contains mostly euchromatin, with only a thin rim of heterochromatin blocks attached to the inner surface of the nuclear envelope (Fig. 24). The nucleolus is Male germ cells of abalone Molluscan Research 111 Fig. 1. Photomicrograph (A) and electron micrographs (B-D) showing various stages of the male germ cells of Haliotis ovina. Sg, Spermatogonia; LSc, leptotene primary spermatocyte; ZSc, zygotene primary spermatocyte; PSc, pachytene primary spermatocyte; DSc, diplotene primary spermatocyte; SSc, secondary spermatocyte; St, spermatid; Sz, spermatozoa. prominent against the background of a rather transparent nucleoplasm. The cytoplasm contains free ribosomes and numerous mitochondria, which appear at the outer surface of the nuclear envelope (Fig. 24). Spermatocytes The primary spermatocytes (PrSc) consist of five stages: leptotene (LSc), zygotene (ZSc), pachytene (PSc), diplotene (DSc) and metaphase (MSc) spermatocytes. The early cells 112 Molluscan Research S. Singhakaew et al. Fig.2. A, High magnification of a spermatogonium (Sg) showing the round nucleus (Nu) with distinct nucleolus (No) and heterochromatin (hc) blocks. Mitochondria (Mi) are numerous on the outer surface of the nuclear envelope (NE). B, A leptotene primary spermatocyte (LSc) contains an oval nucleus with nucleolus. The cytoplasm contains mitochondria, ribosomes (Ri), endoplasmic reticulum (ER) and proacrosomal vesicles (pav). C, A zygotene primary spermatocyte contains an oval nucleus with dense heterochromatin and mitochondria in the cytoplasm. D, A pachytene primary spermatocyte (PSc) contains a nucleus with a bouquet pattern of heterochromatin. Note the presence of ER and mitochondria in the cytoplasm. (LSc, ZSc, PSc) are round, approximately 12-15 um in diameter, whereas the late-staged cells (DSc, MSc) are smaller in size (8-10 um in diameter). Other distinctive differences among various stages ofthe PrSc are the pattern of chromatin condensation and the relative amount of euchromatin versus heterochromatin. Male germ cells of abalone Molluscan Research 113 Leptotene spermatocyte These cells are round, approximately 12-15 um in diameter (Fig. 14,3). The chromatin, which has started to condense into small blocks of heterochromatin, is scattered evenly throughout the nucleus. The nucleolus is still present, but not as prominent as that of the spermatogonium. The cytoplasm contains a few ribosomes, mitochondria and proacrosomal vesicles (Fig. 23). Zygotene spermatocyte These cells are approximately the same size as the LSc (approximately 12 um in diameter; Fig. 18). Under TEM, the heterochromatin blocks are denser than those of the LSc and the nucleolus disappears (Fig. 2C). The cytoplasm contains proacrosomal vesicles and mitochondria, similar to those in the LSc. Pachytene spermatocyte These cells are round and their sizes are smaller than the LSc (approximately 10—12 um in diameter; Fig. 1C). The heterochromatin appears as long threads and thick fibres that are entwined into a ‘bouquet pattern’, and are visible throughout the nucleus (Fig. 2D). The cytoplasm contains proacrosomal vesicles; the mitochondria and rough endoplasmic reticulum are greater in number than the LSc (Fig. 22). Diplotene spermatocyte These cells are similar to the PSc, but are smaller (approximately 8—10 um in diameter). The nucleus of the DSc is also smaller than that of the PSc. The chromatin strands, which are distributed among the denser nucleoplasm, become increasingly thicker than those in the earlier stage (Fig. 34). The cytoplasm contains mitochondria and proacrosomal vesicles, similar to those in the PSc. Metaphase spermatocyte These cells (approximately 8 um in diameter) exhibit thick chromosomes that are arranged close together along the equatorial region and the nuclear membrane completely disappears (Fig. 33). The cytoplasm contains mitochondria, numerous ribosomes and proacrosomal vesicles (Fig. 33). Secondary spermatocyte These cells are round and smaller than the MSc (approximately 6 um in diameter; Fig. 1C). They are rarely observed and occur after meiosis I. The SSc shows thick chromatin blocks composed of criss-crossing fibres, which appear as reticulate chromatin (Fig. 3C). Fewer mitochondria and proacrosomal vesicles are visible in the cytoplasm (Fig. 3C). š Spermatids There are five stages of spermatid (St, ;), differing in size, nuclear shape and chromatin condensation pattern. Spermatid 1 The nuclei of St, are round (approximately 6 um in diameter). The St, can be distinguished by their chromatin, which appears as homogeneous granules that are S. Singhakaew et al. 114 Molluscan Research Fig. 3. 4, A diplotene primary spermatocyte (DSc) contains a round nucleus (Nu) vvith very thick heterochromatin (hc). Note the presence of mitochondria (Mi), ribosomes (Ri), proacrosomal vesicles (pav) and Golgi body in the cytoplasm. B, A metaphase spermatocyte (MSc) exhibits thick chromosomes arranged along the equatorial region. Mitochondria, ribosomes and proacrosomal vesicles are still present. C, A secondary spermatocyte (SSc) with a round nucleus showing reticulate chromatin (ch). D, Spermatid I (St,) showing a round nucleus with homogeneous granular chromatin. gc, Golgi complex. uniformly spaced throughout the nucleus (Fig. 3D). As a result, the whole nucleus appears moderately dense without any intervening transparent area of nucleoplasm. The cytoplasm exhibits fewer organelles, such as mitochondria, which tend to be concentrated on one side of the nucleus (Fig. 3D). Male germ cells of abalone Molluscan Research 115 Spermatid II The general features of St, are similar to those of St,, but the nucleus, which is still round, is smaller in size in St; (approximately 5 um in diameter) and is located centrally within the cell (Fig. 44). The chromatin appears as homogeneous granules that are uniformly spaced throughout the nucleus and condensed into patches. The cytoplasm contains a few mitochondria on one side and ribosomes (Fig. 44). Spermatid III The St, is smaller than St; (approximately 4 um in diameter) and the nucleus maintains a round shape. The chromatin begins to condense into dark blocks, with intervening light areas of nucleoplasm (Fig. 4B). The cytoplasm contains few ribosomes, mitochondria and proacrosomal vesicles (Fig. 48). Spermatid IV The St, is smaller (approximately 3 um in diameter), but still appears round in shape. The chromatin of St, is almost completely condensed, resulting in an opaque nucleus (Fig. 4C). At this stage, there is a continued loss of cytoplasm, a decrease in nuclear size and condensation of nuclear material. The cytoplasm contains numerous ribosomes, a few large mitochondria and centrioles (Fig. 4C). Spermatid V The St; is the smallest among ther spermiogenic cells (approximately 2 um in diameter), but is still round in shape. The chromatin of St, is almost completely condensed (Fig. 4D). Fusion of proacrosomal vesicles into an acrosome that appears slightly indented at the anterior region of the nucleus can be seen (Fig. 4D). The cytoplasm contains numerous ribosomes and a few mitochondria (Fig. 4D). The mitochondria are fewer, but are larger in size, and occupy the posterior border of the nucleus (Fig. 4D). Spermatozoa Mature spermatozoa are detached from trabeculae and are arranged in rows further from the former stages, whereas their long tails are pointing outwards and are mingled with those of another spermatogenic unit (Fig. 12). The nucleus (1 x 4 um in size) is fully elongated and slightly tapered at the anterior end (Fig. 54). The chromatin is completely condensed and the anterior portion of the head is covered by an acrosome that appears as an inverted cup, the concavity of which separates it from the anterior border of the indented nucleus (Fig. 54). This subacrosomal space contains a crystalline acrosomal core embedded in more homogeneous material (Fig. 5C). The acrosomal matrix appears homogeneous, with varying degrees of electron opacity (Fig. 54). The nuclear chromatin is completely condensed (Fig. 5C). The head of each spermatozoon comprises a barrel-shaped nucleus (Fig. 54). At the posterior border of the nucleus, there are proximal and distal centrioles that are surrounded by a ring of five round mitochondria with cristae (Fig. 54.3). The tail, or flagellum, consists of a 9 + 2 arrangement of microtubules and is surrounded by a plasma membrane (Fig. 52). Discussion Most light microscopic studies on Haliotis have not categorised the various stages of spermatogenesis, apart from suggesting that there are four broad classes: spermatogonia, 116 Molluscan Research S. Singhakaew et al. Fig. 4. 4, Spermatid II (St;) shows more condensed chromatin (ch) in a round nucleus (Nu). The cytoplasm exhibits fewer organelles. B, Spermatid III (St;) contains a round nucleus. Chromatin is condensed into dark blocks. Mitochondria (Mi) begin to form a cluster. C, Spermatid IV (St,) appears round in shape. The nucleus contains almost completely condensed chromatin. Mitochondria are located at the posterior border of the nucleus. Note the presence of centrioles (ce). D, Spermatid V (St;) contains a nucleus with almost completely condensed chromatin. Note the fusion of proacrosomal vesicles (pav) into an acrosome (Ac). Mitochondria are larger and occupy the posterior border of the nucleus. Ri, Ribosome. spermatocytes, spermatids and spermatozoa (Tomita 1967; Takashima ef al. 1978). Apisawetakan er al. (1997) studied the gametogenic processes in H. asinina using the light microscope and classified spermatogenesis into 13 stages: spermatogonium, five stages of primary spermatocyte, secondary spermatocyte, four stages of spermatid and spermatozoa Male germ cells of abalone Molluscan Research 117 Fig. 5. Spermatozoa (Sz). A, Mature spermatozoa showing a fully elongated nucleus (Nu), acrosome (Ac) with subacrosomal space (S), proximal and distal centrioles (pc and dc, respectively) and a ring of mitochondria (Mi). B, High-power magnification of a ring of five mitochondria. C, High-power magnification of an acrosome. D, The tail (T) or flagellum consists of 9 2 arrangement of microtubules surrounded by a plasma membrane (pm). ce, Centriole. (immature and mature). Sobhon et al. (2001) reported 14 stages of spermatogenesis based on their ultrastructural study: spermatogonium, six stages of primary spermatocyte, secondary spermatocyte, four stages of spermatid and spermatozoa (immature and mature). The present study determined 13 stages of spermatogenesis in H. ovina: the spermatogonium, five stages of the primary spermatocyte, the secondary spermatocyte, five stages of the spermatid and the spermatozoa. We could not observe the diakinetic stage of the primary spermatocyte, only the metaphase spermatocyte. 118 Molluscan Research S. Singhakaew et al. Like H. asinina (Sobhon et al. 1999, 2001), the spermatogonium of H. ovina is the earliest cell with a large nucleus containing a relatively large amount of euchromatin and a prominent nucleolus and little cytoplasm with free ribosomes and mitochondria. Spermatogonia divide mitotically to give rise to primary spermatocytes that pass through four stages of the first meiotic division. These prophase cells exhibit different forms of chromatin condensation. One remarkable characteristic of LSc, ZSc and PSc of both H. ovina and H. asinina is the presence of multiple proacrosomal vesicles in the cytoplasm. These vesicles begin to be synthesised in LSc and increase in PSc (Sobhon et al. 2001). In the MSc, the vesicles still appear quite numerous, whereas the chromatin becomes highly condensed into very large chromosomes. Thus, in this primitive gastropod, the proacrosomal vesicles are synthesised early in the LSc stage, even though they may be fused to form acrosomes much later in the late spermatid stages (Young and DeMartini 1970; Hodgson and Bernard 1986; Healy ef al. 1998; Apisawetakan et al. 2000; Panasophonkul 2000; Sobhon er al. 2001). The fusion of multiple proacrosomal vesicles has been reported in several groups of vetigastropods, such as the trochids (Azevedo et al. 1985; Hodgson ef al. 1990), bivalves (Hodgson and Bernard 1986) and scaphopods (Hou and Maxwell 1991). Although the very earliest stage of proacrosomal vesicle production was not observed in the present study, the ultimate source of these vesicles is undoubtedly the Golgi complex, as demonstrated previously for other vetigastropods (Azevedo et al. 1985; Healy and Harasewych 1992) and bivalves (Hodgson and Bernard 1986). Secondary spermatocytes, similar to those of H. asinina (Sobhon et al. 2001), have heterochromatin that exhibits a checkerboard pattern. As mentioned earlier, there are five stages of spermatids in H. ovina, whereas Sobhon et al. (2001) described only four stages of spermatids in H. asinina. These classifications are based on size, chromatin granulation and condensation. Successive stages vary in size from 6 um in St, to 2 um in Stş. The major differences in the ultrastructure of developing spermatids of H. ovina and H. asinina are: (1) the formation of the acrosome begins in St, and is completed in St, for H. asinina (Sobhon et al. 2001), whereas this occurs at a later stage (St;) in H. ovina; (2) all stages of spermatids in H. ovina retain a round shape, whereas they vary from round and oval (St, s) to ellipsoid (St,) in H. asinina (Sobhon er al. 2001); (3) the acrosome of H. ovina appears slightly indented at the anterior region of the nucleus, whereas in H. asinina the acrosome only touches the nuclear envelope at the anterior end of the nucleus (Sobhon et al. 2001); and (4) the St, of H. asinina appears to be in a more advanced stage than that of H. ovina (i.e. there is a concentration of mitochondria in the posterior border of the nucleus and the centriole starts to form the axonemal complex of the tail (Sobhon er al. 2001); in H. ovina, only a concentration of mitochondria was found). There are two stages of spermatozoa (Sz, 5) in H. asinina (Apisawetakan et al. 2000; Sobhon ef al. 2001), whereas there is only one stage of spermatozoa in H. ovina. The Sz, is an immature spermatozoon with an acrosome and a short, developing tail, whereas Sz, is a mature spermatozoon. i The results of the present study show that the spermatozoa of H. ovina are very similar to those of H. asinina, except for the morphology of the sperm head. Haliotis ovina spermatozoa have a round to barrel-shaped head, whereas those of H. asinina are ellipsoid (Apisawetakan et al. 2000). In general, examination, to date, of spermatozoa in haliotid species shows that there are three types of sperm head: (/) short and globular or barrel-shaped (H. ovina; the present study); (2) ellipsoid (H. laevigata (Healy et al. 1998), H. diversicolor supertexta (Gwo et al. 1997) and H. aquatilis (Shiroya and Sakai 1992)); and (3) long and bullet-shaped (H. discus (Sakai et al. 1982; Usui, 1987), H. rufescens Male germ cells of abalone Molluscan Research 119 (Lewis et al. 1980) and H. midae (Hodgson and Foster 1992)). Consequently, the sperm nuclei can also be classified into three types: round, ellipsoid and elongated. The chromatin of H. ovina, similar to that of H. asinina (Apisawetakan et al. 2000; Sobhon et al. 2001), appears to be of a granular type. This granular pattern of chromatin condensation can also be observed in other primitive gastropods, such as trochids (Hodgson ef al. 1990; Healy 1996), scaphopods (Dufresne-Dube et al. 1993) and bivalves (Bozzo et al. 1993; Casas and Subirana 1994; Johnson et al. 1996). In H. ovina sperm, the acrosome is situated at the apex of the nucleus, as in H. asinina and also in many bivalves (Kubo 1977; Kubo and Ishikawa 1978; Apisawetakan et al. 2000; Sobhon et al. 2001). The acrosome of H. ovina has an inverted cup shape, similar to those of H. asinina, H. laevigata, H. diversicolor supertexta and H. aquatilis (Gwo et al. 1997; Healy et al. 1998; Apisawetakan et al. 2000; Shiroya and Sakai 1992). In contrast, the acrosomes of H. discus, H. rufescens and H. midae are much longer and narrower (Lewis et al. 1980; Sakai et al. 1982; Hodgson and Foster 1992). The acrosomal core consists of a crystalline-like axis embedded within a moderately dense matrix that occupies the whole subacrosomal space. The core is much shorter. The crystalline material probably consists of actin filaments and associated proteins, as reported in other molluscs (Baccetti and Afzelius 1976; Shiroya et al. 1986; Tilney et al. 1987; Healy 1989). This acrosomal core may participate in the extension of the acrosomal process during the acrosomal reaction and fertilisation (Apisawetakan et al. 2000). The tail ofa H. ovina sperm consists of five globular mitochondria located at the posterior end of the nucleus surrounding a pair of centrioles. A long axoneme stretches backwards from the distal centriole, which is surrounded by mitochondria. The axoneme core consists of 9 + 2 doublets of microtubules surrounded by a plasma membrane. This type of tail and midpiece was also observed in H. asinina (Apisawetakan et al. 2000; Sobhon et al. 2001), H. laevigata (Healy et al. 1998), H. aquatilis (Shiroya and Sakai 1992), H. diversicolor supertexta (Gwo et al. 1997) and also in several other vetigastropods (Azevedo et al. 1985; Healy 1989; Hodgson et al. 1990; Healy and Harasewych 1992) and in many bivalves (Hodgson and Bernard 1990; Bozzo et al. 1993; Casas and Subirana 1994; Healy 1996; Johnson et al. 1996), all of which reproduce by external fertilisation. It is apparent that the sperm of those molluscs with external fertilisation do not require larger quantities of energy than those of molluscs with internal fertilisation. Such sperm usually have midpieces that contain large cylindrical or helical mitochondria (Jaramillo et al. 1986; Gallardo and Garrido 1989; Amor and Durfort 1990; Sretarugsa et al. 1991; Caceres et al. 1994). Gwo et al. (1997) suggested that large species of Haliotidae usually possessed long sperm heads and small species contain short sperm heads. If this is the case, then H. ovina sperm should be classified into the latter group. Sperm of H. ovina bear certain similarities to those of H. asinina, except for the shape of the sperm head. Further studies need to be performed on the ultrastructure of haliotid sperm. We anticipate that continued comparative work in this field will help to shed further light not only on species relationships within the Haliotidae, but also on the validity of the many proposed subgenera. Acknowledgment This investigation was supported by the Thailand Research Fund BRG/04/2543. References Amor, M. J., and Durfort, M. (1990). Changes in nuclear structure during eupyrene spermatogenesis in Murex brandaris. Molecular Reproduction and Development 25, 348—356. 120 Molluscan Research S. Singhakaew et al. Apisawetakan, S., Thongkukiatkul, A., Wanichanon, C., Linthong, V., Kruatrachue, M., Upatham, E. S., Poomthong, T., and Sobhon, P. (1997). The gametogenic processes in a tropical abalone, Haliotis asinina Linnaeus. Journal of the Science Society of Thailand 23, 225-240. Apisawetakan, S., Sobhon, P, Wanichanon, C., Linthong, V. Kruatrachue, M., Upatham, E. S., Jarayabhand, P., Pumthong, T., and Nugranad, J. (2000). Ultrastructure of spermatozoa in the testis of Haliotis asinina Linnaeus. Journal of Medical and Applied Malacology 10, 101—109. Azevedo, C., Lobo-da-Cunha, A., and Oliveira, E. (1985). Ultrastructure of the spermatozoon in Gibbula umbilicalis (Gastropoda, Prosobranchia), with species reference to acrosomal formation. Journal of Submicroscopic Cytology 17, 609—614. Baccetti, B., and Afzelius, B. (1976). “The Biology of the Sperm Cell”. Monograph of Developmental Biology, Vol. 10. (Karger: Basel.) Bozzo, M. G., Ribes, E., Sagrista, E., Poquet, M., and Durfort, M. (1993). Fine structure of spermatozoa of Crassostrea gigas (Mollusca, Bivalvia). Molecular Reproduction and Development 34, 206—211. Bussarawit, S., Kawinlertwathana, P., and Nateewathana, A. (1990). Preliminary study on reproductive biology of the abalone. Kasetsart Journal (Natural Science) 24, 529-539. Caceres, C., Ribes, E., Muller, S., Cornudella, L., and Chiva, M. (1994). Characterization of chromatin-condensing protein during spermiogenesis in neogastropoda mollusc (Murex brandaris). Molecular Reproduction and Development 38, 440—452. Casas, M. T., and Subirana, J. A. (1994). Chromatin condensation and acrosome development during spermiogenesis of Ensis ensis (Mollusca, Bivalvia). Molecular Reproduction and Development 37, 223-228. Crofts, D. R. (1929). “Haliotis” Liverpool Marine Biological Committee Memoir 29. (Liverpool University Press: Liverpool.) Dufresne-Dube, J., Picheral, B., and Guerrier, P. (1993). An ultrastructure analysis of Dentalium vulgare (Mollusca, Scaphopoda) gametes with special reference to early events at fertilization. Journal of Ultrastructural Research 83, 242—257. Fallu, R. (1991). ‘Abalone Farming.’ (Blackwell Science: Oxford.) Gallardo, C. S., and Garrido, O. A. (1989). Spermiogenesis and sperm morphology in the marine gastropod Nucella crassilabrum with an account of morphometric patterns of spermatozoa variation in the family Muricidae. Invertebrate Reproduction and Development 15, 163-170. Gwo, J.-C., Chang, H.-H., and Jong, K.-J. (1997). Ultrastructure of the spermatozoa of the small abalone, Sulculus diversicolor supertexta (Mollusca, Gastropoda, Haliotidae). Journal of Submicroscopic Cytology and Pathology 29, 239-244. Healy, J. M. (1989). Ultrastructure of spermiogenesis in the gastropod Calliotropis glyptus Watson (Prosobranchia: Trochidae) with special reference to the embedded acrosome. Gamete Research 24, 9-19. Healy, J. M. (1996). Molluscan sperm ultrastructure correlation with taxonomic units within the Gastropoda, Cephalopoda and Bivalvia. In ‘Origin and Evolutionary Radiation of the Mollusca’. (Ed. J. Taylor.) pp. 99-113. (Oxford University Press: London.) Healy, J. M., and Harasewych, M. G. (1992). Spermatogenesis in Perotrochus quayanus (Fischer and Bernardi) (Gastropoda: Pleurotomariidae). The Nautilus 106, 1—14. Healy, J. M., Beames, K. P., and Barclay, D. B. (1998). Spermatozoa of the Australian ‘greenlip’ abalone (Haliotis laevigata Donovan): ultrastructure and comparison with other gastropods, especially other Haliotidae (Vetigastropoda, Mollusca). Invertebrate Reproduction and Development 34, 197—206. Hodgson, A. N., and Bernard, R. T. F. (1986). Ultrastructure of the sperm and spermatogenesis of three species of Mytilidae (Mollusca, Bivalvia). Gamete Research 15, 123-135. Hodgson, A. N., and Foster, G. G. (1992). Structure of the sperm of some South African archaeogastropods (Mollusca) from the superfamilies Haliotidae, Fissurelloidea and Trochoidea. Marine Biology 113: 89-97. Hodgson, A. N., Heller, J., and Bernard, R. T. F. (1990). Ultrastructure of the sperm and spermatogenesis in five South African species of the trochid genus Oxystele (Mollusca, Prosobranchia). Molecular Reproduction and Development 25, 263-271. Hou, S. T., and Maxwell, W. L. (1991). Ultrastructural studies of spermatogenesis in Antalis entalis (Scaphopoda, Mollusca). Philosophical Transactions of the Royal Society of London Series B 333, 101-110. Jaramillo, R., Garrio, O., and Jorquera, B. (1986). Ultrastructural analysis of spermiogenesis and sperm morphology in Cholrus giganteus (Lesson, 1928) (Prosobranchia: Muricidae). Veliger 29, 217-225. Male germ cells of abalone Molluscan Research 121 Johnson, M. J., Casse, N., and Pennecc, M. (1996). Spermatogenesis in the endosymbiont-bearing bivalve Loripes lucinalis (Veneroida: Lucinidae). Molecular Reproduction and Development 45, 476—484. Kubo, M. (1977). The formation of a temporary-acrosome in the spermatozoon of Laternula limicola (Bivalvia, Mollusca). Journal of Ultrastructural Research 61, 140—148. Kubo, M., and Ishikawa, M. (1978). Organizing process of the temporary-acrosome in spermatogenesis of the bivalve Lyonsia ventricosa. Journal of Submicroscopic Cytology 10, 411—421. Lewis, C. A., Leighton, D. L., and Vacquire, V. D. (1980). Morphology of abalone spermatozoa before and after the acrosome reaction. Journal of Ultrastructural Research 72, 39—46. Nateewathana, A., and Hylleberge, J. (1986). A survey on Thai abalone around Phuket Island and feasibility study of abalone culture in Thailand. Thai Fisheries Gazette 39, 177—190. Panasophonkul, S. (2000). “Spermatogenesis and chromatin condensation of male germ cells of a marine oyster, Saccostrea forskali Gmelin? MSc Thesis. (Mahidol University: Bangkok, Thailand.) Sakai, Y. T., Shiroya, Y., and Haino-Fukushima, K. (1982). Fine structural changes in the acrosome reaction of the Japanese abalone, Haliotis discus. Development, Growth and Differentiation 24, 531—542. Shiroya, Y., and Sakai, Y. T. (1992). Coiled filamentous structure in the sperm acrosome of Gastropoda, Sulculus aquatilis and Turbo cornutus. Development, Growth and Differentiation 34, 509—516. Shiroya, Y., Hosoya, H., Mabuchi, I., and Sakai, Y. T. (1986). Actin filament bundle in the acrosome of abalone spermatozoa. Journal of Experimental Zoology 239, 105—115. Singhagraiwan, T. (1989). “The experiment on breeding and nursing of donkey's ear abalone (Haliotis asinina Linne)? (The Eastern Marine Fisheries Development Center, Marine Fisheries Division, Department of Fisheries, Ministry of Agriculture and Cooperatives: Thailand.) Sobhon, P, Apisawetakan, S., Chanpoo, M., Wanichanon, C., Linthong, V, Thongkukiatkul, A., Jarayabhand, P., Kruatrachue, M., Upatham, E. S., and Pumthong, T. (1999). Classification of germ cells, reproductive cycle and maturation of gonad in Haliotis asinina Linnaeus. Science Asia 25, 3-21. Sobhon, P., Apisawetakan, S., Linthong, V., Pankao, V., Wanichanon, C., Meepool A., Kruatrachue, M., Upatham, E. S., and Poomthong, T. (2001). Ultrastructure of the differentiating male germ cells in Haliotis asinina Linnaeus. Invertebrate Reproduction and Development 39, 55—66. Sretarugsa, P, Ngowsiri, U., Kruatrachue, M., Sophon, P, Chavadej, J., and Upatham, E. S. (1991). Spermiogenesis in Achatina fulica as revealed by electron microscopy. Journal of Medical and Applied Malacology 3, 7-18. Takashima, F., Okuno, M., Nichimura, K., and Nomura, M. (1978). Gametogenesis and reproductive cycle in Haliotis diversicolor diversicolor Reeve. Journal of the Tokyo University of Fisheries 1, 1—8. Tilney, L. G., Fukui, Y., and Derosier, D. J. (1987). Movement ofthe actin filament bundle in Mytilus sperm: a new mechanism is proposed. Journal of Cell Biology 104, 981—993. Tomita, K. (1967). The maturation of the testis of the abalone, Haliotis discus hannai Ino, in Rebun Island, Hokkaido, Japan. Scientific Report of Hokkaido Fisheries Experimental Station 7, 1—7. Tookvinart, S., Leknim, W., Donyadol, Y., Preeda-Lampabutra, Y., and Perngmak, P. (1986). “Survey on species and distribution of abalone (Haliotis spp.) in Surajthani, Nakornsrithammaraj and Songkhla province.’ (National Institute of Coastal Aquaculture, Department of Fisheries, Ministry of Agriculture and Cooperatives: Thailand.) Usui, N. (1987). Formation of the cylindrical structure during the acrosome reaction of abalone spermatozoa. Gamete Research 16, 37-45. Young, J. S., and DeMartini, J. D. (1970). The reproductive cycle, gonadal histology and gametogenesis of the red abalone, Haliotis rufescens (Swainson). California Fish and Game 56, 298—309. http://www.publish.csiro.au/journals/mr ope exu Gv | m i E S ang Mid PN A rə b rə al. vum ida ER IT SCELUS IUE a SEU) slo aoa MT Te 14 oji Seta 5” Sce Fo WARS 1.416 ferai apa: PETS LUDERE "D —77 HORT aM Vna? uio) d BG Montis tte bons apte pn ih əə a mee : Mns W. prinio Sarl stud yəti 4 Misa) re A ex pitpücd ARM aig Gölladinad T Ve Do AU MANN ado afl Yo ly Xə irs "4 i "Sisi gts} best cA nnd hérserip i4 mn hi ATA gil 1 T^ n i e SM abs & ROTY .