Introduction. Robert J. Labanowski 1 * and Andrew J. Lowin 1

Transcription

1 Herpetology Notes, volume 4: (2011) (published online on 19 April 2011) A reptile survey in a dry deciduous forest fragment in northern Madagascar showing new records for the little-known snake Pararhadinaea melanogaster and a range extension for the skink Amphiglossus tanysoma Robert J. Labanowski 1 * and Andrew J. Lowin 1 Abstract. A small area within a fragment of unprotected secondary dry deciduous forest named Antsolipa, located between the protected areas of Montagne d Ambre National Park and Ankarana Special Reserve in northern Madagascar, was surveyed for its reptile fauna over a ten-week period between July and September 2009, during the dry season. A combination of active searching, opportunistic collection as well as pitfall and funnel trapping yielded a total of 19 reptile species including 2 listed as vulnerable on the IUCN Red List of Threatened Species and 7 species on the CITES appendices. As this forest was previously unstudied, this survey provides the first inventory of reptile species for the area, also providing a new locality record for a little known species of colubrid snake, Pararhadinaea melanogaster, of which only six specimens were previously known from a few locations in Northern Madagascar. It is also the first locality where multiple specimens of this species have been observed. The discovery of a skink thought to be Amphiglossus tanysoma is also important, as this would be a northern range extension for this species. The fragmented forests in this area are under increasing threat from logging and clearing, and as few studies have been carried out in these unprotected areas their conservational importance remains poorly known. It is hoped that the results of this survey may help to highlight the rich species diversity contained within these forests, and hopefully lead to some form of official protection of what little remains of these potentially important habitats. Keywords. Conservation, Threatened, IUCN, CITES, Colubrid, Lizard, Scincidae. Introduction Madagascar is well known for its endemic fauna and has outstanding herpetological diversity in comparison to other African countries (Glaw and Vences, 2000). Of the known reptile species inhabiting the country, 92% are endemic, with many being endemic to very small regions, some even confined to individual national parks (Glaw and Vences, 2007). Northern Madagascar is no exception and is continuously establishing itself as an important centre of herpetological diversity and endemism, which has included the discovery of many undescribed herpetological taxa (Megson et al., 2009). Northern Madagascar s herpetofauna is currently conserved by several protected areas including Montagne d Ambre National Park and Ankarana Special Reserve, now separated by savannah and agricultural land. Although a number of herpetological studies have been conducted in northern Madagascar, they have mainly Society for Environmental Exploration/Frontier, Rivington Street, London EC2A 3QP, United Kingdom; address: Research@frontier.ac.uk * Corresponding author. focused on these protected areas of relatively undisturbed forest, and the current information regarding nonprotected areas and secondary habitats remains scarce (Andreone et al, 2003). Patches of fragmented forests remain between the protected reserves, remnants of a once continuous lowland corridor of dry or transitional forest that would have linked them (Ramanamanjato et al. 1999; D Cruze et al., 2006, 2007, 2008). These forests fragments have yet to be thoroughly studied in order to assess the extent to which, if at all, they still act as tenuous habitat corridors, facilitating gene flow for faunal populations. Therefore these forests appear to harbour their own unique herpetofaunal communities and may act as habitat islands for species which are not found in both protected areas, but whose ranges instead overlap in the intervening area. This study hopes to contribute to the scarce literature regarding the herpetological composition of these unprotected forest fragments and highlight the importance of these habitats to a number of rare and regionally endemic species, providing information that can then be used to assess future conservation priorities.

