BEHAVIOR AND ECOLOGY OF THE
ROCK IGUANA CYCLURA CARINATA
JOHN BURTON IVERSON III
A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
To the peoples of the Turks and Caicos Islands, for their
assistance in making this study possible, with the hope that they
might better see the uniqueness of their iguana and deem it worthy
I wish to express my sincere gratitude to Walter Auffenberg,
chairman of my doctoral committee, for his constant aid and encourage-
ment during this study. Thanks also are due to the other members of
my committee: John Kaufmann, Frank Nordlie, Carter Gilbert, and
I am particularly grateful to the New York Zoological Society for
providing funds for the field work. Without its support the study would
have been infeasible. Acknowledgment is also given to the University of
Florida and the Florida State Museum for support and space for the
duration of my studies.
Of the many people in the Caicos Islands who made this study
possible, special recognition is due C. W. (Liam) Maguire and Bill and
Ginny Cowles of the Meridian Club, Pine Cay, for their generosity in
providing housing, innumerable meals, access to invaluable maps, and
many other courtesies during the study period.
Special thanks are also due Francoise de Rouvray for breaking my
monotonous bean, raisin, peanut butter, and cracker diet with incompar-
able French cuisine, and to Gaston Decker for extending every kindness
to me while on Pine Cay. They enriched my visits to the islands more
th.r. any other tw'o persons. For similar courtesies I also thank George
and iarou Nipanich. I am also indebted to many other Turks and Caicos
islanders for information and help throughout my study.
Don Correll, Bob Anderson, and Walter Auffenberg supplemented my
plant collections and identified most of the plants. Don Buden supplied
a list of birds recorded from the lurks and Caicos Islands. Arthropod
identifications were made by Bob Woodruff (insects), and Martin Muma
(solpugids and scorpions).
Walter Auffenberg and David Auth unselfishly allowed me access to
their Caicos Islands field notes. Diderot Gicca and Tom Wiewandt
generously shared with me their field experiences with other species of
Cyclura. Aerial photography of the study islands was made possible by
Bill and Ginny Cowles. Numerous others have, in one way or another,
added to the successful completion of this work; I apologize for my
failure to acknowledge each one individually.
My wife, Sheila, typed the numerous manuscript drafts, endured
my frequent absences from home, and supported me in many ways during
all phases of the work. All illustrations are the work of the author.
TABLE OF CONTENTS
ACKNOWLEDGEMENTS .. . . . ........
ABSTRACT . . . . . . . . . .
INTRODUCTION . . .
Purpose . . . . .
Systematic Relationships and
Study Area . . . .
Materials and Methods .
MORPHOLOGY . . . . . . . . . . . . . .
HABITAT . . . . . . . . . . . . . . .
Climate. .......... . . . . ......
So il . . . . . . . . . . . . . . .
Vegetation . . . . . . . . . . . . .
Burrows . . . . . . . . . . . .
REPRODUCTION . . . . . . . . . . . . . .
Male Sexual Cycle. . . . . . . . . . . .
Female Sexual Cycle . . . . . . . .. .
Courtship and Mating .. . . .
Neszing .. .... ..............................
Eggs, Incubation and Hatching. . . . ... . . ..
Fertility, Prenatal and Natal Mortality .. . . .....
General Seasonal Reproductive Pattern .. . . .....
Reproductive Effort and Strategy .. . . ........
GROW . . . . . . . .
General Pattern . . . .
Factors Affecting Growt . . . .
Longevity . . . ...........
Regeneration . . . . . . .
Comparisons With Other Iguanines . ..
FOOD AND FEEDING ....... . . . .
D ie t . . . . . . . . .
. . . . 1
. . . . 1
. . . . . 2
. . . . . 3
. . . . . 9
. . . . . 10
. . . . 136
. . 136
. . . 7 1
. . 172
. . 172
. . 173
. . 180
Feeding Behavior ....... . . . . ..... 190
Gastrointestinal Tract Anatomy . . . . . . 202
Digestive Efficiency ..... . . . . . . .. 211
Fat Bodies .... . . . . . . . . . 213
Food as a Limiting Factor. . . . . . . ..... 217
Herbivory as a Feeding Strategy. . . ... ...... . . 220
ACTIVITY AND MOVEMENTS . ... ..... . . . . . 224
Locootion . . . . .... ..... . . . . 224
Retreats ...... .. .. . . ....... .. . 225
Activity Cycles. . . ... ..... ... . . . 225
SOCIAL ORGANIZATION. . . . ... . . . . . . . 255
Displays . . . . . . . .. .. . . . . 255
Dominance Relationships. ... . . . . . . . 272
INTERSPECIFIC COACTIONS. . . ... .. . . . . . 286
Natural Cohabitants .. . . . . .. ... . . 286
Introduced Cohabitants ..... . . . . . . 287
Broken Tail Frequency. . . . . . . . . . . 292
Epifauna ....... . . . . . . . . . 298
Endofauna . . . . ...... . . . . . 299
DENSITY AND DEMOGRAPHY .... . . . . . . . 305
Sex Ratio. . . . . . . .. .. ....... ..... . 305
Characteristics of Juvenile Population . . . . ... 305
Characteristics of Adult Populations . . . . .. . 314
Life Tables . . . .. . . . . . . . 328
Life History Strategy. . . . . . . . . .. 332
OUTLOOK. . .................. .... ...... 337
Populations on Turks and Caicos Islands. . . . . ... 337
Basic Lizard Requirements. . . . . . . . 347
Population Control ................ .. . . 349
Conclusions . .. . ............ . .. .. 352
LITER TURE CITED . . . . . . . . .... . 353
IOCGRAPHICAL SKETCH ......... . . . . . .. . 379
Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
BEHAVIOR AND ECOLOGY OF THE
ROCK IGUANA CYCLURA CARINATA
John Burton Iverson III
Chairman: Walter Auffenberg
Major Department: Zoology
The natural history and social behavior of the rock iguana, Cyclura
carinata, was studied during 25 weeks between September, 1973, and June,
1976, on several small cays in the Turks and Caicos Islands, British
West Indies, and in captive enclosures in Gainesville, Florida.
Reproductive cycles were synchronized to climatic cycles. Testes
sizes were maximal in April and May (the end of the dry season).
Vitellogenesis began in January (beginning of the dry season); cvulation
and mating occurred in early May. Courtship was typical of most iguanid
lizards; the mating system was probably primarily monogamous with serial
polygyny among some males.
A single annual clutch of two to nine eggs was laid in early June
in the terminal portion of the female's retreat burrow. Clutcn size was
positively correlated with female size. Clutch weight averaged about
25% of preoviposition weight. Females defended their nest burrows for
several days to several weeks after nesting, but were not territorial
during the remainder of the year. Hatching occurred after approximately
90 days. Neonates averaged 79.8 mm SV and 14.6 gm. Juvenile growth
rate averaged 19 2 mm SV/year. Males reached s-xual maturity at
approximately 220 mm SV and 375-475 gm, at an age of about seven years;
females, 185-200 mm, 200-300 gm, and six to seven years. Adult males
averaged 276 mm SV and 935 gm; adult females, 225 mm and 475 gm. The
largest male measured 360 mm SV and 1864 gm; largest female, 292 mm and
1135 gm. Adults grew between 2 and 17 mm SV/year.
Rock iguanas were primarily herbivorous at all ages, and fed
arboreally and terrestrially. Five partial septa partition the proximal
colon region; this modification presumably increases digestive efficiency.
Huge oxyurid nematode populations inhabited the colon, and their relation-
ship with the iguana may be mutualistic rather than parasitic. Abdominal
fat deposits were negatively correlated with gonadal activity.
Home range was correlated with body size and habitat productivity,
and averaged 980 m2 for females, 1260 m2 for subdominant males, and 1590
m2 for dominant males. Diurnal activity was bimodal during warmer
months; high midday temperatures resulted in reduced activity. Winter
activity was generally reduced and peaked at midday. Iguanas basked
following emergence; most of the remainder of the day was spent feed-
ing, interacting with other lizards, and shuttling in the sun-shade
mosaic for thermoregulatory purposes. Basking preceded the termi-
nation uf daily activity
Rock iguanas performed ritualized "signature" displays or head-
bobs. ?ostural adjustments were among the modifiers producing variation
in this display in different behavioral contexts. Displays were important
to territory declaration, male challenging, and sex recognition. Males
were territorial year-round in the field, but developed dominance
hierarchies in captivity. Dominant males displayed less than subor-
dinates. Male territorial defense probably evolved as a mechanism
guaranteeing access to 1) food resources, and 2) females for breed-
Sex ratio was 1:1 in all age classes. Survivorship was positively
correlated with body size. Juvenile densities reached 90.3/ha in
optimum habitat, adults exceeded 31.1/ha. Biomass estimates were 5.15
kg/ha for juveniles and 17.01 kg/ha for adults. Life table data
indicate replacement rate was near unity, and mean generation time
was 14.0 years. In its life history strategy, C. carinata is a
predominately K-selected species.
Cyclura are vulnerable to few native predators, but suffer
drastic population reductions with the introduction of feral mammals.
Virtually the entire iguana population of one study island (> 15,000
individuals) was extirpated by dogs and cats during the course of this
study. The status of Cyclura carinata in the Turks and Caicos Islands
was assessed and a management program congruent with the most critical
aspects of the life history of the rock iguana was recommended.
The genus Cyclura is one of the most poorly known lizard genera in
North America, despite being one of the most conspicuous members of the
West Indian Fauna. Although a sizeable literature exists regarding the
genus, most of the references are faunal lists or general works and
contain only anecdotal information; no thorough autecological study of
any of the species has been made.
Few New World lizards have experienced as intimate a relationship
with man, and suffered more from the interaction, than Cyclura. Major
populations have been extirpated within historic times. Probably no
presently existing population is entirely free from pressures induced
by man and his animals. It is increasingly clear that a thorough
knowledge of the habits and adaptations of our fauna is necessary if we
are to be able to insure their survival in the face of continuing and
increasing cohabitation with humans. It was for these reasons that
study was begun on a still relatively common species: Cyclura carinata,
the Turks Island iguana. From September, 1973, through June, 1976,
C. carinata was studied both in the field in the Turks and Caicos
Islands and in the laboratory.