—. "ou ñ 4 " Epis a, J^ n Meet s , * € nO A^» — Voge. poisis I reef == : 1 xit =. uM ex vote 5j E TW: CSIRO PUBLISHING www.publish.csiro.au/journals/mr Molluscan Research, 2003, 23, 123-148 Gastropod phylogeny based on six segments from four genes representing coding or non-coding and mitochondrial or nuclear DNA D. J. Colgan^?, W E Ponder^, E. Beacham^ and J. M. Macaranas^ ^The Australian Museum, 6 College Street, Sydney, NSW 2010, Australia. BTo whom correspondence should be addressed. Email: donc@austmus.gov.au Abstract Significant differences remain between gastropod phylogenetic hypotheses based on morphological and molecular datasets. We collected additional data from three gene segments (28S rDNA expansion region DI (36 taxa plus two from GenBank), cytochrome c oxidase subunit 1 (35 species plus one from Genbank) and small nuclear RNA U2 (24 species)). These were combined with data available for the same species for histone H3 and two other segments of 28S rDNA. Analyses of these data using cladistic, maximum likelihood or Bayesian methodologies were conducted in an attempt to resolve some of the differences between current hypotheses of gastropod relationships based on morphological and molecular data. The results were of particular interest in four areas. (7) Patellogastropoda in most analyses are included in a derived clade with some Vetigastropoda. In an analysis with Nautilus as the sole outgroup, transversions weighted threefold as costly as transitions and, with third codon position data ignored, Patellogastropoda are excluded from an otherwise monophyletic Gastropoda. (2) Cocculiniformia was never monophyletic in our analyses, although this possibility is not statistically rejected. (3) Nerita, the only representative of Neritopsina in this dataset, is placed anomalously in most analyses, but is, in a few cases, shown as a sister-group to the Apogastropoda, in accord with some morphological hypotheses. (4) Heterobranchia is rarely monophyletic in our analyses owing to the variable placement of the architectonicoid Philippea. This genus, even judged by the high levels of divergence within Heterobranchia, has undergone extreme rates of substitution. The Euthyneura is invariably monophyletic and nearly always included in a clade with the valvatoidean Cornirostra as its sister-group. Additional keywords: DNA sequence, heterobranch, multiple gene, Neritopsina, patellogastropod. Introduction Recent phylogenetic investigations of gastropods have used a variety of different datasets, ranging from shell morphology (including fossils, protoconch morphology and shell structure; Bandel 1997; Fryda, 1999; Wagner 2002; for a review, see Wagner 2001), ultrastructure (Ponder and Lindberg 1997; Künz and Haszprunar 2001), development (Freeman and Lundelius 1992; van den Bigelaar 1996; van den Bigelaar and Haszprunar 1996; Lindberg and Guralnick 2001), pallial cavity structures (Haszprunar 1988a; Ponder and Lindberg 1996, 1997; Lindberg and Ponder 2001), overall morphology (Haszprunar 1988a; Ponder and Lindberg 1996, 1997; Barker 2001), mitochondrial gene order (Kurabayashi and Ueshima 2000a, 20005) and molecular sequences. These studies show general agreement for the recognition of five major groups within Recent gastropods: (7) the Patellogastropoda (or Docoglossa), being the true limpets; (2) the Vetigastropoda (trochids, haliotids, fissurellids etc.); (3) the Neritopsina (or Neritimorpha), the nerites and relatives; (4) the Caenogastropoda (most of the “mesogastropods”, including the architaenioglossan taxa, and the neogastropods); and (5) the Heterobranchia (or Heterostropha as used by some palaeontologists (Bandel 1997), © Malacological Society of Australasia 2003 10.107 1/MR03002 1323-5818/03/020123 124 Molluscan Research D. J. Colgan et al. whereas others (Ponder and Warén 1988) use Heterostropha as a paraphyletic subgroup within the Heterobranchia). The Caenogastropoda and Heterobranchia form the Apogastropoda in Ponder and Lindberg's (1997) sense. This taxon was originally introduced by Salvini-Plawen and Haszprunar (1987) to include the caenogastropods and only the basal heterobranchs, or “allogastropods”, in their usage, making it paraphyletic. An additional extant group, the Cocculiniformia (for a review, see Haszprunar 1998), is sometimes recognised and placed near the base of the gastropods. The first three groups and the Cocculiniformia comprise the paraphyletic 'archaeogastropods', whereas the first four groups comprise the paraphyletic “prosobranchs”. The Heterobranchia, as first recognised by Haszprunar (1985a), contains the euthyneurans, a grouping of two taxa previously given subclass rank (the opisthobranchs and the pulmonates), as well as a number of assumed more “primitive” members, all of which were included within the “prosobranchs” at some time. These latter groups comprise the paraphyletic *Heterostropha' (sensu Ponder and Warén 1988) or “Allogastropoda” (Haszprunar 19855; Bandel 1997). Hickman (1988) advocated restricting the usage of the name Archaeogastropoda to the Vetigastropoda, whereas some (Bandel 1982, 1997; Bandel and Geldmacher 1996) formally use the same name for a group encompassing the patellogastropods together with the vetigastropods and the Cocculiniformia (i.e. excluding the Neritopsina), based primarily on their possession of a similar larval shell. However, the artificial nature of such a grouping is recognised (Fryda 1999). In stark contrast with this view is the idea that the patellogastropods represent the sister-taxon to the rest of the gastropods, a result suggested by most recent morphological studies (Golikov and Starobogatov 1975; Graham 1985; Haszprunar 1988a; Ponder and Lindberg 1996, 1997; Sasaki 1998; Barker 2001). It was on this basis that Ponder and Lindberg (1997) introduced the Eogastropoda, to encompass the Patellogastropods and their assumed coiled ancestors, and Orthogastropoda for the remainder of the gastropods. Gastropods date from the early Cambrian, although there is considerable speculation over which of the small univalve fossils in that period represent torted gastropods or monoplacophorans. For example, Parkhaev (2001) assigns Helcionelloidea s.s. to Gastropoda, whereas members of this group are often considered to be monoplacophorans (Wen 1990; Peel 1991). j Only recently have individual molecular investigations of gastropod phylogeny included examples from nearly all critical taxa in the class (McArthur and Koop 1999; Colgan et al. 2000). The genes now available for a broadly representative set of taxa include histone H3 (Colgan et al. 2000) and four regions of the 28S rDNA gene. These are the 27 (Tillier et al. 1992, 1994; McArthur and Koop 1999), Dó (Rosenberg et al. 1997; McArthur and Koop 1999) and D4—5 (“28SA” in Colgan et al. 2000) expansion regions and a section near the D7 area including a new expansion region (“28SB” in Colgan ef al. 2000). With some notable exceptions, major groups are supported or weakly contradicted in molecular investigations. Nevertheless, the most comprehensive analysis (Colgan et al. 2000) is based on relatively short sequences (less than 900 aligned base positions), so it is not surprising that there are still disagreements between molecular and morphological (Haszprunar 1988a; Ponder and Lindberg 1997) assessments of gastropod relationships (Fig. 1). The collection of more molecular and morphological data offers the best chance of resolving such disagreements, although data from fossils are also being examined rigorously and may well assist in further elucidating gastropod phylogeny (Wagner 2001). Gastropod molecular phylogenetics Molluscan Research 125 Ischnochiton Polyplacophora Trichomya Anadara Bivalvia | Nautilus Cephalopoda Cellana Eogastropoda Notoacmaea Patellogastropoda Montfortula Perotrochus Austrocochlea Lepetodrilus Notocrater Pseudococculinidae Depressigyra Peltospiridae B Leptopoma j " E Bellamya Architaenioglossa | Campanile Nodilittorina d Serpulorbis Strombus Cypraea Cabestana Ataxocerithium Epitonium Nassarius Dicathais b Mitra Neogastropoda Conus Cancellaria Zeacumantus d PEN Heterostropha] Onchidium Salinator Hedleyoconcha Siphonaria Ophicardelus Aplysia 4 = Bullina Ophisthobranchia Coccopigya Cocculinoidea Nerita Neritopsina Vetigastropoda Orthogastropoda Apogastropoda Sorbeoconcha g o o o. o Iz] = o g D o c o g o Pulmonata Heterobranchia Euthyneura Fig. 1. Gastropod phylogeny based on the morphological analyses of Ponder and Lindberg (1997). Families not included in their analyses are-placed according to their general taxonomic classification (i.e. Cancellariidae and Mitridae in Neogastropoda, Onchidiidae, Siphonariidae, Ellobiidae and Charopidae in the Pulmonata). Taxa studied by Ponder and Lindberg (1997) but not here are pruned from the tree. Names of higher categories follow Ponder and Lindberg (1997). Differences of this topology from the Haszprunar (1988) topology are indicated. An upper case letter on a clade in the Ponder and Lindberg (1997) tree indicates that it is shown at the point specified by the corresponding lower case letter in the Haszprunar tree. In the present paper, we report analyses of an enlarged molecular dataset principally addressed to four areas of disagreement with morphologically based hypotheses, namely: (7) the position of the Patellogastropoda; (2) the monophyly and relationships of Cocculiniformia; (3) the relationships of Neritopsina; and (4) the monophyly of the Heterobranchia. Parts of two genes not previously used in overall gastropod phylogeny (cytochrome c oxidase subunit 1 (CO/) and small nuclear U2 RNA (snU2 RNA)) have been sequenced and new sequences from the 27 28S rDNA expansion region have been collected for most species studied in Colgan et al. (2000). The new genes.extend the range of gene. types used in gastropod phylogeny because they are respectively mitochondrial coding (CO1) and nuclear non-coding (snU2 RNA). The snU2 RNA is a component of the 126 Molluscan Research D. J. Colgan et al. spliceosome (Guthrie and Patterson 1988) that has previously provided useful characters for higher-level phylogenetic studies of Arthropoda (Colgan et al. 1998) and Polychaeta (Brown et al. 1999). The division of Gastropoda into Eogastropoda and Orthogastropoda has not been supported in comprehensive molecular studies to date. In Colgan et al. (2000), Patellogastropoda plus a cocculiniform limpet (Cocculinoidea: Coccopigya) was a sister-group to the remainder of the studied species. A similar topology is seen in analyses of the approximately 450 (aligned) base pairs of /8S rDNA in the compiled dataset of Harasewych and McArthur (2000), where Patellogastropoda plus a monophyletic Cocculiniformia is a sister-group to all other gastropods. In McArthur and Koop (1999), where Cocculiformia are not represented, Patellogastropoda is a sister-group to Apogastropoda. Statistically, no placement of Patellogastropoda as a sister-group to another major clade has a significantly higher likelihood than any other in the McArthur and Koop (1999) analyses. Monophyly of 'cocculiniform' limpets is one of the major areas of disagreement between the Haszprunar (19885) and Ponder and Lindberg (1996, 1997) morphological topologies. The two main constituent groups, Cocculinoidea and Lepetelloidea, comprise a monophyletic Cocculiniformia (Haszprunar 1988a; 1998) or are diphyletic (Ponder and Lindberg 1996, 1997). The molecular datasets also disagree. Cocculiniformia are diphyletic in Colgan er al.(2000), Notocrater houbricki (Lepetelloidea) being placed in Vetigastropoda and Coccopigya hispida (Cocculinoidea) with Neritopsina in accordance with Ponder and Lindberg (1997). Cocculiniformia represented by Cocculina messingi (Cocculinoidea) and N. houbricki are monophyletic (albeit with weak support) in Harasewych and McArthur (2000), using partial 78S rDNA data, and are a sister-group to Patellogastropoda. The relationship of the Neritopsina (or Neritimorpha) to the rest of the gastropods is unresolved with two main scenarios: they are either a sister-group to the vetigastropods and the apogastropods or a sister-group to the apogastropods. Fryda (1999), Bandel and Fryda (1999) and Bandel (2000) have discussed the fossil evidence for the evolution of this group since its first undoubted appearance in the Late Silurian-Devonian (428—374 million years ago; Fryda and Blodgett 2001), although earlier origins have been argued. Within the Heterobranchia, Euthyneura, comprising two of the three classes (Opisthobranchia and Pulmonata) in Thiele’s (1925, 1929-31) classification, are monophyletic in most recent molecular analyses where few taxa are used, but not in studies with larger taxonomic samples (Thollesson 1999; Dayrat et al. 2001). In morphological analyses involving Recent taxa, Heterobranchia is the sister-clade to Caenogastropoda (Haszprunar 1988a; Ponder and Lindberg 1996, 1997). In McArthur and Koop (1999), the only included heterostrophan (Valvata sp.) is a sister-taxon to Euthyneura. In Colgan et al. (2000), two representatives of the group are included. They are monophyletic but not a sister-group to Euthyneura, possibly owing to the high degree of autapomorphy in their sequences, this being particularly notable for the architectonicoid species Philippea lutea (see below). Materials and methods Materials and molecular methods The taxa used, collection and registration data of specimens, methods and tissue types used for DNA extraction are given in Colgan et al. (2000). Our naming of higher groups follows Ponder and Lindberg (1997). Specimen voucher numbers are given in Table 1. Gastropod molecular phylogenetics Molluscan Research 127 Primer pairs for U2 are given in Colgan et al. (1998). The primers for CO7 were as follows: Cox AF, CWAATCAYAAAGATATTGGAAC (41); Cox AR, AATATAVVACTTCVVGGGTGACC (725); and Cox 623R, GGTAARTYTATTGTAATAGCWCC (623). The figures in parentheses refer to the 3’ end of the oligonucleotide in the sequence of Lumbricus terrestris (Boore and Brown 1995; GenBank accession LTU24570). Primers Cox AF and AR were used together to produce a 690 bp product. Where a product was not obtained using this pair, Cox 623R was used with Cox AF to give a 626 bp product. The primers for the D/ expansion region were as follows: 288 DIF, ACCCSCTGAAYTTAAGCAT (43); 28S DIR, AACTCTCTCMTTCARAGTTC (406). Figures in parentheses refer to the 3 end of the oligonucleotide in the mouse 28S rDNA sequence (X00525; Hassouna ef al. 1984). Primer DIF was taken from Macarthur and Koop (1999) and DIR was designed from an alignment of Ascaris lumbricoides (U94751), Drosophila melanogaster (M21017, M29800) and Mus musculus (X00525) sequences. The basic polymerase chain reaction (PCR) profile was as follows: 959C for 5 min, 43—549C for 45 s, 72°C for 1 min for one cycle; 95°C for 30 s, 43—54?C for 45 s, 72°C for 1 min for 30-34 cycles; and 95°C for 30 s, 45—60°C for 45 s, 72°C for 5 min for the final cycle. The annealing temperatures varied according to the primers’ specificity for the different taxa and were usually 50-52°C for U2, 52-54°C for D1 28S rDNA and 43-45?C for CO1. To obtain PCR products from difficult samples, 20 uL GeneReleaser™ (Bioventures, Murfreesboro, TN, USA) was added to the DNA template and microwaved for 6 min. The remaining PCR mix was immediately added and cycling commenced. Reaction products were resolved on 2% agarose gels containing ethidium bromide. All single band products were purified using the QIA quickTM PCR Purification Kit (Qiagen, Venlo, The Netherlands) and, where multiple band products were obtained, the correct sized band was excised from 2% low-melt agarose in TAE buffer (0.04 M Tris , 0.001 M EDTA (pH 8.0), 0.02 M glacial acetic acid) and purified using the same kit. Products were sequenced in both directions by the Applied Biosystems (ABI*; Norwalk, CT, USA)310 DNA Sequencing System using the DyeDeoxy™ Terminator sequencing method (Big Dye?" version | or 2; ABI), according to the manufacturer's instructions. The consensus sequence for each individual was obtained by reconciling forward and reverse sequences using Sequence Navigator (ABI). The GenBank accession numbers of sequences used in Colgan et al. (2000) are AF033716-AF033794 for 2584 rRNA and 2888 rRNA and AF033675-AF033715 for H3. Nautilus pompilius CO1 data are from AF000054 (Carlini and Graves 1999). Sequences for Viviparidae (U75863) and Cornirostridae (U75862) were obtained from GenBank (McArthur and Koop 1999). The GenBank accession numbers for the new sequences are AY296815—AY296850 for CO1, AY296873—AY296909 for zə DI and AY296851— AY296872 for snU2 RNA. Phylogenetic analysis Sequences were aligned using the default values in CLUSTAL W (Thompson ef al. 1994). The CLUSTAL output was inspected and, where required, indels were edited by hand. The CO/ and U2 sequences required no modification, but some regions of the D/ segment were. altered. These were specified as regions of uncertain alignment and were not used for analyses. All bases in the 2854 and H3 alignments from Colgan et al. (2000) were used. The region of uncertain alignment in 28S rDNA B in Colgan er al. (2000) was not used. The alignments are available from the authors and as Accessory Material from the journal's website. PAUP* 4.0b10 (Swofford 2000) was used for maximum parsimony analysis by conducting heuristic searches (100 iterations with random input order) with the default options (unless otherwise stated below). All characters were assumed unordered and indels treated as unknown in all reported analyses. One hundred bootstrap pseudoreplicates (Felsenstein 1985) were conducted with 20 random input replicates at each replicate to estimate support for nodes. Bremer decay indices were calculated using AutoDecay version 3.03 (Eriksson and Wikstróm 1996) Maximum likelihood analyses of a reduced taxon set were performed using 10 random addition sequences for heuristic searches with the following settings. The number of substitution types was two, with the transition/transversion ratio estimated by maximum likelihood. Empirical nucleotide frequencies were assumed. The assumed proportion of invariable sites was zero, with a gamma discrete approximation (with shape parameter 0.5) using four rate categories. Another ML analysis was conducted with the same assumptions except that the maximum parsimony trees were used to start the analysis, branch swapping was by subtree pruning and reconnection, and five replicates were used. Quartet puzzling (100 000 steps) was performed using PAUP with the same likelihood settings, entering the transition/transversion ratio estimated during the likelihood searches by hand. A Bayesian analysis was conducted with the program MrBayes (Huelsenbeck and Ronquist 2001) using the same likelihood parameters as the maximum likelihood analysis and sampling a tree every 100 steps . J. Colgan et dl. Molluscan Research 128 ¿n en 102 “n IOO IIceoco 0Iceoco 6076072 CFII£OIN INO 80c£0c 2 LOC£0CO €£9888 NNSA £IPCV-0C4 `L 9026070 999888 VNNST1 88ISL'IN ZNINN TOT-SAH L S0ct0c o Yüct0c2 8L9S88 NNSA £0c£0c O TOTEOT O l0ceoc o LIGI ‘EPA wnayoquids 2]updumz) (ççgl ‘Aqiamos) snypuiipoqns snjupumn2poz (9061 319qoyl) sisuaSunpSupns ipnoi n&uv]Iag (OPS *dno[o1e19) vpionjiad DiuodojdoT €68] DADDY DSOJU2IUDA]D DIIN (1SgI 'surpy ^y) vmod ropju2020418ny- €96[ “ƏKpg sppiu sni[20410424 8861 ‘UvaTOW sisuoonf snjipojodoT (PEST prune wz Aond) psoan. p[nii0f1uo]i $661 '[oKmosere] P ULITIN 2/214Q1101| 4210420]0N 9861 “IIENSTEIN vpidsiy paSıdoəəo?) 6861 Əuonoq 29 uo1eA, snyyıqo/3 vIASIssaidaq ` (9181 ‘SPOOM `L) 1p4ord pəptuopo)o0N (Z081 'uəl[oH) v2r42s0uma) pup]]9O OSL OOPS] snipjnarqo42s SNJUNDN (6181 5io1eure T) pgnsanp DAWOYIIAT (6€81 'so&eusoq) pzzədbu) p.upppuy (0F81 “qioAoS) sipajsnp (pisppaoutosp) uogpf2ouwosp ovpipruedure;) *eopropruedure;) eeprue|[neg “vəprornə? _BSSO[BZOIUDLIOAN, guSuoooaqios ƏPDpLIPdIAIA “pƏpiouip||ndury eveprioudo[o;) *eopro1oqdo[oÁ7) ,?sso[SoruoejTqo1y, vpodouise3ouov;) ƏBPDHƏN "vopIo)HoN vursdoj.IoN. oepiqoo1[ *voprouoo1] oepiLreurojoJno[q *voproeurojono[d oepi[ripojodo' *eoprojripojodo 7 ovpi[[o1nssr,q *eopro[[o1nsst,q vpodonseSnoA əpprurnəəoəopnəsq *eaprourno2020pnosq əpprurnəəo?) “pəproulinəəo?) eIULIOJIUI[n220/) oepridsoj[oq *eopro[ejduiooN VdOdOULSVOOHLAO 9vpirjo'T “əpproəpuroV *eur[[oeN 9epi][o1eq *vopro[[ojeq LUPA epodoujseso[[ojed YAOdOULSVDOA VdOdOULSVO VO9pIO[HnEN “PpineN epro[pneN. VCOdO' IVHdHO ƏuptInÁN ‘BPIAN oeprory *eoproory eydioworad VIATVAIG ovpruojrqoouuosp *euruojooutgos vVje»uo[ooN V*IOHdOOV Id XTIOd vjep BUISSIN "ou uəoturoods səroəds Suidnoi8 1op10-19Q81H SHEJƏp əqənoA pue sərəəds *uoneonisse[) "1 AQEL 129 Molluscan Research Gastropod molecular phylogenetics X9) ən) ur poouo19joa1 pue 1oquinu uorssoooe Aq porjroods oue xueguor) woy soouonbog "UOXE) 1Eü) 10J pa1oos jou sem ouo3 pəlyiəəds əy} Jey} sojeorpur uuin[oo gyep SursstJA, ou ur AUD uy `(uonoəlloO egroruung, ‘L ‘OA 'uojigurgseA “O1SIH PLINN JO UMN [RUOTEN 'JANS feuegsug *umosnjq pue[suoonQ) “WO :uojgur[oAA pue eə9Z MON JO umesnjq [euoneN “ZNINN "SAVOTTOy SB OI suomnerAa1qqe IIO “üməsniy ueresny oY} JO U01909 [eorgo[ooe[eur oy} ur uoneisrao1 nəq) Aq porjnuoptr o1e səjdures “pərşiəəds əsrAvtəulo sso[u(] i —— —””— UT UU o e ¿n ¿n cn e en en e ¿n ¿n 89090IN INO I£ce0c o 0£c€0c O 6CC€0c O 8tc€0c O LTTEOT O 9cce0c O Scceoc o vcceoc O £cce0c O ccceoc O Icceoc o 0cceoc o 61c£0cO 81c£0cO LITEOT O 9ITEOT O SITEOT O tITEOT O £Iceoc O TITEOT'O (¿S81 9JJI9Jd) jap vupuo2otojparr (IZ81 ‘oessni94) snjpuao snjapavonjdo) (££81 'preuimer) əy Aonğ)) poipurjez piupuoydiş (8481 *oxoeuos) ppyyos 40jmuipg "ds wunipipoug TESI 'p1eumep X Aon) pupijní yə pisiidy (S781 '&e15) pour vung (Cc81 »joreureT) zəy? pəddilmq (v $6] *uo1eseT) ppzonyyəd najsoa.107) 8S] *snoeuur] sau snuo2 6F81 ‘Aqiamos pyoynpun (p421dpp«g) vi40]]22uD7) TIS] 5oreure T puzi2umono DANN (1621 “ut|əuuD) DJ1q4O sinijpoiq (6781 “Iddıyruq ur səyunq) mpapuoanq (pt4pp24b21]d) SNLIDSSDN (C681 's9q1o4) wnunisaynl “yə uniuojidg "ds wnnpia2oxpb]y 1181 “məq 142]8uads nupjsaqv?) SSLI 'sneouur] snpnuup (ptubləuoyy) vavidty 86,1 'snoeuurT snupnin] (1xo4nui0u07)) squio4Jg "ds siq4ojndaag (9781 eID) pyprospftun vuronjipoN oepidoien? ‘eroydoyeurwoj Ais Əepitqollq *voprorqo[a eepirreuoudis *eoprorreuoudts əvpıroqrqdury ‘eeprojoqiydury eepriprqouo ?jeuouiing eeprisA[dy *eopidsejoN. eepriunepÁH “səpidsereudə?) gruougiqoulstudo piməuKunq erouerqo1919H evproruojoejrqory “eƏpiooruo1o91Iuo1y 9eprsoiru10;) *eoprojeA[eA. “pudonsoə)əH, vrqoueiqo19]19H əppıuo?) *eaprouo?) eepiLv[[o2ue;) *eopronre[[ooue;) 9EepEDIA eeprounjq oepruroong epodo.nsE30əN eqouooooqiog evpriuojid *eoprourque oeprsdorgiuo;) *eopro1oudig ,Dsso[Souajq, eqouooooqiog eepi[[ougs *eoprouuom voprooe1dÁ) 'oeproe1dA? dEPIQUIONS *eoproquiodg SEPNOULIIA *?opIOJ9ULIQA 9epruLiogrT *eoprourogrT epodonseSosdAH 130 Molluscan Research D. J. Colgan et al. along a 100 000 step Markov chain. Only two simultaneous chains were run, owing to computer memory restrictions, and the first 43 300 steps were discarded because convergence to an area of stable likelihood did not occur until this time. All analyses listed below, except (ix)-(xi), were conducted with maximum parsimony. (i All data, excluding areas of uncertain alignment in the 285 rDNA. (ii) All data, excluding areas of uncertain alignment in the 28S rDNA and third codon positions. (iii) All data, excluding areas of uncertain alignment in the 28S rDNA, transversions weighted threefold as costly as transitions. (iv) All data, excluding areas of uncertain alignment in the 28S rDNA and third position data, transversions weighted threefold as costly as transitions. (v) All data, excluding areas of uncertain alignment in the 28S rDNA, transversions weighted threefold as costly as transitions; Nautilus was used as the only outgroup. (vi) All data, excluding areas of uncertain alignment in the 28S rDNA and third codon position data, transversions weighted threefold as costly as transitions; Nautilus was used as the only outgroup. (vii) All data, excluding areas of uncertain alignment in the 28S rDNA, and excluding Philippea lutea. (viii) All data, excluding areas of uncertain alignment in the 28S rDNA and third codon position data, and excluding Philippea lutea. (ix) A maximum likelihood analysis of a reduced dataset using random taxon addition and excluding areas of uncertain alignment in the 28S rDNA. (x) A maximum likelihood analysis using maximum parsimony starting trees excluding areas of uncertain alignment in the 285 rDNA. (xi) A Bayesian analysis of all data excluding areas of uncertain alignment. Although analyses of individual genes are not reported in detail, they were conducted to confirm that there was no cross-contamination between sequences within this study or the inclusion of sequences from other material treated in this laboratory. MacCLADE (Maddison and Maddison 1992) was used to set character types and substitution weights and to compare trees with the *winning-sites test” (Prager and Wilson 1988) using the “compare two tree files’ option. Files containing all trees from each analysis were used. One tree (or set of trees) was preferred to another if the number of characters for which it had less steps than the alternative tree(s) was significantly greater than the number of such characters for the alternative. Probabilities were estimated using a two-tailed normal approximation. Results The number of aligned bases excluding the areas of uncertain alignment, the number of variable characters and the number of parsimony informative characters (in order for the individual gene segments) were: D] 28S rDNA, 317, 154 and 113 respectively; 2854, 255, 111 and 71 respectively; 28SB, 256, 73 and 47 respectively; CO1, 567, 400 and 318 respectively; histone H3 274, 121 and 107 respectively; and snU2 RNA 131, 56 and 41 respectively. J The mean GC content of the studied species for s U2 RNA was lower (46.11%) than the other non-coding RNAs and notable differences were observed between the three 28S rDNA segments: 57.35% for the D/ region, 50.67% for 28SA and 48.67% for the 2888 re- gion. The GC content was 58.42% for histone H3. The percentage of GC in CO1 was very low (38.36%), but increased to 45.37% with the exclusion of third-position bases, consist- ent with the usual pattern for animal mitochondrial coding DNA (Wirth et al. 1999). Chi-square tests suggest that the percentage nucleotide composition was homogeneous within the studied species for H3 (P = 0.571), U2 (P = 1) and all 28S segments (P = 1 for D1 28S rDNA, 2834 and 28SB), but was significantly inhomogeneous for CO! (P < 0.001). This inhomogeneity was due to third codon position data; when these data were omitted, the hypothesis of compositional homogeneity was not rejected (P = 0.999). The average transition to transversion ratios for the six gene segments based on the Kimura two-parameter distance, and ignoring the areas of uncertain alignment and pairwise Gastropod molecular phylogenetics Molluscan Research 131 comparisons without transversions, were: 1.792 + 1.560 for U2 (range range 0-13.31); 1.260 + 0.600 for H3 (range 0.348—5.260); 1.276 + 0.874 for DI 28S rDNA (range 0—12.036); 3.017 + 2.790 for 28SA (range 0—26.065); 2.555 + 3.602 for 28SB (range 0-18.622); and 1.063 + 0.306 for CO1 (range 0.456—-2.345). Ratios were also examined omitting the third position of H3 and COI. For H3, the average was 1.980 + 1.028 (range 0.279—9.793). For CO1, the