Title Natural History of a Madagascan Gecko Author(S) Ikeuchi, Isami; Mori

Title Natural History of a Madagascan Gecko

Author(s) Ikeuchi, Isami; Mori, Akira

Citation Current Herpetology (2014), 33(2): 161-170

Issue Date 2014-08

URL http://hdl.handle.net/2433/198818

Type Journal Article

Textversion publisher

Kyoto University Natural History of a Madagascan Gecko Blaesodactylus ambonihazo in a Dry Deciduous Forest Author(s): Isami Ikeuchi and Akira Mori Source: Current Herpetology, 33(2):161-170. Published By: The Herpetological Society of Japan DOI: http://dx.doi.org/10.5358/hsj.33.161 URL: http://www.bioone.org/doi/full/10.5358/hsj.33.161

BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published by nonprofit societies, associations, museums, institutions, and presses. Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use. Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries or rights and permissions requests should be directed to the individual publisher as copyright holder.

BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research libraries, and research funders in the common goal of maximizing access to critical research. Current Herpetology 33(2): 161–170, August 2014 doi 10.5358/hsj.33.161 © 2014 by The Herpetological Society of Japan

Natural History of a Madagascan Gecko Blaesodactylus ambonihazo in a Dry Deciduous Forest

IǈƵǑƿ IKEUCHI1,2 Ƶǃƺ AǁƿǇƵ MORI1*

1Department of Zoology, Graduate School of Science, Kyoto University, Sakyo, Kyoto 606–8502, JAPAN 2Present address: Takinocho 2–19–3–202, Nagaokakyo, Kyoto 617–0817, JAPAN

Abstract: Several aspects of natural history of Blaesodactylus ambonihazo, a gecko distributed in a dry forest of northwestern Madagascar, were investi- gated in the rainy and dry seasons between 2000 and 2006. There were no signiÖcant seUual siWe di÷erences in snout-vent length, and no seUual diJor- phisJ was detected either in head width or body Jass. FeJales ceased oogenesis during the rainy season and were recrudescent at the beginning of the dry season. They probably lay eggs in the Jiddle of the dry season. Hatchlings were observed at the beginning of the rainy season and probably reach adult siWe in the subseNuent dry season. At night geckos perched on tree trunks and buildings and eUhibited typical sit-and-wait foraging. During the day they retreated to shelters, Jainly crevices between buttress roots. The gecko was therJally passive to environJental teJperatures, showing a wide range of cloacal teJperatures 0 , but they selected relatively higher substrate teJperatures at low air teJperature, possibly for therJo- regulation. ecause Jultiple individuals were observed on single trees, hoJe ranges of the gecko were presuJably overlapping each other, and no obvious territorial behavior was observed. The absence of Jale-biased seUual diJor- phisJ, which suggests little Jale-Jale coJpetition for Jating, also supports the absence of territoriality of B. ambonihazo. oJparison with a syntopic diurnal gecko, Phelsuma kochi, which shows ecological characters siJilar to B. ambonihazo but has eUclusive hoJe ranges, Jay clarify ecological corre- lates associated with teJporal niche partitioning.

Key words: Gekkonidae; Blaesodactylus ambonihazo; Madagascar; Field obser- vation; Ankarafantsika National Park

IǃǉǇǄƺǊƹǉƿǄǃ 2013). Gekkonidae, the largest family of Gekkota, is comprised of approximately 930 Gekkota is a large monophyletic group of species and is mainly distributed in the Old Squamata, which accounts for approximately World (Bauer, 2013). Because of their high one-fourth of all species of lizards (Bauer, species richness, their ecology and life history are likely to be diverse. Nonetheless, informa- * Corresponding author. Tel: +81–75–753–4075; tion on their natural history based on quanti- Fax: +81–75–753–4075; S@SHUDëDKCNARDQU@SHNMRHRK@BJHMFENQSGD E-mail address: [email protected] majority of the species, impeding comparative 162 Current Herpetol. 33(2) 2014