The goals of the study were: 1) to obtain data on the behavior,
adaptations, life history, and community role of this species; 2) to
determine what factors limit population size; 3) to propose methods by
which the species might survive cohabitation with man, without threat
of extinction; and 4) to provide an indepth charter study for extrap-
olation to similar investigations of other less common and more
endangered species of Cyclura in the West Indies.
Systematic Relationships and Distribution
The genus Cyclura is most closely related to the three largest
herbivorous iguanins lizard genera inhabiting continental North and
Central America: Ctenosaura, Iguana, and Sauromalus. On the basis of
myology and osteology Cyclura is most similar to Ctenosaura and next
most similar to Iguana (Avery and Tanner, 1971). Endemic to the
Antilles and Bahamas, Cyclura apparently evolved from a pre-Ctenosaura
stock isolated in the islands. The genera Amblyrhynchus, Brachylophus,
Conolophus, Ctenosaura, Cyclura, Dipsosaurus, Enyaliosaurus, Iguana, and
Sauromalus comprise the subfamily Iguaninae (Avery and Tanner, 1971).
There are seven extant species comprising the genus Cyclura,
distributed from the northern Bahamas through the Greater Antilles
(Schwartz and Thomas, 1976). Pleistocene remains are known from
several islands within the present range (review in Carey, 1975).
Cyclura carinata is closely allied to C. rileyi (San Salvador,
Bahamas). The two species share a number of characters (lack of enlarged,
tubercular, median frontal shields; presence of dorsal crest, with
elongate spines, interrupted on shoulders and rump; maximum size less
than 1 meter, etc.) and are both distributed in the southern half of the
The relationship of Cyclura ricordi (Hispaniola) to C. carinata
deserves additional study despite the larger size (commonly exceeding
one meter total length) of the former. The two species have similar
dorsal crests, pigmentation patterns, and are the only two species of
Cyclura without enlarged supracephalic scales.
The Turks and Caicos Islands lie to the southeast of the Bahama
Islands between latitude 210 and 220 N and longitude 710 and 72031' W
(Figure 1). Although politically distinct from the Bahama Islands,
this British Crown Colony is geologically part of the Bahaman Archi-
pelago. The islands are situated approximately 150 km north of
Hispaniola and 890 km southeast of Miami, Florida. They consist of
two groups of islands separated by a 35.5 km wide, deep water (> 2200 m)
channel, the Turks Island Passage. The Turks Islands lie east of the
passage, the Caicos, to the west. The former consist of two inhabited
islands (Grand Turk and Salt Cay), six uninhabited cays and numerous
The Caicos Islands lie along the perimeter of the Caicos Bank, a
northwest-to-southeast-lying triangular shoal with a base and altitude
of approximately 125 and 75 km, respectively. Depths frequently exceed
180 m within 2 km of the reefs surrounding the Turks and Caicos banks.
Each of the banks is surrounded by depths of at least 1800 m. The
highest elevation in the islands barely exceeds 85 m (on Providenciales).
The present surface area of the islands, as calculated by the Turks
and Caicos Survey Department, is approximately 500 km, with only
27 km2 in the Turks Islands group. Recent evidence indicates that
shoaling sand is filling many of the cuts, linking previously separated
cays via sandy isthmuses (Noble and Klingel, 1932).
_ _I_ _1 _~_1_1__1__
Although these changes are usually not rapid, the effects of
tropical storms can be. When Hurricane Donna passed directly over the
Caicos Islands in 1960 the cut between Pine and Water Cay was blocked
by sand, connecting the islands. Shoaling since then has nearly
filled the cut. These changes are now and probably have been in the
past, very important to the dispersal of fauna and flora between islands
of the bank.
My studies of Cyclura carinata in the Turks and Caicos Islands
were predominately based in the West Caicos Cays. The major islands
comprising this group are, from north to south, Parrot, Dellis, Fort
George, Pine, Water, and Little Water Cay (see inset, Figure 2).
Studies reported here were confined to the latter four. These islands
are phenetically very similar. Each has its leeward sandy beaches,
windward rocky coasts and mangrove swamps, inland brackish water lakes,
northeast-to-southeast lying karst ridges paralleled by sandy dunes,
maximum elevations of eight meters, vegetation increasing in luxuriance
from west to east, and dense Cyclura populations.
Study was concentrated on Pine Cay, a privately owned island of
nearly 350 ha (Figure 2). The average elevation on Pine Cay is less
than three meters. Only three points on the island exceed six meters;
the highest point on the island barely exceeds eight meters. Extraor-
dinary in the Bahama Archipeligo, Pine Cay lies over an impermeable basin
preventing salt water intrusion. Maintained solely by rainfall, a
freshwater lens sometimes exceeding 15 m in thickness occupies this
banin. Average storage capacity has been estimated at over 510 million
liters. The water table is exposed in seven inland lakes. This
availability of fresh water is in part responsible for the presence
Figure 2. Topographic map of Pine Cay, Caicos Islands, showing
locations of principal study sites: 1.) SW Blind, and 2.) Ridge
Area. Letters A through F indicate sectors of flush transect route
(dotted line). Other trails and roads are not illustrated. Dashed
line encloses limits of Hotel grounds. Unnumbered shaded areas
on Pine Cay of an extensive stand of Caribbean pine (Pinus caribaea)
found elsewhere in the Bahamas only on North Caicos, Grand Bahama,
Abaco, Andros, and New Providence (Rabb and Hayden, 1957). Vegetation
on Pine Cay varies from sparse on the most recent westerly sand beach
ridges to dense on the geologically older coral ridges to the east with
better developed soils (see Vegetation).
During the summer of 1973, construction began on a private resort
hotel, the Meridian Club, on Pine Cay. I was thus presented with an
opportunity to study the consequences of increased human interference
on theretofore relatively undisturbed iguana populations.
The Turks and Caicos fauna is part of a larger southern Bahaman
faunal group which includes the islands southeast of the Crooked Island
passage plus Rum Cay and Watling's Island farther north. At least 19
reptiles are endemic to the area, and although strongly divergent, most
have a Greater Antillean ancestry (Schwartz, 1968). This is despite
the fact that the Bahaman platform has had its closest geological
relationship with the North American tectonic plate since Triassic
times (Freeland and Dietz, 1971).
Among the amphibians, only Hyla reaches into the southern Bahamas
from the north and then only to the Mayaguana passage. None are
associated with Cyclura carinata. Over 115 species of birds have been
identified on the Turks and Caicos Islands (Buden, pers. comm.).
Differentiation of at least tuo of these has occurred within the
southern Bahanan faunal region: Columbina passerina and Calliphlox
evelynae (Bond, 1961). The avian species most important to the biology
of Cyclura will be discussed under sections concerning specific relation-
ships (see INTERSPECIFIC COACTIONS and FOOD AND FEEDING). Among the
depauperate mammalian fauna only the bat Monophyllus redmani has
apparently diverged within the southern Bahamas (Buden, 1975).
Materials and Methods
Preliminary evaluations of Cyclura populations in the Turks and
Caicos Islands were made by Walter Auffenberg between August 4 and 14,
1973- Those investigations revealed Pine Cay in the West Caicos Islands
to be the most feasible island for study for several reasons, including
the availability of water and adequate lodging, and the presence of an
extremely dense Cyclura population. Specific study sites on Pine Cay
were designated during my first visit in September, 1973.
Eleven trips to the Caicos have been made, with 163 days spent on
Pine Cay and nearby islands (September 15-29, December 8-22, 1973; March
2-16, June 1-July 20, August 29-September 10, November 26-December 6,
1974; April 1-14, July 29-August 16, November 4-11, 1975; and May 1 to 8,
June 8 to 15, 1976). Twelve additional days were spent observing and
assessing Cyclura populations on numerous eastern islands of the Caicos
Bank. Field notes from the Caicos islands were made available by Walter
Auffenberg (48 study days from September, 1973 through 1975) and David
Auth (112 days from June, 1974 through February, 1976). Copies of all
of our field notes are filed in the Herpetological Library of the Florida
A site of approximately 0.9 ha in the southwestern corner of Pine
Cay was chosen as my major study area to be used mainly for behavioral
observations with as little interference and manipulation as possible on
my part (Figure 3). The area was initially chosen due to the abundance
of lizard spoor and manure on open areas, and the generally good visi-
bility below the canopy.
An abundance of signs typically reflects lizard density in the
immediate area. Since Cyclura carinata on Pine Cay were extremely shy,
it was necessary to construct a blind in this southwestern study site.
Lizards generally ignored the blind after it was in place for only one
day. A total of 47 days (287 hours of actual observation) was spent in
this bline during the study period. In order to accurately quantify
observed lizard movements, yellow marker flags, spaced six meters apart,
were placed in a measured grid system around the blind. Each marker was
assigned a letter N (north of the blind) or S (south), and a number
(related to the distance from the blind). Thus a lizard at "N-5 West
three meters" was located three meters west of marker number "N-5". This
greatly increased the speed and accuracy of data recording, and allowed me
to tabulate precise lizard movements. Over the course of field work, this
entire study area was very thoroughly and accurately mapped. Surveyors and
topographic maps, prepared under the direction of Liam Maguire of the
Meridian Club and based on numbered cement lot markers on Pine Cay, and
aerial photos (personally taken) made possible precise mapping of lizard
and other landmark locations.
Field notes were recorded on a cassette tape recorder each day and
transcribed in the field notebook at night. A Minolta SRT 101 35 mm camera,
SonyAV 3400 portable video camera and recorder, and CAFST/602 Super 8 movie
camera were all employed to record lizard behavior. In excess of 1400 slides,
110 minutes of videc and 50 feet of movie film were taken. The utmost care
and patience was always taken at the southwestern blind to minimize inter-
Figure 3. SW Blind Study Area
i i BANK
\ x \
\0 + 4
1/ + '
SV\ 3UNVEGETATED APEAS
-3- CO OUR INTERVALS (m
S\-R ROCKY CLIFF
0\ ~ EACH
D DC', NANT MALE RETREATS
LAK OilE:ER RETREAIS
L /+ DISTANCE MARES
i / + DISTANCE MARKERS
ference with the natural system. Early in the study the resident popu-
lation was marked without handling by means of a "Paint Rifle", a large
plastic syringe firmly mounted on the end of a 1 m, 2 cm diameter, dowel.