average was 1.372 + 1.376 (range 0.279—21.186). Incongruence length difference analysis for the data set for analysis (7) with 100 replicates returned a probability of 0.27 that the phylogenetic implications of the various gene segments do not differ. For other analyses, the probabilities are: (ii) 0.99; (iii) 0.99; (iv) 0.01; (v) 0.99; (vi) 0.99; (vii) 0.01; and (viii) 0.62. Summaries of maximum parsimony analyses for individual genes are given in Table 2. Generally, few clades receive bootstrap support greater than 50% in these analyses, so the results are not discussed in detail. Details of the various analyses, including the number of maximum parsimony trees, the consistency index, the length of the trees and the minimum length of trees satisfying the Ponder and Lindberg (1997) topology, are presented in Table 3. The maximum parsimony trees for analyses (1), (ii), (iii), (vi) and (ix) are presented in Figs 2-6. The results for other analyses are compared with these figures below. Their bootstrap supported clades are indicated in Table 2. Figure 2 shows one of two maximum parsimony trees for analysis (i). The topology of the second differs only above point A on the figure (with bolded branches found in both trees). The topology with Philippea removed (analysis (vii)) is the same as for analysis (i) (omitting Philippea) at nodes below the arrows on Fig. 2. The topologies differ above the arrowheads, notably in that Opisthobranchia and Pulmonata are monophyletic sister-groups in analysis (vii). Bootstrap supported clades are the same as for analysis (7) and have similar percentages. The topology for analysis (viii) is the same as for analysis (ii) (with the removal of Philippea) at nodes below the arrow on Fig. 3, except that Patellogastropoda is a sister-group to the clade comprising (Montfortula, Austrocochlea, Notocrater and Lepetodrilus). Branches above the arrow seen in the strict consensus for analysis (ii) and for analysis (viii) are indicated in bold. Figure 4 shows one of the two maximum parsimony trees for analysis (iii). The second tree differed in the placement of some taxa in the area between Nodilittorina and Bellamya, with Nerita shown in the equivalent position to Bellamya in the first tree. In analysis (v) Nautilus, Coccopigya and the other gastropods form a basal trichotomy. The gastropods are then divided into: (a) Euthyneura; and (b) the remaining taxa. Group (b) is further divided: (b.1) Leptopoma and Campanile; and (b.2) other Caenogastropoda, Vetigastropoda, Patellogastropoda and Philippea. The topology for analysis (iv) is similar to that for analysis (vi). Addition of the other outgroup taxa in analysis (iv) places the root of the tree at the position marked by A in Fig. 5. The maximum likelihood topology of the reduced taxon set is shown in Fig. 6. The estimated transition/transversion ratio on which this is based was 1.64. The estimated ratio for analysis (x) was 1.62. Analysis (x) produced two trees differing only in some placements within Caenogastropoda. The primary dichotomy lay between a group comprising Patellogastropoda, Notocrater and the Vetigastropoda except Perotrochus and the other gastropods. Caenogastropoda was monophyletic except that it anomalously included Nerita and that Leptopoma was grouped with Nautilus plus Philippea as a sister-group to a clade comprising Coccopigya, Pleurotomaria and Depressigyra. Euthyneura was monophyletic Molluscan Research . J. Colgan et al. 132 -eyguouu[nq [nq fenouAqing “Ing “prqəubiqorəşəH 39H “epodonseSofrəyeq Wd ‘Leag ATH "A[&ydouour qeəoidiəər ojeorpur Apussooou jou soop ,snjd, Áq exe} jo Surdnoi8 oq L `u)8uə[ pue xəpur Aouojsrsuoo 119} *səən Auotursred uinturxeur jo 1oqumu əy} moys suumn[oo oq Eme a ae ə o üə — IA cus UR ll U MN — — eer TSS 1170 sLI€ £$C0 ESI £0€0 OL 0090 $06 9£c0 LTO | Sce0 dul 0690 coc Sy9'0 Tes L6v'0 %18 pruipuoydiş ydəəxə yüzi “9462 MF “9469 vaddyiygq snjd 9I Arg snjd ing “0616 pəddiyiiya snjd Arg ‘%ps (vaddy iy “pduoyoluy) SES (ma&S1sso4doqg *n31d0220;)) %LS ni4puoijdig 1dooxo [nq “0484 ni4puoudis ydəəxə € INA ‘%1 L ME “0609 (pu(8issəidəq *D481d02207)) “oy (puuodoldə'] ‘snjunoy) ‘%16 voddipigq sn|d Arg %15 Qunipipoug) 401pui[pS) $516 pi/ouooo4ə]pəl] oz 1dooxo şəH “046€ (əl1updunpoO ‘puodojdaT) $o498 (bəyifəooodisny “pmyuofiuoyy) 19489 Wd oze ZL (snj4pojodo'T 49]42010N| *ro]2020418ny “oymyaofruoyy) “oğ (snquioug ‘vuodordaT) T 969€ Wd % gs (snyəpupəndo ‘unipiyouo “byouooodəlpəH) 10662 (snjopavonjdo) unipioug) ‘%p9 (snuoO ‘snquong) ‘%78 Quniuojid7 “siqaondaəg) teL (uni14o20xv]y *uniuonds siq40]ndaag ‘vuiomjpoN) :% S (nplapojodo7] *vopioo2041sny ) ç tos c9 (snpipojadoT *vapo02041sny “pmyaofluoyy) :%001 Wd ‘ATS SUNON snjd epodoujser) 8 | %06 (tununaəooxp]y “tuntuold:] *siq40]ndaog “puluoyunpoN)) "906001 Wd 009 Sç (Goypuypg unipiouQ) ‘%85 (Pany untuond:] *siq40]nda28) :o5€8 yüz :%001 Wd :%0/ ATA %95 (n81d02207) *pui|ng) ‘%78 (snjopaponjdo “uünipiqouq) 1001 “o4çç p43)doəəo?) su|d MJ “0462 081d0220;) pue nso411407) sn|d MA “0406 (suo?) *snquio4Ig) S991] ‘ON sopve[o poiioddns-deujsyoog sjuoauidos [enprAimpur 10} sj[nso.1 jo A1£urums 'ZəlquL (uonisod putu) ou) 709 102 en (uonisod pitu) ou) £H £H (peurquioo) SQZ asc VS8C Id S8c sısÁjeuy 133 Molluscan Research Gastropod molecular phylogenetics 'pojonpuoo sosÁ[eue ou JO s|I81Əp 10} 1X9} 0} Təyəşl "10070 > dx PUB 600 > dx "utunjoo ,d}BIAOP [EUUON, ən) ul `səən S1əqpur] pue 1opuog TE ur JoJOYs siə]opreuo Jo 1oquinu sy} Aq pəətofloy SLIEP Jey} 10] ASo|odo] S1oqpurT pue Jopuog oy} SurÁJsnus sooy pje ur uer sisA[eue porjroods oy} JO səən TE ur 1ə3100$ s1ojoe1eqo Jo 1oquinu oq SMOYS uuin[oo 189] sojs-GutuutA SU L £9'I FII NA ISSS L6T 0 I ICT 101 öll 9666 Loc 0 9 1971 vL $6 S9çç 86c0 r Ara It $9 EPIT 9607 £ST0 01 ***|0€ € I8 yol IIPS SLTS 8670 Y xxx09 p LC £L ESET LOET Orr'0 5 s SET TSI LLI £rv6 0876 S8C0 I ***VL `S 9c L8 ELTH telr Sep 0 [4! 2.04 LTI DIG CVS0T 6SSOI LLCO I ***9F t ST 89 LET ELC LrF 0 r ***8Ç€ S c6 6LI £096 cess 8670 [4 9]eIA9p [RULION, Sois SuruurA uyöuəi S1oqpurT/1epuod yuq I2 SƏƏLL SƏSAIEUE [EnpIAIDUI 10] SINSHEIS "€ AQEL eIULIOJIUI[R2207) eIQgouelqolo9]9H epodouseSou1iQ/epodonsesoq (na) (ua) Ga) (A) (a) (u) (u) (2) sisÁ[euy 134 Molluscan Research D. J. Colgan et al. Ischnochiton 100 Trichomya SO cnanges 15 Anadara Nautilus 3 Philippea Heterostropha - part Leptopoma Architaenioglossa - part 4 56 Perotrochus Vetigastropoda - part 7 Depressigyra Peltospiridae 5 Coccopigya Cocculinoidea Nerita Neritidae 3 Cypraea Zeacumantus t Bellamya Architaenioglossa — part | $ 6 Campanile I 4 Strombus S ç Y Cabestana ° 3 Ataxocerithium o A baz Cancellaria* m 4 Nodilittorina S 3 Dicathais* o 0 Serpulorbis o 4 2 Epitonium ó Nassarius* Mitra* 2 Conus* Cornirostra Heterostropha —part 3 Onchidium S Hedleyoconcha 97 zs Siphonaria Pulmonata E 9 8 Ophicardelus £ Pog Salinator 3 Bullina i C 20 Aplysia Opisthobranchia 100 Cellana 62 Notoacmaea Patellogastropoda 3 Notocrater Pseudococculinidae 3 Montfortula 2 Austrocochlea ba Vetigastropoda- part 3 Lepetodrilus Fig. 2. One of two maximum parsimony trees for all data, excluding areas of uncertain alignment (analysis (i)). The topology of both is identical below point A. The arrowheads indicate topological similarities with analysis (vii) detailed in the text. Bold branches above A are also seen in the strict consensus of analysis (i) and analysis (vii). Bootstrap percentages above 50% are shown above a branch. Bremer decay indices are shown below the branch. Branches without figures have indices of 1. Asterisks on genus names indicate membership of Neogastropoda. Note that the brackets on the right-hand side do not necessarily show monophyletic clades. with Cornirostra its sister-group. The clades with puzzling support more than 5096 were: (Nautilus, Philippea) 55%; ((Irichomya, Anadara) 8496; (Cellana, Notoacmaea) 5196; ((Austrocochlea, Lepetodrilus) 68%; Montfortula) 61%; ((Perotrochus, Depressigyra) 65%; Coccopigya) 56%; (Ataxocerithium, Cancellaria) 60%; Pulmonata 52%; Ophisthobranchia (here Aplysia, Bullina) 69%; (Euthyneura plus Cornirostra) 60%. The topology of the Bayesian analysis (xi) with /schnochiton as the outgroup is broadly similar to the results of other analyses, albeit with many instances of high “posterior probabilities" of clades that are unexpected on morphological grounds but that are shown Gastropod molecular phylogenetics Molluscan Research 135 Ischnochiton Trichomya 18 Anadara Nautilus Coccopigya Cocculinoidea Leptopoma Architaenioglossa — part Campanile Bellamya Architaenioglossa -part Nodilittorina 60 Serpulorbis 2 3 Epitonium 2 Ataxocerithium 4 Cypraea 2 Zeacumantus 0 Mitra* Dicathais* Conus* Nassarius* Strombus Cabestana Cancellaria* Cornirostra Heterostropha -part 53 Onchidium Ophicardelus Salinator Pulmonata Hedleyoconcha 14 Siphonaria Bullina Opisthobranchia 100 — Cellana = Nice Patellogastropoda] 2 0 Philippea Heterostropha - part Montfortula h 92| 68 yos Vetigastropoda -part | a Notocrater Pseudococculinidae 4 Lepetodrilus i E Perotrochus MOET TELE part] 2 Depressigyra Peltospiridae Nerita Neritidae Caenogastropoda 93 Euthyneura Fig.3. The strict consensus tree for analysis (ii), all data excluding areas of uncertain alignment and third positions. Bootstrap percentages above 50% are shown above a branch. Bremer decay indices are shown below the branch. Branches without figures have indices of 1. Moving branch A to branch a makes the topology of this analysis identical to that for analysis (viii) at points below the arrowhead. All Euthyneura except Siphomaria are included in a clade with bootstrap support of 54%, although not found in the maximum parsimony tree. Asterisks on genus names indicate membership of Neogastropoda. Note that the brackets on the righ-hand side do not necessarily show monophyletic clades. in parsimony or maximum likelihood analyses without great support. For instance, the anomalous pairing of Nerita with Cypraea has a posterior probability of 0.99; the pairing of Strombus with Nassarius in a group including most Neogastropods has a probability of 0.70 and the pairing of the remaining neogastropod Cancellaria with Ataxocerithium has a probability of 0.99. Patellogastropoda was monophyletic with high bootstrap support in all parsimony analyses, but was shown in a basal position only in analysis (vi) (Fig. 5). In other analyses, the group was included in a clade with some vetigastropods and Notocrater (analyses (i), (vii); Fig. 2), with this group plus Depressigyra (analysis (x)), paired with Philippea (analysis (iij) as a sister-group to a clade comprising other heterobranchs, caenogastropods, Coccopigya, Nerita and Nautilus (Fig. 4); or paired as sisters with Philippea (analyses (ii), (v); Fig. 3), Nautilus (analysis (iv)) or Notocrater, Montfortula, 136 Molluscan Research D. J. Colgan et al. Ischnochiton 100 Trichomya 74 Anadara 9 Depressigyra Peltospiridae 9 Pleurotomaria 9| 94 Montfortula i 31 57 Austrocochlea ..... 7 Lepetodrilus 10 Notocrater Pseudococculinidae Nautilus dö 12 Coccopigya Cocculinoidea Cornirostra Heterostropha- part Onchidium 73 54] 8 Hedleyoconcha S 22 3 Salinator Pulmonata - part ° 2| 2 Ophicardelus > 2 98 [3 əə Opisthobranchia = 33 Sibhonaria Pulmonata — part | !!! Leptopoma Architaenioglossa - part 2 Campanile t Nodilittorina ° Serpulorbis A 4 Til? Epitonium U Nassarius* e 9 Strombus 9 4 12 Cabestana E. Ataxocerithium Oo 2İ 2 Cancellaria” e A 2 Dicathais* ° 5 Conus* o Zeacumantus 7 7 Mitra” Nerita Neritidae 12 Cypraea Caenogastropoda- part 7 Bellamya Architaenioglossa — pan | 100 52 [7439 Not Apalan Patellogastropoda | 14 Philippea Heterostropha — part — 100 changes Fig. 4. One of two maximum parsimony trees for all data, excluding areas of uncertain alignment with transition/transversion weighting (analysis (iii)). Bootstrap percentages above 50% are shown above a branch. Asterisks on genus names indicate membership of Neogastropoda. Note that the brackets on the right-hand side do not necessarily show monophyletic clades. Austrocochlea and Lepetodrilus (analysis (viii)) within a grouping of all vetigastropods and hot-vent taxa except Coccopigya. In no case did the pairing of Patellogastropoda as sister-groups with any other taxon receive bootstrap support greater than 50%. In analysis (xi), Patellogastropoda is a sister-group to the grouping of all vetigastropods and hot-vent taxa except Coccopigya with a posterior probability of 0.52. Imposing the constraint that Eogastropoda and Orthogastropoda were monophyletic sister-groups required 32 more steps (P ~ 0.11 using the winning-sites test). Gastropod molecular phylogenetics Molluscan Research 137 Nautilus 100 Cellana 125 Notoacmaea Montfortula Vetigastropoda — part | 94 Austrocochlea 55| 26 Notocrater Pseudococculinidae 3 Lepetodrilus Vetigastropoda -part Depressigyra Peltospiridae Perotrochus Vetigastropoda - part Coccopigya Cocculinoidea Cornirostra Philippea Heterostropha 60 Onchidium Ophicardelus Hedleyoconcha = 13 Salinator Bullina = 98 i 13 Aplysia Opisthobranchia 23 Siphonaria Pulmonata- part B Leptopoma Architaenioglossa Bellamya f Nodilittorina 70 Serpulorbis 5 LS3T 7 Epitonium 3 Ataxocerithium Strombus Cypraea Cabestana Zeacumantus Nassarius* A => Dicathais* 3 Mitra* Conus* Cancellaria* Campanile Nerita Neritidae Patellogastropoda Pulmonata- part Euthyneura Caenogastropoda Fig. 5. The strict consensus tree for analysis (vi), all data excluding areas of uncertain alignment and third positions, with Nautilus as the only outgroup and transition/transversion weighting. Bootstrap percentages above 50% are shown above `a branch. Bremer decay indices are shown below the branch. Branches without figures have indices of 1. “A” indicates the root when the other outgroup taxa are added (analysis (iv)). The other differences between the strict consensus trees of these two analyses is that branch B moves to b and branch C to c in analysis (iv). Asterisks on genus names indicate membership of Neogastropoda. Note that the brackets on the right-hand side do not necessarily show monophyletic clades. Cocculiniformia was never monophyletic in our analyses. Coccopigya was a sister-group to Euthyneura plus Cornirostra in analyses (ii), (iv), (vi), (viii), (ix), (x) and (xi), with significant bootstrap support in analyses (i7) and (viii) and a posterior probability of 100 in analysis (xi). It was a sister-group to Nautilus in analysis (iii) and Perotrochus plus Depressigyra in analysis (i) and analysis (vii) and formed one member of a basal trichotomy with Nautilus in analysis (v). Notocrater was always closely associated with a group of Vetigastropoda (Montfortula, Austrocochlea and Lepetodrilus). This group of four taxa was monophyletic with high support in all analyses except analyses (i) and (vii), where it also included the Patellogastropoda (as a sister-group to Notocrater). Even in analyses (i) and (vii), although the group of four was not shown in maximum parsimony trees, it received 138 Molluscan Research D. J. Colgan et al. Ischnochiton 62 Trichomya Philippea Heterostropha- part 71 Cellana Notoacmaea Patellogastropoda Coccopigya S Cornirostra Heterostropha- part 3 zu 81 Onchidium ° 57 Hedleyoconcha Pulmonata | = Aplysia Opisthobranchia | 5 | Nerita ut m Cypraea ” Neritidae œ Nodilittorina ° Ataxocerithium o. Nassarius” £ Conus* E Leptopoma ^ 5 5 Bellamya Architaenioglossa | S Campanile S Montfortula Ç Austrocochlea Vetigastropoda- part | ° Notocrater Pseudococculinidae ule Perotrochus Vetigastropoda- part Depressigyra Peltospiridae Nautilus — 0.05 substitutions/site Fig. 6. Maximum likelihood tree of a reduced taxon set, including all data, excluding areas of uncertain alignment. Asterisks on species names indicate membership of Neogastropoda. The figures above branches indicate puzzling support of more than 5096: Nodilittorina plus Nassarius (5490), Ataxocerithium plus Conus (60%) and Montfortula plus Austrocochlea (64%) received puzzling support more than 50%, despite not appearing as clades in the likelihood tree. Note that the brackets on the right-hand side do not necessarily show monophyletic clades. bootstrap support of more than 60%. Relationships within the group varied, with Notocrater being found as a sister-group to each of the three other members in at least one analysis. Imposing the constraint that Cocculiniformia is monophyletic required 18 more steps (P = 0.10 using the winning-sites test). Euthyneura was monophyletic in all analyses with high bootstrap support (Figs 2-6; Table 2). Pulmonata and Opisthobranchia were both monophyletic only in analyses (vii), (x) and (xi), here with high posterior probabilities for each clade. In some other analyses (i and ii), Opisthobranchia was paraphyletic with respect to Pulmonata, but the groups were intermingled in analyses (iii), (iv), (v), (vi) and (vii), with bootstrap support for a clade of all Euthyneura except Siphonaria in analysis (iv) (8596) and analysis (vi) (80%). In all analyses, Cornirostra was closely associated with Euthyneura, being shown with high bootstrap support or posterior probability as the sister-group to this clade except for analyses (iv) and (vi). In these analyses, Cornirostra was paired with Philippea as a sister-group to Euthyneura to form a monophyletic Heterobranchia, with bootstrap support of 67% in analysis (iv). Generally, Heterobranchia was disrupted by the association (not bootstrap supported) of Philippea with other taxa: Nautilus in analyses (i) and (x); and Patellogastropoda in analyses (ii), (iii) and (v). Imposing the constraint that Heterobranchia is monophyletic required 23 more steps (P = 0.22 using the winning-sites test). The genetic divergence of the Heterobranchia as measured by the distance of terminals from the root in maximum parsimony analyses is striking, although less pronounced in likelihood analysis. Gastropod molecular phylogenetics Molluscan Research 139 In each analysis except analyses (v) and (vii), the great majority of the Caenogastropoda and Heterobranchia grouped to form a recognisable but weakly supported *apogastropodan clade’. Examples of the exclusion of Philippea from this are listed above. Lepotopoma was excluded in analyses (i) and (vii). Unexpectedly included taxa are Nerita in analysis (i), (iv), (viii), (ix) and (x), Nautilus and Coccopigya in analyses (ii) and (iii), and Coccopigya in analysis (vi). In analysis (v), Euthyneura plus Cornirostra was a sister-group to all other gastropods except Coccopigya. In analysis (vii), five unexpected taxa (Nautilus, Nerita, Pterotrochus, Depressigyra and Coccopigya) disrupt the *apogastropod' lineage. None of the unexpected inclusions or exclusions had bootstrap support greater than 5096. Discussion The present analyses used data from six gene segments from four loci to address four ofthe major differences between morphological (Haszprunar 1988a; Ponder and Lindberg 1997) and molecular (Tillier et al. 1992, 1994; Rosenberg et al. 1997; McArthur and Koop 1999; Colgan et al. 2000; Harasewych and McArthur 2000) understanding of gastropod relationships. These were: (1) the position of Patellogastropoda; (2) the relationships of members of Cocculiniformia; (3) the relationships of members of Neritopsina; and (4) the monophyly of Heterobranchia. From the molecular perspective, the division of Gastropoda into Eogastropoda and Orthogastropoda remains an open question, despite the addition of more data in the present paper. The position of Patellogastropoda varies in present analyses as in previous molecular investigations (Rosenberg et al. 1997; McArthur and Koop 1999; Colgan et al. 2000). Long-branch length attraction (Felsenstein 1978; Lyons-Weiler and Hoelzer 1997; Siddall and Whiting 1999; Stiller and Hall 1999; Philippe and Germot 2000) may be a possible explanation for the pairing of Patellogastropoda with groups that would be unexpected on morphological grounds. Morphological support for the eogastropod/orthogastropod division is strong, but not incontestable. Character states supporting the division of Gastropoda into Eogastropoda and Orthogastropoda should be synapomorphic in the latter group and plesiomorphic or having an autapomorphy not derived from the state in Orthogastropoda in Patellogastropoda. The potential number of such characteristics is impressive. Fifteen changed state between the nodes uniting all gastropods and all orthogastropods in the Ponder and Lindberg (1997) topology. However, five have consistency indices less than 0.4 and one (adult operculum) is not applicable to Patellogastropoda, although they possess a larval operculum. Among the other nine characteristics, only three (18, the presence of a hypobranchial gland; 60, the bending plane of the radula; and 98, statocyst position) have possibly derived states in all major orthogastropod clades. In the remaining six characteristics, some or most Vetigastropoda share the same derived state as Patellogastropoda or have the symplesiomorphic state for Gastropoda. The situation with some characteristics is ambiguous; for example, with the hypobranchial gland. Sasaki (1998; his character 25) confirmed that a hypobranchial gland is absent in patellogastropods and noted its absence in Nautilus (although the nidamental glands may be homologous; Salvini-Plawen 1990). Sasaki (1998) also stated that the gland is absent in Fissurellidae on the basis of his own observations and Neomphalidae (fide McLean 1981), although it was recorded as present in these taxa by Ponder and Lindberg (1997). Fretter and Graham (1962) recorded a hypobranchial gland in Diodora and implied its presence in Emarginula (both fissurellids). Haszprunar (19895) observed a small hypobranchial gland in males of Pseudorimula (a fissurellid) but not in females. Israelsson 140 Molluscan Research D. J. Colgan et al. (1998) recorded a hypobranchial gland in Pachydermia (related to Neomphalus) and one has also been recorded in Melanodrymia (Haszprunar 19894), but the presence or absence of one is not noted in Neomphalus by Fretter et al. (1981). Sasaki (1998) bases his statement regarding the absence of a hypobranchial gland on McLean (1981), who says that a thick folded gland, as seen in haliotids and trochids, is absent but that there are *...scattered subepithelial gland cells ... comparable to .... the Fissurellidae in which gland cells are present in the mantle skirt but do not form a discrete organ with a folded surface”. Given the presence of an undisputed gland in closely related taxa, what is present in Neomphalus is certainly a reduced hypobranchial gland, similar reductions being seen in many other gastropods, even within genera. These observations do not discount the possibility that the hypobranchial glands are secondarily absent in patellogastropods. They are absent in Scaphopoda, and possibly Cephalopoda, and a possibly homologous gland is present in Monoplacophora (Lemche and Wingstrand 1959; Wingstrand 1985; Haszprunar 1997). One of the characteristics supporting the monophyly of the orthogastropods, the flexoglossage condition of the radula (Haszprunar 1988a; Salvini-Plawen 1988; Ponder and Lindberg 1997), is now thought to be plesiomorphic, owing to some lateral bending of the radula being found in chitons (Guralnick and Smith 1999). Guralnick and Smith (1999) suggest that the stereoglossate condition of the radula in patellogastropods is secondary. The radular stroke of living monoplacophorans has not been examined, so the condition of their radula can only be inferred. However, Guralnick and Smith (1999) argue that it is also probably flexoglossate with the structure of the radula most like that of lepetid patellogastropods. Available data also suggest a flexoglossate condition in cephalopods, scaphopods and in ‘aplacophorans’ (Guralnick and Smith 1999). Thus, on the basis of these findings, the stereoglossate condition appears to be an autapomorphy of the patellogastropods. There are, however, some plesiomorphic states retained in the patellogastropod radula that are not found in other gastropods (Guralnick and Smith 1999). Since the publication of Ponder and Lindberg (1997), two new datasets add additional weight to the basal position of the patellogastropods. These relate to the buccal cartilages and the fine structure of the cephalic tentacles. Only the number of buccal cartilages present in the odontophore was scored by Ponder and Lindberg (1997). Sasaki (1998) and Guralnick and Smith (1999) have attempted to homologise the cartilages. Guralnick and Smith (1999) used position and shape as a primary means of tracking the evolution of the buccal cartilages. They argue that a medial pair of cartilages is plesiomorphic for Mollusca, as also (probably, therefore being secondarily absent in “aplacophorans”) are the dorsolateral (= anterolateral of Sasaki (1998)) cartilages. More likely, assuming 'aplacophorans' are basal (e.g. Haszprunar 2000), the cartilages are probably synapomorphic of Testaria (sensu Haszprunar 2000). Whereas Sasaki (1998) considered these latter cartilages to be autapomorphies of patellogastropods, Guralnick and Smith (1999) argued that they were homologues of the dorsolateral cartilages of chitons and monoplacophorans. In these latter taxa, the two pairs of cartilages are attached by a connective tissue sheath, the space between being the hollow vesicles seen in those groups. Dorsolateral cartilages are absent in all Apogastropoda. A pair of dorsal cartilages is found in chitons and these are absent in modern Monoplacophora, but present in some patellogastropods. In addition, there are two pairs of posterior cartilages in chitons and the patellid patellogastropods (absent, presumably lost, in some of the more modified patellogastropods and in living Monoplacophora; Guralnick and Smith 1999). A single posterior pair is found in some vetigastropods and neritopsines. In patellogastropods, the subradular membrane is not associated with the medial cartilages Gastropod molecular phylogenetics Molluscan Research 141 as it is in other gastropods, but is, instead, associated with the plesiomorphic dorsolateral (and dorsal cartilages when present), lying well above the medial cartilages. Künz and Haszprunar (2001) showed that the fine structure of the cephalic tentacles of patellogastropods differs significantly from that of vetigastropods and neritopsines and that they share ciliary features observed in bivalves and 'aplacophorans'. A similar configuration (stiff cilia with a more or less homogeneous pattern of microtubules) is unknown in most other gastropods, although somewhat similar cilia are known from the tentacles of the pulmonate Lymnaea (Emery 1992). In addition, the ciliary tufts of patellogastropods have several ciliary types, whereas in the other two groups the ciliary morphology is much more uniform. Further, patellogastropods, vetigastropods and neritopsines all show differences in their sensory elements, supporting and mucous cells. Other recent datasets that are less well resolved, but also appear to show that the patellogastropods are distinct from the vetigastropods and other gastropods, include larval musculature and the development of adult muscles (Wanninger et al. 1999) and sperm ultrastructure (e.g. Hodgson and Morton 1998). In other characteristics (e.g. cleavage pattern (van den Bigelaar and Haszprunar 1996), larval morphology and ciliation (Hadfield et al. 1997)), the patellogastropods and vetigastropods share assumed plesiomorphic conditions. A significant problem for the patellogastropod ancestors being the sister-group to the orthogastropods is the lack of undoubted patellogastropods or obvious coiled ancestors in the early fossil record. The oldest undoubted patellogastropod has been confirmed recently (on the basis of shell structure) from the Triassic (Hedegaard ef al. 1997). Recognition of such, probably coiled, ancestors will be difficult, but Wagner (2002) very tentatively suggests ‘euomphalinaes’ as candidates. If this was the case, the split in the two main gastropod lineages occurred in the Late Cambrian, around 510 million years ago. Cocculiniformia are not monophyletic in our analyses. Notocrater (Pseudocculinidae) is strongly associated with Vetigastropoda, as found by Ponder and Lindberg (1997). Coccopigya is variously associated in derived positions (e.g. with Heterobranchia or Depressigyra and Perotrochus). The pairing with Nerita at the base of Orthogastropoda observed by Ponder and Lindberg (1997) is not found in any of our analyses. There are no morphological characteristics directly suggesting that Cocculinidae and Heterobranchia are sister-groups, although the sinistral protoconch coiling found in all members of the latter group has its analogue in at least some Cocculinidae. The pairing of Coccopigya with two other deep-sea taxa (Depressigyra and Perotrochus) appears to be coincidental because Notocrater and Lepetodrilus are also found in this environment. Heterobranchia is rarely monophyletic in our analyses, owing to the variable placement of Philippea. Euthyneura is monophyletic in all analyses. Within Euthyneura, Opisthobranchia and Pulmonata are rarely monophyletic, concurring with Dayrat ef al. (2001). They found that Opisthobranchia is paraphyletic with respect to Pulmonata, albeit that the nodes suggesting this observation had low bootstrap support. The clade comprising Euthyneura plus Cornirostra is strongly supported in our analyses, supporting the suggestion of Haszprunar (19884) that Valvatoidea is closer to Euthyneura than is Architectonicoidea. Confirmation of this will require data from other members of the family because Philippea is undoubtedly highly autapomorphic and its placement appears to depend on long-branch attraction. Questions of monophyly or paraphyly of Heterostropha, which includes the well-established families | Omalogyridae, Pyramidellidae, Valvatoidea, Architectonicoidea and Rissoellidae, as well as a number of recently created Recent and fossil families, will be a fruitful area for further research. 