J. Zool., Lond.

(2000) 251, 187±197 # 2000 The Zoological Society of London Printed in the United Kingdom

Abundance, demographics and body condition of a

translocated population of St Lucia whiptail lizards
(Cnemidophorus vanzoi)

H. C. Dickinson1* and J. E. Fa2{

1
FitzPatrick Institute, University of Cape Town, Cape Town, South Africa
2
Durrell Wildlife Conservation Trust, Les AugreÁs Manor, Jersey JE3 5BP, U.K. E-mail: jfa@durrell.org
(Accepted 21 July 1999)

Abstract
The whiptail lizard Cnemidophorus vanzoi is a large-bodied teiid found only on two islands (Maria Major
and Maria Minor), off St Lucia, West Indies. In May 1995, 42 lizards from Maria Major were introduced
to the smaller uninhabited Praslin Island on the same coastline. Three years post-release, we studied
abundance, demography and morphometrics of the translocated lizard population, during a 6-month
period covering wet and dry seasons. Age, sex, snout±vent length (SVL), body mass (BM), tail length, tail
regeneration, and overall condition (moulting, reproductive condition, cuts, external parasites) of 107
animals caught during the study are analysed in the present paper. Comparisons are also made with the
source population. A body condition index (CI (BM/SVL)), sex ratio (adult males : adult females), age
ratio (adults : juveniles), and sexual size dimorphism ratio (SVL adult male : SVL adult females) were
calculated for the study population. Distance sampling and mark±re-sight surveys were used to estimate
population size and lizard density. A total of 155 ‹ 26 individuals were estimated. The lizard population
was found to have a high growth rate (r = 0.97±3.95). There were signi®cant seasonal changes in lizard
abundance. Seasonal differences in lizard numbers, BM and CI suggest either severe resource limitation
during the dry season, or selective aestivation. A high frequency of tail autotomy may point to intense
intraspeci®c competition as the island is relatively free from main predators such as the black rat Rattus
rattus. Sex ratio, sexual size dimorphism and sexual dichromatism all indicate a territorial species in a
generally non-territorial family (Teiidae). Some adult males seem to maintain juvenile colours. It is
suggested that the introduced population has successfully colonized its new environment and that no
signi®cant change in the animals condition or size has occurred during the 3 years since translocation.