DW@LHM@SHNMRNESGDHQDBNKNFHB@KCHUDQRHëB@SHNM MƵǉƻǇƿƵǂǈ Ƶǃƺ MƻǉƾǄƺǈ Madagascar is one of the global hotspots that has a great species diversity of endemic The study was conducted in a tropical reptiles (Myers et al., 2000; Goodman and d e c i d u o u s f o r e s t a t A m p i j o r o a , A n k a r a f a n t s i k a Benstead, 2005). Gekkos are not exceptional: National Park (16q19'S, 46q49'E) from October more than 10% of gekkonids, most of which 2000 to January 2001, from June to July 2003, are endemic, occur in Madagascar (Glaw and from October 2003 to June 2004, and from Vences, 2007). However, in spite of their tre- November 2005 to February 2006. The dry mendous diversity and high endemism, eco- season usually lasts from May to October, and logical and behavioral studies on wild geckos the rainy season starts gradually in November in Madagascar based on quantitative data are (See Fig. 1 in Ikeuchi et al., 2012). The study extremely limited (e.g., Vences et al., 2004; area included a camping site (approximately Ikeuchi et al., 2005, 2012). Appropriate knowl- 250×100 m), Jardin Botanique A (JBA: edge of natural history of each species is approximately 550×450 m), trails from the essential to understand the patterns of their camping site to JBA (approximately 1200 m), adaptive radiation. In addition, because origi- and trails around Lake Ravelobe (approxi- nal forests of Madagascar, which are major mately 5 km). The vegetation consisted of a G@AHS@SR NE SGD FDBJNR BTQQDMSKX RTðDQ EQNL deciduous canopy of 10–15 m high and fairly RDUDQDEQ@FLDMS@SHNMAXëQDR HKKDF@KKNFFHMF sparse understory. and deforestation for agriculture (Ganzhorn, We walked through the study area at least #TëKR RTBGJMNVKDCFDHRBQHSHB@K twice per week in the day (0700–1800 h) and SNDU@KT@SDSGDHMìTDMBDNECDENQDRS@SHNMNM at night (2000–0400 h), searching for geckos their survival and establish conservation mea- on trees and buildings up to ca. 4 m high. sures and management strategies of them. When a gecko was located, we attempted to In the present study we examined several collect it with a noose or by hand. We imme- aspects of natural history of a gecko, diately measured its cloacal temperature Blaesodactylus ambonihazo, which is known (CT), air temperature at 1 m above the ground only from a dry forest of northwestern (AT), and substrate temperature at the exact Madagascar (Bauer et al., 2011). The taxo- spot where the gecko was perching (ST) using nomic history of geckos in Blaesodactylus a quick-reading thermistor. When geckos were was recently reviewed by Greenbaum et al. perching in high places (approximately more 3GDXBNMëQLDCSGDRHRSDQFQNTOQDK@- than 2 m), we measured substrate temperature tionship and reciprocal monophyly of the at the spot as close as possible. Then we mea- Malagasy Blaesodactylus and the mainland sured its snout-vent length (SVL), head width African Homopholis. Greenbaum et al. (2007) (HW), and body mass (BM). Sex was deter- also reported a specimen that is genetically mined as male if a gecko had a swollen tail highly divergent from all other recognized base (i.e., hemipeneal bulge). Gravidity of species of Blaesodactylus. Subsequently, Bauer EDL@KDR V@R BNMëQLDC AX NARDQU@SHNM @MC et al. (2011) made detailed examinations of palpation of eggs in the abdomen. Individuals this form both morphologically and geneti- with SVL smaller than that of males and cally and described B. ambonihazo as the gravid females were considered juveniles (see fourth member of this genus from a dry forest Results). After measurement, the individuals in Ankarafantsika National Park. In the pres- were marked by toe clipping and were released ent study we describe its size composition, at the site where they were found. We recorded female reproductive traits, thermal character- habitat type (tree or building) for each gecko’s istics, spatial usage, and foraging mode based perch site and whether it was exposed (exposed NM@KNMF SDQLëDKCRSTCX or in shelter). If the perch site was a tree, substrate characters (trunk, branch, crevice IKEUCHI & MORI—NATURAL HISTORY OF BLAESODACTYLUS AMBONIHAZO 163 between trunk, crevice between buttress TRHMF*DMC@KK¥RBNQQDK@SHNMBNDîBHDMS roots, or hole), perch height (PH), perch diameter (PD), and diameter at the breast RƻǈǊǂǉǈ height (DBH; 1.2 m high) of the tree were recorded. If a marked gecko was re-sighted on Size composition and female reproductive trait @SQDDCHðDQDMSEQNLSGDOQDUHNTRNMD SGDCHR- Because the minimum SVL of males and tance between the trees was measured. gravid females were 66.6 mm and 85.3 mm, To examine foraging mode, focal animal respectively, individuals whose SVL was less observations were conducted for six geckos, than 66.6 mm were considered juveniles. The which were perching on trees in forest, at two smallest individuals (SVL=44.0 and night (2100–0100 h) from March to June 44.2 mm) each had an obvious umbilicus and 2004 using an infrared video camera. We VDQDBNMRHCDQDC@RG@SBGKHMFR 2DWT@KCHðDQ- continuously observed their behavior at least ence in SVL was not detected (Mann-Whitney for 17 min standing at least 5 m away from U-test; U=3734.5, P=0.70; Table 1). Head trees where they were perching. An individual VHCSG@MC!,VDQDANSGRHFMHëB@MSKXBNQQD- V@RBK@RRHëDC@RDHSGDQRS@SHNM@QXNQLNUHMF lated with SVL (log transformed data; HW, Stationary included postures and actions in r2=0.95, P<0.0001, BM, r2=0.98, P<0.0001), which the gecko remained in the same posi- @MC SGDQD VDQD MN RHFMHëB@MS CHðDQDMBDR tion (e.g., remaining motionless, moving its bet ween sexes both i n H W a nd BM (A NCOVA , head, jaws, tongue, limbs, or tail, and turning log transformed data; HW, elevation, df=1, around by 180 degrees). Moving included 170, F=0.85, P=0.36, slope, df=1, 170, actions that accompany the change of loca- F=0.90, P=0.35; BM, elevation, df=1, 170, tion of the gecko (e.g., crawling, walking, F=0.00001, P=0.99, slope, df=1, 170, F= running, jumping, and jerking forwards). 0.00001, P=0.99). Small individuals (SVL< When the gecko changed its behavior from stationary to moving, or vice versa, the time TƵƸǂƻ 1. Morphological measurements of Blaesodactylus ambonihazo. SVL: snout-vent length, was recorded. To facilitate a comparison with HW: head width, BM: body mass. Mean±SD are previous studies (e.g., Cooper, 2005; Bauer, shown. Figures in parentheses are sample size and 2007; Perry, 2007), we calculated moves per range. minute (MPM) and percent of time spent moving (PTM) (Huey and Pianka, 1981). Sex SVL (mm) HW (mm) BM (g) An ANCOVA was used to examine sexual Male 96.9±9.94 19.6±2.04 22.5±7.38 CHðDQDMBDR NE ANCX RHYD 25+ @R BNU@QH@SD (92, 66.6–112) (92, 14.3–23) (91, 6.8–40.4) and thermal usage (CT as covariate) after SDRSR ENQ RHFMHëB@MBD NE SGD QDFQDRRHNM 3GD Female 97.2±10.6 19.5±2.14 23.1±7.58 CHðDQDMBDR ADSVDDM "3 @MC BNQQDRONMCHMF (84, 70.5–115) (84, 14.8–23.5) (83, 6.7–36.4) AT or ST were examined at night and in the day separately by paired t-test. Mann-Whitney U-test was used to analyze spatial usage (PH, /# @MC #!' CHðDQDMBDR ADSVDDM C@X @MC MHFGS 2DWT@KCHðDQDMBDRNE25+ CHTQM@K@MC nocturnal spatial usages, and moving distance were also tested using Mann-Whitney U-test. The correlation between DBH and the number of individuals sharing the same tree and Fƿƽ. 1. The seasonal change of the number of the correlation between maximum moving Blaesodactylus ambonihazo BK@RRHëDC AX RMNTS CHRS@MBD@MCSGDC@XREQNLSGDëQRSSNK@RS vent length. Data in 2000–2001, 2003–2004, and observations of an individual were analyzed 2005–2006 were combined. 164 Current Herpetol. 33(2) 2014

