Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Journal of Zoology. Print ISSN 0952-8369 Living with your food: geckos in termitaria of Cantão L. J. Vitt1, D. B. Shepard1, J. P. Caldwell1, G. H. C. Vieira2, F. G. R. França2 & G. R. Colli2 1 Sam Noble Oklahoma Museum of Natural History and Department of Zoology, University of Oklahoma, Norman, OK, USA 2 Departamento de Zoologia, Universidade de Brası́lia, Brası́lia, DF, Brazil Keywords ectotherm thermoregulation; diets; microhabitat specificity, lizard ecology, termitaria. Correspondence Laurie J. Vitt, Sam Noble Oklahoma Museum of Natural History, 2401 Chautauqua Avenue, Norman, OK 73072, USA. Tel: +1 405 325 5002; Fax: +1 405 325 7699 Email: [email protected] Received 11 August 2006; accepted 27 September 2006 doi:10.1111/j.1469-7998.2006.00273.x Abstract We tested three non-exclusive hypotheses that the lizard, Gymnodactylus carvalhoi, lives in termitaria to avoid thermal extremes, to avoid predators, or because of an abundance of food (dietary specialist). We first confirm that these geckos are restricted to termitaria in the region studied. Body temperatures (Tb) of geckos averaged below environmental temperatures during day outside of termitaria and above outside temperatures at night; Tb averaged only slightly higher than temperatures inside termitaria. We conclude that thermal constraints in Cerrado habitats lacking rocks restrict Gymnodactylus to termite nests. High frequencies of tail loss and the presence of many potential predators within termitaria suggest high encounter rates with predators, indicating that predation pressure does not restrict these geckos to termite nests. Dietary data indicate that G. carvalhoi is a termite specialist. Published data indicate that other Gymnodactylus species and populations are also termite specialists, even though several live primarily outside termitaria (in crevices and under rocks). An evolutionary history of termite specialization and low thermal requirements in the clade (Gymnodactylus) predispose them to feed on termites within the termitaria. Introduction Energy gained by an organism from food is used for maintenance, growth and, if sexually mature, reproduction (Congdon, Dunham & Tinkle, 1982; Congdon, 1989; Schwarzkopf, 1994; Doughty & Shine, 1997). To maintain a positive energy balance, energy gain resulting from prey ingestion must exceed energetic costs of searching for, pursuing and capturing prey. However, this simplified view of predator–prey interactions excludes potential risk associated with foraging behavior (e.g. Shine, Schwarzkopf & Caley, 1996). Foraging individuals may increase their exposure to predators (direct fitness cost) or may be constrained in their ability to harvest available prey because of predation risk (Stamps, 1977; Werner et al., 1983; Cooper, 1997; Cooper et al., 2003). In addition, physiological constraints may limit an individual’s ability to forage in some habitat patches or during certain times (e.g. Grant, 1990). Terrestrial ectotherms are primarily constrained by their inability to tolerate changes in moisture and temperature (Shine, 1988; Congdon, 1989; Hutchison & Dupre, 1992; Qualls & Andrews, 1999), which affect behavioral and physiological processes (Huey & Stevenson, 1979; Troyer, 1987; Van Damme, Bauwens & Verheyen, 1991). These constraints may restrict individuals to microhabitats that meet the physiological requirements of the species. We examined the ecology of Gymnodactylus carvalhoi (Gekkonidae), a lizard that appears to be almost entirely restricted to termite nests constructed of hard-packed soil in some areas of the Brazilian Cerrado. The vast grasslands known as Cerrado experience both daily and seasonal shifts in temperature and moisture (Franco, 2002). Little rainfall and high temperatures characterize a prolonged dry season extending from about May through October, depending upon locality. The wet season is characterized by sporadic, sometimes intense rainfall and less extreme temperatures. Frequent fires affect vegetation structure as well as nutrient fluxes (Miranda, Bustamante & Miranda, 2002). Reptile and amphibian diversity is high in Cerrado (Colli, Bastos & Araujo, 2002) and many species use termite nests as refugia. In habitats where lizards have been studied, Cerrado contains from 22 to 75 termite species in four families (Colli, Constantino & Costa, 2006), many of which live in large, compacted terrestrial termitaria (Lacher et al., 1986). Many additional Cerrado termite species are known (Mathews, 1977; Gontijo & Domingos, 1991; Constantino, 1998, 2005). Termites are important in tropical ecosystems for a wide range of reasons, varying from food sources for amphibians and reptiles (e.g. Vitt & Colli, 1994; Colli et al., 2006) to production of atmospheric gases (Zimmerman et al., 1982). In both Australian deserts and Brazilian Cerrado, lizard diversity is correlated with termite diversity (Morton & James, 1988; Pianka, 1989; Colli et al., 2006). Nevertheless, c 2007 The Authors. Journal compilation c 2007 The Zoological Society of London Journal of