2 114 Robert J. Labanowski & Andrew J. Lowin of species collected and cover as many habitats and microhabitats as possible, as there are few published field studies regarding the herpetofauna of the location (Megson et al., 2009). Figure 1. Location of the study site ( S, E) situated near the village of Tsarikabany between Montagne d Ambre National Park and Ankarana Special Reserve. Study site The study site is situated in an area of fragmented secondary dry deciduous forest known as Antsolipa (E S , elevation: 456 metres), situated approximately 3 kilometres northwest of the village of Tsarakibany between two protected reserves: Montagne d Ambre National Park to the north, and Ankarana Special Reserve to the south, both of which contain a number of their own endemic reptile species (Glaw and Vences, 2007). Antsolipa was once part of a larger continuous forest that would have connected the two reserves. The region is subject to distinct seasonal variation, with a wet season lasting between November and March with a dry season for the rest of the year. The area surrounding the forest is highly ecologically degraded and made up almost entirely of savannah grassland and agricultural land as well as several other small, unprotected forest fragments. Anthropogenic disturbance is also apparent within the forest, with evidence of recent logging, ongoing slash-and-burn clearing and animal traps in the immediate area surrounding the study site. However, the area is of scientific interest due to its location between the two diverse reserves and may well be acting as transitional zone. Materials and methods A range of methods were used including pitfall traps, funnel traps, active searching and opportunistic collection. The reason for applying this range of techniques was to maximise the number Trap line A single trap line was set up with a total length of 150 m. The trap line followed a disused path into the forest that formed a Z-shape, dividing the trap line into three equal sections each measuring 50 m, which covered different elevations and gradients. The trap line consisted of a drift fence together with pitfall traps as used by Raxworthy and Nussbaum (1994), and sunk into the ground at 5 m intervals, with the drift fence passing over the centre of each bucket. Handles were removed from the buckets and small holes were made in the bottom to allow for drainage. Wire mesh double-ended funnel traps, similar to those described by Fitch (1951) were placed along the length of the trap line at 10 m intervals positioned tightly against the drift fence on alternating sides. The funnel traps consisted of fine wire mesh cylinders with an inverted funnel at each end, allowing easy entry but making it difficult to find an exit once inside. The reason for using funnel traps in conjunction with pitfall traps is that pitfall traps are biased towards catching small ground-dwelling species that are unable to climb out. Funnel traps therefore increase the chances of catching larger species, which would otherwise be able to easily escape from the pitfall traps. The trap line was left in place for 55 days during which time it was checked both in the early morning and again in late afternoon. Captures were identified and morphological data taken before being photographed and marked with a small amount of nail varnish to prevent re-recording. Animals were then released in the area close to where they had been captured. Active searching Diurnal and nocturnal searches were carried out along several transects (paths) within the surrounding forest and along the forest edge. These lasted between 30 minutes and 2 hours wherein all unmarked reptiles were collected, placed in ventilated containers and taken back to the camp for recording before being marked and released where they had been found. This method was particularly important for collecting chameleons and geckos, neither of which were likely to be captured by the trap line. Opportunistic collection Opportunistic collections were included in the survey simply because they provide a large proportion of reptile captures and include cryptic and rare species that may not be captured during normal survey methods. Any unmarked reptile encountered at any time during the 10-week survey period was collected and dealt with as described above in Active searching. Results A total of 19 reptile species were recorded over the survey period, consisting of 7 lizards (3 Chamaeleonidae, 2 Gekkonidae and 2 Scincidae), and 12 snakes (2 Boidae and 10 Colubridae). All these species are endemic to Madagascar, with one species,

