THERMOREGULATIO N IN CHELONIANS 259

of some poikilothermal animals. lzv. Nauchn. Ins/. Im. Ward, F. P., Hohmann, C. J., Ulrich, J. F. and Hill, S. E. Lesgafl a 17-1 8, 313-372 (in Russian). ( 1976). Seasonal microhabitat selections of spotted Sturbaum, B. A. (1982). Temperature regulation in turtles. turtles (Chemmrs gu//a/a) in Maryland elucidated by Compara1ive Biochemis1n• and Physiology 72A, 6 l 5-620. radioisotope tracking. Herpe10/ogica 32, 60-64. Swingland, I. R. and Frazier, J. G. (1980). The conflict Williams, T. L. ( 1981 ). Experimental analysis of the gait and between feeding and overheating in the Alda bran giant frequency of locomotion in the tortoise. with a simple tortoise. In Amlaner, C. J. and MacDonald, D. W. mathematical description. .Jouma/ of' Phrsio/og1· 310, (Eds) A handbook of bio1e/eme/1T and radio Tracking, 307-320. 61 1-615. Pergamon Press, Oxford and New York. Woodbury, A. M. and Hardy, R. ( 1948). Studies of the desert Swingland, I. R. and Stubbs, D. (1985). The ecology of a tortoise. Gopherus agassi::ii. Ecological Monographs 1 8 , Mediterranean tortoise (Tes/udo hermanni): Repro­ 145-200. duction . .Journal of Zoolog1·. London 205, 595-610. Wright, J., Steer, E. and Hailey, A. (in press). Habitat Whitlow, G. C. and Balazs, G. H. (1982). Basking behavior separation in tortoises. and the consequences for of the Hawaiian green turtle, Che/onia m1•das. Pacifi c activity and thermoregulation. Can. J. Zoo/. Science 36, 129-139.

HERPETOLOGICAL JOURNAL, Vol. I, pp. 259-263 (1988)

HYPOTHESIS:

A GEOLOGICAL BASIS FOR SOME HERPETOFAUNAL DISJUNCTIONS IN THE SOUTHWEST PACIFIC, WITH SPECIAL REFERENCE TO VANUATU

AARON M. BAUER*

Depanme111 of Zoologl' and Museum of Venebrale Zoologr. Universi/.I' of California. Berkeley. California, USA. 94720. *Prese111 address: Biology Depanmenl, Villanova Universil.1'. Villanova. Pennsylvania 19085. USA.

(Accep1ed21. 9.87)

INTRODUCTION composite regions comes from magnetic anomalies, heat flow st udies, bathymetry and st ratigraphy. Vanuatu is generally considered to form part of the Terranes of the North American Cordillera (Irving, Outer Melanesian Arc, which runs from New Guinea 1979; Coney e1 al. , 1980) and the central Philippines through the Solomons and eventually to Fiji and (McCabe et al., 1982, 1985) have been particularly well Tonga (Holloway, 1979). Within the arc, however, documented. Terranes ofthe former have been used in Vanuatu is relatively isolated by the surrounding an att empt to explain the biogeography of bolito­ Johnson Trough, Vitiaz Trench and Vanuatu Trench glossine salamanders (Hendrickson, 1986) and it is (Coleman, 1970). A spreading ridge is present to the possible that some of the peculiarities of animal east ofthe group. Recent work (Chase, 1971) indicates distribution in Vanuatu may also be explained by its that until 6-8 mybp Vanuatu lay north of Fiji and past fragmentation and movement. Tonga, along the border of a wider arc, adjacent to the Vitiaz Arc (Carney and Macfarlane, 1982). Until the THE VANUATUAN FAUNA AND Miocene Vanuatu maintained its position relative to ITS AFFINITIES Fiji. At that time it began to drift. Subsequent arc bending and counterclockwise rotation has resulted in The herpetofauna of Vanuatu has traditionally been its present position (Gibbons, 1985). This theory is regarded as depauperate Paupuan (Baker, 1928, 1929; consistent with the current thought that many Darlington, 1948), having suffered attrition through orogenic belts are actually composites of smaller filter effects that have reduced diversity away from the blocks (terranes) of varied origins (Craw, 1982). In New Guinea source area. This 'impoverished' fa una general, support for the terrane accretion origin of has, however, been supplemented by species of 260 AARON M. BAUER

