On South Georgia: Some Implications of Shell Size, Shell Shape, and Site Isolation in a Singular Sub-Antarctic Land Snail P.J.A

Antarctic Science 23(5), 442–448 (2011) & Antarctic Science Ltd 2011 doi:10.1017/S0954102011000289 Notodiscus (Charopidae) on South Georgia: some implications of shell size, shell shape, and site isolation in a singular sub-Antarctic land snail P.J.A. PUGH1 and R.I. LEWIS SMITH2 1Department of Life Sciences, Anglia Ruskin University, East Road, Cambridge CB1 1PT, UK 2Centre for Antarctic Plant Ecology and Diversity, Alton Road, Moffat DG10 9LB, UK [email protected]

Abstract: Multivariate analysis shows that shells of Notodiscus sp. (Charopidae: Pulmonata) reported from South Georgia are smaller and proportionately taller than, but otherwise similar to, populations of Notodiscus hookeri (Reeve) from Iles Crozet and Iles Kerguelen. The origin of this solitary, and spatially limited, South Georgia population is enigmatic. It is conﬁned to a remarkably small coastal lowland site which was glaciated at Last Glacial Maximum, precluding a Tertiary relict origin, and on the leeward north-east coast, ruling out post- glacial ocean rafting. The site is close to the King Edward Point settlement, yet the absence of any logistics connections with the Iles Crozet or Iles Kerguelen mitigates against anthropogenic introduction. The close proximity of the population to nests of blue-eyed shag (Phalacrocorax atriceps), Dominican gull (Larus dominicanus) and light-mantled sooty albatross (Phoebetria palpebrata) could imply the snail was originally introduced to South Georgia via these ocean transiting seabirds.

Received 15 September 2010, accepted 14 February 2011, ﬁrst published online 15 April 2011

Key words: colonization, Gastropoda, Notodiscus, Southern Ocean islands, zoochory

Introduction Amsterdam, Marion and possibly Prince Edward Island (Prince Edward Islands) (Fig. 1), and containing a single The islands of the Southern Ocean (Fig. 1) harbour only variable species, N. hookeri Reeve, 1854 (Solem 1968, 41 species of endemic and introduced terrestrial slugs and Smith 1992, Madec & Bellido 2007). The primary focus of snails (Gastropoda: Mollusca) (Pugh & Scott 2002). Twenty- this study is a morphometric comparison of the snails on seven species occur on the South Pacific Province (SPP) South Georgia with those of N. hookeri collected on the Iles Antipodes, Bounty, Campbell, Chatham, Macquarie and Crozet and Iles Kerguelen by Madec & Bellido (2007). Snares islands, while the South Indian Province (SIP) Amsterdam, Crozet, Heard, Kerguelen and Prince Edward islands are colonized by two exotics and a single endemic Methods species - Notodiscus hookeri (Reeve) (Charopidae) (Fig. 1). Field observations The impoverished South Atlantic Province (SAP) fauna is confined to three slugs and snails introduced to the Falkland The snail occurs c. 1 km north of Hope Point, King Edward Islands, and a single snail on South Georgia. Cove, Cumberland East Bay, South Georgia (54816'S, This South Georgian snail, discovered by RILS in January 36823'W) (Fig. 2). The site, c. 25 m above sea level and 1970, was first reported as an unidentified terrestrial mollusc 15–20 m inland from a narrow pebble beach, was the lower (Smith & Walton 1975), and subsequently tentatively ascribed 5 m of a high, vertical shaded cliff with a south-east aspect. to the genus Stephanoda (Endodontidae) or Notodiscus Ledges and crevices on this cliff face were colonized (Charopidae). Stephanoda (sensu Block 1984, Headland by liverworts (including Pachyglossa dissitifolia), lichens 1984), was based on a 1978 personal communication from (Cladonia spp., Rhizocarpon geographicum, Tephromela F.C. Naggs (British Museum, London) and Notodiscus (sensu atra), mosses (Andreaea spp., Bartramia patens, Grimmia, Pugh & Scott 2002), on further personal communications Pohlia nutans, Racomitrium spp.), and small vascular plants from F.C. Naggs (British Museum, London 1981), M. Vogel (Cystopteris fragilis, Deschampsia antarctica, Festuca (Philipps-Universita¨t, Hamburg 1985), the late A. Solem contracta, Phleum alpinum). Moist areas of rock were (Field Museum of Natural History, Chicago 1986), and coated by a film of microalgae and cyanobacteria, and P. Mordan (Natural History Museum, London 1999). We the snails appeared to be feeding on these organisms. The subsequently confirmed Notodiscus (Pugh et al. unpublished), soil and scree at the base of the cliff were covered by a genus previously known only from the SIP sub-Antarctic Acaena magellanica and A. tenera, scattered Festuca and islands of Kerguelen, Possession (Iles Crozet), Heard, mosses (especially Syntrichia robusta). Many empty shells