Key words: Cnemidophorus vanzoi, translocation, body condition index, St Lucia

INTRODUCTION recipient habitat needs to be assessed according to the

In the face of the global reduction of biodiversity,
translocation is an important conservation technique.
We de®ne translocation as the intentional release of
animals to the wild to establish, re-establish, or augment
a population (IUCN, 1987). Viable, free-ranging popu-
lations of a species which were locally extinct or
extirpated in the wild can result, requiring only minimal
long-term management. Translocation success is judged
by population persistence, which is affected by founder
numbers, the species' ecological attributes (e.g. intrinsic
rate of increase) and the condition of the recipient
habitat (B. Grif®ths et al., 1989). The condition of the
* Present address: Durrell Wildlife Conservation Trust, Les AugreÁs
Manor, Jersey JE3 5BP, U.K.
{All correspondence to: J. E. Fa, Durrell Wildlife Conservation Trust,
Les AugreÁs Manor, Jersey JE3 5BP, U.K.
species' ecological requirements at the time of release,
and by the temporal variation of crucial resources post-
release. The founding population must increase in
number quickly to withstand ¯uctuations in both envir-
onment and demography. Thus, translocations will not
succeed if seasonally ¯uctuating resources curtail the
survival and reproductive potential of small colonizing
populations (Gilpin & SouleÂ, 1986). Understanding the
impact of periods of resource shortage and of severe
environmental conditions is crucial for planning an
effective translocation. These types of studies exist
(Miura & Maruyama, 1986; Ianson, 1989; Sweitzer &
Berger, 1993; Kennish, 1997), but there are few studies
demonstrating how these variables in¯uence the ecology
and population dynamics of translocated populations.
Comparing demographic traits between translocated
and wild populations will help managers determine
188 H. C. Dickinson and J. E. Fa
when a population has become an established viable
entity (Miller et al., 1999). The aim of this paper is to
describe the morphometrics and demographic status of
St Lucia whiptail lizards Cnemidophorus vanzoi trans-
located from a natural population to another island
previously uninhabited by the species. The translocated
St Lucia whiptail lizard population is subjected to
seasonally ¯uctuating resources in the new environment.
We also describe the translocated population size,
3 years after release, and study the temporal changes in
body condition and demography of the lizards.
Wherever possible, comparisons with the source popula-
tion animals are made.
The lizard is a ground-dwelling, diurnal, primarily
insectivorous macroteiid (Vitt & Breitenbach, 1993). It
is a single island endemic and the sole representative of
its genus in the Lesser Antilles (Swartz & Henderson,
1991). The species was suspected to have inhabited
mainland St Lucia, but no evidence exists for this.
Before the translocation in 1995, the lizard inhabited
only two small offshore islands (Maria Major and
Maria Minor). Due to the perceived risk from ®res and
introduced predators from the main island, the Jersey
Wildlife Preservation Trust, U.K. and the St Lucia
Forestry Department, translocated Maria Major lizards
to Praslin Island. After initial eradication of rats from
Praslin (Johnston, Anthony & Bloxam, 1994), and
ecological comparisons of the islands made to assess
suitability, 28 lizards (13 adult males, 12 adult females,
two juvenile males, and one juvenile female) were
released onto Praslin in May 1995. The discovery of a
single mongoose Herpestes sp. shortly after the lizard
translocation, may have caused the loss of a signi®cant
number of lizards, since only one adult male was
detected (Q. Bloxam, pers. comm.). After removal of
the mongoose, another 14 lizards (seven adult males, six
adult females, and one juvenile female) from Maria
Major were released onto Praslin. No other releases
have taken place since then.
METHODS AND ANALYSIS
Study area
Praslin Island (13852.2'N, 60853.1'W), in Praslin Bay, is
a 1.1 ha island located 220 m off the east coast of
St Lucia and 21 km north of Maria Major (Fig. 1). It is
a privately owned island leased to the St Lucia Forestry
Department. The island slopes along the east±west axis
from a maximum peak of c. 15 m. It is composed of
conglomerate mud¯ow (Applied Technology &
Management Inc., 1996). The eastern coast is severely
undercut by wave action and there is evidence of a
landslide on the north-eastern side. The vegetation has
recovered since goats were removed in 1991, and further
since the eradication of rats in March 1993 (Johnston
et al., 1994).
Climate in the region has a distinct wet (June±
December) and dry (January±May) season. During the
Atlantic
Ocean 0 10 km
Caribbean Sea
St Lucia Cap Estate
Gros
S. AMERICA Islet
CASTRIES
Roseau
Praslin Island
Dennery
Soufrière Mt Gimie 950 m
Micoud
Laborie
Vieux Fort
Viex Fort
Maria Islands
Fig. 1. Location of source and translocated populations of
St Lucia whiptail lizards Cnemidophorus vanzoi.
study, average precipitation at Hewanorra Interna-
tional Airport (Vieux Fort) was: October 5.2 mm,
November 4.7 mm, December 3.2 mm, January
2.5 mm, February 0.6 mm, March 0.8 mm. During the
wet season there was 400 mm of rain and in the dry
season there was 111.6 mm. Average maximum and
minimum temperatures were 31.1 8C and 25.4 8C during
the wet season, and 30.3 8C and 24.8 8C during the dry
season. Hurricanes and tropical storms are a minor
threat to the island between August and September
(Applied Technology & Management Inc., 1996).
Although there are no data on food resources for the
translocated lizards during the 2 seasons, higher insect
abundance is likely to be correlated with the increase in
primary production during the wet season, as seen in
other tropical environments (Janzen & Schoener, 1968;
Tanaka & Tanaka, 1982; Wilson, 1991). The wet
season on Praslin is thus considered a high resource
period.
Praslin is covered by 5 main habitat types: (1) mixed
woodland, (2) manchineel woodland, (3) scrub, (4)
bunch grassland, (5) mixed grassland. Mixed woodland
covers approx. 40% of the island. The canopy layer is
comprised of Pithecellobium unguis-cati, Tabebuia
heterophylla and Cornutia pyramidata. The shrub layer,
dominated by Rauvol®a viridis, has a herb layer com-
posed of Spermacoce ernestii, S. verticillata and Ruellia
tuberosa. In contrast, the manchineel woodland consists
of Hippomane manchielle trees with a sparse shrub layer
of R. viridis and Tabernaemontana citrifolia, and little, if
any, herb layer. Bromeliads Tillandsia utriculata are
numerous. Bunchgrass vegetation is found along the
 Population studies of a translocated lizard
crest and north-eastern part of the island and consists
entirely of the tussock grass Schizachyrium microsta-
chyum. Mixed grassland, which is found on the south-
eastern portion of the island, is primarily dominated by
2 herbs (Stachytarpheta jamaicensis and Enicostema
verticullatum) and the grasses Cyperus sphacelatus,
Fibristylis ovata and Digitaria sp. The vines Centrosema
virginianum and Alysicarpus vaginalis are common in
both grassland types.
Morphometry and demography of the Praslin lizards
The Praslin lizard population was censused and indivi-
duals collected during the wet (October±December
1997) and dry seasons (January±March 1998). Lizards
were caught using a hand-held noose made from a
4.5 kg breaking strain, mono®lament line. Sex was
determined by the number of scales above the vent
(male n = 3, female n > 3). Adult status of males was
determined by coloration. Adult males have a sulphur
yellow ventral surface and turquoise tail and ¯anks.
Juvenile males and females are predominantly brown
with cream ventral surfaces. Adult females were categor-
ized as having snout±vent length (SVL) > 76 mm since
they display no secondary sexual characteristics. This
SVL measurement is used by Vitt & Breitenbach (1993)
to de®ne an `average' female Cnemidophorus based on a
review of the literature of the genus. SVL and tail length
were measured with 150 mm callipers or ruler (for tails
> 150 mm), to the nearest 0.5 mm. Body mass (BM) was
determined with a 100-g hand-held spring balance to the
nearest 0.5 g. The general condition of each individual
was noted (i.e. moulting, gravid, cuts, external para-
sites). Information on tail regeneration was also
collected; 4 individuals who dropped their tails during
capture were not included in the analysis.
Morphometric and demographic data were analysed
by age-sex group (i.e. adult males, adult females, juve-
nile males, juvenile females) and pooled across months
(n = 6) and seasons (n = 2). Body size differences were
tested by the 3 main habitat types. Unpublished data on
the founder population (1995, n = 29) and individuals on
Maria Major (Buley, Proir & Gibson, 1997; n = 20) were
pooled and used for source and translocated population
comparisons. Group means are in¯uenced by age struc-
ture bias as there was no ®eld method to determine age
and individuals were not permanently marked.
Mean SVL and BM, age ratios (adults : juveniles), sex
ratios (adult males : adult females), size dimorphism
ratio (SVL adult male : SVL adult females; Stamps,
1983), and a condition index (CI, length-corrected mass;
Wilson, 1991) were calculated. Size dimorphism ratio
employed length instead of mass to avoid bias created
by differences in reproductive status (i.e. gravid females;
Stamps, 1983). CI calculations were restricted to adult
males and adult non-gravid females with no evidence of
tail regeneration (Wilson, 1991) to minimize effects of
high juvenile growth rates (Christian, Grif®ths &
Bedford, 1996).
189
Abundance estimates
Two methods for population size estimation were used
in the present study: (1) a Petersen mark±re-sight survey
(consisting of 1 marking and 1 re-sight event; Krebs,
1989) alongside collection of morphometric data, and
(2) distance sampling, an extension of ®nite population
sampling (Buckland et al., 1996).
For the PETERSEN estimates, lizards caught for
morphometric measurements were individually marked
using colour paint combinations (yellow, orange and
red, `Cryla' heavy body acrylic co-polymer emulsion;
Bracknell, Berkshire, U.K.) on the dorsal surface
between forelegs, hindlegs, midsection and head. Re-
surveying for marked lizards was conducted the
following day. Locations of re-sighted marked lizards
were recorded to the nearest transect line (see below).
The program PETERSEN, in Krebs (1989), was
employed. The equation used is assumed to be unbiased
for small populations and that an individual is counted
only once during re-sighting. For population estimates
involving distance sampling, 9 transect lines spaced
20 m apart were laid at 3508 (magnetic north), perpendi-
cular to habitat changes (Fig. 2). The ®rst line was
randomly located from the west tip of the island. White
painted stones were numbered and placed every 3.0 m
along each line. The total length of transect line was
405.5 m. All transects were walked once per day
(between 09:00 and 11:30 on sunny days without
morning rain) during the last week of each month
(October 1997±March 1998). When a lizard was sighted,
its exact location was ¯agged by pegging a marker on
the spot recording age and sex. Sex was estimated from
colour, length of tail, thickness of the base of the tail
and behaviour. Individuals were regarded carefully for
distinguishing marks and direction of movement to
ensure they were counted only once from the line. All
sightings were made from the line. If another lizard not
previously viewed from the line was disturbed whilst
placing the marker, its location was not marked. When
all transects had been surveyed, the lines were re-walked
and exact location of the lizard was measured to the
line. Perpendicular distances were measured on an east±
west bearing to the line, and the north±south distance
was measured to the closest marker stone.
Abundance estimated from transect data was calcu-
lated using Program DISTANCE (version 2.1; Laake
et al., 1994). DISTANCE ®ts a probability density
function (pfd) to model the decrease in detection of
objects at greater distances from the line. Hazard rate
models (key, cosine adjustment functions, and a
maximum of 2 ®tted parameters) and half-normal
models (key, hermite and cosine adjustment functions,
and a maximum of 2 ®tted parameters were run on each
data set. A truncation distance of 10 m was used in all
analyses. Models were selected through comparison of
Akaike's information criterion, Chi-P value and per
cent coef®cient of variation (Buckland et al., 1996).
Data were analysed by month and season. Raw data
from wet season months and dry season months were
190 H. C. Dickinson and J. E. Fa