presumed New Caledonian ongm (Leio/opisma sequently, diffe rences in the snail fa una (Solem, 1958) atropunctatum) as well as by endemics, chiefly have also been indicated. In all cases the richer fauna associated with a localised radiation of the scincid occurs in New Caledonia and shows distinct affinities genus Emoia (Medway and Marshall, 1975). Although with New Zealand and especially Australia, in contrast this ev aluation of the Vanuatuan reptile fa una can to the poorer, Papuan fauna of Vanuatu. Cheesman explain the origin of most native species, it is limited in (1957) noted breaks within the fl ora of Vanuatu and that it accounts for the taxa present by invoking suggested that the southern islands, Aneityium, equilibrium theory (McArthur and Wilson, 1963, Erromanga and Tanna had probably been connected 1967), an approach more nearly appropriate for with New Caledonia during the Cretaceous, when both biogeographical problem ov er ecological time than for split off from Queensland. She believed that those ov er geological time. Recent advances in the subsequent division occurred in the Late Miocene or geology and paleogeography of the southwest Pacific Early Pliocene. Solem ( 1958) presented conflicting (e.g. Chase, 197 1) suggest explanations for the ev idence fr om land snail distributions and pointed out distribution of certain 'problem' groups of amphibians the existence of the 6400m Vanuatu Trench separating and reptiles in Pacific biogeography (e.g. Platymantis the two regions. Among animal groups, only highly and Perochirus) while maintaining the applicability of vagile forms and some parasitic groups such as the existing hypothesis for the origin of more recently Diptera pupipara (Maa and Marshall, 1981) show a arrived and vagile groups. Gibbons ( 1985) first strong affinity across the Vanuatu Trench. Current suggested the application of knowledge of the views of plate tectonics clearly indicate that Vanuatu changing paleoposition of Vanuatu to herpetological and New Caledonia have never been in contact and problems. This paper expands on his views and generally account for the differences in the fa una!, and suggests that initial reference to historical biogeo­ to some extent, floral affinities of the two areas. graphical implications for distribution may provide As noted by Roux (1913) and Gibbons (1985), the fruitful results in the analysis of the fauna I origins and Loyalty Islands, located approximately lOOkm offthe relations of older island groups. east coast of New Caledonia and 150km from the Like that of other Outer Arc elements, the fa una of southern islands of Vanuatu, possess an intermediate Vanuatu shows Papuan affinities with limited local herpetofauna best characterised as a depauperate New endemism. Mega- and microchiroptera are the only Caledonian fa una with some typical Papuan elements native mammals, and fu lly half of the land birds are and very low endemism. Brown ( 1956) considered the shared with New Guinea (Medway and Marshall, Loyaltys as the link between New Caledonia and the 1975). Further, the number of bird species in Vanuatu rest of Southern Melanesia. Berlioz ( 1962) also noted is in accord with the predictions of the McArthur­ the Vanuatuan influence in the Loyaltys' av ifauna. Wilson model. The herpetofauna is particularly poor The age of the emergent Loyalty Islands is unknown, in comparison with neighbouring New Caledonia or but appears to be very recent, perhaps Pleistocene. ev en with Fiji, fu rther to the east. The Earl of Thus the herpetofauna is probably only of recent Cranbrook (Medway and Marshall, 1975; Cranbrook origin, being derived via overwater dispersal, although and Pickering, 1981; Cran brook, 1985) recognised a parts of Mare may have been emergent for a much Vanuatuan terrestrial reptilian fa una of 20 species. In longer time and may, at one time, have been much addition, there is one introduced species of hylid frog, closer to the New Caledonian source area. Litorea aurea (Meday and Marshall, 1975; Fischthal The relationship of Vanuatu to other elements of the and Kunzt, 1967; Tyler, 1979). To this list may be Outer Melanesian Arc, notably the Solomon Islands added several additional endemic Emoia (W. C. and Fiji, is more problematic. As mentioned, the Brown, in press) and perhaps Crocodylus porosus, herpetofauna of Vanuatu is slightly poorer than that of identified from the trackways on Espiritu Santo Fiji, which is more distant from the Papuan source (Baker, 1928, 1929). Further, Gehyra vorax, now area (elements of the Outer Arc have never been generally recognised as specifically distinct from broadly connected by land, thus overwater dispersal is G. oceanica (contra Burt and Burt, 1932), has also been likely to have been responsible for the introduction of recorded from Vanuatu (Baker, 1928, 1929; Bou lenger, the fa una of the eastern elements of the arc). The Fijian 1885). The Fijian iguana, Brachylophus fa sciatus has herpetofauna, as reviewed by Pernetta and Watling been introduced to the fa una recently by man (pers. ( 1979) consisted of 22 species of reptile and three comm., W. C. Brown). Of Vanuatuan reptiles amphibians (excluding the semi-marine Crocodylus Gekko vittatus(not included by Medway and Marshall, porosus, Derrick, 1965). The addition of newly 1975) and Gehyra mutilata are known from single described species of Emoia (Brown and Gibbons, specimens collected by Baker (1928, 1929) and 1986), Brachylophus (Gibbons, 1981) and Leiolop isma probably do not constitute native species. Similarly, (Zug, 1985), as well as the inclusion of Ramphotyphlops the snake Ramphotyp/ops bra minus is likewise a recent jlaviventerand the removal of introduced fo rms (Bufo addition, probably transported by man (Meday and marinus. Hemidactylus ji'enatus, H. garnotii and Marshall, 1975). I accept the remaining taxa Hemiphyllodactylus typus) yields a native fauna of (approximately 21 species) as constituting the native about 27 species, including the two ranid frogs, herpetofauna of Vanuatu. Platymantis vitianus and P. vitiensis. The existence of A distinct break between the faunas of New the iguanine. Brachylophus, present in Fiji, Tonga and Caledonia and Vanuatu, especially in regard to the Wallis Island, is generally explained by ov erwater herpetofauna of these areas, was noted by Roux dispersal from the Americas (Cogger, 1974; Gibbons, (1913), Baker (1928) and Sternfield (1920). Sub- 1981, 1985). If this is the case, its absence in Vanuatu HERPETOFAUNAL DISJUNCTIONS IN THE SW PACIFIC 261