442

Downloaded from https://www.cambridge.org/core. Open University Library, on 22 Jan 2020 at 00:05:42, subject to the Cambridge Core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/S0954102011000289 NOTODISCUS ON SOUTH GEORGIA 443

Fig. 1. Location of South Georgia (SG) relative to the South Paciﬁc and South Indian Ocean islands of Amsterdam (Am), Crozet (Cr), Heard (Hd), Kerguelen (Kr), Prince Edward (PE) and St. Paul (SP). were observed at the base of the cliff face while live snails were conﬁned to c.30m2 of the cliff and particularly abundant at 1–3 m above the cliff base on mosses, in

Fig. 3. Notodiscus sp. a. Dorsal. b. Lateral. c. Ventral. Scale bars: 1 mm.

crevices and pockets of soil on ledges. Associated invertebrates included Enchytraeidae (Oligochaeta), mites (Acarina) and beetles (Coleoptera: Hydromedion and Perimylops spp.). Similar habitat, ﬂora and fauna occur Fig. 2. Offset: South Georgia showing Cumberland East Bay throughout much of the north-east coast of South Georgia (arrow). Main map: Cumberland East Bay showing collection yet many years of detailed ecological surveys and searches site (K) and the settlements of Grytviken whaling station (G) speciﬁcally for the snail have so far failed to locate and King Edward Point (K). evidence of other populations.

Table I. Simple comparison of shells from Iles Kerguelen, Iles Crozet wet conditions. Specimens kept for several weeks in a Perspex (after Madec & Bellido 2007) and South Georgia (current study). container at laboratory temperature (c. 15–188C) at South Measurements (after Madec & Bellido 2007) Georgia appeared to be more active at night. RILS fixed a Island AB CD BG BE EG EF total of 40 live specimens in 70% laboratory ethanol for Kerguelen mean 4.32 1.62 2.47 0.10 2.35 0.39 anatomical study, although some subsequently dried out. ± SD 0.37 0.15 0.19 0.04 0.18 0.05 Crozet mean 4.29 1.56 2.59 0.18 2.41 0.47 Specimen preparation ± SD 0.63 0.30 0.38 0.06 0.35 0.12 We cleaned several of the dried snails collected in March 1981 South Georgia mean 3.11 1.30 2.00 0.22 1.78 0.36 and prepared them for scanning electron microscopy. Shells ±SD 0.24 0.17 0.21 0.07 0.19 0.09 were soaked in absolute ethanol for 48 hours and cleaned Note: notable South Georgia values are italic (low)orbold (high). of adherent debris in low viscosity ethanol, to minimize mechanical risk for fragile shells, in an ultrasound bath at Little can be gleaned from the published literature about 47 kHz delivered in 15 second ‘bursts’ to prevent alcohol the ecology of Nothodiscus hookeri throughout its sub- ignition. Loose debris was removed with a trimmed no. 3 Antarctic distribution. It is widespread from near sea level artist’s brush. Cleaned individual shells were mounted on to . 700 m altitude in the sub-Antarctic islands of the conventional aluminium SEM stubs via double-sided adhesive Indian Ocean. Madec & Bellido’s (2007) samples from Iles carbon disks (a clean background for imaging) and a small Crozet were collected between 10 and 70 m, while those spot of colloidal silver (for optimum grounding), then gold- from Iles Kerguelen were from c. 190–700 m altitude. From coated in a Biorad SC502 sputter-coating unit. We imaged Dall’s (1876) and Dell’s (1964) accounts the Kerguelen the shells in a Leica-Cambridge S360 scanning electron populations appear to be associated with stony habitats and microscope configured for maximum depth of field at moss (i.e. fellfield) and also with the megaherb Pringlea a 10 kV acceleration voltage, 120 pA probe current and a antiscorbutica in wetter sheltered habitats. On Marion 20 mm working distance, capturing secondary electron Island N. hookeri typically frequents fellfield habitats at images via an electronic frame store (Fig. 3). lower altitudes (RILS, personal observation). However, no accounts refer to the snails occurring on cliff faces, as on Morphometric analysis South Georgia, although such a habitat could be construed as ‘‘vertical fellfield’’. We briefly dried the 36 ethanol fixed specimens, collected Subsequent summer (January 1981 and February 2000) in 2000, on laboratory tissue, mounted them aperture observations by RILS suggested the snails aggregated uppermost on a small bead of ‘Blu-Tack’ pressure sensitive (5–20 per dm2 over several square metres) in crevices and adhesive and measured via a dissection microscope fitted beneath overhangs during sunny, dry weather but dispersed with calibrated eyepiece graticule at x30. Our analysis focused onto open rock and compact moss cushions during overcast or on six principal measurements of maximal diameter (AB),

