ACTIVITY, MOVEMENT AND HOME RANGE
OF VIRGINIA OPOSSUM (DIDELPHIS VIRGINIANA)

IN FLORIDA

Jan Ryseri

ABSTRACT

The activity and spatial organization ofthe Virginia opossum (Didelphis virginiana), a solitary didelphid
marsupial, were studied in north-central Florida. Opossums were strictly nocturnal, and the activity level peaked
between 2000 h and 0200 h. Activity was influenced by temperature. Below 8°C, duration of nightly activity
decreased strongly with decreasing temperature. Above 8°C, activity tended to decrease with increasing
temperature. Average nightly activity did not differ among months from August to December or between males
and females. Opossums lived in well defined and at least seasonally stable home ranges with moderate to
extensive home range overlap. Females tended to more exclusive space use and greater stability of home ranges
than mates. Males often shifted home ranges at the end of mating seasons. Three different measures of home
range size were presented. Average values for males and females were 141.6 ha and 64.4 ha, respectively
(corrected minimum convex polygon). Individual home range sizes of males were influenced by habitat  being
larger in presumably poorer habitats, and by body size. Average distances moved per hour were 234 m in males
and 178 m in females. Nightly movements of males and females averaged 1835 m and 1465 m, respectively
(maximums 4665 m and 3973 m). Distances traveled were associated with temperature, home range size or body
size. Very limited data from other regions suggest greater space use by opossums in the studied population.

RESUMEN

En el centro-norte de Florida, se estudi6 la actividad y organizaci6n espacial de la comadreja de Virginia
(Didelphis virginiana), el cual es un marsupial did6lfido solitario. Las comadrejas fueron estrictamente
nocturnas y el nivel de actividad liegi a su maximo entre las 2000 hr y las 0200 hr. La actividad estuvo
influenciada por la temperatura. Bajo los 8 ° C, la duraci6n de la actividad nocturna disminuy6 fuertemente con
una disminuci6n de la temperatura. Sobre 8 ° C, la actividad tendid a decrecer a medida que la temperatura
aumentaba. El promedio de actividad nocturna no difiri6 entre meses, enlre agosto y diciembre, ni entre hembras
y machos. Las comadrejas habitaron dmbitos de hogar bien definidos y estables at menos estacionalmente con

1 Department of Zoology, University of Bern, Baltzerstrasse 3,3012 Bern. Switzeriand. and Florida Muscurn of Natural History, P. 0. Box 117800,
University of Florida, Gainesville FL 32611-7800, U.SA Present address: Doribergstrasse 1,3550 Langnau, Switze,land.

RYSER, J. 1995. Activity, movement, and home range of Virginia opossums (Didelphis virginiana) in Florida.
Bull. Florida Mus. Nat. Hist. 38, PL II(6):177-194.



178 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 38, PT. II(6)

moderada a extensa sobreposici6n de dmbitos de hogan Las hembras tendieron a ser mas exclusivistas en el uso
de espacio y a presentar una mayor  estabilidad en los dnibitos de hogar comparado con los machos. Los machos
frecuentemente cambiaron de dmbito de hogar al ttrmino de la temporada de montas. Se presentaron tres
diferentes medidas de tamafio de ambito de hogar. Los valores promedio para machos y hembras fueron 141.6 ha
y 64.4 ha, respectivamente (poligono convexo minimo corregido). El tamafto de los ambitos de hogar
individuales de los machos estuvo influenciado por: el habitat  siendo mAs grandes en habitats supuestamente Inds
pobres, siendo ademas influenciados por el tamaao corporal. Las distancias promedio recorridas porhora fueron
234 m en machos y 178 m en hembras. Los movimientos nocturnos de machos y hembras promediaron 1835 m
y 1465 m, respectivamente (midmos: 4665 m y 3973 m). Las distancias recorridas fueron asociadas con la
temperaturg tamaao de imbito de hogar o tamaho corporal. Datos muy limitados de otras regiones sugieren que
esta poblaci6n de comadrejas utiliza un mayor espacio.

INTRODUCTION

The spatial distribution of an animal population is ultimately shaped by
environmental factors, such as the quality and distribution of resources in space and
time, intraspecific competition, and predation pressure (Clutton-Brock and Harvey
1978; Davies and Houston 1984; Wrangham and Rubenstein 1986). The spatial
distribution, on the other hand, influences the social organization including the mating
system, which in turn affects morphological, behavioral and other traits such as sexual
size dimorphism (Clutton-Brock and Harvey 1978; Rubenstein and Wrangham, 1986;
Ryser 1992).

The spatial organization of the Virginia opossum has been subject to uncertainty.
In his review, Gardner (1982) termed opossums solitary wanderers that rarely remain
in any area and do not maintain separate home ranges. More recent studies (Gillette
1980; Allen et al. 1985; Seidensticker et al. 1987), however, suggested that opossums
live in well defined and at least seasonally stable home ranges.

