Behavioral Ecology Vol. 12 No. 5: 626-632
© 2001 International Society for Behavioral Ecology
Territoriality and male reproductive success in arctic ground squirrels
Museum of Zoology and Department of Biology, The University of Michigan, Ann Arbor, MI 48109, USA
Both authors are now at the Museum of Vertebrate Zoology and Department of Integrative Biology, University of California, Berkeley, CA 94720. Address correspondence to E.A. Lacey. E-mail: ealacey{at}socrates.berkeley.edu .
Received 18 April 2000; revised 8 October 2000; accepted 18 January 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although territorial defense is a common form of reproductive competition among male vertebrates, the exact reproductive consequences of this behavior are often poorly understood. To explore relationships between territoriality and reproductive success in a nongroup-living mammal, we quantified patterns of space use, mating success, and fertilization success for males in a free-living population of arctic ground squirrels (Spermophilus parryii plesius). Because litters of this species are sired almost exclusively by a female's first mate, we predicted that territory ownership would be associated with first access to estrous females. During the 2-week period when mating occurred, each male in the study population attempted to defend a distinct portion of the habitat, although the success of this defense varied among individuals. Twenty-six of 28 females monitored mated with the male on whose territory they resided. However, the majority of females observed throughout estrus (65%; n = 20) also mated with at least one other male, indicating that territory ownership was not associated with exclusive access to females. In contrast, territory ownership was significantly associated with first access to estrous females; 20 (71.4%) of 28 females mated first with the male on whose territory they resided. In this regard, the behavior of S. parryii plesius parallels that of socially monogamous birds in which territorial defense by males functions to deter extrapair copulations by females. Although territorial defense represents an important component of male reproductive success in arctic ground squirrels, other aspects of male behavior (e.g., the ability to dominate agonistic interactions on the day of a female's estrus) are also critical. We suggest that future studies of vertebrate mating systems will benefit by viewing such defense as only one of multiple axes along which conspecific males compete for access to females.
Key words: extrapair copulation, ground squirrels, reproductive success, Spermophilus parryii plesius, territoriality.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Territorial defense by males is a conspicuous component of many vertebrate mating systems (Clutton-Brock, 1989
; Davies, 1991;
Ligon, 1999). For species in
which each male defends an area occupied by one or more females, such defense
is generally assumed to prevent reproductive competitors from gaining access
to those females (Clutton-Brock,
1989; Emlen and Oring,
1977), thereby increasing the copulatory success of the territory
owner relative to other males. Studies of sperm competition (e.g.,
Birkhead and Møller,
1992; Ginsberg and Huck,
1989; Møller and
Birkhead, 1989), however, suggest that not all copulations are
equally likely to result in offspring. As a result, both copulatory and
fertilization success must be considered when evaluating the adaptive
significance of male behavior, including territorial defense associated with
mating.
Relationships between territorial defense and male copulatory and
fertilization success have been examined for a number of avian species,
frequently in the context of extrapair copulations in socially monogamous
birds (e.g., Birkhead and Møller,
1992; Møller,
1990; Rodrigues,
1998). These studies have revealed that, in many species, male
defense of a territory does not preclude the female resident on that territory
from engaging in genetically effective copulations with other males
(Birkhead and Møller,
1992). The well documented occurrence of extrapair fertilization
suggests that male reproductive success must be influenced by more than just
territory ownership, raising the possibility that factors such as dominance
relationships among individual males or the reproductive interests of females
are also important determinants of male success.
In comparison, the reproductive consequences of male territorial defense
have received relatively little attention in mammals. Because social monogamy
is rare among mammalian species
(Clutton-Brock, 1989;
Kleiman, 1977), a parallel
literature on extrapair copulations has not developed. Although a number of
studies have examined patterns of male fertilization success in group-living
mammals in which adults of both sexes defend a common territory (e.g.,
black-tailed prairie dogs: Hoogland,
1995; European marmots:
Goossens et al., 1998; dwarf
mongooses; Keane et al.,
1994), fewer data are available for mating systems in which the
spatial distributions of opposite-sexed adults overlap but in which males and
females do not form explicit social bonds. Because the latter type of system
appears to be relatively common among mammals
(Clutton-Brock, 1989),
understanding patterns of copulatory and fertilization success in this context
is critical to understanding the adaptive significance of territorial defense
as a form of male reproductive competition.