Zoology 272 (2007) 321–328 321 Geckos in termitaria L. J. Vitt et al. termite diversity in the Cerrado does not appear to drive lizard diversity (Colli et al., 2006). We first establish that the Cerrado endemic gecko, G. carvalhoi, is restricted to termite nests in the study area. Next, we ask why G. carvalhoi is restricted to termite nests even though no other sympatric lizard species are. We analyze temperature, tail loss frequency and dietary data to determine whether temperature, predation, diet or a combination of these factors account for the restriction of these lizards to termite nests. Finally, we comment on evolution within the genus Gymnodactylus and consider the evolution of microhabitat specialization in an ecological and historical context (e.g. Vitt et al., 2003; Vitt & Pianka, 2005). Methods Field data collection The study took place at the Parque Estadual do Cantão, located just east of the Araguaia River in western Tocantins state, municipality of Caseara, Brazil (approximate coordinates are: 09118 0 33.200 S 49157 0 27.600 W). The study area is part of the Cerrado Biome, consisting of open grasslands with stunted trees and large numbers of termite nests. We used drift-fence arrays with a combination of pitfall and funnel traps to sample most lizards and snakes in the area. We also methodically broke apart 50 termite nests over a period of 8 days; subsequent observations revealed that termites immediately began reconstructing their nests. Before opening each nest, we measured nest width on x and y axes at ground level and height at the center. Nests approximate the shape of elliptical cones; thus, we used these measurements to estimate the volume of nests based on the formula for an elliptical cone: V ¼ ðpÞðaÞðbÞðh=3Þ where a is the x axis of the base, b is the y axis of the base and h is the nest height. We recognize that our volume estimates are minimal estimates because the tops of nests are rounded off rather than peaked like cones, and nests varied in the extent to which they extended underground. We recorded the location of each nest with respect to trees (in full sun, partly shaded, fully shaded, hereafter, nest aspect). We counted all nest openings large enough for a small vertebrate to enter. We measured four surface temperatures of each nest (approximating N, E, S and W) using TM a Rayteks ST Pro (Raytek Corp., Santa Cruz, CA, USA) hand-held infrared noncontact thermometer. We recorded temperatures inside of up to three holes entering the nests. For these, we attempted to obtain temperatures as deep in the nest as possible. We then broke apart each nest and recorded all vertebrates encountered within the nest. We recorded the number, sex and relative size of Gymnodactylus in each termitarium, when present. When Gymnodactylus were found, we captured them and immediately took cloacal temperature (Tb) with a Miller–Weber rapid register thermometer. The time lag between first disturbing a nest as we broke into it and the time we found geckos varied consider322 ably. In some instances, we worked on individual nests for more than 10 min before finding and capturing a gecko. To determine whether the time necessary to find geckos influenced their Tb, we made two relevant comparisons. First, because we often found more than one gecko in a nest, we used an ANOVA to compare Tb between all geckos caught first, second or third. This included geckos caught in nests containing only a single gecko (they would be the first). Second, we used a paired t-test to compare Tb of only those geckos with paired values (e.g. two geckos captured consecutively in the same nest). If time to capture affected gecko Tb, then we would expect geckos caught later relative to others to have higher Tb. To gain more insight into temperature fluctuations of the thermal environments on and within termite nests, we selected four moderately large termitaria and monitored temperature every 10 min over a 13-day period (October 7–19) with electronic dataloggers (TidBits; Onset Computer Corp.). A single TidBit was placed on the top of each nest to provide time series of surface temperatures. Three TidBits with strings attached were dropped into holes in each nest. Because these lizards live inside nests and are sometimes active on nest surfaces, we consider these to estimate operative temperatures (Te) in representative gecko microhabitats. We did not attempt to validate the use of TidBits for two reasons: (1) several recent studies have shown that TidBits are useful as proxies for copper models of smallbodied lizards (Vitt & Sartorius, 1999; Shine & Kearney, 2001; Vitt et al., 2005) and (2) the most relevant TidBit temperatures were those within the termitaria where most physical variables affecting lizard temperatures (wind speed, insolation, lizard posture) were at best only minimally relevant. We broke apart these four nests after our temperature-sampling period; all contained Gymnodactylus. To aid in interpretation of our temperature data, we assembled data on body temperatures of other Neotropical lizards (L. J. Vitt, Ávila-Pires, S.S. Sartorius & P.A. Zani publ. data and unpubl. data) collected by some of the authors. Laboratory data collection We collected 69 G. carvalhoi from 25 termitaria for morphological and diet analyses. These were taken back to our field laboratory and humanely euthanized within 1–2 h after capture. We measured snout–vent length (SVL), unregenerated portion of the tail (tail base) and regenerated portion of the tail (if any) to 1 mm with a plastic rule, and head width (widest point), head length (from the anterior edge of the tympanic opening to the snout), head height (at deepest point), body width and height (mid body), hindleg length from the body anterior to the limb to the tip of the longest toe and foreleg length from the body posterior to the limb to the tip of the longest toe to 0.01 mm with digital calipers for each individual lizard before fixation. We weighed lizards to 0.01 g on Acculab digital balances (Acculab, Edgewood, NY, USA). Tail measurements allowed us to calculate tail loss frequency within the population. By using only lizards c 2007 The Authors. Journal compilation c 2007 The Zoological Society of London Journal of Zoology 272 (2007) 321–328 L. J. Vitt et al. Geckos in termitaria 4 length width 2 V¼ p 3 2 2 using the program BugRun, a 4th Dimensions-based analysis that produces dietary summaries, calculates mean prey size (length, width and volume) for each lizard, estimates stomach volume based on total prey volume and calculates niche breadth using the inverse of Simpson’s (1949) diversity measure (see Pianka, 1973, 1986): b¼ 1 n P p2i i¼1 where p is the proportional utilization of each prey type i. Niche breadth values (b) vary from 1 (exclusive use of a single prey type) to n (even use of all prey). We log10 transformed all quantitative data to normalize distributions for further analyses. We used linear regression to determine whether prey size and number of prey eaten varied with lizard body size (log10 SVL). To determine whether prey size or number of prey eaten differed between sexes, we used an ANCOVA with log10 SVL as the covariate and sex as the class variable. We also plotted log10 stomach volume with log10 SVL to determine the relationship between stomach volume and lizard size, and to estimate the relative fullness of the lizards sampled (see Huey, Pianka & Vitt, 2001). We made the assumption that a line roughly parallel to the regression slope that intersects upper values approaches the relationship between full stomachs and lizard body size. Results Lizard microhabitats, Tb and thermoregulation Our pitfall trapping established that G. carvalhoi rarely move about on the ground. During 53 days of trapping using 60 arrays containing a total of 240 traps (12 720 trap days), only one Gymnodactylus was caught in a trap. Among 73 individuals captured in this study, one was in a pitfall trap, two were underneath a pile of boards in a field, one was on the ground away from a termite nest at dusk and the remainder (94.5%) were found on or inside termite nests. Several additional geckos (not collected) were observed during the day inside termite nests through holes in the nests, and several others were observed at dusk or at night on or inside termite nests. Those on termite nests immediately fled into the nests when approached. Among the 50 nests opened during the day, 25 (50%) contained geckos. From one to four geckos occurred in nests containing geckos (Fig. 1). Among the 25 nests containing geckos, 14, 10 and one were in sun, partial shade or shade, respectively. Among nests not containing geckos, 16, 8 and one were in sun, partial shade or shade, respectively. The 50 termitaria averaged 61.3 2.3 cm in height (h), 101.9 3.4 cm in long base diameter (a), 92.2 3.3 cm in short base diameter (b) and 0.695 0.083 m3 in volume. The number of holes varied from one to 27 (mean = 11.4 0.8). Termitaria containing geckos had the same number of holes opening to the outside as termitaria without geckos (11 1.0 and 11.6 1.3, respectively, Mann–Whitney U-test, Z=0.01, P= 0.99). The number of holes was not significantly correlated with nest volume (r= 0.24, n = 50, P= 0.10). We used multiple logistic regression to test whether nest volume, the number of holes and nest aspect were good predictors of gecko presence/absence. Gecko presence/absence could not be predicted by nest volume (w2 = 2.07, d.f.= 1, P= 0.15), the number of holes (w2 = 0.49, d.f.= 1, P= 0.48) or nest aspect (w2 = 0.44, d.f.