Here I present results on space use and home range size of opossums in north-
central Florida. The opossum is a nocturnal, solitary, omnivorous marsupial about the
size of a house cat  ranging over most of the eastern United States and south to Costa
Rica, and showing a wide range of ecological tolerance (Gardner 1982). In Florida
opossums have two breeding seasons of about four weeks each in January/February
and May, respectively (Ryser 1992; Sunquist and Eisenberg 1993). The young are born
after a 13-day gestation, weaned at the age of about 100 days (Reynolds 1952; Wright
1989), and are sexually mature by the end of the year. The data were collected during
a two-year radio-tracking study as an essential basis for understanding the opossum
mating system and male mating strategies (Ryser 1990, 1992). Additionally, I
collected quantitative data on activity and movements to investigate geographical and
sex differences in these parameters as well as individual and environmental factors
influencing activity and movement. Despite the opossum's abundance in many parts of
its range, reliable, quantitative data on activity, movement, and home range size are
rare. Gillette (1980) and Allen et al. (1985) presented data on home range size and
Allen et al. (1985) and Seidensticker et al. (1987) on nightly movements. Activity has
been studied in captivity by McManus (1971) and in the field by Seidensticker et al.



RYSER: VIRGINIA OPOSSUMS (DmELPHIS KIRGINIANA) IN FLORIDA 179

(1987). So far there has been no study of an opossum population in Florida, an
environment that differs, particularly in temperature regime, markedly from that of
other studied populations in Wisconsin or Virginia. Moreover, few attempts have been
made to identify the factors that could account for the often considerable individual
variation in space use and activity (Seidensticker et al. 1987; Kissell and Kennedy
1992).

ACKNOWLEDGEMENTS

I am most grateful to John F. Eisenberg and Melvin E. Sunqllist for making this study possible and for
making valuable comments on the manuscript Anonymous reviewers also contributed to the improvement of the
manuscript I also especially thank Debra Wright f6r her cooperation in field work.

STUDY SITE AND METHODS

The study was conducted at the Katharine Ordway Preserve of the University of
Florida  located 40 km east of Gainesville in Putnam County, north-central Florida.
The 37 km2 tract consists of a mosaic of habitat types (see Franz and Hall 1991 for
details): Permanent and temporary lakes and ponds, freshwater marshes, swamp
forests, mesic hardwood hammocks, sand live oak hammocks, sand hill high pine
forests (tree cover dominated by Pinus palustris and Quercus sp., ground cover
dominated by wiregrass Aristida stricta), and grass-dominated areas derived from
former agricultural lands (old fields, pastures, and orange groves). Maximum
elevation is 55 m above sea level.

The climate is characterized by hot, humid summers and cool, mild winters with
freezing occurring at least on a few days or nights. Mean daily maximum and
minimum temperatures are 26.7°C and 14.0°C, respectively, varying from 19.3/5.9°C
in January to 32.5/21.7°C in July and August. Mean annual rainfall is 133 cm, with
60% falling in the summer months May to September (data for Gainesville, from the
National Oceanic and Atmospheric Administration).

Information on the spatial organization of opossums and its dynamics was
gathered by trapping and radio-tracking opossums mostly in an approximately 8-km2
core study area in the center of the preserve from January 1986 to February 1988.
Opossums were trapped in Tomahawk live-traps baited with fish and dry dog food. In
1986 and early 1987, during the weeks prior to the breeding seasons, I attempted to
trap all opossums in the study area, resulting in 1028 trap-nights. From March 1987 to
February 1988, D. Wright and myself conducted four trapping sessions with 140 trap
sites covering most of the preserve. During each trapping session in March/April,
July/August, October/November and December-February we ran the trap lines for 6 to
8 days. Opossums were measured, ear-tagged at their first capture, and pouch young



180 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 38, PT. II(6)

were toe-clipped for permanent identification. Adult animals in the study area were
fitted with radio-collars (Cedar Creek Bioelectronics, Bethel MN), weighing about
35 g. In 1987, radio-tagged individuals included subadults trapped in October/
November, weighing at least about 1 kg. Data were collected by radio-tracking
opossums at night during breeding seasons (see Ryser 1992), by nightly interval-
tracking (see below), and by regularly locating opossums in their dens during the day
at all seasons. A total of 113 opossums (70 males and 43 females) was radio-tagged in
the course of the study.

I studied the activity and movements by locating 28 different radio-tagged males
and females at 1-hour intervals, usually on four consecutive nights per month from
August to December (non-breeding season, except for four opossum-nights of data
from January), yielding 147 opossum-nights of data. While still inactive, the animals
were checked about every 15 to 30 minutes to allow a reasonably exact determination
of the onset of activity. Activity was detected by the changing strength of the radio
signal caused by the animal's movements. Similarly, the end of activity was
determined by a lack of movement and was verified an hour later as well as on the
next day, when the actual den sites were checked. Nightly movements were plotted on
a map (scale 1:6666), and distances traveled were measured as the sum of straight-line
distances between consecutive locations. Two measures were calculated: Mean
distance moved per hour and total distance traveled per night. Factors influencing
these two variables were investigated by means of partial correlation analyses.
Variables examined included nightly minimum temperature (obtained from a weather
station in the preserve), home range size, body size (head and body, millimeters), and
age (months). Age was known based on marking of the animals as pouch young or
juveniles/subadults or was estimated based on tooth wear and appearance. The short
life-span of opossums and seasonal births facilitated age determination (Sunquist and
Eisenberg 1993). The level of significance was set at 0.05 divided by the number of
examined factors (Bonferroni correction), thus 0.01.