and it increased from the beginning of the dry season (Fig. 2). All but one mature females (95%) were gravid in June and July. Twenty-three out of 26 (88%) gravid females had two eggs in their abdomen, and the remaining three had only one egg.

Thermal usage SMHFGS"3NEFDBJNRV@RRHFMHëB@MSKX higher than corresponding AT and ST (paired t-test; AT, t=-12.86, P<0.0001; ST, t=-9.54, P<0.0001; Fig. 3A). However, at low ambient temperatures (less than 22 C) CT of geckos remained higher than AT, whereas CT were Fƿƽ. 2. The percentages of gravid (solid column) close to or lower than ST. During the day, CT and non-gravid females (open column) of V@RRHFMHëB@MSKXGHFGDQSG@MSGDBNQQDRONMC- Blaesodactylus ambonihazo in each two-month ing ST, but lower than the corresponding AT period. The number of individuals is denoted (paired t-test; AT, t=4.70, P<0.0001; ST, above each column. Data in 2000–2001, 2003– t=-5.97, P<0.0001; Fig. 3B). Cloacal temper- 2004, and 2005–2006 were combined. @STQDV@RRHFMHëB@MSKXBNQQDK@SDCVHSG 3@MC ST both at night and in the day (log trans- 50.0 mm) were observed only at the beginning formed data; night, AT, r2=0.92, P<0.0001, of the rainy season (October to November) ST, r2=0.95, P<0.0001; day, AT, r2=0.86, (Fig. 1). The cohort gradually shifted to a P<0.0001, ST, r2=0.89, P<0.0001), and there larger class, and no juveniles (SVL<66.6 mm) VDQDMNRHFMHëB@MSRDWT@KCHðDQDMBDRHM"3@S were observed at the beginning of the dry a given AT or ST both at night and in the day season (April and May). (ANCOVA, log transformed data; night, AT, The percentage of gravid individuals among elevation, df=1, 45, F=2.66, P=0.11, slope, mature females (SVLt85.3 mm) was quite df=1, 45, F=2.5, P=0.12, ST, elevation, df=1, low during the rainy season (November-March), 45, F=0.67, P=0.42, slope, df=1, 45, F=0.72,

Fƿƽ. 3. Relationships between cloacal temperature of Blaesodactylus ambonihazo with air temperature (open circle) and substrate temperature (solid circle). The lines where cloacal temperature is equal to ambient temperature are also shown. A: night, B: day. IKEUCHI & MORI—NATURAL HISTORY OF BLAESODACTYLUS AMBONIHAZO 165

P=0.40; day, AT, elevation, df=1, 52, F=2.96, P=0.09, slope, df=1, 52, F=2.69, P=0.11, ST, elevation, df=1, 52, F=0.38, P=0.54, slope, df=1, 52, F=0.37, P=0.54).