= 2, P= 0.80). External and internal nest temperatures, based on continuous temperature monitoring with TidBits, varied on an hourly and daily basis (Fig. 2), with temperatures inside the nests being considerably lower during the day but higher at night. Temperatures taken on termitaria surfaces and inside 25 20 No. of nests with complete tails in a regression of SVL versus tail length, we were able to generate estimates of original tail lengths for all individuals with broken or regenerated tails to determine the relative position of tail breaks. Most tail breaks occur along cleavage planes in vertebrae (Arnold, 1984), but some secondary tail breaks occur within the regenerated portion lacking vertebrae. Lizards were fixed in 10% formalin and given unique individual tags. Lizards were later transferred to 70% ethanol for permanent storage in the Herpetology Collection of the University of Brası́lia (CHUNB) and the Sam Noble Oklahoma Museum of Natural History (OMNH). Within 1 month following fixation, the stomachs were removed from geckos, prey were identified to family when possible and each prey item was measured for length and width. When invertebrate prey are compressed in lizard stomachs, their shape approaches that of a prolate spheroid. We used the formula for a prolate spheroid to estimate the volume of individual prey items: 15 10 5 0 0 1 2 3 4 No. Gymnodactylus Figure 1 Frequency distribution showing number of geckos Gymnodactylus carvalhoi found in each of 50 termitaria sampled. When nests contained more than two geckos, additional individuals were juveniles. c 2007 The Authors. Journal compilation c 2007 The Zoological Society of London Journal of Zoology 272 (2007) 321–328 323 Geckos in termitaria L. J. Vitt et al. 60 In/bottom In/mid Surface No. of records 50 x− = 35.3 ± 0.39 °C 40 40 30 (26.8− 61.8 °C) n = 196 30 20 10 0 20 25 30 35 40 45 50 55 60 65 60 65 Surface temperatures (°C) 10 60 No. of records 40 50 40 Temperature (°C) 30 (27.0 −49.4 °C) n = 138 20 10 0 25 20 30 35 40 45 50 55 Hole temperatures (°C) 10 Figure 3 Frequency distributions of temperatures recorded with the Rayteks noncontact thermometer on the surfaces and in the holes of termitaria. 60 50 40 30 20 10 60 50 40 30 20 10 0000 0400 0800 1200 1600 2000 0000 Hour of day Figure 2 Daily march of temperatures on surface, inside approximately the middle, and inside near the bottom of four termitaria based on continuous monitoring with TidBits digital temperature recording devices. Repeated bars to the right of each panel depict ranges of Tb for geckos captured inside of the 50 nests that were opened, and the continuous line across each panel indicates the mean value. holes before we broke apart nests varied considerably depending on exposure, but surface temperatures overall were higher than temperatures inside holes (Fig. 3). The mean Tb of 31 captured geckos was 31.3 0.2 1C (28.3–33.8 1C). The order of capture (first, second or third) did not affect gecko Tb (F2,28 = 0.7, P= 0.49) and this 324 x− = 33.3 ± 0.31 °C 30 remains true when only gecko pairs from the same nest were compared (paired t7 = 0.26, P = 0.80). Because these geckos were captured during the day, comparisons with termitaria temperatures are only relevant during those time periods. Nevertheless, G. carvalhoi was able to maintain Tb close to the upper limit of Te inside nests, likely by positioning themselves in exit holes or other nest areas where temperatures were slightly higher. An ANOVA on species’ mean Tb revealed significant differences among lizard families (families represented by less than three species were excluded; F5,60 = 17.0, Po0.001). Gekkonid Tb were not significantly different from those of gymnophthalmids, scincids, polychrotids or tropidurids. Teiid lizards had higher Tb than polychrotids, gymnophthalmids and gekkonids (Po0.05, post-hoc Games–Howell test). However, because most of the lizards represented (Fig. 4) occur in more mesic habitats (e.g. rainforest), we divided species into those living in closed (forest), open (e.g. caatinga, Cerrado) and both habitats (e.g. Ameiva ameiva occurs nearly everywhere in tropical Brazil). A comparison of Tb for lizards in these categories revealed significant differences among categories (F2,63 = 18.7, Po0.0001). Tb of lizards in closed habitats was lower than those in open habitats (Po0.05, post-hoc Games–Howell test), and Tb of lizard species using both open and closed habitats was intermediate (P40.05, posthoc Games–Howell test). Our sample of species’ means did not allow a two-way ANOVA, so these results should be taken only as suggestive (i.e. we cannot sort out the relative effects of phylogeny and habitat type). Most relevant is that Tb of G. carvalhoi in Cerrado (31.3 0.2 1C), Gymnodactylus geckoides in caatinga (32.5 0.6 1C) and Gymnodactylus amarali in Cerrado [30.2 0.5 1C (SE calculated from SD)] c 2007 The Authors. Journal compilation c 2007 The Zoological Society of London Journal of Zoology 272 (2007) 321–328 L. J. Vitt et al. Geckos in termitaria by family 14 by habitat 37 12 36 10 No. of lizards Mean body temperature (°C) 38 35 34 33 8 6 4 32 31 2 30 0 10 29 20 30 40 50 60 70 80 90 100 Per cent of original tail lost 28 Closed habitats Both Open habitats Teiidae Tropiduridae Scincidae Polychrotidae Gekkonidae Gymnophthalmidae 27 Figure 4 Means and SE of lizard Tb by family and among habitat types. See Supplementary Materials Table S1 for data and references. Gekkonid mean Tb is similar to those of gymnophthalmids and polychrotids, most of which live in tropical forest. Gekkonid Tb remain low regardless of habitat type. The mean Tb for Gymnodactylus carvalhoi from this study (open habitat) was 31.3 1C. are considerably lower than mean Tb for lizards