142 Molluscan Research D. J. Colgan et al. The relatively large genetic differentiation of Heterobranchia in maximum parsimony trees suggests that the clade has a long evolutionary history, particularly if substitution rates are even remotely clocklike. The genetic distinction of Heterobranchia is emphasised by mitochondrial DNA genome organisation. In studied Euthyneura (for references, see Kurabayashi and Ueshima 2000a), this is radically different to the arrangement in the caenogastropod Littorina (Widling et al. 1999) but similar to that of Omalogyra (Kurabayashi and Ueshima 20005). Other gene order work on opisthobranchs (Grande 2001; Medina et al. 2001) has yet to be reported in full. To date, the gene order data are based on an extremely small sampling and whether or not Littorina is typical of caenogastropods is unknown. For example, Collins et al. (2001) and Rawlings et al. (2001) report a major gene order rearrangement within the caenogastropod Vermetidae. Sperm structure (Healy 1993) also supports the monophyly of heterobranchs as a whole and Euthyneura. The earliest undoubted heterobranch fossils date from the early Devonian (390-408 million years ago; Fryda and Blodgett 2001), although some taxa included in the subulitoideans are likely heterobranchs and this grouping extends into the Ordovician (Nützel et al. 2000). The affinities of the Heterobranchia (excepting Philippea) are with the Caenogastropoda in a recognisably ‘apogastropodan’ group (Salvini-Plawen and Haszprunar 1987; Haszprunar 1988a; as extended by Ponder and Lindberg 1997), although some taxa are anomalously included or excluded in some analyses. This contrasts with Colgan et al. (2000), where the ‘apogastropod’ group also had unexpected inclusions (Nerita and Nautilus) and Caenogastropoda and Heterobranchia were intermixed. Apogastropoda are monophyletic and comprised of monophyletic Caenogastropoda and Heterobranchia in McArthur and Koop (1999) and Harasewych and McArthur (2000), although these studies include fewer taxa from these latter groups. Caenogastropoda is monophyletic with the exception of the anomalous inclusion of Nerita and/or Nautilus in some analyses and the exclusion of Leptopoma and Cypraea in analyses (i) and (vii). Relationships within Caenogastropoda are not well resolved in the present analyses. In particular, although various sets of two of the five genera of Neogastropoda included in the dataset are found in monophyletic clades in some analyses and four of the five are grouped in the Bayesian analysis (xi), this morphologically strongly supported group is not otherwise shown as closely related, as also found by Harasewych et al. (1997b). Architaenioglossa comprise a number of superfamilies (previously two, now three) not considered close relatives by Ponder and Lindberg (1997). The taxa included here, Bellamya representing Vivipariodea (previously included within what is now considered to be a separate superfamily Ampullarioidea) and Leptopoma representing Cyclophoroidea, are monophyletic only in analysis (ix) based on maximum likelihood and analysis (xi) based on Bayesian likelihood. Bellamya is a member of the Caenogastropoda in all our analyses, but the position of Leptopoma varies widely although remaining within an apogastropodan group, except when all data are considered (analysis (i)), where it is a sister-group to an heterogeneous taxon (Fig. 2). Campanile is always associated with Bellamya or Leptopoma. McArthur and Koop (1999), using partial 28$ rDNA sequences, also found that the architaenioglossans (Ampullaria and Viviparus) were not monophyletic and, unlike our result, that Ampullaria and Campanile were sister-taxa. In their analyses, and with most of our analyses, the architaenioglossans lie within the caenogastropods, as suggested by Ponder and Warén (1988) and demonstrated in the morphological analyses of Ponder and Lindberg (1996, 1997). Alternative hypotheses have been produced on the basis of Gastropod molecular phylogenetics Molluscan Research 143 morphological analyses, notably suggesting that the architaenioglossans are the sister-group to the apogastropods and part of a paraphyletic ‘Archaeogastropoda’ (Haszprunar 1988a) or that they belong to a clade (together with Neritopsina and Neomphaloidea), which is sister-group to the caenogastropods (Barker 2001). The placement of the Neritopsina (or Neritimorpha) remains uncertain. This group, plus the Cocculiniformia (among our studied taxa), formed a sister-clade to all other Orthogastropoda in Ponder and Lindberg’s (1997) preferred topology and in Rosenberg et al. (1997), although in this latter analysis the patellogastropod was included within the apogastropods. In the morphological analysis undertaken by Sasaki (1998), the patellogastropods formed the base of the gastropod clade and Cocculina appeared within a clade containing Neomphalus and the vetigastropods. Neritopsines formed one of four branches in an unresolved Gastropoda in the analysis of Harasewych et al. (1997a) and one branch of a basal trichotomy in McArthur and Koop (1999: fig. 3) and our analysis (vi). As in analyses (ii), (iv) and (viii), the Neritopsina was a sister-group to all other gastropods in maximum parsimony analyses (excluding one 25 bp insert) of the 18S rDNA data of Harasewych and McArthur (2000: fig. 24) and their maximum likelihood analysis (fig. 2C). When two longer inserts were excluded, Neritopsina was a sister-group to Vetigastropoda, this pair being a sister-group to Apogastropoda (Harasewych and McArthur 2000: fig. 23). Nerita was placed within Caenogastropoda in Colgan et al. (2000), in accordance with our analyses (i), (v), (vii), (ix) and (xi) but in contrast with all recent morphological assessments (cf. Bieler 1992). In Barker's (2001) morphological analysis, Neritopsina was the sister-group to a clade consisting of Neomphaloidea + Architaenioglossa. This combination was a sister-group taxon to the caenogastropods. In our analyses (iii) and (vi), Nerita is a sister-group to Apogastropoda, allowing the possibility that larval planktotrophy arose once only in gastropods (cf. Ponder 1991; see also discussion in Ponder and Lindberg 1997: 209—213; and Fryda 2001). The strict consensus of the maximum parsimony trees from analysis (vi) has notable similarities to recent morphological hypotheses. Agreeing with Ponder and Lindberg (1997), Caenogastropoda and Heterobranchia are monophyletic, as is Vetigastropoda, with the predicted inclusion of Notocrater. Patellogastropoda is basal, although excluded from Gastropoda. Although the results ofall analyses should be included in discussions of the phylogenetic implications of our data, we give a little more weight, when results differ, to those of analysis (vi), where, with Nautilus as the only outgroup, third-position data are excluded and transitions and transversions are weighted differently (Fig. 5). Comparison of consistency indices supports the exclusion of data because they are higher in trees including third positions than those excluding them. As judged by the consistency indices, excluding these data reduces the amount of phylogenetic noise. Arguing for differential weighting is the low transition to transversion ratio in the overall data for coding genes CO7 and H3. This ratio increases when third positions are excluded, indicating a substantial degree of saturation. This analysis (as well as some others) has a high probability of homogeneity of phylogenetic inferences from the separate gene data. When the chiton and bivalves are included, the outgroups are not monophyletic in any analysis. The use of Nautilus as the sole outgroup is suggested by the consensus on morphological grounds that Cephalopoda or Monoplacophora are the sister-taxon to Gastropoda (reviewed by Ponder and Lindberg 1997). Although not included in our analysis, scaphopods have recently been shown to be the sister-taxon to the cephalopods (Waller 1998; Haszprunar 2000; Giribet and Wheeler 2002; Wanninger and Haszprunar 144 Molluscan Research D. J. Colgan et al. 2002), in contrast with earlier hypotheses that linked them to the bivalves. This relationship is, however, not apparent in the analysis of Rosenberg et al. (1997). Unfortunately, despite considerable advancement in our knowledge of Palaeozoic fossils in the past decade, the origins of the major gastropod groups remain obscure, although all should have differentiated by the early Ordovician shortly after gastropods evolved (Wagner 2001, 2002). Whereas the considerable extinctions that have occurred during gastropod evolution may account for some ofthe long branch attraction issues encountered (especially between the patellogastropods and the remainder of the gastropods), breaking down some of the long branches encountered in this dataset by the addition of more taxa (Graybeal 1998) may be possible. Despite more molecular data having been incorporated in these present analyses, some major aspects of gastropod phylogeny remain equivocal. Additional genes, gene order data and more refined morphological data will be required to resolve many of these issues, as well as better data on Palaeozoic gastropods. Acknowledgments We are grateful to Diana Picone for collecting more than half the U2 sequences and to the following for providing samples: M. G. Harasewych for tissues of Notocrater, Nautilus and Perotrochus; A. G. McArthur for providing DNA of the vent taxa Lepetodrilus and Depressigyra (with the kind permission of Dr V. Tunnicliffe); B. Marshall for Coccopigya; Fred Wells for Campanile; and Rosalyn Yardin for Anadara DNA. P. Eggler, I. Loch, P. Colman, A. Miller and S. Clark assisted in the collection of material. References Bandel, K. (1982). Morphologie und Bildung der frühontogenetischen Geháuse bei conchiferen Mollusken. Facies 7, 1-198, 6 pls. Bandel, K. (1997). Higher classification and pattern of evolution of the Gastropoda. A synthesis of biological and paleontological data. Courier Forschungsinstitut Senckenberg 201, 57-81. Bandel, K. (2000). The new family Cortinellidae (Gastropoda, Mollusca) connected to a review of the evolutionary history of the subclass Neritimorpha. Neues Jahrbuch für Geologie und Palaontologie Abhandlungen 217, 111—129. Bandel, K., and Fryda, J. (1999). Notes on the evolution and higher classification of the subclass Neritimorpha (Gastropoda) with the description of some new taxa. Geologica et Falaeontologica 33, 219-235. Bandel, K., and Geldmacher, W. (1996). The structure of the shell of Patella crenata connected with suggestions to the classification and evolution of the Archaeogastropoda. Freiberger Forschungshefte C 464, 1-71. [ Barker, G. M. (2001). Gastropods on land: phylogeny, diversity and adaptive morphology. In *The Biology of Terrestrial Molluses”. (Ed. G. M. Barker.) pp. 1-146. (CABI Publishing: Wallingford, UK.) Bieler R. (1992). Gastropod phylogeny and systematics. Annual Review of Ecology and Systematics 23, 311-338. Boore, J. L., and Brown,W. M. (1995). Complete sequence of the mitochondrial DNA of the annelid worm Lumbricus terrestris. Genetics 141, 305—319. Brown, S., Rouse, G., Hutchings, P. A., and Colgan, D. J. (1999). Testing polychaete relationships using DNA sequence data from histone H3, U2 snRNA and 28S rDNA. Australian Journal of Zoology 47, 419—516. Carlini, D. B., and Graves, J. E. (1999). Phylogenetic analysis of cytochrome c oxidase I sequences to determine higher-level relationships within the coleoid cephalopods. Bulletin of Marine Science 64, 57-76. Colgan, D. J., McLauchlan, A., Wilson, G. D. F., Livingston, S. L., Edgecombe, G. D., Macaranas, J., Cassis, G., and Gray, M. R. (1998). Histone H3 and U2 snRNA DNA sequences and arthropod molecular evolution. Australian Journal of Zoology 46, 419—437. Gastropod molecular phylogenetics Molluscan Research 145 Colgan, D. J., Ponder, W. E, and Eggler, P. E. (2000). Gastropod evolutionary rates and phylogenetic relationships assessed using partial 28S rDNA and histone H3 sequences. Zoologica Scripta 29, 29-63. Collins, T. M., Rawlings, T. A., and Bieler, R. (2001). Learning about mechanisms of mtDNA gene-order change from the caenogastropods: major gene order change and gene order homoplasy. In *Abstracts, World Congress of Malacology 2001, Vienna, Austria”. (Eds L. Salvini-Plawen, J., Voltzow, H. Sattmann and G. Steiner.) p. 64. (Unitas Malacologica: Vienna.) (Abstract.) Dayrat, B., Tillier, A., Lecointre, G., and Tillier, S. (2001). New clades of euthyneuran gastropods (Mollusca) from 28S rRNA sequences. Molecular Phyogenetics and Evolution 19, 225-235. Emery, D. G. (1992). Fine structure of the olfactory epithelia of gastropod molluscs. Microscopy Research and Technique 22, 307-324. Erikson, T., and Wikstróm, N. (1996). “AutoDecay”, version 3.0. (Stockholm University Press: Stockholm.) Felsenstein, J. (1978). Cases in which parsimony or compatibility methods will be positively misleading. Systematic Zoology 27, 401-410. Felsenstein, J. (1985). Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, 783-791. Freeman, G., and Lundelius, J. W. (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 Graham, A. (1962). “British Prosobranch Molluscs.’ (Ray Society: London.) Fretter, V., Graham, A., and McLean, J. H. (1981). The anatomy of the Galapagos rift limpet, Neomphalus fretterae. Malacologia 21, 337-361. Fryda, J. (1999). Higher classification of Paleozoic gastropods inferred from their early shell ontogeny. Journal of the Czech Geological Society 44, 137-154. Fryda, J. (2001). On the evolution of larval development in Middle Paleozoic Gastropoda. In “Abstracts, World Congress of Malacology 2001, Vienna, Austria”. (Eds L. Salvini-Plawen, J., Voltzow, H. Sattmann and G. Steiner.) p. 107. (Unitas Malacologica, Vienna.) (Abstract.) Fryda, J., and Blodgett, R. B. (2001). The oldest known heterobranch gastropod, Kuskokwimia gen. nov., from the Early Devonian of west-central Alaska, with notes on the early phylogeny of higher gastropods. Vstnik eského geologického ústavu 76, 39—54. Giribet, G., and Wheeler, W. C. (2002). On bivalve phylogeny: a high-level analysis of the Bivalvia (Mollusca) based on combined morphology and DNA sequence data. Invertebrate Biology, 121, 271-324. Golikov, A., and Starobogatov, Y. I. (1975). Systematics of prosobranch gastropods. Malacologia 11, 185-232. Graham, A. (1985). Evolution within the Gastropoda: Prosobranchia. In “The Mollusca, Evolution”. (Eds E. R. Trueman and M. R. Clark.) pp. 151-186. (Academic Press: New York.) Grande, C., Templado, J., Cevera, J. L., and Zardoya, R. (2001). Complete sequence of the mitochondrial genome of the nudibranch Roboastra europaea (Mollusca, Opisthobranchia). In *Abstracts, World Congress of Malacology 2001, Vienna, Austria”. (Eds L. Salvini-Plawen, J., Voltzow, H. Sattmann and G. Steiner.) p. 130. (Unitas Malacologica: Vienna.) (Abstract.) Graybeal, A. (1998). Is it better to add taxa or characters to a difficult phylogenetic problems? Systematic Biology 47, 9-17. Guralnick, R., and Smith, K. (1999). Historical and biomechanical analysis of integration and dissociation in molluscan feeding, with special emphasis on the true limpets (Patellogastropoda: Gastropoda). Journal of Morphology 241, 175-195. Guthrie, C., and Patterson, B. (1988). Spliceosomal snRNAs. Annual Review of Genetics 22, 387-419. Hadfield, M. G., Strathmann, M. F., and Strathmann, R. R. (1997). Ciliary currents of non-feeding veligers in putative basal clades of gastropods. Invertebrate Biology 116, 313-321. Harasewych, M. G., Adamkewicz, S. L., Blake, J. A., Saudek, S., Spriggs, T., and Bult, C. J. (1997a). Phylogeny and relationships of pleurotomariid gastropods (Mollusca: Gastropoda): an assessment based on partial 18S rDNA and cytochrome c oxidase 1 sequences. Molecular Marine Biology and Biotechnology 6, 1—20. Harasewych, M. G., Adamkewicz, S. L., Blake, J. A., Saudek, D., Spriggs, T., and Bult, C. J. (19972). Neogastropod phylogeny: a molecular perspective. Journal of Molluscan Studies 63, 327-351. Harasewych, M. G., and McArthur, A. G. (2000). A molecular phylogeny of the Patellogastropoda (Mollusca: Gastropoda). Marine Biology 137, 183-194. Hassouna, N., Michot, B., and Bachellerie, J. P. (1984). The complete nucleotide sequence of mouse 28S rRNA gene. Implications for the process of size increase of the large subunit rRNA in higher eukaryotes. Nucleic Acids Research 12, 3563-3583. 146 Molluscan Research D. J. Colgan et al. Haszprunar, G. (1985a). The Heterobranchia: a new concept of the phylogeny and evolution of the higher Gastropoda. Zeitschrift für Zoologische Systematik und Evolutionsforschung 23, 15-37. Haszprunar, G. (19855). The fine morphology of the osphradial sense organs of the Mollusca. Part 2: Allogastropoda (Architectonicidae and Pyramidellidae). Philosophical Transactions of the Royal Society of London B 307, 497-505. Haszprunar, G. (19882). On the origin and evolution of major gastropod groups, with special reference to the Streptoneura (Mollusca). Journal of Molluscan Studies 54, 367-441. Haszprunar, G. (19885). Comparative anatomy of cocculiniform gastropods and its bearing on archaeogastropod systematics. Malacological Review Supplement 4, 64—84. Haszprunar, G. (1989a). The anatomy of Melanodrymia aurantiaca Hickman, a coiled archaeogastropod from the East Pacific hydrothermal vents (Mollusca: Gastropoda). Acta Zoologica 70, 175-186. Haszprunar, G. (19899). New slit-limpets (Scissurellacea and Fissurellacea) from hydrothermal vents. Part 2. Anatomy and relationships. Los Angeles County Museum Contributions in Science 408, 1-17. Haszprunar, G. (1997). Monoplacophora. In ‘Microscopical Anatomy of Invertebrates’. (Eds F. W. Harrison and A. J. Kohn.) pp. 415-457. (Wiley-Liss: New York.) Haszprunar, G. (1998). Superorder Cocculiniformia. In ‘Mollusca: The Southern Synthesis’, Fauna of Australia, Vol. 5B. (Eds P. L. Beesley, G. J. B. Ross and A. Wells.) pp. 653-664. (CSIRO Publishing: Melbourne.) Haszprunar, G. (2000). Is the Aplacophora monophyletic? A cladisitic point of view. American Malacological Bulletin 15, 115—130. Healy, J. M. (1993). Comparative sperm ultrastructure and spermiogenesis in basal heterobranch gastropods (Valvatoidea, Architectonicoidea, Rissoelloidea, Omalogyroidea, Pyramidelloidea) (Mollusca). Zoologica Scripta 22, 263-276. Hedegaard, C., Lindberg, D. R., and Bandel, K. (1997). A patellogastropod limpet from the Triassic St. Cassian Formation of Italy. Lethaia 30, 331—335. Hickman, C. S. (1988). Archaeogastropod evolution, phylogeny and systematics: a re-evaluation. Malacological Review Supplement 4, 17—34. Hodgson, A. N., and Morton, B. (1998). Spermiogenesis and sperm structure in three species of Patelloida and one species of Nipponacmaea (Patellogastropoda: Acmaeoidea). Journal of Molluscan Studies 64, 11-19. Huelsenbeck, J. P, and Ronquist, F. R. (2001). MRBAYES: Bayesian inference of phylogeny. Bioinformatics 17, 754—755. Israelsson, O. (1998). The anatomy of Pachydermia laevis (Archaeogastropoda: “Peltospiridae” ). Journal of Molluscan Studies 64, 93-109. Künz, E., and Haszprunar, G. (2001). Comparative ultrastructure of gastropod cephalic tentacles: Patellogastropoda, Neritaemorphi and Vetigastropoda. Zoologischer Anzeiger 240, 137-165. Kurabayashi, A., and Ueshima, R. (2000a). Complete sequence of the mitochondrial DNA of the primitive opisthobranch gastropod Pupa strigosa: systematic implications of the genome organisation. Molecular Biolology and Evolution 17, 266-277. Kurabayashi, A., and Ueshima, R. (20005). Partial mitochondrial genome organization of the heterostrophan gastropod Omalogyra atomus and its systematic significance. Venus 59, 7-18. Lemche H., and Wingstrand K. G. (1959). The anatomy of Neopilina galatheae Lémche, 1957. Galathea Reports 3, 9—71. Lindberg, D. R., and Guralnick. R. P. (2001). The evolution of early development in gastropod molluscs. In *Abstracts, World Congress of Malacology 2001, Vienna, Austria'. (Eds L. Salvini-Plawen, J., Voltzow, H. Sattmann and G. Steiner.) p. 193. (Unitas Malacologica: Vienna.) (Abstract.) Lindberg, D. R., and Ponder, W. F. (2001). The influence of classification on the evolutionary interpretation of structure: a re-evaluation of the evolution of the pallial cavity of gastropod molluscs. Organisms, Diversity and Evolution 1, 273—299. Lyons-Weiler, J., and Hoelzer G. A. (1997). Escaping from the Felsenstein zone by detecting long branches in phylogenetic data. Molecular Phylogenetics and Evolution 8, 375-384. Maddison, W. P., and Maddison D. R. (1992). ‘MacCLADE’, version 3.0. (Sinauer Associates: Sunderland, MA.) McArthur, A. G., and Koop, B. F. (1999). Partial 28S rDNA sequences and the antiquity of the hydrothermal vent endemic gastropods. Molecular Phylogenetics and Evolution 13, 255-274. McLean, J. H. (1981). The Galapagos rift limpet Neomphalus: relevance to understanding a major Palaeozoic-Mesozoic radiation Malacologia 21, 291—336. Gastropod molecular phylogenetics Molluscan Research 147 Medina, M., Valles, Y., Gosliner, T., and Boore, J. (2001). Mitochondrial evolution of crown gastropods: insight from large subunit sequences and gene order data. In ‘Abstracts, World Congress of Malacology 2001, Vienna, Austria”. (Eds L. Salvini-Plawen, J., Voltzow, H. Sattmann and G. Steiner.) p. 218. (Unitas Malacologica: Vienna.) (Abstract.) Nützel, A., Erwin, D. H., and Mapes, R. H. (2000). Identity and phylogeny of the Late Paleozoic Subulitoidea (Gastropoda). Journal of Paleontology 74, 575—598. Parkhaev, P. Y. (2001). The functional morphology of the Cambrian univalved molluscs: helcionellids. Paleontological Journal 35, 470—475. Peel, J. S. (1991). Functional morphology, evolution and systematics of early Palaeozoic univalved molluscs. Grønlands Geologiske Undersøgelse Bulletin 161, 1—116. Philippe, H., and Germot, A. (2000). Phylogeny of eukaryotes based on ribosomal RNA: long-branch attraction and models of sequence evolution. Molecular Biology and Evolution 17, 830-834. Ponder W. F. (1991). Marine valvatoidean gastropods: implications for early heterobranch phylogeny. Journal of Molluscan Studies 57, 21-32. Ponder, W. F., and Lindberg, D. R. (1996). Gastropod phylogeny: challenges for the '90s. In ‘Origin and Evolutionary Radiation of the Mollusca’. (Ed. J. D. Taylor.) pp. 135—154. (Oxford University Press: Oxford.) Ponder, W. E, and Lindberg, D. R. (1997). Towards a phylogeny of gastropod molluscs: a preliminary analysis using morphological characters. Zoological Journal of the Linnean Society 19, 83-265. Ponder, W. F., and Warén, A. (1988). Classification of the Caenogastropoda and Heterostropha: a list of the family-group and higher category names. Malacological Review Supplement 4, 288—326. Prager, E. M., and Wilson, A. C. (1988). Ancient origin of lactalbumin from lysozyme: analysis of DNA and amino acid sequences. Journal of Molecular Evolution 27, 326-335. Rawlings, T. A., Collins, T. M., and Bieler, R. (2001). A major mitochondrial gene rearrangement among closely related species. Molecular Biology and Evolution 18, 1604—1609. Rosenberg, G., Tillier, S., Tillier, A., Kuncio, G. S., Hanlon, R. T., Masselot, M., and Williams, C. J. (1997). Ribosomal RNA phylogeny of selected major clades in the Mollusca. Journal of Molluscan Studies 63, 301-309. Salvini-Plawen, L. V. (1988). The structure and function of molluscan digestive systems. In “The Mollusca. Form and Function”, Vol. 11 (Eds E. R. Trueman and M. R. Clarke.) pp. 301—379. (Academic Press: Orlando, FL.) Salvini-Plawen, L. V. (1990). Origin, phylogeny and classification of the phylum Mollusca. /berus 9, 1-33. Salvini-Plawen, L. V., and Haszprunar, G. (1987). The Vetigastropoda and the systematics of streptoneurous Gastropoda (Mollusca). Journal of Zoology London 211, 747—770. Sasaki, T. (1998). Comparative anatomy and phylogeny of the recent Archaeogastropoda. The University of Tokyo Bulletin 38, 1—224. Siddall, M. E., and Whiting, M. F. (1999). Long-branch abstractions. Cladistics 15, 9-24. Stiller, J. W., and Hall, B. D. (1999). Long-branch attraction and the rDNA model of early eukaryotic evolution. Molecular Biology and Evolution 16, 1270-1279. Swofford, D. L. (2000) *PAUP*: Phylogenetic Analysis Using Parsimony', version 4.0. (Sinauer Associates: Sunderland, MA.) Thiele, J. (1925). Gastropoda. Bd. 5. In ‘Handbuch der Zoologie’. (Ed. T. Krumbach.) pp. 38-155. (Walter de Gruyter and Co.: Leipzig.) Thiele J. (1929, 1931). ‘Handbuch der systematischen Weichtierkunde’, Vol. 1. (Gustav Fischer: Jena.) Thollesson, M. (1999). Phylogenetic analysis of Euthyneura (Gastropoda) by means of the 16S rRNA gene: use of a “fast” gene for “higher-level” phylogenies. Proceedings of the Royal Society London B 266, 75-83. Thompson, J. D., Higgins, D. G., and Gibson, T. J. (1994). CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position specific gap penalties and weight matrix choice. Nucleic Acids Research 22, 4673-4680. Tillier, S., Masselot, M., Phillipe, H., and Tillier, A. (1992). Phylogénie moléculaire des Gastropoda (Mollusca) fondée sur le sequençage partiel de l'ARN ribosomique 28S. Comptes Rendus Academie de Sciences Serie III 134, 79-85. Tillier, S., Masselot, M., Guerdoux, J., and Tillier, A. (1994). Monophyly of major gastropod taxa tested from partial 28S rRNA sequences, with emphasis on Euthyneura and hot-vent limpets Peltospiroidea. Nautilus 108 (Suppl. 