Herpetologists' League

A Checklist of the Herpetofauna of the Palau Islands (Republic of Belau), Oceania Author(s): Ronald I. Crombie and Gregory K. Pregill Source: Herpetological Monographs, Vol. 13 (1999), pp. 29-80 Published by: Herpetologists' League Stable URL: http://www.jstor.org/stable/1467060 Accessed: 05/10/2008 09:49

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/action/showPublisher?publisherCode=herpetologists.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that promotes the discovery and use of these resources. For more information about JSTOR, please contact [email protected].

Herpetologists' League is collaborating with JSTOR to digitize, preserve and extend access to Herpetological Monographs.

http://www.jstor.org Herpetological Monographs, 13, 1999, 29-80 ? 1999 by The Herpetologists' League, Inc.

A CHECKLIST OF THE HERPETOFAUNA OF THE PALAU ISLANDS (REPUBLIC OF BELAU), OCEANIA

RONALD I. CROMBIE1 AND GREGORY K. PREGILL2 'Division of Amphibians and Reptiles, United States National Museum, Smithsonian Institution, Washington D.C. 20560 2Department of Biology, University of San Diego, San Diego, CA 92110-2492

ABSTRACT: The Palau islands of western Micronesia comprise one of the most physiographically varied archipelagos in the Pacific. There are some 350 islands of volcanic and coralline origin that range from Babeldaob at 333 km2 (about 80% of Palau's total land area) to others, unnamed, that are less than one hectare. A barrier reef encircles most of the archipelago creating a lagoon up to 20 km wide. The majority of the islands are uninhabited, accessible with difficulty only by boat at high tide, and are densely vegetated over steep, rocky terrain. Because of their proximity to the Philippines, New Guinea, the Moluccas, and Borneo, the Palau islands host a significantly richer herpetofauna with more unique taxa than is found on other Pacific islands to the east. As with most other Pacific island groups, however, the herpetofauna has been poorly documented. Since the earliest investigations that began in the nineteen century, Palau's herpetofauna has been uncovered piecemeal. No sustained field effort had been conducted until we began a systematic survey of the islands in 1992. Since then we have collected in Palau every month of the year at least once. This checklist of 46 species of reptiles and amphibians includes almost twice as many taxa as were known from Palau (Owen, 1977) at the beginning of our study. Among these are new species or Palauan records of Gehyra, Hemiphyllodactylus, Lepidodactylus, Nactus, Perochirus, Emoia and Sphenomorphus. Because of the difficulties in surveying so many islands we believe that additional taxa and distributional records await discovery; even the smallest island in Palau can turn up sur- prises. Nonetheless, it is appropriate to consolidate what is now known about this intriguing fauna and environment. We provide an historical summary of previous collecting efforts, an overview of the islands' geology and habitats, and for each species account we include distributional information, taxonomic and systematic discussion, natural history observations, and a comprehensive bibliography. Key words: Micronesia, Palau Islands, Amphibians, Reptiles

PERHAPS THE MOST zoogeographically island group with a more diverse extant significant yet poorly known archipelago in fauna that is closer to potential source ar- the Oceanic realm is the Palau Island eas. Therefore, in 1992 we began extensive group (now the independent Republic of surveys of the extant and extinct verte- Belau) in western Micronesia. The earliest brates of Palau. Although we have accu- herpetological investigations of this area by mulated significant collections and data in the Germans in the mid-1800's revealed our field work to date, each additionaltrip new taxa in genera (Hypsilurus, Spheno- yields important (and often surprising) morphus, and others) otherwise unknown new specimens and distributional infor- from the oceanic islands of the Pacific mation. Nevertheless, given the consider- Plate, and the few subsequent field studies able recent increase of interest in Micro- continued to yield additional novelties. Al- nesian biology and conservation (e.g., though these early collections suggested a Rodda et al., 1991), we feel it is appropri- diverse fauna with unexpectedly high lev- ate to present our working list. els of endemism, they were insufficient for Modern summaries of the Palauan her- addressing the numerous systematic and petofauna (e.g., Cassell et al., 1992; Oto- zoogeographic problems of this region. bed and Maiava, 1994) have had to rely on Because many other islands in the Pa- the incomplete and taxonomically outdat- cific have relatively depauperate extant ed list of Owen (1977). Despite the ad- vertebrate faunas but much larger lists of mitted shortcomings of the data, our list recently extinct taxa (e.g., Steadman, reflects current herpetological taxonomy 1995), Palau provides an ideal opportunity (to August 1998) and it contains nearly to see if this pattern also obtains on an twice as many species as listed by Owen

29 30 HERPETOLOGICAL MONOGRAPHS [No. 13

(1977), with many systematic problems think that the following list adequately duly noted. It is our intention to resolve documents that the discovery phase is far some of these problems with future field from over. work and collections. The strong zoogeo- graphic affinities between Palau and the GEOGRAPHY AND ECOLOGY Papuan region, and additional influences Spelling of Place Names from the Philippines, Borneo, and Ocean- The inconsistent spelling of Palauan ia, further complicate systematic studies. place names is attributable to idiosyncratic This is especially so with respect to the useage by four separate foreign adminis- logistics of obtaining pertinent compara- trations over the past one hundred years. tive material from such a broad area that These administrators made a limited at- includes several political entities not tempt at using native names, and often known for expansive cooperation with sci- misinterpreted their meaning when they entific endeavours. We hope that our ac- did. Moreover, none of these languages cumulated specimens and data-base for (Spanish, German, Japanese, English) is Palau are such that future studies involving adequate to fully capture the phonics of this area will be facilitated. the Palauan tongue, such as the clipped The taxonomic of our list is com- aspect nasal-quality ng. Even within Palau it is prehensive, but we particularly want to common to see the same name spelled two stress the incomplete nature of the distri- or three different ways. There are, for ex- butional data we present. Although we list ample, some 15 spellings for Palau itself, all island records currently available for with the most common alternate being each species, the results of earlier surveys "Belau". The population center, both the plus our own field work have sampled only island and the town, called "Koror" by a small fraction of the multitude of islands practically everybody who lives there is in Palau, with varying degrees of rigor. The spelled "Oreor" on many maps. For all logistics of even briefly sampling islands place names we follow the spelling found spread across nearly six degrees of latitude on the USGS Topographic Maps of the Pa- are daunting and expensive; the survey lau Islands (1983-1984), which are the work itself is grueling and often danger- most detailed available and are used by ous. The Rock Islands in particular are of- most governmental agencies. We note, ten very difficult to land on, much less to however, that some of the spellings on survey adequately for cryptic and elusive these topographic sheets are at odds with amphibians and reptiles. Even the smallest local or casual useage, or are not followed and islands should seemingly insignificant in older (e.g., Anon., 1957; Brigham, 1900; not be ignored, as demonstrated by the re- Bryan, 1971) or recent (Karolle, 1993; cent discoveries of Lepidodactylus pauro- Motteler, 1986) literature. Finally, most of lepis (Ota et al., 1995), Emoia impar, Lep- Palau's smaller islands are unnamed. idodactylus sp. nov., and Nactus sp. (this The following is a list of our island local- paper). Our distributional data are intend- ities and their common alternate spellings. ed to convey only the most general pat- Coordinates are for the center of the island/ terns, many of which will undoubtedly be group (Lat. 0.000 N, Long. 000.000 E) modified with additional field work. We but hope that the data presented Babeldaob (Babelthuap, 7.5 134.5 here are not used to generate or justify Babeldaop) "conservation" efforts Bablomekang 7.14 134.316 counterproductive Beliliou (Peleliu) 7.0 134.25 that will hinder or prevent the much-need- Chemoi Island Group 7.19 134.38 ed continuing data collection that is nec- Euidelchol (Ashakasengu) 7.19 134.36 essary to make sense of the Palauan her- loulomekang 7.134 134.304 Kmekumer Islands 7.19 134.24 petofauna. In short, we emphasize that (island these data are for 46 from Wiles and presented general use, Conry, 1990) with the major caveat that they are not the Malakal (Ngemelachel) 7.33 134.45 "last word" in Palauan herpetology. We Mecherchar (Eil Malk) 7.15 134.36 19991]1I9 HERPETOLOGICALEPTLGCA MONOGRAPHSOORPS3 31

FIG. 1. The islands of Micronesia and the southwest Pacific.

Ngcheangel Atoll (Kayan- 8.07 134.7 Oreor (Koror) 7.34 134.49 gel, Ngerebelas, Nger- Pkulaklim (Orrak) 7.347 134.565 iungs, Orak) Toiracholodochoel 7.339 134.455 Ngeanges (Necos) 7.206 134.370 Uchulameradel 7.507 134.491 Ngeaur (Angaur) 6.9 134.15 Uchulangas 7.340 134.460 Ngebad (Ngabad) 7.019 134.270 Ulebsechel (Auluptagel) 7.31 134.47 Ngebedangel (Ngoba- 7.27 134.33 Ulong (Aulong) Island 7.28 134.29 sangl) Group Ngedbus (Ngesebus) 7.06 134.26 Ngelsibel 7.187 134.344 Southwest Islands: Ngemlis Islands (Baile- 7.12 134.25 chesengel, Iilblau, Fana 5.36 132.22 Ngemlis) Helen 3.0 131.8 Ngercheu (Ngergoi; 7.09 134.28 Merir 4.29 132.31 Pulo Anna 4.66 132.95 Carp) (Palo Anna) 132.22 Ngerdis (Kobasang) 7.35 134.439 Sonsorol 5.31 Ngerduais (Nardueis) 7.355 134.582 Tobi 3.01 131.18 Ngerechong 7.11 134.36 Ngerechur 7.750 134.620 Physical Description Ngerekebesang (Araka- 7.35 1 34AA.4"" besan) ...t. The thousands of islands and atolls in Ngermalk 7.335 134.459 Micronesia have a total land area of only Ngeruchubtang (Ngeru- 7.083 134.327 about 3,227 km2, roughly 0.03% of the sur- gelbtang) face area between 0?-20? N latitude and Islands 7.17 Ngerukeuid (Sev- 135?-170? E The four ar- enty Islands; islands 4, longitude. major 12, 13, 16, 24, 34 of chipelagos of Micronesia, from east to Wiles and Conry, west, are the Gilberts (Kiribati), Marshalls, 1990) Carolines, and Marianas (Fig. 1). Palau is Ngeruktabel (Urukthapel) 7.25 7.365 134. 429 the westernmost group of the Caroline Is- Ngerur N Ngesbokel (Gechbokul) 7.508 134.488 lands, centered near 7?20' latitude and Ngetmeduch 7.357 134.496 134? E longitude. The island of Mindanao 32 HFERPETOLTOGICALHERPETOLOGICAl, MONOGRAPHS [Vo[No. 1312 in the southern Philippines is about 800 km west of Palau, whereas Irian Jaya and the Moluccas are, respectively, to the south and southwest about that same dis- tance; Borneo is an additional 700 km. All of the Gilberts and Marshalls are coral atolls or low islands. The Carolines include both volcanic and coralline islands. Car- tographically, Palau is included in the Car- oline Islands, but the western Carolines (Palau and Yap), as well as the Marianas, are exposed peaks of undersea ridges stretching between Japan and New Guinea and are not located on the Pacific Plate. The Palau islands originated on the now dormant southern section of the Palau- Kyushu Ridge, which formed by about 43 million years ago (mya) in a complex series of rifting and boundary shifts of the Pacific Plate margin during formation of the Phi- lippine Basin (Kroenke, 1984; Seno and Maruyama, 1984). Volcanism along the Pa- lau-Kyushu Ridge ceased in the earliest Miocene about 29 mya. Subsequently, the Palau-Kyushu ridge fractured and the eastern segment ultimately developed into the West Mariana Ridge. Back-arc spread- ing between the the two submerged ridges FIG. 2.-The main islands of the Palau archipelago. (Palau-Kyushu and West Mariana) wid- ened the Philippine basin to the east, and for the last 17 mya the basin has been younger at ca. 4 mya (Springer, 1982). The moving northwest with respect to Eurasia. Carolines are being carried westward to- Thus, Palau, Yap and the Mariana Islands ward the Philippine Plate margin. rest on the eastern margin of the Philip- The Republic of Belau includes the Pa- pine Plate where it contacts the western lau Islands proper and a cluster of remote edge of the Pacific Plate. The Palau, Yap outliers 320 to 480 km to the southwest and Mariana Trenches that mark the sub- (Fig. 2). Each of the latter (Tobi, Sonsorol, duction zone at plate contact are some of Fana, Merir, Pulo Anna and Helen) are the deepest waters on earth. When the is- separated from one another by from 130 lands that are now Palau first emerged to 280 km of open ocean. Few people out- above sea level is uncertain. The oldest or- side of Palau are aware of them, and no ganic limestones are early Miocene (ca. 25 one in the west has ever bothered to give mya), which suggests that the volcanics them a collective name other than the were emergent sometime before that, per- Southwest Islands. The Palau islands haps by late Oligocene (ca. 30 mya). themselves are arranged in a 160 km curv- The eastern Caroline Islands, by con- ing chain from Ngeruangel reef in the trast, are comparatively young and origi- north to Ngeaur Island in the southwest. nated on the Pacific Plate. From Kosrae in The total land area (including mangrove) the east to Woleai (Ulithi Atoll) in the is approximately 415 km2. Most of Palau's west, the eastern Carolines stretch across 350 islands are enclosed in a reef: barrier the Pacific for 2100 km. The Truk (Chuuk) to the west and fringing on the east. Only Islands are estimated to have originated at Ngeaur in the southwest and Ngcheangel ca. 12-14 mya, whereas Kosrae is much Atoll and Ngeruangel reef in the north are 1999]199IEPTLGCA HERPETOLOGICAL MONOGRAPHSOORPS3 33

ang or Malakal, each of which is connected to Oreor by a causeway. The islands are of four geological types: Babeldaob 1) volcanic, 2) high limestone, 3) low lime- stone, reef or atoll, or 4) a combination of 5 km volcanics and limestone. The Military Ge- ology of Palau compiled by the U.S. Army of Corps of Engineers (1956) remains the most thorough description of Palau's geo- logical features. The following summary is based on that compendium. Volcanic is- lands include Babeldaob, western Oreor, Ngerekebesang, Malakal, and numerous tiny islets in the northern end of the la- goon. Contours of the larger volcanic is- lands are generally rolling hills punctuated by steep irregular outcrops or other prom- inent landmarks. Palau's highest point, 242 m Ngerchelchuus in northwest Babeldaob, is one such feature. The largest evident volcanic center is Ngeremeduu Bay on the west coast of Babeldaob. The original cone collapsed leaving an eroded caldera which is now the bay. Poor exposures elsewhere on the volcanic islands obscure most for- mer eruption centers, a possible exception being the volcanic breccias and the incli- nations of pyroclastic rock on Ngerekebes- ang. Coastlines of the volcanic islands are irregular, sometimes steep, and often form small bays. Approximately 125 km of Ba- beldaob's 157 km coastline is bound FIG. 3. The island of Babeldaob, Republic of Palau. by mangrove. The extent of calcareous beach sand on the volcanic islands is very limited, outside the lagoon. The lagoon is approx- and mainly confined to Babeldaob's north- imately 20 km at its widest, covers about east and east coast. Each of the larger vol- 1450 km2, and varies in depth from barely canic islands is dissected by streams, but a meter to over 20 m in a few places. Out- the only perennial flows are on Babeldaob. side the reef ocean depths drop hundreds On Oreor and Ngerekebesang the radial of meters almost immediately. drainages slow to a trickle when rain ceas- Babeldaob (Fig. 3) is the largest island es for more than a week or so. The major at 333 km2, accounting for about 80% of streams and rivers of Babeldaob are more Palau's total land area. The next largest is- or less aligned with the north-south axis of lands are Ngeruktabel (20.2 km2), Beliliou the island, but are linked irregularly with (12.4 km2), Ngeaur (8.3 km2), Mecherchar connecting tributaries. The largest river, (8 km2), Oreor (7.5 kin2), Ulebsechel (4.3 Ngermeskang, flows for about 16 km into km2) and Ngerekebesang (2.3 km2). Most Ngeremeduu Bay from the northeast. The of the other islands are less than 1 km2 and largest body of standing freshwater is Lake many are no more than a few hectares. Ngerdok (approximately 30 hectares) lo- About two-thirds of Palau's 15,500 inhab- cated about 2 km from the east-central itants live on Oreor, and practically all of coast of Babeldaob. the other residents live in the vicinity, ei- The high limestone islands number ther on southern Babeldaob, Ngerekebes- more then 300, but comprise only about 34 34 HERPETOLOGICAL MONOGRAPHS [No.~~~~~~~~~~[No. 13

from southwestern Babeldaob to Beliliou. oNgerur Ngerekebesang O-or J West of this line there is slight westward Ngerchaol tilting. Although there are occasional coves and narrow beaches, the shorelines of the limestone islands are often nearly vertical, Butottoribo Ul. secU tu? , n and almost everywhere are undercut by solution notches horizontal over- Ngebedangel Nler uce forming 0pnf Ngeruktabel hangs 1-5 m deep. Undercutting of the smaller islands gives them the appearance

Euidelchol _, 0 of mushrooms. Ashore, solution features of Ngerukeuidlids sharp, jagged pinnacles, pits and rubble make for treacherous Mercherchar exceedingly footing. Bablomekang I Springs occur on some of the high lime- % Ngemihslids stone islands but there is no surface drain- 55*.* * V Ngercheu Ngerechong age. Caves, sinks, and other subsurface so- lution features are abundant in this highly RockIslands - Palau karstified terrain. On several of the larger t-; Ngedbus 5km Rock Islands numerous interior depres- /l. Ngebad sions extend below sea level to form saline lakes.

