|
FOX SQUIRREL HOME RANGE AND MAST CROPS IN FLORIDA
by
Angela Torres Kantola
Florida Cooperative Fish and Wildlife Research Unit
117 Newins-Ziegler Hall
University of Florida
Gainesville FL 32611
Final Report
Research Work Order No. 9
July 1, 1986
Supported by:
U.S. Fish and Wildlife Service
Washington, D.C.
through the
Florida Cooperative Fish and Wildlife Research Unit
117 Newins-Ziegler Hall
University of Florida
Gainesville FL 32611
FOX SQUIRREL HOME RANGE AND MAST CROPS IN FLORIDA
By
ANGELA TORRES KANTOLA
A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF MASTER OF SCIENCE
UNIVERSITY OF FLORIDA
1986
ACKNOWLEDGMENTS
This study was funded by the Florida State Museum and the U. S.
Fish and Wildlife Service through the Florida Cooperative Fish and
Wildlife Research Unit, University of Florida.
The advice and guidance provided throughout the study by my
committee was most helpful. Many thanks go to my committee chairman and
principal advisor, Dr. Stephen R. Humphrey, Florida State Museum, for
his valuable support and direction. Committee members Drs. John F.
Eisenberg and Melvin E. Sunquist, Florida State Museum, Dr. Michael W.
Collopy, School of Forest Resources and Conservation and unofficial
committee member Dr. George W. Tanner, School of Forest Resources and
Conservation, gave additional helpful suggestions.
The administrative support of supervisors Drs. Thomas J. O'Shea and
Galen B. Rathbun, U. S. Fish and Wildlife Service, and unofficial
committee member Dr. H. Franklin Percival, Florida Cooperative Fish and
Wildlife Research Unit, was very beneficial and most appreciated.
This project would have been impossible without the efforts of
numerous volunteers. 0. Blodgett, R. Newberry, S. Brand, J. Duffey,
R. Pedlow, A. Venables, K. Whitlock, A. Williams, and others spent many
hours in the field, assisting with squirrel trapping and mast
monitoring.
I give special thanks to my husband, Ed for his constant support and
understanding, and to my parents, Dr. and Mrs. Buenaventura Torres, for
their encouragement throughout my academic career.
TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS..... ............................. ... .............. i
LIST OF TABLES...................................................... v
LIST OF FIGURES....................................................vi
ABSTRACT ...........................................................vii
INTRODUCTION........................................................1
STUDY ANIMAL.........................................................4
STUDY AREA... .................................................... 6
METHODS..............................................................9
Mast Monitoring................. ........................9
Nest Counts......................................................12
Capture and Handling..........................................13
Telemetry and Related Equipment................................15
Statistical Methods............................................ 17
RESULTS... .........................................................19
Mast Production........... .......................................19
Home Range.............................. ..... ................... 32
Nests........................................ ..................46
DISCUSSION..........................................................56
Food Resources.................... ............................. 56
Home Range......................................................59
Nests............................................ .............60
LITERATURE CITED..................................................64
BIOGRAPHICAL SKETCH... .............................................68
LIST OF TABLES
Table Page
1 Summary of analysis of variance on transformed counts of
turkey oaks acorns on the Citrus Tract 20
2 Turkey oak acorn production on the Ordway Slope 21
3 Turkey oak acorn production on six areas of the Ordway
Preserve 23
4 Summary of analysis of variance on transformed counts of
turkey oak acorns on six areas of the Ordway Preserve
and on two of those areas within the Squirrel Study Area
(1983) 24
5 Longleaf pine cone production on five areas of the
Ordway Preserve 26
6 Summary of analysis of variance on transformed counts of
longleaf pine cones on five areas of the Ordway Preserve
and on two of those areas within the Squirrel Study Area
(1983 and 1984) 27
7 April temperatures 1982-1985, Gainesville 29
8 Weights and body measurements of captured fox squirrels 30
9 Dates of radiotracking sessions and number of locations/
squirrel in each 33
10 95% and 65% harmonic mean (HM) home range sizes of
individual fox squirrels 34
11 Eveness of fox squirrel home range use as described by
ratios of 65% to 95% HM home range sizes 42
12 Summary of number of nests used by fox squirrels 47
13 Summary of percent of fox squirrel locations in a nest 50
LIST OF FIGURES
Figure Page
1 Distribution of fox squirrels in the southeastern 2
United States
2 Location of the squirrel study area, Ordway Preserve, 8
Putnam County, FL
3 Location of the turkey oak study sites on the Citrus
Tract, Withlacoochee State Forest, Citrus Co., FL 10
4 1983 and 1984 longleaf pine cone and turkey oak acorn
production 25
5 Tracking session dates and seasonal events of
importance to fox squirrels 31
6 Harmonic mean home ranges: tracking session #1 35
7 Harmonic mean home ranges: tracking session #2 36
8 Harmonic mean home ranges: tracking session #3 37
9 Harmonic mean home ranges: tracking session #4 38
10 Harmonic mean home ranges: tracking session #5 39
11 Harmonic mean home ranges: tracking session #6 40
12 Harmonic mean home ranges: tracking sessions 4-6 43
13 Three-dimensional plot of all squirrels' use of the
study area 44
14 Three-dimensional plots of of squirrel home range use
in tracking sessions 4-6 45
15 Air temperature and percent of squirrel locations
in a nest by tracking session 51
vi
16 Mean percent of all squirrel locations in a nest by
tracking period and session 52
17 Frequency and diameter of turkey oaks containing nests
compared with randomly selected turkey oaks in ecotone
and upland 55
Abstract of a Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
FOX SQUIRREL HOME RANGE AND MAST CROPS IN FLORIDA
By
Angela Torres Kantola
August 1986
Chairman: Dr. Stephen R. Humphrey
Major Department: Forest Resources and Conservation
Home ranges, nest use, and primary food resources of Sherman's fox
squirrel (Sciurus niger shermani) were studied on the University of
Florida's Katharine Ordway Research Preserve in Putnam County, Florida.
Sherman's fox squirrel is a distinctive subpopulation with a small
geographic range which is declining in Florida due to habitat loss.
To understand production of its primary food resources, seed
production of longleaf pines (Pinus palustris) and turkey oaks (Quercus
laevis) was monitored for two years on contrasting sites of high
(ecotone) and low (upland) productivity. Mast production varied by
site, tree, year, and tree size. Production was higher in the ecotone,
significantly so in turkey oaks. In the second year, the turkey oak
acorn crop failed entirely.
Patchiness of these primary food resources likely explains the very
large home range size of this animal. Six fox squirrels were
radio-collared and tracked for varying lengths of time over a one-year
viii
period. Home range size was significantly greater in males than in
females. Home range size, position, and intensity of use were probably
affected not only by sex, but also by resource abundance, reproductive
season, and weather factors.
Squirrels used a calculated average of 30 nests/year, mostly
"leaf-nests" as opposed to tree cavities. More time was spent in the
nest during cold and very wet weather, indicating decreased activity
during these times.
2
Squirrel density was estimated to be roughly 12/km Nest counts
appeared to provide a reliable estimate of this density on the squirrel
study area and are recommended for testing in other areas of the
squirrels' range.
Habitat management is of primary importance to the survival of
Sherman's fox squirrel. Preservation and maintenance of natural, mature
longleaf pine forests is needed for successful management.
Chairman
INTRODUCTION
Sherman's fox squirrel, (Sciurus niger shermani) inhabits the
longleaf pine/turkey oak sandhills of the northern 2/3 of Florida's
peninsula and is one of three subspecies of fox squirrel in Florida
(Moore 1956; Fig. 1). The two others are S. n. niger, ranging from the
panhandle to southern Virginia (Hall 1981), and the endangered big
cypress fox squirrel, S. n. avicennia, isolated in southwestern Florida
(Williams and Humphrey 1979). S. n. shermani and S. n. niger are part
of a unique group of fox squirrels of the southeastern coastal plain
(Weigl et al., in press). The southeastern fox squirrels are the
largest and most variably colored tree squirrels in the western
hemisphere (Nowak and Paradiso 1983).
Southeastern fox squirrels have long been recognized as primarily
associated with mature, fire-maintained pine forests (Allen 1871,
Maynard 1872, Chapman 1894, Cory 1896). Now, however, large-scale
forest cutting, conversion to single-stand, short-rotation forestry, and
other land uses have caused a decline in the populations of these
squirrels. While habitat loss is the major reason for decline, Harper
(1927), Moore (1953 and 1954), and J. Reinman (pers. comm.) have
suggested that hunting also may be detrimental to local populations
under certain conditions.
Sherman's fox squirrel is currently under review by the U.S. Fish
and Wildlife Service for listing under the Endangered Species Act as
1
S. N. AVICENNIAi
-a -i
Distribution of fox squirrels in the southeastern
United States. Relic populations and midwestern
subspecies within the range of the map are not shown.
Figure 1.
3
threatened or endangered. It has been placed in Category Two: species
known to be subject to some threat of extinction but for which more data
are needed before a status designation can be made. Based on the trend
of habitat loss, the state of Florida has listed Sherman's fox squirrel
as threatened. However, hunting is still permitted in recognition that
the practice is traditional and not the primary cause of threat.
Wise management of this threatened squirrel requires a more
thorough understanding of the animal's habitat requirements. Toward
this end, the objectives of this study were to determine fox squirrel
home range size, habitat, and nest use, and to monitor mast production
of primary food resources on contrasting sites of high and low
productivity.
STUDY ANIMAL
The only technical literature on Sherman's fox squirrel is a
natural history study by Moore (1957). (More recent studies in North
Carolina (Weigl et al., in press) and Georgia (Hilliard 1979) have
addressed the ecology of S. n. niger.) Contrary to the predictions of
Bergmann's rule, Sherman's is the largest fox squirrel in North America.