(where it has subsequently been introduced) would not Lepidodactylus lugubris, Gehyra oceanica, Lipinia be surprising. noctua, Nactus pelagicus), implying recent dispersal The absence in Vanuatu of frogs of the genus across their present ranges. Finally, there are those Platymantis, however, is perplexing. These ranids are species with broad distribution which have been speciose in the Solomon Islands and are represented by transported by aboriginal or modern man. two endemic species in Fiji. Schmidt ( 1930) hypothesised differential extinction in Vanuatu in order to reconcile this gap with a continental, or PATTERNS OF COLONISATION Papuan, origin of the genus. If Brown's (1952) assertion that the ranids are an old group in the The herpetofauna of Vanuatu exhibits a historical Solomons (and Fiji) is correct, the appearance of complexity that belies its modest diversity. At least Vanuatu at its present position in the Neogene would four broad categories of colonisation by reptiles and seem to explain the absence ofVanuatuan Platymantis. amphibians have characterised the western Pacific. Howev er, Tyler ( 1972) considered the ran ids as having The fi rst, and most complex of these, involves Pleistocene entry into the Solomons, and thus into Fiji. movement of taxa most parsimoniously explained by It seems difficult to reconcile the latter view of known changes in paleogeography of emergent land platymanine origins with the terrane theory of masses. Most of the western Pacific islands of Miocene Vanuatuan origin. The most recent discussion of the or greater age obtained their faunas through prior age of this group (Zweifel and Tyler, 1982) implies that contact with other land masses or through dispersal the platymantines have inhabited the region since the from adjacent regions (some now quite distant). The Miocene. herpetofaunas of New Zealand and New Caledonia As first suggested by Gibbons ( l 985) the presence of consist primarily of taxa ultimately derived from elapids (Ogmodon) in Fiji but not in Vanuatu is continental Gondwanaland (Bauer, 1986). In the case likewise accounted for by the relatively recent shift of of Vanuatu, some fauna! elements (e.g. Perochirus) Vanuatu into a southern position. The distributions of may date from previously existing near-neighbour Ramphotyphlops jlaviventer in Fiji and the Solomon relationships with other island groups while other Islands superficially suggests a similar explanation, groups (Platymantis, elapid snakes) may have been but upon closer examination this seems unlikely given excluded for the same reasons. This type of fauna! the difficulty in locating typhlopids, their ability to be origin, historical in the sense that there are geological transported in soil and the uncertain status of even clues to prior positions, allows for the prediction of alpha level systematics in Ramphotyphlops. The distributed attributes of other animal and plant groups absence of Vanuatu would also have left Fiji closer to a of suitable age. New Zealand/New Caledonian source area for A second category of colonisation by organisms is Leiolopisma and account for the absence of unique that based upon a probabalistic theory of dispersal species of this genus in Vanuatu. with the framework of a static (or in other words, Another anomaly of the Vanuatuan fauna is the recent) geography. Predictions of the equilibrium presence of Perochirus guemheri. This gekkonine is models of island biogeography fit this mode and may known from few specimens (Medway and Marshall, be related to the distribution of Candoia and Emoia in 1975; Brown, l 976) and its congeners occur only on Vanuatu. Guam, Tinian, Ponape, Marcus Island, the Truk Thirdly, wide ranging forms, which because of Islands and Kapingamarangi Atoll, all in Micronesia. biological features such as reproductive characteristics While speculative, a former, northern position of (parthenogenesis, low egg-shell permeability, etc.) Vanuatu might account for the presence of this species (Gibbons, 1985) may become distributed and (Gibbons, 1986), as a position to the north of Fiji established rapidly, without speciation (either because would have 'exposed' Vanuatu to dispersing migrants of maintenance of genetic contact or insufficient time from the northwest. Prior to this suggestion the only or habitat differences). Examples in Vanuatu include existing hypothesis for the disjunct distribution of Lepidodactylus lugubris, Nactus pelagicus and Emoia Perochirus had been the past existence of a continental cyanura. mass connecting virtually all of the islands in the Finally, human assisted transport, either intentional Pacific(Ba ur, 1897). or accidental, places a fo urth level of complexity on For the remainder of the herpetofauna of Vanuatu, distribution. The presence of species such as Litoria a more traditional explanation (i.e. dispersal along a aurea and Ramphotyphlops braminus are known track from New Guinea through the Solomons) of examples of this category of colonisation. Although origin seems on the basis of species diversity to apply. the last two modes are primarily responsible for the Thus, for both Emoia and the boid genus Candoia distribution of reptiles in the eastern Pacific (Ineich, species diversity is highest in the Solomon Islands and 1982; McKeown, 1978), Vanuatu, and the western drops to the south and east. Although speciation has Pacific are influenced by all four. occurred in Vanuatu and elsewhere among Emoia The view of Vanuatu as a recent entrant into the populations, morphological differences are minor and Outer Melanesian Arc seems the most reasonable are not inconsistent with recent expansion in the explanation of apparent herpetofaunal anomalies of Pacific. The remainder of the Vanuatuan species are the region, both on geological and zoological grounds. broadly distributed in the western Pacific (Gehyra In addition to the herpetofaunal examples cited above, vorax) or are pan-Pacific - ( Cryp toblephoius