3 A reptile survey in a dry deciduous forest fragment in northern Madagascar 115 Table 1. Conservation status and distribution of all reptile species recorded during this study. Species Quantity IUCN CITES Relative abundance Chamaeleonidae Ecological distribution Habitat Furcifer oustaleti 36 II A T, AB F, A E Furcifer pardalis 27 II C T, AB F, A E Endemic Furcifer petteri 25 II C T, AB F RE Gekkonidae Phelsuma grandis 2 II R AB F, A E Uroplatus sp. aff. henkeli 14 II C AB F E Scincidae Amphiglossus tanysoma 7 I T F E Madascincus intermedius 6 I T F E Boidae Acrantophis madagascariensis 1 VU I R T F E Sanzinia m.volontany 1 VU I R T A E Lamprophiidae Alluaudina bellyi 1 R T F E Thamnosophis stumpffi 7 I T F RE Dromicodryas quadrilineatus 1 R T A E Ithycyphus miniatus 6 I T, AB F E Leioheterodon madagascariensis 4 I T F, A NE Leioheterodon modestus 1 R T A E Liophidium torquatum 3 I T F E Madagascarophis colubrinus 1 R T F E Pararhadinaea melanogaster 3 R T F RE Phisalixella variabilis 1 R AB F E Abbreviations: Relative Abundance: A=abundant, C=common, I=infrequent, R=rare; Ecological Distribution: AB=arboreal, T=terrestrial; Habitat: F=forest, A=anthropologically disturbed habitat; Endemicity: E=endemic to Madagascar, RE=endemic to the north of Madagascar (regional endemic), N=not endemic to Madagascar. Leioheterodon madagascariensis, also being found on the nearby Comoros where it was introduced (Glaw and Vences, 2000). Of these, 3 species (F. petteri, T. stumpffi and P. melanogaster) are regionally endemic Figure 2. Photograph showing the fragmented nature of Antsolipa and surrounding forests and the ecological degradation of this once forest-covered region. to Northern Madagascar. A complete species list is given in Table 1 along with current IUCN status (IUCN, 2009) and CITES listing (CITES, 2009) of relevant species. Photographic records of some of the species collected during this study are provided in Figure 3. Table 1 also includes ecological data for all recorded species including habitat type (either forested or anthropologically disturbed areas) ecological distribution and relative abundance. The relative abundance was classified using a similar system to that described by Wilson and McCranie (2004), with categories based on frequency of encounters and quantity of individuals found. These were as follows: Abundant (large numbers encountered on a regular basis); Common (encountered on a regular basis); Infrequent (unpredictable encounters, few individuals found); Rare (rarely encountered). Only 1 species (Furcifer oustaleti) was classed as abundant

4 116 Robert J. Labanowski & Andrew J. Lowin D. Table 2. Data collected regarding the new Pararhadinaea melanogaster specimens from Antsolipa Forest. Specimen # Date Habitat Weight (grams) SVL Tail length (mm) Total length (mm) Scale rows Anal scale Subcaudals 1 22/07/09 Forest Divided Divided 2 04/08/09 Forest Divided Divided 3 22/08/09 Forest Divided Divided with 3 species classed as common, 6 infrequent and 6 rare. With regard to habitat, 12 species were found solely in relatively undisturbed areas of forest, with only 3 species found in only heavily anthropologically disturbed areas. 4 species were found to occur in both habitats. A total of 20 reptiles were successfully captured in the pitfall traps, contributing 14% of all reptile captures. 3 species, including both skink and one snake species, were only caught using pitfall traps. The funnel traps were successful in capturing a total of 3 reptiles. Chamaeleonidae were by far the most abundant group found during the survey, with 88 individuals recorded from 3 species, all belonging to the genus Furcifer, contributing 60% of all reptiles and 15% of species recorded. Of these F. oustaleti was most frequently encountered (41%) followed by F. pardalis (31%) and F. petteri (28%). 2 gecko species (Gekkonidae) were recorded making up 11% of recorded reptile species, comprised mainly of an unknown Uroplatus species affiliated with U. henkeli (87%), the other species being Phelsuma grandis (13%). 2 species of fossorial skink (Scincidae) were observed (11% of reptile species) comprised of Amphigossus tanysoma (54%) and Madascincus intermedius (46%). A regionally endemic colubrid species, Thamnosophis stumpffi, was the most commonly encountered snake species making up 25% of snakes recorded (7 individuals), closely followed by the arboreal Ithycyphus miniatus, (21%). 2 species (Acrantophis madagascariensis and Sanzinia madagascariensis volontany) are listed as vulnerable on the IUCN Red List of Threatened species, as well as CITES Index I. Another 5 species are listed as CITES Index II, including all chameleon and gecko species recorded. Discussion Due to the fairly short survey period and the small size of the study area it is very unlikely that all the reptile species present in the area were encountered during this survey. Although it can be assumed that the reptilian diversity of Antsolipa Forest, and most probably the similar surrounding forest fragments, consists of at least 19 confirmed reptile species, the real figure is certainly much higher. Indeed, several unconfirmed reptile species were observed during this study which, due to lack of