Adults weigh just over 1 kg and average 650 mm in length. Pelage color
varies from all-black to all-tan color morphs; 1/16 are all black, 6/16
evenly divided between black over tan and tan over black, and 9/16 are
all tan.
Moore (1957) describes these squirrels as somewhat territorial
with a fairly large home range size. In 1946 he counted eight adult fox
squirrels on his 21-ha study area (described as optimum habitat).
However, he thought this density (38 squirrels/km2) was atypical, and
likely to be much lower in other, unprotected parts of northern Florida.
Investigators on the Katherine Ordway Research Preserve, Putnam Co., FL
have estimated current fox squirrel densities there to be only
8.4-15/km2 (Humphrey et al. 1985, M. Sunquist, pers. comm.).
Principal predators appear to be the bald eagle, red-tailed hawk,
and man (Moore 1957). However, Sherman's fox squirrel exhibits several
behaviors that may help it avoid predation. Individuals apparently
maintain a large number of nests which may provide protection from
predation. Moore (1957) reported several instances of pursued fox
4
5
squirrels taking refuge below ground in gopher tortoise (Gopherus
polphemus) burrows. They also will "hide" in the plume of needles on an
apical twig of a longleaf pine in such a way that the twig bends over
horizontally beneath them and conceals them from predators below. When
pursued, they may leap from tree to tree or jump to the ground from
heights up to 15 m, and escape on the ground where they have been timed
at speeds of 24 kph (Moore 1957).
Longleaf pine (Pinus palustris) seeds and turkey oak (Quercus
laevis) acorns are the squirrel's primary food resources. Fungi,
fruits, insects, and staminate cones may be eaten when pine cones and
acorns are unavailable. Two breeding seasons occur each year and
natality is relatively low (2.3 young per litter) compared to 2.5-3.2
for the much-studied midwestern fox squirrel, S. n. rufiventer. The
lower reproductive rates of S. n. shermani probably are related to the
nutrient-poor Florida sands, which provide less energy to consumers than
the richer soils of the midwest. Several authors have found squirrel
reproductive rates to vary with mast production (Chesemore 1975, Havera
and Nixon 1980, Weigl et al, in press).
STUDY AREA
The study was conducted on the University of Florida's Ordway
Preserve in Putnam County. Roughly 1/3 of this 37-km2 preserve is
longleaf pine/turkey oak sandhill, the typical habitat of Sherman's fox
squirrel. Historically, the longleaf pines of the area have been tapped
for turpentine, subjected to annual burning, and removed by logging.
This reduction in pine density has led to an increase in turkey oak
density. Current densities are reported at 59 pines and 175 oaks >3 cm
diameter-at-breast-height (dbh) per hectare (Humphrey et al. 1985).
The sandhill habitat grades from "upland" downslope into "ecotone".
Upland occurs on dry hilltops of low productivity. Here longleaf pines
and turkey oaks create a fairly open canopy over a moderate amount of
herbaceous vegetation dominated by pineland-threeawn (Aristida stricta).
Similar but denser and more varied vegetation occurs on the lower, more
productive slopes of ecotone, where the predominant longleaf pines and
turkey oaks are interspersed with sand post oak (Quercus stellata var.
margaretta), live oak (Q. virginiana), laurel oak (Q. laurifolia), and
bluejack oak (Q. incana).
The squirrel study area was a sandhill centered in the Ford and AQ
(Anderson Cue) Areas of the Ordway Preserve (Fig. 2). The central
upland graded into ecotone on much of the perimeter. Numbered grid
stakes were placed at 100-m intervals over a 91 ha to facilitate mapping
of animal locations.
CH
0 )
O
V Er
o iU
CL
0 to
01 (
< 0 4-
M U *-*
0
00 0
o*i
U tl )
4^ 3
0 T
-z
0-
- :"
n %
cc
-- ,/ -
/* o---- <
OIJ') ,,
,., "
METHODS
Mast Monitoring
Turkey Oaks
Three data sets on mast production by turkey oaks were analyzed.
First, to understand the spatial, temporal, and tree-to-tree variation
in acorn numbers, I analyzed data collected between 1962 and 1973 by
S. L. Beckwith and R. W. Umber on the "Citrus Tract" in the
Withlacoochee State Forest near Inverness, Florida (Fig. 3). In the
fall of each year they made a complete count of all turkey oak acorns on
the trees and on the ground below the canopy. Trees were sampled on two
260-ha blocks each containing three 8-ha replicates divided into 2-ha
quadrants. Each quadrant ideally contained four sample trees chosen by
the point-center quarter method. Due to low tree density and natural
mortality, not all quadrants contained four trees throughout the study
(Umber 1975). Acorn production was not monitored in 1971 or 1972 and
was only monitored on Block 2 in 1962. In 1963 Block 1 contained 38
sample trees and Block 2 contained 45. By 1973, 33 of the sample trees
remained in Block 1 and 41 remained in Block 2. The results of this
analysis were used to choose sampling procedures and test analytical
methods for the Ordway Preserve studies of turkey oaks.
METHODS
Mast Monitoring
Turkey Oaks
Three data sets on mast production by turkey oaks were analyzed.
First, to understand the spatial, temporal, and tree-to-tree variation
in acorn numbers, I analyzed data collected between 1962 and 1973 by
S. L. Beckwith and R. W. Umber on the "Citrus Tract" in the
Withlacoochee State Forest near Inverness, Florida (Fig. 3). In the
fall of each year they made a complete count of all turkey oak acorns on
the trees and on the ground below the canopy. Trees were sampled on two
260-ha blocks each containing three 8-ha replicates divided into 2-ha
quadrants. Each quadrant ideally contained four sample trees chosen by
the point-center quarter method. Due to low tree density and natural
mortality, not all quadrants contained four trees throughout the study
(Umber 1975). Acorn production was not monitored in 1971 or 1972 and
was only monitored on Block 2 in 1962. In 1963 Block 1 contained 38
sample trees and Block 2 contained 45. By 1973, 33 of the sample trees
remained in Block 1 and 41 remained in Block 2. The results of this
analysis were used to choose sampling procedures and test analytical
methods for the Ordway Preserve studies of turkey oaks.
0 2 4
I II
KM
Figure 3.
INVERNESS
*
Location of turkey oak study sites on the Citrus Tract,
Withlacoochee State Forest, Citrus Co., FL.
To learn whether mast production varied with elevation on the
Ordway Preserve, in the fall of 1983 I counted acorns on 33 trees on a
gradient running from the western hilltop down to the margin of Anderson
Cue Lake. Eleven sets of three trees each were selected at 25-m
intervals and all acorns on these trees and on the ground below them
were counted.
Finally, to monitor mast production on and near the radiotracking
site through the study, I counted acorns in the fall of 1983 and 1984 on
six areas (three ecotone and three upland) within the Ordway Preserve.
Each area contained three, 10-tree replicates, for a total of 180 trees.
The upland areas were designated ECUE, having replicates in the AQ and
Fox Pond Areas, (refer to Fig. 2), WCUE in the AQ and Suggs Areas, and
STUDYU in the AQ Area. The ecotone areas were STUDYE in the Ford Area,
SWCOR in the Rowan Southside Area, and SMITH in the Smith Lake Area.
Acorns were counted in late October and early November of 1983
and 1984 after most had fallen from the tree. Where production was
judged to be moderate to heavy, an area estimated to contain one-quarter
of the groundfall was counted, multiplied by four, and added to any
acorns seen on the tree (using 7x binoculars) to arrive at the total
count. Since this method did not account for acorns removed by animals
or distinguish rotten (infested) or immature acorns, it was intended
only as an index of production over each area. Such an index cannot
accurately describe available energy in terms of total kg of acorns, but
it is adequate for comparison of production among areas.
Longleaf Pines
Longleaf pine cone production was monitored on five areas (two
ecotone and three upland). Pine cone production was twice as variable
12
as acorn production, so each area contained three, 20-tree replicates
for a total of 300 trees monitored. Replicates were in the same general
areas as those of turkey oaks, except in the SMITH area where no pines
were monitored. Pine cones were counted in the winters of 1983 and 1984
after most had fallen to the ground or been cut by squirrels leaving the
cores beneath the tree. Counts made using 7x binoculars in late summer
when most cones remained on the tree proved too inaccurate for use.
To obtain seed counts, pine cones were removed from selected trees
in each area using a tree pruner or a semi-automatic .22-caliber rifle
with telescopic sight. Hollow-point bullets were found to be more
effective in obtaining cones than other bullets. The number of rounds
required to remove each cone depended on position of the cone on the
tree, shooter accuracy, and shooter fatigue, but 8 rounds per cone was
typical.
Nest Counts
Squirrel nests were counted on ten 1-ha plots each in ecotone and
upland portions of the squirrel study area. Counts were made when nests
were most visible in December 1984 and January 1985, after most of the
leaves had fallen from the deciduous oaks. The length of each hectare
was walked in strips 10-20 m wide (depending on tree density) until the
entire area had been checked. At each nest, tree species, dbh, and nest
composition (twigs, leaves, and/or Spanish moss) were recorded. Tree
cavities also were noted but not used in analysis.
Capture and Handling
Southeastern fox squirrels are much more difficult to trap than
midwestern subspecies (Weigl et al., in press). Squirrels were captured
in No. 104 Tomahawk double-door cage live traps. Animals were
particularly difficult to trap when their food resources were abundant.
Trapping effort was nearly continual but most intense in seasons of low
food abundance.