Data for determining home ranges consisted of all available location points of
radio-tagged opossums, that is daytime locations (animals inactive) and nighttime
locations (animals active) of all seasons, except for male locations during the breeding
seasons. Most data were collected in the period just before and during breeding
seasons and on the nights when animals were located hourly. Tracking periods from
the time of radio-tagging until individuals died, were lost, or transmitters were
removed, and thus the time frame for estimation of home ranges of individual
opossums averaged 118 days in males and 133 days in females. Sizes and shapes of
home ranges containing at least 20 location points were calculated using the computer
program McPAAL, version 1.2 (M. Stuwe, National Zoological Park, F.ront Royal VA
22630). I calculated three different measures of home range size based on the
minimum convex polygon and the harmonic mean model (Dixon and Chapman 1980):
(1) the minimum convex polygon (MCP), (2) the MCP with inaccessible areas (lakes
>l ha, wet prairie) subtracted (MCPcorr), and (3) the harmonic mean 95% isopleths
(HM95). A grid of 15x15 cells was chosen for the harmonic mean calculations.



RYSER: VIRGINIA OPOSSUMS (DIDELPHIS VIRGINIANA) IN FLORIDA 181

If opossums clearly shifted their area of activity and did not return to the
previously used area, the new location points were excluded or two separate home
ranges were calculated if data were sufficient. For the calculation of average home
range sizes, the data set was restricted to values based on at least 40 location points in
order to reduce the effect of sample size on home range size. Factors associated with
home range size (MCPcorr) were investigated by a partial correlation analysis.
Examined variables included age, body size, tracking period (days), number of
location points (N), and the percentage of total home range size comprised of sandhill
and of old fields and pastures. These are areas of presumably low productivity with
regard to food abundance. Studies on population densities and habitat use of rodents
(Brand 1987; Jones 1989), and an analysis of habitat use of opossums showing that
these two habitats were used significantly less than expected based on their area (Ryser
1990), supported this notion.

Means were compared for significant differences using student's t-test (two-
tailed). The level of significance was set at 0.05.

0600 -
-

0400 - -

- 22

ES
TI

M
AT

ED
 T

IM
E

0200 - 20
-

2400 - 16
14

2200 - 12

(0.) 3G
n.LV

Y
3dn31 V\InW

INIM

10
1-- -2000 i 8

1800 -

lilli
AUG SEP OCT NOV DEC

Figure 1. Distribution (half-hour-intervals), means (arrows), and extremes (1) ofthe individual onset and end of
nightly activity from August to December. The lower and upper bordering lines represent the time of sunset and
sunrise. Average nightly ' ' temperatures of each month are connected by straight lines. Minimum
temperatures varied from 21-22°C in August  19-20°C in September, 12-14°C in October, 1-16°C in November,
and 0-15°C in December. Samples sizes for the five months are 35, 16, 17, 43, and 32, respectively. Data from
males and females were pooled.



182 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 38, PT. 11(6)

RESULTS

Activity and Movement

Opossums were strictly nocturnal (Figs. 1,2). Figure 1 shows the distribution of
the onset and end of nightiy activity from August to December in relation to sunset
and sunrise and minimum nightly temperatures. Usually opossums emerged about one
hour after sunset and were active for a varying period of time before reentering a den.
However, some animals became active only later at night. Leaving the den late was
more common in cool nights. Minimum temperature at nights when opossums
emerged from dens after 2200 h averaged 8.0°C (n=12) compared to 13.2°C of all
considered nights. The time activity ended varied more strongly than the time of the
onset of activity. The scatter increased in November and December as average nightly
temperatures declined and the variation in temperature increased. Average nightly
activity (* S.D.) from August to December was 8.01 h (* 1.67), 8.81 h (* 1.47), 8.54 h
(* 1.24), 7.85 h (* 3.13) and 8.22 h (* 3.66) and did not differ sigitificantly among
months (Kruskal-Wallis test, df=4, H=4.28, p=O.37).

Maximum activity is reached between 2000 h and 0200 h (Fig. 2), when on
average up to 97.4% of the monitored animals were active (cold nights excluded).

Individual nightly activity varied greatly in duration and tended to be slightly,
but not significantly longer in females than in males (Table 1, t145=0·97, ns). Activity
was neither correlated with home range size (r=0.01, n=147) nor with age (r=-0.03) or
body size (r=-0.04), but was clearly influenced by temperature (Fig. 3). Activity
increased linearly with temperature up to about 8°C (y=2.561+0.828x, n=33,
r2=0.571, p<0.0001). Above this threshold activity was less temperature-dependent,
but tended to decrease with increasing temperature (y=11.053-0.136x, n=114,
r2=0.129, p<0.0001). Based on the former equation, activity is predicted to cease
altogether at nightly minimum temperatures of -3°C. Nightly inactivity was recorded
in only one case (minimum temperature 0°C), but several additional cases occurred
during the mating season in January at nightly lows of -7° to +6°C.