Maximum range exceeded ten m, with accuracy possible to six to seven m.
Lizards were not at all bothered by this technique. By shooting various
colors of paint on various parts of the body, recognition of individuals
was possible. Index cards with outline drawings of the right and left
sides of a lizard were used to record marking and color formation, as
well as sex and natural external morphological data for each lizard.
The cards were modified as parts of the paint pattern were shed and/or
the lizards were remarked. Even after three month absences, remnants of
paint patterns, in conjunction with activity range and behavioral
information, could be used to re-establish the identity of individual
As the study progressed, lizards were occasionally noosed from the
blind for accurate measurement. In order to attract lizards to the blind
for this purpose, as well as to induce lizard interaction, numerous
substances were experimented with as "baits". Among the items tested
were local fruits and flowers, soda and Ritz crackers, peanut butter,
vanilla extract, crab parts, other vertebrate carrion, and sardines.
The latter was the only substance that proved effective as a bait, and
its effect was striking; adult males would occasionally cross the
territories of at least three other territorial males to obtain the
sardines. This bait was, however, less effective during the colder
parts of the year and also when high winds rapidly dissipated the fishy
odor. Foreign (to the study site) lizards, restrained in hardware
cloth cages or on tethers, were sometimes introduced into the activity
ranges of established lizards and the resulting behavioral interactions
Lizards on islands with even minimal human traffic were difficult
to capture by noosing except from blinds, so this technique was not
frequently used on Pine Cay, except at the southwestern blind study
site (hereafter referred to as the SW Blind). The technique was,
however, fairly successful on islands rarely visited by humans.
Juvenile lizards and occasional adults could be secured by chasing
them under rocks or into hollow sticks and logs, from where they could
be removed by hand. Noose snare traps and funnel traps at burrow
mouths in areas of great activity were each experimented with, but
met with only minimal success. Lizards were sometimes obtained by
removing them from their burrows. Nests discovered in this process
were either removed to the laboratory or marked for further observations.
Numerous freshly killed or mortally wounded individuals were also secured
from the mouths of local dogs and cats. In addition, the dried remains
of many mammal-killed Cyclura were collected. An attempt was made to
collect a sample of at least a few lizards during each season of the
year to be used for dissection. Besides those lizards killed by dogs
and cats, the bulk of this sample included lizards removed from the area
on Pine Cay destined for occupation by a hotel and its adjoining beach
cottages. Since most of these lizards would have been killed by bull-
dozer activity, this seemed the best source of material for dissection.
Specimens for dissection were weighed, measured, and preserved in
formalin in the field. Transferred to isopropyl alcohol in the laboratory,
preserved specimens were autopsied for reproductive condition, gut
content, fat bodies, and presence of parasites. Ovaries, testes, and
fat bodies were excised, blotted, and weighed to the nearest 0.01 gm.
In females, all yolked ovarian follicles, oviducal eggs, and corpora
lutea were measured to 0.1 mm with dial calipers. Testes volume was
determined by water displacement.
In order to collect data on growth, lizards located in the area
immediately adjacent to the site chosen for construction of the hotel
were marked and released on the first visit to the study island. On
each of four succeeding visits within a year, no recaptures were made
despite a continued marking program all along the dunes parallel to the
northwest coast. The disappearance rate in this habitat was obvious
on my first return to the island (December, 1973) when not a single
iguana was seen in my original mark-recapture area. A second site was
sought, one less disturbed and more densely populated. In July, 1974,
another mark and recapture program was initiated on an area of approxi-
mately 1.85 ha on the north end of nearby Little Water Cay (Figure 4).
Habitat and vegetation there are virtually identical to the areas of
Open Scrub on Pine Cay (see HABITAT). Growth data were also collected
on numerous juveniles maintained in laboratory on lettuce, cat food,
and bean sprouts.
Body length (snout to vent), tail length and condition, body
weight (measured to the nearest five gm on a Chatillon 3 kg spring
balance), sex (when possible), and any other obvious external morpho-
logical characters (especially in juveniles) were recorded for all
lizards during the mark and recapture program. Maximum height (in mm)
and number of annuli on dorsal spines of the neck, back, and proximal
tail regions were also recorded for each adult. Each lizard was then
0 -C f 0.
o-- 1 (
C C I I ---
--O L -- Q.
zL In a0
u rM a .
l.. E- E m
I 0- r > O
lI C ) C 4- -
LIo 4- LI
-0 C0 c
0 C >
0O 0 C O
SC 4- E
CL 0 O C
Ira r a o L
S- c- O L
0aj ro C >c
-- a- -0
O U -
S*C- 0 '--
0 )0 L C L (o
0 c wMi
+ n 0 0 c -o
;' ON- .
5- -- Io a EI
*- 0 H -4
C C 4-M w
C 0 0 c I
O LI 0)
S- o C
Q 0 vio* 0U
Eo rn a-
*- 0 >0
Oa)C L c
0 ) -- =
0 c -- cO 0
L L C C -
4 Ii m 4- 0
SUO CC E
Sl-- C 0 Co
LI LI L 0 1- 0
S- o aI
-I .- LO
C .- LI
OO LIcC u
LC r 44 0 11
marked with a unique combination of at most two clipped digits, and
released as soon as possible following capture (always within the hour).
At the SW Blind "belts" of various colors modified from cat collars
were secured around several lizards "waists" for later recognition.
In addition to lizard movements observed at the SW Blind, it was
sometimes possible in sparsely vegetated dune sand areas to map a single
lizard's activity range for a single day by his spoor. Since rain and
wind quickly obliterate the tracks, this method is useful only if
mapping is done late in the afternoon (following lizard submergence)
on days lacking high winds or precipitation. A 15 m length of kite
string was several times secured around a captured lizard's abdomen
just anterior to his hind limbs. On release he would drag the string
through the bush and a small portion of his movements could later be
plotted. The success of this method varied, since the string occasion-
ally became entangled in brush and was broken. A three m length was
similarly employed with success to locate captured lizard's home burrows
at the SW Blind. Following lizard submergence, known burrows were
checked for string trailing into them.
Lizards were also tracked using radio telemetry. This technique
was used effectively in lizard displacement and homing experiments as
well as for recording daily movements. All telemetry equipment was
purchased from Mini-Mitter Co., Inc., Indianapolis, Indiana. Transmitters
used were Model L Mini-mitters operating at just over 27 MhZ. A 3-channel
Lafayette HA 420 walkie-talkie fitted with a Mini-Mitter Beat Frequency
Oscillator served as a receiver. A Mini-Mitter Model AF Directional
Antenna w'as employed early in the study; however, technical problems
precluded its use for most of the study. Triangulation by signal
strength received from a whip antenna along right angle transects
proved much more effective, especially in areas of dense vegetation.
The signal with this equipment varied from 100 m in sparsely vegetated
dune areas to just over 50 m in thickly vegetated, rocky areas with
greater relief. The range was entirely sufficient for successful use
with this species.
Transmitters were secured to the previously described cat collar
identification belts with electricians tape and strapped to the lizard's
waist. A short length of twine was tied to the transmitter package atop
the belt. Transmitter retrieval could then be accomplished by simply
pulling the package from the lizard (leaving the belt in place for
identification) as he lay in his burrow at night or early morning.
In this way, a second disturbance of the monitored lizards was eliminated.
The short trailing string did not seem to bother the lizards or hamper
their movements in any way.
Measures of lizard activity were made in basically three ways:
1) direct observation; 2) clocking mechanisms at burrow entrances, and
3) flush transects. To record lizard emergence and submergence times,
an alarm clock was remodeled as a single-event recorder in a technique
similar to that of Arltan (1936). A hole was drilled through the outer
case of each of several inexpensive spring wind clocks directly above
the timing wheel. The gravity operated trigger mechanism consisted of
a small square of cardboard over which an elongate paper clip loop was
suspended. A fine nylon thread attached to the cardboard square was
stretched across the entrance of a lizard burrow. Disturbance of the
thread removed the cardboard allowing the wire loop to drop through
the timing wheel, stopping the clock.
A flush transect method similar to the 'King strip census' (see
Giles, 1971, for discussion) was also utilized for activity patterns as
well as density estimates. The technique consisted of traversing a
permanent census trail at different times throughout the day during each
visit to the study island (Figure 2). Distance along the transect route
where each lizard was flushed and distance from trail to lizard (before
being flushed) were recorded. A total of 103 transects was run during
the course of the study. Data from the flush transects, coupled with
direct data at the SW Blind study site provided the major basis for
analysis of activity patterns and estimates of lizard density. In order
to test and insure the accuracy of this transect method, the transect
route passed through the SW Blind site, from which the most reliable den-
sity data was presumably available. The success of flush transects on
Pine Cay was the result of the character of the island itself as well
as that of the lizard. Cyclura on the study island are very wary and
typically run when a human approaches. Although many lizards were
seen during transects, the location of most was possible only after they
bolted off over noisy leaf litter. This technique could therefore be
successful only in areas where the vegetation was dense enough to prevent
the lizard from seeing the observer until he (the observer) was within
audible range of the noise of his flushing. Such was the case along the
transect route I utilized.
Lizard burrow structure was investigated by digging up several
burrows. Inactive burrows were usually chosen for excavation in order
to minimize interference with lizard populations. However, several
active adult burrows were unearthed in order to obtain the inhabitant
in an attempt to correlate burrow structure with sex and size. Lizards
so obtained were either displaced for homing experiments, brought to the
laboratory for behavioral observation, preserved for dissection, or
released into nearby inactive burrows after marking. The structure,
including length, direction, depth and substrate for all parts of each
excavated burrow was recorded. Eggs unearthed in nature as well as
captive clutches were removed to the laboratory in Gainesville for
incubation in moist sand under constant conditions. In an attempt to
study turnover rates in burrow utilization, all burrows in a randomly
chosen circular area (radius 26 m, area 0.22 ha) along the limestone
ridges of the leeward side of southwestern Pine Cay were surveyed on
six consecutive visits beginning in December, 1973. This site was
designated the Ridge Area (Figure 2).