Groups of 2-20 traps were placed throughout the study area at sites
where squirrels were seen repeatedly. Usually, traps were positioned
beneath trees that contained squirrel nests or appeared to be used by
squirrels for feeding or caching. During warm weather, traps were
covered with Spanish moss (Tillandsia usneoides) to provide shade for
trapped animals. Traps were normally set shortly after sunrise, checked
periodically throughout the day, and closed before dusk. Occasionally,
traps were left open overnight.
Traps were baited with pecans. Peanut butter was later added as a
base to make removal of the pecans difficult and successful trapping
more likely. Late in the study, instant-bonding glue was used for this
purpose, but its success could not be determined since food was abundant
at.this time and only one squirrel was caught. A liquid attractant,
"Squirrel Scent" (Buck Stop Lure Co., Stanton, Mich.), was placed on and
around traps in an effort to increase trapping success. Before trapping
began, traps were prebaited (baited and wired open) for several weeks to
habituate the squirrels to obtaining food from them. The prebaiting
effect was continued throughout the study by wiring open baited traps at
the end of each trapping session.
14
Usually a captured squirrel was transported to an enclosed room in
a nearby pole barn for radio-collaring. The animal was transferred from
the trap to a wooden drugging box and restricted to one end. Vapor from
several drops of Fluothane (Fort Dodge Laboratories Inc., Fort Dodge,
Iowa) on cotton was administered through the top of the box and the
animal was observed through a plexiglass panel. Within a few minutes,
the squirrel could safely be removed with gloved hands and administered
10-15 mg Ketaset (Veterinary Products, Bristol Laboratories, Syracuse,
N.Y.) to further anesthetize it. Booster doses of 5-10 mg Ketaset and
Fluothane vapor were administered as needed. Physical restraint also
was required since the anesthesia did not fully immobilize the animal.
Standard body measurements were made, as well as notes regarding
reproductive and physical condition, wounds, ectoparasites, and pelage
color. The animal was ear-tagged (#1, style 4-1005, National Band and
Tag Co., Newport, Ky.) and then fitted with either a radio or
identifying collar. Radio collars were closed with nuts and bolts
coated with locking compound. Solar radio collar closures were secured
with shrink tubing heated with a soldering iron held 1 cm from the
collar. Radio collars also were wrapped with colored vinyl tape to aid
in visual identification. Two squirrels not radio-collared were fitted
with number 13 ball-chain collars with identifying colored beads.
Finally, wounds (typically nose wounds from the animal "fighting"
the trap) were medicated, and the squirrel was weighed and returned to
the drugging box for holding. Recovery took 30-60 min, after which the
squirrel was transported back to the capture site and released. Animals
captured near dusk usually were held overnight and released the
following morning.
15
Methods of capture and handling were successfully tested on an *
animal living distant from the squirrel study area in March 1984.
Subsequently, five males from the squirrel study area were collared
between 28 March and 3 June 1984. Radio collars were placed on four of
these squirrels (M1-M4). The fifth received an identifying bead collar
but was never re-sighted. M1 and M2 were tracked throughout the study.
Ml's collar came off after 38 days and was replaced four days later.
Both Ml and M2 received solar radio collars on 2 December. M3 was
tracked through 27 October after which his second collar presumably
slipped off (the collar was found still functioning and fastened in a
loop). M3 was never positively identified again and efforts to re-trap
him were unsuccessful. M4 was tracked through 8 September and was found
badly decomposed with collar intact and radio functioning on 14
September. Two females (Fl and F2) were captured in November, after
extensive trapping efforts. Both were tracked through the end of the
study in May 1985. Only three other fox squirrels (all males) were
caught during the study. One apparently died from heat while in the
trap, and a second from unknown causes while under anesthesia. A third
squirrel (male) was caught during an attempt to re-trap Fl and F2 after
the completion of the study.
Telemetry and Related Equipment
Equipment
Two types of radio collars were used. The first was a 2-stage RF
whip antenna transmitter (Cedar Creek Bioelectronics, Bethel, Minn.)
weighing 36 g with a 2/3 A 3 volt lithium battery having an estimated
16
life of 6 months. Eight of these collars were deployed. Three remained
on the animal longer than 6 months and continued to transmit signals.
The second type of radio collar used was a SM-1 solar and nickel-cadmium
(NiCad) battery powered transmitter (AVM Instruments Co., Livermore,
Calif.) weighing 14 g and having an estimated working life of 5 years
with a 60 mA/h 1.2 volt NiCad battery and five 5-mm silicon solar cells.
Two of these collars were placed on animals in December 1984 and were
functioning well at the end of the formal study period in May 1985. The
animals were tracked occasionally until November 1985, after which they
could not be relocated.
The tracking system consisted of a 36-band AVM LA-12 receiver tuned
to 164 MHz, a hand-held RA-2AD yagi antenna (Telonics Inc., Mesa Ariz.)
and lightweight headphones. A vehicle-mounted RA-5A whip antenna also
was used.
Telemetry
The rolling topography of the study area and the need for precise
animal locations rendered triangulation inappropriate for determining
squirrel locations. Instead, a signal was followed until the squirrel
was located visually or precisely by signal strength and direction.
Exact location was recorded by measuring azimuth and estimating distance
of the animal from the nearest numbered grid stake. Locations were
later plotted to the nearest 25 m2 and assigned x and y coordinates.
Time, position (in a tree, nest, or on the ground), and any activity
also were recorded at each location. Nest trees used by each squirrel
were marked and numbered. Animals apparently observed the researcher
and altered their behavior before they could be seen, so behavioral
notes were rare.
17
Squirrels were monitored regularly from June 1984 through May 1985.
Radiotelemetry was divided into six tracking sessions each consisting of
four 2-day (usually consecutive) periods. The four periods in each
session usually were one week apart and spanned a total of 4-6 weeks.
Tracking sessions were separated by 4 weeks of no tracking. To ensure
independence of observations and allow the squirrels to resume normal
behavior after the researcher's intrusions, animals were located once
every 2-4 hours, depending on day length. For each day of tracking,
daylight was divided into three equal periods, and squirrels were
located once each period. To thoroughly sample the day and help
document nest use, squirrels were located early in each period the first
day of tracking and late in each period the second day.
Statistical Methods
Comparisons of mast production data between areas, replicates,
trees, tree size, and year were made using analysis of variance (PROC
GLM of the Statistical Analysis System, Ray 1982). Since many trees
produced no mast, data were transformed by adding one to the count and
computing the natural log.
Home range size was calculated using an IBM microcomputer and the
McPAAL program for the analysis of animal locations (Smithsonian
Institution). Three measures were calculated: 1) the minimum area
(Mohr 1947), also called the minimum convex polygon (MCP), 2) the 95%
probability ellipse (Jennrich and Turner 1969), and 3) the harmonic mean
(HM) (Dixon and Chapman 1980). The first two measures are presented to
18
make the results comparable with other studies. Discussion will focus
primarily on the HM.
Based upon the harmonic mean of an areal distribution, the HM plots
contours encompassing areas of equal activity, or frequency of
occurrence, of an animal within its home range. The HM provides insight
into the intensity of use of the home range, is not as sensitive to
sample size as the MCP, and (unlike the 95% probability ellipse) does
not assume a bivariate normal distribution of animal locations. The
contour encompassing 95% of an animal's locations (95% HM) is here
considered to exclude unusual movements and to represent whole home
range, while the 65% HM approximates a core area of use within the
range.
Another important aspect of home range is the intensity, or eveness
of use. Eveness can be represented by the ratio of the 65% HM to the
95% HM. As this ratio approaches one, the squirrel's eveness of home
range use increases. A three-dimensional representation of squirrel
home range use was produced using PROC G3D in SAS/GRAPH of the
Statistical Analysis System (SAS Institute 1981).
Home ranges were analyzed by both individual and combined
radiotracking sessions. Data from incomplete tracking sessions (Session
2 for M4 and Session 3 for M3 and Fl) were not used in calculating home
range size or eveness of use. T-tests (PROC TTEST of the Statistical
Analysis System, Ray 1982) were used to compare home range size and
eveness of use between the sexes as well as nest counts between ecotone
and upland. The coefficient of determination (Sokal and Rohlf 1981) was
used to measure the correlation between air temperature and percent of
locations squirrels were in a nest.
RESULTS
Mast Production
Turkey Oaks
On the Citrus Tract, the number of acorns produced per tree varied
greatly over both time and location. Umber (1975) reported a high of
49.1 mean acorns/tree in 1965 and a low of 1.0 in 1969. Block 2
consistently outproduced Block 1 with four years of good mast yields to
only one in Block 1, but no biological explanation was offered. The two
blocks were approximately 1.6 km apart, and although different burning
schedules may have been used, no record of burning was kept. Further
analysis by year revealed that acorn production varied between blocks in
all but 1966 and 1967, among replicates in 1963 and 1966, and by dbh
only in 1968. Also by year, no block x replicate or quadrant (nested in
block x replicate) effects were shown. Combined-year analysis (Table 1)
showed variation among years and also among quadrants (nested in block x
replicate). Because the quadrant effect was inconsistent among
replicates and blocks, these did not have significant effects when
quadrants were used as an error term. The effect of tree size (dbh) on
acorn production was insignificant.