Table II. Correlation matrix. Similarity matrix ln AB ln CD ln BG ln BE ln EG ln EF ln AB 1 ln CD 0.846 1 ln BG 0.912 0.874 1 ln BE -0.436 -0.168 -0.200 1 ln EG 0.940 0.848 0.976 -0.403 1 ln EF 0.554 0.705 0.665 0.169 0.590 1

Axis 1 Axis 2 Axis 3 Axis 4 Axis 5 Axis 6 Eigenvalues 4.271 1.209 0.294 0.155 0.070 0.001 Percentage 71.187 20.142 4.897 2.591 1.172 0.011 Cumulative % 71.187 91.329 96.226 98.817 99.989 100 PCA variable loadings Axis 1 Axis 2 Axis 3 Axis 4 Axis 5 Axis 6 ln AB 0.461 -0.164 -0.148 -0.038 -0.858 0.039 ln CD 0.449 0.127 -0.047 0.863 0.188 0.009 ln BG 0.469 0.059 -0.340 -0.325 0.283 -0.689 ln BE -0.153 0.827 -0.492 -0.038 -0.147 0.166 ln EG 0.471 -0.121 -0.194 -0.323 0.356 0.704 ln EF 0.347 0.505 0.762 -0.206 -0.033 -0.013 Signiﬁcant variable loadings (, ± 0.1) are shown in bold.

