This item has the following downloads:
|Table of Contents|
Table of Contents
List of Tables
List of Figures
Methods and materials
Description of the study area
Ecology of reproduction
Population size and density
POPULATION ECOLOGY OF THE GROUND
SKINK, LYGOSOMA LATERAL [SAY]
GARNETT RYLAND BROOKS, JR.
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
I wish to express appreciation to Dr. Coleman J. Goin,
Chairman of my Supervisory Committee, for his direction of my
graduate program and research, and to the other members of my
committee, Drs. Yoneo Sagawa, A. M. Laessle, B. McNab, and E. R.
Jones, for their suggestions concerning the study and for read-
ing and correcting the manuscript.
I also appreciate the help of the following people at the
University of Florida: Mr. George Zug who supplied several speci-
mens; Dr. Carter Gilbert for a valuable specimen and for construc-
tive criticism of portions of the manuscript; Dr. A. M. Laessle and
Mr. Timothy Brown for identification of the plants on the study
area; and Dr. J. N. Layne for the use of his data on the activity
of Lygosoma laterale and for helpful discussions concerning questions
on density and home range.
Finally, I wish to thank the Department of Biology, the
College of Arts and Sciences, the Graduate School of the University
of Florida, and my wife for granting me the privilege of graduate
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ................................................. ii
LIST OF TABLES .................................................. v
LIST OF FIGURES .................................................. vi
I. INTRODUCTION ............................................. 1
II. METHODS AND MATERIALS .................................... 3
III. DESCRIPTION OF THE STUDY AREA ...............
IV. SPATIAL RELATIONSHIPS .................................... 14
Distribution Within the Plot ........................... 14
Home Range ............................................. 16
Territory .............................................. 24
Dispersion ............................................. 30
V. GROWTH ......... .................. ....................... 38
VI. ECOLOGY OF REPRODUCTION ..........................
Sexual Maturity ................................
Breeding Season ................................
Sex Ratio ......................................
Reproductive Potential .........................
VII. POPULATION SIZE AND DENSITY .............................. 55
VIII. POPULATION DYNAMICS ...................................... 59
Method of Determining Year Classes ..................... 59
Age Structure .......................................... 64
Survivorship ........................................... 66
Causes of Mortality .................................... 72
IX. ACTIVITY ................................................. 78
Temperature ............................................ 7P
Moisture ............................................... 82
Time of Day ............................................. S
Season ................................................. R6
X. DISCUSSION ............................................... 87
XI. SUMMARY .................................................. 93
LITERATURE CITED................................................. 97
BIOGRAPHICAL SKETCH............................................. 102
LITS OF TABLES
1. A list of amphibians and reptiles in addition to
Lygosoma laterale found on the study plot............... 13
2. Poisson analysis of the spatial distribution of the
lizards on the plot during June and July, 1961.......... 15
3. Comparison of the number of adults found in pine-
influenced areas with the number in nonpine areas
during June and July, 1961.............................. 18
4. Selected records showing permanence of home range....... 22
5. List of occasions when two lizards were caught at the
same time in the same place............................. 28
6. List of individuals which made a change in their
spatial position........................................ 31
7. List of individuals which evaded capture for an un-
usually long period of time between successive captures. 36
8. The instantaneous relative growth rates (k) for both
sexes of the 1960 hatchling group....................... 42
9. The reproductive potential of the female population on
the plot during the breeding season of 1961............. 53
10. Records of capture-recapture data of the 1961 hatchling
group from July, 1961, to March, 1962................... 54
11. Number and density of skinks on plot between July, 1960,
and August, 1961, by month............................... 57
12. The age composition of the population on the plot from
July, 1960, to March, 1962, listed by sex .............. 65
13. Life table for the 1960 hatchling group of Lygosoma
laterale based on data derived from Figure 14........... 70
14. List of predators of Lygosoma laterale recorded in the
LIST OF FIGURES
1. Diagramatic sketch of lizard feet to illustrate the
numbering system used in this study..................... 4
2. A map of the study plot showing major features of the
vegetation................... ........... ..... . ..... 8
3. Rainfall, average temperature, and number of ground
skinks active per field trip for each month from July,
1960, to March, 1962 ................................... 12
4. Position of adults on the plot captured during June and
July, 1961 ..................................... ...... 17
5. Home ranges of mnles living in an area of the plot bor-
dered by stakes 6A-6C-3C-3A during June, July, and
August, 1961........................................ ... 26
6. Home ranges of females living in an area of the plot bor-
dered by stakes 6A-6C-3C-3A during June, July, and
August, 1961 ............................................ 27
7. Percentage of new lizards in the total catch per month.. 37
8. Theoretical growth curves based on growth rates of the
hatchling group of 1960................................ 41
9. Yearly changes in the size of testes.................... 46
10. Growth of ovarian follicles and the size of oviducal
eggs. ................................................ . 47
11. Snout-vent length at month of first capture of females.. 62
12. Snout-vent length at month of first capture for males... 63
13. The progressive decrease in number of the 1960 hatch-
ling group.............................................. 69
14. Survivorship curve of the 1960 hatchling group of
Lygosoma latcrale...................................... 71
15. Number of ground skinks captured at each recorded field-
shade temperature.............................. ....... 80
LIST OF FIGURES continued
16. Average number of ground skinks captured per field
trip at selected temperature ranges..................... 81
17. The relationship between shade temperature, moisture,
and number of active ground skinks...................... 84
Until recently very little was known about the dynamics of
natural populations of lizards. Gordon's (1956) work on the popu-
lation ecology of Anolis carolinensis, Fitch's (1954) paper on the
life history of Eumeces fasciatus, and Blair's (1960) detailed study
of a population of Sceloporus olivaceous are the major articles con-
taining data on the population ecology of these animals. The last
of these, which is the result of five years of work, well illustrates
the information obtained from such studies. Recent papers concerning
populations of other herptiles are those of Stickel (1950) and Sexton
(1959) on turtles, Pearson (1955), Martof (1956a, 1953b, 1956a, 1956b),
Turner (1960a), Jameson (1955), Bannikov (1950) and Pyburn (1958) on
anurans, Carpenter (1952) on snakes, and Organ (1961) on salamanders.
The present study was undertaken to further our knowledge re-
garding natural populations of lizards. The main objectives were
to investigate the spatial relationships, growth rate, reproductive
potential, age structure, density, and activity of a small population
of Lygosoma laterale (Say). This lizard, the ground skink, was chosen
as the object of study for several reasons: it is extremely abun-
dant in local areas, it is easily captured and marked for future
recognition, it is active the year round in Florida, and it has a
relatively short life span.
The ground skink is one of the more abundant and widely
distributed members of the Scincidae in the southeastern United
States. It is found in deciduous, mesophytic forests, pine woods,
gardens, wooded fields, at the edge of pasture and forest, and in
general, in most wooded areas where there is sufficient cover,
food, and moisture. In these habitats it feeds on the small insects
and spiders which live in the litter layer.
No detailed study of the life history of Lygosoma laterale
has yet been published, although Lewis (1951) studied certain features
of its biology in central Texas, and Johnson (1953) has determined
its reproductive cycle in Louisiana. Several other papers have ap-
peared concerning food habits (Slater, 1949; Hamilton and Pollack,
1961), parasites (Harwood, 1933), population size (Turner, 1960b),
and distribution (Carr, 1940; Smith, 1946; and Conant, 1958).
A statement from Turner's (1960a) study on a frog population
is especially applicable: "Data from such studies," i.e., population
studies on amphibians and reptiles, "are sorely needed, for our
knowledge of vertebrate population ecology is based almost exclusively
on studies of fish, birds, and mammals."
METHODS AND MATERIALS
Field work, involving the capture, mark, and recapture of
lizards, was carried on from July 22, 1960, to April 14, 1962.
During this period collections were made from nearby populations
and preserved for a study of food habits, parasites, and the
reproductive cycle. All lizards were captured exclusively by
hand. Specimens were marked for future recognition by clipping
toes with manicure scissors, and since this species of skink has
five toes on each limb, a large number of different combinations
was available. Toes on the right, rear foot stood for units,
with the first, or most medial, toe representing number one; the
second toe, number two, etc. Combinations of toes were used for
numbers higher than five. Toes on the left, rear foot stood for
tens, with the first representing the number ten; the second,
twenty, etc. Combinations of toes were used for numbers larger
than fifty. Similarly fingers on the right front limb repre-
sented hundreds, and those on the left front limb, thousands.
The numbering system is illustrated in Figure 1. No more than two
toes were removed from one limb, and usually only four or five were
removed from the entire animal. Toes lost by natural conditions
were incorporated as well as possible into the numbering system.
Individuals which had lost other toes after being marked could be
identified by a combination of sex, location, body length, tail
\ A 1000
I /, 500
Figure 1. Diagrammatic sketch of lizard feet to illustrate the
numbering system used in this study.
length, and length of regenerated tail data. No regeneration of
clipped toes was observed. The loss of toes did not seem to hamper
the lizards in their normal activity. In fact, several individuals
were collected on the plot and from other locations which had lost
entire limbs, seemingly with no detrimental effects. An attempt
was made early in the study to mark specimens with paint spots in
addition to clipping toes, but since the paint wore off too quickly
this method of marking was abandoned.
Three body measurements were made in the field: 1) snout-vent
length or the distance from the tip of the snout to the anterior
margin of the vent; 2) tail length or the distance from the posterior
margin of the vent to the tip of the tail; and 3) the length of any
regenerated portion of the tail. Measurements, to the nearest mm,
were made by suspending the lizard near the front limbs between thumb
and index fingers of the collector, and applying a clear mm rule
against the body of the lizard.
The sex of adults was determined by applying gentle pressure
to the region just posterior to the vent. If the individual was
a male, the hemipenes were easily detected, the absence of hemipenes
indicating a female. There is no easily observed difference in
scalation or color between males and females. Because of size the
sex of hatchlings and juveniles could not be determined.
The plot was visited 115 times during the course of the
study. The procedure followed during a trip was to start at either
the north or south border, and work horizontally back and forth
(eastward and then westward) over the plot, thus scrutinizing the
entire area. The location of a lizard, its sex, body measurements,
the air temperature, micro-habitat, time, and weather were recorded
in a field notebook and later transferred to McBee Key Sort cards
(type Ks371N). Temperature was recorded by a centigrade thermometer
placed in the shade one inch above the litter layer.
Statistical methods used in the following analyses were taken
from Snedecor (1956) and Simpson, et al. (1960). The following
abbreviations are used in the text; standard error of the mean as
S.E.m, number as N., and mean as T.