On the Ordway Preserve slope (Table 2), turkey oak production
varied between treatments (top and slope) (P < 0.05) and by dbh
(P < 0.01). Turkey oaks on the hilltop produced a mean of 19.9 acorns,
19
RESULTS
Mast Production
Turkey Oaks
On the Citrus Tract, the number of acorns produced per tree varied
greatly over both time and location. Umber (1975) reported a high of
49.1 mean acorns/tree in 1965 and a low of 1.0 in 1969. Block 2
consistently outproduced Block 1 with four years of good mast yields to
only one in Block 1, but no biological explanation was offered. The two
blocks were approximately 1.6 km apart, and although different burning
schedules may have been used, no record of burning was kept. Further
analysis by year revealed that acorn production varied between blocks in
all but 1966 and 1967, among replicates in 1963 and 1966, and by dbh
only in 1968. Also by year, no block x replicate or quadrant (nested in
block x replicate) effects were shown. Combined-year analysis (Table 1)
showed variation among years and also among quadrants (nested in block x
replicate). Because the quadrant effect was inconsistent among
replicates and blocks, these did not have significant effects when
quadrants were used as an error term. The effect of tree size (dbh) on
acorn production was insignificant.
On the Ordway Preserve slope (Table 2), turkey oak production
varied between treatments (top and slope) (P < 0.05) and by dbh
(P < 0.01). Turkey oaks on the hilltop produced a mean of 19.9 acorns,
19
Table 1. Summary of analysis of variance on transformed
counts of turkey oak
Citrus Co., FL.
acorns on the Citrus Tract,
Source Of SS F PR > F
Dbh
Block (B)
Replicate (R)
B xR
Quadrant (Q) [B x R]
Tree [B x R x Q]
Year (Y)
B xY
RxY
BxRxY
Q x Y [B x R]
Tests of hypotheses using Type
error term:
III MS for Q [B x R] as an
1.81
2.82
B xR
0.1949
0.0863
Tests of hypotheses using Type III MS for Q
an error term:
B x Y
BxY
RxY
B x Rx Y
x Y [B x R] as
47.19
19.85
1.51
1.40
0.0001
0.0001
0.0918
0.1476
0.20
9.88
30.69
3.77
98.10
322.52
332.12
124.15
21.32
17.54
118.08
0.25
12.11
18.82
2.31
6.68
6.59
45.26
19.03
1.45
1.34
0.96
0.6191
0.0005
0.0001
0.1000
0.0001
0.0001
0.0001
0.0001
0.1026
0.1654
0.6148
Table 2. Turkey oak acorn production on the
Ordway slope.
x Acorns/Tree
46.00
31.00
6.00
14.33
2.33
39.67
66.67
16.00
22.33
19.67
19.33
Treatment
Top
Top
Top
Top
Top
Slope
Slope
Slope
Slope
Slope
Slope
Set
1
2
3
4
5
6
7
8
9
10
11
22
compared with 30.6 on the slope. This difference may be correlated with
a greater availability of soil nutrients and water at lower elevations,
but these factors were not measured.
The 180 turkey oaks of the six sampled areas on the Ordway Preserve
yielded a good crop in 1983 (i/tree = 58.5) but almost no acorns were
produced in 1984 (T = 0.18) (Tables 3 and 4, Fig. 4). In 1983
production varied by dbh, among areas, and replicates within area. A
contrast of upland versus ecotone confirmed a variation in production
between these habitat types.
A separate analysis of the two areas on the squirrel study area
showed that acorn production did not differ between areas when
replicates within area were used as an error term. Without the error
term, significant variation was shown. This difference occurs because
the replicates did not vary consistently between areas.
Longleaf Pines
Longleaf pine cone production also was higher in 1983 (x = 27.3)
than in 1984 (7 = 14.0) (Tables 5 and 6, Fig. 4). Analysis of
combined-year data revealed that cone production varied significantly
between years and by dbh. Nested and interaction effects differed for
replicates within areas, trees within areas x replicates, and replicates
x years within areas. Using replicates within areas as an error term,
variation in cone production did not differ among areas. Without the
error term an area effect was shown, so replicates did not vary
consistently among areas. Likewise, areas x year did not differ when
replicates x year within areas were used as an error term. Production
did vary among areas x year without this error term, so replicates did
not vary consistently among areas x year.
Table 3. Turkey oak acorn production on six areas of the Ordway
Preserve.
1983
Acorns/Tree
12.57
7.93
67.47
29.32
120.87
104.90
37.03
87.60
1984
Acorns/Tree
0.25
-0-
-0-
0.08
-0-
0.70
0.67
0.26
Total
Acorns/Tree/Year
6.62
4.33
34.90
15.28
61.46
52.80
18.55
44.27
Treatment
Upland
Upland
Upland
7 Upland
Ecotone
Ecotone
Ecotone
T Ecotone
Area
ECUE
WCUE
STUDY
STUDY
SWCOR
SMITH
--
Table 4. Summary of analysis of variance on transformed
counts of turkey oak acorns on six areas of the
Ordway Preserve and on two of those areas
within the squirrel study area (1983).
Source Of SS F PR > F
All Areas
Dbh
Area (A)
Replicate (R) [A]
41.02
146.10
55.66
33.18
23.64
3.75
0.0001
0.0001
0.0001
Tests of hypotheses using
error term:
A
Contrast:
Upland vs. ecotone 1
Type III MS for R [A] as an
60.83
6.30
13.11
0.0043
0.0035
Squirrel Study Site
14.63
4.63
28.61
R [A]
12.57
3.98
6.14
0.0008
0.0512
0.0004
Test of hypothesis using Type III MS for R [A] as an
error term:
A 0.65 0.4661
140-
120-
100-
80-
60-
40-
20-4
83 84 83 84 83 84
SMITH SWCOR STUDY
-- ECOTONE
~d1
rJL
83 84 83 84 83 84
STUDY ECUE WCUE
UPLAND
REPLICATE PLOT WITHIN AREA BY YEAR
Figure 4.
1983 and 1984 longleaf pine cone and turkey oak acorn
production. In 1984, no acorns were found on the WCUE,
STUDYU, or STUDYE areas and less than one acorn/tree was
found on the ECUE, SWCOR, and SMITH areas. No samples
of longleaf cones were taken at SMITH.
OA
d
Table 5. Longleaf pine cone production on five areas of the Ordway
Preserve.
1983
Cones/Tree
20.85
24.82
27.98
24.55
1984
Cones/Tree
14.37
13.73
13.93
14.01
Total
Cones/Tree/Year
17.61
19.28
21.02
19.30
Ecotone
Ecotone
7 Ecotone
Treatment
Upland
Upland
Upland
7 Upland
Area
ECUE
WCUE
STUDY
STUDY
SWCOR
44.65
17.98
31.32
17.17
11.03
14.10
31.03
14.51
22.77
---
--
--
Table 6. Summary of analysis of variance on transformed
counts of longleaf pine cones on five areas of
the Ordway Preserve and on two of those areas
within the squirrel study area (1983 and 1984).
Source Df SS F PR > F
All areas
Dbh 1 0.39 0.64 0.4249
Area (A) 4 7.64 3.09 0.0164
Replicate (R) [A] 10 45.76 7.40 0.0001
Tree [A x R] 286 614.58 3.48 0.0001
Year (Y) 1 41.93 67.84 0.0001
A x Y 4 9.09 3.68 0.0062
R x Y [A] 10 19.33 3.13 0.0008
Test of hypothesis using Type III MS for R [A] as an
error term:
A 0.42 0.7926
Tests of hypotheses using Type III MS for R x Y [A] as an
error term:
Y 21.69 0.0009
A x Y 1.18 0.3783
Squirrel Study Site
Dbh 1
A 1 0.04 0.07 0.7985
R [A] 4 24.03 10.38 0.0001
Tree [A x R] 113 297.55 4.55 0.0001
Y 1 33.52 57.94 0.0001
A x Y 1 1.57 2.71 0.1026
R x Y [A] 4 6.45 2.79 0.0298
Test of hypothesis using Type III MS for R [A] as an
error term:
A 0.01 0.9405
Tests of hypotheses using Type III MS for R x Y [A] as an
error term:
Y 20.78 0.0103
A x Y 0.97 0.3800
28
A separate analysis of the two sampling areas on the squirrel study
area showed no variation in cone production between areas, or areas x
year. Otherwise, variation within the squirrel study area reflected
that of all areas combined.
Due to difficulties experienced with storing and opening collected
pine cones, few data were collected from these. However, most of the
cones that were examined contained <50% viable seeds.
Seed Production and Temperature
Both longleaf pine and turkey oak seeds require two growing seasons
to mature. Seed production in pines may be adversely affected by
unusually low temperatures or drought during flower development and
immediately prior to pollination (in early spring) (Schopmeyer 1974,
Olson and Boyce 1971). Table 7 summarizes April temperatures for the
area from 1982-1985. Overall, temperatures were lower in April 1983
than any of the other years, and may have contributed significantly to
the turkey oak acorn failure and lower pine cone production in 1984.
Relationship of Squirrels to Food Resources
The capture dates, weights, and body measurements listed in Table 8
provide insight to the squirrels' relationships with their primary food
resources. Of squirrels caught more than once, Ml was at his lowest
weight in early June, at the end of both the summer breeding season
(during which he was sexually mature and active) and the late
spring/early summer period of low food abundance (Fig. 5). M3 was at
his greatest weight in early October, towards the end of the pine
cone-cutting season, but prior to the turkey oak acorn season, which
failed in 1984. Both M1 and M2 had enlarged testes and were seen in
mating chases during the 1984 summer breeding season. However,
Table 7. April temperatures ( C), 1982-1985, Gainesville.
x Monthly Departure
Temp. From Normal
21 -0.1
18.1 -2.79
19.8 -1.08
20.8 -0.09
7 Min.
Temp.