Perch sites We captured 58 and 128 geckos at night and in the day, respectively. At night all but two individuals were observed in exposed condi- SHNMR @MCHMSGDC@X@KKATSëUDFDBJNRVDQDHM shelters (Table 2). Among geckos observed on trees, 91.4% of the individuals perched on trunks at night, and 71.0% used crevices between buttress roots in the day. Fƿƽ. 4. Relationship between diameter of trees at Perch height, PD, and DBH in the day were breast height (1.2 m height) and the number of RHFMHëB@MSKXK@QFDQSG@MSGNRD@SMHFGS,@MM Blaesodactylus ambonihazo observed on the tree. Whitney U-test; PH, U=1141.5, P<0.0001; PD, U=1189, P<0.005; DBH U=1184, P< day, PH, U=1343.5, P=0.70, PD, U=1231.5, 3@AKD 2DWT@KCHðDQDMBDRHM/' /# P=0.27, DBH, U=1359.5, P=0.78, male, n= and DBH were not detected either at night or 52, female, n=54). in the day (Mann-Whitney U-test; night, PH, &DBJNRODQBGDCNMCHðDQDMSSQDDR @MC U=130, P=0.83, PD, U=133.5, P=0.93, DBH, of them (29.5%) were used by at least two indi- U=133, P=0.91, male, n=16, female, n=17; viduals (at most eight individuals). At least two males, two females, and both sexes were TƵƸǂƻ 2. Perching site of Blaesodactylus ambo- collected on 12 trees (12.6%), 11 trees (11.6%), nihazo at night and in the day. and 21 trees (22.1%), respectively. A positive correlation between DBH and the number of Condition Perching site Night Day individuals using the same tree was detected of gecko (Kendall; t=0.32, P<0.0001; Fig. 4). On 13 Exposed Trunk 32 5 trees (13.8%) two or three individuals were Branch 1 0 simultaneously perching at single observations Building (wall, column, girder) 23 0 (19 times). The compositions of the individuals using the same tree were two males (two In shelter Between buttress roots 1 88 Between trunks 1 26 cases), two females (one case), a male and a Hole 0 5 female (10 cases), and an adult and juveniles Building (under cover) 0 4 or unknown sex (6 cases).

TƵƸǂƻ 3. Measurements of trees perched upon Moving distance by Blaesodactylus ambonihazo at night and in the Twenty-four individuals were sighted at least day. PH: perch height, PD: perch diameter, DBH: twice (maximum four times) on trees at an diameter at the breast height. Mean±SD are shown. interval of more than a month. No correla- Figures in parentheses are sample size and range. tion was found between maximum moving Time PH (m) PD (cm) DBH (cm) CHRS@MBD@MCSGDC@XREQNLSGDëQRSSNK@RS observations of an individual (46–918 days) Night 1.09±0.76 49.1±48.5 48.2±48.4 (35, 0.2–3.8) (35, 0.2–150) (35, 0.2–150) (Kendall; t=0.15, P=0.32). Seventeen (10 males, 6 females, and a juvenile, 70.8%) of the Day 1.69±0.89 85.6±50.5 90.8±51.4 re-sighted geckos were found on the same (122, 0.1–4.3) (117, 3–250) (119, 0–210) SQDDRVGDQDSGDXVDQDëQRSB@OSTQDC 3VNNE 166 Current Herpetol. 33(2) 2014 seven individuals that had moved to another it is obvious that this gecko does not greatly tree, returned to the tree they initially used. reduce its activity in the harsh, dry season, The average moving distances±SD of males when virtually no rainfall is available (Ikeuchi and females were 7.42±14.65 m (n=13, range et al., 2012). 0–43.3) and 7.15±10.16 m (n=10, range 0–27.3), We observed the gecko both during the QDRODBSHUDKXMNRDWT@KCHðDQDMBDV@RCDSDBSDC day and at night. However, more than 95% of (Mann-Whitney U-test; U=59.0, P=0.65). geckos sighted in the day were partially or entirely hiding their bodies in shelter and Foraging mode remained motionless. On the other hand, all Six geckos perching on tree trunks were but two geckos sighted at night were perching observed for 17.0 to 50.1 min. Locomotary on trees or buildings while exposing their entire movement was rarely observed. Mean of bodies. These observations indicate that B. MPM and PTM±SD were 0.21±0.21 (range, ambonihazo is a primarily nocturnal gecko, 0.04–0.58) and 0.86±0.96 (range, 0.07–2.27), although some diurnal activity of congeneric respectively. species as well as B. ambonihazo has been described or suggested (Glaw and Vences, DƿǈƹǊǈǈƿǄǃ 2007; Bauer et al., 2011). Detailed focal animal observations would clarify the biologi- Although the climate of the study site has cal role of the possible diurnal activity of these distinct dry and rainy seasons, B. amboni- basically nocturnal geckos. As implied by its hazo seems to be active throughout the year. RODBHëB M@LD ambonihazo, which means The seasonal changes in the frequency of “on tree”, none out of the 186 individuals gravid females and in the size distribution of VDQDENTMCNMSGDFQNTMC BNMëQLHMFHSR the gecko suggest that females cease oogenesis tree-dwelling habit (Bauer et al., 2011). during the rainy season (October to March), Like other nocturnal geckos (Autumn et recrudesce at the beginning of the dry season, al., 1994, 1997), B. ambonihazo was active and probably lay eggs in the middle of the dry over a wide-range of environment tempera- season (June to August). Based on the incu- tures (15–30 C) at night. This gecko is consid- bation period (112–126 days) of B. sakalava ered thermally passive to ambient tempera- reported by Pes (2005), eggs of B. amboni- tures but seems to select a little higher ST at hazo may hatch at the beginning of the low AT for thermoregulation, which is reported subsequent rainy season, which is consistent in some other crepuscular or nocturnal geckos with the occurrence of hatchlings in October- (Rock et al., 2000; Vitt et al., 2007). During November. These reproductive traits of B. the day geckos retreated to narrow shelters, ambonihazo@QDCHðDQDMSEQNLSGNRDNESGD mainly crevices between buttress roots. Because other syntopic lizards, whose oogenesis and the substrate temperatures in shelters are oviposition occur only in the rainy season lower than outside air temperatures in the day, (Oplurus cuvieri; Randriamahazo and Mori, the gecko might stay in cooler sites to thermo- 2001) or continue almost annually with a regulate like other gecko species (Kearney and higher reproductive activity in the rainy Predavec, 2000; Shah et al., 2004; Vitt et season (Phelsuma kochi; Ikeuchi et al., 2005), al., 2007), although we do not have any data which has been also reported in a gecko in for preferred body temperatures of B. ambo- another dry forest of Madagascar (Vences et nihazo. Otherwise, shelters might protect B. al., 2004). Hatchlings of B. ambonihazo, ambonihazo against desiccation, because observed at the beginning of the rainy season, they can provide suitable hydric conditions seem to reach adult size during the subsequent (Schwarzkopf and Alford, 1996), or the gecko dry season. Overall, although our survey may simply retreat for predator avoidance in DðNQSV@RMNSBNMRS@MSSGQNTFGNTSSGDXD@Q the day. IKEUCHI & MORI—NATURAL HISTORY OF BLAESODACTYLUS AMBONIHAZO 167