The botanical nomenclature follows that in "Gray's Manual of
Botany" (Fernald, 1950).
DESCRIPTION OF THE STUDY PLOT
The plot was located in a wooded portion of the University
of Florida campus, Gainesville, Florida. The large number of
ground skinks present, the type of habitat, and convenience made
this area an ideal one for study. The plot was approximately one
and one-fourth acres in size, and measured 120 yards by 50 yards, with
the long axis in a north-south direction (Fig. 2). A grid system,
composed of quadrats 10 yards by 10 yards, was laid out over the
plot. Grid lines running north-south were labeled with letters
(A to F) and lines running east-west by numbers (1 to 13). Wooden
stakes, each bearing a letter and a number, marked the intersections
of grid lines. The stake in the extreme southeast corner of the
plot was selected as 1A. Moving north from this stake, the numbers
went up to 13 (a distance of 120 yards). Moving west, the letters
progressed to F (a distance of 50 yards). Thus stake SC, for example,
was 40 yards due north of line 1, and 20 yards due west of line A.
The plot was bordered on the east side by a dirt road; on the south,
by a paved campus road; and on the southwest, by a thicket of small
trees and shrubs. The north and northwest edges had no natural
barrier but were continuous with further woods.
Large pignut hickories (Carya glabra megacarpa), sweet gums
(Liquidambar styraciflua), laurel oaks (Quercus laurifolia) and a
P C1 N ,
0 o o
A map of the study plot showing major features of the vegetation. Dots show position of
grid stakes; oak trees are indicated by circles, pines by crosses, hickories by triangles, and
gums sy squares. The letters and numbers designate grid lines. The lines represent logs.
few loblolly pines (Pinus taeda)dominated the canopy of the forest.
Approximately 20 per cent of the canopy was made up of the two ever-
greens, laurel oak and loblolly pine. Other trees scattered through-
out the plot were a few water oaks (Q. nigra), live oaks (. virginiana),
and a spanish oak (a. falcata). The canopy was continuous during the
warm months except for an opening at 7A and 7B. As in most deciduous
forests in Florida, the epiphyte, spanish moss (Tillandsia usneoides),
was present on most of the trees.
Smaller trees such as dogwood (Cornus florida), winged elm
(Ulmus alata), cow oak (Q. michauxii), a few ashes (Fraxinum caroliniana),
hop-hornbeam (Ostrya virginiana), and young pignut hickory made up the
understory. Shrubs present consisted of stiff dogwood (C. foemina),
hackberry (Celtis laevigata), red mulberry (Morus rubra), hawthornes
(Crataegus sp.), and a few southern black-haw (Viburnum rufidulum),
sparkleberry (Vaccinium arboreum), and strawberry-bush (Euonymus
Vines present on the plot were Virginia creeper (Parthenocissus
quinquefolia), grape (Vitus rotundifolia), and laurel-leaved green-
brier (Smilax laurifolia). A dense covering of herbaceous vegetation
dominated by Virginia creeper and poison ivy (Rhus radicans) carpeted
the forest floor in most of the northern and eastern quadrats. Other
herbs were violets (Viola walteri) predominantly in the southeast
sector, scattered panic grasses (Panicum spp.), and a few sedges
Since Lygosoma laterale is a ground inhabiting skink, the
litter layer assumes great importance. This layer was composed of
leaves, pine needles, twigs and sticks of assorted sizes, clumps of
fallen spanish moss, and decaying logs. The depth and composition
of the litter layer varied throughout the year; the greatest depth
of leaves occurred during the winter months. Microbial decomposi-
tion increased as spring arrived and by late fall some areas of
forest floor were exceedingly bare. Around the pines, however, a
fairly constant depth of litter was present the year round. The surface
soil of this region, a mixture of Arredondo loamy fine sand and fine
sand, is light brown in color, loose, and well drained (U.S.D.A.
Soil Survey, Alachua County, 1954).
Since the greatest portion of the canopy was deciduous, the
forest floor during the winter months was exposed to the sun, per-
mitting some activity of lizards on warm days. For a review of the
climate of the Gainesville region see Pearson (1955). The rainfall
and average temperature for each month of the study period is given
in Figure 3.
In addition to Lygosoma, a number of other species of reptiles
and amphibians were noticed in the study area. These species are
listed in Table 1.
Figure 3. Rainfall, average temperature, and number of ground skinks active per field trip for each
month from July, 1960, to March, 1962.
z 30 w
(D 26 >
cr 22 c
cr 20 -
I- 18 c
Table 1. A list of amphibians and reptiles in addition to
Lygosoma laterale found on the study plot.
southern leopard frog
green tree frog
southeastern five-lines skink
six-lined race runner
rough green snake
Distribution Within the Plot
After several months of field work it became evident that
the skinks were not uniformly distributed over the plot but were
captured more frequently in certain sectors. To examine the spatial
distribution, the position of 159 adults captured during June and/or
July, 1961, were recorded on a map of the plot (Fig. 4). These
months were chosen because of the large number of adults present,
and because hatching of the 1961 eggs had not yet begun in force.
The number of quadrats containing 0, 1, 2, 3, ..., n number of lizards
was then compared to a Poisson series (Table 2). One characteristic
of a Poisson series is that its variance is equal to its mean. Thus,
if the variance of the counts made on the quadrats is larger than the
mean per plot, a tendency toward aggregation is revealed (Dice, 1952).
Examination of Figure 4 clearly reveals that the lizards were
more abundant in the lower, eastern quadrats and that very few were
found along the southern and middle-western borders. Although the
chi-square value is not significant (P.<.05), a tendency toward ag-
gregation is indicated by the large variance (Table 2). Since there
were habitat differences, as will be pointed out below, within the
plot, a concentration of skinks in favorable locations would be
Table 2. Poisson analysis of the spatial
lizards on the plot during June
distribution of the
and July, 1961.
Number of Number Number
Lizards of Poisson (A E)2 of
Per Quadrat Quadrats Series E Lizards
0 9 4.24 5.344 0
1 11 11,35 0.011 11
2 10 14.89 1.596 20
3 10 13.15 0.755 30
4 11 8.71 0.602 44
5 4 4.62 0.083 20
7 5 3.04 1.264 7
Total 60 60.00 9.655 159
expected, The nonsignificant chi-square value might be due to chance,
that is, by chance the distribution of quadrats containing 0, 1, 2,
..., n number of skinks was not statistically different from that
of a theoretical Poisson series based on 60 quadrats.
To see if vegetational differences influenced distributions,
the plot was divided into areas which contained loblolly pines,
and areas which did not. The portions influenced by pines were arbi-
trarily selected as those areas in which pine needles comprised
most of the litter layer. These different regions are shown in
Figure 4. The area of pine-influenced regions was measured by a
compensating polar planimeter. This area was then subtracted from
the total area of the plot to obtain the nonpine area. A comparison
of the number of skinks living in each region during June and/or
July, 1961, is given in Table 3. The results indicate that in the
study plot, pine-influenced areas provided a more favorable habitat for
ground skinks than nonpine areas.
A comparison of the first recorded position of the hatchlings
of 1961, which in most cases indicates the site of hatching, with a
Poisson series indicated (X2 = 8.54) that the young were nonrandomly
distributed (P. = .01). As was found for the adults, the great
majority of hatchlings were found in the northern and eastern sectors.
The concept of home range has been defined as the area in which
an individual animal normally travels while engaged in its usual daily
activities (Dice, 1952; Burt, 1940). Dice (Ibid.) included breeding
3 4 7 9 10
4. Position of adu and July, 1961.
are enclosed by arked by a cross.
Table 3. Comparison of the number of adults found in pine-influenced
areas with the number in nonpine areas during June and July,
Area in Actual Number Expected Number Chi-
Square Yards of Lizards of Lizards Square
1519.7 60 40.3 9.63
4480.3 99 118.7 3.27
Total Chi-square 12.90*
*A significant value.
sites as a part of the individual's home range, but in this study
movement to a breeding site, or to an egg-laying site, was not included
in the measurement of any home range.
The method for determining size of home range is that described
by Mohr (1947). In this method, each capture point of an individual
is plotted on a map of the area, and the peripheral points are then
connected to form an irregular polygon. The area enclosed by the
polygon represents the "minimum home range." This method is subject
to error for, depending on the worker, polygons of different size can
be constructed by connecting different sets of points of one individual.
There is no definite rule to determine which polygon is most accurate,
but by studying the terrain and vegetational cover, a reasonable
facsimile of the true home range can be obtained. The size of a plotted
home range was determined by the use of a planimeter. Other methods
for estimating the size of home range can be found in Hayne (1950),
Calhoun and Casby (1958), Davis (1953), and Stickel (1954).
Before the average size of a home range can be calculated for
these lizards it must be first determined whether or not they possess
and utilize a definite AREA, i.e., a home range. To do this it is
necessary to find out whether or not the area in which captures are
made increases with an increase in the number of captures. If the
area does increase with an increase in number of captures no measure-
ment of home range is possible and there would be some question as to
whether or not there was any definite "home range" area. If, however,
after a certain number of captures, the area does not increase in size
with an increase in number of captures, a plateau has been reached
and use of the home range concept is justified.
Since in many other studies (Fitch, 1954, 1958; Blair, 1960;
and Tinkle, et al., 1962) it has been found that males and females
have different home range sizes, each sex will be considered separately.
In males, it was found by use of a correlation test that be-
ginning with at least five captures within a 12-month period, the
number of captures and the area over which the captures were made was
positively correlated (r = 0.559; P. = .01). In other words, the
more times an individual was captured, the larger was the area in
which the captures occurred. However, when these data were plotted
on a graph with size of area covered on the ordinate axis and number
of captures on the abcissa, it was noted that a curve drawn through
the points leveled off after reaching the point associated with
eight captures and remained level beyond this point. A correlation
test was then run comparing the area covered with the number of cap-
tures beginning with at least eight captures and ranging to 16. An
r value of 0.100 (P. .05) was obtained, which indicates no correla-
tion between these parameters.
Thus after eight captures of a male the area over which sub-
sequent captures were made remained constant and could be called a
home range. The average size of a home range determined from the
records of 16 males captured eight or more times during a 12-month
period was 62.5 square yards (S.E.m = 6.8).
In females, a correlation test between area covered and
number of captures starting with five captures gave an r value of
0.045 (P. <.05) which indicates no correlation. Thus the home range
of females can be determined from those individuals captured five
or more times. The average size of a home range determined from the
records of 36 females captured five or more times.during a 12 month
period was 17.2 square yards (S.E.m = 2.4).
