Download Living with your food: geckos in termitaria of Canta˜ o

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