Euclidean distances of the first three correlation matrix axes (after Madec & Bellido (2007). Using the Cadima & Jolliffe (1996) protocol to identify any ‘shape’-based components, axis 1 of the centred PCA covariance ‘X-matrix’ (Table III) extracted 83.4 of variance and (except spire elevation) included only positive variable loadings. This left 12.8% of variance on axes 2 and 3 from which we derived a scatter-plot upon which we superimposed Euclidean measurement vectors (Fig. 5). We were naturally wary of applying PCA to population means but ruled out any major axial interactions in terms of (clearly absent) ‘horseshoe-effect’ or ‘arch-effect’ artefacts (Palmer 2008). We verified this via axis 2 vs axis 1 variable Fig. 4. Scatterplot of PCA case scores for a standardized loading least squares linear regressions noting r2 values data/correlation matrix. Data for both Iles Crozet of , .01 for both correlation (size) and covariance (shape) (stippled triangles/dashed line) and Iles Kerguelen (white triangles/solid line) are based on populations (after Madec & matrices. Bellido, 2007), while those for South Georgia (black triangles/dotted line) denote individual shells. Euclidean Results & discussion (loge-) vectors are listed in Table I; vector scaling: 0.4. Both ‘size’ (Fig. 4) and especially ‘shape’ (Fig. 6) PCA scatter-plots show that the Iles Crozet and Iles Kerguelen spire diameter (CD), maximum height (BG), spire elevation snail populations overlap with the South Georgia snails (BE), penultimate whorl height (EF) and maximal aperture confirming that the latter represent a Notodiscus hookeri height (FG) (Table I), as used by Madec & Bellido (2007). population. The South Georgian shells consistently differ We combined measurements of the 36 South Georgia snails from those of both SIP populations in that they are smaller with the summary data for the four Iles Crozet and 13 Iles in all dimensions except spire elevation (BE) which is Kerguelen populations presented in Madec & Bellido (2007). invariably high (Tables I & II). ‘Size’ PCA (Fig. 4) confirms The data were analysed using the Multivariate Statistical that three of the four Crozet and all Kerguelen populations Package (MVSP 3.1 - Kovach 1999), loge- transformed to are ‘large’, i.e. positive with respect to axis 1, along which allow comparison of ‘size’ and ‘shape’ attributes of the Iles all shell dimensions except BE are aligned. The clustergram Crozet, Iles Kerguelen and South Georgia snails via cluster (Fig. 5) shows an entirely coherent Kerguelen/Crozet cluster analysis and principal components analysis (PCA). (EDij , 0.040) and a largely coherent South Georgia cluster Exploring possible ‘size’ attributes via a centred PCA (EDij , 0.034) containing single Kerguelen and Crozet correlation ‘X-matrix’ (Table II)inwhichaxes1and2extract populations. Analysis of ‘shape’ shows all vectors, except 91.3% of total variance, left a residual 8.7% spread across the BE, are positive confirming that axis 1 is indeed a size four remaining axes. We superimposed Euclidean vectors, vector. But whereas the Crozet and Kerguelen populations as measure of factor direction and magnitude, on the axis are clearly discrete, the South Georgian snails are scattered 2/axis 1 scatter-plot (Fig. 4). To further differentiate between across both axes. We cannot however rule out the possibility the Crozet, Kerguelen and South Georgia populations we that this is an artefact of combining data from populations and compiled a clustergram (Fig. 5) computed from a matrix of individual shells within the same analysis.

Fig. 5. Clustergram performed using Euclidean distance (Edij) computed from PCA correlation matrix axes 1, 2 and 3 (96.2 of total variance). Numbers of snails/populations are shown in parentheses.

Table III. Covariance matrix. ln AB ln CD ln BG ln BE ln EG ln EF ln AB 0.019 ln CD 0.011 0.009 ln BG 0.013 0.009 0.010 ln BE -0.004 -0.001 -0.001 0.005 ln EG 0.015 0.009 0.011 -0.003 0.013 ln EF 0.005 0.004 0.004 0.001 0.004 0.004

Axis 1 Axis 2 Axis 3 Axis 4 Axis 5 Axis 6 Eigenvalues 0.051 0.006 0.002 0.001 0.001 0.000 Percentage 83.37 10.026 2.766 2.053 1.772 0.013 Cumulative % 83.37 93.396 96.162 98.215 99.987 100 PCA variable loadings Axis 1 Axis 2 Axis 3 Axis 4 Axis 5 Axis 6 ln AB 0.600 -0.240 -0.034 -0.240 -0.723 0.030 ln CD 0.392 0.285 -0.621 -0.457 0.412 0.010 ln BG 0.441 0.168 0.458 0.061 0.239 -0.712 ln BE -0.107 0.770 0.389 -0.349 -0.237 0.257 ln EG 0.495 -0.097 0.345 0.246 0.373 0.653 ln EF 0.186 0.480 -0.364 0.740 -0.235 -0.022 Signiﬁcant variable loadings (, ± 0.1) are shown in bold.