The difference in home range size between males and females is
significant. The result of a "t" test (t = 2.51; P. = .05) indicates
that males have larger home ranges than females. The average size of
a male home range was 3.6 times as large as that for a female.
Comparisons of home range size between juvenile males and
juvenile females, and between juveniles and adults would have been
interesting, but could not be determined due to the small number of
recaptures of individual juveniles. Only 12 were captured more than
three times as juveniles out of 293 juveniles tagged. Blair (1960)
found that the home range size for juveniles of Sceloporus olivaceous
was small at first but as the individual grew, the size of its home
range also expanded.
Once a home range was established, a skink tended to remain
there throughout its life. Table 4 is a list of selected data show-
ing the permanence of some home ranges. A shift in home range did
occur in a few cases as indirectly indicated by Nos. 483 and 62. A few
individuals, No. 43 for example, tagged at the beginning of the study,
were still in the same range when field trips ended.
Table 4. Selected records showing
permanence of home range.
Time in Same Time Under
Home Range: Observation:
Number Sex Months Days Months Days
* Indicates a shift in
The advantages to an individual possessing a home range
have been attributed to the protective value offered by its familiar-
ity with the terrain, sources of possible danger, location of food
and water supplies, and shelter spots. Factors such as food
and water supplies probably play a small role in the advantages of
a home range for Lygosoma since these supplies are not usually lo-
cated at definite points within the home range. The protective
value, however, of knowing the position of shelter spots and the
shortest route to these spots is well illustrated by the behavior
of individual skinks.
Many times on the study plot skinks were observed before
they were startled. When startled they would unhesitatingly run
for a shelter spot at the base of a tree or under a log. Only in
rare instances would a skink attempt to hide under leaves, making
no attempt to seek shelter elsewhere. That this was not merely
chance choosing of the nearest shelter spot is evidenced by the
observation that on many times a certain skink, easily recognized
due to its stumped tail, would always run, no matter where its posi-
tion in its home range, to a certain tree where it had a shelter
spot between two exposed roots. This behavior was noticed in many
other skinks which could be recognized as certain individuals. A
few had several spots of refuge within their home range and would
run, when startled, toward the one closest in sight.
Other observations which bear out the importance of a home
range revolve around the behavior of liberated skinks. On several
occasions skinks were taken to the laboratory for identification.
When they were returned, always on the following day, and dropped
in their home range, they would immediately run for a shelter spot
previously utilized. On the other hand skinks when captured on
collecting trips off the plot and then released in foreign sur-
roundings would appear hesitant and in most instances burrow under
the leaf mold where dropped. It thus appears that the most obvious
advantage of a home range for the ground skink lies in its protective
In a great variety of animals a portion, or occasionally
all, of a home range is defended by the inhabitant against entry
by other members of the species. The particular method of defense
varies, some animals relying on bluff, others on actual combat.
The area defended is known as a territory and except in a few rare
cases, only the male of the species is known to show this trait.
The best evidence for the presence of a territory is, of course,
direct observation of defense of a home range, but the absence of
overlap between adjacent home ranges can be used as evidence also
To determine if home ranges overlapped, all home ranges of
adults living within a crowded portion of the plot at a given time
were mapped, and the amount of overlap measured. A three-quadrat
square in the eastern sector, 6A-6D-3D-3A, was chosen because of
the large number of lizards known to exist in this area during
June, July, and August, 1961. If a lizard was known to live in one
of these quadrats for two of the three months, its home range was
plotted. Figures 5 and 6 show the results for males and females
respectively. It is readily observed that there was considerably
less overlap in the home ranges of females than in males. Out of
23 female ranges mapped, there were only nine points of overlap,
with only eight square yards being shared. Of 22 male ranges,
however, there were 28 points of overlap. The area contained in
these regions of overlap was approximately 153 square yards, some
19 times more than in females.
One objection to the above procedure is that a few of the
home ranges are based on only three or four captures. The exten-
sive overlap of male home ranges can be illustrated, however, in
another way. If the number of each sex in this area is multiplied
by their respective average size of home range, it is found that
males cover an aggregate of 1375 square yards, whereas females
cover only 396 square yards. Since there are only 900 square
yards in these nine quadrats, the males undoubtedly shared more
space than females.
Further evidence indicating the isolation of individual females
is given in Table 5. This table lists the occurrences when two lizards
were caught in the same place at the same time. Of these 30 pairs,
the sex of both partners was known in 25. Only one of these 25 pairs
Home ranges of males living in an area of the plot bordered
by stakes 6A-6D-3D-3A during June, July, and August, 1961.
Capture points are represented by dots. The letters and
numbers designate grid lines.
D C B A
Figure 6. Home ranges of females living in an area of the plot bordered
by stakes 6A-6D-3D-3A during June, July, and August, 1961.
Capture points are represented by dots. The letters and num-
bers designate grid lines.
Table 5. List of occasions when two lizards were caught at the same time in the same place.
Type of Distance Apart
Interaction in Feet Date Microhabitat
Female with Female
Nov. 5, 1960
Female with Unknown Sex
Female with Unknown
Sep. 24, 1961
Sep. 24, 1961
Mar. 2, 1961
Apr. 16, 1961
under stick matted with spanish moss
in leaves near log
in mat of spanish moss and sticks
in mat of spanish moss and sticks
in leaves at edge of spanish moss mat
in sticks and leaves
in leaves near sticks
in leaves in open
in leaves beside log
in leaves at base of tree
under mat of spanish moss
in copulation in leaves
in mat of spanish moss
in leaves near base of tree
in leaves and twigs
in leaves along side of log
in small brush pile
in leaves in open
in leaves fighting
Table 5 continued.
Type of Distance Apart
Interaction in Feet Date Microhabitat
Male with Male
on top of
mat of spanish moss
at base of tree
of spanish moss
of spanish moss
Male with Hatchling
Nov. 5, 1960 in pine needles near stick
Ohly one instance of fighting was noticed in the field and
only one of the combatants, a female, was captured. Skinks of both
sexes, when confined in terraria, invariably began to fight, one of
them, a large female in most cases, assuming dominance over the
Dispersion within the plot and dispersion outward or inward
from peripheral areas have entirely different effects on the popula-
tion. The first type of movement merely shuffles the lizards within
the plot, while the second type increases or decreases the popula-
tion size. For this reason the two types are reported separately
Dispersion within the plot. As mentioned above, the great
majority of home ranges were permanent. A few cases were recorded,
however, of a shift in home range or movement from one point in the
plot to another. In analysing movement only individuals with two or
more captures at least two months apart were used in calculating
percentages. All movements were assumed to have been in a straight
A distinction was made regarding the age at which dispersion
was detected. Three categories were devised: 1) dispersion of the
lizard after reaching adult size, 2) dispersion as it grew from
juvenile to adult, and 3) dispersion of hatchlings. Table 6 lists
the individuals for which data on dispersion within the plot were
Table 6. List of individuals
which made a change in their spatial
Type of Distance Direction
Number Sex Movement Moved in Yards Moved
4a female A-A* 83.5 S.E.
38 30.7 S.
67 34.1 S.E.
483 28.3 N.W.
69 67.7 N.E.N.
36 37.4 S.E.S.
441 37.4 E.N.E.
14 male 42.7 W.N.W.
308 31.7 S.E.S.
27 20.2 N.W.N.
302 29.8 S.E.S.
62 19.7 N.W.N.
50a female J-A** 29.3 N.W.
447 37.0 E.S.E.
40a 62.4 S.E.S.
426 20.2 S.W.
400 male 37.9 N.
484 53.8 N.
55b unknown H-H*** 27.4 S.E.S.
3098 37.9 S.
Dispersion of adults. Of 181 records for adults taken from
the entire study period, only 12, or 6.6 per cent, showed a shift in
their spatial position. This is undoubtedly a low estimate, for some
movements were probably not detected. The average distance moved
was 38.6 yards. There was little difference in the number of females
and males moving; five of 89 males (5.6 per cent) moved an average
distance of 28.8 yards, and seven of 93 females (7.5 per cent)
moved an average distance of 45.6 yards.
Dispersion of juveniles. A change in position of a lizard
while growing from juvenile size to adult occurred in six of 51
specimens (11.8 per cent). The average distance moved was 40.1 yards.
Dispersion of hatchlines. Only two records out of 24 (8.3 per
cent) revealed movement of hatchlings. One hatchling moved 27.4
yards; the other, 53.8 yards.
The direction of dispersion did not seem to follow any natural
lane, but was essentially random. Seven movements were southeast;
two were northeast; one was southwest; five were northwest; two
were north; one was south, and none were east or west. Any extensive
movement to the east or west would have carried the lizard out of
All movements were one way; no lizard was detected moving
from one spot to another and then later returning to the original
position. This type of travel might occur, however, during the
breeding season by both sexes.
The speed of dispersion could not be measured since in all
but one case, described below, a considerable length of time separ-
ated the successive captures which indicated a change in position.
A gravid female, No. 36, captured 20 feet northwest of stake SC on
June 28, 1961, was found spent four days later, July 2, 1961, four
feet southeast of stake 2B. She had moved 37.5 yards on what might
have been a search for an egg laying site.
Emigration and immigration. Several field trips were de-
voted to marking skinks outside the boundaries of the plot, and
searching for marked lizards which might have emigrated from the
plot. A total of 37 were marked at different times during the
study from areas surrounding the plot. Eight individuals were
marked in the region bordering the northwestern boundary, 14 in the
region bordering the northern boundary, and 15 in the woods on the
eastern side of the dirt road. None of these individuals were ever
recovered within the plot. Likewise no skinks marked within the plot
were ever discovered outside, except for a few individuals whose home
ranges overlapped both regions.
The only nonrestricted pathways available for dispersion were
along the northwestern and northern borders. At these points the plot
was adjacent to further woodland. The paved road at the southern bor-
der prevented movement in this direction. The eastern border, a dirt
road used and maintained by the cross country track team, was kept
free of branches, weeds, etc., and was probably a partial barrier.
No skinks were ever observed in this road, although movement across
it might have occurred whenever a fallen branch or vine provided a
pathway. The small trees and shrubs along the upper southwestern edge
were a partial barrier due to the differences in vegetation structure.
As pointed out by Pearson (1955) one means of detecting im-
migration is to study the percentage of new specimens taken at each
trip. After a period of time one would expect that all the residents
would be marked, and the percentage of new animals would approach
zero. The appearance, therefore, of any unmarked individual would
indicate immigration. This type of analysis has one major drawback;
that is, at what time are all of the permanent residents marked? By
chance, some individuals, fulltime residents of the area, might be
active only during times when the plot was not being searched. One
skink, No. 424 for example, was first marked October 16, 1960, and
was not recaptured until January 23, 1962, still in the same area in
which it was first captured. Table 7 is a list of such lizards in
which an unusually long time separated successive captures. Undoubtedly
these lizards were active between the two dates listed, but this activity
was not observed. Lewis (1951) reported that one Lygosoma he had under
under observation in a study plot was active almost daily for several
The percentage of new skinks in the total catch per month,
plotted against successive months, is shown in Figure 7. The hatchlings
of 1961 were omitted.