Lizards have been categorized into a dichot- this gecko may share trees with other individu- omous paradigm, either sit-and-wait or widely als, even other males. Moreover, male-biased foragers (MPM values of d1.5 and PTM val- sexual dimorphism, which is generally found ues of <10 indicate sit-and-wait, and PTM in lizards in which males have exclusive home of >10 indicates widely foraging, Huey and ranges as a result of sexual selection through Pianka, 1981; Cooper, 2005; Perry, 2007), male-male competition for mating (Ikeuchi et although the presence of another mode has al., 2005; Hibbitts et al., 2007), is completely been documented (Butler, 2005) and contro- absent in B. ambonihazo. This also supports versy still remains. Sit-and-wait foragers detect the possibility of overlapping home ranges moving prey visually while immobile, whereas between males of this gecko. widely forager moves around to detect hidden Although our data suggest the stable activ- prey using chemical cues (Cooper, 2005). ity of B. ambonihazo throughout the year, it is Bauer (2007) suggested that a majority of gek- likely that the above ecological and behavioral kotan species are either sit-and-wait foragers traits, such as home range size and foraging to a large extent (Cooper et al., 1999; Werner mode, may change depending on seasons. and Chou, 2002; Ikeuchi et al., 2005; Aowphol Small sample size of our study precludes con- et al., 2006), or intermediate or mixed forag- ëQL@SHUDRD@RNM@KBNLO@QHRNMRNERTBGSQ@HSR ers using the slow cruise foraging or serial %TQSGDQ RSTCHDR ENBTRHMF NM RD@RNM@K CHðDQ- ambushing (Werner et al., 1997, 2004). ences in ecological and behavioral features Although sample size is limited, B. ambonihazo are necessary to consider appropriate conser- is considered a typical sit-and-wait forager, vation measures. In addition, it should be spending most of the time perching on the noted that B. ambonihazo NESDM TRDC @QSHë- lower part of a tree trunk. cial buildings as well as natural forest trees as Generally, spacing pattern of lizards is perching site at night. This suggests that the @ðDBSDCAXSGD@U@HK@AHKHSXNEQDRNTQBDR%NW gecko may be tolerant, to some extent, to and Shipman, 2003). The number of B. @QSHëBH@K LNCHëB@SHNM NE DMUHQNMLDMSR AX ambonihazo increased on old large trees, such fragmentation of forest. Detailed comparisons as Albizia sp., Ceiba pentandra, and Hura of life history traits, such as growth and survi- crepitans, which would provide more spatial UNQRGHO ADSVDDMM@STQ@K@MC@QSHëBH@KDMUH- resources as shelters (buttress roots and holes) ronments are important to evaluate ecological and foraging sites (Fitzgerald et al., 2002; HLO@BSRNE@QSHëBH@KLNCHëB@SHNMNEDMUHQNM- Reaney and Whiting, 2003). Localized distri- ments bution of resources on large trees might result In contrast to B. ambonihazo, Phelsuma in congregative and extended occupation by kochi, a syntopic diurnal gecko in the this gecko of these trees. At night geckos leave Ampijoroa forest, shows exclusive home their shelters for foraging sites, therefore, their ranges within males, presumably because the home ranges presumably overlap each other dependence on visual modality in diurnal although our data on home range size is insuf- activity allows them to survey a greater dis- ëBHDMS 3GDRO@SH@KCHRSQHATSHNMNEFDBJNGNLD tance and to defend their home ranges against ranges hitherto reported varies among species other males (Ikeuchi et al., 2005). The genus although available information is quite Phelsuma, a speciose diurnal group of geckos, limited. Some species have home ranges that is inferred to be derived from nocturnal overlap both intra- and inter sexually (Semenov ancestors (Taniguchi et al., 1999; Han et al., and Borkin, 1992; Gruber and Henle, 2006; 2004; Nagy et al., 2012). Comparative study Johnston and Bouskila, 2007), whereas others of ecological and behavioral traits of B. are exclusive only between males (Ikeuchi et ambonihazo and P. kochi, which have similar al., 2005; Hibbitts et al., 2007). Blaesodactylus body size, microhabitat (tree-dwelling), and ambonihazo may have a former type because possibly diet (insectivorous) (Ikeuchi et al., 168 Current Herpetol. 33(2) 2014