Beliliou The reef islands and atolls of Palau re- sult from minor uplift of reef rocks and wave building accumulations. The largest FIG. 4.-The Rock Islands of the Palau lagoon. and best developed of Palau's atolls is Ngcheangel (Kayangel), located about 3 km beyond the north end of the Palau 10% of Palau's land area. They include reef. It is more or less oval, about 7 X 3.5 practically all of the islands in the lagoon. km, and embraces four small islands Those in the central and southern part of (beach deposits) on its east side. the lagoon are referred to locally as the Several of Palau's islands are a combi- Rock Islands (Fig. 4). Ngeruktabel and nation of the preceding geological types. Mecherchar are the largest. The Rock Is- High limestone ridges are present on the lands are formed in the Palau Limestone south end of Babeldaob. In a zone south which consists of light-colored, poorly and southwest of Babeldaob volcanic and bedded, porous to very dense raised reef high limestone islands occur adjacently or and lagoonal coralline deposits laid down combined. The two volcanic islands, Nger- in shallow warm water. The age of the Pa- ekebesang and Malakal, are separated by lau Limestone ranges from Miocene in the the high limestone island of Ngerchaol. northern part of the Lagoon and parts of Western Oreor is underlain by volcanics, Beliliou (Bloody Nose Ridge) to Pleisto- whereas the southeast part is a high lime- cene elsewhere. The larger islands are stone ridge. The two larger islands at the characterized by steep, arcuate ridges southwest end of the Palau group, Beliliou flanked by descending skirts of parallel and Ngeaur, consist of high limestone ridg- and/or tangential ridges that once marked es, low platforms, and beach deposits. exposed, successive reef fronts. The small- Today, the Palau islands exist along a er rock islands probably derived from scat- continuum of coastal morphologies deter- tered, irregular deposits in the lagoon. The mined by eustatic changes in sea level dur- limestone islands were raised to their pres- ing the Pleistocene. During the last glacial ent extent by asymmetrical upwarping fol- maximum (18,000 years ago), when sea lowing the Neogene rifting and faulting levels may have depressed as much as 120 that was associated with the tectonic and m, the entire lagoon and most of the ar- volcanic activities in the area. Uplift to as chipelago's 350 islands would have formed much as 200 m elevation occurs on an axis a single island of approximately 2250 km2, 1999] HERPETOLOGICAL MONOGRAPHS 35 more than five times the present area of yield such crops as bananas, cassava, co- emergent land. During periods of higher- conut, papaya, mango, pineapple, and be- than-present sea level, for instance in the tel nut. Urban areas occupy about 1% mid-Holocene (e.g., Allen, 1997), Palau (Otobed and Maiava, 1994). Our impres- was more fragmented than it is today. The sions would put the percentage of agricul- effect of these oscillations on the terres- tural and urban land considerably higher. trial biota was undoubtedly considerable. Besides amphibians and reptiles, the Indeed, the present distributions of Pacific vertebrates of Palau include a fair diversity island amphibians and reptiles in general of freshwater fishes and birds, but a de- must be considered in view of these sea pauperate mammal fauna. The resident level fluctuations (e.g., Gibbons, 1985; freshwater and land birds consist of wide- Gibbons and Clunie, 1986; see also Alli- spread oceanic species, species confined to son, 1996). Micronesia, species found elsewhere only in Indonesia or the Philippines, endemic Habitat and Land Vertebrates species, and a few introductions. Palau's Palau's climate is tropical and rainy. modern avifauna has received much more Mean annual temperature is 27.2 C (81?F) attention than its herpetofauna (Baker, and mean daily temperature varies no 1951; Engbring, 1983, 1988; Engbring and more than 1 degree throughout the year. Pratt, 1985; Pratt et al., 1980), although it On average the islands receive 355 cm would be surprising if these islands have (140") annual rainfall with the wetter not experienced extinctions of native bird months (July and August) receiving 25-50 species similar to what has been discov- cm, the drier (February to April) 15-20 ered elsewhere in Oceania (Steadman, cm. A survey of vegetation and timber re- 1995), including the nearby Marianas sources of Palau was drafted by the U.S. (Steadman, 1999). The endemic land Forest Service (Cole et al., 1987; MacLean mammals known from Palau are the Palau et al., 1988), and some of the vascular Sheath-tailed Bat, Emballonura semicau- plants were listed by Fosberg et al. (1980). data palauensis, the "Marianas" Fruit Bat, A soil survey of Palau by Smith (1983) in- Pteropus marianus pelewensis, and the cludes 1:10,000 scale aerial photo maps of larger Palau Fruit Bat Pteropus pilosus. Babeldaob, Oreor and the causeway is- The Palau Fruit Bat has not been seen lands, Beleliou, and Ngeaur. Cole et al. since the 19th century and is presumed to (1987) estimated that three-fourths of Ba- be extinct (Wiles et al., 1997). Introduced beldaob and the other volcanic islands are taxa include rodents Rattus spp. and Mus covered in native upland forest, in which musculus (see below), as well as the Asiatic Calophyllum, Ficus, Elaeocarpus and musk shrew Suncus murinus, confined to Campnosperma are common genera. In Ngeaur (see Ruedi et al., 1996 for Suncus the past few decades native upland forest zoogeography). The introduced Crab-eat- has been greatly reduced, and most of the ing Macaque, Macaca fascicularis, is most remaining area, especially on Babeldaob, common on Ngeaur, and a few individuals is covered by coarse grasses burned an- also occur on Babeldaob. nually, along with scattered pandanus and A review of Palau's freshwater fishes was ferns. Mangrove (especially Rhizophora compiled by Bright and June (1981), but sp.) accounts for about 11% of Palau's veg- very few of Babeldaob's rivers and streams etation cover, practically all of which is (particularly in the eastern sectors) have found around Babeldaob, Oreor, Malakal, been sampled, much less properly sur- Ngerekebesang, and Beleliou. The luxu- veyed. Undoubtedly endemic, unde- rient limestone forests of the Rock Islands scribed species await discovery (e.g., Par- form impressive, dense stands with cano- enti and Maciolek, 1993). pies reaching 20 m or more on the larger Human impact is most obvious in and islands. Freshwater swamps and marshes around Palau's main habitation centers in Palau comprise about 5% of the area. (Oreor, Ngerekebesang, Malakal, Airai) Cultivated lands comprise about 3-4% and and surrounding settlements and planta- 3626 HERPETOLOGICAL MONOGRAPHS [No. 13 tions. Much of Babeldaob has been defor- 1998; Cassell et al., 1992). The human ested, especially the southern half. Al- population is likely to double in 5-10 years though Palau's present human population (Maiva and BNRD, 1994), and at least one is relatively small (about 15,500), estimates major road is under construction to the of around 40,000 native Palauans were be- north end of Babeldaob where none pres- lieved to be present at the time of first ently exists. Erosion and siltation of some Western contact (Johannes, 1981; Osbor- Babeldaob water courses are already evi- ne, 1966). There is some dispute of that dent. Tourism is burgeoning, and numer- figure, which is based on accounts of the ous hotels are planned or being built in early traders to the islands (Johannes, Oreor, southern and eastern Babeldaob, 1981). Nonetheless, ancient agricultural Malakal, and on the Rock Islands. The terraces and other archaeological indica- government of Palau has proudly and jus- tors suggest that the population was sig- tifiably promoted its marine resources as a nificantly greater than it is today (Craib, tourist attraction. We hope that this does 1983; Osborne, 1966). During this century not come to the detriment of its terrestrial population highs were in the early 1940's ecosystems. when 25,000 Japanese residents and native Palauans lived in the islands (U.S. Army, HISTORICAL SUMMARY 1956; see also Bailey, 1991). The Spaniards were the first Europeans WWII brought substantial devastation to land on Palau, claiming the islands for to parts of Palau. Beliliou, the site of one the King of Spain in 1710. Their presence of the bloodiest battles in the Pacific was uneventful, particularly biologically (Hough, 1990; Ross, 1991), was burned except perhaps for the first introduction of over most of its area. The island has gen- rats. Several European ships were wrecked erated dense second growth forest over in the treacherous waters around Palau in the past 50 years. Ngeaur, Oreor, and the 18th and 19th Century and stories of southern Babeldaob were also heavily the interactions between the marooned bombed by American forces, but of these sailors and Palauans have been amply only Ngeaur was invaded. Despite the im- chronicled (Keate, 1802, 1808; Martin, pact of human occupation and WWII, Pa- 1980). Not surprising, the early Pacific ex- lau still claims some pristine habitat, or ploratory expeditions usually avoided Pa- that which is but moderately disturbed. lauan waters in favor of the better harbors The Rock Islands are practically all pri- on Guam, Yap, and Kosrae, so their natu- mary forest due to their steep slopes and ralists and collectors had little or no op- brutal topography (Johnson, 1972). Yet portunity to sample the Palauan fauna. even there one of the most odious mea- The French ships Astrolabe and Zelee sures of human disturbance is conspicu- briefly contacted the residents of Beliliou ous-introduced Black Rats (Rattus rat- in January 1838 (Rosenman, 1992:162), tus). Our campsite on uninhabited Ulong but we have found no herpetological spec- was overrun by hundreds of rats one night imens that were collected at that time. The in January 1995. During the same month French did, however, make the most sig- in 1993 we trapped along a 200 m forest nificant early contribution to Micronesian transect in Ngaraard State (northern Ba- herpetology during the voyage of the Co- beldaob) and had a capture rate of about quille, 1822-1825. Many of the most com- 73% (19 of 26 traps) in one hour. Rats can mon and widespread Pacific lizards were also be seen commonly in Koror town by described from the island of Oualan (now day. The effect of R. rattus as a predator Kosrae in the eastern Carolines), four of on Palau's herpetofauna and birds has not which also occur in Palau (Gehyra ocean- been documented, but we suspect that it ica, Emoia atrocostata, Lamprolepis sma- is substantial. ragdina, Lipinia noctua). Rapidly growing economic interests in Lessons's (1830) zoological report of the Palau probably will adversely affect the na- Coquil expedition generated confusion on tive fauna in the next few decades (Anon., several levels, however. Brygoo (1986) 1999] HERPETOLOGICALMONOGRAPHS 37 pointed out that the Atlas of plates from scribed Palauan lizards Sternfeld referred the Expedition appeared in 1826 but the to. text not until 1830, so Lesson's taxa must The Japanese period (1914-1944) pro- date from the Atlas. The types of several duced relatively little herpetological re- of the species are no longer extant (Bry- search, as summarized in Utinomi's Biblio- goo, 1986) and Barbour (1911) suggested graphica Micronesica (Fisher, 1947). No that Lesson's type locality (Oualan, now new herpetological taxa were described, Kosrae in the eastern Carolines) was sus- and most of the information was published pect. Loveridge (1948) disagreed and crit- in obscure, generally unavailable journals. icized Barbour's interpretation but we feel Messel and King (1991), in summarizing that Barbour's suspicion was justified for at the crocodilian literature, mentioned two least some of the taxa. The situation is significant papers by Motoda (1937, 1938). complex and a lengthy discussion is not ap- However, we have been unable to locate propriate here since it has no direct impact any extant herpetological collections that on Palau, although we hope to provide were made by the Japanese during their clarification in a future paper. 30-year occupation of Palau. U.S. influence in Palauan Serious herpetological research in Palau herpetology began only after the Germans suceeded began with personnel of the Naval Medi- cal Research Spain in 1899. Unlike the Spanish, Ger- Unit (NAMRU II), who were active in Palau soon after man merchants and diplomats were actu- the islands were secured from the ma- ally resident in Palau. Particularly notable Japanese. Biological terial was collected from November 1944 among them was Carl Semper, who wrote until about 1945 an interesting book on his Palau experi- September (Baker, 1951; USNM records). col- ences (Semper, 1925, English translation catalogue Although efforts centered on Beliliou and Semper and Craig, 1982). Semper was an lecting offshore islands, the team also obtained experienced field naturalist who had made herpetological material from Ngeaur, important collections in the Philippines Ngercheu, and Ngcheangel Atoll. The and elsewhere in Asia. His Palau herpe- NAMRU collections provided the first tological material was studied Wilhelm by distributional data (island and Peters in Berlin, who described new meaningful many localities) for Palau taxa specific amphibians (Peters, 1867a,b, 1874, 1877, 1879). and most of the NAM- Most of the Berlin have no lo- reptiles. Although specimens RU material was in USNM, one data other than the island var- deposited cality group, member of the team, William Lewellen, rendered as "Pelew", "Pelew-In- iously contributed most of his Ngeaur specimens seln" or "Palaos". Additional Palau (CM 29085-29103, collected between 19 material from = in (mostly Angau Ngeaur) November 1944 and 13 January 1945) to the was Senckenberg Museum, Frankfurt, the Carnegie Museum. Sternfeld who reported by (1914, 1920), Until recently, the most extensive and mentioned "another new (1920:385) that, important Palauan herpetological collec- lizard from Palau" was present in the tion was made by H. Adair Fehlmann dur- Senckenberg collection and that it would ing his ichthyological surveys in the mid "also be described here." However, no 1950's (Fehlmann, 1960). Most of the fish new Palauan lizards were described in the data remain unpublished but his discovery 1920 paper or any subsequent publications of the remarkable endemic skink Lipinia by Sternfeld before he was executed by leptosoma appeared early (Brown and the Nazis in 1943, nor are there type spec- Fehlmann, 1958). Fehlmann's herpetolog- imens of any Palauan lizards in the Senck- ical specimens were deposited in the Stan- enberg collection (Mertens, 1967). The ford University collection (now housed at material of this new species alluded to by CAS) and they represent important base- Sternfeld was not located during a visit to line data on distributions before extensive Senckenberg by RIC in May of 1986, and development on Palau. we are unsure which of the still unde- The earliest natural history observations 38 HERPETOLOGICAL MONOGRAPHS [No. 13 on the endemic Palauan frog (Platymantis sufficient data from the Southwest Islands. pelewensis) was by a Palauan, Kenji Atoda This chain of islands provides a potential (1950). Palauan citizens and residents have stepping stone pathway between Palau continued to contribute meaningfully to proper and the Greater Papuan region, herpetology ever since. Although he was particularly during periods of lower sea severely criticized by Messel and King level. The few extant herpetological rec- (1991) for encouraging exploitation and ords from the islands provide many sur- even extermination of crocodiles in Palau, prises, as much for the species that have Robert Owen was the entomologist and not been collected as those that have. Chief Conservation Officer of Palau for Donaldson's (1996) discussion on the strik- many years, and was unfailingly helpful to ing diversity and complexity of relation- visiting researchers. He provided data and ships in the fish fauna convinces us that specimens to many individuals (including the terrestrial fauna will be equally signif- us), and meaningfully aided projects in- icant. volving the Palauan flora and fauna during SPECIES ACCOUNTS his tenure there. Owen's successors, par- ticularly Demei Otobed and Kammen Forty-six species of amphibians and rep- (Taktai) Chin, have continued this fine tra- tiles (39 terrestrial and 7 marine) are treat- dition of cooperation and have made sig- ed in the following section (Table 1). nificant contributions of their own (e.g., Three species are twentieth-century intro- Cassell et al., 1992; Fosberg et al., 1980; ductions, eight species are certain endem- Otobed and Maiava, 1994). ics, and ten others are probably endemic. The latter include several species repre- ZOOGEOGRAPHY sented by only one or a few specimens and Although it is premature to discuss Pa- their status cannot yet be determined. lauan zoogeography in any detail before Other species are known under existing the relationships of many species are un- names but for reasons discussed we be- derstood, summary papers on SW Pacific lieve are distinct species. For each species, biogeography continue to appear (Allison, the localities listed under Distribution are 1996; Brown, 1997; Holloway, 1997 among those islands in Palau where we have col- others), some of which are hampered by lected voucher specimens and/or verified the lack of recent data on Palau. A brief existing museum records. Almost all of the summary is justified. We are confident that island records derive from, or are verified analysis of the data in hand will confirm by, our field work through 1998. Appendix that Palau is an important zoogeographic I is a checklist of species by island. with faunal influences from all crossroads, AMPHIBIA: ANURA four major source areas. Our preliminary data do not confirm any species with un- Bufonidae equivocal Bornean relationships, but the Bufo marinus (Linnaeus, 1758) Oceanic (Gehyra oceanica, Lepidodactylus This large toad is very widely distribut- sp. nov., Perochirus sp.) and Philippine ed, both in its natural range of the Neo- (Lepidodactylus moestus, Mabuya sp.) af- tropics (Zug and Zug, 1979) and in many finities are more certain. Without ques- other parts of the world via introduction tion, the greatest influence on the Palauan (Easteal, 1981; Honegger, 1970). Details herpetofauna is the Greater Papuan region of exactly when and whence Bufo marinus (New Guinea + the Solomons and Moluc- was first introduced into Palau are un- cas, East Wallacea of Brown, 1997 and known (Easteal, 1981:102) but the first others). At least 13 Palauan species are de- museum specimens were collected in the rivatives of this area, and several of the 13 early 1950's. Although the NAMRU 2 or 14 others of uncertain origin will likely teams did not collect any Bufo during their be added to that total with additional re- stay on Beliliou in 1945, it is likely that search. toads were introduced during U.S. military Critically missing from the puzzle are activity in Palau in the post-war years, pos- 199911-991 HERPETOLOGICAL39--R L MONOGRAPHS- 39

TABLE 1.-Amphibians and reptiles of the Palau Is- TABLE 1.-Continued. lands. Those marked by an asterisk (*) have not been reported previously. Cheloniidae Chelonia mydas FROGS Eretmochelys imbricata Bufonidae Dermochelys coriacea Bufo marinus (introduced) Lepidochelys olivacea Ranidae CROCODILES Platymantis pelewensis (endemic) Crocodylidae LIZARDS Crocodylus porosus Agamidae Hypsilurus godeffroyi (status uncertain) Gekkonidae Gehyra sp.* sibly from Guam, New Guinea and/or Aus- Gehyra brevipalmata (endemic) tralia where has been well estab- Gehyra mutilata Bufo Gehyra oceanica lished since the 1930's. Gressitt (1952) Gekko sp. (endemic)* visited Kayangel Atoll in July-November Hemidactylus frenatus 1951 and the toads were well established Hemiphyllodactylus cf. H. typus* then. The of a nov.* by hypothesis post-war (i.e., Lepidodactylus sp. mid-1940's) introduction into Palau is fur- Lepidodactylus lugubris Lepidodactylus moestus ther supported by USNM 123909, collect- Lepidodactylus paurolepis (endemic) ed in July 1946 on nearby Yap. The field Nactus sp. (endemic)* notes of the collector state that, "On Yap Nactus cf. N. pelagicus* (the toad) does not reach the size it does Perochirus sp.* Polychrotidae on Guam and other islands ..." but that Anolis carolinensis (introduced) it was "very common" around Yaptown Scincidae (now Colonia) at that time. Fisher (1948) Carlia cf. C. fusca (introduced)* did not mention toads from Palau, but he Cryptoblepharus sp. (probable endemic)* Emoia atrocostata confirmed that they were common on Emoia caeruleocauda Ponape and all the islands of Yap during Emoia impar* his visits there, at some unspecified time Emoia jakati before 1948. Eugongylus sp. (endemic) Easteal found mu- Lamprolepis Although (1981:117) smaragdina seum for three in Pa- Lipinia leptosoma (endemic) vouchers only islands Lipinia cf. L. noctua* lau-Arakabesan (= Ngerekebesang), Mabuya sp. (endemic) Kayangel (= Ngcheangel Atoll), and Ker- Sphenomorphus sp. nov. (endemic)* rar (= Koror, now Oreor)-Bufo marinus Sphenomorphus scutatus (endemic) Varanidae is widely distributed in the archipelago. It Varanus cf. V indicus is most abundant on the causeway-con- SNAKES nected islands where agriculture (particu- Boidae larly taro patches) and other human activ- Candoia cf. C. carinata ities provide abundant breeding places. Colubridae The is not found on most of the Cerberus rynchops species Dendrelaphis sp. Rock Islands where standing water rapidly Hydrophiidae percolates through the steep, porous lime- Pelamis platurus stone, thus eliminating potential breeding Laticaudidae We have records from Laticauda colubrina places. sight Ulong in when the toads were Typhlopidae January 1995, Ramphotyphlops acuticaudus (endemic) quite common, but several subsequent vis- Ramphotyphlops braminus its to collect vouchers were unsuccessful. TURTLES Rat density was also startlingly reduced Chelidae and we initially suspected that a rat poi- Gen. & sp. indet.* also have Bataguridae (?) soning project might extirpated cf. Chinemys reevesii* the toads. An El Nino-associated drought in late 1997-early 1998 killed off much of 40 HERPETOLOGICALMONOGRAPHS [No. 13 the undergrowth in the disturbed area that the name corrugatus had priority over around the beach that it is unlikely any plicifera. He also commented, ". . . wheth- toads survived this exceedingly dry period er Platymantis plicifera pelewensis Peters, on Ulong. 1867, from the Pelew Islands, is a recog- Our one vouchered Rock Island record nizable race, scarcely concerns us here." is USNM 531973 from Pkulaklim, a high Brown (1952:51) and Inger (1954:351) island just off the Airai Dock area of concluded that the Philippine and New southeastern Babeldaob. The toad was ac- Guinea "corrugatus" were not only specif- tive by day in low undergrowth adjacent to ically distinct, but quite unrelated to one a low swampy area just behind a small another. Inger (1954:355-356) correctly beach on the west side. Taro had recently pointed out that both pelewensis and pa- been planted in the swamp and the toad(s) puensis were more similar to the Philip- may have been intentionally brought in pine Cornufer meyeri (= Platymantis dor- with the agriculture. salis, Brown and Inger, 1964) than to P. Distribution.-Babeldaob, Beliliou, Ma- corrugatus (see also Tyler, 1979). He re- lakal, Ngcheangel Atoll (Ngcheangel), garded the three taxa as distinct species Ngeaur, Ngerekebesang, Oreor, Pkulaklim. and provided some data on pelewensis (size, dorsal and ventral skin texture, dig- Ranidae ital and vocal sac morphology, ova dimen- Platymantis pelewensis Peters, 1867:33 sions). Wilhelm Peters (1867a:33) described Atoda (1950) published an excellent em- Platymantis pliciferus pelewensis from the bryological study of direct development in "Pelewinseln", based on a single specimen the Palauan frog but used no scientific collected by Dr. C. Semper. The holotype name, speculating that it was, "... prob- (ZMB5699) was listed as "not located" in ably a new species of Rana." Alcala (1961: Frost (1985:469) and Peters' extremely 721) referred Atoda's material to Cornufer brief description provided no morpholog- pelewensis (see Zweifel, 1969 for use of ical information, merely indicating that the the name Platymantis rather than Cornu- Palau specimen was nearly identical with fer). Gorham (1965:410-411) confirmed the Philippine Platymantis plicifera Guin- that pelewensis was a valid species and ther, differing only in the coloration of the provided the first illustrations of it (sketch- side of the head (all black in plicifera, up- es of the lateral head, ventral view of the per and lower lips flecked with white in hand and a photo of an entire specimen pelewensis). The name was overlooked in (1965:432, Figs. 3A,B)). Brown (1965:2-3) subsequent reviews of ranids (Boulenger, summarized the known species of "Corn- 1882, 1918), but it reappeared as a valid ufer", also listing pelewensis as a valid spe- species in the more recent literature (Gor- cies. The validity of the Palauan Platyman- ham, 1965; Frost, 1985). Although Ota and tis has remained unchallenged since Matsui (1995) and Kuramoto (1997) re- Brown and Tyler (1968). [Tyler's (1979:82, cently presented some data on the rela- Fig. 6) mention of one species of Rana in tionships of the species, no historical re- Palau is incorrect. This is either a lapsus view has ever been published. We will fill or perhaps a reference to Atoda's (1950) in some of the "lost years". erroneous remark that P pelewensis as "an Kampen (1923:190) transferred plicifera undescribed species of Rana", but Atoda's to the genus Rana (but maintained Platy- paper is not cited.] mantis as a subgenus), where the species Most recently, Ota and Matsui (1995) name became a homonym. He provided described the karyotype and Kuramoto the unnecessary replacement name Rana (1997) analysed morphological, karyologi- rugata and regarded pelewensis as a strict cal, and acoustic information on P pelew- synonym of it. Loveridge (1948:407) criti- ensis to clarify its relationships. All data cized Kampen's arrangement, regarded unequivocally confirmed that pelewensis is Philippine and New Guinea Platymantis as very closely related to P papuensis from subspecifically related, and pointed out New Guinea (e.g., Menzies, 1982a,b) and 1999] HERPETOLOGICAL MONOGRAPHS 41 quite distant from the Philippine P. dor- volved the ecology of female P. pelewensis. salis. We have collected several hundred Collections made from the forest floor more P. pelewensis than our Japanese col- (e.g., Kuramoto, 1997) are usually strongly leagues, material that includes call record- skewed towards males and subadults, but ings, skeletons, and frozen tissue, but we we found that females congregate, often in concur completely with their conclusions. incredible numbers, in the multitude of Our larger sample sizes negate the size dif- natural caves and Japanese-made bunkers ferences between pelewensis and papuen- that are omnipresent in Palau. These bun- sis noted by Kuramoto (1997) and greatly kers usually retain moisture and many expand the limited pattern polymorphism have an abundance of trash on the floor or he reported. crevices in the wall, so we initially sus- Platymantis pelewensis is ubiquitous in pected they may have been an important Palau, at least from Kayangel to Ngeaur, egg deposition area. However, after thor- but Owen (1977) stated that frogs were oughly searching dozens of bunkers (some not found in the Southwest Islands. This repeatedly throughout much of the year) species is abundant in backyards in Koror we have yet to find any indication that re- town and in undisturbed forest on the production occurs there. The bunker frogs Rock Islands. Although Kuramoto (1997: are 90+% female, many of them larger 183) characterized the Palau frog as being than maximum previously reported size, "found among leaf litter on the forest and we initially suspected that they might floor," it is most assuredly not a forest ob- be a second species. However, the few ligate. In fact, based on call records it may males found in samples from more than be most abundant in patches of thick grass 12 "caves" were all immature; calling in disturbed areas, where dozens of males males have been heard near the mouths of can be heard vocalizing among the impen- the caves (but never inside) and the calls etrable, moisture-retaining roots. Call rec- were all typical pelewensis. Apparently ords also confirm the presence of frogs on large females use the bunkers as diurnal many of the Rock Islands, even very small retreats and venture into the forest at pinnacles. We suspect P pelewensis may night for reproduction, presumably at the be salt tolerant as well, since we have male calling site. The bunkers remain con- heard and collected the species at the edg- sistently moist even during the driest es of mangrove swamps and other haline spells, and provide an abundance of food areas. (cave crickets are particularly abundant). Details on the natural history of the Pa- On several occasions we have also found lau frog have been minimal since Atoda's juvenile P. pelewensis in the guts of large (1950) description of the development of females from bunkers. the clutch he discovered. Unfortunately, Distribution.-ENDEMIC. Babeldaob, he did not provide information on where Bailechesengel (Ngemlis Islands; call the eggs were found, if there was an at- only), Beliliou, Malakal, Ngcheangel Atoll tendant frog, or other important details on (Ngcheangel, Ngeriungs), Ngeaur, Nger- direct-developing species. No other clutch cheu, Ngerechur, Ngerekebesang, Ngeru- has ever been found, despite our intense keuid (island 24), Ngeruktabel, Ngetme- field efforts. We suspect, however, that duch, Oreor, Ulebsechel, Ulong. males call from and defend a calling/nest chamber and attend the clutch. We base REPTILIA: SAURIA this on the fact that calling males are often Agamidae nearly impossible to find, particularly dur- Peters, 1867 ing dry periods. On rainy evenings males Hypsilurus godeffroyi may call from exposed sites but in general Originally described from two speci- they are well concealed and several have mens as Lophura (Hypsilurus) Godeffroyi been collected from calling chambers, un- from the "Pelew-Inseln" (Peters 1867b: der detritus but not subterranean. 707, Fig. 1; lectotype ZMB 5891), this spe- One of our more startling discoveries in- cies has never been collected again any- 42 HERPETOLOGICALMONOGRAPHS [No. 13 where in Oceania. Boulenger (1885:295- without a specific locality or collector. The 296) first associated the species name with Germans were also active in the New populations from the Papuan region. In Guinea-Solomons area during this period fact, he did not even include Palau in the and the specimens could have been mis- distribution of the species, listing only labelled, or Palau may have been where "New Ireland, New Britain, Solomon and the specimens were shipped to Europe Fiji Islands, Northern Queensland". Cog- and the shipping port was erroneously in- ger (1975:222) discounted the Australian terpreted as a collecting locality. However, records (and deleted the species from sub- this scenario seems less plausible in view sequent editions of his book) and listed the of the bones collected from Ngeaur. distribution as "Widely distributed . . . Conceivably the species is a recent (i.e., from the Solomon Islands to the Philip- post-European) extinction. It may have pines." However, the species (and the ge- been exterminated by rat predation, per- nus) has never actually been reported from haps coupled with habitat loss in the wake the Philippines and Cogger (personal com- of deforestation, particularly during the munication 1997) confirms that "Philip- Japanese period. pines" was probably a lapsus for Palau. Distribution. -unknown (Ngeaur?), Zug (1991:7-8) regarded the Fijian locality probably endemic. as erroneous. The status of this lizard in Palau is prob- Gekkonidae lematic. Palauans are utterly unfamiliar Gehyra sp. with photographs of this large and distinc- On first inspection, the large Palauan tive species. However, McCoy (1980:14) Gehyra seem simple taxonomically, i.e., emphasized that, in the Solomons at least, the dark, velvety G. brevipalmata with di- H. godeffroyi can be amazingly cryptic, vided terminal lamellae and yellow or or- even where it is common. The only recent ange ventral color, versus the pale, ghostly evidence of the lizard's presence in Palau gray G. oceanica with transverse, undivid- comes from a collection of skeletal re- ed lamellae and a gray (occasionally yel- mains found in a cave(s)/gravel pit (now low) belly. The reality is considerably more flooded) on Ngeaur by Tom Iliffe in complex. Although the two species are March, 1985, and sent to us for examina- usually easily identifiable in the field, even tion. The bones are clearly those of an in areas of absolute syntopy (Ngaraard agamid, and represent a minimum of 11 State, Babeldaob), some small island pop- individuals ranging in approximate snout- ulations are more equivocal. One of us vent length of 70-110 mm; McCoy gives (RIC) misidentified the Ngerur population an average SVL of 120 mm for Solomon of oceanica as brevipalmata on his first vis- Island godeffroyi. Iliffe recovered the it, despite considerable experience with bones along with mollusk shells while sam- both species in the field. pling for archaeological artifacts. The Judging from the number of misidenti- bones show no evidence of deterioration, fied specimens in museum collections, and appear much like macerated skele- preserved material is more confusing still. tons, suggesting that they were deposited The terminal lamellae can appear divided not too long before they were collected. in poorly preserved or dessicated speci- We think it probable that this species still mens of G. oceanica. This apparently exists in Palau and it awaits rediscovery in caused the misidentification of Saipan some forested area. If and when additional oceanica as Peropus papuanus in Dryden material from Palau is collected, we be- and Taylor (1969). lieve it equally probable that the lizard will Although we are reasonably confident prove not to be conspecific with the pop- that the Gehyra with transverse lamellae ulations in the Solomons and New Guinea. from the core islands of Palau are all Other possibilities are that the type lo- oceanica (sensu lato), we are much less cality is erroneous. Peters' types were re- sure about the specimens from the South- ceived from the Godeffroy collection, west Islands. We have no experience in 1999] HERPETOLOGICAL MONOGRAPHS 43 these islands, so no ecological or color data den and Taylor (1969) reported Peropus are available. The few specimens known papuanus (= Gehyra papuana) from Palau (USNM 348770 from Fana; 348762-63 based on one specimen from Ngeaur from Pulo Anna; 348781-82 from Sonso- (EHT-HMS = FMNH 6699), citing Bron- rol; 348788 from Tobi) have unquestion- gersma's (1934) notes on the holotype of ably undivided lamellae, but somewhat G. papuana as justification for their iden- more emarginate than is typical for ocean- tification. Brongersma (1934:174) stated ica (this may be an artifact of preservation that only the basal and terminal lamellae as noted above). The males also have lon- were entire, the rest divided, and that the ger preanal-femoral pore series, extending tail was, "... much depressed ... with a to mid thigh (restricted to the preanal area sharp lateral edge." We have reexamined in Palauan oceanica). King and Horner Dryden and Taylor's specimen, which is (1989) described south-coast Papua New dessicated and tailless so Brongersma's Guinea "oceanica" as Gehyra membrana- characters are difficult or impossible to ap- cruralis but the Southwest Islands speci- ply (see also comments under Gehyra sp.). mens lack the extensive femoral mem- However, several of the proximal lamellae brane of that species. Perhaps the most (not just the basal ones) appear to be un- pertinent comparisons are with Gehyra divided, similar to the digital structure of from the Moluccas (the most proximate G. brevipalmata. King and Horner (1989) landmass to the Southwest Islands), characterized G. papuana as a medium- whence Rooij (1915) records both G. sized species (less than 80 mm SVL) re- oceanica and marginata. Gehyra margin- stricted to New Guinea. We have collected ata is a large species with undivided (but extensively on Ngeaur, finding only G. slightly emarginate) digital lamellae, well brevipalmata and G. mutilata. We are con- distinguished from oceanica by the exten- fident that Dryden and Taylor's specimen sive skin folds along the thighs and the is a misidentified brevipalmata (a species venterolateral body. We have seen no ac- they did not mention), as suggested by tual Gehyra oceanica from the Moluccas Bauer (1994:100). but a series (USNM 237578-83) of ocean- Gehyra brevipalmata is likely one of the ica-sized Gehyra from Halmahera has "old" Palauan species. It occurs widely strongly divided digital lamellae, unlike throughout the archipelago including for- any species previously reported from the est on Rock Islands to buildings in dis- Moluccas and equally distinct from the turbed areas. Its rich, dark, velvety skin is Southwest Islands taxon. remarkably fragile and undamaged speci- Clearly the Gehyra "oceanica" from the mens are rare. When grasped, G. brevi- southwest Pacific are in need of revision. palmata spins and any restraint causes the Rather than refer the Southwest Islands skin to peel back from the subdermal fas- material to this catch-all species, we prefer cia and a completely skinless specimen can to call it Gehyra sp. to emphasize our im- result. pressions of its distinctiveness from Gehyra brevipalmata is particularly "oceanica" on the core islands of Palau. fond of crevices and the species is found Distribution.-Fana, Pulo Anna, Son- commonly on banyan trees, in Pandanus sorol, Tobi. axils, under bamboo sheaths, and in man- made crevices around buildings (under Gehyra brevipalmata (Peters, 1874:159) shutters, behind drain pipes, in corrugated The relationships of this distinctive Pa- tin roofs). In captivity, the species eats lauan endemic [Hemidactylus (Peropus) fruit readily and in the wild we have seen brevipalmata Peters, 1874; Holotype it licking at sweets (tree sap, open soft ZMB7927] are not clear, but we suspect it drink cans and bottles). Crombie and may be close to some of the poorly known Steadman (1987) reported that the stom- New Guinea species with divided lamellae achs of several Gehyra oceanica from the (e.g., papuana, lampei, interstitialis, bal- Cook Islands were "packed with small iola (Rooij, 1915:38; Bauer, 1994)). Dry- seeds" similar to fig, and fruit eating has 44 HERPETOLOGICAL MONOGRAPHS rNo.[No. 13