The mating system of the arctic ground squirrel (Spermophilus parryii
plesius) provides an ideal opportunity to investigate the effects of
territorial defense on both male copulatory and fertilization success. During
the approximately 2-week-long period each year when mating occurs, male S.
parryii plesius defend territories on which multiple females reside
(McLean, 1983). Individual
females exhibit behavioral estrus on only one afternoon per year, and, while
sexually receptive, a female mates with one to four different males
(Lacey et al., 1997).
Paternity analyses based on allozyme and DNA finger-printing data indicate
that >90% of young are sired by the first male with which a female mates
(Lacey et al., 1997). This
pronounced bias in fertilization success does not vary with differences in
mating behavior (e.g., interval between successive copulations); instead, the
primary correlate of male fertilization success appears to be mating order
(Lacey et al., 1997). As a
result, males are expected to compete primarily for access to females that
have not yet mated. If territorial defense by male S. parryii plesius
functions as a form of reproductive competition, then this defense should be
associated with access to unmated, estrous females. Specifically, a male
should typically be the first to mate with the females resident on his
territory.
To test this prediction, we monitored the copulatory success of males in a
free-living population of S. parryii plesius. We observed individual
females throughout behavioral estrus and quantified mating order and
territorial status for each of a female's consorts. These data were then
compared to the results of paternity analyses
(Lacey et al., 1997) to
determine if territory ownership is correlated with fertilization success.
Here, we characterize the relationship between territorial defense, copulatory
success, and paternity in S. parryii plesius and relate these data to
patterns of male territorial defense in other mammals and in birds.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Study population
We studied a population of S. parryii plesius located in the Kluane Game Sanctuary, Yukon, Canada (60°47'N, 137°40'W, elevation 650 m). The study site consisted of an open meadow bounded to the east by Bear Creek and on all other sides by dense vegetation that was not occupied by ground squirrels. Detailed descriptions of the site are provided in Lacey (1991
) and Lacey et
al. (1997). Typically, the
5-ha study site was inhabited by 15-25 adult male and 50-55 adult female
ground squirrels, for a density of 13-16 adults per hectare. We trapped members of the study population using National or Tomahawk live-traps (16.5 x 16.5 x 48.0 cm) baited with peanut butter. All ground squirrels residing on the study site were permanently marked with two numbered ear tags (Monel #1, National Band and Tag Company). To allow visual recognition of individuals during behavioral observations, each animal was also uniquely marked with black hair dye (Lady Clairol #124 natural blue-black). We conducted behavioral observations from 3-m high observation stands (n = 9) positioned around the perimeter of the study site. To facilitate data collection, the study site was gridded into 20-m squares, the corners of which were marked with orange surveyors' flags labeled with a Cartesian coordinate system.
We monitored territorial and reproductive behavior by males during 24
April-3 May 1988, 20 April-11 May 1989, and 23 April-1 May 1990. Animals in
the study population began emerging from hibernation in mid-April, with the
first males typically emerging 7-10 days before the first females
(Lacey, 1991). Because male
emergence was not synchronous, we did not begin recording behavioral data
until approximately 1 week after emergence began, at which point most males
had become active above ground. Females became sexually receptive on their
third or fourth day above ground. Although individual females exhibited
behavioral estrus on only 1 day of the active season
(Lacey et al., 1997), mating
activity in the study population continued for a period of about 2 weeks.
Male territorial defense
Definitions of territoriality frequently include both spatial and social
criteria for territorial defense (reviewed by Kaufman, 1983;
Maher and Lott, 1995). More
specifically, definitions of territoriality often stipulate that individuals
must (1) occupy at least partially non-overlapping portions of the habitat and
(2) display aggressive defense of those areas (e.g.,
Jarman, 1974;
Meier, 1992;
Ostfeld, 1985). Because male
S. parryii plesius have been characterized as territorial during the
mating period (McLean, 1983),
we used these criteria as a starting point for analyses of reproductive
competition among males in our study population. To document the spatial
distribution of males during the mating period, we used scan sampling
(Altmann, 1974) to delineate
the areas occupied by animals resident on the study site during 1988 and 1989.