Distances moved per hour and per night varied widely, but were significantly
greater in males than in females (t140=3.90 and 2.53, respectively, p<0.01 and <0.05,
Table 1). Maximum recorded movements within an hour reached 950 m in males and
790 m in females. Maximum distance traveled per night was nearly 4 km in females
and 4.7 km in males. A partial correlation analysis showed that distance moved per
hour in males was positively associated with home range size and minimum nightly
temperature, whereas in females only body size exerted a significant positive effect
(Table 2). Distance traveled per night is a function of the duration of activity and of
average distances moved per hour and is therefore influenced by the same factors that
influence activity and average hourly movement (Table 2). Thus males with large
home ranges tended to move farther than males with small home ranges and



RYSER: VIRGINIA OPOSSUMS (DIDELPIUS URGINIANA) IN FLORIDA 183

80

%
 a

ct
iv

e 60

40

20

ill'll

1800 2200 0200 0600

Time (EST)
Figure 2. Percentage ofopossums active as a function ofthe time ofnight. • = all nights included, A = only
nights with a minimum temperature >8°C included.

movements decreased with decreasing temperatures. In females, low temperature
negatively affected nightly movements, certainly through its effect on activity.

Nightly foraging movements of opossums, as revealed by radio-tracking at 1-hour
intervals, were usually erratic and unpredictable. However, nightly routes also
included loops and, rarely, straight-line movements. Opossums sometimes moved at a
fairly regular speed during the night while others remained within a small area for
several hours. During a 4-day-period of interval-tracking, individuals with medium to
large home ranges often stayed within a certain but, compared with other months,
changing part of their home range, suggesting that opossums may exploit different
parts of their home range at different times.



184 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 38, PT. II(6)

Table 1. Duration ofactivity and distances traveled per hour and per night ofopossums radio-tracked at 1-hour
intervals. All: all nights, >8'C: nights with minimum temperatures ofmore than 8°C, N=number ofopossum-
nights.

Distance (m) traveled

Activity (h) per hour per night

all >8°C all >8°C all >8°C

Males

Mean 7.86 8.56 234 243 1835 2015
S.D. 2.72 1.91 84.1 81.5 880.8 768.4
Range 0.83-12.0 83-466 100-4665
N 89 69 88 69 88 69

Females

Mean 8.31 9.33 178' 177 1465' 1630
S.D. 2.77 1.56 81.0 76.5 787.6 745.4
Range 0-13.0 60439 100-3973
N 58 45 54 43 54 43

* One case of inactivity excluded.

Spatial Organization

Female opossums lived in fairly well separated and fairly stable home ranges. As
an example of spatial organization, Figure 4 shows female distribution in mid- winter
1987/88. During this period and the second breeding season (April/May) 1987, for
which the data are considered the most complete, 0 to 100% and an average of 43% of
the area of individual female ranges (MCPcorr, n=28) was overlapped by one or more
other female home ranges. On one hand, this figure would be increased by untagged
females living in ranges adjacent to the study area. On the other hand, it overestimates
the actual degree of overlapping use, because the convex polygons do not take into
account spatial or temporal differences in the extent various parts of the home range
are used. Neighboring females often seemed to exploit overlapping areas during
different periods of time, sometimes caused by slight home range shifts, or less
intensively than exclusively used areas. In one case of complete home range overlap,
the individuals were a mother and daughter, and two other cases of complete overlap
involved suspected close relatives. In areas with a presumably greater food abundance,



RYSER: VIRGINIA OPOSSUMS (DIDELPHIS 1,7RGINL4NA) IN FLORIDA 185

20

15-

A
ct

iv
ity

 (
ho

ur
s)

0 ....: .10- :1. .:: ..2,
8 I.0 00 .. :.,.... ...i.. 0 : 01... 1

5-;:. 0

.
.

80 0 .
0

..
.% lilli

-~ 0 5 10 15 20 25

Nightly minimum temperature (°C)
Figure 3. Duration ofactivity as a function ofnightly minimum temperatures.

such as along lakes and in mesic forests, density and home range overlap was
apparently greater than in less productive areas.

Most females stayed in a well defined area until death. Of 19 females tracked for
a period of more than three months, only two clearly shifted their home range about
500-1000 m, and a third continously expanded her range as she grew older. None of
the radio-tagged females dispersed, but two were lost without being retrapped, so were
possible dispersers. One female was apparently trapped and radio-tagged during a
dispersal movement. She disappeared the next day, but one of her small young was
trapped 4.7 km away. One marked female from outside the study area dispersed 2.8



186 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 314 PT. II(6)

Table 2. Partial correlation between mean distance traveled per hour, distance traveled per night or home range
size and various potentially influencing factors (see method for definitions). Level of significance is set at 0.01
(Bonferroni comction).

Males Females

(n=88)* (n=55)
Distance/hour Min.temp. r=0.27 r=0.24

Range size r=0.47** r=-0.1f
Body size r=0.03 r=0.36
Age r=0.01 r=0.13

Distance/night Mintemp. r=0.38 ** r=0.55**
Range size r=0.44 ** r--0.18
Body size r=0.01 r=0.27
Age r=-0.11 r=0.11

(n=33)* (n=33)
Home range size Habitat r=0.40 r=0.31

Body size r==0.40* r=.0.11
Age r=0.14 r=0.26
Track.period r=-0.18 r=0.32
N of locations r=0.57 r=0.30

*

** P<0.01
p<0.001

km. Both dispersal movements occurred when females were carrying young in their
pouch. Vacancies caused by death were rapidly filled by immigrants (unmarked
females) or young. Female young were philopatric and usually established a home
range close to or even within that of their mother. During their first breeding season in
January, female young of 1987 were trapped or located an average of 752 m
(maximum 1950 m, n=12) from their mother. During the study period, female young
were in 1hree cases known to have established a home range within their mother's
range. In one case the young female took over her mother's range when the mother
died after weaning of her young.