FIG. 5.-The distribution of Gehyra mutilata in Oceania. (Also occurs in India, Sri Lanka, Indochina, Philippines, Japan, Madacasgar; after Bauer, 1994). been reported for several other species of later described as Gehyra brevipalmata Gehyra (see Burnett and Nolen, 1996). (Peters, 1874). Distribution.-ENDEMIC. Babeldaob, This species is ubiquitous in Palau and Bablomekang, Beliliou, Euidelchol, liblau is found in a wide variety of habitats. In (Ngemlis Islands) Kmekumer Islands, Ma- the absence of Hemidactylus frenatus it is lakal, Ngeaur, Ngercheu, Ngerduais, Nger- often the commonest house lizard, but it ekebesang, Ngeruktabel, Oreor, Pkulak- is also found in more natural habitats in- lim, Ulebsechel, Ulong. cluding undisturbed forest. Distribution.-Babeldaob, Ioulome- Gehyra mutilata (Wiegmann, 1834) kang, Malakal, Ngcheangel Atoll (Ngche- This gecko is very widely distributed in angel), Ngeaur, Ngerechong, Ngereke- mainland south and southeast Asia, and besang, Ngetmeduch, Oreor, Tobi numerous island groups including Mada- (Southwest Islands). gascar and other Indian Ocean groups, the Indoaustralian Archipelago, the Philippi- Gehyra oceanica (Lesson, 1826) nes, and most of Oceania (Fig. 5; Bauer, Although King and Horner (1989) 1994:92, map 24). Its broad distribution is found Oceanic populations (Fig. 6) of Ge- usually attributed to human transport. hyra oceanica to be homogeneous both Some populations (the Neotropics, New morphologically and karyologically, Beck- Zealand) are clearly post-European intro- on (1992) was able to distinguish morpho- ductions, but the species probably dis- logically Micronesian from Polynesian persed easily with indigenous peoples oceanica. Biochemical data confirm the through much of Oceania. differences between the two populations Peters (1867a) discussed a "variety" of (R. Fisher, 1997), although no one has yet Gehyra mutilata based on a single 60 mm attempted to sort out the confusing syn- specimen collected in the "Pelew-Inseln" onymy of oceanica to determine which by C. Semper which he distinguished from names apply to the two species. typical mutilata by the more extensive dig- Gehyra oceanica has been recorded ital webbing and the small outer postmen- from Oreor (CAS-SU 23747, 23753, CAS tal scales. This was a subadult of what he 122393), but we have collected it only in 199911999] HERPETOLOGICAL MONOGRAPHS 45

FIG. 6.-The distribution of Gehyra oceanica (after Bauer, 1994). the extreme north of Babeldaob and on Basically, Gekko vittatus (in New Guinea peripheral small islands (e.g., Ngerur). We and the Solomons at least, data for the In- have not verified its presence on Oreor de- donesian populations are almost totally spite intensive effort. In Nagaraard State lacking) is dimorphic in color pattern, but (northern Babeldaob) it occurs syntopical- not sexually correlated (McCoy, 1980: ly (often on adjacent trees) with Gehyra plate 7a,b illustrates both forms). The brevipalmata in Ulimang Village, but most common morph has a vivid light ver- brevipalmata seems to be slightly more tebral stripe outlined in black, and a dis- abundant. tinctly banded tail. A more unicolor, ob- Distribution.-Babeldaob, Chemoi, scurely striped lizard is often called Gekko Ngcheangel Atoll (all islands), Ngeanges, vittatus bivittatus (e.g., Boulenger, 1885), Ngerechur, Ngerur, Oreor. but Mertens (1926) pointed out that both morphs occurred at several places in the Gekko sp. New Guinea region, and he referred to This large, attenuate gekkonid is an un- bivittatus as a "Mutation" rather than a described endemic species. It is related to recognizable taxon. Gekko vittatus of New Guinea and the In- Probably because it is obscurely striped, doaustralian Archipelago, and it is fre- the Palau Gekko has occasionally been re- quently identified as vittatus in the litera- ferred to G. vittatus bivittatus (Boulenger, ture and in museum collections. New 1885; Dryden and Taylor, 1969). However, Guinea/Indonesian Gekko vittatus are ex- there are significant meristic and molecu- tremely variable and more than one spe- lar (C. Austin; H. Ota; R. Fisher, personal cies is certainly included in vittatus as cur- communication) differences between the rently defined. The nomenclatural history unstriped Gekko from Palau and those of vittatus and its synonyms is plagued by from the New Guinea region. erroneous type localities, missing type Although common in Palau, Gekko sp. specimens, and other problems that pre- is secretive and easily overlooked. It does clude resolution with currently available inhabit houses, even in Koror town itself, data and specimens. A brief summary as it but it usually haunts the darkest areas, relates to the Palau Gekko is warranted. rather than foraging around lights as many 46 HERPETOLOGICAL MONOGRAPHS [No. 13

FIG. 7.-The distribution of Hemidactylus frenatus in Oceania (after Bauer, 1994). of the smaller species of gekkonids do. experience with Papuan G. vittatus, which More commonly it occurs on tree trunks frequently are found in communal groups and leaves, with little preference other in the field and live well together in cap- than vegetation that provides crevices for tivity. diurnal retreats. Palms, Pandanus, ban- Distribution.-ENDEMIC. Babeldaob, yans, and banana (wild or cultivated) are Bablomekang, Beliliou, Kmekumer Islands favored. Adults are found after dark on (island 46), Malakal, Ngeaur, Ngercheu, leaves, fronds, trunks or branches, appar- Ngerduais, Ngerechur, Ngerekebesang, ently awaiting insect or vertebrate prey. Ngerukeuid Islands (island 24), Oreor, Eggs are adherent, laid in crevices, and the Ulebsechel, Ulong. clutch size is two. Palm, Pandanus, or ba- nana axils are common deposition sites but Hemidactylus frenatus Dumeril and cracks in masonry and other man-made Bibron, 1836 sites are also used. Incubation time is This widespread Southeast Asian weed about 90 days and hatchlings are as ob- species, a quintessential "house lizard" scurely striped as adults. (sensu Mertens 1928), is apparently a re- We suspect that Gekko sp. is a major cent introduction to Palau (Fig. 7). None nocturnal predator in Palau, feeding on ar- of the earlier collections included this thropod prey and small vertebrates. Their highly visible human commensal, and the jaws are quite powerful, capable of inflict- earliest record is a single specimen (CAS- ing a painful bite on human fingers often SU 23728) collected near the Conserva- accompanied by a twisting motion that tion/Entomology Lab on Koror by Fehl- causes bloody lacerations. Subduing even mann in July 1955. Dryden and Taylor sizable vertebrate prey would be well with- (1969) next reported it from Ngeaur. in their abilities. Captive individuals are When introduced into established com- highly territorial and aggressive; males will munities of other gekkonids (native or in- brutalize non-reproductive females and troduced) H. frenatus is often successful in cannibalize juveniles. Fruit is also readily displacing other species from disturbed eaten in captivity. These behavioral char- (lighted) habitats (Case et al., 1994; acteristics are strikingly different from our McCoid, 1996; Petren et al., 1993; Petren 1999]199 HERPETOLOGICALEPTLGCA MONOGRAPHSOORPS4 47

FIG. 8.-The distribution of Hemiphyllodactylus typus in Oceania. (Also occurs in the Mascarenes, through- out southeast Asia including India, Indonesia, and Philippines; after Bauer, 1994.) and Case, 1996). In Palau H. frenatus has several fleeting glimpses of him in shad- not completely displaced Lepidodactylus owy parts of the house. Although the and Gehyra from buildings. On the Na- house was largely constructed from local tional Museum building in Koror, Gehyra material, some lumber was brought in mutilata and two species of Lepidodacty- from Koror and we suspect this one male lus were collected in the mid-1990's, al- came with it. The other population around though H. frenatus was by far the most the church has not dispersed far enough common lizard. Perhaps even more in- to provide him a mate. structive is the situation in Ulimang Vil- Distribution.-Babeldaob, Beliliou, Ma- lage, Ngaraard State, in a section of north- lakal, Ngeaur, Ngercheu, Ngerekebesang, ern Babeldaob that is not yet connected Ngermalk, Oreor, Uchulangas. by road to the southern causeway-con- nected islands. During the time of our Hemiphyllodactylus cf. H. typus Bleeker, field work, Hemidactylus frenatus was al- 1860 most completely confined to the Evangel- This cryptic, easily-overlooked small ical Church, where it was slightly more species was previously unreported from common than Gehyra mutilata around the Palau other than Bauer's (1994:131, map light at night. Despite our extensive field 29) inclusion of Palau in the distribution work in and around this community, we of H. typus (Fig. 8). The only other Mi- found only one other individual of H. fre- cronesian record (Lamberson, 1987 from natus, a single male that vocalized fre- Enewetak, Marshall Islands; also mapped quently in the rafters of a recently con- in Bauer, 1994) is based on a misidentifi- = structed house about one km south of the cation (USNM 205534-35 Lepidodac- village. Although we often heard this in- tylus sp.). Although all other Oceanic pop- dividual over a span of several years, we ulations are all-female, males are present never saw it in the characteristic frenatus in Palau. This complex of species is under foraging area around the lights of the study by G. R. Zug and we defer any tax- house, which were occupied exclusively by onomic conclusions pending completion of Lepidodactylus moestus and occasionally his analysis. Gehyra brevipalmata. We finally caught Hemiphyllodactylus is not a "house liz- 48 HERPETOLOGICALMONOGRAPHS [No. 13 ard" (sensu Mertens, 1928), but it has al, Ngeanges, Ngeaur, Ngercheu, Ngerdis, been found near houses in more sparsely Ngerekebesang, Ngermalk, Ngerur, Nges- populated areas. When active after dark, bokel, Ngetmeduch, Oreor, Pulo Anna, the species has been collected on attenu- Uchulangas. ate leaves of shrubs (including palm fronds), on small and large trees, and on Lepidodactylus moestus (Peters, 1867:13) debris around houses. We also collected it Long mired in the synonymy of the in forest primary (Ngeruktabel). catch-all Lepidodactylus lugubris, this spe- Distribution.-Babeldaob, Ngeanges, cies (Gecko moestus Peters, 1867; Or- Ngeaur, Ngerekebesang, Ngeruktabel, ZMB5610) was recently resurrected by eor, Ulebsechel. Ota et al. (1995). It is undoubtedly a native in it also in Mi- sp. species Palau, but occurs Lepidodactylus cronesia as far east as Arno Atoll in the Bisexual, of the possibly parental species Marshalls (Fig. 9). Lepidodactylus moestus widespread parthenospecies Lepidodacty- is distributed in Palau includ- are known from very widely lus lugubris Micronesia. both disturbed and natural areas. It One bisexual ing distinctively patterned spe- does well even in downtown Koror, cies et 1995: is quite (Radtkey al., plate 1) on the abun- known from in the Tuamotus surviving buildings despite Takapoto dance of It also co- and Arno Atoll in the but it has Hemidactylusfrenatus. Marshalls, exists with L. and the two never been from Palau. We col- lugubris species reported are almost common on the lected one male equally picnic (USNM 531970, pre- on the Museum In se- served with everted low on a pavillion grounds. hemipenes) disturbed areas Klouklubed trunk at the tide line on verely (e.g., palm high Ulong on be more that is similar to the illustration in village Beliliou), lugubris may Radtkey abundant but we have found a few et al. (1995). However, two other only speci- instances where this is so. mens (USNM collected sub- Accumulating 531971-72) evidence et demonstrate to us that it is not (e.g., Radtkey al., 1995, 1996) sequently that moestus is one of the conspecific with the Takapoto/Amo popu- strongly suggests lations. sexual species that gave rise to the parthe- moes- Distribution.-Iilblau (Ngemlis Is- nogenetic lugubris. Lepidodactylus tus is an and vocal lands), Ngeruktabel, Ulong. adaptable, aggressive, species that can survive in some surpris- Lepidodactylus lugubris (Dumeril and ingly harsh environments. We have col- Bibron, 1836) lected the species on the smallest (un- of This widespread, all-female, partheno- named) island at the mouth genetic species has been intensively stud- Ngeremeduu Bay (Ngatpang State, west- ied (Bolger and Case, 1994; Ineich, 1988; ern Babeldaob), a pinnacle barely 4 m2 Ineich and Ota, 1992; Radkey et al., 1995, that supported one Pandanus tree. One 1996) and several distinct species are cer- adult moestus (USNM531969) and a tainly included under this name. The spe- clutch of viable eggs were found in the cies seems to be a relatively recent arrival Pandanus axil. in Palau, regardless of the name applied to Radtkey et al. (1995) illustrated the two it. It occupies the most disturbed areas, distinct morphs of moestus: plain and often syntopic with L. moestus, with which striped. Robert Fisher (personal commu- it may hybridize, producing puzzlingly in- nication) insists that these morphs are ge- termediate specimens (Ota et al., 1995). netically identical. As noted in the account of Hemidacty- Distribution.-Babeldaob, Kmekumer lusfrenatus, this species continues to share Islands (island 46), Malakal, Ngcheangel the building habitat with frenatus in many Atoll (all islands), Neanges, Ngeaur, Nger- areas of Palau. cheu, Ngerduais, Ngerechur, Ngerekebes- Distribution.-Babeldaob, Beliliou, ang, Ngermalk, Ngerukeuid Islands (island loulomekang, Kmekumer Islands, Malak- 4), Ngeruktabel, Ngerur, Ngesbokel, Nget- 1999] HERPETOLOGICAL MONOGRAPHS 49

FIG. 9.-The distribution of Lepidodactyus moestus.

meduch, Oreor, Pulo Anna, Uchulamera- beach habitats, this species is very cryptic del, Uchulangas. and secretive. The adult male was found on a tree in scrubby forest just behind the Lepidodactylus paurolepis Ota et al., beach, and adjacent to a limestone cliff. 1995 No other individuals were seen in 4 hours This recently described species remains of collecting that evening, and ca. 30 hours known only from the small type series col- (day and night) subsequently. The two lected in the Ngerukeuid Islands. Al- hatchlings from Ulebsechel were found though it is a very distinctive gecko, it beneath ground litter, one near the beach would be most surprising if it was indeed and the other at the base of a limestone restricted to this isolated cluster of small ridge inland. islands. We suspect it will be found on oth- Resolution of the systematics of the er islands in the Palau lagoon. Nactus pelagicus complex is still in the Distribution.-ENDEMIC. Ngeru- preliminary stages (see summary in Bauer, keuid Islands (islands 16, 34) 1994:168; Zug, 1998; Zug and Moon, 1995), and this species appears to be un- Nactus sp. described (G. R. Zug, personal communi- Prior to our field work the apparent ab- cation). More specimens and data are sence of Nactus from Palau was enigmatic needed to confirm the status and distri- because the genus is widespread in the bution of the species in Palau. Papuan region and southwest Oceania (see Distribution.-Ulebsechel, Ulong. map 40 in Bauer 1994:169). This absence was simply an artifact of inadequate col- Nactus cf. N. pelagicus (Girard, 1858) lecting. Two species are present in the ar- A single specimen (USNM 348780) chipelago (see also Nactus cf. N. pelagi- from Merir in the Southwest Island Group cus). was referred to N. cf. pelagicus by G. R. Nactus sp. is a bisexual gecko that we Zug (personal communication). The pelag- found only on Ulong (one adult male) and icus group (Fig. 10) is a complex of bisex- Ulebsechel (two hatchlings), but we sus- ual and unisexual, parthenogenetic species pect that its distribution is more extensive. (see summary in Bauer, 1994:169) but the Unlike members of the Nactus pelagicus alpha taxonomy is not yet stabilized. Zug complex, which are often abundant in and Moon (1995) recently resurrected N. 50 HERPETOLOGICAL MONOGRAPHS [No. 13

FIG. 10.-The distribution of the Nactus pelagicus species group (after Bauer, 1994). multicarinatus for the bisexual species of sented by only a few older specimens from Vanuatu and the southern Solomons, but Kapingamarangi Atoll, located more than the situation on New Guinea itself is much 2000 km SSE of Palau. In the summer of more complex. Larger sample sizes from 1996, Donald W. Buden collected a large the Southwest Islands are needed to de- series of scutellatus from Kapingamarangi termine if they are indeed conspecific with (Buden, 1998a,b). We have not yet had the the unisexual pelagicus. The Merir speci- opportunity to examine these specimens in men is a female. detail, and the limited material from Palau Distribution.-Merir. may be inadequate to demonstrate con- specificity in any case. If the Palau speci- Perochirus sp. mens are scutellatus the enormous distri- The genus Perochirus is a Pacific insular butional hiatus could be explained as the endemic, consisting of three species from relictual distribution of an old radiation scattered island groups in Micronesia and created by local extinctions on the inter- Vanuatu (Brown, 1976), and an extinct vening islands. form from 'Eua, Tonga (Pregill, 1993). The Both specimens of Perochirus from most widespread species, Perochirus ate- Fana were collected on trees. Interesting- les, occurs close to Palau in the Carolines, ly, no Gekko were collected from any of Marianas, and Marshalls (Bauer, 1994: the Southwest Islands. Gekko is normally 208, map 49). Although Bauer (1994:283) a conspicuous species where it occurs, and recorded Perochirus ateles from Palau Perochirus may replace it ecologically in with a question mark, there are no docu- the Southwest Islands. mented records of this species from Palau. Distribution.-Fana. A recent collection from the Southwest Islands contains two specimens (USNM Polychrotidae Anolis carolinensis 1832 348771-72) of a large Perochirus from Voigt, Fana. Their proportions and scalation in- This species has been introduced in sev- vite comparison with Perochirus scutella- eral areas of the Pacific, but historical in- tus. Until recently, Perochirus scutellatus formation on the introductions is lacking. was a very poorly known species repre- Some authors have suggested that the 1999] HERPETOLOGICALMONOGRAPHS 5'51