To reduce the distance over which animals were observed, we divided the study
site into four non-overlapping sections, the boundaries of which were
determined largely by the shape of the open meadow in which the ground
squirrels resided. In general, two scan sampling periods per section were
completed daily, typically between 1200 and 1900 h (0700-1200 h was used to
assess the reproductive status of females;
Lacey et al., 1997); a minimum
of 2 h was allowed between successive sampling periods in the same section.
During each sampling period, the locations of all animals (male and female)
visible in that section of the study site were determined at 5-min intervals
for a total of 25 min (n = 6 samples). Locations were recorded to the
nearest meter; trials conducted using objects placed at known locations
revealed this procedure to be accurate to within 2.5 m.
We conducted scan sampling on all days during the approximately 2-week-long
period in which mating occurred. Because spatial relationships among males in
other ground squirrel species are influenced by the presence of estrous
females (e.g., Michener and McLean,
1996), the area occupied by each male in the study population was
determined using only data collected on days when no females were in estrus.
Specifically, for each section of the study site, we divided scan sampling
sessions into those conducted on days when no females in that section were in
estrus versus days when at least one female in that section was in estrus.
Analyses of male spatial relationships on nonestrous days were based on data
collected during 9.3 ± 4.9 (range = 4-20) 25-min scan periods
distributed over 5.8 ± 2.5 (range = 3-12) days per male (n =
16 males).
For each male, we included all positional data recorded on nonestrous days
in spatial analyses to maximize the probability of detecting spatial overlap
between the sexes (see below). Analyses of positional data for five males
resident on the study site in 1988 indicated that using a subset of positional
data records (e.g., only first or last records from a 25-min observation
period) reduced estimates of the area occupied by a male by a mean of 33.1
± 30.5%, leading to underestimation of the degree of overlap between
opposite-sex individuals. Because no more than six positions per individual
were recorded during each scan session and because numerous sampling periods
were conducted over multiple days, we assumed that temporal dependence of
positional data for an individual was limited. To prevent extreme outliers
(e.g., occasional excursions from the home area) from biasing spatial
relationships, we excluded the 5% of data points farthest from an animal's
center of activity (arithmetic mean x and y coordinates)
from our analyses. We analyzed spatial data using the minimum convex polygon
option in the Ranges V software package
(Kenward and Hodder, 1996);
this method of quantifying the area used by an animal is the least sensitive
to interdependence of successive data points
(Hundertmark, 1997). Estimates
of both the number of square meters occupied by a male and the percentage of
overlap between areas occupied by different males were also generated using
Ranges V.
We did not conduct scan sampling of male locations during the 1990 field season due to a shortage of observers. Instead, we estimated the area occupied by a male using the locations at which males were captured during the period between emergence from hibernation and the end of mating activity (4.7 ± 1.2 locations per male, median = 4, range = 3-6 trap records, n = 10 males). To assess the accuracy of this procedure, we compared trapping locations for males resident on the study site during 1988 and 1989 to minimum convex polygons for the same animals generated from scan sampling data. In all cases, trapping locations for an individual fell within the boundaries of that animal's area of activity as determined from scan data. Because trapping data tended to underestimate an individual's area of activity, spatial data from 1990 were used only to assign females to male territories for analyses of male reproductive success (see below); these data were not used to quantify spatial relationships among males on the study site.
We characterized social relationships among males using all-occurrence
sampling (Altmann, 1974) of
male-male encounters during the mating period. Specifically, we were
interested in determining if males actively defended the areas on which they
were resident. For all interactions observed, we recorded the identity of each
participant and the location of the encounter. In addition, we recorded all
occurrences of the following agonistic behaviors
(Lacey, 1991;
McLean, 1983): chases, fights,
parallel runs, and squeal displays. For each interaction, we scored the
participants as "dominant," "subordinate," or
"undetermined" based on the behaviors observed. For example,
individuals that were chased from an encounter site or that fled an encounter
after a fight were considered subordinate. Conversely, individuals that chased
conspecifics or that remained at the location of a fight were scored as
dominant. Not all interactions generated clearly identifiable outcomes;
relationships among males participating in these encounters were scored as
undetermined.