14.5
10.4
12.4
13.6
Monthly Low # of Days With
and Date Min. Temp <7 C
6.1 12 2
4.4 17 11
5.6 6 4
2.8 3 5
Year
1982
1983
1984
1985
t In CL C I o I C
-4 -1 -4 C) --4 -4
I C\J Y)
-4 -4
E
E
-,
c,
,0
a,)
I- E
E
-c
C
a)
cr
_J
S
-4-,
CD
a)
--
4-,
cu
a)
0 LO O o
E EEE E
LC o o t0
Ci Ci CJ o
E E E E
-4 .-- CNJ CM
S:SS S :S
t0C
,- -l
0 0 0
- 0 0 r
O, .-
EE E 4- 4- E
: S LL- u-
Co o cO coo Co coo o o C0o0 Co co co co
S-L LC U SU C S S-.4- > > -
S(v (V QC Q 3 Q(a. 3 0 0 3
:a- COC0 m 0C'j 10 CMJ .--I C CM0 < '> 0
CN CONCO C)C) CMN C CNjO C) CD CM -
1O r-n COO C 0 C) o C
LO O C) rm O O 101 O Cj
C\1 m I Io l o u ) r" i O TO C) -4
LO 10) (.0 1.0 t.0 (D0 '.0 t0 Lo0 LO 10 p
BREEDING
SEASONS
YOUNG
IN NEST
LONGLEAF PINE
CONES AVAILABLE
TURKEY OAK
ACORNS AVAILABLE
1984-a8
TRACKING SESSIONS
2 3
...... ,........
JUN JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY
Tracking-session dates and seasonal events of
importance to Sherman's fox squirrel.
4
...""""".
5 6
*..... .......
Figure 5.
1
"""o""
32
following the turkey oak mast failure in that fall, neither had enlarged
testes when captured in December nor were they observed in mating chases
during what should have been the winter breeding season. In fact, none
of the highly visible mating chases were seen by this investigator or
others questioned during this season in 1984. Following an apparently
good turkey oak acorn crop in 1985, however, M2 and five other squirrels
were observed in a mating chase on 24 November (S. Brand pers. comm.).
These data suggest that the squirrels may respond to mast failures by
not breeding, thereby conserving energy.
Home Range
Tracking-session dates and number of locations/squirrel in each
session are listed in Table 9. Figure 5 shows the tracking-session
dates in relation to important seasonal events and is a useful reference
when comparing home range and nest use between sessions.
The 95% and 65% HM contours for each squirrel in each radiotracking
session are shown in Table 10 and Figures 6-11. A paired-difference
t-test showed that mean squirrel home range size for individual tracking
sessions was significantly larger as measured by the 95% HM than as
measured by the MCP (P < 0.01). Additionally, the 95% HM measured a
greater percentage of cumulative home range size per tracking session
(X = 63.17%) than did the MCP (x = 47.58%).
Mean cumulative home range size (95% HM) was 42.8 ha for males and
16.70 ha for females. These measures include all squirrels, so the
number of locations/squirrel and sessions tracked are not constant. Two
Table 9. Dates of radiotracking sessions and number
of locations/squirrel in each session.
Dates
17Jun84-22Ju184
24Aug84-15Sep84
190ct84-17Nov84
14Dec84-19Jan85
22Feb85-16Mar85
19Apr85-12May85
No.
M1
21
24
24
23
24
24
140
locations/squirrel
M3 M4 Fl F2
22 21 0 0
24 18* 0 0
12* 0 12* 0
0 0 23 23
0 0 24 24
0 0 24 24
58 39 83 71
*Squirrel not tracked through entire session.
Tracking
1
2
3
4
5
6
Total
---
---
LIt
Ii-
C.)
O
tO
CO
CM -d- 0 0
I 1 2 1 .
C C\JY )
-4O -4
N N d
I 1 1 1 1 1 2 2 2 I 2 2 I 2 2
2 1 1 1 1 1 2 2 2 I I 2 2 2 2
Ct)
*-*
n
hO
* .
0"'- i
C4t
0)
to
ro
ko
0n.
t0O
LO
LOT
Vi
20)
S- 1
0)
0. '
2: .
tL CM CO
0m -4
-o 00 -0
2- q -2 rI
-1 -n -4
T-11 1 1 1 1 *
bbio
-4 C l-4 .-4 .-4 r -4 r .-4 -4 r-4 -4 r.-4- .-I1 -_
.-4 (X oXi 00 i LOO .0 M N -4 00 r to C:) td m-
UNNOMOO 000~~0000000
0a) r M) -4 0Co r) o .-4 1r-4 -zM- o ct) 0.) to to -- tO 1- .4- N- cm
C C C m 0) CD 1-4 Ct) co to to t M to CD to M tO
C(i rN tO to O to to .-4 tO (Y) q- r-4 to tO -1 '-A C -4 .-4 CD .)
Co to m mOLs-IC.-) m m .: t LO-C MMO' L-ir-d to O to r-I (n
s t I Ct) C** n to CY) :d- t o-t o td-O to LO No- o
,-I Ir-4 -4 i-f C 2 C2 M O m e L
(17 JUN-22JUL)
Figure 6.
Harmonic mean home ranges:
tracking session #1. Dashed
lines connect outlying
portions of the home range.
(24 AUG-15 SEP)
Figure 7.
Harmonic mean home ranges:
tracking session #2.
(19 OCT-17 NOV)
Figure 8. Harmonic mean home ranges:
tracking session #3.
(14 DEC-19 JAN)
Figure 9.
Harmonic mean home ranges:
tracking session #4.
(22 FEB-16 MAR)
Figure 10.
Harmonic mean home ranges:
tracking session #5
(19 APR-12 MAY)
Figure 11.
Harmonic mean home ranges:
tracking session #6.
males (Ml and M2) and two females (F1 and F2) were tracked through
Sessions 4-6. For this period, the males had a significantly larger
mean home range (95% HM x = 31.91 ha, P < 0.01) and core area (65% HM
x = 8.45 ha, P < 0.05) per tracking session than the females (95% HM
x = 10.99 ha and 65% HM 7 = 2.95 ha). Mean cumulative home range and
core area for sessions 4-6 also were larger (P < 0.01) for the males
(95% HM 7 = 64.97 ha and 65% HM 7 = 12.79 ha) than for the females
(95% HM x = 16.7 ha and 65% HM = 5.68 ha).
The 65% to 95% eveness ratios are listed in Table 11 and can be
seen diagrammatically in the plots of the 95% and 65% HM contours in
Figures 6-11. Considering the great difference between male and female
home-range sizes, one might expect eveness of home-range use to be
greater in females than in males. While mean male (.30) and female
(.28) eveness ratios for tracking sessions 4,5,and 6 were not
significantly different (P > 0.10), cumulative (sessions 4-6) home-range
eveness (sessions 4 through 6) was greater (P = 0.0519) in females
(3 = .34) than in males (T = .20) (Fig. 12).
Three-dimensional diagrams of animal locations also can be used to
depict intensity of home range use. Cumulative use of the study area by
all of the tracked squirrels is shown three-dimensionally in Figure 13.
The X and Y axes represent the x,y grid coordinates and the Z axis
represents number of locations/25-ha grid cell. Squirrels' repeated use
of certain nests results in points very high on the Z axis. Location of
such nests is even more obvious in Figure 14 which shows the home range
use of Ml, M2, Fl, and F2 in sessions 4-6.
The effect of season on home-range size and eveness is difficult to
assess because only two squirrels were tracked for the entire study
42
Table 11. Eveness of fox squirrel home range use as described by ratios
of 65% HM to 95% HM home range size.
Tracking
Session(s) M1 M2 M3 M4 Fl F2 M F All
1 .14 .26 .44 .17 .25 -
2 .20 .19 .19 .19
3 .24 .37 .31 -
4 .13 .41 .19 .05 .27 .12 .19
5 .13 .35 .23 .47 .24 .35 .30
6 .31 .46 .36 .37 .38 .37 .38
1-2 .14 .28 .26 .23 -
1-3 .31 .35 .33 -
1-4 .42 .40 .41 -
1-5 .32 .40 .36 -
1-6 .26 .34 .30 -
2-3 .37 .36 .37 -
2-4 .41 .43 .42 -
2-5 .30 .31 .30 -
2-6 .38 .30 .34 -
3-4 .26 .44 .35 -
3-5 .17 .25 .21 -
3-6 .21 .25 .23 -
4-5 .18 .15 .21 .22 .17 .21 .19
4-6 .15 .25 .35 .33 .20 .34 .27
5-6 .25 .34 .42 .43 .29 .43 .36
(14 DEC-12 MAY)
Figure 12.
Harmonic mean home ranges:
tracking sessions #4-6.
44
i,
au
34-
-0
-II
-) O
OE
S-r-
0r- "
U) C
*- 0
r- I-
r- -0
0
o-
0
- 0
oE
I S- II
O
s- o
I- r-
rS- )
03
0- .
EI*
I II
1 4a>
r--
-C
01-
TRACKING SESSION #4
TRACKING SESSION #5
TRACKING SESSION *a
X (METERS)
Figure 14.
Three-dimensional plots of squirrel home range use in
tracking sessions 4-6. (Square=M1, balloon=M2, club=F1,
and cross=F2.)
Z *
Y (METERS)
46
period. Data in Table 7 indicate a pattern of increased home-range size
in sessions 2 and 3 from session 1. This may have been a response to
the peak of the pine cone-cutting season in session 2 and searches for
acorn-producing oaks in session 3. Individual differences rather than
general trends were prevalent in sessions 4, 5, and 6 for both males and
females. Eveness ratios in Table 9 provide little additional insight
into seasonal changes. The very low eveness ratio for M1 in session 4
does indicate that only a small part of this surprisingly large home
range was used frequently.