2005, unpublished data), would help to clarify 1994. Low cost of locomotion increases perfor- features induced by temporal niche partition- mance at low temperature in a nocturnal lizard. ing, which allow their co-existence. Moreover, Physiological Zoology 67: 238–262. to fully understand the ecological radiation of BƵǊƻǇ, A. M. 2007. The foraging biology of the geckos in the Ampijoroa forest, which harbors Gekkota: life in the middle. p. 371–404. In: S. at least 12 species of geckos (Mori et al., M. Reilly, L. B. McBrayer, and D. B. Miles (eds.), 2006), fundamental natural history studies Lizard Ecology. Cambridge University Press, are essential and should be greatly encour- Cambridge. aged. Such research would provide clues to BƵǊƻǇ, A. M. 2013. Geckos. The Animal Answer unravel the evolutionary mechanisms of the Guide. The Johns Hopkins University Press, DMNQLNTR CHUDQRHëB@SHNM HM FDBJNR NM SGD Baltimore. island of Madagascar. BƵǊƻǇ, A. M., GǂƵǌ, F., GƻƾǇƿǃƽ, P.-S., Ƶǃƺ Vƻǃƹƻǈ, M. 2011. New species of Blaesodactylus AƹǁǃǄǌǂƻƺƽƻǑƻǃǉǈ (Squamata: Gekkonidae) from Ankarafantsika National Park in north-western Madagascar. We are grateful to F. Rakotondraparany, H. Zootaxa 2942: 57–68. 1@JNSNL@M@M@ @MCSGDRS@ðNEKD#lO@QSLDMS BǊǉǂƻǇ, M. A. 2005. Foraging mode of the CD !HNKNFHD MHL@KD %@BTKSl CDR 2BHDMBDR chameleon, Bradypodion pumilum: a challenge 4MHUDQRHSl C¥ MS@M@M@QHUN @MC ,@C@F@RB@Q to the sit-and-wait versus active forager para- National Parks, Ankarafantsika for their digm? Biological Journal of the Linnean cooperation in conducting this research. The Society 84: 797–808. ëQRS@TSGNQSG@MJR3 ,HX@Y@JHENQKHUHMF CǄǄǅƻǇ, JǇ., W. E. 2005. The foraging mode con- supports in Madagascar. Early draft of the troversy: both continuous variation and cluster- manuscript was improved through suggestions ing of foraging movements occur. Journal of NE2 !@QQHAD@T 3GDëDKCRTQUDXV@RO@QSH@KKX Zoology 267: 179–190. supported by a Grant-in-Aid for International CǄǄǅƻǇ, JǇ., W. E., Wƾƿǉƿǃƽ, M. J., Wǎǁ, J. H. V., 2BHDMSHëB 1DRD@QBG ! %HDKC 1DRD@QBG MN Ƶǃƺ MǄǊǉǄǃ, P. ǂƻ. F. N. 1999. Movement- 11691183) from the Japan Ministry of and attack-based indices of foraging mode and Education, Culture, Sports, Science, and ambush foraging in some gekkonid and agamine Technology (MEXT) and the Kyoto University lizards from southern Africa. Amphibia-Reptilia Foundation in 2003. The data analyses were 20: 391–399. partially supported by a Grant-in-Aid for DǊƼƿǂǈ, J. M. 2003. Remaining forest cover. p. (MSDQM@SHNM@K 2BHDMSHëB 1DRD@QBG ! MN 88–96. In: S. M. Goodman and J. P. Benstead 24405008) from MEXT. (eds.), The Natural History of Madagascar. The University of Chicago Press, Chicago. LƿǉƻǇƵǉǊǇƻ Cƿǉƻƺ FƿǉǏƽƻǇƵǂƺ, M., Sƾƿǃƻ, R., Ƶǃƺ LƻǑƹǁƻǇǉ, F. 2002. Radiotelemetric study of habitat use by AǄǌǅƾǄǂ, A., TƾƿǇƵǁƾǊǅǉ, K., NƵƸƾƿǉƵƸƾƵǉƵ, the arboreal snake Hoplocephalus stephensii J., Ƶǃƺ VǄǇƿǈ, H. K. 2006. Foraging ecology of (Elapidae) in eastern Australia. Copeia 2002: the Tokay gecko, Gekko gecko in a residential 321–332. area in Thailand. Amphibia-Reptilia 27: 491– FǄǍ, S. F. Ƶǃƺ SƾƿǅǑƵǃ, P. A. 2003. Social 503. behavior at high and low elevations. p. 310–355. AǊǉǊǑǃ, K., FƵǇǂƻǎ, C. T., EǑǈƾǌƿǂǂƻǇ, M., In: S. F. Fox, J. K. McCoy, and T. A. Baird (eds.), Ƶǃƺ FǊǂǂ, R. J. 1997. Low cost of locomotion Lizard Social Behavior. The Johns Hopkins in the banded gecko: a test of the nocturnality University Press, Baltimore. hypothesis. Physiological Zoology 70: 660– GƵǃǏƾǄǇǃ, J. U., LǄǌǇǎ II, P. P., SƹƾƵǉǏ, G. E., 669. Ƶǃƺ SǄǑǑƻǇ, S. 2001. The biodiversity of AǊǉǊǑǃ, K., WƻƿǃǈǉƵƿǃ, R. B., Ƶǃƺ FǊǂǂ, R. J. Madagascar: one of the world’s hottest hotspots IKEUCHI & MORI—NATURAL HISTORY OF BLAESODACTYLUS AMBONIHAZO 169