The curve drops rapidly at first, but levels off beginning in
August, 1961, and remains fairly level for the remaining months. Assuming
that the great majority of the skinks were tagged in the first year
of study, Figure 7 indicates that immigration was constant, averaging
approximately 8.9 per cent, or about 5.2 lizard immigrants per month.
The high number of new animals taken from March, 1961, to July, 1961,
were composed mostly of young of 1960 which had not been tagged before
Table 7. List of individuals which evaded capture for an unusually
long period of time between successive captures.
Date of First Date of Second Number of Days
Number Sex Capture Capture Between Captures
87 female Aug. 26, 1960 Sep.. 3, 1961 373
26 Aug. 26, 1960 Aug. 15, 1961 354
477 Oct. 9, 1960 July 9, 1961 273
424 Oct. 16, 1960 Jan. 23, 1962 464
402 Dec. 18, 1960 Jan. 25, 1962 403
447 Oct. 1, 1960 Oct. 30, 1961 394
29 male July 29, 1960 June 28, 1961 334
62 Oct. 16, 1960 Oct. 30, 1961 379
445 Sep. 23, 1960 Dec. 17, 1961 420
484 Nov. 5, 1960 Feb. 15, 1962 467
7 8 9 10 I 12 1 2 3 4 5 6 7 8 9 10 11 12 1 23
1960 1961 1962
Figure 7. Percentage of new lizards in the total catch per month. The hatchlings of 1961 were
The growth rate of Lygosoma lateral has never been determined
from field data. Johnson (1953) presented two graphs containing
measurements of preserved specimens from monthly collections, but
made no conclusions concerning rate of growth. Barwick (1959) de-
termined the growth rate of Leiolopisma zelandica, a closely related
form from New Zealand, and the growth rate of several species of
Eumeces has been determined by Fitch (1954), Rodgers and Memmler
(1943), and Breckenridge (1943).
In determining the growth rate, capture-recapture data from the
hatchling group of 1960 were used exclusively; no measurements from lab-
reared young were utilized. A chart for each sex was compiled which
listed the body length of each individual for each month it was cap-
tured. The sample size for each month for each sex is given in Table 8.
If an individual was captured more than once in a month and the snout-
vent measurements were different, an average value was substituted into
the chart. The average snout-vent length for each month was then ob-
tained, and these values used to calculate a second-degree equation to
represent the rate of growth (see Snedecor, 1952, for methods).
The equation which describes the growth curve for females is
Y = 21.2308 + 2.0691X 0.0480X2; that for males is Y = 20.7713 +
1.9473X 0.0491X2. Snout-vent length in mm is represented by Y,
age in months by X. The curves derived from these equations are
shown in Figure 8.
The calculated values of Y from the above equations were used
to determine the instantaneous growth rate between successive time
intervals (Brody, 1945). For convenience each month was considered
to consist of 30 days.
The instantaneous relative growth rate is obtained from the
k = Log Ll Log L2 where
Log L2 and Log Ll are the natural logarithms of snout-vent length,
T2 and T1 are units of time associated with the L's, and k is the
instantaneous relative growth rate. The instantaneous relative
growth rate for each sex is given in Table 8.
As Brody states; "The constant k has a perfectly definite
meaning. It is the instantaneous relative rate of growth for a
given unit of time. Thus, for the growth of a fetus of the albino
rat, from 14 days to birth, the value of k is 0.53; this means that
the instantaneous percentage rate of growth is about 53 per cent
In the months just after hatching, growth is relatively fast
in both sexes (Fig. 8). Females, however, soon began growing at
a faster rate than males, and by adulthood, averaged 3 to 5 mm more
in snout-vent length. Growth continues throughout life, although the
increase in body length is relatively small. Several old females
attained 50 mnun in body length. Although not indicated in the curves
(Fig. 8), very little growth occurred during the winter months.
0 I 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16 17 18 19 20 21 22
AGE IN MONTHS
Theoretical growth curves based on growth rates of the hatchling group of 1960. The upper
curve is that of females; the lower, males.
Table 8. The instantaneous relative growth rates (k) for both
sexes of the 1960 hatchling group.
Age Sample Size Males Females
Months Days Males Females Months Days Months Days
1- 2 30- 60 27 31 0.08 0.003 0.08 0.003
2- 3 60- 90 11 9 0.07 0.002 0.07 0.002
3- 4 90-120 16 27 0.06 0.002 0.06 0.002
4- 5 120-150 9 9 0.05 0.002 0.06 0.002
5- 6 150-180 12 8 0.05 0.002 0.05 0.002
6- 7 180-210 7 5 0.04 0.001 0.04 0.001
7- 8 210-240 11 3 0.04 0.001 0.04 0.001
8- 9 240-270 11 12 0.03 0.001 0.04 0.001
9-10 270-300 15 8 0.03 0.001 0.03 0.001
10-11 300-330 12 7 0.03 0.001 0.03 0.001
11-12 330-360 17 16 0.02 0.001 0.03 0.001
12-13 360-390 19 24 0.02 0.001 0.02 0.001
13-14 390-420 15 9 0.02 0.001 0.02 0.001
14-15 420-450 13 10 0.01 0.02 0.001
15-16 450-480 3 4 0.01 0.01 *
16-17 480-510 3 2 0.01 0.01 *
17-18 510-540 5 1 0.01 0.01 *
18-19 540-570 12 8 ** 0.01 *
19-20 570-600 15 8 ** ** *
20-21 600-630 8 4 ** ** *
21-22 630-660 2 2 ** ** *
* Less than 0.0005.
** Less than 0.005.
ECOLOGY OF REPRODUCTION
Sexually mature males are defined as those producing sperm,
as evidenced by enlarged testes; sexually mature females are those
which contain enlarged ovarian follicles or oviducal eggs. In
Florida both sexes become sexually mature at approximately 35 mm in
snout-vent length. Ground skinks in Louisiana also reach sexual
maturity at 35 mm snout-vent length (Johnson, 1953). Johnson used
the presence of spermatozoa in the testes and/or ducts, and the
presence of enlarged ovarian follicles and/or oviducal eggs as cri-
teria for sexual maturity.
These results do not imply, however, that all individuals
35 mm and above in snout-vent length were sexually mature. Four
females preserved during the breeding season with snout-vent lengths
of 35, 36, 36, and 38 mm were found to have undeveloped follicles.
The same situation is probably applicable in males although no
smears were made to detect sperm. Therefore only a percentage of
those 35 mm and slightly larger contribute to the reproductive pool.
Of 19 female specimens 35-39 mm in snout-vent length preserved during
the breeding season, 15 or 78.9 per cent, were sexually mature. All
specimens 40 mm and over were sexually mature.
Sexual maturity is reached in less than a year in those
hatched in late June, July, or August. Those hatching later in
the year do not enter the reproductive pool until the succeeding
breeding season. Assuming the same hatching date, females will be-
come sexually mature earlier than males since females grow faster
There was no indication of a loss of reproductive powers as
a female aged. On the contrary, larger (older) females produced
more eggs than smaller (younger) females (see below). Blair (1960)
found that females Sceloporus olivaceous reached a reproductive
peak at about four to five years of age, after which their repro-
ductive potential regressed.
The breeding season, or period of sexual activity, for Lygo-
soma in Florida was determined from measurements of the gonads of
preserved specimens. Figure 9 shows the cyclic enlargement and re-
gression of the testes; Figure 10, that of the ovarian follicles.
From this data the breeding season was estimated to occur from Jan-
uary through July for males, and from February through July for
females. Johnson (1953), also from a study of preserved specimens,
concluded that the breeding season in Louisiana was from January
through August for males, and from December through August for females.
Field observations of sexual activity in this lizard have never been
recorded in the literature. The absence of dimorphic pigmentation
may be one of the reasons why this information is lacking. In other
species of skinks, Eumeces laticeps and E. egregius for example, the
males show a change in degree of pigmentation when sexual activity
begins. Thus a handy, external guide to their sexual activity is
Copulation was observed only once. On July 14, 1961, a
copulating pair was noticed on top of the leaves near stake 5E. Air
temperature was 270C, there was a slight breeze, and the sun was
shining brightly. When first observed (9:08 A.M.) the right hemi-
penis was inserted; three minutes later (9:11 A.M.) they broke apart
and started to burrow into the leaves. The male was 39 mm in snout-
vent length and the female was 42 mm. The copulatory position was
similar to that depicted by Barwick (1959) for Leiolopisma zelandica.
Ovarian follicles start to enlarge just after the end of the
breeding season (August-September), but do not increase rapidly in
size until the middle of February (Fig. 10). Between February and
April the follicles increase four fold in width. Oviducal eggs are
present from April through August although by August most have been
Seven of 21 specimens preserved in June and July, 1961, con-
tained enlarged ovarian follicles and extremely extended and thin-
walled oviducts. This condition of the oviducts is characteristic
of females which have just laid eggs. All seven were 42 mm or above
in snout-vent length. The follicles of those from June averaged 1.94
mm in width; those from July, 2.99 mm in width. From Figure 10 it
JAN FEB MAR
(14) (15) (II)
JUL AUG SEP OCT NOV DEC
(11) (9) (8) (7) (8) (13)
Yearly changes in the size of testes. Data taken from pre-
served specimens. The number enclosed by parentheses below
each month is the number of specimens examined. Line 1 re-
presents the length of the right testis; line 2, length of
left testis; line 3, width of right testis; and line 4,
width of left testis.
1 1 1
Growth of ovarian follicles and the size of oviducal eggs.
Data taken from preserved specimens. The numbers) enclosed
in parentheses is the number of specimens examined. In April
through August, the first number represents the number of
specimens examined to determine width of follicles; the second,
of oviducal eggs. The upper line represents width of oviducal
eggs; the lower line, width of ovarian follicles.