morphological data being taken, were excluded from the results as positive identification relied on accurate physical data being collected. These species are thought to include Geckolepis maculata, Phelsuma abbotti, and Blaesodactylus boivini. The fact that this survey was conducted during the dry season, when Malagasy herpetofauna is typically at its least active (Glaw and Vences, 2007), also supports the suggestion that many more seasonal species may be contained within the survey area. Other herpetological studies in northern Madagascar have shown that pitfall trapping was notably more successful during the wet season, with capture rates almost doubling (D Cruze et al., 2007). Further surveying within these forest fragments over a longer time frame is therefore required if the actual number of species is to be more accurately estimated. New herpetofaunal species, many reptilian, are still regularly described each year in Madagascar (Glaw and Vences, 2007). However, it is likely that extinctions are also occurring simultaneously due to human activity. Researchers and conservationists probably overlook these extinction events because the most vulnerable species, those that are restricted to transitional or rare habitat types, are easily missed during regional surveys (Raxworthy and Nussbaum, 2000). Although this survey took place over a short period of time, a relatively large number of reptiles were recorded. Both trapping methods yielded results, with pitfall traps capturing all of the recorded skinks as well as all 3 Pararhadinaea melanogaster specimens (see section below). Funnel traps proved to be a useful addition to the trap line, successfully capturing 3 colubrid snakes (all adult Thamnosophis stumpffi) that would have been large enough to easily escape from the pitfall traps. The

5 A reptile survey in a dry deciduous forest fragment in northern Madagascar 117 A B C Figure 3. Photographic documentation of some of the species encountered during this study; A. Sanzinia madagascariensis volontany; B. Furcifer pardalis; C. Uroplatus sp. aff. henkeli; D. Phisalixella variabilis. D regionally endemic T. stumpffi was the most commonly encountered snake species during this study and is known from several locations in northern Madagascar, including nearby Montagne d Ambre. Interestingly, it has not been found at Montagne d Ambre during two decades of intensive surveying, with the last known specimens from this location collected in 1963 (D Cruze et al., 2008). It is known to inhabit both rainforest and dry forest habitats (Wilme et al., 2006). Both threatened boa species indigenous to northern Madagascar were recorded. A single Acrantophis madagascariensis measuring cm was found in Table 3. Data collected regarding the skink specimens believed to be Amphiglossus tanysoma. Specimen # Date Habitat Weight (grams) SVL Tail length (mm) Total length (mm) Scale rows Mid-body width (mm) Hind leg length (mm) 1 17/07/09 Forest /07/09 Forest (regenerated) /08/09 Forest /08/09 Forest (regenerated) /08/09 Forest /08/09 Forest /07/09 Forest

6 118 Figure 4. Map of northern Madagascar showing the localities where Pararhadinaea melanogaster (including the subspecies, P. melanogaster marojejyensis) have been recorded. the stream at the edge of the forest. A large Sanzinia madagascariensis volontany was found in an area of grassland near the forest boundary. The individual measured cm, making it equal to the maximum recorded size for this species (Glaw and Vences, 2007). These two species, along with Alluaudina bellyi and Pararhadinaea melanogaster, have a limited distribution in northern Madagascar. The fact that the majority of species recorded during this survey were found solely within forested habitat suggests that these species, including the newly reclassified Phisalixella variabilis, whose genus contains among the most arboreal-specialised snake species in Madagascar (Nagy et al. 2010), may be restricted to this habitat type, and are unable to adapt to heavily anthropomorphically disturbed environments. New records for Pararhadinaea melanogaster A particularly noteworthy discovery during this study was the presence of the little-known colubrid species, Pararhadinaea melanogaster (the sole representative of a monotypic genus), that was previously known from only 6 specimens collected from several localities in Northern Madagascar (Franzen et al., 2009). Boettger described the species in 1898, with very little information as to where exactly it was found, other than that it was in Madagascar. Since then, more detailed observations have been made, with specimens collected from Nosy Be, Ankarana Special Reserve, and Daraina Conservation Site, with a subspecies, P. melanogaster marojejyensis, collected from the Marojejy Massif (Franzen et al, 2009), Robert J. Labanowski & Andrew J. Lowin all of which are protected areas. A more recent specimen was