Oceanic Anolis are Cuban in origin (Anolis Lamprolepis and Anolis, which involved a porcatus or Anolis carolinensis porcatus; subadult Anolis we were pursuing in a Oliver and Shaw, 1953; Vance, 1987) but viney tangle near the Port. The anole at- we agree with McKeown (1996) that Pa- tempted to escape by leaping to the cific populations (including Palau) are typ- ground but had barely taken a few steps ical of A. carolinensis of the southeastern before an adult male Lamprolepis dashed United States. down the trunk of a nearby sapling and Although Anolis carolinensis has been in seized it. Palau for at least 30 years (Owen, 1977), Distribution.-Malakal, Oreor. it has not dispersed far from the presumed place of entry, the Port of Palau on Ma- Scincidae lakal. Robert Owen (personal communi- Carlia cf. C. fusca (Gray, 1845) cation to RIC in 1980) said that the lizards Despite its abundance in most disturbed were restricted to the immediate vicinity areas in Palau, Carlia seems to be a very of the Port and adjacent docks at that recent introduction to the islands. The or- time, and their distribution has changed igin and taxonomic status of the propagule little since then. Our field work 15 years remains unclear. Carlia has never been re- later indicates that the species is now corded in the literature from Palau and the found 50-100 meters east and west of the earliest known specimens were collected Port along the road, but it is not abundant. by Tom Fritts in 1988, but we are confi- No obvious dispersal barriers confront this dent that the species was introduced be- species on Malakal, and there is abundant fore then. disturbed habitat all over Malakal and ad- Both the Marianas and Palauan popu- jacent Oreor. Our herpetological surveys lations of Carlia are clearly members of and transects have covered Malakal from the widespread and taxonomically unre- the top of the hill to sea level in several solved fusca group, and the logical as- places from 1993-1997, but yielded no sumption would be that the two areas specimens or even sight records of Anolis were synchronously colonized from the from anywhere other than the narrow same source. However, females and juve- coastal area on the west side of the island. niles in Palau often have strong dorsolat- Then in July 1996, Chris Austin and RIC eral striping, which is rare or nonexistent collected an adult male in thick, scrubby in the Marianas population. Our impres- vegetation in the agricultural area on "Ra- sion is that Palau and Marianas Carlia had dio Tower Hill", just below the band of different origins and may be different spe- forest around the peak, a site we had sam- cies, neither of which is true 'fusca". pled frequently before. The species was Carlia is the dominant terrestrial skink also abundant around the newly construct- on the causeway-connected islands of Pa- ed Coral Reef Research Foundation build- lau, occurring in very high densities in all ing at the Malakal end of the causeway the disturbed areas of southern Babel- across to Koror, and in August 1998 RIC daob, Oreor, and the associated smaller is- found the anole on the edge of Koror town lands, but they are not (yet) on any of the itself. After a long sedentary period, Anolis adjacent Rock Islands. The more distant may finally be dispersing in Palau; it will islands of Ngeaur and Beliliou have be interesting to watch its progress. healthy populations of Carlia, but here We suspect that Lamprolepis smarag- also the lizards do not penetrate forested dina may be somewhat of a controlling areas and are restricted to open grassland agent in the spread of Anolis carolinensis. or the vicinity of human habitation. The The two species often occupy the same parts of northern Babeldaob not connect- habitat, and Palauan A. carolinensis is un- ed by road to the south may be Carlia-free characteristically secretive compared to (at least at the moment). Our extensive ex- other natural and introduced anole popu- perience in the Ulimang-Ngesang area lations we have seen. We have only one yielded a few "possible" Carlia sightings in direct observation of interactions between 1994 and again in 1997, but no vouchers. 52 HERPETOLOGICALMONOGRAPHS [No. 13

Juvenile Mabuya are superficially similar Cryptoblepharus may be present in Palau to striped female and juvenile Carlia and but presently available material is inade- difficult to distinguish when they are on quate for a determination. the move. Extensive sticky-trapping (40+ Distribution.-Babeldaob, Beliliou, Iil- trap-hours) in and around Ulimang village blau (Ngemlis Islands), Kmekumer Islands in August 1998 yielded many Mabuya but (island 46), Ngeaur, Ngebad, Ngerekebes- no Carlia. ang, Ngetmeduch, Oreor. Distribution.-Babeldaob, Beliliou, Ma- lakal, Ngeaur, Ngerekebesang, Ngermalk, Emoia atrocostata (Lesson, 1826) Ngetmeduch, Oreor, Uchulangas. This variable "superspecies" (sensu Brown, 1991) occurs from Taiwan and the Cryptoblepharus sp. Ryukyus to New Guinea, the Moluccas, Mertens (1931, 1933, 1934b, 1964) re- and the southern Malay Peninsula includ- garded Ablepharus (= Cryptoblepharus) ing Christmas Island in the Indian Ocean, boutonii as a polytypic species with more and western Micronesia (Fig. 11). Brown than 40 subspecies ranging from the In- (1991) recognized subspecies from Cape dian Ocean coast of Africa to the eastern York, Australia (australis) and the Solo- limits of the Pacific. Crombie and Stead- mons/Vanuatu area (freycineti). We agree man (1987) cited accumulating evidence with Brown that several taxa are probably that many of these "subspecies" are valid included in his nominate atrocostata, and species, and they suggested that the name that the Micronesian populations are rea- Cryptoblepharus poecilopleurus was appli- sonably distinctive. cable to most of the Oceanic populations. Throughout its broad distribution, Fiji, Vanuatu, and New Caledonia all have Emoia atrocostata invariably inhabits rocky distinct species of Cryptoblepharus (Sad- beaches and mangrove swamps as report- lier, 1986; Zug, 1991 and personal com- ed by Alcala and Brown (1967) in the Phil- munication) but the other Pacific Rim ippines. Some populations are associated populations have not yet been studied in with one habitat over the other, but it is detail. Peters (1879:37) described the Pa- currently unclear if this is merely sampling lauan Cryptoblepharus as Ablepharus ru- bias or an ecological preference with sys- tilus (holotype ZMB 7926) and we suspect tematic implications. that it may be a valid species. Both George In Palau, Emoia atrocostata seems to Zug and Paul Horner are examining vari- have a spotty distribution but we are con- ation in Cryptoblepharus and we have fident that this reflects the difficulty of col- made our material available to them to be lecting individuals amid the multitude of analysed and compared with broader geo- refuges on rocky beaches and in mangrove graphic samples. swamps, rather than actual rarity or a re- In many parts of its range Cryptoble- stricted distribution. For example, we have pharus is a beach species, occurring on seen atrocostata commonly on beach boul- rocks and wrack in the supratidal zone ders near the KB Bridge on Oreor but we (e.g., McCoid et al., 1995). Palauan Cryp- have not yet been successful in collecting toblepharus are much more arboreal, of- them there. In other cases something ten found well away from the beach on more complex is involved. Near the north- tree trunks, vines and lianas (e.g., USNM em tip of Babeldaob in Ngaraard State, 531953-56 from central Oreor). The spe- the species is abundant and easily ob- cies does occur as well in beach situations, served in mangrove swamp near the dock frequently on large trees hiding behind ex- west of Ngebuked on the west side of the foliating bark. On northern Babeldaob peninsula. However, in similar swamps more strongly striped (poecilopleurus-like) south of Ulimang on the east side (less Cryptoblepharus were seen, but not col- than 3 km airline) not a single atrocostata lected, on aerial roots of mangroves. This has been seen in more than 40 person/ particular locality has since been destroyed hours of searching. The species does occur by dredging. More than one species of on the east side of Babeldaob. A single 1999]199 HERPETOLOGICALHEPEOLGIA MONOGRAPHSMNORAHS5 53

FIG. 11.-The distribution of Emoia atrocostata. subadult (USNM 531952) was collected males in Palau have a profusion of tur- under beach wrack at the high tide line quoise or yellow lateral flecks that are lack- directly in front of our base of operations ing in other populations of caeruleocauda at Ulimang in February 1996. Six other in- that we are familiar with. We feel that dividuals (USNM 495091-95) also were Brown's earlier impression was correct, obtained at Melekeok by Gordon Rodda and that additional analysis will demon- and Tom Fritts. Although usually charac- strate that Palauan E. caeruleocauda are terized as a coastal species, Emoia atro- indeed specifically distinct. costata can extend well inland along riv- Although this species is easily displaced erine mangroves, as documented by by Carlia in disturbed habitats, it persists USNM 495090 from the Tabecheding Riv- in backyard garden plots and other shaded, er at Nekkeng, Aimeliik State, Babeldaob. moist areas even in Koror town itself. The Distribution.-Babeldaob, Beliliou, Ma- lizard is primarily terrestrial, but juveniles lakal, Ngeaur, Ngebad, Ngerdis, Ngerek- and subadults in particular frequently as- ebesang, Ngetmeduch, Oreor, Uchulangas, cend low vegetation. Ulebsechel. Distribution.-Babeldaob, Beliliou, Eu- idelchol, Iilblau (Ngemlis Islands) loulo- Emoia caeruleocauda (De Vis, 1892) mekang, Malakal, Mecherchar, Ngchean- Brown (1991) did not fully analyze var- gel Atoll (Ngcheangel, Ngerebelas, Orak), iation in the "superspecies" Emoia caeru- Ngeanges, Ngeaur, Ngebad, Ngebedangel, leocauda. However, many Palau specimens Ngercheu, Ngerechur, Ngerekebesang, in museum collections are identified as Ngeruchubtang, Ngeruktabel, Ngerur, Emoia "pelewensis", apparently an early Ngetmeduch, Oreor, Pkulaklim, Ulebse- Brown manuscript name that was never chel, Ulong. published. Given the broad distribution (Fig. 12) and extensive variation of caeru- Emoia impar (Werner, 1898) leocauda, it is not surprising that Brown Originally described from 12 specimens (1991) deferred in trying to resolve all the collected on Ralum and Mioko Islands in taxonomic problems of this taxon. the Bismarck Archipelago, this species was Palauan caeruleocauda are distinctive in long considered a synonym of Emoia cy- a number of ways. The unicolor ("bronze" anura (e.g., Brown, 1991). It is now clear morph of Crombie and Steadman, 1987) that impar is a distinct species (Ineich and 54 HERPETOLOGICAL MONOGRAPHS [No. 13

FIG. 12.-The distribution of Emoia caeruleocauda "superspecies".

Zug, 1991; Guillaume et al., 1994) and tide. Ngesbokel has a few hectares of flat some data indicate that even more cryptic plateau 2-3 m above the high tide line. sibling species may be recognizable within The vegetation is a mixture of thick grass- the broad Pacific distributions of both cy- es, stunted shrubs and scattered Panda- anura and impar (Bruna et al., 1996). nus. Although we saw a few striped skinks The apparent absence of the ubiquitous on the plateau, we were unable to verify cyanura superspecies (sensu Brown, 1991) their identity in the thick vegetation. How- from the Palau Group was puzzling. How- ever, in the wrack above the high tide line ever, at least E. impar is present in Palau, in a shaded cove that provides the only albeit with a strange distribution, parallel- access to the plateau, we found Emoia im- ing in some ways that of E. jakati. Why par to be abundant at low tide, veritably over driftwood and this odd, disjunct distribution appears in swarming escaping two distinct lineages of Emoia is perplex- into the low vegetation on the slope lead- to the We estimated there ing, but we suspect that both impar and ing up plateau. jakati are relatively recent adventives to were at least a dozen individuals in the Palau, probably via natural dispersal or ac- small area. On some of the Rock Islands cidental transport by Micronesians rather we have found Emoia caeruleocauda in than as post-European introductions. very similar situations. Emoia is on The Palau populations of Emoia impar impar peripheral Ngchean- Atoll as on the two were apparently restricted to small and pe- gel well, occurring only smallest islands, with E. caeru- ripheral islands at the northern (Ngchean- sympatric leocauda on and Orak. The and southern (Southwest Is- Ngerebelas gel Atoll) Southwest Islands have been inciden- lands) extremes of the The only Republic. collected so can be said other had never been found between tally nothing species than to note that the occurs on those two distance of more than species points (a three of the islands. 400 km) until 1996, when Chris Austin Atoll and RIC collected a small series on Distribution.-Ngcheangel (Nger- Nges- ebelas, Orak), Merir, Pulo 4-5 islets at the Ngesbokel, bokel, the largest of tiny Anna, Sonsorol. mouth of Ngeremeduu Bay, Ngatpang State, western Babeldaob. Some of these Emoia jakati (Kopstein, 1926) islands were barely emergent at high tide Although this species is widespread in but connected by broad tidal flats at low the Carolines (Fig. 13), often occurring 19991199IEPTLGCA HERPETOLOGICAL MONOGRAPHSOORPS5 55

FIG. 13.-The distribution of Emoiajakati. sympatrically with members of the Emoia within the fringing reef of Palau is Nger- caeruleocauda and cyanura "superspecies" echur. (sensu Brown, 1991), its distribution in Pa- Based on our experience elsewhere in lau is decidedly strange. Brown's (1991) Micronesia (Pohnpei and surrounding is- treatment of jakati does not lend confi- lands), Emoiajakati is an inhabitant of the dence that all populations he referred to most disturbed and marginal habitats, in- that taxon are indeed conspecific. In fact, cluding around human habitation. The when Loveridge (1945) examined museum Ngerechur population confirms this im- collections for his "Reptiles of the Pacific pression since the species is extremely World" he left many specimens identified abundant around an abandoned house at with the (still unpublished) manuscript the south end of the island, an area of name Emoia mivarti 'fuscolineata," indi- mixed strand vegetation and tall Cocos. cating that he regarded the Oceanic pop- The northern end of the island is higher, ulations as distinct from New Guinea "mi- boulder-strewn and well forested, with varti," a name misapplied to jakati until much evidence of substantial Japanese oc- Brown's revision. cupation during WWII. We did not see E. Like Emoia impar, E. jakati was previ- jakati anywhere in the forest. Picknickers ously known only from Ngcheangel Atoll and overnight visitors often use the house and the Southwest Islands, but nowhere in on the southern end, but there are no per- between. However, in 1996 Chris Austin manent residents at this time. and RIC found the species abundant on Distribution.-Ncheangel Atoll (Ngche- Ngerechur, the northernmost of two small angel), Helen, Ngerechur, Tobi. islands off the northern tip of Babeldaob. At low tide the channel between Ngere- Eugongylus sp. chur and Ngerkeklau is wadable and that Dryden and Taylor (1969) first recorded between Ngerkeklau and the Babeldaob this large scincid from Palau based on a coast (about 0.5 km) is even shallower. The single specimen from Ngeaur. They re- coastal areas of extreme northern Babel- ferred it to Eugongylus mentovarius, a daob have never been collected herpeto- poorly known species previously reported logically, nor has Ngerkeklau, and E. jakati only from the Moluccas. Based on our pre- may indeed occur at either or both, but liminary comparisons of Palauan Eugon- currently the only record for the species gylus with one Moluccan E. mentovarius 56 56 HERPETOLOGICAL MONOGRAPHSMONOGRAPHS [No. 13~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~13

(USNM 237428), we do not consider them are an effective collecting technique that conspecific. However, the systematic rela- yields specimens even in areas where the tionships and content of the genus Eugon- species has not been recorded in visual gylus are so chaotic that we defer a deci- surveys. Wiles and Conry (1990) first not- sion at this time, but offer the following ed this and our collections amply confirm taxonomic summary of the genus as it cur- their observations. Eugongylus are so ef- rently exists. fective at avoiding detection that most Pa- Despite the contributions of B6hme lauans are incredulous that such a sizable, (1976) and Greer (1974), species limits, unfamiliar lizard was collected in their relationships, and even the number of spe- backyard. Sticky traps also yield some in- cies in Eugongylus (including the "genera" teresting natural history information on Phoboscincus and Tachygyia) remain Eugongylus. On at least three occasions largely unresolved. For example, Bohme we have trapped Eugongylus that clearly (1976) did not even mention mentovarius had been attracted to smaller skinks (Car- in his list of the recognized species in the lia and Emoia caeruleocauda) stuck in the genus. He did, however, include sulaensis trap. Two Emoia and one Carlia stuck near (Kopstein, 1927) which was distinguished the edge of the trap were neatly sawed off from mentovarius only in lacking the char- just above the adherent part of their bod- acteristic throat markings and in having a ies, and in another instance a Eugongylus shorter tail. Most recently, Sadlier (1986) that became stuck trying to reach a Carlia tacitly redefined Eugongylus by describing in the center of the trap regurgitated the a new genus for haraldmeieri (Geoscincus) missing pieces of smaller skinks when he and recognizing two species of Phoboscin- was being removed from the trap. Another cus. In so doing, he inferred that Tachy- Eugongylus was completely stuck along gyia is also a valid taxon. the long axis of the trap with an Emoia at Eugongylus has been infrequently col- the far end still in his jaws, thoroughly lected in Palau but it seems to be a com- masticated at midbody. When Eugongylus mon, if cryptic, terrestrial skink in a wide are partially stuck they tend to spin, croc- variety of habitats. Specimens have been odile-like, causing the entire trap to wrap taken in primary forest and mature second around them burrito-fashion. growth on rock islands (Wiles and Conry, On at least three other occasions we ob- 1990; our surveys), in backyard gardens in served Eugongylus that did not get stuck Koror town, and in scrubby forest adjacent trying to pull adherent Emoia or Carlia off to mangrove swamps (our surveys). We the trap, often escaping with a tail or piece found three individuals actively foraging of body. We think these data clearly point and most of the sticky-trapped individuals to Eugongylus being an active, primary (see below) were obviously also active. predator in the leaf litter community, One specimen (USNM531958) was mov- probably feeding on a wide variety of in- ing swiftly and silently with an almost ser- vertebrates and smaller vertebrates. We pentine motion through thick ground cov- would not be surprised if Platymantis, er on the Museum grounds (Koror town) Sphenomorphus, typhlopid snakes, and at 0915 h. Two others (USNM 531961-62) even juvenile Candoia and Dendrelaphis from Ulong were collected in the leaf litter are included in the diet. in forest on the first terrace at 1415 h, and Distribution. -Babeldaob, Beliliou, near the buttress of a large tree near the Kmekumer Islands (island 46), Ngchean- beach at 0710 h during torrential rain. In- gel Atoll (Ngcheangel), Ngeaur, Ngercheu, active individuals were found from 1430- Ngerukeuid Islands (islands 16, 24), Nge- 1510 h under logs and in moist piles of ruktabel, Oreor, Ulebsechel, Ulong. coconut husks in beach forest on Babel- daob (Ngesang and Ulimang, Ngaraard Lamprolepis smaragdina (Lesson, 1826) State). This widespread western Pacific species The secretive habits of Eugongylus give was divided into 8 subspecies by Mertens the impression of rarity, but sticky traps (1929), reflecting the extensive variation 1999]199IEPTLGCA HERPETOLOGICAL MONOGRAPHSOORPS5 57

FIG. 14.-The distribution of Lamprolepis smaragdina. throughout its range (Fig. 14). Dryden and green, and highly prominent on exposed Taylor (1969) referred Palauan specimens Cocos trunks. to the subspecies viridipunctum, but we Palauan Lamprolepis are usually bright prefer to retain the binomial pending a green with varying amounts of enamel more rigorous analysis of variation in the black on the dorsal scale edges. Addition- species. Mertens' subspecific arrangement ally, there is what we have called the "mus- does not accurately reflect the inter- and tard morph" which is most abundant in (or intra-island variation in Oceania and we possibly restricted to) northern Babeldaob. suspect that several recognizable species As the name suggests, "mustard morph" may be subsumed in Mertens' concept of individuals are dull yellow-brown above "smaragdina". and lemon yellow below, without a hint of The large, green tree skink may be the the usual greens. Normal green specimens most familiar reptile in Palau, as it often have been found on trees adjacent to perches conspicuously during the day on "mustard" perch sites and the morph is large backyard trees, even in Koror town. not sexually correlated (males, females, It is a remarkably adaptable species that and juveniles with this coloration have also occurs in forest, but we have only a been collected). few records from true forest habitat. Per- We suspect that Lamprolepis is a major haps the species is simply more difficult to predator on small organisms in Palau. It observe in forest, or it may truly be less has a broad activity period, forages from abundant there. Our impression is that the ground to the canopy, and occurs in Lamprolepis is an "edge" species, most every habitat that supports trees, even low, suited to the tangle of vegetation at the scrubby Pandanus on tiny islets. It feeds forest margin. There is considerable vari- on a wide variety of invertebrates and on ation in habitat preference within the any vertebrate it can subdue, including "species", however. Some populations or geckos and skinks (see also Anolis ac- subspecies (e.g., the Philippine philippin- count). Fruit and other sweets are also ica) are more forest-adapted and crypti- consumed. We observed Lamprolepis on cally colored (brown body and dull green Ngeaur gorging on ripe papaya. In Uli- head) versus small island Oceanic smarag- mang, northern Babeldaob, which sup- dina, which are unmarked, almost electric ports a particularly dense population, 58 HERPETOLOGICAL MONOGRAPHS [No[No. 13