Male-female associations
To determine if spatial relationships among males were associated with
access to unmated females, it was first necessary to identify which females
were resident in each male's area of activity. Although individual females
visited areas occupied by several different males, each female appeared to be
most closely associated with a single male. To quantify this association, we
used both scan sampling data and the locations of females' burrows to assign
individuals to specific male areas. For data from 1988 and 1989, we calculated
the center of activity (arithmetic mean x and y coordinates)
for each female from positional data obtained from scan sampling conducted on
pre- and postestrous days (6.8 ± 5.5 postestrous days per female, range
= 0-15, n = 30 females, data from 1988 and 1989 combined); these
values were superimposed onto maps of male spatial distributions, and each
female was assigned to the male on whose area her center of activity was
located. The location of each female's burrow was then superimposed onto male
distribution maps, and this position was used to assign individuals to male
areas. Comparisons of spatial relationships generated by these procedures
revealed no discrepancies between assignments based on scan sampling data and
those based on burrow locations (n = 38 females). No scan sampling
data were obtained in 1990 (see above), and thus only the locations of
females' burrows were used to assign individuals (n = 18 females) to
male territories for that year.
Male copulatory success
A detailed description of the procedures used to document mating behavior
is given in Lacey et al.
(1997). Briefly, we captured
pre-estrous females each morning and examined their external genitalia to
determine which individuals were likely to be sexually receptive later that
day. We observed sexually receptive females continuously during the afternoon
and evening of estrus; observations typically began at 1400 h (about 3 h
before a female's first copulation) and continued until all focal females had
entered their home burrows for the night (at about 2100 h). Because arctic
ground squirrels frequently mate underground, copulations were detected using
five behavioral criteria developed during observations of above-ground matings
(see also Hoogland, 1995;
Hoogland and Foltz, 1982;
Sherman, 1989). By monitoring
individual females throughout behavioral estrus, we were able to determine the
identity of each of a female's mates, as well as the order in which each male
copulated with a given female.
Data analysis
Behavioral and spatial data from only a single estrous period per female
were included in this study; as a result, each female represented an
independent sample of reproductive behavior. In contrast, because some males
were observed mating with more than one of the females included in this study,
assumptions of statistical independence among estrous periods may not have
been strictly met. Previous work by Lacey et al.
(1997), however, has revealed
no reproductively significant variation in male copulatory behavior,
suggesting that analyses of behavioral data were not confounded by the
identity of a female's mate(s). Because small sample sizes for most analyses
precluded accurate determination of whether data were normally distributed
(Wilkinson, 1986), we used
nonparametric procedures to examine spatial and social relationships among
individuals. Statistical analyses were performed using Statistica 5.1
(StatSoft, Inc.). Throughout the text, means are reported ±1 SD.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Spatial relationships among males
Spatial distributions were determined for 16 males (n = 5 in 1998; n = 11 in 1989) based on scan sampling data collected on nonestrous days during the mating period. The mean number of visual fixes used to map spatial relationships among these animals was 49.0 ± 23.9 (range = 23-82) in 1988 and 26.5 ± 13.0 (range = 6-41) in 1989. Minimum convex polygons for these individuals indicated that, during the mating period, each male in the study population occupied a discrete portion of the study site, the majority of which was not shared with other males (Figure 1). The mean area occupied by an individual was 1415.3 ± 730.8 m2 (range = 488.8-2565.2 m2) in 1988 and 573.3 ± 319.5 m2 (range = 103.1-1151.2 m2) in 1989. This difference between years was significant (F4,10 = 10.8, two-tailed p =.010). The mean percent overlap between areas occupied by adjacent males was 14.6 ± 16.7 (range = 0.0-46.6%) in 1988 and 7.7 ± 19.6 (range = 0.0-89.2%) in 1989; this difference was not significant (F1,29 = 0.775, two-tailed p =.386). The data from 1989 included one male whose area of activity was almost completely encompassed by that of an adjacent male (Figure 1b); when this conspicuous outlier was removed, the mean percent overlap between adjacent males was 4.0 ± 8.6 (range = 0.0-28.8%). This difference between years was significant (F1,28 = 5.21, two-tailed p =.030). The decrease in the mean area occupied by a male from 1988 to 1989 was associated with a nearly twofold increase in the density of males and females on the study site. Based on these data, it is unclear whether males reduced the area over which they were active in response to the increased density of reproductive competitors or the increased density of potential mates. Marked variation in the number of females per male area (range = 1-14; see below), however, suggests that males did not defend a fixed number of potential mates, implying that increased reproductive competition was responsible for the reduction in mean area occupied.