Identification from raising the South Georgian snail to a new specific or even sub-specific taxon. We previously identified the South Georgian snail as a new The following features discriminate the South Georgia Notodiscus species (Pugh et al. unpublished), but hesitate to population from those described from the SIP islands. confirm this here because while raw measurements suggest Adult shell small at 2.5–3.5 mm diameter; 31 to 4 whorls; that the South Georgia snails are smaller than those from 2 umbilicus 0.48–0.70 mm and narrow at c. one sixth of the SIP populations, PCA fails to discriminate the South overall shell diameter (D/U 4.8–6.5). Shape is sub-globose Georgian and SIP snails in terms of shape attributes. with rounded periphery and deeply impressed sutures; Notodiscus hookeri exhibits considerable size variation, aperture wide, ovate to sub-circular, with no barriers and indeed that of the larger ‘subspecies’, N. hookeri margin almost entire. Apical whorls often eroded and heardensis (Dell 1964), was subsequently shown to be the remaining whorls with a fairly heavy periostracum. within the variation of the nominate species and so reduced Sculpture variable, comprising fine radial lines with traces to a synonym (Solem 1968). This factor alone dissuades us of micro-reticulation, to smooth and shining with irregular growth wrinkles. Colour greenish-yellow with a suggestion of irregular darker streaks following the growth lines. The South Georgian snails are thus smaller, more tightly coiled, with (relative to shell diameter) a higher spire and wider umbilicus that their SIP siblings.

Origin Notodiscus hookeri is widespread throughout the SIP islands (Solem 1968, Madec & Bellido 2007), yet the South Georgia population is very limited and probably derived. How then did it become established at its current location? The broad South Indian Ocean distribution of N. hookeri has been suggested as being a remnant of a once widespread (Gondwanan) distribution (Powell 1960). However, both the Fig. 6. Scatterplot of PCA case scores for a non-standardized SIP islands and South Georgia are derived from oceanic crust data/covariance matrix. Data for both Iles Crozet (stippled triangles/dashed line) and Iles Kerguelen (white triangles/ or volcanism (Dalziel et al. 1975, Nur & Ben-Avraham 1982, solid line) are based on populations (after Madec & Bellido, Tanner 1982, Giret 1987), which would preclude a vicariant 2007), while those for South Georgia (black triangles/dotted origin. South Georgia is c. 120 million years old with c.60% line) denote individual shells. Euclidean (loge-) vectors are of current terrain covered by permanent ice. During the listed in Table II; vector scaling: 0.02. Pleistocene (1.8 million to 12 000 years ago) this ice cover