Before analysis of diet in this herbivorous lizard was possible,
it was necessary to collect and identify the majority of the plant species
on Pine Cay. Bob Anderson, graduate student in Botany, University of
Florida, accompanied and assisted me in making preliminary plant
collections and identifications, and vegetation transects, on my first
visit to Pine Cay, in September, 1973. Thereafter, plant samples were
recollected seasonally by myself and Walter Auffenberg, so that as many
vegetative forms (especially reproductive parts) as possible of each
species were available to us for identification. Botanist Don Correll
of Fairchild Tropical Gardens, Miami, Florida, made further plant
collections and identifications during a visit to the Caicos in August,
1974. From these various collections, a reference catalog of seeds,
fruits and leaves was built up for use in identification of lizard
To supplement stomach content analysis from preserved specimens,
311 scats were collected from the SW Blind and Ridge areas for use in
food studies. Walter Auffenberg collected an additional 308 samples
from other areas on Pine Cay. Old scats (those bleached and/or broken
apart) were not collected on Pine Cay in order to eliminate the possi-
bility of sampling a time period other than the current one. However,
all manure encountered on other of the Caicos islands were collected
for analysis. In all, over 670 scats were available for study. Seeds,
elements of arthropod exoskeletons, and most leaves pass through the
lizard's digestive system with almost no alteration in shape, making
food item identification possible from the manure. Records were also
made of species and parts of plants directly observed to be eaten by
Cyclura. Indirect information on diet came from iguana tooth marks
on partially cropped plant parts.
Environmental temperatures were taken with an Atkins Model 3FO
recorder with remote sensing probes. Precipitation was collected in a
standard 5" capacity rain gauge, but was monitored only when an
investigator (David Auth or myself) was on Pine Cay. Climatic infor-
mation collected continuously from 1900 to 1968 at Grand Turk Auxiliary
Airfield, Grand Turk (21026'N, 718'W; elevation 4 meters) and provided
by the National Climatic Center, Asheville, North Carolina, has been
used to supplement data recorded on Pine Cay.
An enclosure approximately 3 x 8 m with walls 1.25 m high, was
constructed in my yard in Gainesville, Florida for captive behavioral
observations. An observation blind was erected adjacent to the pen.
Seven burrows of buried hollow concrete blocks were available in the
pen. The substrate was well vegetated with lawn grasses and small weeds.
Eight clumps of Opuntia sp. were added, and the pen was seeded semi-
monthly with lentil bean and assorted flower seeds. An ad libitum food
supply of lettuce, bananas, commercial cat and dog food, sardines, tomatoes,
and Opuntia blossoms (when available) was provided. A full ten liter water
pan was maintained. Observations were made during mornings and were
terminated during the usual midday reduction in lizard activity. Captive
lizards were brought into the laboratory during the winter. They were kept
in pens similar to those outdoors except in their smaller size and use of
250 Watt infrared lights as heat sources.
Several terms used herein require clarification. The "Study Islands"
include Fort George, Pine, Water, and Little Water Cays; the "Study
Island" refers to only Pine Cay. The principal study sites on Pine
Cay are designated as the SW Blind, and Ridge Area (Figure 2). Lizard
body length or snout-vent length is abbreviated as SV; tail length as TL.
All allusions to lizards, rock iguanas, or iguanas, refer to Cyclura
carinata unless otherwise qualified by scientific name. All material
collected incidental to this study including other vertebrate and
invertebrate specimens, has been placed in the collections of the
Florida State Museum, University of Florida. Statistical methods are
those of Steele and Torrie (1960) unless stated otherwise. Means are
ordinarily given with + one standard deviation. Definitions of
behavioral terms appear in text.
Cyclura carinata Harlan was described in 1825 from "Turk's Island".
The location of Harlan's holotype is unknown. The species can be
diagnosed as follows: Nasals broadly in contact with the rostral.
Frontal, fronto-parietal; and occipital regions covered by small
irregular, strongly keeled scales. Circumorbital scales smaller than
other supracephalic scales. Two to 4 enlarged, vertically arranged
canthal scalses on each side. Dorsal crest broadly interrupted on the
shoulders and rump; nuchal section of the crest with 16 to 20 spines
(up to 2 cm) which are longer than the 45 to 75 spines on the back.
Dorsal crest spines exceed 5 mm in height only in males. Femoral pores
average 21 on each leg (range 18-25). Verticil rows on the tail strongly
keeled, with four vertical scale rows between fifth and sixth rows.
Mid-dorsal tail scales enlarged and sharply keeled. Males less than
80 cm total length, females less than 65 cm.
Two subspecies are recognized. The nominate subspecies can be
distinguished by having the nasals separated by a large wedge-shaped
scale; spinose, hexagonal scales on the outer tibial region, 3 or 4
of which equal the diameter of the tympanum; and fewere than 60 spines
in the dorsal crest. Cyclura c. bartschi described from Booby Cay,
east of Mayaguana Island, Bahamas Island (Cochran, 1931) is character-
ized by nasal scales usually in contact; 4 or more tibial scales equal
to tympanic diameter, and usually more than 60 spines in the dorsal
The following morphological descriptions are based only on specimens
and data obtained in Caicos Islands.
Dentition. The teeth of Cyclura carinata are heterodont. They are
unicuspid at the symphysis and on the premaxillary; the remainder of
the anterior teeth are strongly tricuspid. Median and posterior teeth
are strongly compressed laterally and flared anterlo-posteriorly, with
a multicuspid cutting edge; the large median cusp is flanked by two
or three smaller anterior and one or two posterior cusps. The tooth
formula follows: 0-10 pterygoid; 0 palantine; 4-10 premaxillary; 17-23
maxillary; and 19-28 dentary. The dentition of C. carinata is similar
to the other iguanines (all herbivorous) in both tooth counts (Table 1)
and the presence of increasing cuspidation from anterior to posterior
along the dentary and maxillary (see Avery and Tanner, 1971, for
comparisons). The nearly contiguous, laterally compressed, multi-
cusped, pleurodont teeth forming a continuous shearing surface are well-
adapted for leaf chopping and shearing (Hotton, 1955; Ray, 1965;
Montanucci, 1968). Tooth counts (actually tooth positions) in Cyclura
carinata increase ontogenetically (Figure 5) like those in Ctenosaura,
Iguana and Anolis (Ray, 1965; Montanucci, 1968). Both maxillary and
dentary counts were positively correlated with skull length (r = 0.905
and r = 0.843, respectively).
Skeleton. The basisphenoid is wider than long and slightly to moderately
constricted behind the pterygoid processes; length/width ratio: 0.65-0.72.
Five or more foramina are present on the superficial surface of the rostral
part of the premaxilla. The parietal foramen is small and pierced in the
frontal or the fronto-parietal suture. The parietal is not produced
I t m(-
"o > J0 c
C C a n 0- Nr
I Io to
LL U -
posteriorly. Angular and splenial elements are present. The splenial
is long and narrow. The dorso-ventral borders of Meckel's groove are
fused anterior to the splenial. The antero-ventral alveolar foramen is
partially enclosed by the border of the splenial; the coronoid possesses
a large process overlapping the dentary (Figure 6). 24 presacral
vertebrate are usually present. The postero-ventral angular process is
pointed and elongate, and directed antero-medially. Each presacral
vertebra is provided with zygosphenes and zygantra in addition to
zygopophyses. At least some autotomic caudal vertebrae have two pairs
of transverse processes, the fracture plane passing between them.
Scapular and secondary coracoid fenestrae are well-developed; clavicles
are simple or with posterior hook-like processes; the interclavicle is
T-shaped or arrow shaped; the sternal fontanelle is small or absent;
and four sternal and two xiphisternal ribs are present (Figure 6).
The hyoid apparatus of Cyclura carinata was studied by Avery
and Tanner (1971) and my observations (Figure 6) do not differ from
Hemipenis. The hemipenial structure in the genus Cyclura has not
previously been illustrated. Since the only description available for
the hemipenis in this genus is a brief one by Cope (1896), the organ
in Cyclura carinata is here described. The hemipenis is single-lobed,
subcylindrical to clavate. The sulcus spernaticus is a broad open
groove along the posterior surface of the hemipenis. Calyces cover
the distal half of the hemipenis, whereas the proximal half is covered
with irregular creases. Spines or spinose structures are lacking.
Based on hemipenial structure, Cyclura is most similar to the iguanine
genera Iguana and Amblyrhynchus (Avery and Tanner, 1971).
Figure 6. Skeletal elements of Cyclura carinata: A) Sternum
(ventral view); UF 14354, Middleton Cay. B) Hyoid apparatus
(ventral view); UF 14354. C) Mandible (medial view; tooth
crown structure not illustrated); UF 32678, Middleton Cay.
Lepidosis. Lizards in the genus Cyclura are the only iguanids possess-
ing corneous subdigital combs on digits 2 and 3 (#4 = longest) of the
hind limbs. Subdigital lamellae are multicarinate, not flattened, and
lack a mat of hair-like processes.
Numerous other epidermal scale characteristics of Cyclura carinata
were presented above; additional descriptive data are not here presented
in anticipation of publication of a systematic analysis of the genus
by Schwartz and Carey (in press). As previously stated, the spines
of the mid-dorsal crest exhibit considerable sexual dimorphism in
size. In females, neck spines never exceed 5 mm in height; maximum
recorded back spine height for females was 3 mm. Dorsal spines on the
neck in males begin development (elongation) coincident with the onset
of sexual maturity. Thereafter each time the skin is shed the spine
increases in length and an annular impression is formed around its base.