|
Social relationships among males
Social interactions among males appeared to be influenced more by the
identities of the interacting individuals than by the locations at which
encounters occurred. Of those interactions recorded on nonestrous days that
had identifiable outcomes (n = 44), the number dominated by the male
on whose area of activity the encounter occurred was not significantly
different from random (H0: equal numbers of encounters dominated on
and off the home area) in 1989 (n = 20 of 33 interactions;
2 = 1.48, df = 1, p >.10) or when data from both
years were combined (n = 23 of 44 interactions; 2 =
0.09, df = 1, p >.50); the small number of interactions with
identifiable outcomes during 1988 (n = 11) precluded independent
analysis of data from that year. Similar results were obtained when analyses
were restricted to interactions involving chases, which have been used to
characterize male dominance interactions in other ground squirrel species
(e.g., Davis and Murie, 1985;
Murie and Harris, 1978); the
number of these interactions that occurred on the chasing male's area of
activity was not significantly different from random (n = 20 of 43
interactions, data from both years combined, 2 = 0.21, df = 1,
p >.50). Collectively, these data suggest that males in the study
population did not exhibit site-specific dominance during interactions with
reproductive competitors.
In contrast, when the outcomes of agonistic interactions were examined as a
function of male identity (n = 8 males with >5 interactions each;
data from 1988 and 1989 combined), we found that four animals were
consistently scored as dominant, while four others were consistently scored as
subordinate (Figure 2). Each of
these males interacted with 3.3 ± 1.2 (range = 2-5) different
individuals, suggesting that the tendency to win or lose encounters was not
due to the dominance relationship between a specific dyad of males. This
tendency was even more pronounced among males that were consistently scored as
subordinate; these males interacted with 4.0 ± 1.2 (range = 3-5)
different individuals. For both dominant and subordinate individuals, the
number of interactions that occurred on the focal animal's area of activity
was not significantly different from random (H0: equal numbers of
interactions on and off the home area; 2 
0.44, df = 1,
p >.50). All consistently subordinate individuals engaged in
fights, parallel runs, or squeal displays (6.3 ± 4.0 agonistic
interactions during 13.8 ± 3.4 interactions), indicating that although
all males attempted to defend the area on which they were resident, the
success of this defense varied among individuals
(Figure 2). Thus, although the
behavior of males in the study population appeared to be generally consistent
with spatial and social criteria for territoriality, an individual's tendency
to consistently win or lose agonistic interactions suggested that access to
females may have been influenced by more subtle differences in male
competitive ability.
|
Spatial distribution of females
All burrows occupied by females were located on male territories, and,
conversely, each male's territory contained the burrow system of at least one
female (Figure 1). The mean
number of females per male area in 1988 was 5.4 ± 2.6 (range = 1-8,
n = 5 male areas); in 1989, the mean was 4.8 ± 3.9 females per
male area (range = 2-14, n = 11 male areas). This difference between
years was not significant (Mann-Whitney U = 22.0, n = 5,11,
p =.58). The number of females per male territory was not correlated
with territory size (r =.398, n = 11, p =.254; data
for 1989 only), suggesting that occupation of a larger portion of the habitat
did not increase the number of a male's potential mates.