2), 122-140. 148 Molluscan Research D. J. Colgan et al. van den Bigelaar, J. A. M. (1996). The significance of the early cleavage pattern for the reconstruction of gastropod phylogeny. In *Origins and Evolutionary Radiation of the Mollusca'. (Ed. J. Tayor.) pp. 155-160. (Oxford University Press: Oxford.) van den Bigelaar, J. A. M., and Haszprunar, G. (1996). Cleavage patterns and mesentoblast formation in the Gastropoda: an evolutionary perspective. Evolution 50, 1520-1540. Wagner, P. J. (2001). Gastropod phylogenetics: progress, problems, and implications. Journal of Paleontology 75, 1128-1140. Wagner, P. J. (2002). Phylogenetic relationships of the earliest anisostrophically coiled gastropods. Smithsonian Contributions to Paleobiology 88, i-vi, 1-152. Waller, T. R. (1998). Origin of the molluscan class Bivalvia and a phylogeny of major groups. In ‘Bivalves: an Eon of Evolution. Paleobiological Studies Honoring Norman D. Newell’. (Eds P. A. Johnston and J. W. Haggart.) pp. 1-47. (Calgary University Press: Calgary.) Wanninger, A., and Haszprunar, G. (2002). Muscle development in Antalis entalis (Mollusca, Scaphopoda) and its significance for scaphopod relationships. Journal of Morphology 254, 53-64. Wanninger, A., Ruthensteiner, B., Lobenwein, S., Salvenmoser, W., Dictus, W. J. A. G., and Haszprunar, G. (1999). Development of the musculature in the limpet Patella (Mollusca, Patellogastropoda). Devlopment, Genes and Evolution 209, 226-238. Wen, Y. (1990). The first radiation of shelled molluscs. Palaeontologia Cathayana 5, 139-170. Widling, C. S., Mill, P. J., and Grahame, J. (1999). Partial sequence of the mitochondrial genome of Littorina saxatilis: relevance to gastropod phylogenetics. Journal of Molecular Evolution 48, 348-359. Wingstrand, K. G. (1985). On the anatomy and relationships of Recent Monoplacophora. Galathea Report 16, 7-94. Wirth, T., Le Guellec, R., and Veuille, M. (1999). Directional substitution and evolution of nucleotide content in the Cytochrome Oxidase II gene in earwigs (dermapteran insects). Molecular Biology and Evolution 16, 1645-1653. http://www.publish.csiro.au/journals/mr CSIRO PUBLISHING www.publish.csiro.au/journals/mr Molluscan Research, 2003, 23, 149-158 Reassessment of Australia's oldest freshwater snail, Viviparus (?) albascopularis Etheridge, 1902 (Mollusca: Gastropoda: Viviparidae), from the Lower Cretaceous (Aptian, Wallumbilla Formation) of White Cliffs, New South Wales Benjamin P. Kear^, Robert J. Hamilton-Bruce^*, Brian J. Smith" and Karen L. Govvlett-Holmes? ASouth Australian Museum, North Terrace, Adelaide, SA 5000, Australia. BSchool of Biological, Earth and Environmental Science, University of New South Wales, Sydney, NSW 2052, Australia. CQueen Victoria Museum, Wellington Street, Launceston, Tasmania 7250, Australia. DCSIRO Division of Marine Research, GPO Box 1538, Hobart, Tasmania 7001, Australia. FTo whom correspondence should be addressed. Email: hamilton-bruce.robert@saugov.sa.gov.au Abstract Viviparus (?) albascopularis Etheridge, 1902 is Australia’s oldest documented fossil freshwater gastropod. The taxon was established on the basis of a single opalised shell from the Lower Cretaceous (Aptian) marine deposits of the Wallumbilla Formation (Doncaster Member) at White Cliffs, New South Wales. Reassessment indicates that original placement in the caenogastropod family Viviparidae is justified; however, the specimen is reassigned to the endemic Australian genus Noropala Cotton, 1935 on the basis of shell morphology and close morphometric similarity to extant species. Implications for the origins and current distribution of Australian viviparid taxa are discussed. Introduction The fossil record of Australian Cretaceous non-marine gastropods is depauperate and very poorly known. Nearly all the currently identified material has been recovered from the middle-upper Albian (Lower Cretaceous) fluviatile-lacustrine deposits of the Griman Ck Formation at Lightning Ridge in north-western New South Wales (NSW). This assemblage comprises primarily viviparids, although numerous other groups, including thiarids (Hamilton-Bruce ef al. in press), succinids, camaenids (currently under study) and amphibolids (B. J. Smith, R. J. Hamilton-Bruce and B. P. Kear, unpublished data) are also present (Dettman et al. 1992; Smith 1999; Hamilton-Bruce ef al. 2002). The only other documented Australian Cretaceous non-marine gastropod fossil is a single opalised shell (AM F17456) from the Aptian (Lower Cretaceous) marine sediments of the Wallumbilla Formation (Doncaster Member) at White Cliffs, NSW (Burton and Mason 1998: see figs 1,2 for detailed geological and locality maps of the area). This specimen was described (from a private collection belonging to a Mr H. Y. L. Brown of Adelaide, later acquired by the Australian Museum) by Etheridge (1902) and tentatively assigned to Viviparus (?) albascopularis, recognising similarity to members of the currently extant caenogastropod family Viviparidae. The present paper provides a revised description of the holotype (AM F17456) and only known specimen of K (?) albascopularis Etheridge and reinterprets its taxonomic placement. Implications for the origins and distribution of Australian viviparid taxa are discussed. The Lower Cretaceous (Aptian) opal-bearing deposits of White Cliffs have long been known as a productive locality for fossils. Anderson (1892) briefly remarked on the presence of mollusc remains, crinoids and wood. Jaquet (1893) recorded belemnitid © Malacological Society of Australasia 2003 10.1071/MR03003 1323-5818/03/020149 150 Molluscan Research B. P. Kear et dl. cephalopods and Devonian invertebrate fossils preserved as impressions in large erratic clasts. These were interpreted as a product of reworking from underlying Palaeozoic conglomerates. Etheridge (1897, 1902, 1904) reported the occurrence of bivalves, ammonites, naticid gastropods, plesiosaurs and ichthyosaurs. More recent studies by White (1926), Molnar (1980, 1991) and Kemp (1991) have also identified lungfish and dinosaur remains. Viviparus (?) albascopularis Etheridge is currently the oldest known Australian fossil freshwater gastropod and one of the earliest members of the Viviparidae. Today, this cosmopolitan family comprises various taxa, characterised by medium-large-sized turbiniform shells, possessing a rounded body whorl, moderately high and pointed spire, wide, round aperture and a concentric, horny operculum (Smith 1992). Within Australia, the distribution of the group is limited to a few species occurring in the large drainage basins that span much of the arid centre, northern tropical and coastal regions. The fossil record for Viviparidae is known from the Jurassic-Recent (Viviparus Montfort, 1810), with a tentative report based on an internal shell mold (?Bernicia Cox, 1927), possibly of marine origin, from the Early Carboniferous of England (Brookes-Knight ef al. 1960). The group's Australian record is very sparsely documented. Cotton (1935) described a species of Notopala Cotton, 1935 (N. wanjacalda) from upper Pleistocene sediments along the Murray River, near Sunnyside, South Australia (SA), and noted a second taxon (Notopala sp.) from the same deposit, which showed strong similarity to the extant N. hanleyi (von Frauenfeld, 1862). Dettman et al. (1992) reported viviparid snail shells from the Lower Cretaceous (middle-upper Albian) deposits of Lightning Ridge, NSW, as did Smith (1999), who also recorded representatives of the Naticidae, Thiaridae and Ellobiidae. Recently, Hamilton-Bruce et al. (2002) described a new genus (Albianopalin) and two new species of viviparid from Lightning Ridge, as well as indeterminate material attributable to the currently extant endemic Australian taxon Noropala. Other Australian non-marine gastropod fossils (all of Tertiary age) have been documented by Chapman (1937), McMichael (1968), Archer et al. (1994), Arena (1997) and Pledge et al. (2002). Occurrences from elsewhere in Australasia are rare, particularly in Cretaceous sediments. Some of the few examples include viviparids (genera uncertain) and thiarids (?Melanoides Olivier, 1804) from the Cenomanian-?Santonian (Upper Cretaceous) of New Zealand (Henderson ez al. 2000) and possible thiarids (Pyrgulifera Meek, 1871) from both the Campanian-lower Maastrichtian (Upper Cretaceous) of the Chatham Islands (Stilwell 1998) and ?Campanian of New Caledonia (Henderson ef al. 2000). Material and methods Material registered as the holotype of V (?) albascopularis Etheridge includes a single shell with broken aperture margin and protoconch (AM F17456), preserved entirely in potch (non-precious or common opal). The specimen is derived from an unknown mine locality in the opal-bearing deposits of White Cliffs near Wilcannia in north-western NSW. The lithostratigraphic nomenclature for Lower Cretaceous rocks of the White Cliffs area was recently discussed by Burton and Mason (1998), who placed them within the Doncaster Member of the Wallumbilla Formation (Eromanga Basin), a unit of Aptian-middle Albian (115-approximately 100 million years ago; sensu Lowrie et al. 1980) age. However, the White Cliffs opal-bearing sediments are regarded as representing only the lower Aptian section of the Doncaster Member and comprise predominantly sandy/silty claystone and fine-grained sandstones deposited in a near-shore coastal marine setting (Burger 1988; Burton and Mason 1998). Determinations of palaeolatitude place the White Cliffs area as high as 70°S during the Early Cretaceous (Embleton 1984). Palaeoclimatic indicators for the region also suggest predominantly cool, strongly seasonal conditions with winter freezing (Frakes and Francis 1988, 1990; Sheard 1990; Frakes et al. 1995; De Lurio and Frakes 1999; Henderson etal. 2000). Estimates of sea level isotopic palaeotemperatures in the south-western section of the Reassessment of Viviparus (?) albascopularis Etheridge Molluscan Research 151 Eromanga Basin have yielded averages as low as 12.2°C (Stevens and Clayton 1971; Dettman ef al. 1992). However, Selwood et al. (1994) reported revised isotopic data supporting much cooler ocean temperatures during the Early Cretaceous. Indeed, Pirrie et al. (1995) indicated palaeotemperatures of around 109C based on Early Albian belemnites from the Carnarvon Basin, Western Australia (situated at approximately 45° palaeolatitude during the Cretaceous). In contrast, Huber et al. (1995) and Huber and Hoddell (1996) argued that minimal pole-to-equator thermal gradients existed during much of the Middle-Late Cretaceous. This was also discussed by Henderson et al. (2000), who noted that although palaeotemperatures at 707-809 latitude would certainly have been more equitable than they are today, evidence such as the distinct growth-banding in Australian Cretaceous wood (Dettman et al. 1992), and the presence of potentially ice-rafted quartzite/porphyritic boulders (Frakes and Francis 1988, 1990; Frakes er al. 1995) and glendonites (crystal aggregates pseudomorphing the calcium carbonate hexahydrate mineral ikaite; Sheard 1990; DeLurio and Frakes 1999) attests to the strong seasonality and winter freezing along the inboard extremity of the Australian epicontinental seaway during the Aptian. Shell diameters were measured using the method of Boycott (1928), defined as ‘... the greatest dimension that can be found starting with the edge of the lip to a point on the opposite side of the shell on the last whorl’. Shell measurements were made to the nearest 0.05 mm using dial calipers. The Australian Museum, Sydney is abbreviated as AM throughout. Systematics Class GASTROPODA Superorder CAENOGASTROPODA Superfamily VIVIPARIOIDEA Family VIVIPARIDAE Gray, 1847 Diagnosis Medium to large dextral, turbiniform shells with rounded body whorl, spire moderately high and pointed, aperture wide and round to distinctly lunate—ovate, lirae present or absent and horny, concentric operculum. Remarks The above diagnosis follows Smith (1992), modified to accommodate the presence of a distinctly lunate-ovate aperture and spiral lirae on the last body whorl of the holotype specimen, AM F17456. Viviparid snails are, as their name suggests, live bearing and are found in both lotic and lentic systems throughout the world (Browne 1978). Within Australia, the family is currently represented by the extant native genera Notopala, Larina Adams, 1851 and Centrapala Cotton, 1935 (see Smith 1992) and the introduced Bellamya heudei guangdungensis (Kobelt, 1906), an Asian species now recorded in the wild in NSW (Shea 1994). Australian endemic viviparids have undergone extensive taxonomic revision in recent years (the most inclusive analysis currently being undertaken by W. Ponder of the Australian Museum, personal communication, 2002), with better understanding of intraspecific shell variation and morphometric data resulting in a substantial reduction in the number of accepted species (see Sheldon and Walker 1993). However, many of the key characteristics used to determine interrelationships, such as shell colour and form of the operculum, are usually lost in fossil material; therefore, only structural features of the shell (see below) can be used to assign them to taxa. 152 Molluscan Research B. P. Kear et al. Genus Notopala Cotton, 1935 Notopala Cotton, 1935: 339. Type species (by original designation): Paludina hanleyi von Frauenfeld, 1864. Diagnosis Shell dextral, globose-conic, subumbilicate, up to 5 whorls, ventricose to angulate below the periphery; spiral lirae present on periostracum of last body whorl in some taxa; aperture large and subovate to lunate-ovate, aperture size equal to or greater than height of spire, operculum corneous (unknown in fossil taxa). Remarks Despite the recovery of AM F17456 from a deposit of marine origin, its shell morphology fits within the currently accepted diagnosis of Notopala and Viviparidae with only slight modification (the presence of a distinctly lunate—ovate aperture and calcified spiral lirae on the last body whorl). To justify placement within the family and to establish a basis for both generic reassignment and comparison with extant species, we have applied parts of the morphometric data gathered by Sheldon and Walker (1993). The histogram in Fig. 1 is based on measurements of shell characteristics for each of the living Australian species of Notopala (derived from Sheldon and Walker 1993) and illustrates the morphometric similarity of AM F17456, redescribed herein, to currently existing members of the genus. Notopala albascopularis (Etheridge, 1902) (Fig. 24-D) Viviparus (?) albascopularis Etheridge, 1902: 43; pl. VII, figs 8,9. Material examined Holotype. AM F17456. Type locality and horizon: White Cliffs Opal Field (exact mining claim locality unknown), near Wilcannia, north-western NSW. The deposits form part of the Doncaster Member ofthe Wallumbilla Formation (Rolling Downs Group), Eromanga Basin, and are of Aptian age (see Burton and Mason 1998). This corresponds to the Cyclosporites hughesii-lower-most Crybelosporites striatus spore-pollen Zones and Odontochitina operculata-Diconodinium davidii dinoflagellate zones of Helby et al. (1987). Diagnosis With the features of the genus; irregular, pustular lirae present on the last body whorl; aperture somewhat poorly preserved but distinctly lunate-ovate and markedly adapically situated. Description Shell (Fig. 24—C) dextral, turbiniform, very slightly carinate, subglobose, 21.65 mm high, 20.05 mm maximum diameter, 13.55 mm aperture length, 11.5 mm aperture width, 8.1 mm spire length. Teleconch four complete whorls and broken parts indicating further whorls, probably up to five. Whorls impressed. Relatively evenly spaced fine spiral prosocline growth lines on lower three whorls. Aperture large (13.55 mm high), slightly subangular, distinctly ovate-lunate, markedly adapically situated. Opal replacement of original shell material indicates 12 or more irregular, pustular lirae on the last body whorl (Fig. 2D). Reassessment of Viviparus (?) albascopularis Etheridge Molluscan Research 153 APL SHW APW 4.0 3.0 2.0 1.0 0.0 Eli İd i? "Banded" Shells N. waterhousii ER Notopala sublineata (Cooper) HE N. essingtonensis ll Notopala sublineata (Darling) EH N.alisoni 7 SN. hanleyi N. albascopularis | = Standard deviation * single sample Fig. 1. Histogram showing results of morphometric analysis. Taxa include living species of Notopala (measurements modified from Sheldon and Walker 1993) and N. albascopularis (Etheridge, 1902), AM 17456. APL, aperture length; APW, aperture width; SHW, shell width. Remarks Holotype unique. The presence of irregular, pustular lirae on the last body whorl and a distinctly lunate-ovate, markedly adapically situated aperture separates N. albascopularis (Etheridge, 1902) from other species of Notopala. The generic reassignment of the holotype specimen is highly significant because previously none of the currently living native Australian viviparid genera was known from deposits older than middle-upper Albian (upper-most Lower Cretaceous). This temporal range is now extended back to the Aptian. Discussion The fossil record of Australian non-marine gastropods is very sparsely documented, with the majority of existing reports describing material of Tertiary to Holocene age. Mesozoic specimens have, to date, only been identified from middle to upper Albian (Lower Cretaceous) sediments of the Griman Ck Formation at Lightning Ridge in northern NSW (Dettman et al. 1992; Smith 1999; Hamilton-Bruce ef al. 2002). Therefore, recovery of N. albascopularis (Etheridge, 1902) from the Aptian deposits of the Wallumbilla Formation (Doncaster Member) at White Cliffs gives this taxon the distinction of being both Australia's oldest definitively assigned non-marine gastropod and the earliest recorded representative of the Viviparidae in Australia. Recognition of N. albascopularis as a viviparid also serves to extend the family's temporal range in this region back to at least the later stages of the Early Cretaceous and, given its Jurassic-Recent fossil record in Europe, suggests a possible pre-Jurassic Pangean origin for the group. By the Early Cretaceous, the family had clearly diversified within the Gondwanan region, establishing key endemic taxa that still exist as descendant lineages today. This may be seen, to some extent, in the strong morphological similarity between N. albascopularis and other currently extant Australian 154 Molluscan Research B. P. Kear et al. 10 mm Fig.2. AM 17456 Notopala albascopularis (Etheridge, 1902) in (A) apertural, (B) dorsal and (C) apical views. (D) Magnified section of last body whorl showing lirae (arrowed). Reassessment of Viviparus (?) albascopularis Etheridge Molluscan Research 155 species of Notopala, particularly N. hanleyi and N. sublineata (Conrad, 1850) (see Fig. 1), both of which could potentially represent morphological derivatives. However, further study and the discovery of additional fossil material are required before any definitive phylogenetic relationship can be demonstrated. Although there are numerous records of Cretaceous freshwater bivalves from Australia (McMichael 1957; Ludbrook 1985; Jell and Duncan 1986; Dettman et al. 1992; Hocknull 1997), there are very few for non-marine gastropods of the same period. Indeed, most of the better documented terrestrial units, including the Wonthaggi Formation, Eumeralla Formation and Koonwarra Beds (Korumburra Group) of Victoria (see Jell and Duncan 1986; Rich et al. 1988; Rich and Rich 1989) and Winton Formation of Queensland and SA (see Ludbrook 1985; Dettman et al. 1992; Hocknull 1997), have yet to produce any identifiable gastropod remains. The reasons for this apparent absence are unknown, but could be related to preservational biases, such as shells rapidly breaking up or dissolving after death. However, isolated specimens, such as the holotype of N. albascopularis, seem to have, on occasion, survived transport over considerable distances (perhaps on floating vegetation) prior to eventual burial. This scenario also appears to have been common for many of the other fossils at White Cliffs, which include a high proportion of terrestrial plant remains (Anderson 1892; Jaquet 1893; Etheridge 1902; Newton 1914), freshwater invertebrates (Etheridge 1902; Newton 1914; McMichael 1957; Dettman ef al. 1992) and occasional freshwater/terrestrial vertebrates (White 1926; Molnar 1980, 1991; Kemp 1991), all probably derived from fluviatile input into the near-shore marine depositional environment. Another factor possibly influencing the distribution of Early Cretaceous freshwater gastropods in Australia may have been the strongly seasonal cool to cold climates, which characterised many of the high-latitude continental (Douglas and Williams 1982; Gregory et al. 1989; Dettman et al. 1992; Cantrill 1998) and marine (Frakes and Francis 1988, 1990; Sheard 1990; Frakes er al. 1995; De Lurio and Frakes 1999) environments of the time. Although this may have limited the number of available habitats for non-marine gastropod species, it does not appear to have restricted overall taxonomic diversity or specimen numbers in the few deposits where they occur. Indeed, it is interesting to note that in the Lower Cretaceous freshwater river and lake deposits of Lightning Ridge, viviparids are one of the most common invertebrate faunal elements, far outnumbering other sympatric groups, such as thiarids, ellobiids and naticids (Smith 1999). The reasons for this apparent success are unknown, but could be related to the ability ofthe viviparids to give birth to live young and, thus, secure a competitive advantage over their contemporaries (however, larval brooding is also present in thiarids). Similarly, strict adaptation to freshwater may have enabled Cretaceous viviparids to rapidly colonise available upstream habitats. This contrasts with naticids and thiarids, whose Cretaceous record is largely derived from near-shore marine strata (see Etheridge 1902, Ludbrook 1966; Dettman et al. 1992; Stilwell 1998; Henderson ef al. 2000), and may reflect a preference for more brackish water conditions around estuaries and coastal lagoons. Acknowledgments We thank Robert Jones for generous provision of AM F17456 for study. Philip Ryan examined the statistical methods used and Chris Izzo assisted with measurements and statistical data analysis. This manuscript benefited greatly from the comments of Winston Ponder and Jeffrey Stilwell. The South Australian Museum, Origin Energy, The Advertiser, Coober Pedy Tourism Association and the Waterhouse Club contributed financially to this project. 156 Molluscan Research B. P. Kear et al. References Anderson, W. (1892). Notes on the occurrence of opal in New South Wales. Records of the Geological Survey of New South Wales 3, 29-32. Archer, M., Hand, S., and Godthelp, H. (1994). *Riversleigh: the Story of Animals in Ancient Rainforests of Inland Australia.’ (Reed Books: Sydney.) Arena, D. A. (1997). The palaeontology and geology of Dunsinane Site, Riversleigh. Memoirs of the Queensland Museum 41, 171—179. Boycott, A. E. (1928). Conchometry. Proceeding of the Malacological Society of London 18, 8-31. Brookes-Knight, J., Batten, R. L., Yochelson, E. L., and Cox, L. R. (1960). Supplement Palaeozoic and some Mesozoic Caenogastropoda and Opisthobranchia. In “Treatise on InvertebratePaleontology. Part I Mollusca 1.’ (Ed. R. C. Moore.) pp. 1310-1331. (University of Kansas Press: Lawrence.) Browne, R. A. (1978). Growth, mortality, fecundity, biomass and productivity of four lake populations of the Prosobranch snail, Viviparus georginaus. Ecology 59, 742—750. Burger, D. (1988). Early Cretaceous environments in the Eromanga Basin: palynological evidence from GSQ Wyandra-1 corehole. Memoirs of the Association of Australasian Palaeontologists 5, 173—186. Burton, G. R., and Mason, A. J. (1998). Controls on opal localisation in the White Cliffs area. Quarterly Notes Geological Survey of New South Wales 107, 1—11. Cantrill, D. J. (1998). Early Cretaceous fern foliage from President Head, Snow Island, Antarctica. Alcheringa 22, 241—258. Chapman, F. (1937). Chert limestone with P/anorbis, from the Mount Elder Range, Western Australia. Proceedings of the Royal Society of Victoria 50, 59-68. Cotton, B. C. (1935). Recent Australian Viviparidae and a fossil species. Records of the South Australian Museum 5, 339—344. De Lurio, J. L., and Frakes, L. A. (1999). Glendonites as a palaeoenvironmental tool: implications for Early Cretaceous high latitude climates in Australia. Geochimica et Cosmochimica Acta 63, 1039-1048. Dettman, M. E., Molnar, R. E., Douglas, J. G., Burger, D., Fielding, C., Clifford, H. T., Francis, J., Jell, P., Rich, T., Wade, M., Rich, P. V, Pledge, N., Kemp, A., and Rozefields, S. A. (1992). Australian Cretaceous terrestrial faunas and floras: biostratigraphic and biogeographic implications. Cretaceous Research 13, 207—262. Douglas, J. G., and Williams, G. E. (1982). Southern polar forests: the Early Cretaceous floras of Victoria and their palaeoclimatic significance. Palaeogeography, Palaeoclimatology, Palaeoecology 39, 171-185. Embleton, B. J. J. (1984). Australia’s global setting: past global settings. In ‘Phanerozoic Earth History of Australia’. (Ed. J. J. Veevers.) pp. 11-17. (Clarendon Press: Oxford.) Etheridge, R. (1897). An Australian sauropterygian (Cimoliasaurus), converted into precious opal. Records of the Australian Museum 3, 21-29. Etheridge, R. (1902). A monograph of the Cretaceous invertebrate fauna of New South Wales. Memoirs of the Geological Survey of New South Wales 11, 1-98. Etheridge, R. (1904). A second sauropterygian converted into opal from the Upper Cretaceous of White Cliffs, New South Wales. With indications of ichthyopterygians at the same locality. Records of the Australian Museum 5, 306-316. Frakes, L. A., and Francis, J. E. (1988). A guide to Phanerozoic cold i climates from high-latitude ice-rafting in the Cretaceous. Nature 333, 547—549. Frakes, L. A., and Francis, J. E. (1990). Cretaceous palaeoclimates. In *Cretaceous Resources, Events and Rhythms.’ (Eds R. N. Ginsberg and B. Beaudoin.) pp. 273-287. (Kluwer Academic Publishers: Dortrecht.) Frakes, L. A., Alley, N. E, and Deynoux, M. (1995). Early Cretaceous ice rafting and climate zonation in Australia. International Geology Review 37, 567-583. Gregory, R. T., Douthitt, C. B., Duddy, I. R., Rich, P. V, and Rich, T. H. (1989). Oxygen isotopic composition of carbonate concretions from the Lower Cretaceous of Victoria, Australia: implications for the evolution of meteoric waters on the Australian continent in a paleopolar environment. Earth and Planetary Science Letters 92, 27-42. Hamilton-Bruce, R. J., Kear, B. P., and Smith, B. J. (in press). A new thiarid snail from Lower Cretaceous freshwater deposits of the Griman Creek Formation, Lightning Ridge, New South Wales. A/cheringa. Hamilton-Bruce, R. J., Smith, B. J., and Gowlett-Holmes, K. L. (2002). Descriptions of a new genus and two new species of viviparid snails (Mollusca: Gastropoda: Viviparidae) from the Early Cretaceous Reassessment of Viviparus (?) albascopularis Etheridge Molluscan Research 157 (middle-late Albian) Griman Creek Formation of Lightning Ridge, northern New South Wales. Records of the South Australian Museum 35, 193—203. Helby, R., Morgan, R., and Partridge, A. D. (1987). A palynological zonation of the Australian Mesozoic. Memoirs of the Australasian Association of Palaeontologists 4, 1—94. Henderson, R. A., Crampton, J. S., Dettmann, M. E., Douglas, J. G., Haig, D., Shafik, S., Stilwell, J. D., and Thulborn, R. A. (2000). Biogeographical observations on the Cretaceous biota of Australasia. Memoirs of the Australasian Association of Palaeontologists 23, 355-404. Hocknull, S. A. (1997). Cretaceous freshwater bivalves from Queensland. Memoirs of the Queensland Museum 42, 223-226. Huber, B. T., and Hodell, D. A. (1996). Middle-Late Cretaceous climate of the southern high latitudes: stable isotopic evidence for minimal pole-to-equator thermal gradients. Reply. Geological Society of America Bulletin 108, 1193—1196. Huber, B. T., Hodell, D. A., and Hamilton, C. P. (1995). Middle-Late Cretaceous climate of the southern high latitudes: stable isotopic evidence for minimal pole-to-equator thermal gradients. Geological Society of America Bulletin 107, 1164—1191. Jaquet, J. B. (1893). On the White Cliffs opal-field. Annual Report of the Department of Mines and Agriculture, New South Wales 1892-1893, 140—142. Jell, P. A., and Duncan, P. M. (1986). Invertebrates, mainly insects, from the freshwater, Lower Cretaceous, Koonwarra Fossil Bed (Korumburra Group), South Gippsland, Victoria. Memoirs of the Association of Australasian Palaeontologists 3, 111—205. Kemp, A. (1991). Australian Cenozoic and Mesozoic lungfish. In 'Vertebrate Palaeontology of Australasia. (Eds P. Vickers-Rich, J. M. Monaghan, R. F. Baird and T. H. Rich.) pp. 465-489. (Pioneer Design Studio, Monash University: Melbourne.) Lowrie, W., Alvarez, W., Premoli-Silva, I., and Monechi, S. (1980). Lower Cretaceous magnetic stratigraphy in Umbrian pelagic carbonate rocks. Royal Astronomical Society Geophysical Journal 60, 263-281. Ludbrook, N. H. (1966). Cretaceous biostratigraphy of the Great Artesian Basin in South Australia. Geological Survey of South Australia Bulletin 40, 7-223. Ludbrook, N. H. (1985). Mesozoic non-marine Mollusca (Pelecypoda: Unionidae) from the north of South Australia. Transactions of the Royal Society of South Australia 84, 139-147. McMichael, D. F. (1957). A review of fossil freshwater mussels (Pelecypoda: Unionidae) of Australasia. Proceedings of the Linnean Society of New South Wales 81, 222—242. MeMichael, D. F. (1968). Non-marine Mollusca from Tertiary rocks in Northern Australia. Bureau of Mineral Resources, Geology and Geophysics Bulletin 80, 133—160. Molnar, R. E. (1980). Australian late Mesozoic terrestrial tetropods: some implications. Memoirs de les Société Géologique de France 139, 131—143. Molnar, R. E. (1991). Fossil reptiles in Australia. In *Vertebrate Palaeontology of Australasia. (Eds P. Vickers-Rich, J. M. Monaghan, R. F. Baird and T. H. Rich.) pp. 605—702. (Pioneer Design Studio, Monash University: Melbourne.) Newton, R. B. (1914). On some molluscan remains from the opal deposits (Upper Cretaceous) of New South Wales. Proceedings of the Malacological Society of London 11, 217-235. Pirrie, D., Doyle, P., Marshall, J. D., and Ellis, G. (1995). Cool Cretaceous climates: new data from the Albian of Western Australia. Journal of the Geological Society of London 152, 739—742. Pledge, N. S., Prescott, J. R., and Hutton, J. T. (2002). A late Pleistocene occurrence of Diprotodon at Hallett Cove, South Australia. Transactions of the Royal Society of South Australia 126, 39-44. Rich, T. H., and Rich, P. V. (1989). Polar dinosaurs and biotas of the Early Cretaceous of southeastern Australia. National Geographic Society Research Reports 5, 15—53. Rich, P. V., Rich, T. H., Wagstaff, B. E., McEwan Mason, J., Douthitt, C. B, Gregory, R. T., and Felton, E. A. (1988). Evidence for low temperatures and biologic diversity in Cretaceous high latitudes of Australia. Science 242, 1403-1406. Selwood, B. W., Price, G. D., and Valdes, P. J. (1994). Cooler estimates of Cretaceous temperatures. Nature 370, 453-455. Shea, M. (1994). The Chinese viviparid snail Bellamya heudei guangdungensis (Kobelt, 1906) in Australia (Prosobranchia: Viviparidae). Molluscan Research 15, 3-11. Sheard, M. J. (1990). Glendonites from the southern Eromanga Basin in South Australia: palacoclimatic indicators for Cretaceous ice. The Geological Survey of South Australia, Quarterly Geological Notes 114, 17-23. 158 Molluscan Research B. P. Kear et al. Sheldon, F., and Walker, K. F. (1993). Shell variation in Australian Notopala (Gastropoda: Prosobranchia: Viviparidae). Journal of the Malacological Society of Australia 14, 59-71. Smith, B. J. (1992). Non-marine Mollusca. In ‘Zoological Catalogue of Australia’, Vol. 8. (Ed. W. W. K. Houston.) pp. 1-398. (Australian Government Publishing Service: Canberra.) Smith, E. (1999). “Black opal fossils of Lightning Ridge.’ (Kangaroo Press: Sydney.) Stevens, G. R., and Clayton, R. N. (1971). Oxygen isotope studies on Jurassic and Cretaceous belemnites from New Zealand and their biogeographic significance. New Zealand Journal of Geology and Geophysics 14, 829—897. Stillwell, J. D. (1998). Latest Cretaceous Bivalvia, Gastropoda and Scaphopoda (Mollusca) from the Chatham Islands, South Pacific: systematics and palaeoecology. Alcheringa 22, 29-85. White, E. I. (1926). On the occurrence of the genus Epiceratodus in the Upper Cretaceous of New South Wales. Annals and Magazine of Natural History 17, 677—682. http://www.publish.csiro.au/journals/mr CSIRO PUBLISHING www.publish.csiro.au/journals/mr Molluscan Research, 2003, 23, 159—178 Relationships of Placostylus from Lord Howe Island: an investigation using the mitochondrial cytochrome c oxidase 1 gene Winston E Ponder, Donald J. Colgan^, Dianne M. Gleeson? and Greg H. Sherley" AAustralian Museum, 6 College Street, Sydney, NSVV 2010, Australia. BEcological Genetics Laboratory, Landcare Research, PB 92170, Auckland, Nevv Zealand. CDepartment of Conservation, PO Box 12-416, Wellington, New Zealand. >To whom correspondence should be addressed. Email: winstonp@austmus.gov.au Abstract Large (5-9 cm in length) land snails of the genus Placostylus are found in New Caledonia and the Loyalty Islands, northern New Zealand, the Three Kings Islands just north of New Zealand and on Lord Howe Island. Their presence on Lord Howe, an oceanic island less than 6 million years old, has been an intriguing biogeographical question. Maximum parsimony and maximum likelihood analyses using cytochrome c oxidase subunit I sequence data suggest that the Lord Howe Island and mainland New Zealand taxa are sisters, but that the Three Kings taxon is independently derived, possibly from New Caledonian stock. Placostylus colonies throughout the area of the present study are under considerable threat, with many intraspecific forms and some species threatened and some listed as endangered species. The taxonomic and conservation status of the Lord Howe Island populations are discussed. Additional keywords: biogeography, New Caledonia, New Zealand, south-west Pacific, systematics. Introduction Oceanic islands typically have high levels of endemism and the origin of their species continues to be an intriguing question for evolutionary biologists. Notable examples among vertebrates of such studies include Darwin's finches of the Galapagos Islands, the honeyeaters of Hawaii (Freed et al. 1987) and Brachylophis on Pacific Islands (Gibbons 1981, 1985). There are also spectacular examples of radiations among invertebrates, some of the better known being the Pacific Island land snails, where there have also been massive human-induced extinctions (e.g. Solem 1990; Cowie 1992, 1996). The likelihood of dispersal of individuals and taxa to oceanic islands is dependent on many factors, including the distance from source populations and the size and habitat complexity of the island. However, overriding all these factors is the dispersal ability of the taxon. Dispersal ability can be determined by intrinsic abilities (power of flight, body size and physiology (e.g. resistance to desiccation, tolerance of salt water, habits or habitat preferences; an arboreal species may be more likely to be transported by wind storms than a ground-living or burrowing species)) or extrinsic factors (prevailing winds, presence of suitable dispersal agents etc.). Land snails have no intrinsic means of long-distance dispersal, although small-sized species, in particular, may be dispersed aerially during major storms, accidentally carried by birds (Rees 1965; Vagvolgyi 1975; Kirchner et al. 1997) or rafted on floating vegetation. Successful long distance passive dispersal for most taxa is rare, especially once communities are established (Ward and Thornton 2000). The improbability of oversea dispersal for some taxa has led to hypotheses involving sunken continents or land bridges. © Malacological Society of Australasia 2003 10.1071/MR03001 1323-5818/03/020159 160 Molluscan Research W. Ponder et al. P. fibratus and five others P. bivaricosus Lord Howe Island P. bollonsi». P. ambagiosus P. hongii ə New Zealand Fig. 1. The SW Pacific, showing the locations of the species of Placostylus included in the present analysis. The ellipse around New Caledonia is intended to show the overall range of the six species of Placostylus recognised from that area. The diamond indicates the locality of the New Caledonian specimen of P. fibratus used in the analysis. Lord Howe Island (31?33'S, 159°05 E) lies in the Tasman Sea, is approximately 11 km long and rises to 875 m. It is nearest to eastern Australia, lying 700 km NE of Sydney and 496 km E of Port Macquarie, the nearest point on the coast of New South Wales (NSW). Norfolk Island is 890 km away and Auckland, New Zealand, is 1560 km away (Fig. 1). The highly endemic fauna and flora have elicited much discussion (see Paramonov 1958, 1960, 1963). The island is of volcanic origin, being formed between 6.4 and 6.9 million years ago (McDougall et al. 1981), and lies on the eastern edge of the Lord Howe Rise, a submarine fragment of eastern Australia that separated in the Cretaceous (Cook and Belbin 1978). Drill cores show that the Rise itself has never been above water (Van der Lingen 1973). A chain of submarine seamounts runs due north of Lord Howe Island. These are assumed to be progressively older northwards (McDougall ef al. 1981). Among the highly endemic fauna and flora is the extinct horned tortoise (Meiolania platyceps Owen, 1886) that was apparently terrestrial with poor swimming abilities (Gaffney 1983, 1996) and with close relationships to taxa in eastern Australia and New Caledonia (Walpole Island; Gaffney 1996). Notable among the endemic flora are three endemic genera of palms. There are many indigenous invertebrates, including over 80 species of land snails (Iredale 1944). Prominent among the land snails is P/acostylus bivaricosus (Gaskoin, 1855), which reaches up to approximately 8 cm in length. Hedley (1892) stated that “... Placostylus appears a more fruitful subject of study [for biogeography] than any other molluscan genus inhabiting the same area' and the mystery of the presence of Placostylus on Lord Howe was also discussed by Etheridge (1891). Placostylus relationships based on mtDNA Molluscan Research 161 The distribution of Placostylus was such a conundrum to Hedley (1892) that he hypothesised that the islands containing these snails were “... portions of a shattered continent [which he called the Melanesian Plateau] and are connected by shallow banks formerly dry land'. This hypothesis met strong opposition from at least one notable contemporary, Alfred R. Wallace (see text of letter to Hedley; Wallace 1974). In the most recent taxonomic review of the group (Haas 1935), species attributed to the genus Placostylus are found through some of the western Pacific Islands (New Zealand, New Caledonia, Solomon Islands, Fiji and Vanuatu). Haas (1935), who divided the genus into several subgenera, restricted the typical subgenus to New Caledonia (type species Limax fibratus Martyn, 1784; ICZN Opinion 1662 (ICZN 1992)). Several ‘subgenera’ are recognised that encompass the Pacific Island and New Zealand taxa, but the New Zealand- Lord Howe and New Caledonian (plus Loyalty Islands) taxa, with their large, solid shells, are more similar to one another than to the taxa further north, as noted by Hedley (1892). Because no other similar species are included in this grouping, it is reasonable to assume that the sister-taxon of the Lord Howe species is from either New Zealand, the Three Kings Islands or New Caledonia. The New Caledonian (including Loyalty Islands) Placostylus are still relatively poorly known, both taxonomically and biologically There are many names available in the literature, but no modern revisions have been published. Franc (1956) recognised 19 species but, using anatomical characteristics, Chérel-Mora (1982, 1983) reduced these to four, although most of her work remains unpublished. Dr E. Neubert, who is currently investigating the taxonomy and anatomy of the group in New Caledonia and the region, stated that, of approximately 140 available names, six valid species and approximately 20 geographic subspecies can be recognised in New Caledonia (Neubert 2001). The shell morphology of some of the New Caledonian taxa is very similar to that of Lord Howe Island and New Zealand taxa. The taxon included in the present analysis is P. fibratus (Fig. 2E,F), the type species of Placostylus. Many of the New Caledonian taxa are threatened (Neubert 2001) and some are still used for food. Salas et al. (1997) provide some data on the biology of P. fibratus. While the systematics of the Lord Howe species is poorly understood (see Appendix), a recent survey of extant populations has been performed (Ponder and Chapman 1999), as well as a preliminary study of the genetics (Colgan and Ponder 2001). Species similar to Lord Howe Island Placostylus occur in northern New Zealand (Powell 1947, 19515, Choat and Schiel 1980; Triggs and Sherley 1993; Sherley 1996) and the Three Kings Islands (off northern-most New Zealand; Powell 19514, Brook and Laurenson 1992). Members of the subgenus Maoricolpus Haas, 1935 (type species Bulimus shongii (= hongii) Lesson, 1830) include the species found in northern New Zealand and Lord Howe Island. The Three Kings Islands species (P. bollonsi Suter, 1908) is placed in a monotypic subgenus, namely Basileostylus Haas, 1935. In New Zealand, there are three main taxa. Placostylus hongii (Fig. 24) has been recorded from sites between Whangaroa and Whangarei on the mainland Northland and on offshore islands between Whangaroa and Great Barrier Island (Powell 1979; Browne 1980; Brook and McArdle 1999). Placostylus ambagiosus Suter, 1906 (Fig. 2B,C), comprising ten named extant subspecies (Powell 1947, 19515), is located in the northern-most tip of the Northland Peninsula, where it is confined to tiny remnant populations. The third taxon, P. bollonsi (Fig. 2D), is confined to the Three Kings Islands that lie 60 km NW of Cape Reinga. This group is comprised of one large island and three small islands, all of which have colonies of this snail (Brook and Laurenson 1992). The biota of these islands also 162 Molluscan Research W. Ponder et al. Fig. 2. Shells of Placostylus species (all material from the Australian Museum, Sydney, NSW; dimensions are maximum length). Images not to same scale. A, P. hongii (Lesson), Tauranga-Kawai Point, 6 km N of Whanaki, Northland, New Zealand, C.114955, 77.6 mm. B, P. ambagiosus paraspiritus Powell, paratypes, headland 1 mile S of Cape Maria van Diemen, Northland, C.115002, 64.4 mm. C, P. ambagiosus whareana Powell, paratype, valley to north of Whareana Stream, between Waikuku Beach and Parengarenga, Northland, C.115001, 82 mm. D, P. bollonsi bollonsi (Suter), paratype, Big King Island, Three Kings Islands, C.29117, 91.5 mm. E,F, P. fibratus (Martyn), Isle of Pines, New Caledonia, C.409415, 84 mm (£) and Bourail, New Caledonia, C.409409, 84 mm (7). includes many other endemics. All the New Zealand species of Placostylus have received considerable taxonomic (Powell 1938, 1947, 1948, 1951a, 19515; Climo 1973; Sherley 1996) and, more recently, conservation attention (Brook and Laurenson 1992; Triggs and Sherley 1993, Sherley 1996; Sherley et al. 1998), including the publication of a Recovery Plan (Parrish et al. 1995). Some of the New Zealand taxa are critically endangered: the Three Kings species, P. bollonsi, is confined to two small populations distinguished as separate subspecies Placostylus relationships based on mtDNA Molluscan Research 163 (Powell 1948, 1951a; Climo 1973), one of approximately 130 individuals in a small area of only approximately 1.69 ha and the other of approximately 360 individuals in approximately 2.7 ha (Brook and Laurenson 1992). These populations are restricted by lack of suitable habitat and are not threatened significantly by predators. Placostylus ambagiosus, a species found only in far northern New Zealand, consists of a number of named subspecies (Powell 1938, 1947, 19515) confined to tiny remnant populations that show some degree of genetic structuring using isozyme electrophoresis (Triggs and Sherley 1993). Some of these populations have fewer than 10 living individuals (Parrish ef al. 1995) and all are threatened by introduced predators (rats, birds and pigs). Placostylus hongii, which lives south of P. ambagiosus in Northland, is known from a few mainland locations and some offshore islands and is extant from five mainland populations and three offshore island groups (Brook and McArdle 1999). Translocation of the two mainland species has been undertaken for conservation purposes (Sherley 1994). The broader phylogenetic relationships of Placostylus within the Bulimulidae are still not fully understood. Solem (1959) argued for a relationship between Bothriembryon, the New Hebridean (i.e. Vanuatu) Diplomorpha and Placostylus based on anatomical data, with the latter genus having a northern origin. However, Breure (1979) placed Placostylus in a separate subfamily that he regarded as the sister-taxon to the rest of the family. He hypothesised that members of the Placostylinae reached New Zealand via east Antarctica and moved north. According to Breure (1979), placostylines show little relationship with the buliminids found in the western and southern parts of Australia (Bothrembrion; subfamily Orthalicinae). The present paper provides evidence from molecular data relating to the question of the origin of the Lord Howe Island Placostylus. Because the “subspecific” taxa (see Appendix) on Lord Howe Island are extinct, their status cannot be assessed using this methodology. Material and methods Material A list of the specimens sequenced is given in Table 1. The sequenced specimen of P. fibratus came from a population with intermediate characteristics between P. fibratus fibratus and P. fibratus souvillei Kobelt, 1891 (Dr E. Neubert, personal communication). -- DNA extraction and sequencing Lord Howe Island and New Caledonia specimens were extracted and sequenced in Sydney, whereas New Zealand specimens were extracted and sequenced at Landcare Research, Auckland. For New Zealand specimens, foot muscle tissue was homogenised in 500 uL extraction buffer (0.01 M EDTA, 0.05 M NaCl, 0.5 M Tris-HCl, pH 8.0, 2% sodium dodecyl sulfate). A 10 uL aliquot of 10 mg mL `° ! proteinase K (Boehringer, Mannheim, Germany) was added and the homogenate incubated at 65?C for 1 h. Samples were extracted twice with phenol/chloroform/isoamyl alcohol (25:24: 1), followed by extraction with chloroform/isoamyl alcohol (24:1). The DNA was ethanol precipitated and the pellet rehydrated in 50 uL buffer (10 mM Tris Cl (pH 7.4), 1 mM EDTA (pH 8.0)) following RNAse treatment. A 655 bp region of the cytochrome c oxidase 1 (COI) gene was amplified by polymerase chain reaction (PCR) using primers LCO-1490 (5”-GGTCAACA AATCATA AAGATATTGG-3 P) and HCO-2198 (5 - TA AACTTCAGGGTGACCAAAA A ATCA-3?) designed by Folmer et al. (1994). Amplifications were performed in a volume of 50 uL and consisted of 10 pmol of each primer, 10 mM Tris-Cl, pH 8.3, 1.5 mM MgCl, 50 mM KCl and 0.2 mM of each dNTP. The addition of 2 units Taq polymerase (Boehringer) followed an initial step of 2 min denaturation at 94°C. Cycling consisted of denaturation at 94°C for 1 min, annealing at 50°C (CO/) for 1 min and extension at 72°C for 1 min and 30 s, for 35 cycles. A final cycle | included a 5 min extension at 72?C. The PCR products were purified using the QIAquickTM PCR direct purification kit (Qiagen, Venlo, The Netherlands), according to the manufacturer’s instructions. Direct sequencing of purified products by the W. Ponder et al. Molluscan Research 164 "MSN “Soups "umosnj uemersny əu ur pəsnou si [errojeur JOYONOA 1970 :pueppony *ueqry IN 1 uonoə|loO podompry pueysəz MƏN IY} ut pəöpol st perou put|eoZ MON AL “AIBAO “AQ :opueur “TN 300y “1 ‘pues oansosiq “Dd — o. — — “ I — Hk x... d H,C£ə991 S,0Tecc ‘W OOS ÁAporeurxoudde je siio sjuojJ Jo adojs MS PUO - LIF60rO pIuop9[eO MON smpuaqlf q AO “N ‘Dd ,,6S,60:Z21 SS, LOE ‘SBUTY oou] FS9 EPE LOT €t-Ip SD JSLIuou, `q 4 ^O 'IN 'DG IP, LOTLI ,, CC, 60s b€ 'SSurx ood] LC8 60€ 1071 01-9£ SD [[9^0gq sninq.4p 'q q AO 'IN ‘Da SC, 80sCLT ,, £L, 60s b€ 'SSurx INL OE8 61£ 1071 SE-TE SD [emo snjn42dpo `q q AQ “N ‘Dad ,,Ür,80:C21 ,.90,60-P£ 'sdury PAYL c£8 ETE 1071 0£-9c SD !suo]]oq q q wa PUETEƏZ MƏN 'spuv[s| ssury IAL 1suo]]og q pue[997Z MƏN PULJYHON Wod ‘C SP IZobLI ,,07, CCo SE) SAIOSAY Əlu99S peor] YHON nange3ueu Ay $98-C98 OM put[e9Z MON 'pue[q1oN. 13uoi q W ‘Od SL IPoTLI ,, £0,8ZoP£ “SUL PiEq IL EST 106 CON SC-IC SD [Mog andajsa] *D d AO ‘W "Dd VC, VSoCLI LI, Scope 'neeanepq PES STO CON 0c ‘SISD W ‘Oa ,,FS,SSoCLT ,, 0L SCob€ “BINH 9L LCS 80 TON 61 ‘LI ‘91 SD |[9^oq suaj22uup `p q ^O 'W ‘Dd Lt. CO0s£LT ,, Lr, E Cope S09 “adep YON EPS 8rl ION 9 SD Proq WDM "D q W ‘Od bP ProCL 1.67, 9ToPE “Ed EMENN IRON TIS LL8 TOW ES8-OS8OM PAYLI, "D q A pue|e9Z MƏN "pue[qoN “ou Ivy snsoi8DquiD q a ,,1£,£9681 ,,80, £o 18 “eg YHON L89L8€O 8-Ld a ,, YO, Fo6S1 ,, LT, T EoI E '&yrodoud xoəsiH €L9L8€O Sd d CS, £06S1 ,,0Z, 1£o1£ “YOR SpoN CVLL8EO td 4 ULV, Eo6S1 ,, Ip, [£o T€ Ərousəroj uoosr7 0994862 9 “Ça d 9S, £06S1 ,, 8C, [ol € 'oA10soq s uoudojs 0994862 Cd d ,,6C,te6S1 ,, 6C, CEo TE 10dary 1eou 59104 8€9,8tO Id a PUPISI 90H pio SHSOOLIDAIQ q SonssI] Áj[e900] "ou uməsnyy uouioodg een SITIO] 194} pue pəəuənbəs suourods jo ISI "1 AQEL Placostylus relationships based on mtDNA Molluscan Research 165 primers used for the original amplification was achieved using the Prism Ready Reaction Dye Deoxy Terminator Cycle Sequencing Kit (Applied Biosystems, Norwalk, CT, USA), following the manufacturer's instructions. Sequences were analysed on an automated DNA sequencer (model 377; Applied Biosystems). Two independent PCR products from each specimen were sequenced in both directions. For the Sydney experiments, molecular biological procedures generally followed the methods of Colgan et al. (2000a, 20005). The DNA extraction of foot muscle followed the hexadecyltrimethylammonium bromide procedure of Saghai-Maroof et al. (1984). The Sydney primer pair for CO/ is listed below with degeneracies written with standard TUB codes. Numbers in parentheses indicate the 5” end of the sequence in the complete mitochondrial genome of the earthworm Lumbricus terrestris (Boore and Brown 1995) COXIAFE (19) CWAATCAYA AAGATATTGGAAC; COX IAR, (726) AATATAWACTTCWGGGTGACC. Polymerase chain reaction cycling used the same basic protocols as Colgan et al. (2000a, 20002), with a general reaction mix of 1.0 U Red Hot™ thermostable DNA polymerase, buffer IV (10x, 20 mM (NH4),SO,, 750 mM Tris-HCL, pH 9.0, 0.196 (w/v) Tween; Advanced Biotechnologies, Leatherhead, Surrey, UK), 0.05 mM dNTPs, 2.5 mM MgCl,,12.5 pmol of each primer and 1 uL of a dilution (usually 1:10) DNA sample in a reaction volume of 50 uL overlaid with 30 uL oil. The usual cycling profile was denaturation at 95°C for 4 min, annealing at 45°C for 45 s and extension at 72°C for 30 s (one cycle), followed by 32 cycles of 95°C for 30 s, 45°C for 45 s and 72°C for 30 s and one cycle of 95°C for 30 s, 45°C for 45 s and 72°C for 5 min. The PCR products were purified using the QIAquickTM PCR Purification Kit and sequenced in both directions with an automated DNA sequencer (ABI? 310; Applied Biosystems) using the BigDye™ version 2.0 sequencing kit. Alignment and sequence composition Sequences were edited using Sequence Navigator version 1.0.1 (Anonymous 1994). A compilation of all sequences was aligned using the default values for parameters in CLUSTAL W (Thompson ef al. 1994). It was not necessary to alter this alignment manually.-Data were compiled using MacClade (Maddison and Maddison 1992). Maximum parsimony was conducted in PAUP* 4.0b9 (Swofford 2000) with the default conditions for parsimony analyses with branch and bound searches guaranteed to find the shortest trees. All characteristics were unordered and unweighted. The steepest descent option was not enforced and accelerated transformation for character optimisation was assumed. Zero length branches were collapsed to give polytomies. Gaps were treated as unknown in all analyses. Bootstrap pseudo-sampling was conducted for 100 replicates with branch and bound searching in each. Shortest trees satisfying alternative hypotheses of relationships were sought with the same strategy but using constraints in PAUP. For maximum likelihood analyses in PAUP*, a Hasegawa-Kishino-Yano (HKY) invariant model was used with the following settings. The number of substitution types was two, with the transition transversion ratio estimated by maximum likelihood. Empirical nucleotide frequencies were used. The proportion of invariable sites was estimated via likelihood and the distribution of rates at variable sites was assumed equal. The molecular clock model was not enforced. Other parameters took the default values in PAUP* 4. Comparison of trees using Kishino-Hasegawa likelihood tests assumed a normal approximation, with a two-tailed test. Shimodaira-Hasegawa tests used a one-tailed resampling of estimated log-likelihoods bootstrap with 1000 replicates. Outgroup sequences were obtained from the helicoidean pulmonates Aegista scepasma (Bradybaenidae) (Genbank Accession AB024900; Shimizu and Ueshima 2000), Cepaea nemoralis (Helicidae) (Genbank Accession U23045; Yamazaki ef al. 1997), Euhadra herklotsi (Bradybaenidae) (Genbank Accession Z71701; Yamazaki ef al. 1997) and Albinaria caerulea (Genbank Accession NC. 001761; Hatzoglou et al. 1995). Morphometrics Shell measurements were made using callipers. The discriminant analysis was undertaken using SYSTAT™ version 10 (SPSS, Chicago, IL, USA). Results The locations of samples are shown in Table 1. GenBank accession numbers for the CO/ sequences used are AY 165836—AY 165843, AY 165852 and AY290737—AY 290745. The alignment used in the analysis is available from 166 Molluscan Research W. Ponder et al. the authors. As usual with mitochondrial sequences, there is a major AT bias in nucleotide composition. The mean overall percentages of nucleotides is as follows: A, 26.844; C, 14.653; G, 17.174; T, 41.328 (overall AT average 68.17%). The probability that the nucleotide composition ofthe sequences is homogeneous is very close to 1 (x? = 37.600, d.f. = 63, P = 0.99), suggesting that variation in base composition between taxa has little effect on the analyses. The levels of pairwise difference of the HKY measure of genetic distance within the Placostylus samples vary between 0.0188 (within Lord Howe Island) and 0.6278 (between P. ambagiosus annectens and P bollonsi “north-east”) within New Zealand samples. In Table 2, the averages of the Kimura two-parameter distance are given for comparisons within and between various species. Within species, the averages are less than 0.1, but distances between species rise to 0.2014 for the comparison of Lord Howe Island snails with P. bollonsi. The results are illustrated in Figs 3, 4. The consensus of the maximum parsimony trees (Fig. 3) includes the following strongly supported monophyletic clades: Lord Howe Island specimens (with a decay index of 27), P. ambagiosus, P. ambagiosus + P. hongii and P bollonsi. Notably, the New Zealand samples are split into two distinct lineages. Constraining these to be monophyletic requires five more steps than the unconstrained, most parsimonious trees. The Lord Howe Island samples are sister to the P. ambagiosus + P. hongii clade. This group is, in turn, sister to the New Caledonian sample. Four extra steps are required to make the Lord Howe specimens sister to this to the exclusion of the New Zealand snails. A sister pairing of Lord Howe and New Caledonia specimens is seen in only 6.6% of bootstrap trees compared with 38% of bootstrap replicates showing the pairing of Lord Howe Island with P. ambagiosus + P. hongii. The sister pairing of New Zealand P. ambagiosus + P. hongii with New Caledonia or all the New Zealand taxa with New Caledonia requires, respectively, two and five more steps than the maximum parsimony tree. Comparisons of the likelihoods of the trees satisfying various constraints are shown in Table 3 and the maximum likelihood tree is shown in Fig. 4. In particular, the Kishino— Hasegawa tests reject the possibility that there is a single New Zealand radiation that is Table 2. Average of pairwise genetic distances within and between Placostylus species using the Kimura two-parameter genetic distance = = omma c RR tame Comparisons Average —— — —— E — — `> Within species P. ambagiosus 0.0144 P. bivaricosus 0.0630 P. bollonsi 0.0028 Between species P. ambagiosus/P. bivaricosus 0.1581 P. ambagiosus/P. bollonsi 0.1755 P. ambagiosus/P. fibratus 0.1345 P. ambagiosus/P. hongii 0.0572 P. bivaricosus/P. bollonsi 0.2014 P. bivaricosus/P. fibratus 0.1717 P. bivaricosus/P. hongii 0.1494 P. bollonsi/P. fibratus 0.1849 P. bollonsi/P. hongii 0.1732 P. fibratus/P. hongii 0.1325 ————————————MÁ—ÓRÉÓÉÓÉÓÉÓR—€——— Placostylus relationships based on mtDNA 100 3 100 27 3 a 100 | gg 42 | 2 R 96 7 70 5 53 100 1 14 52 Molluscan Research Aegista Cepaea Albinaria Euhaara LHI1 LHI2 LHI4 Placostylus bivaricosus Lord Howe Island LHI5 LHI6 LHI7 LHI8 LHI9 P. a. annectans Placostylus P. a. lesleyae ambagiosus North Cape area P.a. "irikavva" Northland P. a. watti Placostylus hongii Northland Placostylus fibratus New Caledonia P. b. arbutus Placostylus bollonsi Three Kings Islands P. b. bollonsi P. b. caperatus P. b. 