Nests
Each squirrel used several nests. Most were leaf nests, formed of
twigs, leaves, Spanish moss, and/or pine needles. Very few were cavity
nests in decaying turkey oaks. Nests were used by individual squirrels
an average of 2.65 times each. Most nests (49% of total) were used by a
squirrel only once. 17.3% nests were used twice, 12.5% were used three
times, 15.4% were used 4-7 times, and 5.8% were used 8-11 times.
Nest Numbers
Numbers of nests used are summarized in Table 12. Total number of
nests used by each squirrel varied from 9-28. Total number of nests
used by a squirrel in any one tracking session varied from 3-10
(x=5.76). With the exception of M4, the number of nests used by each
squirrel was relatively constant, as is shown by the mean number of
nests/tracking session and the ratio of total number of nests to number
of locations in Table 10. An average expected number of nests used by a
squirrel over a one-year period may be extrapolated from the latter
Table 12. Summary of number of nests used by fox squirrels.
Individual
# Locations
# Complete
Tracking Session
Total # Nests Used
x Nests/Tracking
Session
# Nests: #Locations
Ml M2
140 138
6
21
4.67
.150
6
28
6.00
.203
M3
58
2
9
4.50
.138
M4
39
1
16
10.00
.410
Fl
83
3
19
6.67
.229
F2 x
71 88.17
3
11
6.00
.155
3.5
17.33
5.76
.214
---
48
figure in the following manner: since squirrels were tracked following
a schedule that would yield a maximum of 144 locations/squirrel/year,
the mean ratio of number of nests to number of locations (.214) is
multiplied by 144, resulting in an expected average of 30 nests/squirrel
over one year. This estimate may be too high if human disturbance by
radiotracking caused the animals to seek refuge in nests they would not
have ordinarily used. On the other hand, it is also possible that
squirrels use more nests than are estimated by a sample of 144
locations/year.
Nest Use Overlap
Some overlap did occur in the nests used. Nine nests (9.5% of all
nests) were used by two of the tracked squirrels. The first user died
before the second animal was located in two of these nests, and the
first animal had not been located in three of the nests in over 4 months
when the second animal was located in them. M2 and M4 more closely
overlapped in their use of two of the nests. M2 was located in the
first of these on 1 May 1984 and again on 21 June. M4 was found in this
same nest on 29 June. M2 was located in the second nest on 6 April
1984, 1 May, 29 June and 22 February 1985. M4 was found in this nest on
5 June 1984 and again on 1 September. A nest "shared" by M1 and Fl was
used by Ml on 15 April 1984 and again on 15 March 1985, then by F1 on 19
April 1985. Because this nest was in an area outside of Ml's primary
area of use in April 1985, he may have been beginning to seeking females
on the area at that time.
A close overlap in nest use occurred on 14 and 15 December 1984
between F1 and F2. At 0920 both were in nests about 100 m apart.
During the afternoon Fl was over 300 m ENE of her first location and F2
49
was near the edge of Lake Rowan, over 300 m SSW of her morning location.
At 1632 F2 was located very near her afternoon location, but Fl was
found in the nest F2 had occupied that morning. In the first two
tracking periods of the next day, Fl was still in the "shared" nest and
F2 was over 300 m away in a nest not far from her two previous
locations. Following this, neither squirrel was relocated in the
"shared" nest.
Percent of Locations in a Nest
The percent of total locations in which squirrels were in a nest
varied considerably. As seen in Table 13, 50.7% of all squirrel
locations were in a nest, but means varied from 32.8-65.1% for
individual squirrels and from 34.4-69.7% for individual tracking
sessions. Figure 15 shows individual and mean percent of locations in a
nest by tracking session as well as mean temperature of tracking days in
each session. A strong negative correlation between temperature and
percent locations is evident. The r2 (coefficient of determination)
value for this relationship is 0.84. Only between sessions 1 and 2 did
temperature and percent locations in a nest vary positively. This may
be explained by heavy rainfall (6.8 cm) on tracking days in session 1
compared with session 2 (0.18 cm), squirrels were usually located in a
nest during heavy rains. Increased pine cone-cutting in session 2 also
may have contributed to decreased nest use.
The negative relationship between temperature and nest use is
further demonstrated by the fact that only during the two "coldest"
tracking sessions (4 and 5) was the percent of locations in a nest
higher during the middle than the last tracking period of the day
(Fig. 16). During these two sessions, peak temperatures were not
Table 13. Summary of percent of fox squirrel locations in a nest.
Individual M1 M2 M3 M4 Fl F2 ALL M F
% of all
Locations
in a Nest 45.0 49.3 32.8 53.8 65.1 60.1 50.7
% of Tracking Session (T. S.)
Locations in a Nest
T. S. #1 42.9 40.9 36.4 66.7 46.5
T. S. #2 37.5 33.3 29.2 38.9 34.4
T. S. #3 41.7 50.0 33.3 75.0 48.6
T. S. #4 56.5 75.0 73.9 73.9 69.7 65.1 73.9
T. S. #5 54.2 54.2 58.3 58.3 56.3 54.2 58.3
T. S. #6 37.5 45.8 58.3 50.0 47.9 41.7 54.2
T. S. #4-6 49.3 57.4 63.4 60.6 57.7 53.2 62.0
S
S.
* .
* S
S
* S
S
S
S
S
S
S
S
S
0
S
S
S
*
*
*
*
*
S
S
* S
S S
5 5
5 0
S
0
S
S
S S
5
S
S
S
S
S
5
0
5
S.
55
S
I I I I I
1 2 3 4 5
JUL SEP NOV JAN
TRACKING SESSION
Figure 15.
Air temperature and percent of squirrel locations
in nests by tracking session. Solid line connects
mean tracking-day temperatures, dotted line, mean
percent of all squirrel locations in a nest.
-26
-24
m
>
m
-22. 7
C
-18
-16
3. a
I
6
MAY
MAR
'"
--
v
1 2 3 4 5
SEP
NOV
JAN
MAR
TRACKING PERIOD AND SESSION
Mean percent of all
tracking period and
B = middle 1/3, and
squirrel
session.
C = last
locations in nests by
A = first 1/3,
1/3 of daylight period.
JUL
ABC
6
MAY
Figure 16.
53
reached until later in the day, and squirrels often stayed in the nest
through period B and became active in period C. During warmer periods,
overall percent of locations in the nest dropped and were at their
lowest during the warm, middle period of the day.
Females had a higher percent of locations in a nest than males in
Sessions 4-6. Most of the difference occurred in period B, when 58.3%
of female locations were in a nest compared to 34.8% of male locations.
Percent of locations in Periods A and C were only slightly higher
(<2.0%) for females than for males.
Nest Location and Composition
Squirrel nests were found in six tree species: turkey oak (68.6%),
longleaf pine (17.7%), live oak (4.9%), post oak (3.9%), laurel oak
(3.9%), and slash pine (Pinus elliottii) (1.0%). Percent of nest uses
was somewhat disproportionate to percent of total nests in turkey oaks
(59%), live oaks (9.3%), and post oaks (8.6%). This resulted from a few
heavily used nests in trees of the latter two species.
As previously mentioned, tree cavities were seldom used by the
tracked squirrels. Only seven cavities were used (7.4% of all nests) a
total of nine times (3.4% of all nests uses). Six of the nine cavity
uses occurred during adverse weather conditions of either cold (5.5 C)
or rain. During this study, squirrels were never observed using "stump
dens" or seeking refuge in gopher tortoise burrows, as reported by Moore
(1957).
Forty-seven nests were counted on 10 ha in ecotone and 27 on 10 ha
in upland, resulting in means of 4.7 and 2.7 nests/hectare for the two
areas, respectively. A T-test revealed this difference to be
significant at P < 0.05. Mean dbh of random and turkey oaks containing
54
nests were significantly different in both ecotone and upland
(P < 0.01), suggesting that squirrels select the larger trees (Fig. 17).
Nest composition (twigs, leaves, Spanish moss and/or longleaf pine
needles) was similar in both habitats for all materials except Spanish
moss, found in 87% of ecotone nests and only 59% of upland nests.
Presumably this difference is due to greater availability of Spanish
moss in the ecotone.
ECOTONE
NEST
.11 1
12 18
UPLAND
NEST
24 30 36 42 48
(1L
12 16 20 24 28 32
I
ECOTONE
RANDOM
II
12 18 24 30 36 42 48
UPLAND
RANDOM
j_.L
12 16 20 24 28 32
DBH (CM)
Figure 17. Frequency and distribution of turkey oaks
containing nests compared with randomly
selected turkey oaks in ecotone and upland.
A
--
Y
_
n-
-~ -
-,I
A
m
DISCUSSION
Food Resources
As shown by this and other studies (Harlow and Eikum 1963, Nixon
et al. 1980, and Christisen and Kearby 1984), mast production is highly
variable. The implication of mast patchiness for consumers is a need
for large areas encompassing diverse species of mast-producing trees.
When one crop fails (such as the turkey oak), others (e.g. live and post
oak) may be available. During the 1984 turkey oak acorn failure, Ml's
eastward home-range shift brought him into the live oak fringe around
Barco Lake. M2's home range included the fringe of Anderson Cue. All
of the tracked squirrels used a variety of habitats.
Numerous factors cause variation in mast production. They may
include climate, tree size, inherent genetic potential, site elevation,
and soil characteristics, previous production, crown size, insects, and
disease. Understanding the effects of these factors can help explain
variability as well as aid in prediction of mast production.