Lamprolepis regularly came onto the was referred to Lipinia by Greer (1974). porch of our house to lick residual soft- Austin (1995, 1998) confirmed this generic drink from the lips of cans. Manthey allocation, but he found the relationships (1985), working mostly with Philippine in- of leptosoma relative to several Papuan dividuals in captivity, confirmed the spe- species equivocal. cies' fondness for sweets and he even re- Brown and Fehlmann (1958) claimed ported egg-eating. that this species was an obligate dweller in Reproduction also reflects the broad the leaf axils of a single species of Panda- ecological tolerance of Lamprolepis. Al- nus, and they reported specimens from though two elongate, leathery eggs are in- Babeldaob, Oreor, and "Toiracholodoch- variable (at least in Palau), we have found oel." The latter is supposedly "a small is- clutches under rocks, inside dead logs, and land at the north end of Malakal" (Brown in the duff of palm axils 3-4 m above the and Fehlmann, 1958:5), but it does not ex- ground. ist on our maps. Gordon Rodda (personal Another aspect of Lamprolepis biology communication) offered that according to that has received very little attention are Demei Otobed "Toiracholodochoel" is ac- the sexually dimorphic heel scales, greatly tually the name of a pass between the two enlarged in males and often more brightly islands, the smaller of which (Ngermalk?) colored than the plantar surface. Gandolfi is apparently the one to which Brown and (1907) first discussed this structure in Fehlmann (1958) referred. In any case, Lamprolepis and provided illustrations of there is no longer any suitable Pandanus sections through the scale, but neither he in this area. There are additional museum nor anyone since has suggested a function specimens (CM 29101-03) from Ngeaur for them. The field notes of William H. collected as long ago as 1944, and we re- Stickel described MCZ 49312 from Toem, cently verified that the species does indeed Irian Jaya, New Guinea as, "Heel scale or- occur there. Most of the recent Ngeaur ange, which is unusual." (Loveridge 1948: specimens were collected from Pandanus 340-341). Manthey (1985:8) provided a near the south end of the island, but sev- color photo of the plantar surfaces of a live eral individuals were seen and collected male and female from the Philippines in (e.g., USNM531945, SAM R4800) from which the enlarged scale in the male is palm axils near the northern tip of the is- only slightly more yellowish than the dead- land. white ventral color. By comparison the Distribution.-ENDEMIC. Babeldaob, scale is usually highly contrasting in the Ngeaur, Oreor, "Toiracholodochoel". greener populations, described in our notes as "lemon yellow", "yellow-orange" Lipinia cf. L. noctua (Lesson, 1826) or even bright orange, against the usually Zweifel (1979:1) noted "the relative uni- yellow-green plantar color. formity of color pattern and scutellation of Distribution.-Babeldaob, Bablome- L. noctua over the Pacific islands, in con- kang, Beliliou, Fana, Kmekumer Islands, trast to the diversity in the much smaller Malakal, Merir, Ngcheangel Atoll (all is- New Guinea region." He did not list the lands), Ngeaur, Ngebad, Ngebedangel, non-New Guinea material he examined Ngercheu, Ngerdis, Ngerechur, Ngereke- but the AMNH collections have noctua besang, Ngermalk, Ngerukeuid Islands (is- from a broad area of Oceania. Our field lands 12, 16, 24), Ngeruktabel, Ngerur, experience with noctua in Oceania are at Ngesbokel, Ngetmeduch, Oreor, Pkulak- odds with Zweifel's suggestion. Although lim, Pulo Anna, Sonsorol, Tobi, Ulebse- scutellation is indeed conservative, we chel, Ulong. have noted significant differences in col- oration, body form, and ecology between Lipinia leptosoma Polynesian and Micronesian noctua. Poly- (Brown and Fehlmann, 1958) nesian noctua tend to be heavier-bodied, Originally described as an endemic ge- short-legged terrestrial or semi-arboreal nus (Aulacoplax) from Palau, leptosoma skinks with a prominent, light occipital 1999] HERPETOLOGICAL MONOGRAPHS 59 spot and a distinct broad vertebral stripe. and central Polynesian population of hu- By contrast, Micronesian noctua are more man-mediated origin. Micronesian "noc- gracile, strongly arboreal skinks (possibly tua" are genetically diverse and probably even obligate palm axil dwellers), often represent pre-human colonization of the with a poorly developed occipital spot and islands. Although Lesson's type locality vertebral stripe, but more distinct dorso- (Oualan, now Kosrae in the Carolines) lateral striping. This pattern is very similar would seem to restrict the name noctua to Palauan Cryptoblepharus. (sensu stricto) to the Micronesian popu- In Palau, Lipinia noctua is found almost lations, the possible unreliability of the lo- exclusively in or very near the axils of cality and the lack of a type specimen palms, and usually uncovered by peeling (Brygoo, 1986:75) further complicate no- dead fronds and fabric from low trees. We menclatural resolution of the problem. have seen a few individuals apparently ac- However, because Lesson's text (1830) tive (or at least visible) outside axils but gives ecological information (on a building they generally slip back into concealment in a sugar cane field) for the type(s) of noc- as soon as they are detected. One excep- tua, perhaps the locality is accurate for this tion was a near-neonate that was active on species. a large dead tree near the beach just south Distribution.-Babeldaob, Malakal, of Ngaramasch, western Ngeaur. The tree Ngeaur, Ngercheu, Ngerekebesang, Oreor, had abundant loose bark which provided Pulo Anna, Ulong. refuge for numerous (10+) active Cryp- toblepharus, and the Lipinia was mistaken Mabuya sp. for a of the former until we ex- juvenile Stemfeld first Ma- amined the more when (1920:395) reported specimen closely buia multicarinata from Palau. He dis- it. Two other adults were col- preserving cussed in detail seven from Feis lected from beneath casuarina bark on specimens (= Fais) and one from Mogomay, Ulit (= northern Ngeaur. We suspect Palauan Lip- Ulithi) Atoll, in the western Carolines just inia (and other Micronesian populations) north of Palau, and concluded may also occupy the axils of tall palms, by saying that the Museum also had similar to the Philippine L. auriculata Senckenberg two from Palau, but no group (Brown and Alcala, 1980; RIC, per- specimens locality or other were sonal observation), but we have found no details provided. The ma- terial from the Carolines included way to sample taller Palauan palms as yet. juve- that the was estab- The preference for palms departs from niles, indicating species lished and the preferred habitat of Polynesian L. noc- reproducing at that time. tua, which are frequently found under Sternfeld (1920:395) summarized the rocks and logs (McKeown, 1996; personal distribution of M. multicarinata as Borneo observation) or active on a variety of low and the Philippines. Brown and Alcala vegetation, but rarely palms (Crombie and (1980) studied Philippine material of Steadman, 1987). From our limited data "multicarinata" and described two smaller on pregnant females, Micronesian noctua species, M. cumingi and indeprensa, that usually produce a single offspring, as op- had long been confused with it. They also posed to the usual litter of 2 (but with a found north-south variation in multicari- range of 1-4) in Hawaii (McKeown, 1996). nata itself, and named the northern Phil- Data from the Cook Islands suggested lit- ippine populations M. multicarinata bo- ter sizes of 1 or 2 (Crombie and Steadman, realis. Ota (1991) extended the northern 1987). distributional limits of M. m. borealis to Christopher C. Austin is currently re- Lanyu Island, off the southeastern coast of viewing morphological and molecular var- Taiwan, noting that it was not found on the iation in the genus Lipinia with significant main island of Taiwan. However, neither samples from Oceania. His recent (Austin, the Bornean nor the Palau/Carolines pop- 1999) mtDNA analysis of the widespread ulations have been studied in light of "noctua" suggests a monophyletic eastern Brown and Alcala's (1980) revision of the 600 HFRPETOIflGTCAI.HERPETOLTOGICAL MONOGRAPHS [No. 13 group, primarily due to limited samples only tentatively refer it to Sphenomorphus from these areas. pending further study. The Palau/Carolines Mabuya represent Interestingly, Tiedemann and Haupl the eastern distributional limit of the mul- (1980:42) listed 5 "Syntypes" [NMW ticarinata group and possibly the eastern 14855 (2), and 16638 (3)] of Lygosoma limit of the Asian members of the genus. (Hinulia)jagorii Peters from "Palaos", pol- Mabuya multifasciata is found in the Mol- lected by Semper in 1874. Quite aside uccas and western New Guinea but its dis- from the fact that these specimens were tribution in this area is poorly document- supposedly collected 10 years after Peters ed; Ingram (1987) concluded that described the species (1864:54), Peters multifasciata was restricted to the western clearly stated that the types ofjagorii were (Irian Jaya) area of New Guinea and that collected by Jagor at "Borongan, ... Sa- the northern Australian record was based mar" and "Tacloban, ... Leyte". Brown on erroneous locality data or an introduc- and Alcala (1980:209) list ZMB 4975 and tion. Some museum material from Palau 4610 as syntypes of Sphenomorphus jago- and the Carolines has been misidentified rii. We know of no Philippine locality as multifasciata. called "Palaos," but that name is another Many of the disjunct populations of Phi- old permutation for Palau, in which case lippine or Papuan vertebrate species in the above "syntypes" collected by Semper Oceania (e.g., Varanus, Crocodylus, Sun- are possibly this species of Sphenomor- cus, Macaca) are often attributed to intro- phus. The Palauan species vaguely resem- ductions by the Japanese during their pre- bles some jagorii, which is a poorly re- World War II occupation of Micronesia solved composite of several to many and much of the rest of insular East Asia. species throughout the Philippine archi- Mabuya, however, clearly pre-dates the pelago (Brown and Alcala, 1980:211; RIC, period of Japanese occupation, and the unpublished data). However, most of the Oceanic populations may represent natu- problematical Philippine populations are ral dispersal from the Philippines and/or in the eastern arc of islands and the far Borneo. Preliminary data from our mate- north (greater Luzon landmass and asso- rial support this proposal, since Palau/Car- ciated island groups) rather than the area olines Mabuya have the large body size most proximate to Palau (Mindanao), typical of multicarinata but scale counts which is occupied by the fairly well de- intermediate between multicarinata and fined S. jagorii grandis. Thinking that the indeprensa. We suspect that these popu- NMW specimens might actually be types, lations represent at least one undescribed but of a species other than jagorii (mean- species. The problem is currently being ing that there might be an available name studied by Hidetoshi Ota and RIC. for the Palau species), we carefully exam- Distribution.-Babeldaob, Beliliou, ined all of Wilhelm Peters' papers (Bauer Ngcheangel Atoll (Ngcheangel), Nger- et al., 1995) and could find no scincid de- cheu, Oreor, Pulo Anna, Sonsorol. scribed from "Palaos" or any permutation thereof. Sphenomorphus sp. nov. We characterize this species as a deni- This moderate-sized skink has only been zen of moist woodland, but this requires found on Koror, Ngermalk, southern Ba- some comment. The "woodland" need not beldaob, and Ngeruktabel, usually in or be primary forest. Most of our specimens near moist wooded areas. It is superficially have come from the small park adjacent to similar to some members of the Spheno- the National Museum, which barely qual- morphus variegatus group, which includes ifies as woodland. We also collected one more than 50 named species that do not specimen (USNM 531965) under a rock in fit conveniently into other groups. The Pa- a roadside ditch at the Mechang dock area lauan species has several distinctive char- (Oreor). The ditch was adjacent to the acters, including bifurcate hemipenes with lawn and driveway of a new house and the long and attenuate terminal elements. We only "woodland" in the area was a small 1999] 1999] HERPETOLOGICAL MONOGRAPHSMONOGRAPHS 61~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~61 patch of dry scrub about 10 m away. On currently assigned to two different (unre- the other extreme, the southern Babel- lated) species groups, but we think it daob specimens (SAM R47736, BPBM equally plausible that they may represent 13046) were collected under a sodden ba- a Palauan radiation of closely related spe- nana spathe in a swampy patch of Panda- cies. nus. Although previously rare in museum We also have a single juvenile from collections (Greer and Parker, 1974, found Ngermalk (USNM 531966), collected by only four specimens), S. scutatus is actu- Kurt Auffenberg in nearly undisturbed ally a common leaf-litter skink on most is- vegetation on the top of the small (north lands. During the rainy season this skink of the causeway) part of this rock island is often active in the leaf litter by day, but that is crossed by the Koror-Malakal cause- during the dry season it is usually encoun- way. Although this individual and some of tered under rocks and logs in shaded ar- our Babeldaob specimens have a more eas. It is found both in forest and in dis- complex pattern than others, they do not turbed areas, including downtown Koror. differ in scalation. The pattern differences Distribution.-ENDEMIC. Babeldaob, may be individual or ontogenetic variation. Malakal, Ngcheangel Atoll (Ngcheangel), Our only Rock Island record (USNM Ngeaur, Ngermalk, Ngeruktabel, Ngerur, 531974 from Ngeruktabel) was collected Oreor, Ulebsechel, Ulong. under a log on the trail to German Light- house at the southern end of the island. Varanidae This trail ascends through lush limestone Varanus cf. V indicus (Daudin, 1802) forest to the lighthouse ruins on the ridge The systematics of the Varanus indicus crest. The Japanese had a shore battery group (Fig. 15) are still quite chaotic de- and a support base in this area and the spite recent research in the Papuan area ruins are prominent landmarks along the that has produced two new species, sub- trail. The specimens was collected about species elevated to species status, and a halfway between the battery (marked by drastically revised taxonomy of the group the enormous gun barrel along the trail) (Bohme et al., 1994; Harvey and Barker, and the ruins of the support base. The Jap- 1998; Sprackland, 1993). None of the re- anese surely cleared some of this forest for visions have included Oceanic material, fields of fire but it is hard to tell now as probably because these populations are the trees are huge and the canopy is high, generally assumed to be introductions by more typical of undisturbed primary for- the Japanese before or during WWII (e.g., est. Owen, 1977). Our experience with Var- Distribution.-ENDEMIC. Babeldaob, anus in Oceania suggests that the situation Ngermalk, Ngeruktabel, Oreor. may not be so simplistic. Uchida (1969) noted that there were museum specimens Sphenomorphus scutatus (Peters, 1867) from Micronesia that long antedated the This Palauan endemic [Lygosoma (Hin- Japanese occupation, and he concluded ulia) scutatum Peters, 1867; syntypes (2) that they must have been a German intro- ZMB5893] is closely related to the small duction. Historical accounts suggest that Papuan species of the solomonis subgroup the Western Caroline (Yap, Ulithi) and Pa- of the S. fasciatus group (Greer and Park- lau populations might be native but that er, 1974). Although S. scutatus and the the eastern Caroline (Ifalik, Kosrae, Pohn- preceding species are strikingly different pei), Marshall (not represented by any in size and scalation, both have the unusu- voucher specimens we have located), and al, attenuate, bifurcate hemipenes (dis- Mariana monitors could be recent intro- cussed under Sphenomorphus sp.). Unfor- ductions. One of the earliest zoological re- tunately, hemipenial morphology is ports on Micronesia (Chamisso, 1821) unknown for virtually all of the multitude mentioned monitors from both Yap and of species currently assigned to Spheno- Palau, and commented on their totem sta- morphus. The two Palauan species are tus in local cultures. On the other hand, in 62 HERPETOLOGICAL MONOGRAPHS [No. 1213

FIG. 15. The distribution of Varanus indicus. the Northern Marianas Pregill (1998) is most definitely not a human commensal. found Varanus only in recent (post-Euro- In Palau, we have been unable to verify pean) strata. the earlier Koror records (AMNH 70652, We are impressed with the morpholog- CAS 152219), and have only one sight re- ical dissimilarity of Micronesian monitors. cord from the island, and that from the The "indicus" on Guam are smaller, paler, rugged limestone Ngermeuangel Penin- and more gracile than Palauan specimens. sula. On Babeldaob we have never seen The common name "mangrove monitor" monitors in the southern part of the island, used for Greater Papuan indicus reflects nor have we received any reports from the its arboreality, but the Palauan populations area by the Filipinos who work the rural are decidedly terrestrial. They take refuge farms. Monitors (bayawok in Tagalog) are in trees when but when ter- pursued only a favored food item among Filipinos and restrial retreats are unavailable. In color- if they occurred in southern Babeldaob we ation and Palauan Varanus are pattern, are confident we would have confirming black with prominent yellow dorsal ro- information and/or from this settes and other with a specimens irregular markings, valuable source. In vivid throat in both adults sparsely populated peach-colored northern Babeldaob we have seen but and are both and juveniles. They longer been unable to collect Varanus in the Uli- bulkier than monitors on Guam. We have area of State. seen a reliable photo of an individual col- mang/Ngesang Ngaraard Gressitt made notes on lected (and eaten) by Filipinos near Uli- (1952, 1953) Varanus on Atoll. Tom Fritts mang, northern Babeldaob. It was laid out Ngcheangel next to a tape measure that was a bit lon- and Gordon Rodda verified that the spe- ger than 6 feet total length. We have seen cies was still present in January 1993, but individuals approaching this length on only on the two larger islands where they Ngeaur. were moderately abundant. The only other Available data on the Marianas/Caro- place Varanus is truly common is Ngeaur, lines populations suggest that they are usu- where 20 or more can usually be seen in ally associated with the most disturbed a few hours walk along the road that cir- habitats (Bates and Abbott, 1958; Dryden, cles the island. They do occur in and near 1965). There is no comparable information Ngaramasch village, but they are most for Yap and Ulithi, but in Palau Varanus abundant in the rugged limestone of the 1999] HERPETOLOGICAL MONOGRAPHS 63

FIG. 16.-The distribution of Candoia carinata.

interior where they have multitudes of ref- the rediscovered lectotype in the Zoolo- uges and are extremely difficult to catch. gisches Forschungsinstitut und Museum Clearly, the status of Oceanic Varanus Alexander Koenig (ZFMK 35503), and re- needs additional study but collection of ported that the type locality, previously un- pertinent material is hampered by the known, was Ambon Island ("Amboina" on CITES status of all varanids. Tissue sam- the specimen label) in the southern Mo- ples from the Oceanic islands would help luccas. They examined over 300 specimens resolve the status of these important, pri- from throughout the species' range (Fig. mary predators, and clarify which popula- 16), and found morphological uniformity tions are introduced versus those that are in the populations from the southern Mo- native (and possibly endemic) species. luccas, Irian Jaya, northern Papua New Distribution. -Babeldaob, Ngcheangel Guinea, New Britain and New Ireland, all Atoll (Ngcheangel, Ngeriungs), Ngeaur, of which they regarded as "typical" C. car- Oreor. inata. Their Palau specimens (n = 29) conformed with these in most respects, SERPENTES but showed several average differences in Boidae some scale counts and tail length ratios. Candoia carinata (Schneider, 1801) Hobart M. Smith has recently taken data Giinther (1863) described Candoia su- on a great many Candoia from throughout perciliosa from Palau, but the taxon was the distribution of the genus (including quickly relegated to the synonymy of the many of our recent Palauan specimens) widespread and variable C. carinata (e.g., and has concluded that the Palauan pop- Boulenger, 1893:107). Our first field im- ulations are quite distinct, but most closely pressions of Palauan Candoia was how related to "true" carinata from the Mol- subjectively "different" they were from uccas (personal communication to RIC, New Guinea/Solomons carinata. Mc- June 1998). We expect him to formally Dowell (1979) examined a small sample of recognize Giinther's superciliosa for this Palauan Candoia in his survey of New species. Guinea boids and found them to be an in- Wynn and Zug (1985) detected some re- teresting mix of his normally allopatric productive differences between Palauan "long tailed" and "short tailed" morphs. Candoia and their larger sample from Bohme et al., (1998) recently described New Guinea, which they suggested were 64 HERPETOLOGICAL MONOGRAPHSMONOCRAPHS [No. 13