'northeast' 167 Fig. 3. The strict consensus of 50 trees found in a branch and bound search of the part cytochrome c oxidase 1 gene. The trees were 556 steps long with a consistency index of 0.678. Bootstrap percentages above 50% are written above branches and Bremer decay indices below. The species names for Placostylus are indicated in bold. 168 Molluscan Research W. Ponder et al. Table3. Comparisons of the likelihoods of trees constrained to show various relationships Tree -InL Diff —In L Kishino— Shimodaira— Hasegavva Hasegawa Unconstrained 3019.61928 NC with NZ 3025.29525 5.67597 0.000* 0.182 NC with LHI 3023.03754 3.41826 0.000* 0.346 LHI with NC + part NZ 3023.02717 3.40789 0.000* 0.339 l ii 0“ U e The first column shows the imposed constraint (if any), the second column shows the negative of the log likelihood of the maximum likelihood (ML) trees, the third column shows the difference between tree likelihoods and the last two columns give the probabilities that the compared trees have the same likelihood under the Kishino-Hasegawa (two-tailed) test and the Shimodaira-Hasegawa (one-tailed) test. The constraints are that: (/) New Caledonia (NC) and all New Zealand (NZ) specimens are a sister pair; (2) New Caledonia and all Lord Howe Island (LHI) snails form a sister pair; and (3) the Lord Howe Island snails are sister to a group comprising the New Caledonian sample and New Zealand P. hongii and P. ambagiosus. sister to the New Caledonian sample (constraint 1) or that this latter specimen and the Lord Howe Island snails are sister taxa (constraint 2). Discussion New Caledonia and New Zealand were separated by the end of the Cretaceous (Hall 2002), but it is unclear as to how much of the Norfolk Ridge remained emergent through the early Tertiary. The oldest basalts on Lord Howe Island were formed only 6.4 million years ago. Thus, the geological evidence indicates that dispersal, rather than vicariance, must account for the presence of Placostylus on Lord Howe Island. While we can only speculate about possible dispersal mechanisms, an analysis such as this can provide evidence for the likely origin of the source for the dispersal event. This does not eliminate the probability that past intermediate populations, for example, on islands that are now submerged seamounts, ridges or plateaus to the north of Lord Howe Island (or the Three Kings Islands), may have facilitated dispersal as “stepping stones”. The Three Kings Islands are comprised of indurated, deformed marine sedimentary and volcanic rocks of probable early Cretaceous age (Haywood and Moore 1987; Brook and Laurenson 1992). Brook and Laurenson (1992) state that “... it is possible that marine conditions persisted in the vicinity of present day Three Kings Islands for much or all of that time” (Cretaceous to early Caenozoic). The present configuration of northern New Zealand (including the Three Kings) was probably developed during the mid-Caenozoic, with a subaerial peninsula developed from about Kaitaia/Doubtless Bay to the Three Kings area in the early Miocene (15-20 million years ago), this degenerating to a few islands by the mid-late Miocene (Brook and Thrasher 1991). The northern-most end of Northland (*North Cape") was, at that time, a separate island and was connected to the rest of Northland subsequent to the early Pliocene (5-3 million years ago; Pillans et al. 1992). Whereas dispersal must be invoked to account for the presence of Placostylus on Lord Howe Island, the Three Kings Islands could, seemingly, have been connected or nearly connected to Northland in the mid-Tertiary and closely adjacent (approximately 10 km) during interglacial periods in the Pleistocene (Brook and Laurenson 1992). Given these data, postulating an origin of P. bollonsi from Northland seems logical and has been argued for by Brook and Laurenson (1992). However, there is little support for this hypothesis from our data, which suggest that the Three Kings taxon and the mainland taxa represent two Placostylus relationships based on mtDNA Molluscan Research 169 Aegista Cepaea Euhadra Albinaria LHI1 LHI2 LHI4 LHI5 LHI8 LHI9 LHI6 LHI7 P. a. annectans P. a. watti P. a. lesleyae P.a. 'tirikawa' P. hongii P. fibratus P. b. arbutus P. b. bollonsi P. b. caperatus P. b. 'northeast' —— 10 changes Fig.4. Phylogram for the cytochrome c oxidase 1 data of the maximum likelihood tree resulting from an heuristic search with 10 replicates of random sequence additions. See Fig. 3 for details of taxa. separate clades, as reflected in the allocation of different subgeneric names. It appears that the Three Kings taxon is the result of either a long overseas dispersal event or (more probably, in our opinion) it represents a surviving relict of a taxon distributed on now submerged islands along the Norfolk Ridge. Our results clearly suggest a sister relationship between the northland P. ambagiosus and P. hongii, separated by the late Pliocene isthmus between what was once a northern island. Whereas there is strong support for the monophyly of each of the main radiations (P. ambagiosus, P. hongii; P. bollonsi and P. bivaricosus), regrettably, the analysis IS weakened by there being only a single sample from New Caledonia. Brook and McArdle (1999) have advocated pre-European Maori transport of Placostylus hongii (which was used as a food item; Haywood and Brook 1981) to and from offshore islands to account for their present distribution. There is no evidence of pre- European contact with Lord Howe Island, so some other form of overseas dispersal must have occurred. This most likely occurred as a result of an individual being carried in a 170 Molluscan Research W. Ponder et al. severe storm to the island. Whereas this may have come from a nearby population on an island (that has since disappeared) to the north, given that there is a chain of seamounts northwards, the results of our analysis suggest that there was a common origin with the mainland New Zealand taxa. Self-fertilisation is common in pulmonate snails, so even a single juvenile could establish a population. Placostylus hatchlings are known to be often arboreal, this habit facilitating such a mode of dispersal of this otherwise ground-dwelling species. Extremely rare aerial dispersal of hatchlings and juveniles, less than 1 cm in length, during major storms seems, to us, to be a more feasible explanation of dispersal than other hypotheses invoking rafting or transport by birds. The results show low levels of genetic divergence between individual Lord Howe Island sequences (Fig. 3) but, on the available small amount of material, no clear pattern emerges (16S rDNA sequences are also available for the Lord Howe Island samples and show substructure between the available populations (Colgan and Ponder 2001)) There is considerable variation in shell morphology within the area containing the material used in the analysis (Ponder and Chapman 1999) but, again, the material sampled is not adequate to address questions relating to whether there is any underlying genetic basis for this variation. The range of the available samples from Lord Howe Island does not cover all the island, with none coming from the hills and mountains in the southern part where Placostylus appears to be extinct. While it remains possible that more southerly populations, from the remote areas around Mounts Lidgbird and Gower, may still be extant, it remains untested whether they are genetically distinct, as suggested by their shell morphology, or whether there is a significant geographic barrier to gene flow in this region. Triggs and Sherley (1993) showed that there was a considerable level of genetic variation between populations of P. ambagiosus. In these cases, small but overlapping differences in shell morphology led to prior creation of several subspecies (Powell 1938, 1947, 1948, 19515; Sherley 1996). The genetic status of New Zealand Placostylus species has been assessed by allozyme electrophoresis (Triggs and Sherley 1993). Evidence of distinct lineages (possibly at the species level) was found in one mainland morphospecies, P. ambagiosus, over a geographic scale of tens of kilometres, with differentiation apparently fostered by Pleistocene and late Pliocene sea level changes. These authors found no significant variation in allozymes in P. bollonsi from the Three Kings Islands, but the results presented herein suggest that at least the north-east population may be distinct. The results clearly discount the Lord Howe Island taxon being a recent introduction from New Zealand but, because we have only one specimen from New Caledonia, where there are probably six Recent species present, the possibility of a relatively recent introduction from that source cannot be discounted. Conservation status of Lord Howe Island Placostylus Etheridge (1889) noted that P/acostylus on Lord Howe Island was *... found everywhere under cover, and in immense numbers’ and that it “... appears to avoid open spaces as a rule, and prefers shady damp situations and the scrubby hill sides where composed of the Coral- sand rock. It is sparingly represented even at the higher altitudes, being reported as seen under the ‘wall’ of Mount Ledgbird.’ Hedley (1891) noted that it was °... all over the island in sheltered places under stones; abundant’. However, this previously abundant animal has been the subject of conservation concern for some time. Iredale (1944) noted that there was a “... new difficulty of recent years, the destruction effected by the rat plague, written up by Hindwood'. More recently, Smithers Placostylus relationships based on mtDNA Molluscan Research 171 et al. (1977) noted that Placostylus is apparently “... existing in only four small colonies, whereas from early reports it was dispersed widely from sea-level up to the mountain tops”. Ponder and Chapman (1999) reported that all the living specimens found in their 1999 survey were located on low, rather flat ground with ample litter. In most cases, the samples were taken on calcarenite soils or sandy soils derived, in large part, from calcarenite. Living specimens were found under well-developed litter. Placostylus seem to prefer the vicinity of the large native fig trees, but not exclusively so. Placostylus bivaricosus is listed as endangered in NSW under the Threatened Species Conservation Act and a Recovery Plan (NSW National Parks and Wildlife Service 2000) has been prepared. Acknowledgments The collection of the New Caledonian P/acostylus was arranged by Dr P. Bouchet, facilitated by Bertrand Richer De Forges and collected by Alain Lapetite. Pam Da Costa sequenced this sample. Michael Murphy, Dean Hiscox and Lisa Menke assisted with the collection of material on Lord Howe Island and the Lord Howe Island Board gave permission to undertake the work. DMG and GHS thank Richard Parrish from the Department of Conservation (New Zealand) for field assistance. Joshua Studdert assisted with preparing the map and the plates and with photographing specimens. Holly Barlow and Joshua Studdert took the measurements used in the shell comparisons. References Anonymous (1994). “Sequence Navigator: DNA and Protein Sequence Comparison Software', version 1.0.1. (Applied Biosystems: Foster City, CA.) Boore, J. L., and Brown, W. M. (1995). Complete sequence of the mitochondrial DNA of the annelid worm Lumbricus terrestris. Genetics 141, 305-319. Breure, A. S. H. (1979). Systematics, phylogeny and zoogeography of Bulimulinae (Mollusca). Zoologische Verhandelingen 168, 1—215, 3 pls. Brook, F. J., and Laurenson, C. M. (1992). Ecology and morphological variation in Placostylus bollonsi (Gastropoda: Bulimulidae) at Three Kings Islands, New Zealand. Records of the Auckland Institute and Museum 29, 135-166. Brook, F. J., and McArdle, B. H. (1999). Morphological variation, biogeography and local extinction of the northern New Zealand landsnail Placostylus hongii (Gastropoda: Bulimulidae). Journal of the Royal Society of New Zealand 29, 407-434. Brook, F. J., and Thrasher, G. P. (1991). Cretaceous and Cenozoic geology of northernmost New Zealand. New Zealand Geological Survey Records 41, 1-42. Browne, G. H. (1980). A note on Placostylus hongii (Lesson) from Fanal Island. Tane 26, 61-62. Chérel-Mora, C. (1982). Geographic variation of the land snails Placostylus (Bulimulidae, Stylommatophora) in New Caledonia (preliminary report). Malacologia 22, 509—510. Cherel-Mora, C. (1983). ‘Variation, géographique et taxonomie des Placostylus (Gastéropodes, Pulmones, Stylommatophores) en Nouvelle-Calédonie.' Doctoral Thesis. (Université P. et M. Curie, Paris.) Choat, J. H., and Schiel, D. R. (1980). Population structure of Placostylus hongii (Gastropoda Paraphantidae (sic)) on the Poor Knights Islands. New Zealand Journal of Zoology 7, 199—205. Climo, F. M. (1973). The systematics, biology and zoogeography of the land snail fauna of Great Island, Three Kings Group, New Zealand. Journal of the Royal Society of New Zealand 3, 565-87. Colgan, D. J., and Ponder, W. F. (2001). *Preliminary Assessment of the Genetics of Placostylus bivaricosus on Lord Howe Island,’ Unpublished report by the Australian Museum, Sydney. (NSW National Parks and Wildlife Service: Coffs Harbour.) Colgan, D. J., Ponder, W. E, and Eggler, P. E. (2000a). Gastropod evolutionary rates and phylogenetic relationships assessed using partial 28S rDNA and histone H3 sequences. Zoologica Scripta 29, 29-63. Colgan, D. J, Zhang C.-G., and Paxton, J. R. (20005). Phylogenetic investigations of the Stephanoberyciformes and Beryciformes, particularly whalefishes (Euteleostei: Cetomimidae), based on partial 12S rDNA and 16S rDNA sequences. Molecular Phylogenetics and Evolution 17, 15-25. 172 Molluscan Research W. Ponder et al. Cook, K. A. W., and Belbin, L. (1978). The southwest Pacific area during the last 90 million years. Journal of the Geological Society of Australia 25, 23-40. Cowie, R. H. (1992). Evolution and extinction of land Partulidae, endemic Pacific Island land snails. Philosophical Transactions of the Royal Society of London 335, 167—191. Cowie, R. H. (1996). Pacific island land snails: relationships, origins, and determinants of diversity. In “The Origin and Evolution of Pacific Island Biotas, New Guinea to Eastern Polynesia: Patterns and Processes”. (Eds A. Keast and S. E. Miller.) pp. 347-372. (SPB Academic Publishing: Amsterdam.) Etheridge, R. (1889). The general zoology of Lord Howe Island: containing also an account of the collections made by the Australian Museum collecting party, Aug.-Sept., 1887. Memoirs of the Australian Museum 2, 1-42, pls 4-5. Etheridge, R. (1891). A much thickened variety of Bulimus bivaricosus (Gaskoin) from Lord Howe Island. Records of the Australian Museum 1, 130—134, pl. 20. Folmer, O., Black, M., Hoeh, W., Lutz, R., and Vrijenhoek, R. (1994). DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology 3, 294—299, Freed, L. A., Conant, S., and Fleischer, R. C. (1987). Evolutionary ecology and radiation of Hawaiian passarine birds. Trends in Ecology and Evolution 2, 196-203. Franc, A. (1956). Mollusques terrestres et fluviatiles de l'archipel Néo-Calédonien. Mémoires du Muséum National d'Histoire Naturelle Série A Zoologie 13, 1—200, 24 pls. Gaffney, E. S. (1983). The cranial morphology of the extinct horned turtle, Meiolania platyceps, from the Pleistocene of Lord Howe Island, Australia. Bulletin of the American Museum of Natural History 175, 361-479. Gaffney, E. S. (1996). The postcranial morphology of Meiolania platyceps and a review of the Meiolaniidae. Bulletin of the American Museum of Natural History 229, 1—166. Gibbons, J. R. H. (1981). The biogeography of Brachylophus (Iguanidae) including the description of a new species, B. vitiensis, from Fiji. Journal of Herpetology 15, 255-273. Gibbons, J. R. H. (1985). The biogeography and evolution of Pacific Island reptiles and amphibians. In “Biology of Australasian Frogs and Reptiles’. (Eds G. Grigg, R. Shine and H. Ehmann.) pp. 125-142. (Royal Zoological Society of NSW: Sydney.) Haas, F. (1935). Beschreibung neuer Untergattungen und Arten von Mollusken. Zoologischer Anzeiger 109, 188-195. Hall, R. (2002). Cenozoic geological and plate tectonic evolution of SE Asia and the SW Pacific: computer- based reconstructions, model and animations. Journal of Asian Earth Sciences 20, 353-431. Hatzoglou, E., Rodakis, G. C., and Lecanidou, R. (1995). Complete sequence and gene organization of the mitochondrial genome of the land snail Albinaria coerulea. Genetics 140, 1353-1366. Haywood, B. W., and Brook, F. J. (1981). Exploitation and redistribution of flax snail (Placostylus) by the prehistoric Maori. New Zealand Journal of Ecology 4, 33-36. Haywood, B. W., and Moore, P. R. (1987). Geology of the Three Kings Islands, northern New Zealand. Records of the Auckland Institute and Museum 24, 215-232. Hedley, C. (1891). The land and freshwater shells of Lord Howe Island. Records of the Australian Museum 1, 134-144. : Hedley, C. (1892). The range of Placostylus: a study in ancient geography. Proceedings of the Linnean Society of New South Wales 7 (2nd series), 335-339. ICZN (1992). Opinion 1662. Limax fibratus Martyn, 1784 and Nerita hebraea Martyn, 1786 (currently Placostylus fibratus and Nerita hebraea; Mollusca, Gastropoda): specific names conserved; and Placostylus Beck, 1837; L. fibratus designated as the type species. Bulletin of Zoological Nomenclature 49, 74—75. Iredale, T. (1944). The land Mollusca of Lord Howe Island. Australian Zoologist 10, 299-334. Kirchner, C., Kratzner, R., and Welter-Schultes, F. W. (1997). Flying snails: how far can Truncatellina (Pulmonata: Vertiginidae) be blown over the sea? Journal of Molluscan Studies 63, 479-487. Maddison, W. P., and Maddison, D. R. (1992). “MacCLADE”, version 3.0. (Sinauer Associates: Sunderland, MA.) McDougall, I., Embleton, B. J. J., and Stone, B. D. (1981). Origin and evolution of Lord Howe Island, Southwest Pacific Ocean. Geological Society of Australia Journal 28, 155—176. Neubert, E. (2001). Placostylus revisited: unraveling the puzzle of the big bulimes of New Caledonia. In "Abstracts, World Congress of Malacology 2001, Vienna, Austria’. (Eds L. Salvini-Plawen, J., Voltzow, H. Sattmann and G. Steiner.) p. 244. (Unitas Malacologica: Vienna.) Placostylus relationships based on mtDNA Molluscan Research 173 New South Wales National Parks and Wildlife Service. (2000). “Lord Howe Island Large Land Snail Placostylus bivaricosus (Gaskoin, 1855) Recovery Plan.’ (NSW National Parks and Wildlife Service: Coffs Harbour.) Paramonov, S. J. (1958). Lord Howe Island, a riddle of the Pacific. Pacific Science 12, 82-91. Paramonov, S. J. (1960). Lord Howe Island, a riddle of the Pacific. Part II. Pacific Science 14, 75-85. Paramonov, S. J. (1963). Lord Howe Island, a riddle of the Pacific. Part III. Pacific Science 17, 361—373. Parrish, R., Sherley, G., and Aviss, M. (1995). “Giant Land Snail Recovery Plan. P/acostylus spp., Paryphanta sp.’ (Department of Conservation: Wellington.) Pillans, B., Pullar, W. A., Selby M. J., and Soons, J. M. (1992). The age and development of the New Zealand landscape. In “Landforms of New Zealand”, 2nd edn. (Eds J. M. Soons and M. J. Selby.) pp. 31-62. (Longman Paul: Auckland.) Ponder, W. E, and Chapman, R. (1999). “Survey of Placostylus bivaricosus on Lord Howe Island.’ Unpublished report by the Australian Museum, Sydney. (NSW National Parks and Wildlife Service: Coffs Harbour.) Powell, A. W. B. (1938). The genus Placostylus in New Zealand. Records of the Auckland Institute and Museum 2, 33-150. Powell, A. W. B. (1947). Distribution of P/acosfylus land snails in northernmost New Zealand. Records of the Auckland Institute and Museum 3, 173-188. Powell, A. W. B. (1948). Land Mollusca of the Three Kings Islands. Records of the Auckland Institute and Museum 3, 273-290. Powell, A. W. B. (1951a). Land Mollusca from four islands of the Three Kings group, with descriptions of three new species. Records of the Auckland Institute and Museum 4, 127-133. Powell, A. W. B. (19515). On further colonies of Placostylus land snails from northermost New Zealand. Records of the Auckland Institute and Museum 4, 134—140. Powell, A. W. B. (1979.) *New Zealand Mollusca. Marine, Land and Freshwater Shells.' (Williams Collins: Auckland.) Rees, W. J. (1965). The aerial dispersal of Mollusca. Proceedings of the Malacological Society of London 36, 269—282. Saghai-Maroof, M. A., Soliman, K. M, Jorgensen, R. A., and Allard, R. W. (1984). Ribosomal DNA spacer length variation in barley: Mendelian inheritance, chromosomal location and population dynamics. Proceedings of the National Academy of Science of the United States of America 81, 8018-8021. Salas, M., Bonnaut, C., Le Bel, S., and Chardonnet, L. (1997). Activity and food intake of captive Placostylus fibratus (Gastropoda: Bulimulidae) in New Caledonia. New Zealand Journal of Zoology 24, 257-264. Sherley, G. H. (1994). Translocations of the Mahoenui Giant Weta Deinacrida n. sp., and Placostylus land snails in New Zealand: what have we learnt? In ‘Reintroduction Biology of Australian and New Zealand Fauna’. (Ed. M. Serena.) pp. 57-63. (Surrey Beatty and Sons: Chipping Norton, NSW.) Sherley, G. H. (1996). Morphological variation in the -shells of Placostylus species (Gastropoda: Bulimulidae) in New Zealand and implications for their conservation. New Zealand Journal of Zoology 23,73-82. Sherley, G. H., Stringer, I. A. N., Parrish, G. R., and Flux, I. (1998). Demography of two landsnail populations (Placostylus ambagiosus, Pulmonata: Bulimulidae) in relation to predator control in the far north of New Zealand. Biological Conservation 84, 83-88. Shimizu, Y., and Ueshima,R. (2000) Historical biogeography and interspecific mtDNA introgression in Euhadra peliomphala. Heredity 85, 84—96. Smith, B. J. (1992). *Non-marine Mollusca. Zoological Catalogue of Australia, 8.' (Ed. W. W. K. Houston.) (Australian Government Publishing Service: Canberra.) Smithers, C., McAlpine, D., Colman, P., and Gray, M. (1977). Island invertebrates. In “Lord Howe Island”. (Ed. N. Smith.) pp. 23-26. (Australian Museum: Sydney.) Solem, A. (1959). Systematics ofthe land and freshwater Mollusca ofthe New Hebrides. Fieldiana Zoology 30, 1—359. Solem, A. (1990). How many Hawaiian land snail species are left? And what can we do for them. Bishop Museum Occasional Papers 30, 27-40. Swofford, D. L. (2000). “PAUP”: Phylogenetic Analysis Using Parsimony”, version 4.0. (Sinauer Associates: Sunderland, MA.) Thompson, J. D., Higgins, D. G., and Gibson, T. J. (1994). CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position specific gap penalties and weight matrix choice. Nucleic Acids Research 22, 4673-4680. 174 Molluscan Research W. Ponder et al. Triggs, S., and Sherley, G. (1993). Allozyme genetic diversity in P/acostylus land snails and implications for conservation. New Zealand Journal of Zoology 20, 19-33. Vagvolgyi, J. (1975). Body size, aerial dispersal, and origin of the Pacific land snail fauna. Systematic Zoology 24, 465-488. Van der Lingen, G. 1. (1973). The Lord Hovve Rise rhyolites. In “Tnitial reports of the Deep Sea Drilling Project, 21”. (Eds R. E. Burns, J. E. Andrews, G. J. van der Lingen, M. Churkin, J. S. Galehouse, G. H. Packman, T. A. Davies, J. P. Kennet, A. R. Edwards, R. P. Von Herzen and I. G. Speeden.) pp. 523—539. (US Government Printing Office: Washington, DC.) Ward, S. A., and Thornton, I. W. B. (2000). Chance and determinism in the development of isolated communities. Global Ecology and Biogeography 9, 7—18. Wallace, A. R. (1974). [letter to Hedley cited in introduction] Proceedings of Linnean Society of NSW 98, 255. Yamazaki, N., Ueshima, R., Terrett, A., Yokobori, S., Kaifu, M., Segawa, R., Kobayashi, T., Numachi, K., Ueda, T., Nishikawa, K., Watanabe, K., and Thomas, R. H. (1997). Evolution of pulmonate gastropod mitochondrial genomes: comparisons of complete gene organizations of Euhadra, Cepaea and Albinaria and implications of unusual tRNA secondary structures Genetics 145, 749—758. Placostylus relationships based on mtDNA Molluscan Research 175 Appendix. The Lord Howe Island Placostylus taxa Placostylus bivaricosus (Gaskoin, 1855) is the only currently recognised species found on Lord Howe Island (Smith 1992). The last review of this group was by Iredale (1944), who noted that the: *variation in this group is worthy of study ... At the present time the various forms may be less visible in the field than in Roy Bell's time, when he indicated no less than six different colonies, separable in the field’. No critical examination of the taxonomy of the Lord Howe Island Placostylus has ever been undertaken, although Smith (1992) catalogued the taxa and details of original names and references can be obtained from that source. Material in the Australian Museum collections appears to support the recognition of at least three, possibly four (see below), Recent and one fossil taxon on the basis of shell morphology (Ponder and Chapman 1999). These can tentatively be treated as “subspecies” as follows. Placostylus bivaricosus bivaricosus (Gaskoin, 1855) Bulimus bivaricosus Gaskoin, 1855. Lord Howe Island (location of type material not known; fide Smith 1992: 108). Placostylus bivaricosus belli Iredale, 1944. Lord Howe Island (holotype Australian Museum C.38973; Fig. 5C). Remarks The typical subspecies (Fig. 54,3) is found on the northern end of the island and in the settlement area as far south as Intermediate Hill. Placostylus bivaricosus belli (Fig. 5C) is based on a small, narrow (shell length 54.6 mm, width 22.2 mm, aperture length 24.3 mm) specimen that can be matched with other occasional small specimens found within the settlement area. Placostylus bivaricosus etheridgei (Hedley, 1891) Bulimus bivaricosus etheridgei Hedley, 1891. Under the wall of Mt Lidgbird (Hedley, 1891). Iredale (1944) states that this species is “apparently from the saddle of Mt Lidgbird' (three syntypes, Australian Museum C.33308; Fig. 5H).