Seed production may be enhanced at lower elevations by greater moisture
and soil nutrient availability. At the Ordway Preserve, this effect was
more evident in turkey oaks than in longleaf pines (which have very deep
taproots). Usually, increased tree size also is associated with higher
mast production. Both tree size and elevation
56
57
affected seed production in the longleaf pines and turkey oaks on the
Ordway Preserve.
Seed production on the Ordway Preserve also was affected by insect
infestation. Larval stages of various insects were found in both pine
cones and acorns. Ebel (1963) found species of the cone moth
(Dioryctria sp.) (commonly called "coneworms") to be the most injurious
insects attacking longleaf pine seeds. No larvae of these species were
identified during this study, but several cones showed typical signs of
having been infested by them. Small orange larvae of the cone midge
(Itonididae) were found in longleaf pine cones. Occurrence of this
insect was reported by Ebel (1963) to be most often associated with
coneworm infestation. Larvae of the cone moth Laspeyresia sp. also were
found in pine cones on the Ordway Preserve. Although not identified at
the time, the larvae found in most of the turkey oak acorns on the
ground on the Ordway Preserve were likely cucurlionid weevils. In a
study in Marion County, Florida, Harlow and Eikum (1963) found
cucurlionid infestation rates of 36.9-81.2% in turkey oak acorns.
Longleaf pines are thought to produce exceptionally heavy seed
crops every 5-7 years (Schopmeyer 1974). The periodicity of turkey oak
acorn crops has not been widely studied, but if like other oaks of the
subgenus Erythrobalanus (red oaks), very heavy or very poor crops rarely
occur consecutively. Seed-bearing is accompanied by a depletion of
stored nutrients and loss of foliage, thus the seed crop of one year
affects that of the next (Matthews 1963). Turkey oaks might be expected
to yield heavy crops as often as every 4 years. Barrett (1931), Gysel
(1956), and Beck and Olson (1968) have reported complementary
seed-bearing years between the red and white oak groups. This has
58
obvious advantages for mast consumers where species of both groups are
found together. At the Ordway Preserve, white oaks (e.g. live and
laurel oaks) are found primarily in lower areas near water. Although
live oak acorn production was not measured directly, many acorns were
observed both on the ground and in the trees within the ranges of Ml and
M4.
The longleaf pine community is dependent on fire for persistence.
The original longleaf pine forests may have burned an average of every
3-5 years (Means and Grow 1985). Suppression of fire, too-frequent
fires (annual), logging, and tapping for turpentine have reduced pine
and increased oak densities over much of the fox squirrel's range.
Properly managed, controlled burning can reverse this trend. After
dropping in October-November, longleaf pine seeds require bare mineral
soil without shade for germination. Once germinated,
(November-December), the seedlings must be protected from fire for one
year (Wakely 1947). Periodic summer burns are recommended to prepare
the seedbed for longleaf pines and help to control turkey oaks.
Due to the extreme variability of seed production, mixed stands of
mast-producing trees are recommended for wildlife (Christisen 1955,
Collins 1961, Goodrum et al. 1971, Christisen and Kearby 1984). The
results of our studies on the primary foods of Sherman's fox squirrel
clearly indicate their need for heterogenous habitat. Longleaf pines
are most important, but mast-producing oaks of both the red and white
oak groups also are necessary. Access to other mast-producing trees
such as hickory also is desirable. (Discarded shells of hickory nuts
were observed several times where fox squirrels were known to have fed).
Lower slopes in ecotonal habitat and the live oak fringes of sandhill
59
lakes are very important due to their higher mast yields and species
diversity.
Home Range
The patchy distribution of the primary food resources of
southeastern fox squirrels may well explain why their home ranges are
larger than predicted on the basis of body size and diet (Weigl et al.
in press). In areas of more uniform resource distribution, western fox
squirrel home ranges average from 0.8-7.0 ha (MCP) (Baumgartner 1943,
Bernard 1972, Adams 1973, Havera and Nixon 1978). In contrast, mean
southeastern fox squirrel ranges average from 19.5-30.3 ha (MCP) in
Georgia (Hilliard 1979), North Carolina (Weigl et al. in press), and
Florida (this study).
In this study, tracking spanned an average of 7 months, and
squirrels were located on average 88 times over 48 tracking days. The
resulting MCP home ranges were 40.0 ha for males and 20.6 ha for
females. Hilliard (1979) tracked S. n. niger in Georgia an average of
1.4 months and located individuals on average 203 times over a period of
32 tracking days. Mean MCP home ranges were 26.4 ha for males and
13.0 ha for females. In North Carolina, Weigl et al. (in press) tracked
S. n. niger over an average of 2 months, locating each individual 78
times. Mean MCP range was 22.8 ha for males and 16.2 ha for females.
The different tracking schedules make it difficult to conclude that
S. n. shermani has a larger home range than S. n. niger, but this may
well be true since S. n. shermani is the largest of the southeastern fox
squirrels (Williams 1977).
60
Home range is clearly larger in males than in females, frequently
2-3 times larger in Florida. Short-term intensity of home-range use may
be similar between the two sexes, but cumulative home range use is more
even in females than in males. Female fox squirrels tend to be more
sedentary (Nixon et al. 1980), while males generally make more
long-distance forays, cover larger areas, and may have greater home
range overlap.
Although usually solitary, the squirrels were not strictly
territorial. Animals overlapped in both home range and nest use. No
evidence of home range defense was observed in these squirrels. With
their smaller home ranges, females might be more likely than males to
defend a core area, particularly when rearing a litter.
Home-range size and use may vary with seasonal food abundance,
reproductive activity, and climate. Home ranges may shift during
shortages of a particular food resource (e.g., Ml). Increased
long-distance forays are characteristic of males before and during the
breeding season. Extremes of temperature and precipitation may cause
decreased activity, thus decreasing home-range size. During this study,
a combination of these and other variables typically acted on squirrel
behavior and home range size, making their effects difficult to
separate.
Nests
Nest use provided additional insight into the ecology of Sherman's
fox squirrel. Percent of time spent in the nest increased both as
temperature decreased and during periods of heavy rain. Only during the
two "coldest" tracking sessions was percent of locations in a nest
higher during the second than the third period of the day, suggesting
that squirrels became active later in the day when temperatures were
warmer. Weigl et al. (in press) also reported later onset of activity
during winter months. Female squirrels spent more time in the nest than
males, providing further evidence of the sedentary tendencies of female
fox squirrels.
Most nests were leaf nests rather than cavities (dens). While
important in colder climates, apparently dens are not required for
survival in Florida. Hilliard (1979) reported similar results but
thought that dens were essential for litter survival and that den trees
limited fox squirrel populations. D. David (pers. comm.) installed
squirrel nest boxes in suitable habitat but reported Sherman's fox
squirrels used them very little. While juveniles may indeed be at some
greater risk of predation without them, this study provides no evidence
that lack of den trees limits squirrel populations in mild southern
climates. Much more evidence exists to indicate that food resources are
the primary limiting factor of southeastern fox squirrel populations.
Turkey oaks containing squirrel nests were larger than randomly
sampled turkey oaks. Allen (1952) and Hilliard (1979) also observed
that fox squirrels appear to select larger trees. Larger trees have
obvious advantages for nest stability and potential size. Large,
well-insulated nests provide the best protection from the elements and
were often used by squirrels during inclement weather.
More nests were found in ecotone than in upland, indicating higher
squirrel densities in ecotone. Advantages of ecotonal habitat are
several: 1) turkey oak acorn production is higher, 2) other important
62
mast producers such as live, laurel, and sand post oaks occur, 3) food
resources in general are more diverse, 4) Spanish moss, important for
winter nest insulation, is more abundant, and 5) the increased ground
and canopy cover can provide squirrels better protection from raptor
predation. Clearly, the ecotonal lower slopes of the sandhill are a
vital part of the fox squirrel's habitat.
The mean number of nests in ecotone and upland combined was
3.7/hectare. If squirrels use 30 nests/year as calculated (although
some overlap in nest use occurs, its effect may be cancelled by the
2
persistence of unused nests), 12.3 squirrels would be expected per km2
Although this number is higher than the 8.4/km2 predicted by Humphrey et
al. (1985), it is consistent both with the number of squirrels observed
by myself on the squirrel study area and with the predictions of other
investigators at the Ordway Preserve (M. Sunquist, pers. comm.).
Fewer squirrels would be expected to occur in less favorable habitat.
Nest counts appeared to provide a reliable estimate of squirrel density
on the study area and should now be tested as an index of squirrel
abundance in other parts of the animal's range. If squirrel
distribution and abundance can be determined from nest counts, informed
management of squirrel populations will be made possible.
How these densities compare to those of other subspecies is
difficult to conclude. In a study of S. n. niger in North Carolina,
2
Weigl et al. (in press) reported a mean density of 5 squirrels/km but
their work was conducted on a greater diversity of sites and may have
better represented average site quality. For S. n. niger in Georgia,
Hilliard (1979) reported a density of 26 squirrels/km2 over the short
duration of his study. High densities during Weigl's study were 28, 22,
63
and 19/km2, which are comparable to that reported by Hilliard. For
S. n. shermani in Florida, Moore (1957) calculated 38 squirrels/km2
based on two years of study near Welaka, Putnam Co., FL, an area
approximately 60 km southeast of the Ordway Preserve.