Male home ranges showed greater overlap than female home ranges, because
male ranges were considerably larger at similar male population density (Fig. 5). In
summer 1987, for which the most reliable data from the non-breeding season are
available, 0% to 100% or an average of 63% (n=12) of the area of male home ranges
was overlapped by other male ranges. It was not unusual to find three or four males
exploiting the same area. In no case were adult males exploiting the same area known
to be related to each other or to females living within their ranges.



RYSER: VIRGINIA OPOSSUMS (DIDELPHIS VIRGINIANA) IN FLORIDA 187

1 km

Figure 4. Distribution offemale home ranges (minimum convex polygons) in the study area in winter 1987/88.
Hatched areas are lakes.

1 km

Figure 5. Distribution ofrnale home ranges in the study area in summer 1987.



188 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 38, PT. II(6)

Table 3. Means  standard deviations, and extremes of male and female home range sizes (ha). MCP=minimum
convex polygon  MCPcorr=MCP with unusable area subtracted, HM95=harmonic mean 95% isopleth. The data
include home range calculations based on more than 40 location points.

Numberof Tracking
location period
points (days) MCP MCI'corr HM95

Males (n=23)

Mean 90.0 142.6 155.9 141.6 170.6
S.D. 46.1 69.8 113.1 103.1 90.9
Range 46-217 34-275 33.7-564.1 31.1-520.4 41.7-433.3

Females (n=23)

Mean 113.9 161.8 73.1 64.4 86.0
S.D. 70.3 112.7 45.8 40.7 46.7
Range 43-285 54-400 8.1-195.2 8.1-186.6 14.6-220.5

Male spatial structure was less stable than that of females, mainly because of
range expansion during breeding seasons, frequent shifting of home ranges at the end
of breeding seasons, and because of natal dispersal. During mating seasons, males
expanded their home range considerably (see Ryser 1992). After breeding seasons,
males retreated again to a smaller range. But only 11 of 19 males for which sufficient
data from breeding and non-breeding seasons are available, retreated to the same area
as before. Eight males established home ranges in another part of the expanded
breeding range, up to 3.8 km away from the previous home range. Among opossums
trapped in July 1987, two males were caught 2.8 and 4.6 km, respectively, away from
their original site of capture during the breeding season in April/May. One male was
apparently trapped and radio-tagged during a dispersal movement at the end of the
breeding season. He had moved 5.3 km when his collar was removed after 5 days.
Between mating seasons, male home ranges were fairly stable. Of 21 adult males
tracked for more than two months during non-breecling seasons, only two clearly
shifted or expanded their home range. Four males were lost during non-breeding
seasons (July, August) and could possibly have dispersed, although a failure of the
transmitters is more likely.

Male young generally dispersed before or during their first breeding season (see
Wright 1989). Contrary to other males, young males that were recorded dispersing
during the breeding season did not expand home ranges prior to dispersal and thus did
not seem to participate in mate searching. Their ranges averaged only 67.5 ha (n=6,
compare with data below).



RYSER: VIRGINIA OPOSSUMS (DIDELPH/SPIRGINIKNA) IN FLORIDA 189

Home Range

Home ranges varied greatly in size, ranging from 8 to 564 ha (Table 3). Male
home ranges were on average about twice the size of female home ranges (MCP:
t44=3.25, p<0.01), irrespective of the method of calculation. HM95 tended to yield
greater values than convex polygons (MCP): Means differed by 10 to 15% (44=0.31
and 0.53 in males and females, respectively, both ns), and in 82% of all cases
individual HM95 were larger than the corresponding MCP. Factors influencing
individual home range size were examined by means of a partial correlation analysis
(Table 2). In males, home range size tended to be positively influenced by habitat (see
method for definition), body size, and number of locations. In females, none of the
examined factors exhibited a coefficient significantly different from zero.

DISCUSSION

Activity

Opossums were strictly nocturnal and, outside mating seasons, none was found
active before sunset or after sunrise. Most opossums left their dens at dusk and stayed
active for a varying period of time before reentering a den, usually well before dawn.
Consequently, the portion of active animals increased rapidly after dusk and reached
its highest level from 2000 h to 0200 h before gradually declining to zero. McManus
(1971) studied activity patterns in outdoor cages and found a similar curve in spring
and summer. However, the increase in activity was slower, and no more than 85%
were active at any time. In the present study, average nightly duration of activity
ranged from 7.85 to 9.33 hours, depending on season, sample, and sex. Seidensticker
et al. (1987) recorded average values of 8.23 to 10.95 hours, depending on season and
sex, but with no significant differences among the groups.