living in open habitats (34.3 0.8 1C), but higher than the mean for all gekkonid species combined (29.5 0.8 1C). Tail-loss frequencies and position of tail breaks Fifty of 69 (72.5%) geckos had lost their tails at least once. For geckos with unbroken tails, tail length was significantly correlated with SVL (adjusted R2 = 0.93, F1,18 =254.8, Po0.0001); the relationship between unbroken complete tail length and SVL was: tail length= 1.29 (SVL)6.843. Percent of the tail broken off varied from 19.6 to 96.5% (mean = 70.9 3.2%), with nearly 20% of geckos losing nearly the entire tail (Fig. 5; i.e. tail break occurring on the first vertebral cleavage plane of the tail). Lizard diets Sixty G. carvalhoi contained a total of 219 prey items in their stomachs, comprising 14 prey categories (Table 1). The diet was dominated numerically and volumetrically by termites, with only spiders and ants contributing much to the remainder of the total diet. Fifty-three of the 60 geckos contained termites. Low niche breadth values (Table 1) indicate that G. carvalhoi is a termite specialist. Individual prey averaged 9.1 0.33 (range 0.80–35.41) mm in length, 3.28 0.11 (range 0.61–11.19) mm in width and 79.7 8.65 (range Figure 5 Frequency distribution of percent of tail lost in Gymnodactylus carvalhoi with missing or regenerated tails. Table 1 Dietary summary for Gymnodactylus carvalhoi from the Cantão region of Brazil based on analysis of 60 stomachs containing prey. Prey type No. % No. Volume % Volume Freq. Crickets and grasshoppers 3 0.38 140.89 1.86 Termites 728 91.8 5881.58 77.76 Beetles 5 0.63 28.62 0.38 Homopterans 6 0.76 143.46 1.9 Flies 4 0.5 119.77 1.58 Hymenopterans (non-ant) 2 0.25 0.63 0.01 Ants 27 3.4 351.42 4.65 Insect egg case 1 0.13 67.24 0.89 Insect larvae 3 0.38 193.65 2.56 Springtails 1 0.13 0.58 0.01 Spiders 9 1.13 456.42 6.03 Millipedes 1 0.13 159.25 2.11 Centipedes 2 0.25 19.34 0.26 Pseudoscorpions 1 0.13 1.01 0.01 SUMS 793 100.00 7563.86 100.00 Niche breadths 1.18 1.63 3 53 3 3 1 2 16 1 2 1 8 1 2 1 — No., number of individuals of a particular prey; Volume, total volume of that prey type in all lizards; Freq., number of individual lizards (out of a possible 60) containing a particular prey type. 0.72–1283.72) mm3 in volume. No relationship was found between gecko SVL and mean prey volume (F1,56 = 2.0, P= 0.16) or number of prey (F1,56 = 3.0, P= 0.09). Larger geckos contained a larger total volume of prey (F1,56 = 5.6, P= 0.02; Fig. 6). Discussion Terrestrial termitaria are abundant in many areas of the Brazilian Cerrado and provide refuge for a wide variety of animals. The endemic Cerrado gecko G. carvalhoi appears largely restricted to these social insect nests in the Cantão region based on our sampling data. Our failure to trap more than a single individual using terrestrial trap systems that c 2007 The Authors. Journal compilation c 2007 The Zoological Society of London Journal of Zoology 272 (2007) 321–328 325 Geckos in termitaria L. J. Vitt et al. Log10 stomach contents volume 4 estimated full stomach 3 2 1 0 −1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Log10 snout – vent length Figure 6 Relationship of lizard body size [snout–vent length (SVL)] with total prey volume for Gymnodactylus carvalhoi. The slope of the relationship, log10 total prey volume = 1.633(log10 SVL) – 0.836, is significantly different from zero (t= 2.36, P= 0.02). Full stomach volume is estimated as the line that intersects the highest points on the regression. proved effective for all other lizard and snake species, even with 12 720 trap days, indicates that these geckos spend little time moving about on the ground. Most likely, they move only when dispersing to find new nests to inhabit. Thus, we confirm that G. carvalhoi are restricted to termitaria as their primary microhabitats in the Cantão region. Populations of G. carvalhoi in some other areas of the Cerrado (e.g. Goiás State), formerly referred to as Gymnodactylus geckoides amarali (see Colli et al., 2003 for ecological data and Vanzolini, 2005 for most recent taxonomy), live in rock crevices and under rocks. Rocky habitats were not available at Cantão as they are at some other Cerrado localities, so restriction to termite nests may result from an absence of alternative appropriate refuges. Alternatively, restriction to rocks in some other areas may result from absence of appropriate termitaria. The possibility also exists that G. carvalhoi as now delineated represents more than a single taxon. Our data on thermal ecology, tail loss and diets suggest that a combination of thermal constraints, limited types of available refugia, and diet preferences explain the restriction of these geckos to termite nests. The hypothesis that predation risk plays a significant role in restricting these geckos to termitaria is not supported by our observations on tail loss. Although high frequencies of tail loss do not necessarily indicate high predation rates (e.g. Schoener, 1979; Schoener & Schoener, 1980), they do suggest that predator attacks are quite common and that many geckos escaped at least one attack