Rocks/sand Scrub Bunchgrass Mixed grassland Manchineel woodland Mixed woodland

L6
L3 L4 L5 L7

L8

L2
L1 L9

0 10 m

Fig. 2. Map of Praslin Island showing transect lines used in population estimation.

combined to provide a larger data set for pdf modelling.
Variance in estimates is primarily attributed to the
variation in encounter rate, calculated from a sample
size equal to the number of transect lines. Independence
is violated by searching the same transect line more than
once. Hence, multiplication of line length by the number
of times the transect line is surveyed provides better
precision of the overall encounter rate (i.e. 5 transect
lines of 10 m each surveyed 3 times provides a sample
size of 5, not 15, and a length of 150 m, not 50 m, for
encounter rate estimation). Vegetation changes between
wet and dry season may have affected detection dis-
tances as foliage and vine cover declined. In mixed
wood, scrub and manchineel habitats, the farthest dis-
tance seen from a speci®c location was measured in
October 1997, January and March 1998. Raw transect
data were strati®ed by habitat type.
RESULTS
and 65.7 ‹ 9.4 mm (range 41.0±75.0, n = 15) for juvenile
females. Mean body mass of juvenile males was
15.5 ‹ 7.1 g (range 7.0±28.0, n = 19) and 7.7 ‹ 2.1 g
(range 3.5±10.5, n = 15) for juvenile females. A two-way
analysis of variance (ANOVA) showed a signi®cant
difference in SVL between ages (F = 210, P = 0.04, d.f. =
100) and sexes (F = 87, P = 0.05, d.f. = 100), but no
interaction between effects (F = 0.7, P = 0.44, d.f. = 100).
The same analysis failed to detect a signi®cant difference
in BM between age and sex (age, F = 18, P = 0.15,
d.f. = 100; sex, F = 10, P = 0.19, d.f. = 100), but found a
signi®cant interaction between effects (F = 5.2, P = 0.03,
d.f. = 100).
Body size did not vary between habitat types
(Kruskal±Wallis test, d.f. = 2, P > 0.4). Pooled SVL and
18
16 Full colour
14 Developing colour
No colour
Frequency (%)

12
Morphometrics 10
8
The smallest male to display adult coloration had a SVL
6
of 81.1 mm. All males SVL > 98.5 mm showed adult
colours (Fig. 3). Mean SVL of adult male lizards was 4
105.3 ‹ 10.2 mm. (range 81.0±2.5, n = 38), but smaller 2
for adult females (87.8 ‹ 5.9.0 mm, range 77.0±98.0, 0
n = 32). Mean body mass of adult males was 70 80 90 100 110 120 130
Snout±vent length (mm)
Snout-vent length (mm)
37.8 ‹ 10.6g (range 10.0±56.5, n = 37) and lighter for
adult females (19.5 ‹ 5.4 g, 11.0±33.0 g, n = 32). The Fig. 3. Presence of adult coloration relative to body size
smallest gravid female measured 78.0 mm. Juvenile male (snout±vent length) for Cnemidophorus vanzoi on Praslin
mean SVL was 79.1 ‹ 11.9 mm (range 62.0±98.5, n = 19) Island.
 Population studies of a translocated lizard 191

30 Table 1. Mass±length relationships for each age-sex group of

(a) Cnemidophorus vanzoi on Praslin Island and pooled age class.
25 Adults demonstrate a linear relation whereas juveniles show
20 an exponential relationship; y = body mass (g), x = SVL (cm);
outliers (n = 6) were removed from data adult males before
15
analysis
10
Males
Juv. male Age-sex class Regression equation r2 n
5 Females
Juv. female
Body mass (g)

0 Adults
35 45 55 65 75 85 95 Males y = 1.05 x ± 75.58 0.89 29
Females y = 0.75 x ± 48.97 0.86 31
60 Pooled y = 0.94 x ± 63.68 0.94 60
(b) Juveniles
50 Males y = 0.66e0.04x 0.95 19
40 Females y = 1.02e0.03x 0.76 15
Pooled y = 0.59e0.39x 0.91 34
30
20
Males
10 Females Table 2. Sex and age ratios of Cnemidophorus vanzoi
calculated from distance sampling data. Ratios were calculated
0
70 80 90 100 110 120 130
by month, season (wet season, October±December 1997; dry
Snout±vent
Snout-vent length
length (mm)
(mm) season, January±March 1998) and pooled across all months

Fig. 4. Body mass and snout±vent length for adult and juve- Month Sex ratio Age ratio
nile Cnemidophorus vanzoi caught between October 1997 and (male : female) (adults : juvenile)
March 1998 on Praslin Island. Outliers (n = 6) removed from
October 0.56 2.40
analysis of adult males. (a) Juveniles; (b) adults. For equations November 0.88 1.16
and sample sizes see Table 2. December 0.72 0.98
January 0.73 1.67
February 0.90 1.10
BM were signi®cantly correlated (rs = 0.97, P < 0.001, March 0.62 1.21
d.f. = 96). When sexes were analysed separately, a sig- Wet season 0.68 1.51
ni®cant correlation was found in adult females Dry season 0.75 1.30
Pooled (all months) 0.71 1.41
(rs = 0.71, P < 0.001, d.f. = 31), but not adult males
(rs = 0.24, P = 0.18, d.f. = 29). Juvenile males and females
showed a signi®cant correlation between SVL and BM
( juvenile males: rs = 0.96, P < 0.001, d.f. = 17; juvenile both ratios; seasonal, Mann±Whitney U-test, P > 0.6,
females: rs = 0.81, P < 0.001, d.f. = 13). Adults demon- d.f. = 1 for both ratios).
strated a linear relationship whereas the juveniles'
relationship was exponential (Table 1, Fig. 4).
Body condition