Activity of opossums showed a temperature-dependence in two directions. At
nightly lows below about 8°C, activity decreased linearly with decreasing temperature,
whereas above this threshold, activity tended to decrease with increasing temperature.
Decreased activity at low temperatures has been reported before in several studies.
Gillette (1980) reported that activity of opossums in Wisconsin decreased rapidly as
temperatures dropped below freezing, and that opossums did not leave their dens at
temperatures below -15°C. Fitch and Sandidge (1953) and Fitch and Shirer (1970)
found that opossums in Kansas stayed in dens for several days during extremely cold
weather and that they sometimes foraged during the daytime after such cold periods.
With activity starting to decrease already at 8°C and, as predicted by the regression,
ceasing altogether at -3°C, opossums in Florida seem to react more sensitively to cold
temperatures than opossums in areas with harsher climates.



190 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 38, PT. II(6)

Decreasing foraging activity at low temperatures is likely to be the consequence
of increased energy expenditure and, at the same time, decreased foraging success,
particularly concerning invertebrates and ectotherm vertebrates, leading to a negative
energy balance at some low temperature. Increasing energetic costs already at
moderately cold temperatures are suggested by the fact that opossums are not well
adapted to the cold. Their pelage is sparse and has relatively poor insulative properties
(Scholander et al. 1950 cited in McManus 1971; Hsu et al. 1988). Opossums start
showing behavioral thermoregulation (e.g. shivering) already at around 10°C
(McManus 1969; Dills 1972), and they are unable to maintain average body
temperatures below 3°C (McManus 1969). In north-central Florida, the generally mild
winter temperatures certainly do not seriously limit the opossum's energy intake, but
the favorable conditions may allow the animals to reduce activity already at less cold
temperatures compared to opossums in more northern parts of their range with longer
lasting freezing temperatures.

The decrease of activity with increasing temperatures above 8°C is most likely
related to other factors associated with temperature. However, average duration of
nightly activity did not differ among months, excluding the possibility that decreasing
activity at higher temperatures was a seasonal effect. Perhaps it was greater food
abundance at higher temperatures which allowed opossums to reduce activity in warm
nights.

The time of the onset of nightly activity did not show much individual variation,
as most animals left their dens within about an hour. However, a few individuals
started activity distinctly later at night, and the onset of activity shifted seasonally
according to the time of sunset. Due to varying individual duration of activity, the
temporal distribution of the end of activity varied to a much greater degree than did
the distribution of the onset. In November and December, apparently as a consequence
of the much greater variation in nightly temperatures, including low values close to
freezing, the patterns of activity clearly deviated from that in August to October.
However, despite similar temperature regime, opossums behaved differently in the two
months. In November, the pattern of the onset of activity corresponded to that of the
earlier months, but increased individual variation in the duration of activity led to a
great scatter in the time activity ended. Average nightly activity, however, was only
slightly lower than in the previous months, because the long nights allowed opossums
in mild nights to stay active for up to 13 hours. In December, on the other hand,
increased individual variation in activity was expressed primarily as variation in the
onset of activity.

Other factors beside temperature that may be responsible for the individual
variation in nightly activity may include the physical condition of the animals,
previous and current weather conditions, and foraging success. Opossums seemed to
stay active longer than expected following periods of cold nights. Individual
differences also seemed to exist, as it was often the same animal that stayed active the
longest, and as opossums sometimes left and entered dens in the same order on each of
the four nights of consecutive tracking.



RYSER: VIRGINIA OPOSSUMS(D/DELPHIS URGINIANA) IN FLORIDA 191

Spatial Organization

The opossum is not necessarily nomadic, as suggested by Hunsaker (1977),
Gardner (1982), and others. Under stable environmental conditions, the opossum's
spatial organization is characterized by fairly even spacing among individuals and at
least seasonal stability of the spatial structure. There are sexual differences, however,
with females having smaller home ranges and tending to more exclusive use and
greater stability of home ranges than males. The impression of earlier capture-
recapture studies of an unstable spatial structure may have been  among others, the
result of the males' expansion of home ranges during, and frequent shifting of home
ranges after mating seasons (Ryser 1992), which could cause males to be trapped
several kilometers from their usual area of activity.

The data from this study suggest that opossums have a highly flexible strategy of
space use. On one hand, females exploit much of their home range exclusively, which
allows the exploitation of predictable resources with minimal competition. On the
other hand  mutual tolerance allows the opossum, as an onmivorous, opportunistic
feeder, to jointly exploit high quality areas or clumped food supplies, such as fruiting
trees, amphibian breeding ponds, or animal carcasses. In May 1987, an amphibian
breeding pond was visited by four opossums almost every night. At close distance,
however, overt aggressive behavior prevailed. Opossums also frequently shifted
foraging areas within home ranges after days or weeks, probably to react to temporal
and spatial changes in the distribution and abundance of food. Although I could not
compare seasonal home ranges due to a lack of adequate data, there seemed to be no
obvious shifts of home range related to particular seasons. Except for cases of
aggression at common feeding places, there was no evidence for physical interactions
among neighbouring individuals, suggesting that exclusive areas in female ranges
were maintained by mutual avoidance. A similar spatial system was found by Charles-
Dominique (1983) in several South American didelphids and by Sunquist et al. (1987)
in Didelphis marsupialis.