on its way out. Oryx 35: 346–348. the temperate gecko Christinus marmoratus. GǂƵǌ, F. Ƶǃƺ Vƻǃƹƻǈ, M. 2007. A Field Guide to Ecology 81: 2984–2996. the Amphibians and Reptiles of Madagascar. MǄǇƿ, A., IǁƻǊƹƾƿ, I., Ƶǃƺ HƵǈƻƽƵǌƵ, M. 2006. 3rd Edition. Vences & Glaw Verlag, Cologne. Herpetofauna of Ampijoroa, Ankarafantsika GǄǄƺǑƵǃ, S. M. Ƶǃƺ BƻǃǈǉƻƵƺ, J. P. 2005. Strict Nature Reserve, a dry forest in northwest- Updated estimates of biotic diversity and ern Madagascar. Herpetological Natural History endemism for Madagascar. Oryx 39: 73–77. 10: 31–60. GǇƻƻǃƸƵǊǑ, E., BƵǊƻǇ, A. M., Ƶǃƺ JƵƹǁǑƵǃ, MǎƻǇǈ, N., MƿǉǉƻǇǑƻƿƻǇ, R. A., MƿǉǉƻǇǑƻƿƻǇ, T. R. 2007. Homopholis and Blaesodactylus C. G., ƺƵ FǄǃǈƻƹƵ, G. A. B., Ƶǃƺ Kƻǃǉ, J. (Squamata: Gekkonidae) revisited: new insights 2000. Biodiversity hotspots for conservation from a molecular phylogeny. African Journal of priorities. Nature 403: 853–858. Herpetology 56: 101–114. NƵƽǎ, Z. T., SǄǃƻǉ, G., GǂƵǌ, F., Ƶǃƺ Vƻǃƹƻǈ, GǇǊƸƻǇ, B. Ƶǃƺ Hƻǃǂƻ * 3GDDðDBSNE M. 2012. First large-scale DNA barcoding as- movement on survival—a new method with an sessment of reptiles in the biodiversity hotspot application in the arboreal gecko Gehyra of Madagascar, based on newly designed COI variegata. p. 43–46. In: M. Vences, J. Köhler, T. primers. PLoS ONE 7: e34506. Ziegler, and W. Böhme (eds.), Herpetologia PƻǇǇǎ, G. 2007. Movement patterns in lizards: Bonnensis II. Proceedings of the 13th Congress measurement, modality, and behavioral corre- of the Societas Europaea Herpetologica. Societas lates. p. 13–48. In: S. M. Reilly, L. B. McBrayer, Europaea Herpetologica, Bonn. and D. B. Miles (eds.), Lizard Ecology. HƵǃ, D., ZƾǄǊ, K., Ƶǃƺ BƵǊƻǇ, A. M. 2004. Cambridge University Press, Cambridge. Phylogenetic relationships among gekkotan Pƻǈ, T. 2005. Births and hatchings. Plzen Czech lizards inferred from C-mos nuclear DNA Republic. EAZA News. 52: 17. RDPTDMBDR@MC@MDVBK@RRHëB@SHNMNESGD RƵǃƺǇƿƵǑƵƾƵǏǄ, H. J. A. R. Ƶǃƺ MǄǇƿ, A. 2001. Gekkota. Biological Journal of the Linnean Egg-laying activities and reproductive traits in Society 83: 353–368. females of Oplurus cuvieri cuvieri. Journal of HƿƸƸƿǉǉǈ, T. J., Wƾƿǉƿǃƽ, M. J., Ƶǃƺ SǉǊƵǇǉ-FǄǍ, Herpetology 35: 209–217. D. M. 2007. Shouting the odds: vocalization RƻƵǃƻǎ, L. T. Ƶǃƺ Wƾƿǉƿǃƽ, M. J. 2003. Picking signals status in a lizard. Behavioral Ecology a tree: habitat use by the tree agama, and Sociobiology 61: 1169–1176. Acanthocercus atricollis atricollis, in South HǊƻǎ, R. B. Ƶǃƺ PƿƵǃǁƵ, E. R. 1981. Ecological Africa. African Zoology 38: 273–278. consequences of foraging mode. Ecology 62: RǄƹǁ, J., AǃƺǇƻǌǈ, R. M., Ƶǃƺ CǇƻƻ, A. 2000. 991–999. $ðDBSR NE QDOQNCTBSHUD BNMCHSHNM RD@RNM @MC IǁƻǊƹƾƿ, I., HƵǈƻƽƵǌƵ, M., Ƶǃƺ MǄǇƿ, A. 2012. site on selected temperatures of a viviparous Characteristics of sleeping sites and timing of gecko. Physiological and Biochemical Zoology depature from them in a Madagascan diurnal 73: 344–355. gecko, Lygodactylus tolampyae. Current SƹƾǌƵǇǏǁǄǅƼ, L. Ƶǃƺ AǂƼǄǇƺ, R. A. 1996. Herpetology 31: 107–116. Desiccation and shelter-site use in a tropical IǁƻǊƹƾƿ, I., MǄǇƿ, A., Ƶǃƺ HƵǈƻƽƵǌƵ, M. 2005. amphibian: comparing toads with physical Natural history of Phelsuma madagascariensis models. Functional Ecology 10: 193–200. kochi from a dry forest in Madagascar. SƻǑƻǃǄǋ, D. V. Ƶǃƺ BǄǇǁƿǃ, L. J. 1992. On the Amphibia-Reptilia. 26: 475–483. ecology of Przewalsky’s gecko (Teratoscincus JǄƾǃǈǉǄǃ, G. Ƶǃƺ BǄǊǈǁƿǂƵ, A. 2007. Sexual przewalskii) in the Transaltai Gobi, Mongolia. dimorphism and ecology of the gecko, Asiatic Herpetological Research 4: 99–112. Ptyodactylus guttatus. Journal of Herpetology SƾƵƾ, B., Sƾƿǃƻ, R., HǊƺǈǄǃ, S., Ƶǃƺ KƻƵǇǃƻǎ, 41: 506–513. M. 2004. Experimental analysis of retreat-site KƻƵǇǃƻǎ, M. Ƶǃƺ PǇƻƺƵǋƻƹ, M. 2000. Do noc- selection by thick-tailed geckos Nephrurus turnal ectotherms thermoregulate? A study of milii. Austral Ecology 29: 547–552. 170 Current Herpetol. 33(2) 2014