Territoriality and male copulatory success
Copulatory records for 28 females (1989, n = 13; 1990, n
= 15) were used to test the hypothesis that territoriality was associated with
access to unmated females. Twenty of these females were monitored continuously
throughout behavioral estrus; although copulatory records for the remaining
eight females were not complete, these animals were observed continuously
through the end of their first mating. Because previous research
(Lacey et al., 1997) has
revealed that the number of mates per female did not differ among years, data
for 1989 and 1990 were combined for the following analyses.
Twenty-six (92.9%) of the 28 females monitored mated with the male on whose
territory they resided (n = 12 males); 20 (76.9%) of these females
mated first with the resident male, indicating that territory ownership was
significantly associated with access to unmated females (2 =
5.2, df = 1, p <.025). Males did not, however, have exclusive
access to the females resident on their territories; 13 (65.5%) of the 20
females monitored throughout estrus mated with at least one nonresident male,
and 51.1% of all copulations detected (n = 47) involved nonresident
males.
All first copulations by females occurred on the male territory on which they resided. On the afternoon of estrus, the resident male appeared to actively defend the receptive female(s) on his territory. During this period, the distance between the resident male's center of activity and that of the estrous female was significantly less than on nonestrous days (Wilcoxon signed-rank, z = 2.56, n = 11, p =.010). In contrast, the rate of agonistic interactions between the resident and other males was significantly greater on estrous days (estrous: 1.8 ± 0.7 interactions/h, range = 0.5-3.4, based on 3.8 ± 1.3 h of observation; nonestrous: 0.5 ± 0.4 interactions/h, range = 0.2-0.7, based on 11.8 ± 4.3 h of observation; Wilcoxon signed-rank, z = 2.7, n = 10, p =.007), with such interactions generally occurring within 10 m of the receptive female. Collectively, these observations suggest that males shifted from defense of an area to direct defense of resident females on afternoons when those females were in estrus.
First copulations with nonresident males (1989, n = 3; 1990,
n = 5) occurred when the resident male was actively displaced by a
neighboring animal. Several hours before the onset of behavioral estrus, the
neighboring male began making incursions onto the resident's territory; the
locations visited by the nonresident were outside of that animal's typical
area of activity. Agonistic interactions were observed whenever the
nonresident and resident male encountered each other; in this context, the
number of interactions dominated by non-residents (22 of 43 interactions) was
significantly greater than expected given the percentage of interactions
dominated by nonresidents on nonestrous days (33.3%, n = 63
interactions, 2 = 4.1, df = 1, p <.05).
Displacement of the resident male lasted only until the neighboring animal had copulated with the female; after copulation, the neighboring animal returned to the area in which he was usually active. All displacements of resident males occurred before a female's first copulation; this association between displacement of the resident male and mating order was significant (8 of 28 first copulations; 0 of 13 second copulations; Fisher's Exact test p =.03). Pairwise comparisons revealed no tendency for the body weights of displacing males to be greater than those of the residents that they displaced (Wilcoxon signed-rank, z = 0.14, n = 8, p =.889); insufficient data prevented similar comparisons of other male traits such as age or dominance status on nonestrous days.
Territoriality and fertilization success
Lacey et al. (1997)
presented data on paternity for 16 litters of arctic ground squirrels reared
by females that were observed continuously throughout behavioral estrus.