by losing their tails. Tail loss may occur during intrasexual encounters in some lizards (e.g. Vitt et al., 1974; Vitt & Zani, 1997), but we have not observed such encounters in Gymnodactylus. Individuals often raise their tails as a defensive display to expose the black and white-banded underside, similar to other lizard species (Congdon, Vitt & 326 King, 1974; Colli et al., 2003). This behavior presumably distracts predator attacks away from the head and body toward the expendable tail. Such displays may deter predator attacks rather than elicit them (Dial, 1986), but we have no evidence for this in G. carvalhoi. Raising the tail over the body in Coleodactylus brachystoma, another Cerrado gecko, presumably makes lizards resemble sympatric scorpions, thereby reducing predation (Brandão & Motta, 2005). Lizard-eating snakes, large scolopendromorph centipedes, therophosid spiders and large amblypygids occur inside termitaria and are potential predators on Gymnodactylus. If anything, crepuscular habits of G. carvalhoi while on the surface of nests could reduce predation by lizard-eating birds, most of which forage during morning. Whether a social (e.g. Fox & Rostker, 1982; Fox, Heger & Delay, 1990; Martin & Salvador, 1993) or demographic (e.g. Niewiarowski et al., 1997) cost of tail loss exists in G. carvalhoi remains unknown. Te within termitaria remained relatively constant throughout the day and was within the range of Tb observed in G. carvalhoi (Fig. 2). However, Te on termitaria surfaces were much higher during the day and considerably lower during the night than gecko Tb, suggesting that thermal extremes (high and low) in the external environment contribute partially to the restriction of G. carvalhoi to termitaria. Although the underlying causes of low Tb in G. carvalhoi remain unknown, our comparisons of Tb among lizard families and among habitat types shed some light on this. We conclude that G. carvalhoi is living in a microhabitat (inside of termitaria) that buffers these lizards from external temperature extremes that most gekkonids would not be able to tolerate. These findings are consistent with our temperature data on termitaria and the rarity of these lizards in terrestrial traps. Thus, temperature is one reason why G. carvalhoi inhabits these social insect nests. However, other species and populations of Gymnodactylus live in non-exposed microhabitats in open habitats (under rocks for G. geckoides and Goiás populations of G. carvalhoi; Vitt, 1995; Colli et al., 2003). In the Cantão region, surface rock microhabitats are not available. In caatinga where G. geckoides occurs and rocky areas of Goiás, large termite nests are rare. The high frequency of geckos containing prey in their stomachs, and the relative fullness of stomachs suggest that prey resources are easily available and that these geckos maintain a positive energy balance (e.g. Huey et al., 2001). All three measures of dietary importance, numbers, volumes and frequencies of use, indicate that G. carvalhoi is, for the most part, a termite specialist. Because these lizards eat other invertebrates as well, the possibility exists that their apparent specialization on termites results from superabundance of termites in microhabitats in which geckos are restricted. Spiders and ants, the only other prey categories contributing substantially to the diet, are also common in termitaria. Termites dominate the diets of G. geckoides (caatinga) and Goiás populations of G. carvalhoi (Cerrado) as well (Vitt, 1995; Colli et al., 2003). Open habitats in warm environments present numerous challenges to small terrestrial vertebrates, including high c 2007 The Authors. Journal compilation c 2007 The Zoological Society of London Journal of Zoology 272 (2007) 321–328 L. J. Vitt et al. diversity and abundance of predators, thermal extremes and great fluctuations in humidity and rainfall. Nevertheless, the diversity and abundance of lizards is often high in these habitats (e.g. Pianka, 1975; Colli et al., 2002; Pianka & Vitt, 2003). Gekkonid lizards in most habitats avoid thermal extremes, diurnal predators (birds in particular) and interactions with voracious autarchoglossan lizards by restricting activity to night (e.g. Vitt et al., 2003). Gymnodactylus carvalhoi appears to avoid thermal extremes and likely some predators by living in terrestrial termitaria where an abundant food supply offsets risk associated with foraging outside of the nest and in which a moderate thermal environment provides refuge from external temperature extremes. Use of termites may reflect an evolutionary history of termite specialization in Gymnodactylus geckos, whereas relatively low Tb compared with other open habitat lizards likely reflects an evolutionary history of low Tb in gekkonids in general. Acknowledgements We first thank the Brazilian Government, particularly Conselho Nacional de Desenvolvimento Cientı́fico e Tecnológico [CNPq] (Portaria MCT no. 649, de 23/12/2004) and the Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis [IBAMA] (permit no. 0217/ 2004-CGFU/LIC) for issuing permits and helping with often complex bureaucratic logistics. The State Government of Tocantins allowed us to work in Cantão and provided logistic support. We thank NATURATINS (Instituto Natureza do Tocantins) for providing logistic support at Parque Estadual do Cantão and permits (No. 018/2005). IACUC Assurance Number A3240-01, University of Oklahoma, applies to this study. This work was supported by the National Science Foundation under Grant No. 0415430. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation. References Arnold, E.N. (1984). Evolutionary aspects of tail shedding in lizards and their relatives. J. Nat. His. 18, 127–169. Brandão, R.A. & Motta, P.C. (2005). Circumstantial evidences for mimicry of scorpions by the Neotropical gecko Coleodactylus brachystoma (Squamata, Gekkonidae) in the Cerrados of central Brazil. Phyllomedusa 4, 139–145. Colli, G.R., Bastos, R.P. & Araujo, A.F.B. (2002). The character and dynamics of the Cerrado herpetofauna. In The Cerrados of Brazil: ecology and natural history of a Neotropical savanna: 223–241. Oliveira, P.S. & Marquis, R.J. (Eds). New York: Columbia University Press. Colli, G.R., Constantino, R. & Costa, G.C. (2006). Lizards and termites revisited. Aust. Ecol. 31, 417–424. Colli, G.R., Mesquita, D.O., Rodrigues, P.V.V. & Kitayama, K. (2003). Ecology of the gecko Gymnodactylus geckoides amarali in a Neotropical savanna. J. Herpetol. 37, 694–706. Geckos in termitaria Congdon, J.D. (1989). Proximate and evolutionary constraints on energy relations of reptiles. Physiol. Zool. 62, 356–373. Congdon, J.D., Dunham, A.E. & Tinkle, D.W. (1982). Energy budgets and life histories of reptiles. In Biology of the reptilia: 233–271. Gans, C. (Ed.). New York: Academic Press. Congdon, J.D., Vitt, L.J. & King, W.W. (1974). Geckos: adaptive significance and energetics of tail autotomy. Science 184, 1379–1380. Constantino, R. (1998). Catalog of the living termites of the New World (Insecta: Isoptera). Arq. Zool. (S. Paulo) 35, 135–231. Constantino, R. (2005). Padrões de diversidade e endemismo de térmitas no bioma Cerrado. In Biodiversidade, Ecologia e Conservação do Cerrado: 319–333. Scariot, A., SouzaSilva, J.C. & Felfili, J.M. (Eds). Brası́lia: Ministério de Meio Ambiente. Cooper, W.E. Jr. (1997). Factors affecting risk and cost of escape by the broad-headed skink (Eumeces laticeps): predator speed, directness of approach, and female presence. Herpetologica 53, 464–474. Cooper, W.E. Jr., Pérez-Mellado, V., Baird, T., Baird, T.A., Caldwell, J.P. & Vitt, L.J. (2003). Effects of risk, cost, and their interaction on optimal escape by nonrefuging Bonaire whiptail lizards, Cnemidophorus murinus. Behav. Ecol. 14, 288–293. Dial, B.E. (1986). Tail display in two species of iguanid lizards: a test of the ‘‘predator signal’’ hypothesis. Am. Nat. 127, 103–111. Doughty, P. & Shine, R. (1997). Detecting life history tradeoffs: measuring energy stores in ‘‘capital’’ breeders reveal costs of reproduction. Oecologia 110, 508–513. Fox, S.F., Heger, N.A. & Delay, L.S. (1990). Social cost of tail loss in Uta stansburiana: lizard tails as status-signalling badges. Anim. Behav. 39, 549–554. Fox, S.F. & Rostker, M.A. (1982). Social cost of tail loss in Uta stansburiana. Science 218, 692–693. Franco, A.C. (2002). Ecophysiology of woody plants. In The Cerrados of Brazil: ecology and natural history of a Neotropical savanna: 178–197. Oliveira, P.S. & Marquis, R.J. (Eds). New York: Columbia University Press. Gontijo, T.A. & Domingos, D.J. (1991). Guild distribution of some termites from cerrado vegetation in south-east Brazil. J. Tropical Ecol. 7, 523–529. Grant, B.W. (1990). Trade-offs in activity time and physiological performance for thermoregulating desert lizards, Sceloporus merriami. Ecology 71, 2323–2333. Huey, R.B., Pianka, E.R. & Vitt, L.J. (2001). How often do lizards ‘‘run on empty’’? Funct. Ecol. 82, 1–7. Huey, R.B. & Stevenson, R.D. (1979). Integrating thermal physiology and ecology of ectotherms: a discussion of approaches. Am. Zool. 19, 357–366. Hutchison, V.H. & Dupre, R.K. (1992). Thermoregulation. In Environmental physiology of the amphibia: 206–249. Feder, M.E. & Burggren, W. (Eds). Chicago: University of Chicago Press. Lacher, T.E. Jr., Egler, I., Alho, C.J.R. & Mares, M.A. (1986). Termite community composition and mound c 2007 The Authors. Journal compilation c 2007 The Zoological Society of London Journal of Zoology 272 (2007) 321–328 327 Geckos in termitaria L. J. Vitt et al. characteristics in two grassland formations in central Brazil. Biotropica 18, 356–359. Martin, J.A. & Salvador, A. (1993). Tail loss reduces mating success in the Iberian rock-lizard, Lacerta monticola. Behav. Ecol. Sociobiol. 