boutoni, it also appears to explain disjunctions in the 262 AARON M. BAUER distribution of a wide variety of groups in the Outer American Museum of Natural History. Bulle1in of 1he Arc (e.g. the Cryptotympanini [lnsecta: Cicadoidea] American Museum of Na/Ural His101y 63, 461-597. Duffe l;>, 1986). The analysis of fa una! and floral Carney, J. N. and Macfarlane, A. ( 1982). Geological evidence elements in light of this information may yield further bearing on the Miocence to Recent structural evolution ev idence of Micronesian, or simply northern, affinities of the New Hebrides Arc. Tec/Onophysics 87, 147- 175. in the Vanuatuan fauna, particularly in groups ofpre­ Chase, C. G. ( 1971). Tectonic history of the Fiji Plateau. Neogene origin or arrival in the western Pacific. Bulletin of !he Geological Socie1y of America 82, 3087-3 110. Cheesman, L. E. ( 1957). Biogeographical significance of Aneityum Island, New Hebrides. Nature 180, 903-904. Cogger, H. G. ( 1974 ). Voyage of the banded iguana. ACKNOWLEDGEMENTS Aus1rafian Na1ural His1orr 18, 144- 149. Coleman, P. J. ( 1970). Geology of the Solomon and New I thank Ors. Kevin Padian, Tony Russell, David Hebrides Islands, as part of the Melanesian re-entrant, Wake and Marvalee Wake for their insight and southest Pacific. Pacifi c Science 24, 289-3 14. comments on the ideas discussed in this paper. Jane Coney, P. J .. Jones, D. L. and Monger, J. W. H. (1980). Larson typed the several manifestations of the Cordilleran suspect terranes. Na1ure 288, 329-333. manuscript. This paper originally constituted an Cran brook, Earl of ( 1985). The herpetofauna of Vanuata appendix in a doctoral dissertation submitted to the [sic] (Pacific Ocean). Bulle1in of1he Bri1ish Herpelological Department of Zoology at the University of Socie1_1' II, 13. California, Berkeley. Cran brook, Earl ofand Pickering, R. ( 1981). A check list and The ideas presented in this paper were stimulated by preliminary key to the lizards of Vanuatu. Naika 4, work authored by the late J. R. H. Gibbons of the 3-5. University of the South Pacific, Suva, Fiji. This paper Craw, R. C. (1982). Phylogenetics, areas, geology and the biogeography of Croizat: a radical view. S1·s1ema1ic represents an expansion and interpretation of concepts - Zoology 31, 304-3 16. introduced by Dr. Gibbons (1985) and is dedicated to o his memory. Darlington, P. J. ( 1948). The geographical distributi n of cold-blooded vertebrates. Quar1erl_1· Review of Biologr 23, 105- 123. Derrick, R. A. (1965). The FijiIsland s. Suva: Government REFERENCES Printer. Baker, J. R. (1928). The non-marine vertebrate fauna of the Duffe ls, J. P. (1986). Biogeography of lndo-Pacific New Hebrides. Annals and Magazine ofNatur al His1ory Cicadoidea: a tentative recognition of areas of endemism. 10(2), 294-302. C!adislics 2, 318-336. Baker, J. R. ( 1929). Man and animals in the New Hebrides. Fischthal, J. H. and Kuntz, R. E. ( 1967). Digenetic trematodes London: George Routledge & Sons. of amphibians and reptiles from Fiji, New Hebrides and British Solomon Islands. Proceedings of 1he Helmi/1/ho­ Bauer, A. M. ( 1986). Systematics, biogeography and logical Sociely of Washing/on 32, 244-251. evolutionary morphology of the Carphodactylini (Reptilia:Gekkonidae). Ph.D. dissertation, University Gibbons, J. R. H. ( 1981 ). The biogeography of Brachylophus of California, Berkeley. including the description of a new species, B. vitiensis, Baur, G. ( 1897). New observations on the origin of the from Fiji. Journal of Herpelology 15, 255-273. Galapagos islands with remarks on the geological age Gibbons, J. R. H. ( 1985). The biogeography and evolution of of the Pacific Ocean. No. II. Th e American Naturalist Pacific island reptiles and amphibians. In The biology October 1897, 864-896. of Auslralasian fr ogs and rep liles, 125-142. Grigg, G., Berlioz, J. ( 1962). Les caracti::res de la faune avienne en Shine, R. and Ehmann, H. (Eds). Chipping Norton: Nouvelle-Caledonie. Comptes Rendus de la Socit?!e de Surrey Beatty & Sons Pty. Ltd. Biogeographie 39, 65-69. Hendrickson, D. A. ( 1986). Congruence of bolitoglossine Boulenger, G. A. (1885). Catalogue of the lizards in the biogeography and phylogeny with geologic history: British Museum, volume !. London: Taylor & Francis. paleotransport on displaced suspect terranes. Cladistics 2, 113-129. Brown, W. C. ( 1952). The amphibians of the Solomon Islands. Bulletin of the Museum of Comparative Zoology 107, Holloway, J. D. (1979). A survey of !he Lepidoplera, 1-64. biogeography and ecology of New Caledonia. The Hague: Dr. W. Junk Publishers. Brown, W. C. ( 1956). The distribution of terrestrial reptiles in the islands of the Pacific Basin. Proceedings of the Ineich, I. (1982). Contribution it l'etude des reptiles tettestres Eighth Pacific Science Congress ll!A , 1479-1491. de Polynesie Franc;;aise: taxonomie, ecologie et biogeo­ graphie. Diplome thesis, Universite de Science et Brown, W. C. (In press). A revision of the genus Emoia Techniques du Languedoc. (Reptilia:Scincidae). Occasional Papers of the California Academy of Sciences. Irving, E. ( 1979). Paleopoles and paleolatitudes of North American and speculation about displaced terraines. Brown, W. C. and Gibbons, J. R. H. (1986). Species of the Canadian Journal of Earth Sciences 16, 669-694. Emoia samoensis group uf lizards (Scincidae) from the Fiji Islands, with descriptions of two new species. Maa, T. C. and Marshall, A. G. (1981). Diptera puparia of Proceedings of !he California Academy of Sciences 44, the New Hebrides: taxonomy, zoogeography, host 41-53. association and ecology. Quarterly Journal of the Burt, C. E. and Burt, M. D. ( 1932). Herpetological results Taiwan Museum 34, 213-232. of the Whitney South Sea Expedition, VI. Pacific island MacArthur, R. H. and Wilson, E. 0. (1963). An equilibrium amphibians and reptiles in the Collection of the theory of insular biogeography Evolu1ion 17, 373-387. HERPETOFAUNAL DISJUNCTIONS IN THE SW PACIFIC 263