Demographics and dimorphism
The mean sex ratio and age ratio of individuals caught
on Praslin Island was 1.2 males to one female, and two
adults to one juvenile. Observations of juveniles were
highest at the end of the wet season (December, 1997;
Table 2). Sex ratios and age ratios calculated from
distance sampling methods averaged 0.71 and 1.41,
respectively. Bias in sampling methods may affect mean
ratio determination. Juvenile males sighted from a
distance may have been misidenti®ed as females, re-
sulting in a negatively biased sex ratio and positively
biased age ratio. Males may have been easier to catch
with a noose, producing a positively biased sex ratio
(55% of captures were male). Averaging the two esti-
mates gives a sex ratio of approx. 1.0 and age ratio of
approx. 1.7. Size dimorphism (mean male to female
SVL) was calculated as 1.20. No signi®cant difference
was observed in monthly or seasonal ratios (monthly,
Kruskal±Wallis non-parametric test, P > 0.6, d.f. = 5 for
The CI index of adult males was signi®cantly different
from those of adult females (t = 6.1, P < 0.001,
d.f. = 37). Females and juvenile males showed a signi®-
cant decline in CI between the wet season and dry
season (Mann±Whitney U-test; females, Z = 2.2,
P = 0.03; d.f. = 15; juv. males, Z = 3.0, P = 0.002,
d.f. = 12; Fig. 5). The per cent decrease in mean body
mass between seasons was 18% in adult males, 21% in
adult females and 53% in juvenile males. Juvenile
females showed a mean seasonal increase of 6% but this
is an artefact of the small sample size during the wet
season (n = 2). The CI was only signi®cantly correlated
with precipitation in juvenile males (rs = 0.74, d.f. = 12,
P < 0.01; all other age-sex groups P > 0.5).
Evidence of tail regeneration was observed in 35% of
all caught individuals (n = 107). The highest frequency
occurred in adult females (40%), then adult males (38%)
followed by juvenile males (14%) and juvenile females
(8%). Adults had signi®cantly more tail breaks than
juveniles (w2 = 11.9, d.f. = 1, P = 0.001), but there was no
192 H. C. Dickinson and J. E. Fa

3.5
Island and Maria Minor populations (n = 5). However,
0.64
0.50
Wet season SVL and BM measurements of Maria Minor individuals
3 Dry season are much less than average measurements for the
species; individuals appear smaller and are less brightly
0.32*
coloured (Buley et al., 1997).
Condition index

2.5 0.46*
0.45*
2
1.5
0.18*
1 (males = 60%, females = 35%; juvenile males = 5%), and
0.5
0
Males Females Juv.
Juv. males
,ales
Fig. 5. Mean condition index (CI) for each age-sex group on
Praslin Island. Condition indices were calculated for indivi-
duals without evidence of tail autotomy (males, n = 22;
females, n = 17; juvenile ( juv.) males, n = 14; juvenile females,
n = 12). Wet season, October±December 1997; dry season,
January±March 1998. Numbers, 1 s d; *, signi®cant (P < 0.05)
seasonal differences.
difference in tail break frequency between sexes
(w2 = 0.03, d.f. = 1, P = 0.87) or seasons (w2 = 2.2, d.f. = 1,
P = 0.14).
Population comparisons
There were no signi®cant differences in body propor-
tions between adult populations on Praslin Island
(n = 69) and those on Maria Major (n = 30; Mann±
Whitney U-test, SVL; P > 0.24 for all comparisons).
Masses of adult males were also similar (Mann±Whitney
U-test, Z = 0.18, P = 0.86, d.f. = 44), but adult females
were initially found to be heavier on Maria Major
(Mann±Whitney U-test, Z = 4.41, P < 0.001, d.f. = 28).
After controlling for the reproductive state of the Maria
Major population (40% were gravid), there was no
signi®cant difference between female BM. Sample sizes
were too small to test for differences between the Praslin
Tail break frequency was not signi®cantly different
between the source and translocated population (n = 49;
w2 = 0.5, d.f. = 1, P = 0.48). The source population
0.01 0.23 (n = 49) demonstrated 41% frequency of tail autotomy
similar to Praslin Island individuals, adults showed a
signi®cantly higher frequency than juveniles (w2 = 16.2,
d.f. = 1, P < 0.001) but not different between sexes
Juv. females (w2 = 1.8, d.f. = 1, P = 0.18). Condition indices did not
differ signi®cantly between the translocated population
and the source population (Mann±Whitney U-test;
males, Z = 0.53, P = 0.60, d.f. = 43; females, Z = 0.22,
P = 0.82, d.f. = 19).
Abundance estimates
A mean population size of 151.5 ‹ 25.9 was estimated
from pooled overall abundance estimates (range 118±
166) as they did not differ signi®cantly (Tukey multiple
comparison, Q = 1.26, P > 0.05, d.f. = 2). Lizard abun-
dance was signi®cantly different between wet and dry
season (w2 = 11.4, P < 0.001, d.f. = 1).
Abundance estimates generated by the different
models for data on the same month did not vary
signi®cantly (described by ssd; Tables 3 & 4). Monthly
abundance estimates pooled across seasons were not
signi®cantly different (Mann±Whitney U-test, Z = 0.87,
P = 0.4, d.f. = 1). Newman±Keuls multiple comparison
(MSE = 15.2, d.f. = 4, P < 0.05) showed that abundance
estimates for November±January (Q = 2.3) and
December±February (Q < 1.0) were similar. October
differed signi®cantly from March (Q = 31.8) and both
were signi®cantly different from the groups above
(October, Q < 11.5 & Q = 23.3; March, Q = 8.6 &
Q < 17.9, respectively). Seasonal population estimates
calculated from pooled raw data showed low variance

Table 3. Monthly abundance estimates of Cnemidophorus vanzoi in Praslin Island using DISTANCE estimates

Numbers Density Coef®cient Detection Encounter

Month counted Abundance (lizards/ha) of variation Sample s d probability (%) (lizards/m) Model

October 212 196.54 178.67 19.78 15 93 60.13 Hazard rate, cosine

adjustment
November 189 142.50 129.55 16.38 9.6 44 0.11 Half normal
December 98 105.71 96.10 15.78 0.8 41 0.08 Half normal
January 145 151.82 138.02 11.71 9.2 32 0.03 Half normal
February 154 105.12 95.56 9.58 4.5 39 0.07 Half normal
March 125 72.68 66.07 18.41 4.6 45 0.06 Hazard rate
Weta 499 149.18 135.62 15.26 4.8 40 0.11 Hazard rate
Dryb 424 96.37 87.61 8.51 5.8 42 0.07 Hazard rate
Allc 923 118.27 107.52 9.32 1.7 41 0.09 Hazard rate
a
Data pooled over the wet season, October±December 1997.
b
Data pooled over the dry season, January±March 1998.
c
Data pooled October 1997±March 1998.
 Population studies of a translocated lizard 193

Table 4. Monthly abundance estimates of Cnemidophorus vanzoi in Praslin Island using PETERSEN estimates