After 2 sometimess) sloughs of the skin, the back spines begin develop-
ment in exactly the same pattern as those of the neck. Back spines
are therefore shorter and possess two fewer annuli than neck spines on
most individuals. The relationship of the spines to sexual recognition
and their annuli to growth and age are discussed later (see GROWTH).
In old lizards the back spines approach or even exceed the neck spines
in height, and the annuli become inconspicuous as the spines become
smooth due to wear. The middorsal spines on the anterior portion of
the tail in male iguanas also elongate with age albeit very slowly.
A maximum height of 18 mm is attained in extremely old individuals.
Color. Rock iguanas exhibit great individual and population color
variability. Patterns tend to be darker (gray, brown, or green,
on population) immediately following emergence and preceding sub-
mergence each day. The midday coloration is typically light gray to
bleached green (again related to population). This changing albedo
is presumably an adaptation to increase absorptive radiation in sub-
optimal ambient temperature regimes (Cowles and Bogert, 1944). The
change in color is most rapid during the morning basking period and
has been timed at less than one hour. The relationships of changing
albedo and skin reflectivity to thermoregulation are under study
(Auth, in preparation). Seasonal color changes, related to hormonal
levels, have been observed in numerous iguanid lizards (Carpenter, 1967;
Ferguson, 1976) however, no such changes were evident in the studied
Inter-population color differences on the Caicos bank are very
pronounced. Lizards from the eastern populations (Middleton and Long
Cays) are very drab gray, dorsally and ventrally, with only slight
changes in darkness of the gray. Lizards from Fort George and Pine
Cays vary from a brownish to greenish-gray following emergence, to a
nearly white-gray at midday. The venter is gray to blue-gray. A very
abrupt pigmentation change occurs between Pine Cay and Water Cay to
the south. Lizards from both Water Cay and Little Water Cay are the
most brilliantly colored of Caicos populations. Basically a dark gray-
green on emergence, during basking these lizards become a pale yellow-
green. The typically lighter venter in these lizards is blue-gray
flushed with yellow. The yellow coloration extends over the tail and
on the ventral surfaces of the limbs and digits. Pine and Water Cays
have apparently been separated throughout historic time by a deep water
channel, allowing the maintenance of very different color morphs.
However, the channel was blocked with sand and Pine and Water Cays were
linked by a sandy isthmus as a result of tropical hurricane Donna in
1960. Iguanas have not yet breached the open sand barrier of the
isthmus, but will undoubtedly do so as primary succession provides
the necessary shade and food plants. The effects of genetic inter-
change on the integrity of the two color morphs will be interesting.
Cyclura carinata also exhibits a variable pattern of narrow
vertical stripes on the dorsal lateral body surface. The stripes are
most obvious near the anterior mid-dorsal line and become increasingly
less apparent laterally and posteriorly. The stripes typically number
9 or 10, and are areas where the pigmentation is simply lighter than
adjacent regions. The pattern is most obvious in hatchlings and
juveniles, and typically very obscure in adults.
Anatomy. Morphology of the thyroid gland in Cyclura carinata is
discussed by Lynn et al. (1966). Digestive tract anatomy is discussed
under FOODS AND FEEDING.
The Turks and Caicos Islands experience a typical tropical maritime
climate. East-southeasterly tradewinds predominate, averaging 22.5 km
per hour throughout the year, and annual air temperature- fluctuationsare
not great (Figure 7). Recorded annual air temperature extremes between
1960 and 1968 on Grand Turk (110 km to the east of the study island) were
15.6C and 36.OC. However, microclimatic temperatures show much greater
fluctuations both daily and seasonally. Temperatures on light coral
sand substrates ranged from 12.5C to 60.OC on the study island; those
on darker soils were considerably higher. Substrate temperature
differentials between exposed and shaded areas were greatest during mid-
day in July through August with a maximum recorded difference of 25C
(31-56C). Twenty degree divergences are typical of summer days without
midday precipitation. The maximum recorded winter (November-December)
shaded-exposed substrate differential was 18.5C.
To provide a general picture of the diurnal thermal environment,
microclimatic temperature data from optimum days (unaffected by overcast
or rainy conditions) on the study site were averaged by half hour for two
periods, representing the longest and shortest days of the year (Figure
8). Data from six days in late June, 1974, and six in December, 1973,
were used. These differences in summer and winter temperatures are in
part due to an annual fluctuation in day length (photoperiod) of nearly
Figure 7. Average monthly temperature regime in Turks and Caicos
Islands. Substrate temperatures were recorded at SW Blind, Pine
Cay during study period; remaining temperatures collected between
1900 and 1968 at Grand Turk Airfield, Grand Turk.
< MEAN DAII.Y MAX i
MEAN DAILY MI
J F A M J J A S N D
II__ _ _
Figure 8. Average diurnal microclimatic temperatures for June and
December at SW Blind study site. Data averaged by half hour for
six storm-free days in June, and four in December. Arrows indicate
sunrise and sunset.
- 30 SHADED SUBSTRATE
I I I I I I I I I I I I I I 1
6 9 12 15 18
I SN SUN
2 25 CM BELOW GROUND
5 6 7 8 9 10 11 12 13 14 15 16 17 18 EST
TIME of DAY
three hours (Figure 9). A more detailed analysis of the thermal environ-
ment and its relation to the biology of Cyclura carinata on Water Cay,
Caicos Islands, is found in Auth (in preparation).
Since it was possible for me to monitor rainfall only while on the
study island, data collected between 1900 and 1968 at the Grand Turk
Auxiliary Airfield (Latitude 21026'N, longitude 71008'W; elevation four
meters) are presented here, unless otherwise indicated. A well-defined
dry season exists from January through May with monthly rainfall averag-
ing only 13.5 mm (Figure 10). The wet season usually begins and ends
rather abruptly in June and January, respectively, and rainfall averages
62.8 mm monthly. In addition to amount of rainfall, the percentage of
days per month experiencing some rainfall also increases substantially
during the wet season (Table 2). Average annual rainfall on Grand Turk
is 597 mm. However, based on 458 mm recorded on Pine Cay in 194 days
over three years, the annual estimate is approximately 862 mm, somewhat
higher than on Grand Turk. Comparison of the vegetation, soil moisture,
and general weather conditions across the bank on several occasions, in
addition to information supplied by locals indicate that rainfall is
greater on the more western islands of the bank, but probably does not
exceed 760 mm annually.
Changes in rainfall, temperature, and photoperiod are the major
aspects of climatic seasonality. The effects of each of these are
reflected in lizard activity. Days are often cool enough to curtail
lizard activity in the winter, but almost never so in summer. The
climate in the western cays of the Caicos Islands is one of a warm dry
spring (March to May), followed by a hot, wet summer (June to October)
o0 If o 0
which grades into an even wetter, but still warm autumn (September to
December), and is followed by cool, relatively dry winter (January to
March). Abrupt changes in temperature and precipitation characterize
the transition between the wet and dry seasons.
Reflecting their reef origin, the Turks and Caicos Islands are
characterized by honeycomb limestone formations usually covered with
coral sand. The rocks are typically exposed along the windward shores,
with broad sandy beaches along the leeward. Inland accumulations of
humus provide a rich nutrient source and support more luxuriant
vegetation (see Vegetation).
Iguanas frequent rocky and sandy areas as well as the more elevated
humus areas, but generally avoid the lower areas with thicker soils due
to the proximity of the water table to the surface and their inability
to dig through the root mass.
The spatial relationships of the principal vegetation types on
Pine Cay are illustrated in Figure 11 and listed in Tables 3 and 4. The
vegetation on Pine Cay corresponds to the "Evergreen Bushland" of
Beard (1949), and is physiognomically very similar to that found on
Anegada in the British Virgin Islands by D'Arcy (1975). The same genera
(if not species) are represented in similar habitats on both islands.
Although D'Arcy (1975) divided the flora of Anegada into only four
primarily edaphic communities (shorelines, sandy plains, limestone
\tt ,;. '*JI
/*.-* .y ..-
/ /:::*."" "" '
//::::. r-.,', l ;:U o u ,
.,.:-;..>,.i y 3 U 0
.. / z ui a
0 o o
0 CU 0- 0.
4 -> 4-
U >- L.
- c -
uo C C
. .- 4 4
M u T ro
CU *- -
L) (j cL
- 1- 1 -.-
E *- O
a.U 0 0I
1- 0 -3 a
u C "
C 0 L- CU
44 U_ CUJ- 4-f
CU (U o -i
+4--I- CU Q- C
. 30 0 -
0 -- C0 U
4- i- C- 44
- E 1.
2 c Mn -
( 0 0 (an
M 1 0- E0 C E
C 4I- .- .o lU .- _
>- CU 0 o 0 Co >-
U 0 m m- C
> -C F- -C u
CU CU a 4- CU CU C: 4- c
CU CU CU 0 0- L. -^ CUJ
O-C U >-
J. 4- fC U +4 TO CU +0 -
f-1) 0i n/ V/C CD 2 Cj V) rN
S4I -- u E } a)
Y C)- 0 -
(D *- a)E' o > : >
C- *- ECO 3 C c
-.. E O 0 C 0
m CU a) :3 2. 0-
C C E I` E -
*-> u 2 !0 -C or-
C UI- )0 j C 3 U 0- --
SC *- I n- 0 a n a a
0 :O .
u -- C-- Ln f -O L E
L. C0 U- C 0 -
O OC CO ) -- > -
E I L Z OL c a N I JU
CC a 0 L C-
C ,- CL 0
7- C) < U
-J IC C
Cl 0 O Ci a ca
C C..- C, C *
JC >- 0 Un
ri U- -l U,
CM C -- C1
>1 C:CCC: F
131 ro W C)
r3 o 03 CL
o- O- O
L Q >. E 0
0 E E C
u: E: 03 '
- 0 0 0
C 0 (U
L. L- >
>- 3 0 0
< 1- 0 0
oo o on
L 4-. .-- L
< < CL C
C C -
0 *- C :3
-o 0 om
U E L- )
=3 =3 >
oc M O m
i 0 in
a i *-
L L >
>- -o -.
vin X 0 0
0 -, U 0
- 0 0
L L" ..C
-- in n
o i m
o U L-
0 (] r- o-
0 ~o /i f
c L 0 c
0 0 L-
(;u u L
x x x x x
viCL X X X X X X X X X
~ 1. L X X X
n 0 C f 0 <
CO c u
Su0 0 M D m /-
r- o .- x x
-au I '- 0
O _g _- 0 -n
-V U V1 I 3 (U (
H < l r3 m X X X X 1<
mo4 m L w w
3 V- O V O .-- .- V -
Co "/I VI (V *- (V 0 V :L
o U VI Vi V- E k
'V L- O V ( V C (V *- V- O
O IV Iv c Va (V I- Vi 43 U
0L 0I 43 T C > L L V (V
aV .0 0 i 4 C 43 V - VI
> 0- >- C > 0 0VI 4- ov
(Vc - 0 3 U 4 4 L Vi > 0
H C) C C)] U U LU LU LL L3 (i
S -ro o
1 0 E0
c o o (
S '- U (U
C( C] Cr CJ
- 0 (U
E a t >
o o c =
c 4- ) -
CL L >
C (U (U
a a 0 c
E U 0 )E
C C (
I) (U i
E U c Q
- U 14
C a >- -
^ u c -
m- n naj
2 4 i0 I
S O O
-C -T C N
U 0 *-
0- 0 0 0
u c >t
C' t" m Q
-_ -C l
C C- Ca 0
3- 0 3 0
S CO C
0 .C *-
0. 0. 0. C.