MacA rthur, R. H. and Wilson, E. 0. (1967). The 1heory of 79-169. Sarasin, F. and Roux, J. (Eds.). Wiesbaden: island biogeography. Princeton: Princeton University C. W. Kreidels Verlag. Press. Schmidt, K. P. ( 1930). Essay on the zoogeography of the McCabe, R., Almasco, J. and Diegor, W. (1982). Geologic Pacific islands. In .Jungle Islands, 275-292. Shurcliff, S. N., and paleomagnetic evidences for a possible Miocene New York: Putnams. collision in western Panay, central Philippines. Geolog1• Solem, A. ( 1958). Biogeography of the New Hebrides. 10, 325-329. Na1ure 81 , 1253-1255. McCabe, R., Almasco, J. and Yumui, G. (1985). Terranes of Sternfeld, R. ( 1920). Zu Tiergeographie Papuasiens und der the central Philippines. In Tec1onos1ra1egraphic terranes pazifischen lnselwelt. Abhandlung herausgegeben von der of cirucm-Pacifi c region, 42 1-436. Howell, D. H. (Ed). Senckenbergischen Natwforschenden Gesel/schaft 36, Circum-Pacific Energy and Mineral Resources, Earth 373-436. Sciences. volume I. Tyler, M. J. ( 1972). An analysis of the lower vertebrate fa una I McKeown, S. (1978). /-la1rniian rep1iles and amphibians. relationships of Australia and New Guinea. In Bridge Honolulu: Oriental Publishing Company. and barrier: the natural and cultural history of Torres Medway, Lord and Marshall, A. G. (1975). Terrestrial Strait, 23 1-256. Walker, D. (Ed.). Research School of vertebrates of the New Hebrides: origin and distribution. Pacific Studies Publication BG/3. Philosophical Transac1 ions of1he Royal Society of London Tyler, M. J. ( 1979). The introduction and current distribution 8272, 423-465. in the New Hebrides of the Australian hylid frog Litoria Pernetta, J. C. and Watling, D. (1979). The introduced and aurea. Copeia 1979, 355-356. native vertebrates of Fiji. Pacifi c Science 32, 223-243. Zug, G. R. ( 1985). A new skink (Reptilia: Sauria: Leiolopisma) Roux, J. (1913). Les reptiles de la Nouvell-Caledonie et des fr om Fiji. Proceedings of the Biological Society of lies Loyalty. l n Nova Caledonia. Zoolgie. volume /, Washington 98, 22 1-231.