Female reproductive success in polygynous mammals such as the opossum is
usually limited by access to adequate resources, such as food and shelter (Trivers
1972). Male reproductive success, on the other hand, is largely limited by access to
mates (Ryser 1992). Male spatial distribution is therefore expected to be influenced not
only by ecological factors, but also by the mating system and the distribution of
females. The sexual difference in space use, which has also been observed in many
other mammals (Harestad and Bunnell 1979), therefore, most likely is related to a
great degree to the male's reproductive strategy, which includes covering numerous
female home ranges during mating seasons (Ryser 1992). The notion that sex
hormones influence greater space use in males is supported by the fact that a male
castrated prior to maturity exhibited a home range size (MCP 68.6 ha) typical of a
female. Some of the sexual difference in home range sizes, however, is also expected



192 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL. 38, PT. II(6)

to be due to the larger body size of male opossums (means 2.73 vs. 2.08 kg) and
consequently greater energetic needs (Harestad and Bunnell 1979; Reiss 1988)

Home Range and Movement

The minimum convex polygon has often been criticized for its sample-size bias,
sensitivity to extreme locations, linear borders, and because it may include large
unused areas (Jennrich and Turner 1969; Dixon and Chapman 1980). In the present
study, however, unused areas consisted mostly of lakes and wet prairie which could
easily be excluded (resulting in MCPcorr). Extreme locations were very rare and did
not represent a serious bias. The harmonic mean isopleths, on the other hand, are
related to the intensity of use and can define home ranges of any shape while being
largely insensitive to extreme locations (Dixon and Chapman 1980). However,
although the limits of the 95% isopleths roughly followed the distribution of data
points, the model was not able to closely delimit the areas of activity, and it could
include large unused areas while excluding particular location points. This makes the
method difficult to use in quantitative studies of habitat use (Ryser 1990). HM95 were
also in most cases larger than convex polygons, the difference being considerable in
cases with clumped location points. The same effect was also observed by Spencer and
Barrett (1984). Additionally, a closer examination of the results showed that HM95
were only slightly less sensitive to varying samples sizes than were MCPs (Ryser
1990).

Therefore, I consider the corrected minimum convex polygon the most reliable
home range size estimate. With average values of 64.4 and 141.6 ha for females and
males, respectively, they are larger than values reported by Gillette (1980) from
Wisconsin (females: 51 ha, n=9; males: 108 ha, n=5) and by Allen et al. (1985) from
Georgia (females: 22 ha, n=5; males: 78 ha, n=2). However, these were based on 95%
convex polygons and a modified minimum area method, respectively, so possibly
underestimated home range sizes compared to this study. Sunquist and Eisenberg
(1993) found a similar mean home range size (58.7 ha, n=18) for female opossums on
the Ordway preserve. Values based on earlier capture-recapture studies (see Gardner
1982) can not be considered comparable to those of radio-tracking studies.
Quantitative data on nightly movements of opossums are equally rare. Allen et al.
(1985) and Sunquist et al. (1987, for D. marsupia/is), both using 2-hour tracking-
intervals, reported values of 1278 (males) and 1026 m (females), and 1376 (males) and
1025 m (females), respectively. Seidensticker et al. (1987) reported values of 1137 to
1550 m depending on sex and season. The opossums in Florida thus tended to travel
farther in search of food, which corresponds to the possibly larger home range sizes
and may reflect a lower overall food density in the study area. The notion that
differences in space use are correlated with habitat productivity is supported by the
finding that the proportion of sandhills and former agricultural lands, areas of



RYSER: VIRGINIA OPOSSUMS (DIDELPHIS URGINIANA) IN FLORIDA 193

presumably low food abundance, was positively associated with male and slightly
positively associated with female home range sizes, suggesting that home ranges are
larger in poor habitats where opossums are expected to range more widely to satisfy
their metabolic needs. Additionally, a quantitative study of habitat use showed that
opossums avoided sandhills and former agricultural lands compared to the proportion
of these habitats (Ryser 1990). Kissell and Kennedy (1992) in Tennessee, on the other
hand, did not find a correlation between habitat parameters and the number of
opossums trapped.

The positive relationship of body size with home range size in males may also be
the result of increasing energetic needs of large individuals. The same explanation
may hold for the influence of body size on movements in females. The correlation
between home range size and movement in males may be related to the habitat. As was
shown, large home ranges tended to be associated with a greater portion of presumably
poor habitat in the range, which could force these opossums to travel farther to
compensate for the lower food density. In the future, field experiments involving the
feeding of selected individuals might help elucidate the relationships between temporal
and spatial food availability, energetic needs and home range size, movements, or
activity.

LITERATURE CITED

Allen. C. H., R. L Marchinton, and W. M. Lentz 1985. Movements  habitat use and denning of opossums in the
Georgia Piedmont Amer. Midi. Nat  113:408-412.

Brand, S. 1987. Small mammal communities and vegetative structure along a moisture gradienL M.S. thesis,
Univ. Florida, Gainesville.

Charles-Dominique, P. 1983. Ecology and social adaptations in didelphid manupials: Comparison with
eutherians of similar ecology. Pp. 395-422 in J.F. Eisenberg, and D. G. Kleiman, eds. Advances in the
study ofmammalian behavior. Spec. Publ. Amer. Soc. Mamm. 7.

Clutton-Brock, T. H., and P. H. Harvey. 1978. Mammals  resources and reproductive strategies. Nature
273:191-195.

Dills, G. 1972. Telemetered thermai responses of a specimen of the Virginia opossum J. Alabama Acad. Sci.
43:55-62.