- Qa 0
E O E
a a- C ro
0 0 0 C/3 (/3 C
0 E 0
o a O
E o -- rm
3- 3- E
0 a 0
QO 3- 3-
3n 3,3 3,3 03
S 0 0 --
2 0 0N
X L C N
plains, and vegetation near man), it seem justified on Pine Cay to
distinguish additional plant communities within at least two of his
provinces. A more thorough discussion of the components of the vegetation
types on Pine Cay and their relationships appearsin Auffenberg (in
Cyclura carinata inhabits all of Pine Cay's plant commu:' t;s except
the Marsh and portions of the Mixed Woodland vegetation types. These
are areas where the proximity of the water table to the ground surface
precludes the possibility of burrow construction and vegetation is too
dense for movement. Cyclura reaches its maximum adult abundance on
Pine Cay in areas of Rocky Coppice. Young lizards are more commonly
observed in areas of Open and Dense Scrub.
Unless natural shelters in the form of abandoned land crab
(Cardisoma guanhumi) burrows, or honeycomb limestone or coral formations
are available, Cyclura carinata digs its own burrows. Since natural
retreats are not abundant on the study islands, most lizards dig their
burrows. Large mounds of vegetation, dirt, and rocks piled up by
bulldozers while clearing roads are often used for retreats by lizards
on Pine Cay. These sites also offer excellent elevated basking sites.
Adult male iguanas exhibit great burrow tenacity. This is reflected
in the greater average length, depth, and complexity of burrows dug by
males versus those dug by females (Table 5). Sexual differences in
burrow depth are also due, in part, to the need for shallow incubating
chambers for successful nesting in females; burrow length differences
+1 r + -z
-- -uO *
Q o 0o
n c (U-. OU c a) )
.0 u oum0
cU - m
0 0 II 0 + -r
c . nO
CU Z in Z *UCn-
o L C
LC- C -
Q- ~ v (U U
are also related to sexual dimorphism in size. Though adults of both
sexes commonly use the same burrow, strongly territorial adult males
never share them, unless used as a retreat after having been frightened.
Although territorial adult male lizards near the SW Blind maintained
a single retreat, they were nevertheless aware of all burrows (and
other potential retreats) both within their own territories and well
into those of adjacent males.
Even subadult lizards showed familiarity with burrows far outside
their normal activity ranges. On September 20, 1973, adjacent to the
Ridge Area on Pine Cay, a 178 mm SV lizard was removed from its burrow.
Following marking, it was released at the site of its burrow. It ran
directly and non-stop 16.1 m, where it entered another active burrow.
The second burrow was found to contain the marked lizard, in addition to
its resident male (SV 229 mm). When released after marking, this resident
immediately ran over the ridge and directly to the burrow first excavated.
The lizard first marked sought refuge in dense vegetation when released.
Both lizards were clearly aware of the exact location of the other's
Blocking of a territorial male's primary burrow causes him to shift
his retreat. Removal of the obstacle, after one to three days, is
followed by his return to the primary burrow. The effect of permanent
destruction of the primary burrow was not determined. The short term
blockage seemed to cause no change in activity range location, but a
shift is suspected if the burrow is destroyed.
Juveniles and subadults frequently shift burrows. One subadult
non-territorial male, carrying a transmitter, in the SW Blind area,
utilized three different burrows on three consecutive nights. However,
in some areas the unavailability of auxiliary retreats necessitates the
continual use of a single site. A hatchling took up residence in a small
hollow tree limb in a brush pile on September I, 1974, within only a few
days of emergence. It still occupied the site in April, 1975, when last
observed. Since it was never again seen on the study site, it is
presumed it was killed.
The following patterns in burrow usage were observed in the Ridge
Area. Large burrows active at the beginning of the study remained
active throughout the study period. There was, however, considerable
small burrow turnover, probably due to 1) predation on juveniles; 2)
lack of burrow tenacity in juveniles; and, 3) periodic excavation of
burrows by terrestrial crabs. Since the burrow is central to all
activity in Cyclura carinata, its importance cannot be overemphasized.
Not surprisingly, burrow construction in all species of Cyclura
seems to be related to the substrate conditions. In areas where sandy
dune situations predominate, as on the typically small, low islands of
the Bahaman archipelago, constructed burrows are used. However, on
islands dominated by cavernicolous limestone and coral exposures the
possibility of burrow excavation by iguanas may be precluded. Natural
retreats are then employed (see Retreats under ACTIVITY AND MOVEMENTS).
The presence of at least some friable soil for nest construction
appears to be a very important factor ultimately limiting distribution
and numbers of Cyclura populations. In the species of Cyclura for which
the nest site has been described (Foble, 1923; Wiewandt, in preparation),
a nest burrow was excavated in well-insolated soil areas, sometimes
several km from the female's normal activity range (see Discussion
under REPRODUCTION). In addition to protection, burrows could serve
a thermoregulatory function, however lizards do not use them as heat
sinks, even under extreme conditions.
Adult Cyclura typically maintain their burrows by removing sand,
leaves, and other debris during the middle of the day when general lizard
activity is reduced due to high environmental temperatures. However,
this is not a daily routine; a period of several days generally
separates burrow cleaning activities. In removing debris, the lizard
enters the burrow head first and kicks material backward as it moves
further inside. Exiting the burrow head first, it turns and repeats
the procedure several times.
The complete excavation of a burrow was never observed, so t'- time
required is unknown. However, a young adult male completely excavated a
new burrow near the SW Blind during my absence between July 20 and
September 1, 1974.
Neonate lizards usually utilize natural cavities or spaces cleared
beneath rocks,roots, logs, etc., for retreats. In open sandy areas,
they will excavate their own burrows. These burrows are usually shallow
(< 15 cm deep) and short (< 50 cm), with diameters between 2.5 and 3 cm.
Some have short side branches but these may be added by crabs.
Since many young lizards frequent more than one burrow and many
small burrows are actively maintained by crabs, small burrows are not
useful in estimating lizard density. Data on locations of large active
burrows can, however, provide fairly accurate estimates of adult lizard
density (see DENSITY AND DEMOGRAPHY).
Burrow dimensions and complexity are quite variable (Figure 12),
dependent in part on the surface and subsurface soil texture. The
longest burrow excavated was in dune sand on Pine Cay, and measured
6.4 m. It was occupied by both a female and a male when opened. One
of the biggest burrows encountered was an active one which I could not
completely excavate. The burrow was in an area of deep dune sand nearly
devoid of vegetation. The 16 cm diameter of the burrow suggested that
it had been dug by a male. After 5.0 m of horizontal excavation, the
burrow was 1.5 m below the sand surface. It continued downward with the
same slope at least one m farther. Digging was suspended due to constant
collapsing of the excavation pit.
Burrow orientation on Pine and Water Cay (Figure 13), appears to
be related more to micro-topographic features than to intentional
directional construction. More burrows generally open to the northwest
and southeast on these two islands, because these retreats lie perpen-
dicular to the generally northeast-to-southwest lying ridges (Figures
2 and 12). Cyclura prefers digging under the limestone exposures in
the sides of these ridges to digging in open flat sandy areas. In the
latter areas, burrows are typically excavated into the sides of small
hummocks formed by the accumulation of blown sand around the bases of
clumps of vegetation. An obvious advantage in avoiding flat areas of
open sand for burrow construction is the reduction of the possibility
of entombment by burrow collapse.
As in the burrows of the Florida gopher, Gopherus polyphemus
(Young and GoFF, 1939), there are in nearly every Cyclura burrow,
numerous arthropod inquilines, many apparently obligatory. The most
Figure 12. Variation in burrow morphology and orientation in adult
Cyclura carinata. Soil depth (in cm) to top of burrow is indicated;
changes in depth are uniform between measurements. Letters adjacent
to burrow entrance (solid circles) indicate sex of inhabitant.
Asterisks mark burrows used for nesting in previous yeers.
13 15 15
Figure 13. Orientation of burrow entrances in Cyclura carinata,
recorded as compass direction into burrow.
common and seemingly most obligatory were camel crickets (Ceuthophilus?).
Also found were solpugids (genus Ammotrechella), scorpionids, amblypygids
(Tarantula marginepennis), nymphal cicadas (Ollanta caicosensis), and
numerous unidentified spiders.