HERPETOLOGICAL JOURNAL, Vol. I, pp. 263-271 (1988)

POPULATION ECOLOGY AND CONSERVATION OF TORTOISES: THE ESTIMATION OF DENSITY, AND DYNAMICS OF A SMALL POPULATION

ADRIAN HAILEY*

Depar1111e11t of Zoolog1·. Universit r of Th essaloniki. Greece 540 06.

*Present address: Depanment of Physio/ogr. The Medical College of Sr. Bar1holome1r's Hospira!. Chanerhouse Square. London EC IM 6BQ.

(A ccepted 30.6.87)

ABSTRACT

Part 1: Line transect methods were used in three areas with known densities of Testudo hermanni. This enabled calculation of the proportion of the population available forfi nding (PA F); that is, excluding tortoises in refuges or in thick cover. At times of peak daily activity PAF varied between about 0.1 and 0.3, and was greater in spring than in summer. Males had higher PAF than fe males on summer evenings, the main courtship period, and PAF of males was greater in a population with a male-biased sex ratio. Transect sampling is useful for estimating the density of tortoise populations, provided that the low PA F is taken into account. Values of effective transect width are given for a variety of habitat types. Part 2: A small population ofT. graeca at Alyki was studied between 1980-1986; its size was estimated at about 25 adults, with an even sex ratio. Immature animals were recruited into the adult population, which was stable or increasing during this period. Adult size and juvenile growth rate were similar to other coastal populations of T. graeca in the region. The implications for the conservation of endangered tortoises are discussed.

INTRODUCTION Basic information on tortoise populations is needed to The threat to Mediterranean tortoise populations evaluate this threat (Honegger, 1981; Groombridge, from collection for the pet trade (Lambert, 1969, 1979, 1982 ). This paper considers two aspects of the 1981 a) has now ended. The problem of habitat loss is, population ecology of tortoises which have special however, still present (Lambert, 1984; Cheylan, 1984). relevance to conservation.