FIG. 17.-Locality records for Cerberus rynchops (after Gyi, 1970). correlated with latitude. Harlow and Shine who characterized it as "usually found in (1992) amplified Wynn and Zug's repro- the brackish water of tidal rivers, creeks, ductive information and added dietary and estuaries, but is equally at home in the data based on dissection of museum ma- upper reaches of rivers, and in salt waters terial and published observations on cap- along the coasts." tive specimens. They further noted that Although he apparently did not examine all natural data on Can- virtually history any material from Palau, Gyi included Pa- doia were derived from preserved muse- lau as the eastern distributional limit of the um material, and captive specimens, very nominate subspecies. He recognized the few actual field observations, all from the based on four New Guinea subspecies novaeguineae region. from Merauke in the Irian Candoia is common and widely distrib- specimens Jaya of New Guinea. uted in Palau and it is found in both dis- portion Cerberus is common in but often turbed habitats (even in Koror town and Palau, difficult to collect since it is nocturnal and on Malakal) and in forest. The species is usually reported to be arboreal and noc- the mangrove swamps it prefers are inhos- turnal, and indeed most of our specimens pitable after dark. The species is occasion- were so found. Still, several juveniles were ally found in the open ocean or along found active on the ground in the early beaches, particularly near river or stream morning near the beach in northern Ba- mouths. One specimen (USNM 507563) beldaob. An adult also was found sunning from northern Babeldaob was found on a limestone ledge at 1130 h on Nger- crawling across strand vegetation between duais (a small island separated from the the beach and an inland mangrove swamp southeast coast of Babeldaob by mangrove at 0710 h. and a stream). Reproduction in Philippine Cerberus Distribution.-Babeldaob, Beliliou, Ma- rynchlops was described by Gorman et al., lakal, Ngeaur, Ngercheu, Ngerduais, Nger- (1981). For the Palauan population we Islands echur, Ngerekebesang, Ngerukeuid have data on a single neonate with an um- Or- (island 13) Ngeruktabel, Ngetmeduch, bilical scar (USNM 531967, SVL 232 mm, eor, Pkulaklim. Tail 62 mm, live weight 11.3 gm) collected Colubridae 29 Jan 1995. This individual emerged from Cerberus rynchops (Schneider, 1799) a crevice in the mud bank of a mangrove The most recent review of this wide- swamp on Oreor when one of us (RIC) spread species (Fig. 17) was by Gyi (1970), stepped and sank into the ooze. We 1999] HERPETOLOGICAL MONOGRAPHS 65 thought that there might be an underwater blue populations) even 3-5 m above entrance, but further manipulation of the ground on palm fronds or in the axils. area produced no further neonates, al- Distribution.-Babeldaob, Beliliou, Ma- though other juveniles were seen in adja- lakal, Ngeanges, Ngeaur, Ngemlis Islands cent pools of the swamp. (sight record, Bailechesengel), Ngercheu, Distribution.-Babeldaob, Ngerekebes- Ngerduais, Ngerekebesang, Ngerukeuid ang, Oreor. Islands (island 24), Ngeruktabel, Oreor, Ulebsechel, Ulong. Dendrelaphis sp. Peters (1867a) described Dendrophis Hydrophiidae striolatus (ZMB 5450) that was collected Pelamis platurus (Linnaeus, 1766) from an unspecified locality on Palau; he This pan-tropical, pelagic species prob- suggested the new species was most close- ably rarely if ever enters the waters inside ly related to the Australian D. punctulatus. the Palau reef. The only known specimen Boulenger (1894:87) synonymized striola- from the Palau area (CAS 122587) was col- tus with lineolatus and his arrangement lected "ca. 2 miles off SW side of Anguar was followed by Dryden and Taylor (1969) (sic) Id.," outside the reef. Pelamis gives in their discussion of three specimens (two birth to live young while swimming at sea from Ngeaur, one from Oreor). Meise and so, unlike Laticauda, does not need to Hennig (1932) retained striolatus as a valid make landfall to lay eggs. Owen (1977) re- subspecies of lineolatus but Mertens ported that Pelamis shows up occasionally (1934a) referred to the Palau species as D. in Palauan waters after storms. punctulatus striolatus. McDowell (1984:6) Distribution. -Ngeaur? also called the Palau snakes Dendrelaphis punctulatus (with lineolatus as a synonym Laticaudidae but no mention of striolatus), based on a Laticauda colubrina (Schneider, 1799) very small sample size. This is the common sea snake in Palau, Despite McDowell's careful analysis of although some other species could be ex- the New Guinea material, we are not con- pected as occasional vagrants. Laticauda vinced that his placement of the Palauan colubrina is very common inside the reef populations is correct. Our collections con- and is often seen by divers in a variety of tain two distinct forms of Dendrelaphis: shallow and deep-water situations. Juve- the common, greenish, striped variety with niles are often trapped in tidal pools at low a yellow belly occurs in the southern part tide, where we have observed reef herons of the archipelago, but a larger, unmarked (Egretta sacra) feeding on them. All of our blue form with a pale blue venter is found Laticauda vouchers were collected on- in northern Babeldaob. Our samples do shore. not demonstrate any meristic differences As the name implies, sea kraits are not between the two morphs. as tied to the ocean as the true sea snakes. Occasional "blue" individuals are found Laticauda must come ashore to lay eggs among the Philippine Dendrelaphis cau- (unlike the live-bearing sea snakes) and dolineatus (RIC, personal observation), they often spend enough time on land to but there does not seem to be any geo- accumulate ticks (Booth et al., 1997; Das, graphic correlation in the variant. The 1992; Pernetta, 1977). Although sea caves "melanistic" Dendrelaphis salomonis from and uninhabited, precipitous rocky islands Misima Island (McDowell, 1984) may be are favoured reproductive areas where a similar situation. large aggregations of snakes may be found Palauan Dendrelaphis are very ecologi- (Saint Girons, 1964; Stuebing, 1988; cally tolerant, surviving quite well in Koror Stuebing et al., 1990), individuals may town as well as in primary forest on the come ashore singly and climb into beach Rock Islands. They are diurnal and pri- wrack to shed, rest, and perhaps even marily terrestrial, but usually sleep on bask. Shed skins and inactive individuals fences, low bushes, or (especially in the have been found in exposed root masses 66 66 HERPETOLOGICAL MONOGRAPHS [No. 13~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~13 of undercut beach trees in Ngaraard State, we have never collected it in primary for- northern Babeldaob. In this same area, a est, or the more forested parts of the larg- large male (USNM 507503) was found ac- er islands. We have collected it at the for- tively moving through strand vegetation 10 est edge at Nekkeng, Babeldaob, m from the sea at 2035 h. At this site (just sympatrically with R. braminus. The two south of Ulimang Village), there is only a species were collected microsyntopically in narrow strip of "dry land" between the sawdust piles on the grounds of the Mu- beach and an inland mangrove swamp. La- seum in Koror town (R. Owen, personal ticauda has not been found in mangrove communication, based on the following swamps in Palau, although, we collected a specimens: acuticaudus-USNM 207022, DOR subadult adjacent to mangrove 207024, 207026-28; braminus-USNM (USNM 531992; 350 mm SVL) in north 207023, 207025) and we have collected Oreor just south of the causeway to Nget- the two species under adjacent rocks in meduch and the KB Bridge. Mangrove is the same area. Unlike Owen's series, bra- extensive to the east of the road, and the minus is now much more abundant every- specimen was collected on that side of the where than acuticaudus. road; to the west mangrove is scattered, Although we have never found this spe- and opens to the shallows of the lagoon. cies on any of the many Rock Islands sur- Distribution.-Babeldaob, Beliliou, Ma- veyed, Ota's record (KUZ 20006-07) for lakal, Ngelsibel, Ngerduais, Ngerechur, Ngercheu is undeniable. It is possible that Ngerukeuid Islands (islands 13, 24), Oreor, this population was accidentally transport- Pkulaklim, Ulong. ed there in construction material for the Carp Island Resort. Typhlopidae Distribution.-ENDEMIC. Babeldaob, Ramphotyphlops acuticaudus Ngeaur, Ngercheu, Oreor. (Peters, 1877:416) This species (Typhlops acuticaudus Pe- Ramphotyphlops braminus ters, 1877; ZMB9127) is endemic to Palau (Daudin, 1803) but it is very close in coloration and exter- This polyploid, unisexual, parthenoge- nal morphology to the Papuan R. flaviven- netic species (Wynn et al., 1987) is the ter (McDowell, 1974). In fact, one muse- most successful disperser in the snake um specimen from Ngeaur (CM 29087) world, and its broad distribution in the was identified as flaviventer by Neil D. IndoPacific rivals (and parallels in some Richmond, a specialist in typhlopid snakes, ways) that of some of the house geckos. in 1967. Addison H. Wynn has examined Unlike some frogs (e.g., Bufo marinus, most of our recent Palauan typhlopid ma- Rana catesbeiana) that were intentionally terial and he confirms the lack of diagnos- introduced in new places for insect con- tic characters separating acuticaudatus trol, Ramphotyphlops braminus is appar- from flaviventer. Van Wallach (personal ently a passive disperser, albeit likely with communication) also dissected and exam- human assistance. The most probable ined the everted hemipenes of one of our mechanism is in the root balls of orna- acuticaudus specimens (SDSNH 68437) mental (more recently) or food (historical- and feels that the species is distinct from ly) plants transported by humans. Al- flaviventer. However, he (Wallach, 1996) though the natural range of braminus is did not discuss acuticaudus in his recent unknown, it occurs from the Indian sub- review of the flaviventer group of Ram- continent to most of mainland Southeast photyphlops. We retain acuticaudus as a Asia (Hahn, 1980:40). Presumably the spe- valid species, related to flaviventer, pend- cies originated somewhere in this area. ing the results of future studies. The salinity tolerance of braminus has Although Ramphotyphlops acuticaudus never been tested, but nonetheless it is a seems to require slightly more mesic con- superb island colonizer. It occurs on many ditions than R. braminus, R. acuticaudus Indian Ocean islands and has even is certainly not a forest obligate. In fact, reached the coasts of Africa. It is ubiqui- 1999] HERPETOLOGICALMONOGRAPHS 67 tous on the close-in East Asian archipela- salinity, as well as abundant insect prey for gos (Indonesia, Philippines), and it pre- a small snake like R. braminus. Assisted by sumably dispersed from there to the high-energy storm tracks, a propagule (= Pacific islands and even to the coast of a single individual of a parthenogenetic Mexico. The species' distribution in species) could easily cross the 800 km Oceania is spotty and it did not reach the from Mindanao to Palau. In fact, late Ho- Neotropics by "island hopping" across the locene fossil remains from the Marianas entire Pacific. Until very recently, the (Pregill, 1988) indicate the likely pre-hu- Mexican distribution centered around man (and certainly pre-European) pres- Acapulco, Guerrero (and associated trade ence of a braminus-like blindsnake. Al- routes inland), which was the eastern ter- though Dryden and Taylor (1969) minus of the "Manila Galleon" route that described Typhlops pseudosaurus [re- tained as a valid of transported goods from the Philippines to species Typhlina (= from the New World colonies during 18th and Ramphotyphlops) by Hahn, 1980] Harmon RIC examined the 19th Century (Dixon Village, Guam, Spanish occupation of T. and Hendricks, 1979). The contin- holotype pseudosaurus (FMNH species and considers it to R. ues to in Mexico 189357) be brami- disperse (Mancilla-Mor- nus. The was preserved as it was eno and Ramirez-Bautista, 1998). It has specimen to shed, and aberrant even turned in the preparing having up recently Caribbean head sutures on one side accounted for and Hodge, 1997) and additional (Censky Dryden and Taylor's (1969) supposed di- is inevitable. range expansion agnostic characters. Natural dispersal from the mainland to In Palau, the limited historical and eco- closer island is not out of the groups ques- logical evidence suggest that Ramphotyph- tion. Much has been made of "arboreal" lops braminus is a recent arrival, and R. behavior in the normally fossorial lepto- acuticaudus is a much older autochthon. typhlopid and typhlopid snakes, but com- The earliest records for Palauan braminus mon sense would indicate that these obli- are Fehlmann's (mid-1950's), who found gate ant and termite feeders should be the species commonly on Oreor (25 spec- able to utilize above-ground "soil" in the imens) but nowhere else, despite the fact arboreal nests and runways of their prey. that he spent far more time on Babeldaob, Our field observations and anecdotal lit- where he only collected R. acuticaudus. erature reports provide preliminary data He collected both species on Oreor, in- that this "arboreality" may be much more cluding one locality where both were common than previously expected (Das found syntopically (1 acuticaudus and 11 and Wallach, 1998; Gaulke, 1995). Taylor braminus). Ramphotyphlops braminus is (1922:59-64) suggested that some of the now widespread in Palau and is particular- larger endemic Philippine species of Typh- ly common in disturbed areas on the lops may occur exclusively in the extensive causeway-connected islands. Still, we have root masses of aerial ferns (Asplenium), collected it from the northern end of Ba- which can provide substantial "fossorial" beldaob (not connected to the south by road in the and habitat in the forest canopy. On several oc- 1990's), Ngcheangel Atoll, There are no records from the casions we have found Ramphotyphlops Ngeaur. braminus a meter or more above Rock Islands or Beliliou, but there is one climbing from the Beliliou satellite of the on live coconut trees specimen ground (after collected in 1955 Robert dark). We have also found several other Ngedbus, by Owen (CAS-SU 19077). of and species leptotyphlopid typhlopid Distribution.-Babeldaob, Malakal, snakes R. 2 m or (including braminus) Ngcheangel Atoll (Ngcheangel), more above the in Ngeaur, ground dead, usually Ngedbus, Ngetmeduch, Oreor. ant- or termite-ridden trees. The punky interior of dead coconut REPTILIA: TESTUDINES trees knocked into the ocean by storms is Chelidae, gen. and sp. indet. an obvious dispersal mechanism. There is Aoki (1977) first reported a freshwater insulation from the dehydrating effects of turtle from Palau, but his photograph was 68 HERPETOLOGICALMONOGRAPHS rNo.[No. B13 not adequate to confidently identify it to introductions elsewhere). As noted in the genus or species and no voucher speci- chelid account, exotic turtles are a poten- mens are known to exist. tially serious threat to the poorly docu- Apparently this turtle is (or was) a re- mented endemic freshwater fish and in- cent introduction from New Guinea, prob- vertebrate fauna, and serious effort should ably a species of Elseya or Emydura as be made to document the status of these suggested by Aoki (1977). Nothing is turtles in Palau. Further import should be known of the distribution or status of the prohibited, and extant populations (if any) species on Palau and most Palauans are identified and eradicated while control is unaware of its existence. Residents of still possible. Ngerekebesang, however, indicated that Distribution. -unknown. turtles could be found "commonly" in taro patches on the eastern end of the island. Cheloniidae Until a voucher specimen is available to Chelonia mydas (Linnaeus, 1758) verify the identification, it would be spec- Eretmochelys imbricata (Linnaeus, 1766) ulative to what this suggest impact (if any) Green sea turtles and introduction have if the turtles are (Chelonia mydas) might Hawksbills imbricata) are indeed established and How- (Eretmochelys reproducing. common in Palauan waters, and for cen- since turtles are often ever, top predators turies were native resi- in the could af- they exploited by aquatic ecosystems species dents for their and shells fect native inverte- meat, eggs, (Jo- aquatic vegetation, hannes, 1981; 1981). There are no or fishes. McCoy, brates, current censuses of sites within the Distribution.-unknown. nesting lagoon. Sandy beaches are limited and Bataguridae nesting sites are restricted accordingly. cf. Chinemys reevesii Hawksbills nest in the Ngerukeuid group (Seventy Islands), and Greens utilize sites Etpison (1997:211) published photo- on some of the small islands of another freshwater turtle from surrounding graphs on and doubtless oth- an in Palau. R. Babeldaob, Beliliou, unspecified locality George er isolated within the lagoon. The and Carl H. Ernst examined the places Zug pho- most site for Greens are and are sure that important nesting tograph they reasonably on Helen and Merir in the Southwest Is- it is an Asian a sub- batagurid (formerly lands of Hidetoshi Ota (Pritchard, 1981; Wilson, 1969). family Emydidae). (per- Leatherbacks coriacea) and sonal that it (Dermochelys communication) suggested Olive turtles oliva- could be and we so re- Ridley (Lepidochelys Chinemys reevesii, also occur in Palau but are uncom- fer it on that basis. Without a voucher cea) mon to Greens and Hawksbills. no further conclusions can be compared specimen Distribution in Palau.-Scattered but made. Palau It is unclear if either of the freshwater probably throughout including Southwest Islands. turtles (see previous account) are estab- lished in Palau, and origin of the introduc- REPTILIA: CROCODYLIA tions is likewise unknown. In Hawaii, sev- eral Asian turtles were imported and Crocodylus sp. 1801 released by the Chinese community as a Crocodylus porosus Schneider, food source, and at least two of them are Salt water crocodiles are native to Palau now established and breeding (McKeown, (Motoda, 1937, 1938), and supposedly 1996). Palau has a significant Chinese were much more numerous 30 years ago community and there has been a sizeable than they are now (Kimura, 1968a,b; Wil- recent influx of people and economic sup- son, 1964). They occur elsewhere in the port from Taiwan. This seems a plausible Caroline Islands only as rare waifs. A 12- possible source of the Etpison (1997) tur- foot individual was caught on Pohnpei in tle. Turtle herpetoculture is not practiced 1971 after three pigs teathered near the in Palau (but it is a major cause of exotic beach mysteriously disappeared over a six- 1999] HERPETOLOGICALMONOGRAPHS 69 month period (Allen, 1974). In Palau, a mained alive. Escapees apparently are the non-fatal attack in 1964 and the subse- source of the supposed other species in quent capture of a crocodile nearly 15 feet Palau. We would also like to point out that long was followed by the death of a local despite Messel and King's vehement de- spear fisherman in December, 1965 nial of a native freshwater croc in Palau, it (Brown, 1971; Wilson, 1965). Subsequent- is hard to imagine that the islands would ly, a crocodile eradication program was ini- not have (or had) one on purely zoogeo- tiated, and during the next 15 years various graphic grounds. It has long been thought parties of Japanese, Australians, and Pa- that Crocodylus mindorensis and C. no- lauans hunted and killed crocodiles. Un- vaeguinea are closely related, even con- fortunately, records of kills were inade- specific by some authors (see discussion in quately maintained, and there is no Hall, 1989). Although palustrine crocodiles accurate figure of the number of croco- can (and do) cross freshwater barriers, diles taken, nor were scientific vouchers most areas (especially islands, even large saved. Estimates run from a few hundred ones like Borneo) harbour only one fresh- to several thousand killed during the water Crocodylus. If in fact C. mindorensis 1970's according to Messel and King and C. novaeguinea are sister taxa then (1991), who reviewed the history of croc- only two plausible dispersal routes could odile populations in Palau and conducted explain the distribution: island-hopping the only systematic biological survey. Mes- through the islands west of New Guinea, sel and King made their survey in June, or by a direct oversea route between the 1991, and concluded that a generous es- Philippines and New Guinea. The island- timate of Palau's total number of saltwater hopping route would have to bypass Bor- crocodiles was less than 150. At that time neo, which is inhabited by another fresh- the majority of their actual sightings were water species, C. raninus (Ross, 1992). On made on Beliliou and various river systems the oversea route the only major landfall on Babeldoab (tributaries feeding Ngere- between New Guinea and the Philippines meduu Bay, others along the east coast is Palau. Given the strong Papuan influ- south to Airai) as well as Ngerdok Lake. ences (and lesser Philippine ones) in the Crocodiles are also known to occupy the Palauan herpetofauna, including species various marine lakes in the Rock Islands, that have great difficulty in crossing ma- but Messel and King (1991) dismissed the rine barriers (e.g., frogs), we think that the likelihood that they breed there because of presence of freshwater crocodiles in Palau insufficient places to lay eggs, and the fact is not as unthinkable as Messel and King that larger breeding adults would have insist. Note that we do not suggest that the limited access to these lakes because of the species is there, merely that it is an in- narrow crevices and channels that connect triguing possibility, particularly given the them to the lagoon. Today, a crocodile abundance of anecdotal reports and local farm in north Koror harbors about 40 in- lore. Unfortunately, it is a moot point since dividuals of various size classes of local C. CITES restrictions prevent collection of porosus. Supposedly, this enterprise is in specimens and data that would resolve the place to build a breeding stock for even- problem. Even more regrettable is that tual release of adults into protected areas. none of the crocodiles killed during the Messel and King (1991) also emphati- "control" period was saved for science. cally state that C. porosus is the only spe- We made no attempt to search actively cies of crocodile in Palau, despite rumors for crocodiles, C. porosus or otherwise, and opinions that have persisted for de- but note that ample habitat exists in which cades that another species, perhaps two, crocodiles could breed. We photographed occur (Kimura, 1968a,b; Thyssen, 1988). a two-foot juvenile that was brought into Crocodylus novaeguineae and the Philip- the Coral Reef Research Foundation on pine C. mindorensis were imported during Malakal in January, 1995 that was taken the 1930's for a farming venture, but by from nearby Koror. Indeed, our experienc- the end of WWII few if any of these re- es with Crocodylus in Palau are somewhat 70 HERPETOLOGICAL MONOGRAPHS [No. 13 at odds with Messel and King's data. With- ZMB collections, and Andreas Schluter provided a out even for we list of the Stuttgart holdings. particularly looking them, As the for the second Palauan and repository largest have seen hatchlings, juveniles, large herpetological collection, The California Academy of adults virtually everywhere we have trav- Science endured our frequent visits with surprising eled in Palau, even quite close to Koror tolerence. Walter Brown, Bob Drewes, Jacques Gau- Town itself. Messel and King (1991) re- thier, Al Leviton, Heidi Robeck and especially Jens counted the difficulties in for Vindum came through with space, assistance and surveying courtesies well above and the call. crocodiles, for that tidal fluctua- beyond example Jens Vindum and Carol Prince were cordial hosts tions limit survey time and that navigating whose warm hospitality was always appreciated, even through the slackwater of mangrove is per- if it cast doubt upon their sanity. ilous at night. We agree, but while we do For basic in-country logistics, Demei Otobed, Tim not diminish the efforts of their it Taunton, Chris Regel, and the Shiro family of Nga- survey, raard Babeldaob deserve mention. The would us if the number of croc- State, special surprise late Harson Shiro was particularly generous in shar- odiles in Palau was not significantly greater ing his knowledge of Palau, especially of the Ulimang than 150. area. Distribution.-Babeldaob, Beliliou, Or- The Coral Reef Research Foundation (funded by eor. the National Cancer Institute) also was a source of immeasurable assistance to us on so many fronts that Acknowledgments.-This project would have been it is difficult to thank them adequately for it and their impossible without the cooperation and assistance of companionship. To Pat Colin, Lori Colin, and Larry numerous Palauan citizens and residents. We have Sharron, plus their staff Matt, Emilio, Nathan, and enjoyed more than the usual cordial welcome in Pa- Lolita, our appreciation is bottomless. lau, and we owe a huge debt of gratitude to many At various times we were assisted in the field or people. Our governmental contacts in Palau were aided in other ways by Tim Asis, Kurt Auffenberg, Robert Owen (early in the planning stages), Demei Chris Austin, Eric Carlson, Ginny Carter, Ken Otobed, the late David Idib, Noah Idechong, and Creighton, Bill Dunson, Mandy Etpison, Robert most recently Kammen (Taktai) Chin, each of whom Fisher, Tom Fritts, Ben Imperial, The New Koror so greatly facilitated our activities that any success we Hotel, George Phocas, Gordon Rodda, Ron Leidich have had must be credited to their enlightened pol- and Sam's Dive Tours, Mary McCutcheon, Susan icies and cooperation. Schwartz, Dave Steadman, Ken Tighe, and Gary In attempting to locate and verify identifications on Wiles and the staff of the Guam Division of Aquatic all earlier Palau collections, we received cooperation and Wildlife Resources. from a great many individuals and institutions. We The Pacific Savings Bank in Koror came to our aid are unable to acknowledge all of the collections re- on countless occasions by providing communications, sponding to verify that they did not have Palauan ma- transportation, sanctuary, and numerous other cour- terial, but we appreciate the information nonetheless. tesies. President Tim Taunton along with Debbie The collections and their staffs following provided Rengiil and John De Vivo deserve banner acknowl- or other lists, loaned and/or printouts specimens edgment for their unstinting assistance and good made available for us acro- working space (museum cheer. We apologize for the persistent assumption in follow Leviton et al., 1978): nyms Koror that we must be employed by the bank. We AMNH-R. Zweifel, J. Cole, C. Myers, D. Frost, L. are confident, however, that the PSB will survive this Ford slur and continue to prosper. ANSP-E. Malnate We thank Leslie Overstreet for bibliographic assis- BMNH-N. Arnold, C. McCarthy, B. Clambe tance, and Fred Kraus and Chris Austin for com- BPBM-A. Allison, C. Kishinami ments on earlier drafts of the manuscript. Angela Kay BYU-W. Tanner Kepler provided invaluable information and speci- CM-J. McCoy, E. Censky mens from the Southwest Islands. Hidetoshi Ota FMNH-H. Voris, A. Resetar, C. Redhead shared herpetological records (KUZ) for Ngercheu KUZ-H. Ota Island. George Zug and Carl Ernst helped with the MCZ-E. E. Williams, J. Rosado, J. Cadle photo identification of the batagurid turtle. Gordon MNHN-A. Dubois, R. Bour, E. Brygoo Rodda and David Steadman provided expert reviews MVZ-H. Greene, D. Wake of the final draft, uncovering several lapses on our SAM-M. C. Austin Hutchinson, part. SMF-K. G. K6hler Klemmer, We are at last grateful to the National Geographic SMNS-A. Hallermann Schluter, J. Society for funding this project (NGS 5032-93) and UMMZ-A. G. Schneider Kluge, field biology in general. Additional support was gen- We have not yet had the oportunity to examine the erously provided by the USFWS. RIC's museum important early Palau collections in Germany and work and the early permit negotiations in Palau were Amsterdam, but Aaron Bauer and Rainer Glinther supported by Smithsonian Research Opportunities provided us with a great deal of information on the funds. 1999] HERPETOLOGICAL MONOGRAPHS 71

LITERATURE CITED Part 1. Australia and Oceania. Das Tierreich, Teil- band 109, Verlag Walter de Gruyter, Berlin, New ALCALA,A. C. 1961. Breeding behavior and early de- York, U.S.A. velopment of frogs of Negros, Philippine Islands. BAUER,A. M., R. GUNTHER,AND M. KLIPFEL.1995. 1961:679-726. Copeia The Herpetological Contributions of Wilhelm C. ALCALA,A. C., AND W. C. BROWN. 1967. Population H. Peters (1815-1883). Society for the Study of ecology of the tropical scincoid lizard, Emoia atro- Amphibians and Reptiles, Facsimile Reprints in costata, in the 1967:596-604. Philippines. Copeia Herpetology. ALLEN, G. R. 1974. The marine crocodile, Crocody- BECKON, W. N. 1992. The giant Pacific geckos of the eastern Caroline lus porosus, from Ponape, Islands, genus Gehyra: morphological variation, distribu- with notes on food habits of crocodiles from the tion and biogeography. Copeia 1992:443-460. Palau 1974:553. Archipelago. Copeia BOHME, W. 1976. Uber die Gattung Eugongylus Fit- M. S. 1997. Coastal climatic ALLEN, morphogenesis, zinger, mit Beschreibung einer neuen. Art (Repti- and Cook Island 124-146 In trends, prehistory. Pp. lia: Scincidae). Bonner Zoologische Beitrage 27(3/ P. V. Kirch and T. L. Hunt Historical Ecol- (Eds.), 4):245-251. in the Pacific Islands. Yale ogy University Press, BOHME,W, H-G HORN, AND T. ZIEGLER.1994. Zur New U.S.A. Haven, Connecticut, Taxonomie der Pazifikwarane (Varanus-indicus- ALLISON, A. 1996. of and Zoogeography amphibians komplex): Revalidierung von Varanus doreanus (A. of New Guinea and the Pacific reptiles region. Pp. B. Meyerr, 1874) mit Beschreibung einer neuen 407-436 In A. Keast and S. E. Miller (Eds.), The Unterart. Salamandra 30(2):119-142. and Evolution of Pacific Island Biotas, New Origin BOHME,W, H. M. SMITH, RYBAK,F. VAN BREU- Guinea to Eastern Patterns and Pro- J. Polynesia: KELEN,AND D. CHISZAR.1998. The lectotype and cesses. Acad. Pub., Amsterdam, Netherlands. type locality of Candoia carinata (Reptilia, ANON. 1957. NIS Gazeteer, West Pacific islands. Serpen- tes). Contemporary 1998(2):1-5. Central D.C., Herpetolology Intelligence Agency, Washington, BOLGER,D. T., AND T. CASE. 1994. ecol- U.S.A. J. Divergent of clones of the asexual . 1998. from a Mie-Palau ogy sympatric gecko, Lep- Report Joint Survey 100:397-405. for Environmental Conservation. Mie Prefectural idodactylus lugubris. Oecologia BOOTH, W. E., K. M. WONG, A. S. KAMARIAH,S. Association for Higher Education. Tsu, Japan. (In CHOY, AND I. DAS. 1997. A survey of the flora and Japanese). fauna of Pulau Brunei Darussalam. San- AOKI, R. 1977. The occurrence of a short-necked Punyit, dakania 9:55-66. chelid in the Palau Islands. Japanese Journal of BOULENGER,G. A. 1882. of the Batrachia Herpetology 7(2):32-33. (In Japanese). Catalogue Salientia s. Ecaudata in the collection of the British ATODA, K. 1950. Metamorphosis of the "non-aquatic Museum. and London. frog" of the Palau Islands, western Carolines. Pa- Taylor Francis, . 1885. of the lizards of the British cific Science 4(3):202-207. Catalogue Museum Vol. 1. AUSTIN, C. C. 1995. Molecular and morphological (Natural History), Geckonidae, evolution in South Pacific scincid lizards: morpho- Eublepharidae, Uroplatidae, Pygopodidae, Agami- dae. and London. logical conservatism and phylogenetic relationships Taylor Francis, . 1894. of the snakes in the British of Papuan Lipinia (Scincidae). Herpetologica Catalogue 51(3):291-300. Museum (Natural History), Vol. 2. Containing the conclusion of the Colubridae and . 1998. Phylogenetic relationships of Lipinia Aglyphae. Taylor (Scincidae) from New Guinea based on DNA se- Francis, London. . 1918. quence variation from the mitochondrial 12s rRNA Remarks on the batrachian genera and nuclear C-MOS genes. Hamadryad 23(2):93- Cornufer, Tschudi, Platymantis, Gthr., Simomantis, 102. g.n., and Staurois, Cope. Annals and Magazine .1999. Lizards took express train to Polynesia. Natural History (ser. 9) 1:372-375. Nature 397(6715):113-114. BRIGHAM,W. T. 1900. An index to the islands of the BAILEY, D. E. 1991. WWII Wrecks of Palau. North Pacific Ocean. Memoires Bernice P. Bishop Mu- Valley Diver Publications. Redding, California, seum 1(2):87-256. U.S.A. BRIGHT, G. R., AND J. A. JUNE. 1981. Freshwater BAKER, J. K. 1979. The rainbow skink, Lampropholis fishes of Palau, Caroline Islands. Micronesica 17(1- delicata, in Hawaii. Pacific Science 33(2):207-212. 2):107-111. BAKER, R. H. 1951. The avifauna of Micronesia, its BRONGERSMA,L. D. 1934. Contributions to Indo- origin, evolution, and distribution. University Kan- Australian herpetology. Zoologische Mededeelingen sas Publications Museum of Natural History 3(1): 17:161-251. 1-359. BROWN,J. 1971. Man-eating crocodile plague in Pa- BARBOUR, T. 1911. New lizards and a new toad from lau. Argosy Magazine, pp. 34-38. the Dutch East Indies, with notes on other species. BROWN,W. C. 1952. The amphibians of the Solomon Proceedings Biological Society Washington 24:15- Islands. Bulletin Museum Comparative Zoology 22. 107(1):3-64. BATES,M., AND D. ABBOTT. 1958. Coral Island: Por- . 1965. New frogs of the genus Cornufer (Ran- trait of an atoll. Charles Scribner's Sons, New York. idae) from the Solomon Islands. Breviora (218):1- New York, U.S.A. 16. BAUER, A. M. 1994. Gekkonidae (Reptilia, Sauria). . 1976. Review of the genus Perochirus (Gek- 72 HERPETOLOGICAL MONOGRAPHS [No. 13