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC
(8) (12) (6) (6-3) (3-4) (7-4) (6-3) (6-4) (16) (9) (9) (7)
appears possible that these follicles could have developed into
oviducal eggs by late August. Johnson found in several females 44 mm
or more in snout-vent length both oviducal eggs and ovarian follicles
2 mm or more in diameter. Number 432 on the plot was recorded as
gravid on March 23 and 27, 1961, and again on July 21, 1961. Four
months separated these dates, which is ample time to lay one clutch
and develop another. From these data it is evident that a small per-
centage of the larger females do produce two clutches of eggs per
Of 303 adults sexed in the study area, 152 were males and 151
were females, an almost perfect 1:1 ratio. The sex ratio of young
was determined from specimens 34 mm or below in snout-vent length
collected from July, 1961, to November, 1961. There were 41 males
and 39 females.
The number of males and females within the hatchling group
of 1960 for 20 consecutive months is given in Table 12. The number
of males and females decreased at approximately the same rate, neither
sex showing a higher mortality rate than the other. In contrast, a
differential mortality rate between the sexes has been found for
several other species of lizards. Blair (1960) observed that male
Sceloporus olivaceous suffered higher attrition than females. Hirth
(1962) found a higher mortality rate among male sub-adults than
among females for Ameiva quadrilineata and Basiliscus vittatus.
To determine the reproductive potential of the population
two main types of data were required: 1) the number of sexually
mature females alive on the plot during the breeding season, and 2)
the average number of eggs produced per female. The first was obtained
from capture-recapture results and preserved material; the second, ex-
clusively from preserved material. Of secondary consideration is the
possibility of a female producing more than one clutch of eggs in one
Reproductive potential of the 1961 breeding population. There
were 54 females from the 1960 hatchling group present on the plot in
the spring of 1961. Possibly all had reached sexual maturity. Thirteen
of these 54 were gravid, and eight of these were 40 mm or above in
snout-vent length, the other five were 39 mm or less. This left 41
individuals whose sexual maturity was in doubt. Eight of these 41
were over 39 mm in snout-vent length and thus may be presumed to have
been sexually mature. The other 33 were between 35 and 39 mm in snout-
vent length. As indicated before, 78.9 per cent of those females
with a snout-vent length of 35 mm to 39 mm which were dissected proved
to be sexually mature. Hence it will be assumed that 26 or 78.9 per
cent of the 33 under consideration were sexually mature. Thus out of
the 54 females, 47 are estimated to have been sexually mature in the
spring of 1961.
There were 57 females present on the plot which had hatched
prior to 1960. Since all were 40 mm or above in snout-vent length,
all were assumed to have been sexually mature. Of these 57, 30 were
recorded as being gravid, and the other 27 were alive for a consider-
able length of time during the breeding season. Fifteen individuals
over 42 mm in snout-vent length could possibly have laid two clutches
The distinction made between the two size groups is important
because the average number of eggs produced by individuals of the two
groups was different. The 15 preserved specimens between 35 mm and
39 mm in snout-vent length, all hatched in 1960, contained an average
of 2.13 eggs, whereas the average for specimens over 39 mm was 2.82
eggs. This separation of body sizes is arbitrary and does not imply
complete separation of those lizards hatched in 1960 (1960 young)
from those hatched in preceding years (pre-1960 young). Some of
those 40 mm in snout-vent length were 1960 young, but none below 40
mm were pre-1960 young. The average number of eggs in those individuals
which indicated a developing second clutch was 2.57.
A positive correlation (r = 0.640; P. = .01) was found to
exist between snout-vent length and number of eggs. That is, larger
females tend to produce more eggs than smaller females. Johnson (1953)
found no correlation between body length and number of eggs, but he
did not include in his computation (as determined from Figure 13 in
his paper) any specimen 39 mm or less in snout-vent length.
The estimated number of eggs produced by each group of females,
and the total number of eggs produced in the 1961 breeding season is
given in Table 9.
There were no data available concerning mortality of Lygosoma
eggs. A measure of egg survival, however, was obtained by dividing
the number of young tagged in the fall of 1961 by the potential number
of eggs laid in the preceding summer. Between August, 1961, and March,
1962, 109 hatchlings were tagged. The survival rate based on this
figure was 35.2 per cent, that is, a little over one out of every
three eggs hatched and the hatchling survived till capture.
The number of hatchlings tagged is probably a low estimate of
the true number on the plot. There were 183 individuals hatched dur-
ing 1960 that were tagged during the study, yet 79 were tagged after
March, 1961. In other words only 104 of these 1960 young were tagged
between July, 1960, and March, 1961.
The modified Lincoln index described by Ilayne (1949) was used
to achieve an estimate based on the capture-recapture data of the
young hatched during 1961. In this method the population estimate is
based upon the increase in the proportion tagged which appear in suc-
ceeding catches. The capture-recapture results are given in Table 10.
The reader is referred to Ilayne's paper for methods and also for pos-
sible sources of error. A population estimate of 130 hatchlings was
computed. Using this estimate the survival rate of the eggs was about
41.9 per cent.
In comparison with the survival rate found for eggs and young
of Sceloporus olivaceous by Blair (1960) both of the above estimates
are extremely high. Blair found that under normal conditions the po-
tential production was about 4000 individuals but only about 160 were
required to reach adulthood to maintain a stable population. Verte-
brate predators were the most important cause of nest failure, but
desiccation of the eggs, failure of the young to escape from the shell,
and failure of eggs to hatch also were instrumental in causing egg
mortality (Blair, 1960). These factors were probably responsible for
egg mortality in this study also. In the laboratory only two eggs out
of 27 incubated were lost, both from a fungus infection. None were
Table 9. The reproductive potential of the female population
on the plot during the breeding season of 1961.
Number of Sexually Average Number of Number of Eggs
Mature Females Eggs Produced Produced by Group
Females hatched during 1960
40 mm and above
16 2.82 45.12
39 mm and below
31 2.13 66.03
Females hatched prior to 1960
Number laying one clutch
57 2.82 160.74
Number laying a second clutch
15 2.57 38.55
Estimated number of eggs produced in 1961 310.44
Records of capture-recapture data of the 1961 hatchling
group from July, 1961, to March, 1962. The formulae be-
low were taken from Hayne (1949); P represents population
size, the other symbols are defined in the headings of the
Proportion of Total Number
Month Number of Captures Catch Previously Previously
of Previously Total Handled Handled
Capture New Handled (w) (Y) (x)
August 17 0 17 0.00 0
September 44 6 50 0.12 17
October 10 15 25 0.60 61
November 12 15 27 0.56 71
December 2 14 16 0.88 83
January 10 14 24 0.58 85
February 7 13 20 0.65 95
March 5 8 13 0.62 102
Y xyw = 6499.5
P = 129.7
POPULATION SIZE AND DENSITY
One of the objectives of this study was to determine the
density of Lygosoma in a favorable habitat and to compare this den-
sity with that of other populations of Lygosoma and with that of other
In determining density two different figures were computed.
One, minimal density, was based on the entire area of the plot plus
a buffer strip (explained below). The other, maximal density, was
based on the area remaining after subtracting regions of unfavorable
habitat from the area of the plot plus buffer strip.
In computing maximal density, unfavorable areas were considered
as those quadrats or portions of quadrats, in which no lizards, or only
one or two, were captured. Six quadrats (1A-2A-2D-1D, 1F-2F-2E-1E,
6F-7F-7E-6E, 7A-8A-8B-7B), or 600 square yards, were so designated.
The importance of adding a buffer strip to a study area was
was proposed by Dice (1938). This buffer stip is equal to one-half
of the diameter of the average home range and should be added to the
plot at points Ohere the plot is bordered by continuous habitat. By
adding this strip, the extension of home ranges of individuals living
on the inner periphery onto area outside the plot is considered in
computing density. Along the southwestern, northwestern, northern,
and northeastern edges of the plot, the wooded habitat was continuous
along the border. Skinks residing along these edges could possibly
range out of the plot at these points. One-half of the diameter of
the average home range was added between 13A and 13F, 7F and 13F, 1F
and 5F, and between 8A and 13A. The diameter of the average home
range was determined by averaging the male and female data used in
the section on home range. The diameter was 6.3 yards.
The minimal density computation was based on the area of the
plot (6000 square yards) plus the area of the buffer strip (640 square
yards), or on a total of 6640 square yards. The maximal density
computation was based on the total area minus the unfavorable area
(600 square yards) or on 6040 square yards.
The number of individuals estimated to have been alive on the
plot for each month from July, 1960, to August, 1961, is given in the
fourth column of Table 11. The number of lizards per 100 square yards
and per acre is also given in Table 11. At the end of the study, 499
lizards had been tagged within the plot.
As would be expected, the population underwent a cyclic change
in number of individuals. A peak density occurred in August when most
of the hatchlings appeared. The population then started to decline
in numbers, the most drastic loss occurring during the spring months
(see the section on Survivorship). The low point in population size
occurred in July. The highest density recorded on the plot was in
August, 1960, when there were 263 lizards per acre.
Turner (1960b), in a study in which different methods for de-
termining population size were compared, used Lygosoma laterale as the
Table 11. Number and density of skinks on plot between July, 1960, and August, 1961, by month.
Uncorrected* Density Corrected** Density
Number Number per 100 sq. yds. per 100 sq. yds. acre
Month Adults Young Total Adults Young Total Adults Young Total Total
July 178 178 2.68 2.68 2.95 2.95 131
August 174 183 357 2.62 2.76 5.38 2.88 3.03 5.91
September 327 4.92 5.41 240
October 321 4.83 5.31 236
November 284 4.28 4.70 209
December 278 4.19 4.60 204
January 272 4.10 4.50 200
February 270 4.07 4.47 199
March 263 3.96 4.35 193
April 247 3.72 4.09 182
May 230 3.46 3.81 169
June 213 3.21 3.53 157
July 185 185 2.79 2.79 3.06 3.06 136
August 147 130*** 277 2.21 1.96 4.17 2.43 2.15 4.59 204
*Density based on entire area of plot plus buffer strip, see text.
**Density based on favorable area of plot plus buffer strip, see text.
***Based on estimate derived from Hayne's method, see Table 10 and text.
test animal. He used two methods, a capture-recapture analysis known
as the Schumacher method (see Schumacher and Eschmeyer, 1943, and also
DeLury, 1958) and a removal method described by Zippin (1958), to es-
timate the size of an isolated population of the ground skink in Louis-
iana. An estimate of 175 lizards and 100 lizards was obtained from the
two methods respectively. Although the size of the study area was not
mentioned, a very crude approximation of the area involved was obtained
by utilizing Figure 1 in his paper. From this approximation an estimate
of the density was computed using both estimates of population size.
The area of the habitat around the skinks was estimated to be
1160 square meters or about 1388 square yards. Using the population
size determined by Schumacher's method, a density of 12.6 lizards per
100 square yards is found.