Lizards Re-sighted Abundance

Month marked (M) with marks Total re-sighted estimate 95% CI Distribution

October 11 3 61 185 81±408 Poisson

November 20 10 90 173 118±284 Binomial
December 14 6 65 140 70±273 Poisson
January 20 7 60 159 98±328 Binomial
February 20 3 64 340 149±749 Poisson
March 19 9 34 69 48±116 Binomial

14 DISCUSSION
October 1997–January 1998
12
Visibilty increase (m)

January–March 1998 In tropical lizards, life-history strategies have evolved to

10
8 nounced wet±dry seasons (James & Shine, 1988; Vitt &
6
4 by lower precipitation and vegetation cover, increased
2 temperatures and lower abundance of invertebrate prey.
0
Scrub
Shrub Manchineel
Manchlineel Mixed woodland
Mixed wood
Fig. 6. Visibility measurements taken from permanent points
in three habitat types on Praslin Island.
and seasonal abundance estimates were signi®cantly
different between wet and dry seasons (w2 = 11.4,
P < 0.001, d.f. = 1; Tables 3 & 4).
Signi®cant changes in visibility appeared between
habitat types (repeated measures ANOVA, F = 15.6,
P < 0.001, d.f. = 3) during October±January, and
January±March (Fig. 6). Because strati®cation by
habitat type resulted in smaller data sets, the observed
variation in abundance estimates was greater. Estimates
by habitat types were not signi®cantly different from
those generated from transect counts, for the wet season
(w2 = 0.9, P = 0.34, d.f. = 1), or overall (w2 = 2.8, P = 0.10,
d.f. = 1), but were signi®cantly different for the dry
season (w2 = 4.3, P = 0.04, d.f. = 1).
PETERSEN estimates for each month showed large
con®dence intervals re¯ecting the small sample sizes
(Krebs, 1989). Comparisons between DISTANCE-
generated and PETERSEN-generated abundance
estimates were similar for all months except February,
where the number of re-sights were low (Tables 3 & 4).
However, mark±re-sight surveys only estimate the
catchable population and theoretically would tend to
underestimate population size relative to estimates pro-
vided by DISTANCE. Depending on the estimate used,
an 8 to 11 fold increase in abundance has resulted from
a minimum founder population of 15 individuals (seven
pairs plus one adult male) over 3 years. Using simple
compound interest equations incorporating 14±20
founders compounding one to two times annually
(representing one to two annual reproductive periods),
the growth rate ranges from 0.97 to 3.95.
accommodate resource ¯uctuations associated with pro-
Blackburn, 1991; Wilson, 1991; Christian & Bedford,
1995). On Praslin Island, the dry season is characterized
Correspondingly, all age-sex groups of the translocated
lizards exhibited a drop in general body condition
during the dry season. This effect has also been observed
in West Indian anoles, where a drop in body mass was
correlated with a signi®cant reduction in invertebrate
availability during the dry season (Lister, 1981; Stamps
& Tanaka, 1981; Floyd & Jenssen, 1983; A. D. Grif®ths
& Christian, 1996). Lower food supplies during the dry
season can also result in reduced growth and activity
(Huey, Pianka & Hoffman, 1977; Dunham, Tinkle &
Gibbons, 1978; Andrews, 1982; Taylor, 1986; Paulissen,
1988; Howland, Vitt & Topez, 1990; A. D. Grif®ths &
Christian, 1996; Smith, 1996), increased mortality, and
lower reproductive success (Tinkle, 1970; Bradshaw,
1986; Marken-Lichenbelt, 1993).
Strong seasonal changes in fat reserves and seasonal
cycling of lipids are common in lizards, with the highest
fat reserves recorded after the wet season (Vitt, 1983;
Taylor, 1986; Dearing & Schall, 1994). Caudal fat
reserves supplement poor-quality diets and can affect
survival during periods of low food availability
(Sweitzer & Berger, 1993; Perez-Mellado, Court &
Laconic, 1997). Low CI are also associated with reduced
growth rates and activity in lizards during dry seasons
in arid environments (A. D. Grif®ths & Christian, 1996;
Smith, 1996). Although no measure of seasonal growth
rate could be determined from our data, the linear
relation between BM and SVL in adults, compared with
the exponential trend in juveniles, suggest that the two
are part of a typically asymptotic growth curve. This
has also been shown for the other insular species,
C. murinus (Dearing & Schall, 1994). The signi®cant
seasonal difference in abundance and the non-signi®cant
difference in active individuals with evidence of auto-
tomized tails (P = 0.14) supports the suggestion of
Casas-Andreu & Gurrola-Hidalgo (1993) that indivi-
duals with suf®cient fat reserves aestivate whilst others
must remain active to search for scarcer food resources
(in C. lineatissimus and C. communis). Aestivation and
194 H. C. Dickinson and J. E. Fa
reduced activity levels are also behavioural responses to
dehydration to reduce evaporative water loss (in
Sceloporus undulatus; Crowley, 1987).
Providing a good estimation of population size
proves challenging and methods designed for reptiles
suffer shortcomings (House & Spellerberg, 1983;
Cueller, 1993; for a review of methods see Turner, 1977;
Vogt & Hine, 1982; Heyer et al., 1994). The mean
overall abundance on Praslin Island is greater than that
predicted (n = 64) by an equation developed from
species-area relationships (Williamson, 1981) or extra-
polated from body size vs density graphs (n = 13) for
vertebrate poikilotherms (Peters, 1983). The signi®cant
seasonal decrease in abundance is taken as a true
decrease as detectability of lizards will have remained
the same or increased during the dry season as a result
of the decline in vegetation cover.
Seasonal CI (length-corrected mass) were signi®cantly
different in adult females and juvenile males. In Chlamy-
dosaurus kingii, male CI were constant whilst female CI
were variable across seasons (A. D. Grif®ths & Chris-
tian, 1996) and female C. ocellifer show a constant
decline in body fat during the dry season (Vitt, 1983).
Changes in male CI are related to energetically expen-
sive activities associated with the breeding season
(Fleming & Hooker, 1975; Howland et al., 1990; Shine,
1990). Tropical Cnemidophorus species have reproduc-
tive seasons of various duration and frequency (Vitt &
Breitenbach, 1993; Dearing & Schall, 1994) but gener-
ally occur during the wet season (Andrews, 1989). From
the calculated age ratios and abiotic conditions neces-
sary for incubation (Vitt & Breitenbach, 1993) and that