recorded (but not collected as a voucher specimen) in a small forest fragment near the village of Tsarakibany (Frontier, unpublished data), a few kilometres southeast of the Antsolipa study site, with similar unprotected status. Although P. melanogaster has been observed at other localities in northern Madagascar, only a single specimen has been recorded at each of these localities (Franzen et al., 2009). The area of unprotected forest in which this study was conducted currently contains the highest known density of P. melanogaster, with 3 individuals captured within a very small (less than 100 m²) trap site area in a short period. Therefore, it can be assumed that this forest contains a significant population of these secretive snakes. The physical characteristics of the 3 newly recorded specimens comply with those of previously identified individuals as described by Franzen et al. (2009). Unfortunately, due to lack of necessary permits, no voucher specimens were taken, but detailed head drawings and photographic evidence can be found in Figures 5 and 6, respectively. Although most species recorded in this survey do not seem to be in immediate danger from a conservational perspective, as their ranges include one or more protected areas, the conservation status of Pararhadinaea melanogaster is currently unknown due to the extreme rarity of captures and lack of data regarding the ecology of this snake. Although most of the records of P. melanogaster come from within protected areas across northern Madagascar, their abundance in these localities is unknown. However, considering the extent of intensive surveys carried out in these areas, its abundance appears to be very low. It has been suggested that the apparent rarity of this species may be due to its secretive habits rather than its actual rareness and that the extent of occurrence of P. melanogaster is not necessarily small, possibly occuring across northern Madagascar (Franzen et al., 2009). The results from this survey support the suggestion that P. melanogaster may occur across much of northern Madagascar. The fact that this survey recorded a relatively high density of P. melanogaster compared to its other known localities may suggest that its preferred habitat lies within these unprotected western dry forests. Data regarding the 3 new specimens are shown in Table 2. Range extension for Amphiglossus tanysoma The discovery of a skink thought to be Amphiglossus tanysoma is also noteworthy, as its presence within the study site would be a range extension for this species,

7 A reptile survey in a dry deciduous forest fragment in northern Madagascar 119 A B Figure 5. Pararhadinaea melanogaster specimen from Antsolipa Forest. Drawing of head in (a) dorsal and (b) lateral view. R. Labanowski. previously recorded only as far north as Antsirasira, near Ambanja in the Sambirano region (Glaw and Vences, 2007), an area biogeographically similar to the Antsolipa study site. Due to the lack of permits required Figure 6. Pararhadinaea melanogaster specimen observed in Antsolipa Forest.

8 120 Robert J. Labanowski & Andrew J. Lowin Figure 7. Amphiglossus species, believed to be A. tanysoma, based on its morphometric data and physical appearance. to collect voucher specimens or take DNA samples, the identification of the 7 individuals recorded believed to be A. tanysoma cannot be confirmed. However, the morphometric data and physical appearance of these individuals comply with that of two Amphiglossus species: A. tanysoma and A. phaeurus, the latter not recorded in northern Madagascar. Considering this, it is most likely that the species recorded in this study are the more northerly A. tanysoma. Morphometric data for these specimens is shown in Table 3. Photographic evidence of one individual is shown in Figure 7. Further action Madagascar s forests are among the most biologically rich and unique in the world, (Harper et al., 2007) but those outside the boundaries of protected reserves are continuously under threat. During the short period of this study the area of Antsolipa Forest where the survey was conducted was reduced significantly due to logging and slash-and-burn clearance. These unprotected forests have proved to contain diverse herpetofauna including threatened, rare and relatively unknown reptile species, with potentially many more that are currently undescribed. Yet the alarming rate at which these forest fragments are being cleared, including the locality where this study was conducted, is continuously decreasing the chances of new species being discovered, as well as destroying potentially important habitats for many rare and little-known species. It is important that these unprotected forest fragments continue to be studied in order to properly assess their importance from a conservational perspective. The fact that many threatened and rarely seen species have been found in this relatively small area should prompt further studies and the subsequent design of conservation management strategies. A Figure 8. Photographic documentation of the ongoing slash-and-burn clearance at the study site, Antsolipa Forest; A. Clearing a patch of forest to turn into agricultural land; B. Recently burned B