konidae). Occasional Papers California Academy The Chancellor of the Empire, Count Romanzoff, Sciences (126):1-14. in the Ship Rurick, Vol. 3. Longman, Hurst, Rees, . 1991. Lizards of the genus Emoia (Scincidae) Orme, and Brown, London. with observations on their evolution and biogeog- COGGER,H. G. 1975. Reptiles and Amphibians of raphy. Memoirs California Academy Sciences (15): Australia. A. H. and A. W. Reed. Sydney. 1-94. COLE, T. G., M. C. FALANRUW,C. D. MACLEAN,C. . 1997. Biogeography of amphibians in the is- D. WHITESELL,AND A. M. AMBACHER.1987. Veg- lands of the Southwest Pacific. Proceedings Cali- etation survey of the Republic of Palau. Res. Bull. fornia Academy of Sciences 50(2):21-38. PSW-22, Pacific Southwest Forest and Range Ex- BROWN,W. C., AND A. C. ALCALA.1980. Philippine periment Station, Forest Service, USDA, Berkeley, lizards of the family Scincidae. Silliman University California, U.S.A. Natural Science Monograph Series (2):1-264. CRAIB,J. L. 1983. Micronesian prehistory: an archae- BROWN,W. C., AND H. A. FEHLMANN.1958. A new ological overview. Science 219:922-927. genus and species of arboreal scincid lizard from CROMBIE,R. I., AND D. W. STEADMAN.1987. The the Palau Islands. Occasional Papers Natural His- lizards of Rarotonga and Mangaia, Cook Island tory Museum Stanford University (6):1-7. Group, Oceania. Pacific Science 40(1-4):44-57. BROWN,W. C., AND R. F. INGER.1964. The taxonom- DAS, I. 1992. Notes on egg sizes and ticks of the ic status of the frog Cornufer dorsalis A. Dumeril. yellow-lipped sea krait (Laticauda colubrinus) on Copeia 1964:450-451. Palau Punytt, Brunei Darussalam. Hamadryad 17: BROWN,W. C., AND M. J. TYLER.1968. Frogs of the 45-46. genus Platymantis (Ranidae) from New Britain DAS, I., AND V. WALLACH. 1998. Scolecophidian ar- with of new Bi- descriptions species. Proceedings boreality revisited. Herpetological Review 29:15- ological Society Washington 81:69-86. 16. E. R. N. FISHER,AND T. CASE. 1996. BRUNA, M., J. DIXON, J. R., AND F. S. HENDRICKS. 1979. The and evolution de- Morphological genetic appear wormsnakes (family Typhlopidae) of the Neotrop- coupled in Pacific skinks (Squamata; Scincidae; ics, exclusive of the Antilles. Zoologische Verhan- Emoia). Proceedings Royal Society London B 263: dlungen. (173):1-39. 681-688. DONALDSON, T. J. 1996. Fishes of the remote South- E. H. 1971. Guide to names in the BRYAN, place west Palau Islands: a zoogeographic perspective. Trust Territories of the Pacific Islands (the Mar- Pacific Science 50(3):285-308. Caroline and Mariana Pacific shall, Is.). Honolulu, DRYDEN, G. L. 1965. The food and feeding habits of Information Bernice P. Scientific Center, Bishop Varanus indicus on Guam. Micronesica 2(1):73-76. Museum. DRYDEN, G. L., AND E. H. TAYLOR. 1969. Reptiles BRYGOO, E. R. 1987. Les de Scincides types (Rep- from the Mariana and Caroline Islands. University tiles, Sauriens) du Museum National d'Histoire Na- Kansas Science Bulletin 48(8):269-279. turelle. Bulletin Museum Na- Catalogue critique. EASTEAL, S. 1981. The of introductions of tional d'Histoire Naturelle Paris (ser. 4, sect. A, history Bufo marinus (Amphibia: Anura); a natural exper- suppl.) 7(3):1-126. iment in evolution. Biological Journal Linnean So- BUDEN, D. E. 1998. Morphological variation and dis- ciety London 16(2):93-113. tributional ecology of the giant Micronesian gecko ENGBRING, 1983. Avifauna of the Southwest Is- (Perochirus scutellatus) of Kapingamarangi Atoll. J. Pacific Science 52:250-258. lands of Palau. Atoll Research Bulletin (267):1-22. . 1988. Field Guide to the Birds of Palau. . 1998. The reptiles of Kapingamarangi Atoll, Micronesia. Atoll Research Bulletin 453:1-8. Conservation Office, Koror, Palau. AND H. D. PRATT. 1985. BURNETT,S., AND NOLEN. 1996. Fruit eating by ENBRING, J., Endangered J. birds of Micronesia: their and future the gecko Gehyra dubia in Townsville. Memoirs history, status, Bird Conservation 2:71-105. Queensland Museum 39(2):364. prospects. ETPISON, M. T. 1997. Palau-Portrait of Paradise. CASE, T. J., D. T. BOLGER,AND K. PETREN. 1994. Invasions and competitive displacement among Neco Marine Corp., Koror. house geckos in the tropical Pacific. Ecology 75: FEHLMANN,H. A. 1960. Ecological distribution of 464-487. fishes in a stream drainage in the Palau Islands. Ph.D. Stanford CASSELL,J. D., D. OTOBED,AND H. ADELBAI.1992. Dissertation, University. Stanford, Comprehensive Conservation Strategy for the Re- California. public of Palau. A review of the Palau conservation FISHER, H. I. 1947. Utinomi's Bibliographica Micro- program and recommendations for additional pro- nesica: chordate section. Pacific Science 1(3):129- gram policies. Div. Conservation/Entomology, Bu- 150. reau of Resources and Development. Koror, Palau. . 1948. Locality records of Pacific island rep- CENSKY,E. J., AND K. HODGE. 1997. Geographic dis- tiles and amphibians. Copeia 1948:69. tribution: Ramphotyphlops braminus. Herpetolog- FISHER, R. N. 1997. Dispersal and evolution of the ical Review 28:210. Pacific Basin gekkonid lizards Gehyra oceanica and CHAMISSO. 1821. A Voyage of Discovery, into the Gehyra mutilata. Evolution 51:906-921. South Sea and Beering's Straits, for the Purpose of FOSBERG, F. R., D. OTOBED, M. SACHET, R. OLIVER, Exploring a North-East Passage, Undertaken in the A. CULCIE, AND J. E. CANFIELD. 1980. Vascular Years 1815-1818, At the Expense of His Highness plants of Palau with vernacular names. Department 1999] HERPETOLOGICAL MONOGRAPHS 73

Botany, Smithsonian Institution, Washington, D.C., species of blue-tailed monitor lizard (genus Var- U.S.A. anus) from Halmahera Island, Moluccas. Herpe- FROST, D. R. (Ed.). 1985. Amphibian Species of the tologica 54(1):34-44. World: a Taxonomic and Geographic Reference. HOLLOWAY, J. D. 1997. Sundaland, Sulawesi and Allen Press, Inc., Lawrence, Kansas. eastwards: a zoogeographic perspective. Malayan GANDOLFI,H. 1907. Ein sekundarer Geschlechtsun- Nature Journal 50:207-227. terschied bei Lygosoma smaragdinum (Less). Zool- HONEGGER, R. E. 1970. Eine Krote erobert die Welt. ogischer Anzeiger 32(7):186-188. Natur und Museum 100(10):447-453. GAULKE,M. 1995. Observations on arboreality in a HOUGH, F. 0. 1990. The Assault on Peleliu. Marine Philippine blind snake. Asiatic Herpetol. Research Corps Monographs No. 25. The Battery Press (6):45-48. Nashville, Tenessee. GIBBONS, J. R. H. 1985. The biogeography and evo- INEICH, I. 1988. Mise en 6vidence d'un complexe lution of Pacific island reptiles and amphibians. Pp. unisexue-bisexu6 chez le gecko Lepidodactylus lu- 125-142 In G. Grigg and R. Shine (Eds.), Biology gubris (Sauria, Lacertilia) en Polynesie fran?aise. of Australasian Frogs and Lizards, Royal Society C. R. Acad6mic Sciences Paris, t. 307, Ser. III: New South Wales, Surrey Beatty and Sons, Sydney, 271-277. Australia. INEICH,I., AND H. OTA. 1992. Additional remarks on GIBBONS,J. R. H., AND F. G. A. CLUNIE. 1986. Seal the unisexual-bisexual complex of the gecko Lepi- level changes and Pacific Prehistory. Journal Pacific dodactylus lugubris, in Takapoto Atoll, French History 21(2):58-82. Polynesia. Bulletin College Sciences University of GORHAM,S. W. 1965. Fiji frogs (with synopses of the the Ryukyus 53:31-39. genera Cornufer and Platymantis). Zoologische INEICH,I., AND G. R. ZUG. 1991. Nomenclatural sta- Beitrage 11(3):381-435. tus of Emoia cyanura (Lacertilia, Sciencidae) pop- GORMAN,G. C., P. LICHT,AND F. MCCOLLUM.1981. ulations in the central Pacific. Copeia 1991:1132- Annual reproductive patterns in three species of 1136. marine snakes from the central Philippines. Journal INGER, R. F. 1954. Systematics and zoogeography of Herpetology 15(3):335-354. Philippine Amphibia. Fieldiana: Zoology 33(4): GREER,A. E. 1974. The generic relationships of the 183-531. scincid lizard and its relatives. genus Leiolopisma INGRAM, G. 1987. Does the skink, Mabuya multifas- Australian Journal Ser. Zoology, Suppl. (31):1-67. ciata, occur in Australia? Northern Territory Nat- A. AND F. PARKER.1974. The GREER, E., fasciatus uralist (10):11-12. of (Lacertilia: Scin- species group Sphenomorphus JOHANNES, R. E. 1981. Words of the Lagoon. Fishing cidae): notes on eight previously described species and Marine Lore in the Palau District of Micro- and of three new New descriptions species. Papua nesia. University California Press, Berkeley, Cali- Guinea Scientific 25:31-61, Society Proceedings fornia, U.S.A. 1973(1974). JOHNSON, S. P. 1972. Palau: exploring the limestone GRESSITT, L. 1952. of Atoll, J. Description Kayangel islands. National Parks and Conservation Palau Islands. Atoll Research Bulletin 14:1-6. Magazine 46(7):4-8. . 1953. Notes on Ngaruangl and Kayangel P. N. VAN. 1923. The of the Indo- Atolls, Palau Islands. Atoll Research Bulletin 21:1-5. KAMPEN, Amphibia Australian E. Leiden. GUILLAUME,C.-P., I. INEICH, AND S. BOISSINOT. Archipelago. J. Brill, KAROLLE, B. G. 1993. Atlas of Micronesia, 2nd ed. 1994. Allozyme evidence for specific status of the Bess Press, Honolulu. two French Polynesian skink species in the genus G. 1802. An Account of the Pelew Emoia (Reptilia: Lacertilia). Copeia 1994:1042- KEATE, Islands, 1047. Situated in the Western Part of the Pacific Ocean, from the Journals of GUNTHER, A. 1863. Third account of new species of Composed Captain Henry snakes in the collection of the British Museum. An- Wilson and Some of his Officers Who, in August nals and Magazine Natural History (ser. 3) 12(71): 1783, were there Shipwrecked in the Antelope, a 348-365. Packet Belonging to the Honourable East India GYI, K. 1970. A revision of colubrid snakes of the Company. G. Nicol, London. subfamily Homalopsinae. University Kansas Publi- . 1808. The Voyage of Captain Henry Wilson, cations Museum Natural History 20(2):47-223. Principally Relating to his Shipwreck on the Pelew HAHN, D. E. 1980. Liste der Rezenten Amphibien Island Including the History of Prince Lee Boo. und Reptilien. Anomalepididae, Leptotyphlopidae, Charles Spear, Windsor, Vermont, U.S.A. Typhlopidae. Das Tierreich, Berlin, Lief. 101. KIMURA, W 1968a. Crocodiles of Palau Islands. Ata- HALL, P. M. 1989. Variation in Geographic isolates of gawa Tropical Garden and Alligator Farm, Re- the New Guinea crocodile (Crocodylus novaegui- search Reports (1):1-40. (In Japanese). neae Schmidt) compared with the similar, allopatric .1968b. On the hatching of crocodile eggs of Philippine crocodile (C. mindorensis Schmidt). Palau. Atagawa Tropical Garden and Alligator Copeia 1989:71-80. Farm, Research Reports (2):1-32. (In Japanese). HARLOW,P., AND R. SHINE. 1992. Food habits and KING, M. B., AND P. HORNER. 1989. Karyotypic evo- reproductive biology of the Pacific island boas lution in Gehyra (Gekkonidae; Reptilia). V. A new (Candoia). Journal Herpetology 26(1):60-66. species from Papua New Guinea and the distribu- HARVEY,M. B., AND D. G. BARKER.1998. A new tion and morphometrics of Gehyra oceanica (Les- 74 HERPETOLOGICAL MONOGRAPHS [No. 13

son). The Beagle, Records Northern Territory Mu- New Guinea. American Museum Novitates (2775): seum Arts and Sciences 6(1):169-178. 1-28. KOPSTEIN,F. 1926. Reptilien von den Molukken und McKEOWN, S. 1996. A Field Guide to Reptiles and den benachbarten Inseln. Zool. Meded. Leiden 9: Amphibians in the Hawaiian Islands. Diamond 71-112. Head Publications, Inc. Los Osos, California, KROENKE,L. 1984. Cenozoic tectonic development U.S.A. of the Southwest Pacific. UN ESCAP. CCOP/ MAIAVA,I., AND BUREAUOF NATURALRESOURCES SoPac. Technical Bulletin 6:1-122. AND DEVELOPMENT(BNRD). 1994. Republic of KURAMOTO,M. 1997. Relationships of the Palau frog, Palau: National environmental management strat- Platymantis pelewensis (Anura: Ranidae): morpho- egy. South Pacific Environment Programme, Apia, logical, karyological, and acoustic evidence. Copeia Western Samoa. 1977:183-187. MANTHEY,U. 1985. Der Smaragdskink, Lamprolepis LAMBERSON, J. 0. 1987. Reptiles of Enewetak Atoll. smaragdina (Lesson)-ein eierfressender Baum- Pp. 325-329 In D. M. Devaney et al. (Eds.), The skink. Sauria 7(1):5-11. Natural History of Enewetak Atoll, Vol. 2. Bioge- MARTIN,K. R. (Ed). 1980. "Naked and a Prisoner": ography and Systematics, U.S. Dept. Energy, Oak Captain Edward C. Barnard's Narrative of Ship- Ridge, Tennessee, U.S.A. wreck in Palau 1832-1833. Kendall Whaling Mu- LESSON, R. P. 1830. Zoologie. In L. I. Duperrey seum, Sharon, Massachusetts, U.S.A. (Ed.), Voyage Autour du Monde, Excete par Ordre MEISE, W, AND W. HENNIG. 1932. Die Schlangen- du Roi sur la Corvette de sa Majeste, La Coquille, gattung Dendrophis. Zoologischer Anzeiger 99(11/ Pendant les Annees 1822-1825. Arthur Bertrand, 12):273-297. France. Paris, MENZIES, J. I. 1982a. Systematics of Platymantis pa- LEVITON,A. E., W. E. DUELLMAN,AND T. H. FRITTS. puensis (Amphibia: Ranidae) and related species of 1978. Museum Review acronyms. Herpetological the New Guinea region. British Journal Herpetol- 9(1):5-9. ogy 6:236-240. LOVERIDGE,A. 1945. Reptiles of the Pacific World. . 1982b. The voices of some male Platymantis MacMillan Co., New York. species of the New Guinea region. British Journal . 1948. New Guinean and reptiles amphibians Herpetology 6:241-245. in the Museum of and the Comparative Zoology MERTENS, R. 1924. Herpetologische Mitteilungen. United States National Museum. Bulletin Museum III. Zur Nomenklatur einiger Scinciden. Sencken- Comparative Zoology 101(2):307-430. bergiana 6(3/4):180-181. MANCILLA-MORENO,M., AND RAMIREZ-BAUTISTA, . 1926. Uber die Rassen einiger indo-austral- A. 1998. distribution: Geographic Ramphotyphlops ischer Reptilien. Senckenbergiana 8(5/6):272-279. braminus. Review 29(1):54. Herpetological .1928. Cber die Einwirkung der Kulturland- MACLEAN, D. C., T. G. COLE, C. D. WHITESELL, schaft auf die Verbreitung der Amphibien und AND K. E. McDUFFIE. 1988. Timber Resources of Der Garten 1(5/6):195- of Palau. USDA Resource Reptilien. Zoologischer Babelthaup, Republic 203. Bulletin PSW 23:1-8. . 1929. Die Rassen des Dasia M. 1996. Putative interactions of Smaragdskinkes, McCOID, J. geckos Lesson. in the southern Mariana Islands. Micronesica smaragdinum Zoologischer Anzeiger 84(9/ 10):209-220. 28(2):193-202. . 1931. boutonii und MCCOID, M. G. H. RODDA, R. A. HENSLEY,AND Ablepharus (Desjardin) J., seine Variation. T. H. FRITTS. 1995. Habitat selection by geographische Zoologischer Jahr- Crypto- blicher blepharus poecilopleurus (Scincidae) in the Mari- 61(1/2):63-210. die Rassen ana Islands. Micronesica 28(1):103-108. . 1933. Weitere Mitteilungen uber von boutonii I. Zoolo- McCoy, M. 1980. Reptiles of the Solomon Islands. Ablepharus (Desjardin), Anzeiger 105(3/4):92-96. Wau Ecology Institute Handbook 7. gischer McCoY, M. A. 1981. Subsistence hunting of turtles 1934a. Die Schlangengattung Dendrelaphis in the western Pacific: the Caroline Islands. Pp. Boulenger in systematischer und zoogeographisch- 275-280 In Karen A. Bjorndal (Ed.), Biology and er Beziehung, Archiv Naturgeschichte Systematis- Conservation of Sea Turtles. Smithsonian Institu- che Zoologischer. N.F., 3(2):187-204. tion Press, Washington D.C., U.S.A. . 1934b. Weitere Mitteilungen uber die Ras- McDOWELL, S. B. 1974. A catalogue of the snakes sen von Ablepharus boutonii (Desjardin), II. Zool- of New Guinea and the Solomons, with special ref- ogischer Anzeiger 108(1/2):40-43. erence to those in the Bernice P. Bishop Museum, . 1964. Weitere Mitteilungen uber die Rassen Part I. Scolecophidia. Journal Herpetology 8(1):1- von Ablepharus boutonii, III. Zoologischer Anzeig- 57. er 173(2):99-110. . 1979. A catalogue of the snakes of New .1967. Die herpetologische Sektion des Na- Guinea and the Solomons, with special reference tur-Museums und Forschungs-Institute Sencken- to those in the Bernice P. Bishop Museum, Part berg in Frankfurt a. M. nebst einem Verzeichnis III. Boinae and Acrochordoidea. Journal Herpetol- ihrer Typen. Senckenbergianna Biologica 118(A): ogy 13(1):1-92. 1-106. . 1984. Results of the Archbold Expeditions. MESSEL, H., AND F. W. KING. 1991. Survey of the No. 112. The snakes of the Huon Peninsula, Papua crocodile populations of the Republic of Palau, 1999] HERPETOLOGICAL MONOGRAPHS 75