copulation was not observed during the study, only
post-breeding season males at an already low CI may
have been sampled. Hence, the low CI may have been
maintained by, rather than re¯ecting, a seasonal change
in resources.
If females were receptive only at the beginning of the
wet season, intense male±male competition would be
expected. Sexual size dimorphism has been used to infer
levels of male aggression (Molina-Borja, Padron-
Fumero & Alfonso-Martin, 1997) and territoriality
(Stamps, 1983), where larger males have a competitive
advantage defending sites or procuring mates (Turner,
Jenrich & Weintraub, 1969; Case, 1978; Thorpe & Baez,
1987). Generally, teiid lizards are non-territorial
(Molina-Borja et al., 1997; Stamps, 1983) as expected
due to patchy invertebrate distributions. The large size
dimorphism ratio observed in C. vanzoi (1.20) re¯ects
other non-territorial, sexually dimorphic species within
the genus (C. arubensis 1.5, C. lemniscatus, 1.22 and
C. murinus, 1.1; Stamps, 1983).
Sexual size dimorphism in C. vanzoi as a result of
intrasexual competition is supported by two results.
First, the frequency of tail autotomy is an indirect
measure of intraspeci®c aggression and predation
(Schall, 1974; Wilson, 1991). Adults had a signi®cantly
higher frequency of tail breaks than juveniles, although
there was no signi®cant difference between sexes. Adult
males may incur injury when competing for or guarding
mates. Cnemidophorus vanzoi has been observed to
guard gravid females (S. Lesmond, pers. comm.). Mate
guarding occurs in other teiid lizard species (e.g.
Eumeces laticeps, Vitt & Cooper, 1984) resulting in
sexual size dimorphism in the absence of territoriality.
Tail autotomy in females may result from oviposition
site defence. Juveniles (which are assumed to be non-
breeders) would escape this reproduction-related, intra-
sexual aggression. Secondly, 16% of males above the
minimum SVL associated with adult coloration main-
tained juvenile colours. Dearing & Shall (1994) have
reported this phenomenon in C. arubensis and
C. murinus (Netherland Antilles) and suggest that
retaining juvenile coloration avoids aggressive
interactions. Both these species demonstrate a low re-
productive rate (limited access to receptive females) and
extreme sexual dimorphism. Aggressive interactions
between males and females were observed on Praslin
Island, usually resulting in the exclusion of one
individual from the vicinity.
Alternatively, tail break frequency may relate to
predation pressure. Six black rats Rattus rattus were
caught on Praslin Island during the ®eld season (see also
Johnston et al., 1994). The nocturnal behaviour of rats
could lead to depredation on lizards whilst in burrows
or crevices (Atkinson, 1985). However, no lizard
remains were found in the stomach contents of two
trapped rats. Regardless of the cause of autotomy, tail
loss is energetically expensive. In addition to the loss of
fat reserves and predation avoidance behaviour,
re-growth incurs diversion of resources from growth
and reproduction and can affect social status, home-
range size and thermoregulatory capacity (Arnold,
1988; Salvador et al., 1996; Perez-Mellado et al., 1997).
Wilson (1991) found that predation decreased during
the dry season as reduced activity levels decreased
predator±prey interactions in Uta stansburiana.
Cnemidophorus vanzoi does not display the general
trend of smaller sized, high density insular populations
described for teiids and lacertids (Case, 1983; Case &
Bolger, 1991). Compared with other island species
(C. arubensis & C. murinus) and two mainland species
(C. lemniscatus; central and north South America;
C. ocellifer, South America; Vitt & Breitenbach, 1993)
within the Lemniscatus species group the reverse trend
appears; the island species are larger. However, the
limited data on the C. vanzoi population on Maria
Minor shows smaller means than the populations found
on Maria Major and Praslin Island. Possible explana-
tions (beyond the intuitive bias in the sample size)
include an adaptive radiation or genetic drift. Morpho-
logical changes on Maria Minor could be related to a
harsher environment (e.g. exposure to elements, lack of
habitat diversity), acute seasonal variation in resources
(e.g. invertebrate density, alternate food sources), and
selection on physiological characteristics (e.g. decreased
surface to volume ratio, increased melanism). The
estimated densities of C. vanzoi on Praslin Island (this
study) and Maria Major (Brice & Bloxam, 1995; Howes,
1995) are much lower than other island species within
 Population studies of a translocated lizard
the genus (C. arubensis and C. murinus) suggesting a
lower carrying capacity of these islands.
This translocation provided an opportunity to collect
data on a little studied species, to investigate the condi-
tion and demography following a colonizing event and
has provided a baseline for future monitoring. The
ultimate goal of the conservation initiative was to
increase the survival potential of the species by reducing
extinction risk (Q. Bloxam, pers. comm.) and C. vanzoi
now exists on three small offshore islands in St Lucia,
West Indies. Small populations and endemic insular
species are associated with greater extinction risk from
the magni®ed effects of demographic, environmental,
genetic and catastrophic stochasticity (Shaffer, 1987).
Environmental variation and demographic stochasticity
particularly impact small populations in the short term
(Leigh, 1981; Pimm, et al., 1988). The small size of
Praslin Island combined with habitat ¯uctuations that
result in seasonal physiological stress could affect the
activity, growth and reproduction in C. vanzoi.
However, the seasonal change in the condition of
individuals implies seasonal resource ¯uctuations on
Praslin Island but this does not appear to have affected
population growth.
No change in body size or condition between the
source and translocated population has occurred during
the 3 years since translocation, probably indicating no
lasting effects of the translocation event. Abundance
estimates showed that a small founder population
successfully populated Praslin Island in 3 years. The
calculated high growth rate in the population suggests
that it is reproducing successfully in its new environ-
ment. This study also suggests that demographic and