extended to over 100 km offshore, inundating all fjords Oxychilus and Vitrea (Zonitidae) (Holdgate 1965, Barker and bays, including Cumberland East Bay (Dalziel 1984, 1979, Smith 1992, Cooper & Ryan 1994, Preece 1995, Clapperton et al. 1989, Tingey 1991). The montane ice-free 2001, Martins 1999). This, albeit indirect, evidence further refugia, which persisted throughout the Pleistocene on mitigates against N. hookeri on South Georgia being an mainland Antarctica, South Shetland and South Orkney anthropogenic exotic, i.e. human-mediated, introduction. islands (Convey et al. 2008, 2009), were probably absent Post-glacial avian dispersal is postulated for some on South Georgia (Clapperton et al. 1989), and there were Mollusca, including Helicarionidae, Succineidae and certainly no coastal refugia that could have supported land Vitrinidae on isolated islands (Rees 1965, Vagvolgyi snails. Extensive glaciation and the absence of co-occurring 1975) and, indeed, for N. hookeri on Marion Island (Falla Tertiary species imply that the South Georgia N. hookeri 1960, Dell 1964, Van Zinderen Bakker 1971). Gittenberger population, like all other South Georgian invertebrates, et al. (2006), using molecular phylogenetics, ascribe the represents post-glacial (late Pleistocene or Holocene) endemic Clausiliidae on the Trista da Cunha–Gough Island ‘waif’ colonists (Pugh & Convey 2000). group in the mid-South Atlantic to Balea, a genus The distribution of Notodiscus records require minimum otherwise known only from Palaearctic islands in the oceanic transits to South Georgia of between c.4400km North Atlantic Ocean. They suggested passive dispersal, (from Prince Edward Islands) and c. 6000 km (Heard Island) probably involving an avian vector. RILS found the South via rafting on ocean currents, anthropogenic introduction or Georgia snails close to a number of nesting seabirds (blue- seabirds. Airborne drift or ballooning is an improbable vector eyed shags (Phalacrocorax atriceps (King)), Dominican gulls for minute flightless arthropods (Pugh 2003), let alone a (Larus dominicanus Lichtenstein), and light mantled sooty relatively large snail. Although rafting has been postulated for albatrosses (Phoebetria palpebrata (Forster))), which also some South Pacific Charopidae (e.g. Smith & Djajasasmita breed on other SIP and SPP islands (e.g. Harrison 1983). 1988, Preece 1995), all native Southern Ocean land snail However, only P. palebrata is known to circumnavigate the records relate to vicariant distributions (Pugh & Scott 2002) in Southern Ocean and possibly make landfall at any of these which habitat, and a particular altitude, are incongruent with island groups. While we are unable to prove a three way ocean rafting as a viable vector. Furthermore, the Antarctic location/snail/seabird connection, avian zoochory remains the Circum-Polar Current, or West Wind Drift, would carry most parsimonious, indeed only, solution to account for this ocean-borne colonists to the west coast of South Georgia not highly disjunct distribution. to east coast Cumberland Bay. This leaves introduction by anthropogenic and avian vectors as the only alternative modes of introduction. Conclusions The distributions of flightless exotic insects (two We have confirmed our original suggestion that the South Carabidae and one Perimylopidae - Coleoptera) and mites Georgia land snail is a member of the genus Notodiscus which on South Georgia are confined by glaciers (Ernsting et al. was previously known only from the South Indian Ocean 1995, 1999, Pugh 1997, Brandjes et al. 1999). These islands with Notodiscus hookeri heardensis (sensu Dell 1964) arthropods, and the high proportion (11 of 56) of exotics now reduced to a synonym (representing a locally endemic among the Southern Ocean terrestrial molluscs (Pugh & ‘giant’) of the nominal N. hookeri. We believe that the South Scott 2002), could not rule out this single Notodiscus site Georgian snail may simply be a local ‘tall dwarf’ variant of record being regarded as an anthropogenic introduction. the nominal species. Only a comprehensive molecular The collection site is 3 km from the abandoned Grytviken analysis can reveal the true affinities of the South Georgia whaling station which operated from 1904 to 1965, and snail and its relationships with the other Southern Ocean 1.5 km from the small South Georgia Government settlement Notodiscus populations. at King Edward Point, which has been occupied since 1912 (Headland 1984). However, exotic invertebrates on isolated islands are Acknowledgements usually devolved from embarkation ports (Barker 1979, Smith 1992, Smith & Stanisic 1998) which, in this case, We would like to thank F.C. Naggs and P. Mordan, Natural would be in northern Europe or South America. There are History Museum, London; M. Vogel, Philipps-Universita¨t, no records of sealing, or indeed other, vessels having made Hamburg Marburg and (the late) A. Solem, Field Museum transits between South Georgia and the SIP islands (Bonner of Natural History, Chicago for providing comments on the 1968, Headland 1984). Furthermore, anthropogenic import possible identity and affinities of the South Georgian snail; of terrestrial molluscs to tropical, temperate and sub-polar R.K. Headland, formerly of the British Antarctic Survey, islands is usually attributed to a limited range of ‘exotic’ or Cambridge, for collecting additional specimens; B.J. Smith, ‘invasive’ species, particularly the temperate genera Arion formerly of the Queen Victoria Museum, Launceston, (Arionidae), Helix, Otala, Theba (Helicidae), Deroceras Tasmania, for assistance in specimen identification; and (Limacidae), Milax (Milacidae) Testacella (Testacellidae), two anonymous referees for their invaluable comments.