- : Placostylus etheridgei ‘Brazier’, 1889. Nomen nudum (introduced as a name on the caption to a figure in Etheridge, 1889). Placostylus bivaricosus royi Iredale, 1944. Little Slope, base of eastern side of Mount Gower (holotype, Australian Museum C.38974; Fig. 51). Remarks This taxon is found in the southern mountains. It is variable in size, although often large, and has a thin shell with a weakly developed apertural varix and the surface of the shell is sculptured with oblique rugae. It is probably extinct, no living specimens having been collected for several decades. There is some suggestion from museum material of overlap, or perhaps merging, of typical bivaricosus and etheridgei in the area around Intermediate Hill. Unfortunately, the populations in this area are now either extremely reduced or, more likely, extinct. Both the available names for this taxon are from populations around the base of Mount Lidgbird and Mount Gower, or in the Erskine Valley, which lies between the two mountains. These have a significantly larger shell than samples from the top of Mount Gower (Fig. 5D,F; see below). 176 Molluscan Research W. Ponder et al. Fig.5. Shells of Placostylus from Lord Howe Island (all material from the Australian Museum, Sydney, NSW; dimensions are maximum length). Images not to same scale. 4,5, P. bivaricosus bivaricosus (Gaskoin), Lord Howe Island, C.114669, 56.6 mm (A) and C.114666, 60.7 mm (B). C, Holotype of P. bivaricosus belli Iredale, Lord Howe Island, C.38973, 54.6 mm. D,F, P. aff. bivaricosus etheridgei (Hedley), Mount Gower (C.114677), 54.0 mm (D) and 62.2 mm (F). E, P. bivaricosus cuniculinsulae (Cox), Rabbit (= Blackburn) Island, Lord Howe Island, C.118151, 45.6 mm. G, P. bivaricosus solidus (Etheridge), syntype, Lord Howe Island, C.171111, 74.8 mm. HI, P. bivaricosus etheridgei (Hedley), syntype of Bulinus etheridgei Hedley, Mt Lidgbird, Lord Howe Island, C.33308, 67.2 mm (H) and holotype of P. bivaricosus royi Iredale, Little Slope, Lord Howe Island, C.38974, 77.2 mm (1). Placostylus relationships based on mtDNA Molluscan Research 177 Placostylus bivaricosus cuniculinsulae (Cox, 1872) Bulimus (Placostylus) cuniculinsulae Cox, 1872. Rabbit Island (= Blackburn Island), Lord Howe Island (holotype, Natural History Museum, London, 1888.12.1 1.47-8). Remarks This small form was found only on Blackburn Island and is now extinct (Iredale 1944). Etheridge (1889) noted that there was a great deal of variation in shell morphology and that it seemed “... more in keeping with the facts to regard the Rabbit Island shell simply as a variety'. However, the material in the Australian Museum from Blackburn Island all possesses a distinctive, thin shell and is smaller than any adults seen from the main island (Figs 5E, 6; Table 4). Placostylus bivaricosus solidus (Brazier in Etheridge, 1889) Bulimus bivaricosus solida Brazier in Etheridge, 1889. (Seven syntypes, Australian Museum C.171111 (Fig. 5G), C.171110). Remarks This large, heavy form is found in calcarenite along the eastern side of the settlement area of the island. Smith (1992) indicates that the original introduction of the name is a nomen nudum. However, there is a statement differentiating the taxon in the original introduction of the name (p. 27), so it is valid. In addition, Etheridge (1891) described this taxon in greater detail and provided figures. There are also fossil specimens that appear to be indistinguishable from the Recent material (including one of the syntypes in C.171111), although specimens can usually be readily separated into the solida morph or the Recent morph. Unfortunately, there has not yet been any attempt to systematically collect fossil Placostylus and relate shell morphology to stratigraphy. A comparison of three basic shell measurements (Table 4) indicates that the Recent taxa discriminate well (Fig. 6), but mainly on size rather than shape differences. There also appears to be justification for the recognition of an additional unnamed and recently extinct taxon that lived on the top of Mount Gower (Australian Museum C.I 14677; Fig. 5D,F) and on a razor-back spur to the south of the summit (Australian Museum C.114679). These two lots have similar shell dimensions to typical P. bivaricosus (Fig. 6; Table 4), but have the Fig.6. Plotofthe first and second discriminant scores from the measurement data summarised in Table 4. The ellipses have a confidence level of P = 0.683. (V), Placostylus bivaricosus cuniculinsulae; (+), P. bivaricosus bivaricosus; (©), P. bivaricosus etheridgei; (O), P. bivaricosus aff. etheridgei (top of Mt Gower); (x), P. bivaricosus solidus. Score 2 178 Molluscan Research W. Ponder et al. Table 4. Shell measurements of Placostylus from Lord Howe Island Data show the range and mean + s.d. e—a Shell length (mm) P. bivaricosus bivaricosus (n = 40) P. bivaricosus etheridgei (n = 19) P. bivaricosus cuniculinsulae (n ^ 9) Summit of Mount Gower (n = 11) P. bivaricosus solidus (n = 15) 46.0-61.9, 53.4 + 3.5 65.5-77.1, 71.5 x 4.1 40.0-45.7, 43.3 + 1.8 51.9-65.1, 57.6 + 4.2 61.1—79.0, 71.4 + 6.2 Shell width (mm) 21.3-27.5, 23.8 + 1.5 28.0-35.3, 31.9 + 2.3 19.4-21.9, 20.3 + 0.8 23.7-27.8,25.4 + 1.2 26.8-36.6, 32.6 + 3.1 Aperture length (mm) 23.9-32.4, 26.9 € 2.1 31.8-41.1,35.9 + 2.9 20.3-22.9, 21.4 + 0.9 26.5-31.9,28.3 + 1.8 31.5-41.0, 37.2 € 2.7 Specimens measured: P 5. bivaricosus (C118724; C114653; C.78759; C.114656); P. b. solidus (C.171111; C.171110; C.114648; C.47513); P. b. etheridgei (C.33308; C.38973; C.114672; C.114675); P. b. cuniculinsulae (C.118151; C.114646; C.114647); Mount Gower (C.114679; C.114677). Aperture length is the external, not internal, length. colour and surface sculpture of P. bivaricosus etheridgei. Given their isolation on the summit, they were probably genetically distinct from the populations around the base of the southern mountains and the typical lowland form. http://www.publish.csiro.au/journals/mr CSIRO PUBLISHING ÇO T ROLE edu 7 — — a www.publish.csiro.au/journals/mr Molluscan Research, 2003, 23, 179-183 Short contribution Changes in tissue composition during larval development of the blacklip pearl oyster, Pinctada margaritifera (L.) Jan M. Strugnell^P€ and Paul C. Southgate^ Agchool of Marine Biology and Aquaculture, James Cook University, Townsville, Qld 4811, Australia. BMolecular Evolution, Department of Zoology, Oxford University, South Parks Rd, Oxford OX1 3PS, UK. CTo whom correspondence should be addressed. Email: jan.strugnell@merton.oxford.ac.uk Abstract This paper reports on the changes in proximate composition (i.e. protein, lipid and carbohydrate) of tropical blacklip pearl oyster Pinctada margaritifera (L., 1758), larvae throughout development. Protein was the largest component of dried larval tissues. Mean protein, lipid and carbohydrate contents all decreased from Day 1 to Day 4. Lipid loss between Day 1 and Day 4 contributed 56% of the total energy utilised during this period, whereas protein contributed almost 40%. Between Day 18 and Day 21, the accumulation of lipid contributed almost 7096 of the total energy gain per larva during this period, suggesting that lipid may be the primary energy reserve utilised during metamorphosis. Patterns of energy reserve composition, utilisation and accumulation within P. margaritifera larvae were comparable to those reported for temperate species. Introduction Fundamental differences exist between tropical and temperate marine environments that directly influence the organisms living within them. Tropical marine environments are characterised by surface-water temperatures of 20—30*C, relatively low nutrient loads and correspondingly low phytoplankton levels (Nybakken 1982). In contrast, temperate marine environments are characterised by seasonal, but relatively high, nutrient loads, high phytoplankton levels and water temperatures fluctuating between 10 and 20°C in a seasonal fashion (Nybakken 1982). On this basis, the rates of energy-metabolism in temperate bivalve molluscs cannot be assumed to pertain to tropical species. However, studies reporting on changes in proximate composition during larval development of bivalves have focused on temperate species (Holland and Spencer 1973; Bayne ef al. 1975; Gallager and Mann 1986; Gallager et al. 1986; Whyte et al. 1987, 1990, 1991). The hypothesis tested in this study was that patterns of energy-reserve utilisation and accumulation in larvae of the tropical blacklip pearl oyster, Pinctada margaritifera (L., 1758), differ from those reported for temperate species. This was examined by determining changes in proximate composition (i.e. protein, lipid and carbohydrate) during larval development. Materials and methods Spawning induction and larval rearing of P. margaritifera followed standard procedures (Southgate and Beer 1997). Ten hatchery-conditioned P. margaritifera broodstock were induced to spawn using thermal shock. Progeny from each of the females were randomly distributed across larval batches. Larvae were reared in six 500-L fibreglass tanks and water was changed every 48 h. Larvae were fed a diet consisting of the two prymnesiophytes, Pavlova salina and Isochrysis aff. galbana clone T-ISO, and the diatom Chaetoceros muelleri (Southgate and Beer 1997). A sample of approximately 15000 larvae was taken for © Malacological Society of Australasia 2003 10.1071/MR02018 1323-5818/03/020179 180 Molluscan Research J. M. Strugnell and P. C. Southgate proximate analysis at each water change from each tank and stored in liquid nitrogen to await analysis of protein, lipid and carbohydrate content (Mann and Gallager 1985; Baethgen and Alley 1989). Prior to analysis, larval samples were freeze-dried. The experiment ended after 21 days, when 5096 of larvae were “eyed”. Water temperature was measured at 0800 hours and 1800 hours each day and ranged from 25?C to 29?C with a mean of 27.2?C. Proximate compositional data were used to calculate total energy values using caloric equivalents of 36.42, 23.86 and 17.16 kJ g ! for lipid, protein and carbohydrate respectively (Brett and Groves 1979). Results and discussion Pinctada margaritifera larvae increased in mean (+ SEM, n = 30) shell length from 76 + l um on Day 1 (24 h after fertilisation) to 213 + 5 um on Day 21. Mean (+ SEM, n = 30) larval dry weight increased from 556.6 + 1.1 ng larva! on Day 1 to 2793.4 + 40.0 ng larva! on Day 21. Mean survival of larvae to Day 21 was 13.2 + 5.096. Changes in mean protein, lipid and carbohydrate content of P. margaritifera larvae are shown in Fig. 1. Protein was the largest component of dried larval tissues. Mean protein content almost halved from 133.7 + 5.1 ng larva! on Day 1 to 67.3 + 3.1 ng larva”! on Day 4 (+ SEM, n = 6). The most notable increase in mean protein content of larval tissues occurred between Day 12 (133.5 + 17.7 ng larva !) and Day 18 (395.2 + 12.9 ng larva !). Subsequently, the increase in protein content slowed considerably; the tissues of 21-day-old larvae possessed 417.7 + 16.78 ng larva ! of protein. Lipid was the second largest component of dried larval tissue during development. Mean lipid content decreased by greater than 75% from 73.3 + 9.1 ng larva! to 11.7 + 1.1 ng larva! between Day 1 and Day 4. Subsequently, larval lipid content increased to 96.8 + 11.6 ng larva! on Day 15, declined to 76.9 + 5.2 ng larva! on Day 18, then increased rapidly to 140.6 + 2.2 ng larva! on Day 21. Carbohydrate was the smallest component of dried larval tissues. Mean carbohydrate content decreased by about 80% between Day 1 (13.5 + 4.9 ng larva ') and Day 4 (2.7 + 0.8 ng larva !). Subsequently, carbohydrate content increased to 63.1 + 31.1 ng larva! on Day 21. Lipid loss between Day 1 and Day 4 contributed 56% of the total energy utilised during this period, whereas protein contributed almost 40%. The energy contributed by carbohydrate was very small, less than 10% of the amount contributed by lipid. Between Day 18 and Day 21, the accumulation of lipid contributed almost 7096 of the total energy gain per larva during this period. The energy available from lipid was more than four times that available from protein and carbohydrate, each of which contributed about 1596 of the total energy value at this time. Patterns of energy reserve composition, utilisation and accumulation within P. margaritifera larvae were found to be comparable to those reported for temperate species (Holland and Spencer 1973; Gallager et al. 1986; Whyte et al. 1987). As reported in all previous studies with bivalves (Holland 1978; Whyte et al. 1989), protein was the largest organic component of P. margaritifera larvae and showed the largest increase during larval development. Protein forms the bulk of the structural organic components of larval tissues (Holland 1978; Whyte et al. 1989). Also in accordance with previous studies, lipid was the second largest organic component of P. margaritifera larvae and carbohydrate content was relatively small and changed little during larval development. The marked decline in protein, lipid and carbohydrate content of P. margaritifera larvae between Days 1 and 4 is similar to that reported for the larvae of temperate scallops (Patinopecten yessoensis and Crassadoma gigantea), where protein, lipid and carbohydrate were reported to be catabolised simultaneously and linearly with time during embryonic Tissue composition of pearl oyster larvae Molluscan Research 181 500 ---- Carbohydrate — Lipid prT = /F q — - - Protein 400 Tissue composition (ng larva”1) Age (days from fertilisation) Fig. 1. Changes in mean (+ SEM) protein, lipid and carbohydrate content of P. margaritifera larvae (ng larva!) during development (n = 6, Days 1 to 9; n = 3, Days 12 to 21). development (Whyte ef al. 1990, 1991). This decline is thought to result from utilisation of endogenous reserves, provided to the egg by the female parent, to fuel the transition to an exogenous mode of feeding (Bayne et al. 1975; Mann and Gallager 1985). Utilisation of protein, lipid and carbohydrate by P. margaritifera between Days 1 and 4 shows that larval development during this period is an energetically expensive process; approximately 66% of the total energy content of 1-day-old larvae was utilised by Day 4. Similarly, 56.996 of the energy content of fertilised eggs was reported to be utilised during the first 72 h after fertilisation in scallop (Crassodoma gigantea) larvae (Whyte et al. 1990). Over half the energy utilised by P. margaritifera larvae between Days 1 and 4 was contributed by lipid, suggesting that lipid is the primary energy reserve during this period. Similarly, embryogenesis of the oyster (Crassostrea virginica) and clam (Mercenaria mercenaria) has been reported to be fuelled primarily (55-96% and 50-6576 respectively) by parentally derived lipid (Gallager and Mann 1986). A number of studies with temperate bivalves only investigated changes in the lipid content of tissues during larval development and disregarded changes in protein or carbohydrate contents (Gallager and Mann 1986; Gallager ef al. 1986; Napolitano et al. 1988; Delaunay et al. 1992, 1993). However, in P. margaritifera larvae, protein was found to contribute 40% of the total energy utilised between Days 1 and 4, indicating it to be an important secondary energy source during this period. Similarly, Whyte et al. (1990, 1991) reported protein to contribute a significant portion of the total energy expended during the embryogenesis of the scallops, Patinopecten yessoensis and Crassodoma gigantea (44.9% and 43.5% respectively). Such values were only slightly lower than the energy contribution reported for lipid of 47.6% and 4% respectively (Whyte et al. 1990, 1991). Although the overall energy contribution from carbohydrate in P. margaritifera tissues was small between Days 1 and 4, it accounted for an 80% depletion of its initial energy content, which is comparable to the decline in lipid (8670). Similar depletion of carbohydrate has been reported during embryogenesis of scallops (Whyte et al. 1990, 1991). In comparison, the decline in protein content of P. margaritifera larvae between Days 1 and 4 accounted for only 50% of the initial energy content. This reflects the important structural role of proteins in keeping the larval body intact (Holland 1978). 182 Molluscan Research J. M. Strugnell and P. C. Southgate Larval lipid and carbohydrate contents increased markedly between Days 18 and 21. During the same period, the rate of protein accumulation declined considerably. Although both lipid and carbohydrate contents increased by almost 50% between Days 18 and 21, accumulation of lipid contributed more than four times the total energy gain (per larva) than either protein or carbohydrate. This strongly suggests that lipid may be the primary energy reserve utilised during metamorphosis. Although this assumption is supported by the results of similar studies with temperate species (Holland and Spencer 1973; Whyte et al. 1987), further study is required to confirm this role for lipid in P. margaritifera. These results provide a strong basis for future study by clearly indicating changes in proximate composition and energy-reserve utilisation and accumulation during larval development of P. margaritifera larvae. This research has increased our understanding of the energetics of bivalve larvae, being the first study of its kind to investigate a tropical species. These findings have significant practical application and will be useful in the development of more efficient hatchery techniques for pearl oysters; for example, in formulating diets that best provide the biochemical requirements of larvae to maximise growth and survival and to minimise the time to metamorphosis. References Baethgen, W. E., and Alley, M. N. (1989). A manual colorimetric procedure for measuring ammonium nitrogen in soil and plant Kjeldahl digests. Communications in Soil Science and Plant Analysis 20, 961—969. Bayne, B. L., Gabbott, P. A., and Widdows, J. (1975). Some effects of stress in the adult on the eggs and larvae of Mytilus edulis L. Journal of the Marine Biological Association of the United Kingdom 55, 675—689. Brett, J. R., and Groves, T. D. (1979). Physiological energetics. In “Fish Physiology, Vol. III’. (Eds W. S. Hoar and D. J. Randall.) pp. 279—351. (Academic Press: New York, USA.) Delaunay, F., Marty, Y., Moal, J., and Samain, J. F. (1992). Growth and lipid class composition of Pecten maximus (L.) larvae grown under hatchery conditions. Journal of Experimental Marine Biology and Ecology 163, 209-219. Delaunay, F., Marty, Y., Moal, J., and Samain, J. F. (1993). The effects of monospecific algal diets on growth and fatty acid composition of Pecten maximus (L.) larvae. Journal of Experimental Marine Biology and Ecology 173, 163-179. ; Gallager, S. M., and Mann, R. (1986). Growth and survival of larvae of Mercenaria mercenaria (L.) and Crassostrea virginica (Gmelin) relative to broodstock conditioning and lipid content of eggs. Aquaculture 56, 105—121. Gallager, S. M., Mann R., and Sasaki, G. C. (1986). Lipid as an index of growth and viability in three species of bivalve larvae. Aquaculture 56, 81-103. Holland, D. L. (1978). Lipid reserves and energy metabolism in the larvae of benthic marine invertebrates. In “Biochemical and Biophysical Perspectives in Marine Biology”. (Eds P. C. Malins and J. R. Sargent.) pp. 85-123. (Academic Press: London, UK.) Holland, D. L., and Spencer, B. E. (1973). Biochemical changes in fed and starved oysters, Ostrea edulis L. during larval development, metamorphosis and early spat growth. Journal of the Marine Biological Association of the United Kingdom 53, 287—298. Mann, R., and Gallager, S. M. (1985). Physiological and biochemical energetics of larvae of Teredo navalis L., and Bankia gouldi (Bartsch) (Bivalvia: Teredinidae). Journal of Experimental Marine Biology and Ecology 85, 211-228. Napolitano, G. E., Ratnayake, W. M. N., and Ackman, R. G. (1988). Fatty acid components of larval Ostrea edulis (L.); importance of triacylglycerols as a fatty acid reserve. Comparative Biochemistry and Physiology 90B(4), 8975—883. Nybakken, J. W. (1982). “Marine Biology: An Ecological Approach.’ (Harper and Row: New York, USA.) Southgate, P. C., and Beer, A. C. (1997). Hatchery and early nursery culture of the blacklip pearl oyster (Pinctada margaritifera L.). Journal of Shellfish Research 16(2), 561—567. Tissue composition of pearl oyster larvae Molluscan Research 183 Whyte, J. N. C., Bourne, N., and Hodgson, C. A. (1987). Assessment of biochemical composition and energy reserves in larvae of the scallop Patinopecten yessoensis. Journal of Experimental Marine Biology and Ecology 113, 113—124. Whyte, J. N. C., Bourne, N., and Hodgson, C. A. (1989). Influence of algal diets on biochemical composition and energy reserves in Patinopecten yessoensis (Jay) larvae. Aquaculture 78, 333-347. Whyte, J. N. C., Bourne, N., and Ginther, N. G. (1990). Biochemical and energy changes during embryogenesis in the rock scallop Crassodoma gigantea. Marine Biology 106, 239—244. VVhyte, J. N. C., Bourne, N., and Ginther, N. G. (1991). Depletion of nutrient reserves during embryogenesis in the scallop Patinopecten yessoensis (Jay). Journal of Experimental Marine Biology and Ecology 149, 67-79. http://www.publish.csiro.au/journals/mr tai mo kündən i mafic A E y 15 asw LUN d 55.5 wrea) rail ABD Pipes n d (drieuom Ty l A belua PR p VE bi je 57 ae ^ 120] də vəh C dii gren dp “play iya pn * aint petra n ———.— 2... Rid Mite o Bun ə -—. k areari vene Thea maie — Droit 2.47. inii Qhapaqqa : porma (URSINI and xg .25.3..1.0 sona - “during "arvat develurerarm ui Emarri dario. This veron Ranke feted nd undersea pling vd the auerwenicn of edito kevan sn the (rtr abs of d rut Te aueia a Iopisnt apate Tise Arein kavei .. néeuem arısının [d ill bu neeful an the 227... Tuus cium fut exp, m təyimiləlinz chow 2 tem mente: - boeno — . — cmi RN Wt wx vii AL NT even — 5 200... s dd wñ 5—.— L : ' 25. kəsi t am Pie MM n — E bə eo R.A 20 TM — ül gi — wbul, se Datis, uf ferius silah 1. f: weed o ta, Mes — —. x. vəsə dı hot L XM "eue 1T pfi. mua dod duae b Me np vv Hun id D. f Ñami) pp rro ndo Pria Ons resi nə s E p 227-55-53... 0... itat 2541751515. bn Mey xir x biles in e - yi r T E İk b, sm À n 5— səsi, S babe as. : ui 2 Yeptuaus & h *** Available at bargain prices *** BACK ISSUES of MOLLUSCAN RESEARCH AND JOURNAL OF THE MALACOLOGICAL SOCIETY OF AUSTRALIA See http://www.booksofnature.com for details or write to: Books of Nature P.O. Box 345 Lindfield NSW 2070 Australia phone/fax: + 61 (0)2 9415 8098 email: books@booksofnature.com Back issues from Volume 22 are available from CSIRO PUBLISHING, PO Box 1139, Collingwood, Victoria 3066, Australia. Tel: +61 (0)3 9662 7666; fax: +61 (0)3 9662 7555; email: publishing.sales@csiro.au y singu The Malacological Society of Australasia Ltd The prime objective of The Malacological Society of Australasia (MSA) is to promote the study of molluscs. Our members include amateurs, students and scientists interested in the study, conservation and appreciation of molluscs. On joining the Society you will receive Molluscan Research and Australasian Shell News and may attend meetings of the Society wherever held by branches throughout Australasia. You will also be supporting and furthering the interests of malacology in the Australasian region. Branches of the Society meet regularly in Melbourne, Sydney and Brisbane. Many other societies with similar interests are affiliate members. The association of affiliate members with the Society gives them a more effective voice on issues of common concern. Membership rates Ordinary member (Australasia, SE Asia, Western Pacific) AU$70.00 Ordinary member (rest of the world) 100.00 Extra family members 5.00 Affiliated Society 100.00 Student members 45.00 Further information about The Malacological Society of Australasia can be found on-line at: http://www.amonline.net.au/malsoc. Application forms are also available on-line from the above web site or from the Membership Secretary (see below). Please send payment in Australian dollars (cheque, bank draft or credit card) to: The Membership Secretary, 45 Castlewood Drive, Castle Hill, NSW 2154, Australia. Telephone/Facsimile: +61 (0)2 9899 5719 Email: chrisann@swavley.com.au Molluscan Research Molluscan Research, previously known as the Journal of the Malacological Society of Australasia, is a publication for authoritative scientific papers dealing with the Phylum Mollusca and related topics. Three issues are published in each annual volume. General and theoretical papers relating to molluscs are welcome. Papers concerning specific geographical areas or new taxa should normally focus on the Indo-West Pacific region, as well as Australasia and the Southern Ocean. Enquiries regarding back issues prior to Volume 22 should be addressed to: Capricornica Publications, PO Box 345, Lindfield, NSW 2070, Australia. Phone/fax: +61 (0)2 9415 8098 Email: caprica@capricornica.com Editorial enquiries regarding Molluscan Research should be addressed to: Dr W. F Ponder, Australian Museum, 6 College St., Sydney, NSW 2010, Australia. Email: winstonp@austmus.gov.au Instructions to authors are available in the first issue of each volume and on the journal’s web site at: http://www.publish.csiro.au/ journals/mr/ Museum Victoria . Molluscan Research Contents Volume 23 Number 2 2003 Embryonic and larval development of Pinctada margaritifera (Linnaeus, 1758) M. S. Doroudi and P. C. Southgate 101 Ultrastructure of male germ cells in the testes of abalone, Haliotis ovina Gmelin S. Singhakaew, V. Seehabutr, M. Kruatrachue, P. Sretarugsa and S. Romratanapun 109 Gastropod phylogeny based on six segments from four genes representing coding or non-coding and mitochondrial or nuclear DNA D. J. Colgan, W. F. Ponder, E. Beacham and J. M. Macaranas 123 Reassessment of Australia's oldest freshwater snail, Viviparus (?) albascopularis Etheridge, 1902 (Mollusca : Gastropoda: Viviparidae), from the Lower Cretaceous (Aptian, Wallumbilla Formation) of White Cliffs, New South Wales B. P. Kear, R. J. Hamilton-Bruce, B. J. Smith and K. L. Gowlett-Holmes 149 Relationships of Placostylus from Lord Howe Island: an investigation using the mitochondrial cytochrome c oxidase 1 gene W. F. Ponder, D. J. Colgan, D. M. Gleeson and G. H. Sherley 159 Short contribution Changes in tissue composition during larval development of the blacklip pearl oyster, Pinctada margaritifera (L.) J. M. Strugnell and P. C. Southgate 179 www.publish.csiro.au/journals/mr (n) CSIRO Published by CSIRO PUBLISHING for the Malacological Society of Australasia ISSN 1323-5818 } ^ b Í fi 790 | 0 [