Davies, N. B., and k I. Houston. 1984. Territory economics. Pp. 149-169 in J. R. Krebs and N. B. Davies, eds.
Behavioural ecology. An evolutionary approach Blackwell Sci. Publ., Oxford

Dixon, K. R., and J. A Chapman. 1980. Harmonic mean measure of animal activity areas. Ecology 61:1040-
1044.

Fitch, H. S., and L L Sandidge. 1953. Ecology ofthe opossum on a natural area in northeastcrn Kansas. Univ.
Kansas Publ. Mus. Nat. }list  7:305-338.

Fitch, H. S., and H. W. Shirer. 1970. A radiotelemetric study of spatial relationships in the opossum Amen Midi.
Nat. 48:170-186.

Franz, R, and D. W. Hall. 1991. Vegetative communities and annotated plant lists for the Katharine Ordway
Preserve - Swisher Memorial Sanctuary, Putnam County, Florida  Ordway Preserve Research Series,
Report No. 3, Florida Mus. Nat. Hist, Univ. Florida Gainesville.

Gardner, A L 1982. Virginia opossum (Didelphis virginiana). Pp. 3-36 in J. AChapman and G. A Feldhamer,
eds. Wild mammals ofNorth America. John Hopkins Univ. Press  Baltimore.

Gillette, L N. 1980. Movement patterns ofradio-tagged opossums in Wisconsin Amer. Midi. Nat  104:1-12.
Harestad, A. S., and F. L  Bunnell. 1979. Home range and body weight - a reevaluation. Ecology 60:389-402.



194 BULLETIN FLORIDA MUSEUM NATURAL HISTORY VOL 38, PT. 11(6)

Hsu, M.. J. D. Harder, and S. I. Lustick 1988. Seasonal energetics of opossums (Didelphis virginiana) in Ohio.
Comp. Biochem Physiol. 9OA.441-443

Hunsaker, D. 1977. Ecology of New World marsupials. Pp. 95-156 in D. Hunsaker, ed. The biology of
marsupials. Academic Press, New York.

Jennrich. R. I., and F. B. Turner. 1969. Measurement of non-circular home range. J. Theor. Biol. 22:227-237.
Jones, C. A 1989. Microhabitat use by Podomys =flondamis in the high pine lands of Putnam Co., Florida

PAD.dissertation, Univ. Florida  Gainesville.
Kissell. & E.. and M. L Kennedy. 1992. Ecologic relationships of co·«curring populations of opossums

(Didelphis virginiana) and raccoons (Procyon lotor) in Tennessee. J. Mamm  73:808-813.
MeManus, J. J. 1969. Temperature regulation in the opossum, Didelphis marsupialis virginiana. J. Mamm.

50:550-557.
McManus, J. J. 1971. Activity ofcaptive opossums, Didelphis marsupialis virginiana. J. Marna 52:846-848.
Reiss, M. 1988. Scaling ofhome range size: Body size, metal»lic needs and ecology. Trends Ecol  Evol. 3:85-86.
Reynolds, H. C. 1952. Studiesonthe reproduction in the opossum (Didelphis virginiana). Univ. California Publ.

Zool. 52:223-284
Rubenstein, D. I., and R. W. Wrangham, eds. 1986. Ecological aspects of social evolution. Princeton Univ. Press

Princeton,.New Jersey.
Ryser, J. 1990. The mating system, ecology, and biology of the Virginia opossum, Didelphis virginiana, in

north-central Florida Ph D. thesis, Univ. Bern, Switzerland.
Ryser, J. 1992. The r-nting system and male mating success ofthe Virginia opossum (Didelphis virginiana) in

Florida. J. Zool. 228:127-139.
Scholander, P. F., V. Walters, R. Hock, and L Irving- 1950. Body insulation of some arctic and tropical

mammals and birds. Biol. Bull. 99:225-236
Seidensticker, J.. M. A O'Connell and A. J. T. Johnsingh. 1987. Virginia opossum. Pp. 246-261 in M. Novak. J.

A Baker, M. E. Obbard and B. Malloch, eds. Wild Furbearer Management and Conservation in North
America. Ministry Nat. Res., Ontario, Canada

Spencer, W. D., and R. H. Barrett. 1984. An evaluation ofthe hannonic mean measure for defining carnivore
activity areas. Acta Zool. Fenn. 171:255-259.

Sunquist, M. E., S. N. Austal and F. C. Sunquist 1987. Movement patterns and home range in the common
opossum (Didelphis marsupialls). J. Mamm. 68:173-176.

SunquisL M. E., and J. F. Eisenber* 1993. Reproductive strategies of female Didelphis. Bull. Florida Mus. Nat
Hist, Biol. Sci. 36(4):109-140

Trivers, R L 1972. Parental investment and sexual selection. Pp. 136-179 in B. Campell, ed- Sexual selection
and the descent of'man, 1871-1971. Aldine, Chicago.

Wrangham, R. W., and D. I. Rubenstein. 1986. Social evolution in birds and mammals. Pp. 452-470 in D. I.
Rubenstein, and K W. Wrangham, eds. Ecological aspects of social evolution. Princeton Univ. Press
Princeton, New Jersey.

Wright, D. D. 1989. Mortality and dispersal of juvenile opossums, Didelphis virginiana. M.S. thesis, Univ.
Florida, Gainesville.