References HOLDGATE, M.W. 1965. The fauna of the Tristan da Cunha Islands. Part III. Philosophical Transactions of the Royal Society of London, B249, BARKER, G.M. 1979. The introduced slugs of New Zealand. New Zealand 361–402. Journal of Zoology, 6, 411–437. KOVACH, W.L. 1999. MVSP - A Multi-Variate Statistical Package for BLOCK, W. 1984. Terrestrial microbiology, invertebrates and ecosystems. Windows, version 3.1. Pentraeth: Kovach Computing Services. In LAWS, R.M., ed. Antarctic ecology. Vol. 1. London: Academic Press, MADEC,L.&BELLIDO, A. 2007. Spatial variation of shell morphometrics in 163–236. the subantarctic land snail Notodiscus hookeri from Crozet and BONNER, W.N. 1968. The fur seal of South Georgia. British Antarctic Kerguelen islands. Polar Biology, 30, 1571–1578. Survey Scientific Reports, No. 56, 81 pp. MARTINS, A.M.F. 1999. Evolution and distribution of the terrestrial BRANDJES, G.J., BLOCK,W.&ERNSTING, G. 1999. Spatial dynamics of two molluscs of the Azores. Bulletin of the Malacological Society of introduced species of carabid beetles on the sub-Antarctic island of London, 33, 5–6. South Georgia. Polar Biology, 21, 326–334. NUR,A.&BEN-AVRAHAM, Z. 1982. Oceanic plateaus, the fragments of CADIMA, J.F.C.L. & JOLLIFFE, I.T. 1996. Size- and shape-related principal continents, and mountain building. Journal of Geophysical Research, component analysis. Biometrics, 52, 710–716. 87, 3644–3661. CLAPPERTON, C.M., SUGDEN, D.E., BIRNIE,J.&WILSON, M.J. 1989. PALMER, M. 2008. Ordination webpage. http://ordination.okstate.edu Late-glacial and Holocene glacier fluctuations and environmental POWELL, A.W.B. 1960. Antarctic and subantarctic Mollusca. Records of the change on South Georgia, Southern Ocean. Quaternary Research, 31, Auckland Institute and Museum, 5, 117–193. 210–228. PREECE, R.C. 1995. Systematic review of the land snails of the Pitcairn CONVEY, P., GIBSON, J.A.E., HILLENBRAND, C.-D., HODGSON, D.A., Islands. Biological Journal of the Linnean Society of London, 56, PUGH, P.J.A., SMELLIE, J.L. & STEVENS, M.I. 2008. Antarctic terrestrial 273–307. life - challenging the history of the frozen continent? Biological PREECE, R.C. 2001. Introduced land molluscs on the islands of the Tristan Reviews, 83, 103–117. da Cunha–Gough group (South Atlantic). Journal of Conchology, 37, CONVEY,P.,STEVENS, M.I., HODGSON, D.A., SMELLIE, J.L., HILLENBRAND,C.-D., 253–259. BARNES, D.K.A., CLARKE,A.,PUGH, P.J.A., LINSE,K.&CARY, S.C. 2009. PUGH, P.J.A. 1997. Acarine colonisation of Antarctica and the islands of Exploring biological constraints on the glacial history of Antarctica. the Southern Ocean: the role of zoohoria. Polar Record, 33, 113–122. Quaternary Science Reviews, 28, 3035–3048. PUGH, P.J.A. 2003. Have mites (Acarina: Arachnida) colonised Antarctica COOPER,J.