Male Sexual Cycle
The male testicular cycle closely parallels the female reproductive
cycle. Since testes weight has been shown to be an accurate indicator
of testicular activity in lizards (Hahn, 1964; Marshall and Hook, 1960;
Ballinger, 1973), no histological analysis was made. Testes reach
maximum sizes in April and May, declining rapidly in June and July to
complete regression in early fall (Table 6). Testicular size maxima are
coincident with breeding in Cyclura carinata. A similar cycle (early
spring maximums and late summer minimums) is typical of at least the
Sexual maturity in male C. carinata was determined by testicular
enlargement between January and June, or actual observation of attempted
or successful matings in May. The smallest male with enlarged testes
measured 216 mm SV (weight approximately 440 grams). The largest males
without testicular enlargement were 222 mm SV (435 gm live weight) and
210 mm SV (365 gm). A 219 mm SV captive male which attempted copula-
tions on numerous occasions during May, 1975, was the smallest male
observed to exhibit courtship behavior. These data indicate sexual
maturity in males is reached at approximately 220 mm SV and a body
weight of 375 to 475 gm. Growth data indicate this requires about seven
years (see GROWTH). Adult males averaged 276.3 mm SV (191-360) and
935.1 gm body weight (256-1864).
Table 6. Seasonal variation in proportional testes weight
(testes weight in gm x 10-5/sV in mm) in male Cyclura
carinata over 216 mm SV.
Date N 7 S.D. Range
March 6 1 867 ---
May 10 1 1358 ---
June 30 7 178 141.13 48 454
August 15 2 117 66.47 70 164
September 15 1 47 ---
December 15 2 117 36.77 151 203
December 30 4 165.5 115.37 62 305
Female Sexual Cycle
Ovaries are difficult to distinguish from testes in lizards less
than one year of age, however the obvious presence of oviducts in
females allows accurate sexing of even hatchlings. Immature ovaries
contain numerous follicles up to 4 mm in diameter. Maximum follicular
diameter in mature females between oviposition and spring recrudescence
is also approximately 4 mm (Figure 14). Yolking of follicles begins
between January 1 and March 1 and continues until ovulation in early
May. Maximum follicular diameters reach 30 mm at that time.
Seasonal combined ovarian weights reflect the same trends as
follicular diameters (Figure 15), with ovarian weight maxima (63.7 gm
in a 248 mm SV lizard with five enlarged follicles) in mature females
occurring in early May and post-reproductive minima (generally less than
0.35 gm) from July through December. The maximum combined ovarian
weight recorded in an immature (184 mm SV) female was 0.275 gm in March.
Immature weights were normally considerably less than this. Perhaps
her ovaries were developing in preparation for breeding the following
year when she would surely be of adult size.
Oviducal eggs were not present before at least the last of April
and were noted only as late as June 9. Yolked follicles and oviducal
eggs were never found in the same female and no yolked follicles (except
an occasional atretic one) were found in post-ovipository females.
Fresh corpora hemoragica averaged 6 mm in diameter, but regression was
rapid and corpora lutea were usually unidentifiable by early September
Of only two females with oviducal eggs, one showed evidence of
transuterine migration of ova. The right reproductive tract of that
J F M A M J J A O N D
Figure 14. Seasonal variation in maximum follicle diameter (mm) in
ovaries of female Cyclura carinata. Small dots indicate individual
females; large dots, two or more lizards. M and N mark mating and
nesting seasons, respectively.
Figure 15. Seasonal changes in proportional ovarian weight (grams
x 10-4/SV in mm) in mature female Cyclura carinata. Each point
represents one individual lizard.
2 00 00
J F M A M J J A S 0 N D
( ) II(
(0 IO J CJ CW 4
female included two corpora lutea in the ovary and three eggs in the
oviduct; the left tract bore four corpora lutea and three oviducal
eggs. Cuellar (1970) has previously discussed this phenomenon in
Minimum sizes for mature females (as indicated by the presence of
yolked ovarian follicles, oviducal eggs or corpora lutea) were 184 mm SV
(250 gm; with enlarged yolked follicles), 190 mm SV (280 gm; corpora
lutea present), 194 mm SV (live weight unknown; with corpora lutea),
and 197 mm SV (205 gm; with corpora lutea). Maturity of young adult
and subadult females collected between September and January could not
be accurately determined due to the regressed state of the ovaries
even in obviously adult females. The largest immature females collected
between January and August were 210 mm SV (approximately 320 gm), 191
mm SV (240 gm) and 184 mm SV (weight unknown). Maturity in females at
185-200 mm SV (200-300 gm) at an age of about six years (see GROWTH)
is indicated. Nearly all females presumably oviposit their first
clutch of eggs between age six and seven. Adult females averaged 225.4
mm SV (190-292) and 475.6 gm (205-1135) body weight.
Courtship and Mating
From late March through April, males became increasingly interested
in females. During April, males were frequently observed following
females on the ground. At this time, the resident territory-holding male
typically approached within one meter of the female, but did not display
or otherwise court the female. The approach usually caused the female
to headbob (submissively; see SOCIAL ORGANIZATION), move a short distance
away, stop, and again bob to the male. The male often reciprocated with
a headbob, and again approached the female closely and the sequence was
repeated (up to a dozen times). The impression was that the male was
attempting to insure his proximity to a female as the mating season
approached. Perhaps this precourtship interest is necessary for the
induction of receptivity as in Drosophila melanogaster (see Parker, 1974,
p. 158) and Anolis carolinensis (Crews, 1975b).
Resident males terminated their female-directed activity to displace
neighboring males from their (the resident's) territory (hence away from
the female) at this time of the year, just as they did during the
remainder of the year. However, it was my subjective opinion that the
males were more willing to defend at this time of the year than at any
other. There was no question that the general level of activity was
greatest at this time of the year (see ACTIVITY AND MOVEMENTS). Follow-
ing defense, the resident male returned to the female and resumed his
interest in her. The precourtship approach behavior could be a mechanism
by which a female can determine the extent of the territory and the
effectiveness of its defense by a potential mate.
In some cases, the male was so persistent in his attempts to
closely approach the female, that she ultimately moved out of his
territory and into that of a neighboring male. However, females some-
times also moved between males' territories without this pressure from
the resident males. Perhaps female Cyclura were making active choices
among courting partners at this time, as Trivers (1976) suggests for
The interest of males in females increased through April, pre-
sumably until the female tolerated the close approach of a male and
active courtship could proceed. Unfortunately, neither active court-
ship nor copulation was observed in nature. However, they were fre-
quently observed in captivity and the basic behavioral components are
undoubtedly the same as those in non-captives.
Courtship was observed nearly 100 times (three times to completion)
in the captive enclosure in Gainesville during nine days between April 24
and May 9, 1975 (Table 7). During this period, four mature females
and at least two of four males were confined together in a pen at any
one time. The lizards were captured April 5-9, 1975, on the study
islands. The females and smallest male (#4) were released into the
captive enclosure on April 17, while the other males were maintained
separately in the laboratory. These males were introduced into the pen
at various times between April 24 and May 9 to investigate dominance
relationships (see SOCIAL ORGANIZATION).
Successful mating behavior can be divided into four basic stages:
approaching, displaying, neck holding, and copulating. Figure 17
illustrates the male and female components of courtship behavior.
Approach. In captivity the male and female did not always exchange
headbobs prior to the male's approach to the female. In the field, the
female always bobbed submissively at the appearance of the male; the
male often did not reciprocate (see SOCIAL ORGANIZATION). The court-
ing male typically approached the female laterally and from behind
(their bodies aligned at approximately a 30 to 450 angle). The profile
of the approaching male once within 0.5 m of the female was very low to
the ground. The body was dorsally-ventrally flattened and the chin was
kept less than 5 cm from the substrate. A few times the male turned
his head to inspect the female's hip area, but usually approached until
his head was lateral to the anterior trunk region of the female. At
Table 7. Average number of courtships per day attempted by each
male of each female in Gainesville enclosure during nine days
between April 24 and May 9, 1975.
#1 #3 #4
FEMALES1 (mm) 349 286 219
#1 235 3.0 2.333 4.75
#2 235 1.67 1.222 0
#3 216 1.33 1.222 0.75
#4 185 -- 1.000 0.75
Days male 3 9 4
IAll females were present everyday.
Figure 17. Behavioral components of courtship in Cyclura carinata.
Display -. Bob
------- -(submissive; on O's approach)
(with or Complete rejection
without bob) --- (moves away)
(while approaching) Rejection
(runs away; usual ly with open-
Straddle and neck bi te
I Return bite
Release neck hold -
(with open-mouthed bobs)
Ta il I maneuver
Refusal to arch tail
S(possible only if 0 small)
Opposition of cloacae
IT IITROM ISSION
TERMIIIIIArlO OF MATING
this point the female often moved away, rejecting the male's advances.
If she did not (or could not, as in some captive situations) the male
began his courtship display and continued to approach the female.
Display or Vibratory Phase. The courtship display of Cyclura
carinata w'as stereotyped and very similar to that of most other iguanid
and agamid lizards that have been studied (Carpenter, 1967; Gonzales,
1974). The male continued his approach toward the female while rapidly
vibrating his head vertically. I have termed this behavior vibratoryy'
display due to the low amplitude of the headbobs. A typical display
action pattern (DAP; see Carpenter and Grubitz, 1961) of this head move-
ment appears in Figure 18. Four videotaped displays, each including four
oscillations of the head, averaged 0.36 seconds (0.32-0.40). Additional
amplitudes required proportionately longer during several other displays.
Males sometimes performed displays with open mouths.
This vibratory courtship display is synonomous with the "jiggling"
display described by Evans (1938) and Ruibal (1967); the "courtship
nod", by Carpenter (1962b); and the "shudder sequence", by Ferguson (1970)
for other lizards. In each case, the display consisted of a sequence of
shallow, rapid headbobs. Male Ctenosaura pectinata (Evans, 1951) and
Amblyrhynchus cristatus (Carptenter, 1966) also exhibit a similar court-
ship display. Sauromalus obesus (Berry, 1974) is the only iguanine
lizard studied to date that lacks this display.