Caroline Islands, Pacific Ocean, 8-24 June 1991. A gischen Museums. Monatsberichte der koniglich Report to the Government of Palau. Koror, Palau. Akademie der Wissenschaften Berlin, 159-164. MOTODA,S. 1937. Crocodiles of Palau. Botany and . 1877. Herpetologische Notizen. II. Bemer- Zoology, Special Issue 5:131-138 (In Japanese). kungen tiber neue oder weniger bekannte Amphi- MOTODA,S. 1938. Crocodiles of Palau-a follow-up bien. Monatsberichte der kbniglich Akademie der report. Botany and Zoology, Special Issue 6:83-86. Wissenschaften Berlin, 415-423. (In Japanese). . 1879. Neue oder weniger bekannte Eidech- MOTTELER,L. S. 1986. Pacific island names. A map senarten aus der Familie der Scinci (Eumeces and name guide to the new Pacific. Bernice P. Giintheri, Euprepes notabilis, Ablepharus rutilus). Bishop Museum Miscellaneous Publications 34:1- Sitzungsberichte der Gesellschaft Naturforschen- 91. den Freunde Berlin, 35-37. OLIVER,J. A., AND C. E. SHAW. 1953. The amphib- PETREN, K., D. T. BOLGER,AND T. J. CASE. 1993. ians and reptiles of the Hawaiian Islands. Zoologica Mechanisms in the competitive success of an in- 38(2):65-95. vading sexual gecko over an asexual native. Science OSBORNE,D. 1966. The archaeology of the Palau Is- 259:354-358. lands. Bernice P. Bishop Museum Bulletin 230:1- PETREN,K., AND T. J. CASE. 1996. An experimental 497. demonstration of exploitation competition in an on- OTA, H. 1991. Taxonomic status of Mabuya multi- going invasion. Ecology 77:118-132. H. P. L. AND D. carinata (Gray, 1845) (Scincidae: Squamata: Rep- PRATT, D., J. ENGBRING, BROWN, tilia) from Taiwan, with comments on the herpe- G. BERRETT. 1980. Notes on the taxonomy, natural tofauna of Lanyu Island. Bulletin College Sciences, history, and status of the resident birds of Palau. University Ryukyus (51):11-18. Condor 82:117-131. OTA, H., R. N. FISHER, I. INEICH, AND T. J. CASE. PREGILL, G. K. 1993. Fossil lizards from the late Qua- of 'Eua, Pacific Science 47:101-114. 1995. Geckos of the genus Lepidodactylus (Squa- ternary Tonga. mata: Reptilia) in Micronesia: description of a new . 1998. Squamate reptiles from prehistoric in 1998:64-75. species and reevaluation of the status of Gecko sites the Mariana Islands. Copeia P. C. H. 1981. Marine turtles of Micro- moestus Peters, 1867. Copeia 1995:183-195. PRITCHARD, nesia. 263-274 In K. A. OTA, H., AND M. MATSUI.1995. Karyotype of a ranid Pp. Bjorndal (Ed.), Biology and Conservation of Sea Turtles. Smithsonian In- frog, Platymantis pelewensis, from Belau, Micro- stitution D.C., U.S.A. nesia, with comments on its systematic implica- Press, Washington R. S. C. R. N. tions. Pacific Science 49(3):296-300. RADTKEY, R., DONNELLAN, FISHER, C. MORITZ,K. A. HANLEY,AND T. CASE. 1995. OTOBED, D. O., AND I. A. MAIAVA.1994. Republic J. When species collide: the origin and spread of an of Palau: State of the environment report. South asexual of Pacific Environment Western species gecko. Proceedings Royal Society Programme, Apia, London B 259:145-152. Samoa. RADTKEY,R. R., B. BECKER,R. D. MILLER,R. RIB- OWEN, R. P. 1977. Terrestrial vertebrate fauna of the LET, AND T. CASE. 1996. Variation and evolution Palau Islands. Report from the Office of Chief J. of Class I Mhc in sexual and parthenogenetic geck- Conservationist, Trust Territory of the Pacific Is- os. Proceedings Royal Society London, London B Koror. lands, 263:1023-1032. PARENTI,L. R., AND J. A. MACIOLEK.1993. New si- RODDA, G. H., T. H. FRITTS,AND J. D. REICHEL. from and Palau, Micronesia, cydiine gobies Ponape 1991. The distributional patterns of reptiles and with comments on of the Si- systematics subfamily amphibians in the Mariana Islands. Micronesica 24: cydiinae (Teleostei: Gobiidae). Bulletin Marine Sci- 195-210. ence 53(3):945-972. RooIJ, N. DE. 1915. The reptiles of the Indo-Austra- 1977. Observations on the habits and PERNETTA,J. C. lian Archipelago. I. Lacertilia, Chelonia, Emydo- of the sea snakes Laticauda colubrina morphology sauria. E. J. Brill Ltd., Leiden. (Schneider) in Fiji. Canadian Journal Zoology 55: ROSENMAN,H. 1992. Two Voyages to the South Seas. 1612-1619. University of Hawaii Press, Honolulu, Hawaii, PETERS,W. 1867a. Herpetologische Notizen. Mon- U.S.A. atsberichte der k6niglich Akademie der Wissen- Ross, C. A. 1992. Designation of a lectotype for schaften Berlin, 13-37. Crocodylus raninus S. Muller and Schlegel (Rep- . 1867b. Uber Flederthiere (Pteropus Gouldii, tilia: Crocodylidae), the Borneo crocodile. Pro- Rhinolophus Deckenii, Vespertilio lobipes, Vespe- ceedings Biological Society Washington 105(2): rugo Temminckii) und Amphibien (Hypsilurus 400-402. Godeffroyi, Lygosoma scutatum, Stenostoma nari- ROSS, B. D. 1991. Peleliu: Tragic Triumph. Random rostre, Onychocephalus unguirostris, Ahaetulla House, New York. polylepis, Pseudechis scutellatus, Hoplobatrachus RUEDI, M., C. COURVOISIER,P. VOGEL,AND F. M. Reinhardtii, Hyla coriacea.) Monatsberichte der CATZEFLIS. 1996. Genetic differentiation and zoo- koniglich Akademie der Wissenschaften Berlin, geography of Asian Suncus murinus (Mammalia: 703-712. Soricidae). Biological Journal Linnean Society Lon- . 1874. Uber neue Reptilien (Peropus, Agama, don 57:307-316. Euprepes, Lygosoma, Typhlops, Heterolepis) der SADLIER, R. 1986. A review of the lizards of New herpetologischen Sammlung des Berliner Zoolo- Caledonia. Records Australian Museum 39:1-66. 76HEPTLGCLMNGAH[N.176 HERPETOLOGICAL MONOGRAPHS [No. 13

SAINT GIRONS, H. 1964. Notes sur l'ecologie et la monitor lizard, Varanus indicus (Daudin), as a rat structure des populations des Laticaudinae (Ser- control agent on Ifaluk, western Caroline Islands. pentes, Hydrophiidae) en Nouvelle Caledonia. La Journal Faculty Agriculture, Kyushu University Terre et la Vie (2):185-214. 15(3):311-330. SCHNEIDER,J. G. 1801. Historiae amphibiorum na- UNITED STATESARMY CHIEF OF ENGINEERS.1956. turalis et literariae, II. Jena. Military Geology of Palau Islands, Caroline Islands. SEMPER,C. 1873. Die Palau-Inseln im Stillen Ocean: U.S. Army Forces Far East, and Eighth U.S. Army Reiseerlebnisse, Brockhaus, Leipzig. (Rear). SEMPER,C., AND R. D. CRAIG. 1982. The Palau Is- VANCE,T. 1987. The Cuban green anole (Anolis por- lands in the Pacific Ocean. Micronesian Area Re- catus): a colonizing species. Bulletin Maryland search Center, University Guam, Mangilao. Herpetological Society 23(3):105-108. SENO, T., AND S. MARUYAMA.1984. Paleogeographic WALLACH, V. 1996. The systematic status of the Ram- reconstruction and origin of the Philippine Sea. Pp. photyphlops flaviventer (Peters) complex (Serpen- 53-84 In R. L. Carlson and K. Kobayashi (Eds.), tes: Typhlopidae). Amphibia-Reptilia 17:341-359. Geodynamics of Back-Arc Regions. Tectonophysics WERNER, F. 1898. Vorlaufige Mittheilungen uber die 102. von Herrn Prof. F. Dahl im Bismarckarchipel ge- SMITH, C. W. 1983. Soil survey of Islands of Palau, sammelten Reptilien und Batrachier. Zoologischer Republic of Palau. USDA Soil Conservation Ser- Anzeiger 21(57):552-556. vice. WILES, G. J., AND P. J. CONRY.1990. Terrestrial ver- SPRACKLAND, R. G. 1993. Rediscovery of a Solomon tebrates of the Ngerukewid Islands Wildlife Pre- Islands monitor lizard (Varanus indicus spinulosus serve, Palau Islands. Micronesica 23(1):41-66. Mertens, 1941). The Vivarium 4(5):25-27. WILES, G. J., J. ENGBRING, AND D. OTOBED. 1997. SPRINGER, V. G. 1982. Pacific Plate biogeography, Abundance, biology, and human exploitation of with special reference to shorefishes. Smithsonian bats in the Palau Islands. Journal Zoology, London Contributions Zoology (367):1-182. 241:203-227. STEADMAN, D. W. 1995. Prehistoric extinctions of Pa- WILSON, P. T. 1964. Report of crocodile attack in the cific Island birds: biodiversity meets zooarchaeol- western Caroline Islands. Micronesica 1:151-153. Science 267:1123-1131. ogy. WILSON, P. T. 1965. Large crocodile taken in the Pa- . 1999. The prehistory of Vertebrates, espe- lau Islands. Micronesica 1:87. on and North- cially birds, Tinian, Aguiguan, Rota, WILSON, P. T. 1969. Trust Territory of the Pacific Is- ern Mariana Islands. Micronesica 31(2):59-00. lands. Pp. 687-695. In F. E. Firth (Ed.), The En- STERNFELD,R. 1914. Die der Deutschen schlangen cyclopedia of Marine Resources, Van Nostrand in der Sudsee. 1-47 In Die Fau- Schutzgebiete Pp. Reinhold Co., New York. na der Deutschen Kolonien, Reihe VI, Deutsche WYNN,A. H., C. COLE,AND A. L. GARDNER.1987. Schutzgebiete in der Sudsee, Heft 1; R. Friedlan- J. Apparent triploidy in the unisexual Brahminy blind der & Sohn, Berlin, Germany. snake, Ramphotyphlops braminus. American Mu- . 1920. Zur Tiergeographie Papuasiens und seum Novitates (2868):1-7. der Pazifischen Inselwelt. Abhandlungen, Senck- WYNN, A. H., AND G. R. ZUG. 1985. Observations on enbergishen Naturforschenden Gessellschaft 36: the of Candoia carinata 375-436. reproductive biology (Ser- Boidae). The Snake 17:15-24. STUEBING, R. 1988. Island romance: the biology of pentes, G. R. 1991. The lizards of natural the yellow-lipped sea krait. Malayan Naturalist ZUG, Fiji: history and Bernice P. Museum Bul- 41(3+4):9-11. systematics. Bishop STUEBING,R. B., E. LADING, AND F. S. P. LIEW. letin Zoology (2):1-136. G. R. 1998. Australian of the Nac- 1990. Snake Island of Pulau Tiga Park. Sabah Parks ZUG, populations Memoirs Mu- Publications (11):1-26. tus pelagicus complex. Queensland TAYLOR, E. H. 1922. The snakes of the Philippine seum 42(2):613-626. Islands. Dept. Agriculture and Natural Resources, ZUG, G. R., AND B. R. MOON. 1995. Systematics of Bureau Science, Manila, Pub. 16:1-312. the Pacific slender-toed geckos, Nactus pelagicus TIEDEMANN,F., AND M. HAUPL. 1980. Typenkatalog complex: Oceania, Vanuatu, and Solomon Islands der Herpetologischen Sammlung. Kataloge der populations. Herpetologica 51(1):77-90. wissenschaftlichen Sammlungen des Naturhisto- ZUG, G. R., AND P. B. ZUG. 1979. The marine toad, risches Museum Wien, Austria. Vertebrata II. Rep- Bufo marinus: a natural history resume of native tilia. 4:5-79. populations. Smithsonian Contributions Zoology THYSSEN, M. 1988. A Guidebook to the Palau Is- (284):1-58. lands. Neco Marine, Koror. ZWEIFEL, R. G. 1969. Frogs of the genus Platymantis TYLER, M. J. 1979. Herpetofaunal relationships of (Ranidae) in New Guinea, with the description of South America with Australia. Pp. 73-106 In W. E. a new species. American Museum Novitates Duellman (Ed.), The South American Herpetofau- (2374):1-19. na: its Origin, Evolution and Dispersal. Museum . 1975. Two new frogs of the genus Platyman- Natural History, University Kansas Monographs 7. tis (Ranidae) from New Britain. American Museum UCHIDA, T. 1969. Rat control procedures on the Pa- Novitates (2582):1-7. cific islands, with special reference to the efficiency . 1979. Variation in the scincid lizard Lipinia of biological control agents. I. Appraisal of the noctua and notes on other Lipinia from the New 1999] HERPETOLOGICAL MONOGRAPHS 77

Guinea region. American Museum Novitates *Crocodylus porosus (2676):1-21. CHEMOI GROUP APPENDIX I (ca. 0.5 km NW of Mecherchar) Distributions by Island Gehyra oceanica (* = island records NOT verified by RIC/GKP EUIDELCHOL ISLAND collections) (ca. 0.5 km N of Mecherchar) BABELDAOB Emoia caeruleocauda Gehyra brevipalmata Bufo marinus Platymantis pelewensis IOULOMEKANG Gehyra brevipalmata (ca. 0.8 km SSW of Bablomekang) Gehyra mutilata Gehyra oceanica Gehyra mutilata Gekko sp. Lepidodactylus lugubris Hemidactylus frenatus Emoia caeruleocauda Hemiphyllodactylus cf. typus KMEKUMER ISLAND GROUP Lepidodactylus lugubris Lepidodactylus moestus (ca. 2 km W of Ngerukeuid Group) Carlia cf. C. fusca (*specimens from island "46" of Wiles and Conry, Cryptoblepharus sp. 1990) Emoia atrocostata Emoia caeruleocauda Gehyra brevipalmata *Gekko sp. Eugongylus sp. lugubris Lamprolepis smaragdina Lepidodactylus *Lepidodactylus moestus Lipinia leptosoma *Cryptoblepharus sp. Lipinia cf. L. noctua *Eugongylus sp. Mabuya sp. Lamprolepis smaragdina *Sphenomorphus sp. nov. Sphenomorphus scutatus MALAKAL *Varanus cf. v. indicus (CAS 152219) Candoia carinata Bufo marinus Cerberus rynchops Platymantis pelewensis Dendrelaphis sp. Gehyra brevipalmata Laticauda colubrina Gehyra mutilata Gekko Ramphotyphlops acuticaudus sp. Ramphotyphlops braminus Hemidactylus frenatus *Eretmochelys imbricata (CAS 154193) Lepidodactylus lugubris *Crocodylus porosus Lepidodactylus moestus Anolis carolinensis BABLOMEKANG Carlia cf. C. fusca (ca. 2.5 km W of Mecherchar) Emoia atrocostata Emoia caeruleocauda Gehyra brevipalmata Lamprolepis smaragdina Gekko sp. Lipinia cf. L. noctua. Lamprolepis smaragdina Sphenomorphus scutatus BF,.TJTJOU Candoia carinata Dendrelaphis sp. Bufo marinus Laticauda colubrina Platymantis pelewensis Ramphotyphlops braminus *Gehyra brevipalmata Gekko sp. MECHERCHAR Hemidactylus frenatus Emoia caeruleocauda Lepidodactylus lugubris Carlia cf. C. fusca NGCHEANGEL ATOLL *Cryptoblepharus sp. NGCHEANGEL ISLAND *Emoia atrocostata *Emoia caeruleocauda *Bufo marinus *Eugongylus sp. (CAS 122407-08) *Platymantis pelewensis Lamprolepis smaragdina *Gehyra mutilata *Mabuya sp. (CAS 122595-97) *Gehyra oceanica *Candoia carinata *Lepidodactylus moestus *Dendrelaphis sp. *Emoia caeruleocauda *Laticauda colubrina *Emoia jakati 78 HERPETOLOGICAL MONOGRAPHS No.[No. 13

*Eugongylus sp. (CAS 122409) NGEBAD *Lamprolepis smaragdina *Cryptoblepharus sp. *Mabuya sp. *Emoia atrocostata *Sphenomorphus scutatus *Emoia caeruleocauda *Varanus indicus *Lamprolepis smaragdina *Ramphotyphlops braminus NGEBEDANGEL NGEREBELAS ISLAND (0.1 km NW of Ngeruktabel) *Gehyra oceanica Emoia caeruleocauda *Lepidodactylus moestus Lamprolepis smaragdina *Emoia caeruleocauda *Emoia impar NGEDBUS *Lamprolepis smaragdina (0.6 km N of Beliliou) NGERIUNGS ISLAND *Ramphotyphlops braminus CAS 19077

*Platymantis pelewensis NGELSIBEL *Gehyra oceanica (ca. 1 km N of Mecherchar) *Lepidodactylus moestus Laticauda colubrina *Lamprolepis smaragdina *Varanus cf. v. indicus NGEMLIS ISLANDS ORAK ISLAND BAILECHESENGEL

oceanica Platymantis pelewensis (calling 1/97) Gehyra Dendrelaphis sp. (sight record 1/97) Lepidodactylus moestus Emoia caeruleocauda IILBLAU Emoia impar Gehyra brevipalmata Lamprolepis smaragdina Lepidodactylus sp. Cryptoblepharus sp. NGEANGES *Emoia caeruleocauda (sight record 1/97) Gehyra oceanica NGERCHEU Hemiphyllodactylus cf. H. typus Lepidodactylus lugubris (4 km N of Beliliou; all specimens KUZ) Lepidodactylus moestus *Platymantis pelewensis Emoia caeruleocauda *Gehyra brevipalmata Dendrelaphis sp. *Gekko sp. *Hemidactylus frenatus NGEAUR *Lepidodactylus lugubris moestus Bufo marinus *Lepidodactylus *Emoia caeruleocauda Platymantis pelewensis sp. *Hypsilurus godeffroyi (?) *Eugongylus *Lamprolepis smaragdina Gehyra brevipalmata of. L. noctua Gehyra mutilata *Lipinia Gekko *Mabuya sp. sp. *Candoia carinata Hemidactylus frenatus Hemiphyllodactylus cf. H. typus *Dendrelaphis sp. *Ramphotyphlops acuticaudus Lepidodactylus lugubris Lepidodactylus moestus NGERDIS Carlia cf. C. fusca (off Palau Pacific Resort, Ngerekebesang; connected Cryptoblepharus sp. Emoia atrocostata to mainland by jetty) Emoia caeruleocauda Lepidodactylus lugubris *Eugongylus sp. *Emoia atrocostata (CAS 184016) Lamprolepis smaragdina *Lamprolepis smaragdina (CAS 184017) leptosoma Lipinia NGERDUAIS Lipinia cf. L. noctua Sphenomorphus scutatus (off SE Babeldaob, 100 m SW Garudawisei Point) Varanus cf. v. indicus Gehyra brevipalmata Candoia carinata Lepidodactylus moestus Dendrelaphis sp. Gekko sp. *Pelamis platurus (CAS 122587) Candoia carinata *Ramphotyphlops acuticaudus (CM 29087) Dendrelaphis sp. Ramphotyphlops braminus *Laticauda colubrina 1999] HERPETOLOGICAL MONOGRAPHS 79

NGERECHONG ISLAND 12 (2.3 km S of Mecherchar) *Lamprolepis smaragdina mutilata Gehyra ISLAND 13 NGERECHUR *Candoia carinata (ca. 2 km off N point Babeldaob) *Laticauda colubrina Platymantis pelewensis ISLAND 16 Gehyra oceanica Gekko sp. *Lepidodactylus paurolepis Lepidodactylus moestus *Eugongylus sp. Emoia caeruleocauda *Lamprolepis smaragdina Emoia jakati ISLAND 34 Lamprolepis smaragdina Candoia carinata *Lepidodactylus paurolepis Laticauda colubrina NGERUKTABEL NGEREKEBESANG Platymantis pelewensis Bufo marinus Gehyra brevipalmata cf. H. Platymantis pelewensis Hemiphyllodactylus typus Gehyra brevipalmata Lepidodactylus moestus Gehyra mutilata Lepidodactylus sp. Gekko sp. Emoia caeruleocauda Hemidactylus frenatus Eugongylus sp. Hemiphyllodactylus cf. H. typus Lamprolepis smaragdina scutatus Lepidodactylus lugubris Sphenomorphus Sphenomorphus sp. nov. Lepidodactylus moestus Carlia cf. C. fusca Candoia carinata Cryptoblepharus sp. Dendrelaphis sp. Emoia atrocostata Emoia caeruleocauda NGERUR Lamprolepis smaragdina (0.8 km W of Ngerekebesang) Lipinia cf. L. noctua Gehyra oceanica carinata Candoia Lepidodactylus lugubris *Cerberus rynchops Lepidodactylus moestus Dendrelaphis sp. Emoia caeruleocauda Lamprolepis smaragdina NGERMALK Sphenomorphus scutatus (between Oreor and Malakal, connected by causeway) NGESBOKEL Hemidactylus frenatus (Ngeremeduu Bay, west coast of Babeldaob) Lepidodactylus lugubris Lepidodactylus lugubris Lepidodactylus moestus Lepidodactylus moestus Carlia cf. C. fusca Emoia impar Lamprolepis smaragdina Lamprolepis smaragdina Sphenomorphus sp. nov. Sphenomorphus scutatus NGETMEDUCH (ca. 0.2 km north of Oreor, connected by causeway) NGERUCHUBTANG Platymantis pelewensis (4 km SSW of Ngerechong) Gehyra mutilata *Emoia caeruleocauda Lepidodactylus lugubris Lepidodactylus moestus NGERUKEUID ISLAND GROUP Carlia cf. C. fusca NGERUKEUID ISLAND (Island 24) Cryptoblepharus sp. *Platymantis pelewensis Emoia atrocostata *Gekko sp. Emoia caeruleocauda *Eugongylus sp. Lamprolepis smaragdina *Lamprolepis smaragdina Candoia carinata *Dendrelaphis sp. Ramphotyphlops braminus *Laticauda colubrina OREOR ISLAND 4 Bufo marinus *Lepidodactylus moestus Platymantis pelewensis 80 HERPETOLOGICAL MONOGRAPHS [No.No. 13

Gehyra brevipalmata Gekko sp. Gehyra mutilata Hemiphyllodactylus cf. H. typus *Gehyra oceanica (CAS-SU 23747, 122393) Nactus sp. Gekko sp. Emoia atrocostata Hemidactylus frenatus Emoia caeruleocauda Hemiphyllodactylus cf. H. typus Eugongylus sp. Lepidodactylus lugubris Lamprolepis smaragdina Lepidodactylus moestus Sphenomorphus scutatus Anolis carolinensis Dendrelaphis sp. Carlia cf. C. fusca Cryptoblepharus sp. ULONG Emoia atrocostata Platymantis pelewensis Emoia caeruleocauda Gehyra brevipalmata Eugongylus sp. Gekko sp. Lamprolepis smaragdina Lepidodactylus sp. Lipinia cf. L. noctua Nactus sp. *Lipinia leptosoma Emoia caeruleocauda Mabuya sp. Eugongylus sp. Sphenomorphus sp. nov. Lamprolepis smaragdina Sphenomorphus scutatus Lipinia cf. L. noctua *Varanus cf. v. indicus (CAS 154150) Sphenomorphus scutatus Candoia carinata Dendrelaphis sp. Cerberus rynchops Laticauda colubrina Dendrelaphis sp. Laticauda colubrina SOUTHWEST ISLANDS Ramphotyphlops acuticaudus FANA ISLAND Ramphotyphlops braminus Crocodylus porosus *Gehyra sp. *Eretmochelys imbricata (CAS 154194) *Perochirus sp. *Lamprolepis smaragdina PKULAKLIM (ORRAK) HELEN (Helen Reef) (ca. 0.6 km SW of Ngerduais) Bufo marinus *Emoia jakati Gehyra brevipalmata MERIR ISLAND Emoia caeruleocauda Lamprolepis smaragdina *Nactus cf. N. pelagicus Candoia carinata *Emoia impar Laticauda colubrina *Lamprolepis smaragdina PULO ANNA TOIRACHOLODOCHOEL (island name from Brown and Fehlmann, 1958:5 as *Gehyra sp. "a small island at the north end of Malakal") *Lepidodactylus lugubris moestus *Lipinia leptosoma *Lepidodactylus *Emoia sp. UCHULAMERADEL *Emoia impar (southernmost of eastern islands, mouth of *Lamprolepis smaragdina *Lipinia cf. L. noctua Ngeremeduu Bay) *Mabuya sp. Lepidodactylus moestus SONSOROL ISLAND UCHULANGAS *Gehyra sp. to Oreor and (connected Ngermalk by causeway) *Emoia sp. Hemidactylus frenatus *Emoia impar Lepidodactylus lugubris *Lamprolepis smaragdina Lepidodactylus moestus *Mabuya sp. Carlia cf. C. fusca Emoia atrocostata TOBI ISLAND *Gehyra sp. ULEBSECHEL *Gehyra mutilata Platymantis pelewensis *Emoia jakati Gehyra brevipalmata *Lamprolepis smaragdina