Only 10-20% of the original longleaf pine sandhill habitat remains
in Florida. Populations of Sherman's fox squirrel have declined as a
result of this habitat loss. Further habitat destruction will surely
result in continued decline, and perhaps even extiction of Sherman's fox
squirrel unless substantial refuges are established. Due to this
squirrel's large home range size and patchy food resource distribution,
refuges on the order of several square miles, not just small parcels,
will be required to sustain viable squirrel populations. Given that
typical squirrel habitat occurs on approximately 1/3 (12km2) of the
37km2 Ordway Preserve, and that squirrels may also use additional, less
favorable habitat, I estimate the population of fox squirrels on the
Ordway Preserve to be about 200 animals. The Ordway Preserve, then, is
a good example of a minimum refuge (>10mi ) necessary to sustain viable
fox squirrel populations. Preservation and maintenance of large areas
of heterogenous, natural sandhill habitat such as the Ordway Preserve
are key to the survival of Sherman's fox squirrel and perhaps numerous
other species of the sandhills as well.
LITERATURE CITED
Adams, C. E. 1976. Measurement and characteristics of fox squirrel,
Sciurus niger rufiventer home ranges. Am. Midl. Nat. 95:211-215.
Allen, J. A. 1871. On the mammals and winter birds of East Florida
with an examination of certain assumed specific characters in birds
and a sketch of the bird fauna of Eastern North America. Bull.
Mus. Comp. Zool. 2:161-540.
Allen, J. M. 1952. Gray and fox squirrel management in Indiana.
Indiana Dept. of Conserv., P-R Bull. No. 1. 101pp.
Barrett, L. I. 1931. Influence of forest litter on the germination and
early survival of chestnut oak, Quercus montana Willd. Ecology
12:476-484.
Baumgartner, L. L. 1943. Fox squirrels in Ohio. J. Wildl. Manage.
7:193-202.
Beck, D. E. and D. F. Olson. 1968. Seed production in southern
Appalachian oak stands. USDA For. Serv. Res. Note SE-91.
Southeast. For. Exper. Sta., Asheville, N. C. 7pp.
Bernard, R. J. 1972. Social organization of the western fox squirrel.
M. S. Thesis, Michigan State Univ., East Lansing, Mich. 41pp.
Chapman, F. C. 1894. Remarks on certain land mammals from Florida with
a list of the species known to occur in the State. Bull. Amer.
Mus. Nat. Hist. 6:333-346.
Chesemore, D. L. 1975. Ecology of fox and gray squirrels (Sciurus
niger and Sciurus carolinensis) in Oklahoma. Ph.D Thesis, Oklahoma
State Univ., Stillwater, Okla. 348pp.
Christisen, D. M. 1955. Yield of seed by oaks in the Missouri Ozarks.
J. Forestry. 53:439-441.
Christisen, D. M. and W. H. Kearby. 1984. Mast measurement and
production in Missouri (with special reference to acorns).
Terrestrial Series No. 13. Missouri Dept. of Conserv. Jefferson
City, Mo. 34pp.
Collins, J. 0. 1961. Ten year acorn mast production study. Louisiana
Wildl. and Fish. Comm. P-R Projects 24-R and 29-R.
Cory, C. B. 1896. Hunting and fishing in Florida. Estes and Lauriat,
Boston, Mass. 304pp.
Dixon, K. R. and J. A. Chapman. 1980. Harmonic mean measure of animal
activity areas. Ecology 61:1040-1044.
Ebel, B. H. 1963. Insects affecting seed production of slash and
longleaf pines: their identification and biological annotation.
USDA For. Serv. Res. Paper SE-6. Southeast. For. Exper. Sta.,
Asheville, N. C. 24pp.
Goodrum, P. D., V. H. Reid, and C. E. Boyd. 1971. Acorn yields,
characteristics, and management criteria of oaks for wildlife.
J. Wildl. Manage. 35:520-532.
Gysel, L. W. 1956. Measurement of acorn crops. For. Sci. 2:305-313.
Hall, E. R. 1981. The mammals of North America. 2nd ed. John Wiley and
Sons, New York, N. Y. 1:1-606+90.
Harlow, R. F. and R. L. Eikum. 1963. The effect of stand density on
the acorn production of turkey oaks. Proc. Annu. Conf.
Southeastern Assoc. Fish and Wildl. Agencies 17:126-133.
Harper, F. 1927. The mammals of the Okefenokee Swamp Region of
Georgia. Proc. Boston Soc. Nat. Hist. 38:191-396.
Havera, S. P. and C. M. Nixon. 1978. Interaction among adult female
fox squirrels during their winter breeding season. Trans. Ill.
State Acad. Sci. 71:24-38.
-------- 1980. Winter feeding of fox and gray
squirrel populations. J. Wildl. Manage. 44:41-55.
Hilliard, T. H. 1979. Radiotelemetry of fox squirrels in the Georgia
coastal plain. M. S. Thesis, Univ. Georgia, Athens, Ga. 121pp.
Humphrey, S. R., J. F. Eisenberg, and R. Franz. 1985. Possibilities
for restoring wildlife of a longleaf pine savanna in an abandoned
citrus grove. Wildl. Soc. Bull. 13:487-496.
Jennrich, R. I. and F. B. Turner. 1969. Measurement of non-circular
home range. J. Theor. Biol. 22:227-237.
Matthews, J. D. 1963. Factors affecting the production of seed by
forest trees. For. Abstracts 24:i-xiii.
Maynard, C. J. 1872 Catalog of the mammals of Florida with notes on
their habits, distribution, etc... Bull. Essex Inst. 4:135-150.
Means, D. B. and G. Grow. 1985. The endangered lonleaf pine community.
ENFO Report, Environmental Information Center of the Florida
Conservation Foundation, Inc. 85-4. 12pp.
Mohr, C. 0. 1947. Table of equivalent populations of North American
small mammals. Am. Midi. Nat. 37:223-249.
Moore, J. C. 1953. The fox squirrel in Florida: variation and natural
history. Ph.D Dissertation, Univ. Florida, Gainesville, Fla.
203pp.
-------- 1954. Fox squirrel receptionists. Everglades Nat. Hist.
2:152-160.
------- 1956. Variation in the fox squirrel in Florida.
Am. Midl. Nat. 55:41-65.
--------. 1957. The natural history of the fox squirrel Sciurus
niger shermani. Bull. Amer. Mus. Nat. Hist. 113:1-71.
Nixon, C. M., S. P. Havera, and L. P. Hansen. 1980. Initial response
of squirrels to forest changes associated with selection cutting.
Wildly. Soc. Bull. 7:298-306.
--------, M. W. McClain, and L. P. Hansen. 1980. Six years of hickory
seed yields in southeastern Ohio. J. Wildl. Manage. 44:534-539.
Nowak, P. M. and J. L. Paradise. 1983. Walker's mammals of the world.
4th ed. Vol II. The John Hopkins Univ. Press, Baltimore, Md.
1362pp.
Olson, D. F., Jr., and S. G. Boyce. 1971. Factors affecting acorn
production and germination and early growth of seedlings and
seedling sprouts. In Oak Symposium Proc., USDA Northeast For.
Exper. Sta., Upper Darby, Pa. 161pp.
Ray, A. A. 1982. SAS user's guide: statistics. SAS Inst., Inc.,
Cary, N. C. 584pp.
SAS Institute. 1981. SAS/GRAPH user's guide. SAS Inst., Inc., Cary
N. C. 126pp.
Schopmeyer, C. S. 1974. Seeds of woody plants in the United States.
USDA For. Serv. Agric. Handbook No. 450. 883pp.
Sokal, R. R. and F. J. Rohlf. 1981 Biometry. 2nd ed. W. H. Freeman
and Co., San Francisco, Calif. 859pp.
Umber, R. W. 1975. Impact of pine plantation succession on sandhill
wildlife in central Florida. M. S. Thesis, Univ. Florida,
Gainesville, Fla. 64pp.
67
Wakely, P. C. 1947. The 1947 cone crop and forest fires. The Forest
Farmer 6:5.
Weigl, P. D., M. A. Steele, L. J. Sherman, J. C. Ha, and T. S. Sharpe.
In press. The ecology of the fox squirrel (Sciurus niger in North
Carolina: implications for survival in the Southeast. Tall
Timbers.
Williams, A. I. 1977. Energetic determinants of size in the fox
squirrel, Sciurus niger. M. A. Thesis, Wake Forest Univ.,
Winston-Salem, N. C. 97pp.
Williams, K. S. and S. R. Humphrey. 1979. Distribution and status of
the endangered big cypress fox squirrel (Sciurus niger avicennia)
in Florida. Florida Sci. 42:201-205.
BIOGRAPHICAL SKETCH
Angela Torres Kantola was born in Jones, Oklahoma, in 1961. She
graduated from Mason High School, Tulsa, in 1979, and from Oklahoma
State University, Stillwater, in 1983 with a B. S. in wildlife ecology.
She was married to Edward C. Kantola in July, 1983.
During her graduate education at the University of Florida, Angela
was employed as a biologist (wildlife) with the U. S. Fish and Wildlife
Service Sirenia Project through the Service's Cooperative Education
Program. Angela received the M. S. in forest resources and conservation
with a concentration in wildlife ecology in August 1986.
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a thesis for the degree of Master of
Science.
Stephen R. Hump rey,/Chairman
Associate Professor, Forest
Resources and Conservation
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a thesis for the degree of Master of
Science.
icha W. Collopy -
Associate Professor, Forest
Resources and Conservation
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a thesis for the degree of Master of
Science.
freesor, Forest R sources
and Conservation
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a thesis for the degree of Master of
Science.
/em n E. Sunqrifst
Associate Professor, Forest
Resources and Conservation
This thesis was submitted to the Graduate Faculty of the School of
Forest Resources and Conservation in the College of Agriculture and to
the Graduate School and was accepted as partial fulfillment of the
requirements for the degree of Master of Science.
Director, Forest Resource and
Conservation
C-.
Dean, Graduate School
August 1986
|
RSS
HIDE