Turner (1960b) stated that the removal method probably gave
a low estimate of the population size. Thus the actual density was
probably closer to 12 lizards per 100 square yards than to 7. Both
densities, however, are much higher than that observed in this study.
In a study of Leiolopisma zelandica by Barwick (1959) a popula-
tion of at least 200 individuals were found within a small cemetery.
This population size was equivalent to a population of about 900 skinks
per acre, some four and one-half times as many as found in any one month
in this study (see Table 11). It is not clear, however, whether 200
individuals were alive at one period of time or whether this number
represents the total number of skinks tagged throughout the study. Even
if the second suggestion is correct, Barwick's population was still
larger than the one observed in this study.
The importance of survivorship and life table information in
population studies has been well emphasized by Deevey (1947). Very
little of this type of information can be found, however, in ecological
studies of reptiles and amphibians. Bannikov determined the population
structure for the toad, Bombina bombina (1950), and the salamander,
Ranodon sibiricus (1949). Turner (1960a) was able to compute minimal
survival rates for Rana pretiosa, and Organ (1961) successfully pre-
pared survivorship curves and life tables for five species of the sala-
mander genus Desmognathus. The survival rate for the lizard Sceloporus
olivaceous was determined by Blair (1960) but he did not include in his
book survivorship curves or life tables.
Method of Determining Year Classes
As Organ (1961) has pointed out, size-frequency histograms have
been used in the past in an attempt to separate age classes within a
given sample. This method easily separates young or newly hatched in-
dividuals from adults, but in most cases it is less successful in separ-
ating older year classes within the adult size group. The danger in
this method is illustrated by the mistake of Taylor (1935) who divided
a large preserved sample of Eumeces skiltonianus into 16 age groups.
It was later found by Rodgers and Memmler (1943) that the normal life
span of this lizard was only five to six years.
In this study an attempt was made to separate year classes by
correlating the snout-vent length at first capture, the month of first
capture, and the known growth rate.
In examining the growth rate of the female hatchlings of 1960
it was found that it took, on the average, 18 months to reach 42.4 mm
in snout-vent length. Thus if it takes 18 months to reach 42.4 mm, an
individual 44 mm or above tagged in August-October, 1960, must have
been hatched in August, 1958, or before, not in 1959. Those individuals
between 36 mm and 42 mm tagged in August-October, 1960, were 1959 young.
For male 1960 hatchlings it took 18 months to reach 40.0 mm in snout-
vent length. Thus any individual 41 mm or above tagged in August-
October, 1960, was hatched in August, 1958, or before, and any between
36 mm and 39 mm would be 1959 young.
By utilizing these growth data the longevity can also be estimated.
As mentioned above, it is possible to show that some individuals were
hatched in 1958, or before. These individuals would have been at least
24 months old when tagged in August, 1960. Six of these individuals
were still alive in March, 1962, 43 months after hatching. Since field
trips ended in early April, 1962, it is possible that a few individuals
hatched prior to 1959 might have lived for several more months. There-
fore the minimal estimate of the life span of Lygosoma laterale is be-
tween three years and seven months and four years. The longevity
record for Lygosoma casuarinae, the Tasmanian Skink, in the Philadel-
phia Zoological Garden was five years and four months (Conant and
To determine how many year classes were represented, and how
many specimens were in each, the sexed, pre-1960 hatchlings, each sex
separately, were taken and individual snout-vent lengths at the time
of first capture plotted against month of first capture. The results
are shown in Figures 11 and 12 for females and males respectively.
Two size groups are recognizable in Figure 11, the larger size group
composed of individuals hatched prior to 1959, and the smaller group
composed of 1959 hatchlings. Those specimens enclosed in the trapezoid
were considered as borderline cases. The record of each was closely
examined in respect to future growth. On the basis of this examina-
tion each specimen was assigned to either one or the other of the two
groups. In Figure 12 there were more borderline cases, but two main
groups were still discernable. Those enclosed in the trapezoid were
treated as in Figure 11. The group designated as pre-1959 hatchlings
contained mostly individuals hatched in 1958, a few hatched in 1958,
and possibly a few hatched in 1956. The hatchlings of 1960 were easily
separated from pre-1960 hatchlings on the basis of size.
A small group of pre-1960 hatched individuals whose sex was not
determined in the field was assigned to year classes and sex by compar-
ing their size and time of capture with Figures 11 and 12. Only 11
were unassignable as to sex. These 11 were divided so that six were
considered as females and five as males. All 11 were hatched in 1959.
One other small group of individuals remained to be classified.
This group consisted of those captured for the first time after July,
1961. Those individuals tagged as new through July, 1961, were
o= PRE-1959 YOUNG
*= 1959 YOUNG
50 o o x= BORDER LINE CASE, SEE TEXT
49 o o
48 0 0 o o o
| 47 o
- 46 oo00 o o o oo
0 45 o0 oo oo ooo oo 0
-.J 44 o 0 oo 0o o o o 0 0
,W 43- x xxx x xxx * *
-- 42- ** 0
th 41- eeee 0 ee e
JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG
Figure 11. Snout-vent length at month of first capture of females.
o= PRE-1959 YOUNG
*= 1959 YOUNG
x = BORDER LINE CASE, SEE TEXT
0 0000 0 000
/ X xxX xxx
x xx x x x
x x xx 90
* ***** *
JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN JUL AUG
Figure 12. Snout-vent length at month of first capture for males.
considered to have been residents on the plot since the study began
in July, 1960. July, 1961, was chosen as the cut-off month since
after this month the per cent of new individuals appearing in the
total catch was fairly constant for the remainder of the study
(Fig. 7). Those assigned to the pre-1959 or 1959 groups (as above)
were considered as immigrants. Those assigned to the 1960 group
were examined as to their position in the plot when tagged. Indi-
viduals which were tagged on the inner periphery of the plot were
considered as immigrants, whereas those found well within the plot
were considered as residents which had, by chance, evaded previous
capture. All individuals delegated as immigrants were counted as
part of the population at the month of their first capture.
There are many sources of error in the above methods, especially
in the allocation of individuals to age groups. The chance of error
concerning the placement of immigrants is somewhat less because
there was no evidence that immigration occurred more frequently than
emigration. Immigration and emigration were probably of equal magni-
tude. The results based on the above methods, however, provide the
best estimates of the age structure of the population available from
The age composition of the population from July, 1960, to
March, 1962, is given in Table 12. Of the 357 individuals estimated
to have been alive in August, 1960, 16.8 per cent were pre-1959
Table 12. The age composition of the population on the study
area from July, 1960, to March, 1962, listed by sex.
Pre-1959 1959 1960 1961
Month Male Female Male Female Male Female Unsexed Unsexed Total
* Indicates insertion point of immigrantss.
** Data insufficient to warren inclusion.
hatchlings, 31.9 per cent were 1959 hatchlings, and 51.3 per cent
were 1960 hatchlings. In August, 1961, there were only 277 lizards
estimated to have been alive; 6.9 per cent were pre-1959 hatchlings,
14.8 per cent were 1959 hatchlines, 31.4 per cent were 1960 hatch-
lings, and 46.9 per cent were 1961 hatchlings. In both years the
hatchling group comprised approximately one half of the population.
Also the one year old age group at each August comprised 31 to 32 per
cent of their respective populations. Even though the population
decreased in size, the percentage of corresponding age groups, when
compared between years, remained fairly constant.
During the breeding season at least three and possibly four
year classes contribute to the reproductive pool. In 1961, for
example, lizards hatched in 1960, 1959, 1958, and possibly 1957 were
available for reproductive functions.
The sex ratio within each age group remained constant at one
The turnover, or survival, of different age groups within the
population, and of the original population itself, may be calculated
by dividing the original number of marked lizards into the number re-
covered after a period of time. The quotient, expressed as a percen-
tage, is an estimate of the minimal rate of survival over the period
of time selected. To compute an absolute rate of survival the number
of marked lizards lost from the original population by emigration and
the number of marked individuals inactive at the time of census would
have to be known. Mortality rates may, of course, be obtained by sub-
tracting the figure for rate of survival from 100. Minimal survivor-
ship rates of the population can be determined from the data given in
The overall survival rate of the original population from August,
1960, to August, 1961, was 41.2 per cent. The survival rate of the
population including the 1961 hatchling group was 77.6 per cent. A
survival rate of 100 per cent would indicate a constant population
size from year to year. The rate of survival for the different year
classes between August, 1960, and August, 1961, was 31.7 per cent for
the pre-1959 group, 35.9 per cent for the 1959 group, and 47.5 per cent
for the 1960 group.
If the study had been continued for several more months the
survival rate of the population would probably have been much higher
than 77.6 per cent, since more hatchlings of 1961 would have been dis-
covered (see page 51).
The effect of sex on rate of survival was not consistent within
the different age groups. The rate of survival of females and males
from August, 1960, to August, 1961, was 31.3 per cent and 32.1 per
cent respectively within the pre-1959 group, and 43.9 per cent and 28.1
per cent respectively within the 1959 group. The rate of survival for
each sex in the 1960 hatchling group was determined by dividing the
group of unsexed individuals so that 26 were considered as males and
27 as females. The resultant survival rate for females was 46.1 per
cent; that for males, 48.9 per cent. The rate of survival of all
females (including 27 of the unsexed individuals) from August, 1960,
to August, 1961, was 42.7 per cent. The rate of survival of all males
(including 26 of the unsexed individuals) over the same period of time
was 39.7 per cent.
A graph of the progressive decrease in numbers of the 1960
hatchling group, based on an initial population of 1000, is given in
Figure 13. Only the data for the first 19 months are based on field
observations, the rest of the curve is speculative, being based on
the assumptions that the average life span is 3.8 years (see page 60)
and that the population decreased at a constant rate. A life table
for the 1960 hatchling group based on the data obtained from Figure
13 is given in Table 13. The abbreviations employed in the headings
of Table 13 are defined as follows: x represents age in months, x' re-
presents age as per cent deviation from the mean length of life, dx
represents the number dying in age interval of 1000 hatched, lx repre-
sents the number surviving to beginning of age interval out of 1000
hatched, l000qx represents the mortality rate per 1000 alive at the
beginning of age interval, and ex represents the expectation of future
life in months. For methods of preparing a life table see Dublin, cte
al. (1949). A survivorship curve based on the data in Table 13 is
given in Figure 14.
The survivorship curve of the 1960 hatchling group given in
Figure 14 closely approaches the Type II curve described by Deevey
(1947). A Type II curve was defined by him as being "diagonal (when
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46
AGE IN MONTHS
The progressive decrease in number of 1960 hatchling group. The data for the first 19 months are
based on field observations. The remainder of the curve is speculative, based on an estimated
3.85 year life span and a constantly declining population.