Fifteen of these litters were sired exclusively by the female's first mate,
with paternity of the 16th litter divided among the female's first and third
mates. Eleven of these females had mated first with the male on whose
territory they resided. The five remaining females (including the mother whose
litter was of mixed paternity) mated first with a neighboring male. The
percentage of young sired by a female's first mate did not differ between
these groups (Mann-Whitney U = 0, n = 5,11, p =
1.0), indicating that first copulations resulted in fertilization success for
either resident or nonresident males.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The behavior of male S. parryii plesius during the mating period was consistent with traditional criteria for territoriality in that each male occupied a discrete portion of the habitat that he attempted to defend against incursions by reproductive competitors. Territory ownership was significantly associated with male reproductive success. Paternity analyses (Lacey et al., 1997
) indicated
that litters in the study population were sired almost exclusively by a
female's first mate; in the present study, more than two-thirds of the females
monitored during estrus mated first with the male on whose territory they
resided. Territoriality did not, however, provide males with exclusive access
to females; nearly half of the copulations detected involved nonresident
males, indicating that territory ownership did not prevent reproductive
competitors from gaining access to females. Instead, territoriality appeared
to increase fertilization success by allowing a male to mate first with the
females on his territory, thereby providing the resident male with the
copulations that were most likely to result in offspring. Territorial defense alone was not sufficient to prevent reproductive competitors from gaining access to sexually receptive females. On the afternoon of estrus, the resident male actively attempted to deter competitors from gaining access to the estrous female(s) on his territory. These efforts were not always successful, and about 28% of first copulations involved nonresident males that had succeeded in displacing the resident male, thereby gaining access to the receptive female. Observations of male-male interactions on nonestrous days revealed that individuals either consistently won or consistently lost agonistic encounters (Figure 2). Although our data did not allow us to assess the relative reproductive success of winners versus losers, we suspect that males that consistently lost agonistic encounters on nonestrous days were more likely to be displaced by neighboring animales on days when females were receptive. Thus, defense of a portion of the habitat may represent only one aspect of reproductive competition among males; layered on top of this defense may be more subtle differences in competitive ability that significantly affect reproductive success.
The role of female behavior in shaping reproductive competition among males
was not addressed in this study. The locations of females' burrows provided an
unambiguous means of assigning females to male areas, but individual females
did not restrict their activity to the area defended by a single male, and
most females interacted with multiple males before and on the day of estrus
(Lacey, 1991). Although
receptive females interacted most frequently with the male on whose area they
resided (Lacey, 1991), females
appeared to have the opportunity to assess different prospective mates and,
potentially, to influence the number and order of males with which they mated.
The behavior of female California and Richardson's ground squirrels has been
shown to change on the day of estrus in ways that appear to increase contact
with potential mates (e.g., increased area of female activity;
Boellstorff et al., 1994;
Michener and McLean, 1996),
suggesting that females may play a critical role in determining the copulatory
and fertilization success of males. The behavior of female S. parryii
plesius and its effects on male reproductive success will be examined in
greater detail in a future publication.
Interspecific variation in ground squirrel mating systems
The relationship between territorial defense and fertilization success
reported here has important implications for inter-specific differences in
ground squirrel mating systems. In particular, patterns of male fertilization
success may help explain why reproductive competition in some species of
Spermophilus is highly spatially defined (i.e., territorial;
Dobson, 1984;
Schwagmeyer, 1990), whereas
other species show no evidence of site-specific defense of potential mates.
Although both ecological and phylogenetic correlates for this behavioral
difference have been considered (e.g.,
Dobson, 1984;
Schwagmeyer, 1990), none has
been found to provide a robust explanation for the observed variation in male
site-specificity during the mating period.
Previous efforts to explain variation in ground squirrel mating systems
have not considered the roles of sperm precedence and differential
fertilization success in shaping patterns of reproductive competition.
Although multiple mating by females appears to be common among ground
squirrels (Boellstorff et al.,
1994; Murie, 1995;
Lacey et al., 1997;
Michener and McLean, 1996;
Schwagmeyer, 1990;
Sherman, 1989), patterns of
fertilization success vary markedly among species
(Foltz and Schwagmeyer, 1989;
Lacey et al., 1997;
Sherman, 1989). These
differences may contribute to interspecific variation in male behavior by
creating asymmetries in the reproductive benefits received from copulations.
For example, for species in which multiple paternity of litters is common
(e.g., S. beldingi: Hanken and
Sherman, 1981; S. tridecemlineatus:
Foltz and Schwagmeyer, 1989),
a comparatively large proportion of copulations will yield offspring, and thus
males should behave in ways that allow them to mate with as many females as
possible. In contrast, for species in which paternity is more restricted
(e.g., S. parryii plesius: Lacey
et al., 1997), males should adopt behaviors that increase their
chances of achieveing the few genetically effective copulations available;
increased site specificity of male activity, including defense of a specific
subset of females, may represent one response to this challenge.