environmental stochasticity may have a reduced effect
on extinction risk in this population compared to the
risk associated with catastrophic stochasticity (e.g.
hurricanes or predator introductions).
Acknowledgements
We are most grateful to Wildlife Preservation Trust
Canada, Glenmore Highlands Inc., Trademark Trust
and British Airways Assisting Nature Conservation for
®nancial assistance. Thanks to the St Lucia Forestry
Department for permission to work on the island and for
constant logistic support during the ®eld study. Back-
ground information on the translocation was kindly
made available by Q. Bloxam and K. Buley. HD would
like to thank Mr Stephen Lesmond, the people of Praslin
Village, the Fox Grove Inn, H. & D. Dobson and
D. Gough. P. Ryan, S. Buckland, M. Day, assisted
with ideas, comments and improvements to a previous
draft.
REFERENCES
Andrews, R. M. (1982). Patterns of growth and reproduction in
reptiles. In Biology of the Reptilia 13: 273±307. Gans, C. &
Plough, F. H. (Eds). New York: Academic Press.
195
Andrews, R. M. (1989). Intra-population variation in age of
sexual maturity of the tropical lizard Anolis limifrons in
Panama. Copeia 1989: 751±753.
Anthony, D. (1993). The ecology of the Maria Island ground lizard.
MSc thesis, University of Kent at Canterbury, U.K.
Arnold, E. N. (1988). Caudal autotomy as a defence. In Biology of
the Reptilia 16, Ecology B: Defence and life history: 235±273.
Gans, C. & Huey, R. B. (Eds). New York: Academic Press.
Atkinson, I. (1985). Effects of rodents on islands. In Conservation
of island birds: 35±81. Moors, P. J. (Ed.). Cambridge: Cam-
bridge International Council for Bird Preservation Tech. Publ.
No. 3.
Baskin, J. N. & Williams, E. E. (1966). The lesser Antillean
Ameiva: re-evaluation, zoogeography and the effects of preda-
tion. Stud. Fauna Curacao 23: 144±176.
Bradshaw, S. D. (1986). Ecophysiology of desert reptiles. Orlando,
FL: Academic Press.
Brice, S. & Bloxam, Q. (1995). Understanding translocation possi-
bilities for a teiid lizard in St. Lucia: distribution, density of the
source population and ecological aspects of the target habitat.
Internal Report. United Kingdom: Jersey Wildlife Preservation
Trust.
Buckland, S. T., Anderson D. R., Burnam, K. P. & Laake, J. L.
(1996). Distance sampling: estimating abundance of biological
populations. London: Chapman & Hall.
Buley, K., Prior, K. A. & Gibson, R. C. (1997). In search of the
St. Lucia racer, Liophis ornatus, with further observations on
the reptile fauna of the Maria Islands, St. Lucia, West Indies.
Dodo J. Wildl. Preserv. Trusts 33: 104±117.
Casas-Andreu, G. & Gurrala-Hidalgo, M. A. (1993). Comparative
ecology of two species of Cnemidophorus in coastal Jalisco,
Mexico. In Biology of the whiptail lizard (genus Cnemido-
phorus): 133±150. Wright, J. W. & Vitt, L. J. (Eds). Norman,
OK: Herpetological League Spec. Publ. No. 3.
Case, T. J. (1978). A general explanation for insular body size
trends in terrestrial vertebrates. Ecology 59: 1±18.
Case, T. J. (1983). Sympatry and size similarity in Cnemidophorus.
In Lizard ecology: 297±325. Huey, R. B., Pianka, E. R. &
Schoener, T. W. (Eds). London: Harvard University Press.
Case, T. J., Bolger, D. T. & Richman, A. D. (1992). Reptilian
extinctions: the last ten thousand years. In Conservation
biology: 91±126. Fiedier, P. L. & Jain, S. K. (Eds). London:
Chapman & Hall.
Christian, K. A. & Bedford, G. S. (1995). Seasonal changes in
thermoregulation by the frillneck lizard, Chlamydosaurus kingii,
in tropical Australia. Ecology 76: 124±132.
Christian, K. A., Grif®ths, A. D. & Bedford, G. S. (1996).
Physiological ecology of the frillneck lizards in a seasonal
tropical environment. Oecologia (Berl.) 106: 49±56.
Corke, D. (1987). Reptile conservation on the Maria Islands
(St. Lucia, West Indies). Biol. Conserv. 44: 263±279.
Crowley, S. R. (1987). The effect of desiccation upon the preferred
body temperature and activity level of the lizard Sceloporus
undulatus. Copeia 1987: 25±32.
Cuellar, O. (1993). Lizard population ecology: A long term
community study. Bull. Ecol. 24: 109±14.
Dearing, M. D. & Schall, J. J. (1994). Atypical reproduction and
sexual dimorphism of the tropical Bonaire Island whiptail
lizard, Cnemidophorus murinus. Copeia 1994: 760±766.
Dunham, A. E., Tinkle, D. W. & Gibbons, J. W. (1978). Body size
in island lizards: A cautionary tale. Ecology 59: 1230±1238.
Fleming, T. H. & Hooker, R. S. (1975). Anolis cupreus: the
response of a lizard to tropical seasonality. Ecology 56: 1243±
1261.
Floyd, H. B. & Jenssen, T. A. (1983). Food habits of the Jamaican
lizard Anolis opalinus: resource partitioning and seasonal effects
examined. Copeia 1983: 319±331.
Gilpin, M. & SouleÂ, M. E. (1986). Minimum viable populations:
processes of species extinctions. In Conservation biology:
196 H. C. Dickinson and J. E. Fa
science of scarcity and diversity: 19±34. SouleÂ, M. E. (Ed.).
Sunderland, MA: Sinauer Associates.
Grif®ths, A. D. & Christian, K. A. (1996). Diet and habitat use of
frillneck lizards in a seasonal tropical environment. Oecologia
(Berl.) 106: 39±48.
Grif®ths, B., Scott, J. M., Carpenter, J. W. & Reed, C. (1989).
Translocation as a species conservation tool: status and
strategy. Science 345: 447±480.
Heyer, W. R., Donnelly, M. A., Mcdiarmid, R. W., Hayck, L. C.
& Foster. M. S. (Eds) (1994). Measuring and monitoring
biological diversity: standard methods for amphibians.
Washington: Smithsonian Institute Press.
House, S. M. & Spellerberg, I. M. (1983). Ecology and conserva-
tion of the sand lizard (Lacerta agilis) habitat in southern
England. J. appl. Ecol. 20: 417±437.
Howes, L. (1995). Computer simulation for predicting the success of
Cnemidophorus vanzoi translocation to Praslin Island, St. Lucia.
Diploma thesis, Canterbury: University of Kent.
Howland, J. M., Vitt, L. J. & Topez, P. T. (1990). Life on the
edge: The ecology and life history of the tropidurine iguanid
lizard Uranoscodon superciliosum. Can. J. Zool. 68: 1366±1373.
Huey, R. B., Pianka, E. R. & Hoffman, J. A. (1977). Seasonal
variation in the thermoregulatory behaviour and body
temperature of diurnal Kalahari lizards. Ecology 58: 1066±
1075.
Ianson, G. R. (1989). Mortality of mountain hares in relation to
body size and age. J. Zool. (Lond.) 219: 676±680.
IUCN (1987). Position statement on the translocation of living