Life table for a hatchling group
on data derived from Figure 13.
months. Headings of columns are
of Lygosoma laterale based
Mean length of life, 13.29
defined in the text.
x x' dx lx l000qx ex
-100 -50 0 +50 +-100 4150 +200 +250
PER CENT DEVIATION FROM MEAN
LENGTH OF LIFE (X')
Survivorship curve of the 1960 hatchling group of
the logarithm of the number of survivors is plotted against age),
implying a constant mortality rate for all age groups, or no one
age as a favorable time of dying." In comparison a Type I curve,
"the negatively skew rectangular, is shown by members of a cohort
which, having been born at the same time, die more or less simul-
taneously after a life span which is presumably characteristic of
the species." A Type III curve, "the positively skew rectangular,
shows extremely heavy mortality beginning early in life, but the
few individuals which survive to advanced ages have a relatively high
expectation of further life." In order to compare survivorship curves
of different species the age should be expressed as percentage devia-
tion from the mean (Deevey).
The survivorship curve of Lygosoma laterale is not typically
diagonal but shows a slight shift towards a Type I curve.
An individual which survives the first year has an expectation
of eight more months of life; one which survives the first two years
has an expectation of six more months of life (Table 13). Thus, if
an individual survives the first breeding season it has a good chance
of partaking in another.
Causes of Mortality
Actual occurrences of mortality are difficult to observe in
any field study since only a fraction of the life span of any one
individual is spent by the collector in the individual's home range.
The small size and secretive behavior of Lygosoma make observations
of this kind even more difficult. Evidence of predation pressure,
however, is more easily detected. In most lizards the tail is
broken when seized by a predator or other organism, a new tail being
regenerated from the stump if the individual escapes. In the ground
,skink a regenerated portion is easily detected due to a difference
in color and scalation between it and the stump. The presence of a
regenerated portion of the tail would thus indicate either unsatisfied
aggression by a predator, or autophagy.
The possibility of some ground skinks feeding on their own
tails (autophagy) or on the tails of other individuals (canabalism)
will be considered first. Carr (1940) reported finding "tails in the
stomachs of 4 specimens," of Lygosoma, "each of which recently lost
its tail," and furthermore, "the stub remaining on the animal plus
the portion in the stomach constituted a tail of the proper facies
and dimensions." lie interpreted these findings as evidence for auto-
phagy. No evidence was found in this study to support such a conclusion.
Over 100 individuals were maintained in the laboratory during the period
of the study, with, in some cases, ten or more in a single terrarium,
and no evidence of one lizard ever breaking its own tail, or that of
another, was ever observed even though numerous fights were noticed.
Also, 381 intestinal tracts of ground skinks taken from varied habitats
and collected over the period of a year were examined, and none contained
any portion of a Lygosoma. Two tracts did contain patches of shed skin
from the limbs.
This leaves the possibility that predators are responsible
for lost portions of tails. On the plot, 90.5 per cent of all speci-
mens with a snout-vent length of 35 mm or more possessed regenerated
tails. Only 22.2 per cent of the individuals below 35 mm had regener-
ated tails. The probably reason for the difference in percentages is
that larger lizards because of their age had been exposed to predators
more often than juveniles. One of two ring-necked snakes, Diadophus
punctatus, collected on the plot contained remains of a Lygosoma tail
in its intestine, but no other body part was found. As further evi-
dence of predation, many specimens, both preserved and on the plot,
were missing toes, and in a few cases an entire limb was missing.
Corollary evidence as to the effect of predation on the numbers
of lizards possessing regenerated tails may be found in a study by
Rand (1954). In a study of an island and a mainland population of
Cnemidophorus lemniscatus he found that the ratio of regenerated tails
on the island population to those from the mainland was 1:8. He at-
tributed this ratio to greater predation pressure on the mainland.
Literature records of predation on Lygosoma laterale are listed
in Table 14. Those species of predators known to occur on or near the
plot are marked by an asterisk. Other species not recorded previously
as predators of Lygosoma are the armadillo, Dasyipus noviemcencyus (W.
Wirtz, personal communication), and the milk snake, Lampropeltis
doliata (T. Brown, personal communication). Animals that are known
to feed on other small lizards and which might occasionally prey on
Lygosoma include bluejays, Cyanocitta cristata, mockingbirds, Mimus
polyglottos, young rat snakes, Elaphe obsoleta quadrivittata, coral
snakes, Micrurus fulvius, moles, Scelopus aquaticus, shrews, Blarina
brevicauda, opossums, Didelphis marsupiolis, and the house cat, Felix
Emigration should also be considered as a form of mortal ity
since it results in the loss of individuals from the population.
During the course of field work on the plot one adult ground
skink was found dead of undeterminable cause under a log, and one
adult and two hatchlings were killed in capture.
There was no evidence to consider parasitism as a cause of death.
Tapeworms, Cyclotaenia americana, were found in 53.2 per cent of all
preserved specimens 35 mm and above in snout-vent length. Nematodes,
consisting of two or three unidentified species, were found in 50.9
per cent, and 27.5 per cent were parasitized by a digenetic fluke,
Brachyocoelium sp. All of the above parasites were found in the in-
testinal tract. Three lizards on the plot were parasitized by unidenti-
fied ticks. In no case did a host seem to have been adversely affected
by the presence of one or more species of parasite. f S A
Table 14. List of predators of Lygosoma lateral recorded
literature. Those known to inhabit the area of
plot are marked by an asterisk.
Hamilton and Pollack (1956
(eastern hognose snake)
(pygmy rattle snake)
(canebrake rattle snake)
(broad headed skink)
(southeastern five-lined skink)
Hamilton and Pollack (1956)
Hamilton and Pollack (1956)
Hamilton and Pollack (1956)
Hamilton and Pollack (1956)
Hamilton and Pollack (1956)
Hamilton and Pollack (1955)
Hamilton and Pollack (1961)
Hamilton and Pollack (1961)
- I IIlII II -- ------ I -~-- I-I---- I
Table 14. Continued.
Cryptotis parva floridana
(Florida short-tailed shrew)
(black widow spider)
The common name of Lygosoma laterale, the ground skink, is very
appropriate since this lizard spends its entire life cycle in associa-
tion with the litter layer and surface soil. When not active, an in-
dividual remains quiescent under a mat of spanish moss, a rotting log,
fallen leaves, or had burrowed into loose soil beneath the litter layer.
The most favorable retreat is under leaf mold at the base of trees and
logs. When active, a ground skink prowls over and under leaves, twigs,
mats of spanish moss, and vines in search of small insects and spiders.
The presence or absence or activity is, of course, determined
by a complex of factors, both internal and external. Only those environ-
mental factors which were easily correlated with activity in the field
will be considered; the effect of environmental changes on the hormonal
regulation of the body is beyond the scope of this study.
Field trips were nonrandom in nature. Most trips were made only
when success was expected. Thus the quantitative results given in com-
paring amount of activity and different factors must be regarded as
Since lizards are ectothermic, environmental temperatures are
important in regulating their activity. To determine the preferred
temperature range for activity, the number of skinks captured at each
recorded field, shade temperature was counted. The results are given
in Figure 15. Figure 16 shows the average number of active skinks
per field trip for arbitrarily selected temperature ranges.
From both figures it is evident that most activity occurred
between 250 and 300 C. Fitch (1956b) found an average cloacal tempera-
ture of 28.80 C for 16 ground skinks in Kansas. In his study the air
temperatures ranged from 22.8* C to 28.50 C with but one exception;
one skink was active at an air temperature of 14.70 C.
The activity of skinks at low shade temperatures was probably
due to higher temperatures in sunlit spots. All skinks active on
days when the shade temperature was 120 C to 180 C were captured in
sunny spots where the air temperature, one inch above the litter
layer, was always 19 C or above. Basking was observed only on sunny
and partly cloudy days during late fall and winter.
The number of active skinks decreased considerably more when
the shade temperature reached above 300 C than when it dropped below
250 C (Fig. 15). This might be explained by the fact that in lizards
the preferred optimum body temperature is closer to the critical
maximum than to the critical minimum (Cowles and Bogert, 1944). Also
at lower shade temperatures an individual could bask at intervals and
thus raise its body temperature to its optimum for activity.
It must be realized that the shade temperature at the point
of highest activity does not represent the ecological optimum tempera-
ture of Lygosoma laterale. Cole (1943) has shown that the body
20- Yi H( T
12 14 16 18 20 22 24 26 28 30 32 34
SHADE TEMPERATURE TWO INCHES ABOVE
GROUND IN DEGREES CENTIGRADE
Figure 15. Number of ground skinks captured at each recorded shade
0- 16- 19-
15 18 21
SHADE TEMPERATURE TWO INCHES
ABOVE GROUND IN DEGREES
Figure 16. Average number of ground skinks captured per' field trip
at selected temperature ranges.
k I I ,I I , I I I I ,
temperature of a lizard is very little affected by that of the sur-
rounding air; the substratum and incident radiation being the more
important temperature determining factors. However, since the average
body temperature found by Fitch (1956b) falls within the range of
shade temperature which contained the greatest amount of activity,
the ecological optimum temperature of Lygosoma lateral probably occurs
within the temperature range of 250 C to 300 C.
At each field trip the moisture content of the litter layer
was recorded in one of two ways: 1) "moist," if the litter layer
contained any moisture as judged by touch, and 2) "dry." if the litter
layer was dry down to the surface soil.
The average number of active lizards per field trip under
"moist" conditions was 21.5 (S.E.m = 1.5); under "dry," 11,1 (S.E.m = 1.5).
The average number active per hour of field trip was 8.2 for "moist" and
6.7 for "dry."
Figure 18 shows the relationship between shade temperature,
moisture content of the litter layer, and the number of ground skinks
active. Between 25* C and 300 C (the area blocked off in the graph)
there were 34 "moist" trips with an average of 25.8 skinks active
(S.E.m = 1.5), and 15 "dry" trips with an average of 15.3 active
skinks (S.E.m = 2.1). Within this temperature range "moist" conditions
were much more favorable for activity than "dry." Whether moisture
affects the individual directly, or indirectly by affecting the
Figure 17. The relationship between shade temperature, moisture,
and number of active ground skinks.
o = DRY
* = MOIST
8 10 12 14 16 18 20 22 24 26 28 30 32 34
SHADE TEMPERATURE IN DEGREES CENTIGRADE
activity of food items, is not known. At temperatures below 250 C
the difference in number of active skinks between "moist" and "dry"
conditions was less. "Moist" conditions, however, still resulted
in more active skinks; an average of 12.3 S.E.m = 2.7) were active
under "moist" trips, 7.2 (S.E.m = 1.6) "dry."