TƵǃƿƽǊƹƾƿ, Y., HƿǈƵǉǄǑƿ, O., YǄǈƾƿƺƵ, M., Ƶǃƺ equatorial gecko Cnemaspis kendallii in TǄǁǊǃƵƽƵ, F. 1999. Evolution of visual pig- Singapore (Sauria: Gekkoninae). Raùes Bulletin ments in geckos. FEBS Letters 445: 36–40. of Zoology 50: 185–196. Vƻǃƹƻǈ, M., GƵǂѯǃ, P., MƿǇƵǑǄǃǉƻǈ, K., Ƶǃƺ WƻǇǃƻǇ, Y. L., OǁƵƺƵ, S., OǉƵ, H., PƻǇǇǎ, G., Vƿƻƿǉƻǈ, D. R. 2004. Weak expression of Ƶǃƺ TǄǁǊǃƵƽƵ 2 5@QHDC@MCìTBST@S- reproductive seasonality in a dwarf gecko ing foraging modes in nocturnal lizards of the (Lygodactylus verticillatus) from arid south- Family Gekkonidae. Asiatic Herpetological western Madagascar. Journal of Arid Research 7: 153–165. Environment 56: 329–338. WƻǇǃƻǇ, Y. L., TƵǁƵƾƵǈƾƿ, H., YƵǈǊǁƵǌƵ, Vƿǉǉ, L. J., SƾƻǅƵǇƺ, D. B., CƵǂƺǌƻǂǂ, J. P., Y., Ƶǃƺ OǉƵ, H. 2004. The varied foraging VƿƻƿǇƵ, G. H. C., FǇƵǃѳƵ, F. G. R., Ƶǃƺ CǄǂǂƿ, mode of the subtropical eublepharid gecko G. R. 2007. Living with your food: geckos in Goniurosaurus kuroiwae orientalis. Journal of termitaria of Cantão. Journal of Zoology 272: Natural History 38: 119–134. 321–328. WƻǇǃƻǇ, Y. L. Ƶǃƺ CƾǄǊ, L. M. 2002. Observations on the ecology of the arrhythmic Accepted: 30 July 2014