&RYAN, P.G. 1994. Management plan for the Gough Island and the islands of the Southern Ocean via air currents? Polar Record, Wildlife Reserve. Edinburgh, Tristan da Cunha: Government of Tristan 39, 239–244. da Cunha, 96 pp. PUGH, P.J.A. & CONVEY, P. 2000. Scotia Arc Acari: antiquity and origin. DALL, W.H. 1876. Mollusks. Contributions to the Natural History of Zoological Journal of the Linnean Society of London, 130, 207–217. Kerguelen Island made in connection with the United States Transit- PUGH, P.J.A. & SCOTT, B. 2002. Biodiversity and biogeography of non- of-Venus Expedition, 1874-75. Bulletin of the United States National marine Mollusca on the islands of the Southern Ocean. Journal of Museum, 3, 1–122. Natural History, 36, 927–952. DALZIEL, I.W.D. 1984. Tectonic evolution of a forearc terrain, Southern REES, W.J. 1965. The aerial dispersal of Mollusca. Proceedings of the Scotia Ridge, Antarctica. Geological Society of America Special Paper, Malacological Society of London, 36, 269–282. 200, 1–32. SMITH, B.J. & DJAJASASMITA, M. 1988. The land molluscs of the Krakatau DALZIEL, I.W.D., DOTT, R.H., WINN, R.D. & BRUHN, R.L. 1975. Tectonic Islands, Indonesia. Philosophical Transactions of the Royal Society of relations of South Georgia Island to the southernmost Andes. Bulletin of London, B323, 379–400. the Geological Society of America, 86, 1034–1040. SMITH, B.J. & STANISIC, J. 1998. Pulmonata. In BEESLEY, P.L., ROSS, G.J.B. DELL, R.K. 1964. Land snails from Subantarctic islands. Transactions of &WELLS, A., eds. Mollusca: the southern synthesis, fauna of Australia, the Royal Society of New Zealand, 4, 167–173. Vol. 5. Melbourne: CSIRO Publishing, 1037–1125. ERNSTING, G., BLOCK, W., MACALISTER, H.E. & TODD, C.M. 1995. SMITH, R.I.L. & WALTON, D.W.H. 1975. South Georgia, subantarctic. The invasion of the carnivorous carabid beetle Trechisibus antarcticus Ecological Bulletins, 20, 399–423. on South Georgia (sub-Antarctic) and its effect on the endemic SMITH, V.R. 1992. Terrestrial slug recorded from sub-Antarctic Marion herbivorous beetle Hydromedion spasutum (sic). Oecologia, 103, Island. Journal of Molluscan Studies, 58, 80–81. 24–42. SOLEM, A. 1968. The subantarctic land snail, Notodiscus hookeri (Reeve, ERNSTING, G., BRANDJES, G.J., BLOCK,W.&ISAAKS, J.A. 1999. Life history 1854) (Pulmonata, Endodontidae). Proceedings of the Malacological consequences for a subantarctic beetle: evaluating the contribution of Society of London, 38, 251–265. direct and indirect effects. Journal of Animal Ecology, 68, 741–752. TANNER, P.G.W. 1982. Geological evolution of South Georgia. In FALLA, R.A. 1960. Ocean birds as dispersal agents. Proceedings of the CRADDOCK, C., ed. Antarctic geoscience. Madison, WI: University of Royal Society, B152, 655–659. Wisconsin Press, 167–176. GIRET, A. 1987. Geologie des Terres Australes Francaises. Comite´ TINGEY, R.J. 1991. The geology of Antarctica. Oxford: Clarendon Press, National Française des Recherches Antarctiques, 58, 17–41. 680 pp. GITTENBERGER, E., GROENBERG, D.S.J., KOKSHOORN,B.&PREECE, R.C. 2006. VAGVOLGYI, J. 1975. Body size, aerial dispersal, and the origin of the Molecular trails from hitch-hiking snails. Nature, 439, 409. Pacific land snail fauna. Systematic Zoology, 24, 465–488. HARRISON, P. 1983. Seabirds, an identification guide. Boston, MA: VAN ZINDEREN BAKKER, E.M. 1971. Introduction. In VAN ZINDEREN BAKKER Houghton Mifflin, 448 pp. SR, E.M., WINTERBOTTOM, J.M. & DYER, R.A., eds. Marion and Prince HEADLAND, R. 1984. The island of South Georgia. Cambridge: Cambridge Edward Islands. Report on the South African Biological & Geological University Press, 293 pp. Expedition 1965–1966. Cape Town: A.A. Balkema, 1–15.