organisms: Introductions, re-introductions, and re-stocking.
Gland, Switzerland: IUCN Council.
James, C. D. & Shine, R. (1988). Life-history strategies of
Australian lizards: a comparison between the tropics and the
temperate zone. Oecologia (Berl.) 755: 307±316.
Janzen, D. H. & Schoener, T. A. (1968). Differences in insect
abundance and diversity between wetter and drier sites during a
tropical dry season. Ecology 49: 96±110.
Johnston, J. P., Anthony, D. & Bloxam, Q. (1994). Eradication of
rats from Praslin Island, St. Lucia. Dodo J. Wildl. Pres. Trust.
30: 114±118.
Kennish, R. (1997). Seasonal patterns of food availability: in¯u-
ences on the reproductive output and body condition of the
herbivorous crab, Grapsus albolineatus. Oecologia (Berl.) 109:
207±218.
Krebs, C. J. (1989). Ecological methodology. New York: Harper
Collins.
Laake, J. L., Buckland, S. T., Anderson, D. R. & Burnam, K. P.
(1994). Distance user's guide version 2.1. Colorado: Colorado
State University.
Leigh, E. G. (1981. The average lifetime of a population in a
varying environment. J. Theor. Biol. 90: 231±239.
Lister, B. C. (1981). Seasonal niche relationships of rainforest
anoles. Ecology 62: 1548±1560.
Marken-Lichtenbelt van, W. D. (1993). Optimal foraging of a
herbivorous lizard, the green iguana, in a seasonal environ-
ment. Oecologia (Berl.) 95: 246±256.
Miller, B., Ralls, K., Reading, R. P., Scott, J. M. & Estes, J.
(1999). Biological and technical considerations of carnivore
translocation: a review. Anim. Conserv. 2: 59±68.
Miura, S. & Maruyama, N. (1986). Winter weight loss in Japanese
serow. J. Wildl. Manage. 50: 336±338.
Molina-Borja, M., Padron-Fumero, M. & Alfonso-Martin, M. T.
(1997). Intrapopulation variability in morphology, coloration
and body size in two races of the lacertid lizard, Gallotia galloti.
J. Herpetol. 31: 499±507.
Paulissen, M. S. (1988). Ontogenetic and seasonal shifts in micro-
habitat use by the lizard Cnemidophorus sexlineatus. Copeia
1988: 1021±1029.
Peters, R. H. (1983). The ecological implications of body size.
Cambridge: Cambridge University Press.
Perez-Mellado, V., Court, C. & Laconic, P. (1997). Tail autotomy
and extinction in Mediterranean lizards. A preliminary study
of continental and insular populations. J. Zool. (Lond.) 243:
533±541.
Pimm, S. L., Jones, H. L. & Diamond, J. (1988). On the risk of
extinction. Am. Nat. 132: 757±785.
Salvador, A., Martin, J., Lopez, P. & Veiga, J. P. (1996). Long
term effects of tail loss on home range size and access to
females in male lizards (Psammodromus algirus). Copeia 1996:
208±209.
Schall, J. J. (1974). Population structure of the Aruban whiptail
lizard, Cnemidophorus arubensis, in varied habitats. Herpetolo-
gica 30: 38±44.
Shaffer, M. (1987). Minimum viable populations: coping with
uncertainty. In: Viable populations for conservation: 69±86.
SouleÂ, M. E. (Ed.). Cambridge: Cambridge University Press.
Shine, R. (1990). Function and evolution of the frill of the
frillneck lizard, Chlamydosaurus kingii (Sauria: Agamidae).
Biol. J. Linn. Soc. 40: 11±20.
Smith, G. R. (1996). Habitat use and its effect on body size
distribution in a population of the tree lizard, Urosaurus
ornatus. J. Herpetol. 30: 528±530.
Stamps, J. A. (1983). Sexual selection, sexual dimorphism and
territoriality. In Lizard ecology: 169±204. Huey, R. B., Pianka,
E. R. & Schoener T. W. (Eds). Cambridge, MA: Harvard
University Press.
Stamps, J. A. & Tanaka, S. (1981). The relationship between food
and social behaviour in juvenile lizards (Anolis aeneus). Copeia
1981: 422±434.
Swartz, A. & Hendersen, R. W. (1991). Amphibians and reptiles of
the West Indies: description, distribution and natural history:
385±386. Florida: University of Florida Press.
Sweitzer, R. A. & Berger, J. (1993). Seasonal dynamics of mass
and body condition in Great Basin porcupines (Erethizon
dorsatum). J. Mammal. 74: 198±203.
Tanaka, L. K. & Tanaka, S. K. (1982). Rainfall and seasonal
changes in arthropod abundance on a tropical oceanic island.
Biotropica 14: 114±123.
Taylor, J. A. (1986). Seasonal energy storage in the Australian
lizard, Ctenotus taeniolatus. Copeia 1986: 445±453.
Thorpe, R. & Baez, M. (1987). Geographic variation within an
island: Univariate and multivariate contouring of scalation,
size and shape of the lizard Gallotia galloti. Evolution 41: 256±
268.
Tinkle, D.W. (1970). Comments on laboratory survivorship in
two species of the lizard genus Uta. Copeia 1970: 381±383.
Turner, F. B., Jenrich, R. I. & Weintraub, J. D. (1969). Home
ranges and body size of lizards. Ecology 50: 1076±1081.
Turner, F. B. (1977). The dynamics of populations of squamates,
crocodilians and rhynchocephalians. In Biology of the Reptilia
7: 157±262. Gans, C. & Tinkle, D. W. (Eds). New York:
Academic Press.
Vitt, L. J. (1983). Reproduction and sexual dimorphism in the
tropical teiid lizard, Cnemidophorus ocellifer. Copeia 1983: 359±
366.
Vitt, L. J. & Blackburn, S. R. (1991). Ecology and life history of
the viviparous lizard Mabuya bistriata (Scincidae) in the Brazi-
lian Amazon. Copeia 1991: 916±927.
Vitt, L. J. & Breitenbach, G. L. (1993). Life histories and
reproductive tactics among lizards in the genus Cnemidophorus
(Sauria: Teiidae). In Biology of the whiptail lizard (genus
Cnemidophorus): 211±239. Wright, J. W. & Vitt, L. J. (Eds).
Norman, OK: Herpetology League Spec. Publ. No.3.
Vitt, L. J. & Cooper, W. E., Jr (1984). The evolution of sexual
dimorphism in the skink Eumeces laticeps: an example of sexual
selection. Can. J. Zool. 63: 995±1002.
Vogt, R. C. & Hine, R. L. (1982. Evaluation of techniques for
assessment of amphibian and reptile populations in Wisconsin.
In Herpetological communities: a symposium of the society for
 Population studies of a translocated lizard 197

the study of amphibians and reptiles: 201±217. Scott, N. J. (Ed.).
Washington, DC: U.S. Department of Interior, Wildlife
Research Report No. 13.
Williamson, M. (1981). Island populations. Oxford: Oxford
University Press.
Wilson, B. S. (1991). Latitudinal variation in activity season
mortality rates of the lizard Uta stansburiana. Ecol. Monogr.
61: 393±414.
Wright, J. W. (1993). Evolution of the lizards of the genus
Cnemidophorus. In Biology of the whiptail lizard (Genus Cnemi-
dophorus): 27±81. Wright, J. W. & Vitt, L. J. (Eds). Norman,
OK: Herpetological League Spec. Publ. No. 3.