Females rejected male courtship advances by a stereotyped rejection
display or by simply running away. Either behavior was accompanied by
vigorous headbobbing, typically with an open mouth. In the rejection
display (Figure 19), the female compressed her body laterally, postured
high on all four legs with her back strongly arched and her tail on the
.- 8 c
U 01 -
C 1c n3
L0 C -
L CU .
ground, and headbobbed rapidly (usually with open mouth). In the field,
as in many of the captive courtships, this display is probably sufficient
to thwart the males' courting interests. However, captive males often
continued the courtship sequence despite this display. If the female
maintained her position during the male's vibratory display and approach,
the male terminated the display, raised up and gripped the female's neck
in his teeth.
Many iguanid lizard species have a similar rejection posture
(Fitch, 1956; Clarke, 1963; Carpenter, 1967). Noble and Bradley (1933)
first described and illustrated this behavior but were unable to deter-
mine its significance. Within the subfamily Iguaninae,Sauromalus obesus
(Berry, 1974) also uses this rejection posture, while Amblyrhynchus
cristatus apparently does not (Carpenter, 1967); other members have
not been sufficiently studied.
Only one homosexual courtship (with the smallest male) was
observed and it was terminated by the courting male following his
Neck holding. Following the vibratory phase of courtship, the
male moved to straddle the female and secure a mouth grip on the dorsal
region of the female's neck. The female could still terminate the
courtship at this point by turning and biting at the male's head and/or
struggling free. Females sometimes attempted to walk away, dragging the
male for short distances.
Once the female's neck was securely held, the male attempted to
maneuver his tail under that of the female to align the cloacae. The
receptive female then arched her tail upward to facilitate the male's
tail maneuvering. She could apparently still reject the male's advances,
by refusing to arch her tail, only if the courting male was smaller than
herself; large males could force their tails under the female's tail
even if she did not arch it. This was observed once in captivity during
the forced copulation (rape) of one of the females (discussed later) by
a large male. The refu-al of a female to arch her tail to allow
cloacal apposition was observed six times when a small male (219 mm SV)
had obtained a neck grip on a larger female. Once the cloacae were
finally opposed, the neck grip was maintained until hemipenial with-
drawal following copulation. The mating posture was exactly like that
pictured for Amblyrhynchus by Schmidt (1935).
Copulation. Once the cloacae were opposed, intromission followed
immediately and lasted from 40 to 76 seconds (Mean = 58.7; N = 3). No
thrusting by the male was observed. Wiggling of the distal portion of
the tail by the female immediately preceded withdrawal of the penis by
the male. Uncoupling was rapid; the male released his neck grip and
backed away from the femle. His pelvis remained elevated from the
ground until his hemipenis was fully retracted. Typically both lizards
then moved very little and rested quietly. Their activity during the
remainder of the day was generally depressed.
In all iguanine lizards except Iguana iguana, for which data are
available (Table 9) copulation (i.e. female receptivity) occurs approx-
imately one month before oviposition. A distinct relationship with the
fer.,ale ovarian cycle is implied. It is suggested, but not known, that
receptivity in Cyclura carinate immediately precedes ovulation as it
does in Anolis carolinensis (Crews, 1973a). and perhaps Crotaphytus
wislizeni (Parker and Pianka, 1976).
It could not be determined if female receptivity is terminated by
copulation as Crews (1973b) has shown to be the case in Anolis
carolinensis, but it is strongly suspected. One captive female was
copulated with twice by the same male in a period of three days, however
she was an unwilling partner in the second mating. The male was able
to "rape" the female due to his much larger size (286 mm SV versus
235 mm) and the confinement of the enclosure. Since no female Cyclura
was observed to accept a male more than once in captivity (mated females
always rejected further courtship attempts), I suspect that females
copulate only once, with only one male, during each breeding season.
Males, however, will copulate more than once during the breeding
season, but apparently only once during any one day. On April 24, male
#3 had attempted to mate once (with another female) before he copulated
successfully with female #1. He did not attempt courtship again that
day. The following day he unsuccessfully courted three times. The
next day he attempted to breed eleven times (with two females) before
he forcibly mated #1 female. He did not try to court again that day,
but did on several following days. On May 9, #3 male copulated success-
fully with the first female (#3) he courted. She was very receptive and
moved to him immediately after his morning emergence. He did not attempt
courtship for the remainder of the day. It is suggested that males in
the field are able to mate with more than one female if they are avail-
In captivity, female #1 accepted male #3 on April 24, and was
raped (by the same male) on April 26; female #3 accepted male #3 on
May 9. The other two females were not observed to accept males, but
both shelled eggs. It is therefore presumed that they also mated. The
earliest courtship attempt was observed on April 24; none were attempted in
two days of observation prior to that. Since the lizards involved were
captured less than three weeks prior to that time, a two to three week
mating season, centered about the first of May, is indicated.
Post-Copulatory Behavior. Captive males frequently dragged their
cloacal regions on the substrate both prior to and following courtship
behavior. Two captive males also exhibited a strange modification of
normal defecatory behavior during the breeding season. The pelvic
region was elevated above the ground in typical defecation posture;
however, the abdominal contractions necessary to egest the fecal pellet
resulted in version of the hemipenes. The organs remained everted until
after the scat was deposited. As they were then withdrawn a small
quantity of white viscous fluid (semen?) dripped from the region of
each retracting hemipenis. 'I could not determine the significance of
Probably the most significant post-copulatory behavior noted in
the captive situation was when the male drove off males that courted
the female with which he had copulated. Unfortunately, this behavior
was clearly observed after only one of the captive matings. Following
the other two copulations, other males did not attempt to court the
Two other mlaes were present in the enclosure at 1237 hours on
April 26, 1975, when male #3 copulated with female f#. Male #2 was
old, senile, and subdominant to #3; he was never observed to court
any females. Male #4 was very young (first or second year post-maturity);
he courted frequently (Table 7) but was never accepted by a female.
Approximately eight minutes after #3's copulation, male #4 courted
female A4 twice in less than 30 seconds and was rejected each time
following the vibratory display. Male #3, resting in the shade, showed
no interest in this activity. However, at 1249, #4 approached female
#1 very rapidly and after an abbreviated vibratory display, jumped up
and bit at her neck. Simultaneous with her attempts to reject him,
#3 postured (see SOCIAL ORGANIZATION) and charged from his resting
place and chased #4 away from "his" female. Following the short chase,
#4 bobbed submissively, as #3 moved back to his resting place. Within
30 seconds, #4 again moved to court #1 and again #3 postured, charged,
drove #4 from the vicinity of the female, and moved into the shade to
rest. This same sequence was repeated five times over the following
five minutes; 20 to 60 seconds separated each sequence. At that time
(about 1256 hrs), male #2 moved to the feeding station and this seemed
to stifle #4's advances temporarily.
No interaction occurred until 1321 when #4 resumed his courting
interest in #1 and #3 again chased him away from her. The sequence was
repeated six more times, each separated by 60 to 150 seconds, over the
next nine minutes. Male #4 dragged his cloaca on the substrate three
times during the following nine minutes and retreated to a burrow for
the remainder of the day at 1341 hrs. Number 3 male did not attempt
to "guard" any female during any other day of observation. This pro-
tective behavior thus appears to be restricted to the day of copulation.
Assuming females are receptive for only a short time during the
ovarian cycle, the practice oF guarding a recently-copulated female by
a male in the field would help guarantee that he, alone, contributed to
that female's entire brood that year. Parker (1974) has discussed the
significance of this behavior in his theoretical treatment of female-
guarding as a male time-investment strategy.
The mating system of Cyclura carinata is apparently primarily
monogamous with serial polygyny among some males (see Wilson, 1975).
There appears to be pair-formation during at least the breeding season.
Two females at the SW Blind lived within the home range of the same
respective males for two years until all were killed by dogs and cats.
Monogamy and/or polygyny seem to be the predominant systems among the
iguanines. Although no breeding observations were made, pair bonding
and monogamy have been suggested for Cyclura pinguis (Carey, 1975).
Johnson (1965) noted male-female pairing in Sauromalus obesus, but did
not observe courtship. In a more complete behavioral study, Berry (1974)
found a polygynous system with pair formation for the same species.
Male Mona Island Cyclura cornuta are generally polygynous but some
are apparently promiscuous; females are monogamous, polyandrous, or
promiscuous (Wiewandt, in preparation). Street (1952) implied that
Cyclura nubila is promiscuous, but no supporting data were presented.
His observations were based on tame, unrestrained lizards inhabiting a
golf course in Cuba, but may not be representative of field populations.
Alvarez del Toro's (1972) observations of Iguana iguana suggest a
polygynous mating system for the species. However, Peracca (1891,
cited in Noble and Bradley, 1933) observed that captive male iguanas
mated with more than one female, but did so more with one female than
another. He also noted multiple insemination of females (up to eleven
times). This suggests a polygamous if not totally promiscuous system,
but it may have been an artifact of the captive situation. Mueller
(1972) reported that the species is polygamous in Columbia. Further
studies of courtship behavior as well as all aspects of social behavior
are sorely needed for lizards of this subfamily if meaningful comparisons
are to be made.
Female Cyclura carinata oviposit their single annual clutch in
early June. Two gravid Pine Cay females had fully shelled oviducal eggs
on June 3 and 9 and four recently deposited nests were excavated on June
5, June 7, and June 9 (2) on Pine and Water Cays. A female at the SW
Blind site laid her eggs June 6. A female captured June 5 on Water Cay
nested in an observation pen four days later. Captive lizards tended
to nest later than native lizards, perhaps due to subnatural thermal
regimes. A Pine Cay female collected April 4 deposited her clutch on
June 12 in Gainesville. A Water Cay female, captive for nearly one year,
dropped her eggs on July 25, 1975 (David Auth, personal communication).
Soil for burrowing was not available in the pen in 1975 and the eggs
were deposited on the enclosure's substrate. The female had ceased
feeding two weeks previous to oviposition and resumed immediately after
Since copulation was not observed in the field, I have no way of
knowing with certainty the relation of a female's nest site to the
territory of the male with which she copulated. However, burrows, or
branches thereof, offering optimum nesting conditions (see later) are
limited in many areas due to substrate conditions and high lizard
densities. Therefore, female nest site location is probably not
closely determined by the location of the territory of the male with
which she copulated, but rather by nest site availability. A clustered