Lygosoma laterale thus seems to prefer warm-moist conditions
over warm-dry, cool-moist over cool-dry, but warm-dry over cool-moist
The data shown in Figure 3 indicated that there is some cor-
relation between activity and rainfall in a few months (October and
December, 1960; February, October, and November, 1961) but the more
obvious correlation is between activity and temperature.
Time of Day
Observations during the course of the study indicated that
throughout the year more skinks were active during the afternoon than
during the morning. Lizards were active though at all times of the
day at all seasons. In general, skinks were active throughout the
day in spring, mostly in the morning and later afternoon in summer,
and usually in the afternoon during fall and winter. Four field
trips were made to the plot at night in 1961; one in April, two in
June, and one in July. No skinks were found active on the surface but
several were uncovered when logs were being moved. These individuals
immediately ran for cover.
The earliest recorded time of activity on the plot was at 810
on July 14, 1961, in the southeastern corner; the latest, at 1820 on
June 15, 1961, in the southwestern corner. Individual ground skinks
have been observed by Dr. J. N. Layne (personal communications) to
be active at 615 in June, 1958, at 615 on July 5, 1962, at 640 on
September 19, 1958, at 700 on April 21, 1959, and before sunup on
May 27, 1958. One was observed active at 2000 on June 28, 1959. All
of these observations were made in his study plot located in a turkey
oak woods. Myers (1959) observed a ground skink in Missouri crawling
in leaves after midnight on June 13, 1954.
The average number of active lizards per field trip for each
of the seasons was as follows: spring, 26.0 (S.E.m = 3.2, N. = 12);
summer, 21.3 (S.E.m = 1.7, N. = 28); fall, 12.8 (S.E.*m = 2.2, N. = 27);
and winter, 16.2 (S.E.m = 2.3, N. = 22). A large number of skinks
active during an exceptional warm spell in February raised the winter
average (Fig. 3). The large number of skinks active during spring
months was probably due to a combination of factors, chief of which
were suitable temperature and moisture conditions and the influence
of the breeding season. That the reproductive urge influenced activity
is indicated by the differential number of males and females active in
winter and spring (Table 16). The low number of skinks active during
the fall was due to unfavorable temperatures. As one example, on
October 21, 1961, the shade temperature was 170 C and no lizards were
active. The next day, October 22, 1961, the shade temperature was 240 C
and 12 lizards were captured. Both trips were recorded as being "moist."
An interesting discovery during this study was the unusual
distribution of individuals within the plot. A large number of skinks
were found associated with areas characterized by a litter layer com-
posed mainly of oak, hickory, and pine leaves. Fewer skinks than ex-
pected were found in other types of litter.
This difference in concentration may be due to one or many of
several reasons. As was mentioned on page 16, the litter layer beneath
pine trees was much more constant in depth throughout the year in com-
parison with that beneath oaks and hickories. The mixture of pine
needles, oak, hickory, and other leaves was also much less compacted
than other litters. Skinks escaped much easier in pine-oak-hickory
litter than in other types. The constancy in depth alone undoubtedly
provided more favorable cover, a more dependent supply of food items,
and a more stable moisture content to individual ground skinks.
A difference in the p1H of the humus might also have influenced
the skink density either by affecting the skinks directly or by af-
fecting the supply of certain food items. Unfortunately no chemical
determinations were made of the humus during the study. Heatwole (1961),
however, found a difference in the pH between surface samples of an
oak-pine-aspen forest and an oak-hickory forest. The pH of the former,
taken at a depth of 2.5 cm was 4.7; that of the latter, taken at the
surface, was 6.0. Both forests were in Michigan.
The distribution of the skinks may also be approached from
another point of view. The oak-hickory-pine litter as opposed to
other types of litter might cautiously be regarded as analogous to
an ecotonal effect. An ecotone is the transition or tension zone
between two major communities (Allee, et al., 1949). In this analogy
the two major communities would be pine litter and oak-hickory litter.
There is a tendency in an ecotone for an increased variety and density
of organisms over that found in the adjoining communities (Odum, 1959).
The reasons behind this effect have been described by Allee, et al.:
"The ecological reality of the ecotone is attested by the fact that,
in addition to organisms penetrating this boundary area from both
communities involved and living therein for all or a regular part of
their lives, there are other organisms that find the biotic and
physical environment of the ecotone more stimulating than the condi-
tions prevailing in either community." Thus in the plot, the mixture
of different leaves possibly resulted in a greater variety and density
of food items being available which, in combination with the increased
cover, resulted in a high density of ground skinks.
It is interesting to speculate on the future of the lizard
population if the woods remain unmolested. If undisturbed the woods
will probably succeed to a mesic hammock or similar vegetation. The
present vegetation is dominated by loblolly pines, various oaks and
pignut hickory. No small pines or pine seedlings were found in the
area yet both Quercus and Carya were represented by seedlings and
saplings. The trend toward a more mesic habitat was indicated by the
presence of seedling magnolias, Magnolia grandiflora, scattered
throughout the plot.
As the pines die and are replaced by other trees will the
population density of skinks rise, fall, or remain stable? A vague
answer can be supplied by comparing the amount and character of the
litter layer between the present and the future woods. Dominant
trees of a mesic hammock are magnolias, pignut hickory, and laurel
oak (C. Monk, personal communication). Decomposition of magnolia
leaves is relatively slow and the litter layer remains deep the
year round (A. M. Laessle, personal communication; personal observa-
tion). Thus perhaps a decrease in density might occur during a transi-
tion period but an increase would probably occur once the magnolias
In connection with the above, a mesic hammock dominated by
oaks and magnolias was visited several times and was found to have
a very high density of ground skinks. On one trip lasting 35 minutes,
12 skinks were captured and approximately that many escaped. Subse-
quent trips to this hammock revealed large numbers of skinks in an
area much smaller than the study plot.
The difference in the average size of home range between males
and females is not unique among lizards. The home ranges of adult
males of Eumeces fasciatus (Fitch, 1954), Uta stansburiana (Tinkle,
et al., 1962), Sceloporus olivaceous (Blair, 1960), and Anolis carolinen-
sis (Gordon, 1956) were found to be larger than those of females. Bar-
wick (1959), however, found that the home range of Leiolopisma
zelandica did not differ in size or shape between adult males and
females, and Fitch (1958) found that females of Cnemidophorus sexlineatus
had a slightly larger home range than males.
The behavior of males in searching for mates is thought by
Blair (1960) to be the basic reason for adult males having larger
ranges than females. Another possibility will be discussed below.
The nonoverlapping of female home ranges indicated female terri-
toriality. This evidence, however, is not conclusive; observation of
actual defense of an area by the female inhabitant is needed for de-
The possibility of females and not males exhibiting territorial-
ity is without precedence among the Scincidae. A thorough search of
the literature failed to reveal any record of territoriality by females
of scincoid species.
Many examples of territoriality by males among iguanid lizards
have been reported (Fitch, 1940, 1956a; Blair, 1960; Evans, 1951;
Schmidt, 1935) and it is among the iguanids that the only cases of
female territoriality have been found. Evans (1938) reported that
both females and males of the Cuban lizard, Anolis sagrei, defend a
territory. The presence of a foreign female within the territory
elicits defense behavior by the resident female. The resident male,
however, prevents any attack and courts the strange female. Rodolfo
Ruibal (personal communication) has observed females of Anolis
sagrei and of A. allisoni showing aggressive behavior.
The biological advantage bestowed upon a female because of a
territorial habit might basically be one of energy conservation.
Dice (1952) believes that "the habit of living in territories results,
... in a minimum amount of fighting among the members of a species
compared to what might take place if each individual roamed about
widely over the habitat." A territorial habit also is a regulatory
mechanism in the community. A territory "prevents the over utiliza-
tion and depletion of food and other sources in the ecosystem, thereby
lessening for the species concerned the danger of recurrent crises
caused by lack of one or more items essential for its existence"
The apparent lack of a territory among males may in part be
responsible for the larger size of male home ranges. An individual
male should require more space than a territorial female since he
shares space with other males and with females.
The above discussion adequately expresses the need for further
study on the spatial relations of individuals within a high density
The position of Lygosoma in the trophic structure of the com-
munity is of importance since it occurred in such large numbers. The
four other species of lizards on the plot were not nearly as abundant
as Lygosoma. Anolis carolinensis was common along the southern edge
of the plot but only a few individuals were noticed elsewhere. The 10
to 15 individuals of Cnemidophorus sexlineatus observed were confined
to the southern border. A few individuals of Eumeces laticeps and E.
inexpectatus were observed in the interior of the plot.
Lygosoma filled a niche in the community as a secondary and
tertiary consumer. A study of the intestinal tracts of 328 skinks
revealed a diet composed of a wide variety of organisms. Insects, be-
longing to 10 orders, were found in 90 per cent of the tracts. Coleop-
terans found in 24 per cent of the guts, dipterans in 20 per cent,
hemipterans in 22 per cent, collembolans in 17 per cent, and lepidop-
teran larvae in 18 per cent comprised the greater portion of the in-
sects eaten. Other types of insects taken were: hymenopternas, orthop-
terans, isopterans, neuropterans, and dermapterans. Spiders were a
large source of food, being found in 48 per cent of the tracts. Other
inhabitants of the litter layer were also eaten: snails were found in
10 per cent of the guts, isopods in 19 per cent, and millipedes,
centipedes, earthworms, and acarinids all in less than four per cent.
Ground skinks in return served as a food supply for many other animals
(see Table 14).
The population ecology of Lygosoma laterale, the ground skink,
was studied from July, 1960, to April, 1962, a period of 22 months.
The study plot, 120 yards by 50 yards in size, was located
in a wooded area on the campus of the University of Florida, Gaines-
ville. A mixture of hardwoods, mostly oaks and hickory, and loblolly
pines dominated the vegetation.
All captures of skinks were made by hand. Lizards were marked
by removing toes; each individual having a different number. A total
of 115 field trips were made during the study. Collections were made
from other populations for study of the reproductive cycle and para-
Within the plot, ground skinks were more abundant in pine-
influenced areas than in nonpine-influenced areas. Areas in which
the litter layer contained a mixture of pine needles and other leaves
were considered as pine influenced. Areas in which the litter con-
tained no pine needles were considered as nonpine-influenced regions.
Litter containing pine needles provided a more favorable habitat due
to the constancy in depth, moisture content, and possibly food supply.
Skinks were found to occupy home ranges throughout their lives.
The average size of a home range for males was more than three times