32, 185–189. Mathews, A.G.A. (1977). Studies on termites from the Mato Grosso State, Brazil. Rio de Janeiro: Academia Brasileira de Ciências. Miranda, H.S., Bustamante, M.M.C. & Miranda, A.C. (2002). The fire factor. In The Cerrados of Brazil: ecology and natural history of a Neotropical savanna: 51–68. Oliveira, P.S. & Marquis, R.J. (Eds). New York: Columbia University Press. Morton, S.R. & James, C.D. (1988). The diversity and abundance of lizards in arid Australia — a new hypothesis. Am. Nat. 132, 237–256. Niewiarowski, P.H., Congdon, J.D., Dunham, A.E., Vitt, L.J. & Tinkle, D.W. (1997). Tales of lizard tails: effects of tail autotomy on subsequent survival and growth of free-ranging hatchling Uta stansburiana. Can. J. Zool. 75, 542–548. Pianka, E.R. (1973). The structure of lizard communities. Ann. Rev. Ecol. Syst. 4, 53–74. Pianka, E.R. (1975). Niche relations of desert lizards. In Ecology and evolution of communities: 292–314. Cody, M. & Diamond, J. (Eds). Cambridge, MA: Harvard University Press. Pianka, E.R. (1986). Ecology and natural history of desert lizards. Princeton, NJ: Princeton University Press. Pianka, E.R. (1989). Desert lizard diversity: additional comments and some data. Am. Nat. 134, 344–364. Pianka, E.R. & Vitt, L.J. (2003). Lizards: windows to the evolution of diversity. Berkeley, CA: University of California Press. Qualls, C.P. & Andrews, R.M. (1999). Maternal body volume constrains water uptake by lizard eggs in utero. Funct. Ecol. 13, 845–851. Schoener, T.W. (1979). Inferring the properties of predation and other injury-producing agents from injury frequencies. Ecology 60, 1110–1115. Schoener, T.W. & Schoener, A. (1980). Ecological and demographic correlates of injury rates in some Bahamian Anolis lizards. Copeia 1980, 839–850. Schwarzkopf, L. (1994). Measuring trade-offs: a review of studies of costs of reproduction in lizards. In Lizard ecology: historical and experimental perspectives: 7–29. Vitt, L.J. & Pianka, E.R. (Eds). Princeton, NJ: Princeton University Press. Shine, R. (1988). Constraints on reproductive investment: a comparison between aquatic and terrestrial snakes. Evolution 42, 17–27. Shine, R. & Kearney, M. (2001). Field studies of reptile thermoregulation: how well do physical models predict operative temperatures? Funct. Ecol. 15, 282–288. Shine, R., Schwarzkopf, L. & Caley, M.J. (1996). Energy, risk, and reptilian reproductive effort: a reply to Niewiarowski and Dunham. Evolution 50, 2111–2114. Simpson, E.H. (1949). Measurement of diversity. Nature 163, 688. Stamps, J.A. (1977). The relationship between resource competition, risk, and aggression in a tropical territorial lizard. Ecology 58, 349–358. Troyer, K. (1987). Small differences in daytime body temperature affect digestion of natural food in a herbivorous 328 lizard (Iguana iguana). Comp. Biochem. Physiol. A 87, 623–626. Van Damme, R., Bauwens, D. & Verheyen, R.F. (1991). The thermal dependence of feeding behaviour, food consumption and gut-passage in the lizard Lacerta vivipara Jacquin. Funct. Ecol. 5, 507–517. Vanzolini, P.E. (2005). On Gymnodactylus amarali Barbour, 1925, with the description of a new species (Sauria, Gekkonidae). An. Acad. Brasil. Cieˆnc. 77, 595–611. Vitt, L.J. (1995). The ecology of tropical lizards in the caatinga of northeast Brazil. Occ. Pap. Oklahoma Mus. Nat. His. 1, 1–29. Vitt, L.J., Caldwell, J.P., Sartorius, S.S., Cooper, W.E. Jr., Baird, T.A., Baird, T.D. & Pérez-Mellado, V. (2005). Pushing the edge: extended activity as an alternative to risky body temperatures in a herbivorous teiid lizard (Cnemidophorus murinus: Squamata). Funct. Ecol. 19, 152–158. Vitt, L.J. & Colli, G.R. (1994). Geographical ecology of a Neotropical lizard: Ameiva ameiva (Teiidae) in Brazil. Can. J. Zool. 72, 1986–2008. Vitt, L.J., Congdon, J.D., Platz, J.E. & Hulse, A.C. (1974). Territorial aggressive encounters and tail breaks in the lizard Sceloporus magister. Copeia 1974, 990–993. Vitt, L.J. & Pianka, E.R. (2005). Deep history impacts present-day ecology and biodiversity. Proc. Natl. Acad. Sci. USA 102, 7877–7881. Vitt, L.J., Pianka, E.R., Cooper, W.E. Jr. & Schwenk, K. (2003). History and the global ecology of squamate reptiles. Am. Nat. 162, 44–60. Vitt, L.J. & Sartorius, S.S. (1999). HOBO’s, Tidbits and lizard models: the utility of electronic devices in field studies of ectotherm thermoregulation. Funct. Ecol. 13, 670–674. Vitt, L.J. & Zani, P.A. (1997). Ecology of the nocturnal lizard Thecadactylus rapicauda (Sauria: Gekkonidae) in the Amazon region. Herpetologica 53, 165–179. Werner, E.E., Gilliam, J.F., Hall, D.J. & Mittelbach, G.G. (1983). An experimental test of the effects of predation risk on habitat use in fish. Ecology 64, 1540–1548. Zimmerman, P.R., Greenberg, J.P., Wandiga, S.O. & Crutzen, P.J. (1982). Termites: a potentially large source of atmospheric methane, carbon dioxide, and molecular hydrogen. Science 218, 563–565. Supplementary material The following material is available for this article online: Table S1 Body temperatures (Tb) of Neotropical lizards grouped by family. This material is available as part of the online article from http://www.blackwell-synergy.com/doi/abs/10.1111/ j.1469-7998.2006.00273.x primary_article Please note: Blackwell Publishing are not responsible for the content or functionality of any supplementary materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article. c 2007 The Authors. Journal compilation c 2007 The Zoological Society of London Journal of Zoology 272 (2007) 321–328