9 A reptile survey in a dry deciduous forest fragment in northern Madagascar 121 Acknowledgements. I would like to thank all of the Frontier staff and research assistants who worked so hard collecting data, the villagers and chief of Tsarakibany, also Frank Glaw, Miguel Vences and Aurélien Miralles for their help and guidance as well as the Malagasy authorities, in particular the Ministère des Eaux et Forêts for providing the necessary research permits. References Andreone, F., Glaw, F., Nussbaum, R.A., Raxworthy, C.J., Vences, M., and Randrianirina, J.E. (2003): The amphibians and reptiles of Nosy Be (NW Madagascar) and nearby islands: a case study of diversity and conservation of an insular fauna. J. Nat. Hist. 37, Boettger, O. (1898): Katalog der Reptilien-Sammlung im Museum der Senckenbergischen Naturforschenden Gesellschaft in Frankfurt am Main. II. Teil (Schlangen), Knauer (Frankfurt am Main). D Cruze, N.C., Green, K.E., Robinson, J.E., Gardner, C.J. (2006): A rapid assessment of the amphibians and reptiles of an unprotected area of dry deciduous forest in north Madagascar. Herpetol. Bull. 96, D Cruze, N., Sabel, J., Green, K., Dawson, J., Gardner, C., Robinson, J., Starkie, G., Vences, M., Glaw, F. (2007): The first comprehensive survey of amphibians and reptiles at Montagne des Français, Madagascar. Herpetol. Cons. Biol. 2, 2: D Cruze, N., Köhler, J., Franzen, M., Glaw, F. (2008): A conservation assessment of the amphibians and reptiles of the Forêt d Ambre Special Reserve, north Madagascar. Madagascar Conservation and Development. 3, 1: Fitch, H.S. (1951): A simplified type of funnel trap for reptiles. Herpetologica. 7, Franzen, M., Jones, J., Raselimananam A.P., Nagy, Z.T., D Cruze, N., Glaw, F., Vences, M. (2009): A new black-bellied snake (Pseudoxyrhophiinae: Liophidium) from western Madagascar, with notes on the genus Pararhadinaea. Amphib. Reptil. 30, Frontier-Madagascar, Durkin, L. and Steer, M.D. Report 26: Herpetological Surveys of Forest Fragments between Montagne d Ambre National Park and Ankarana Special Reserve, Madagascar. Frontier-Madagascar Environmental Research, Society for Environmental Exploration, UK. Glaw, F. and Vences, M. (2000): Current counts of species diversity and endemism of Malagasy amphibians and reptiles. Diversité et Endémism à Madagascar Glaw, F. and M. Vences (2007): A Field Guide to the Amphibians and Reptiles of Madagascar, 3 rd Edition. Köln, Vences and Glaw Verlag. Glaw, F., Nagy, Z.T., Köhler, J., Franzen, M., Vences, M. (2009): Phylogenetic relationships of a new species of pseudoxyrhophiine snake (Reptilia: Lamprophiidae: Thamnosophis) suggest a biogeographical link between western and northern Madagascar. Organis. Div. Evol. 9, Harper, G.J., Steininger, M.K, Tucker, C.J., Juhn, D., Hawkins, F. (2007): Fifty years of deforestation and forest fragmentation in Madagascar. Environ. Conserv. 34, Hawkins, A., Chapman, P., Ganzhorn, J., Bloxam, Q., Barlow, S., Tonge, S. (1990): Vertebrate conservation in Ankarana Special Reserve, northern Madagascar. Biol. Conserv. 54, 2: Lehtinen, R.M., Ramanamanjato, J., Raveloarison, J.G. (2003): Edge effects and extinction proneness in a herpetofauna from Madagascar. Biodiv. Cons. 12, Megson, S., Mitchell, P., Köhler, J., Marsh, C., Franzen, M., Glaw, F., D Cruze, N. (2009): A comprehensive survey of amphibians and reptiles in the extreme north of Madagascar. Herpetol. Notes. 2, Nagy, Z.T., Glaw, F. & Vences, M. (2010) Systematics of the snake genera Stenophis and Lycodryas from Madagascar and the Comoros. Zoologica Scripta. Ramanamanjato, J.B., Nussbaum, R.A. and Raxworthy, C.J. (1999): A new species of Mabuya Fitzinger (Squamata: Scincidae: Lygosominae) from northern Madagascar. Occas. Pap. Mus. Zool. Univ. Michigan. 728, Raxworthy, C.J., Nussbaum, R.A. (1994): A rainforest survey of amphibians, reptiles and small mammals at Montagne d Ambre, Madagascar. Biol. Conserv. 69, 1: Vences, M. and Glaw, F. (2003): Phylogeography, systematics and conservation status of boid snakes from Madagascar (Sanzinia and Acrantophis). Deutsche Gesellschaft für Herpetologie und Terrarienkunde. 39, Vences, M., Glaw, F., Mercurio, V., Andreone, F. (2004): Review of the Malagasy tree snakes of the genus Stenophis (Colubridae). Deutsche Gesellschaft für Herpetologie und Terrarienkunde. 40, 2: Wilmé, L., Goodman, S.M., Ganzhorn, J.U. (2006): Biogeographic evolution of Madagascar s microendemic biota. Science. 312, Wilson, L.D. and McCranie, J.R. (2004): The herpetofauna of Parque Nacional El Cusuco, Honduras (Reptilia, Amphibia). Herpetol. Bull. 87, Accepted by Zoltán T. Nagy