This argument assumes that fertilization success is determined, at least in
part, by factors other than male behavior. This assumption seems to be met in
S. parryii plesius, in which variation in male copulatory behavior
has no effect on paternity (Lacey et al.,
1997). Differences in male behavior, however, have been shown to
influence fertilization success in other ground squirrel species
(Foltz and Schwagmeyer, 1989;
Schwagmeyer and Foltz, 1990;
Sherman, 1989), making it more
difficult to identify causal relationships between probability of
fertilization and copulatory behavior. The factors affecting sperm precedence
and fertilization success are complex, and detailed physiological studies of
ground squirrels are needed to determine precisely how paternity is controlled
(Davies, 1991). Nevertheless,
the striking differences in paternity and reproductive competition outlined
here suggest that sperm precedence may be an important determinant of male
behavior.
Parallels with extrapair fertilizations in birds
Relationships between territoriality and male fertilization success have
been examined for a number of other mammals, many of which are species in
which individuals form discrete social groups and adults of both sexes defend
a common territory (e.g., dwarf mongooses:
Rood, 1983; alpine marmots:
Goossens et al., 1998;
black-tailed prairie dogs: Hoogland, 1985; chimpanzees:
Goodall, 1986). In these
societies, reproductive success is often shared among the males within a
group, but extragroup copulations appear to be rare (e.g.,
Arnold, 1990;
Keane et al., 1994; but see
Goossens et al., 1998;
Hoogland, 1995;
Morin et al., 1994). Because
such groups are frequently composed of close relatives (e.g.,
Rood, 1983;
Hoogland, 1995;
Goosens et al., 1998), even
males that do not sire young may gain limited reproductive success via
inclusive fitness benefits. In contrast, male S. parryii plesius are
not group-living, and neighboring males are unlikely to be related
(Lacey, 1991). As a result,
young that are sired by reproductive competitors yield no fitness benefits to
resident males.
In this regard, reproductive competition among male S. parryii
plesius more closely resembles that in avian species in which individual
males actively attempt to deter extrapair copulations by the female(s)
residing on their territory (e.g., Birkhead
and Møller, 1992;
Langmore, 1996;
Tobias and Seddon, 2000).
Although arctic ground squirrels do not form a distinct intersexual bond
comparable to that found in many birds, males in both types of systems face
the common challenge of preventing genetically effective copulations by
reproductive competitors. Male reproductive success in a number of avian
species appears to be associated with territory ownership (e.g.,
Currie et al., 1999;
Gibbs et al., 1990;
Sundberg, 1994), although, as
in the present study, defense of a portion of the habitat is not always
effective in preventing nonresident males from siring young.
Several recent studies of male reproductive competition and fertilization
success in birds have partitioned the general phenomenon of territorial
defense into a number of more specific aspects of male behavior such as
patrolling territory boundaries, vocal advertisement of territory ownership,
and changes in territory size in response to female receptivity
(Langmore, 1996;
Rodrigues, 1998;
Tobias and Seddon, 2000). The
specific aspects of male behavior most closely associated with reproductive
success vary among species (e.g., Gil et
al., 1999; cf. Tobias and
Seddon, 2000), but one common theme of these studies is that
traditional views of mating territories as fixed entities capable of deterring
reproductive competitors are overly simplistic. Clearly, defense of a portion
of the habitat represents only one of the suite of behaviors that males in
these species use to prevent reproductive competitors from gaining access to
receptive females. Our findings from arctic ground squirrels suggest that the
same is true of territoriality in mammalian species. Thus, rather than
regarding territoriality as a comprehensive description of male reproductive
behavior, future studies of avian and mammalian mating systems will benefit by
viewing territorial defense as only one of the many axes along which males
compete for access to reproductive partners.
|
ACKNOWLEDGEMENTS |
|---|
Permission to conduct field work in the Kluane Game Sanctuary was provided by the Heritage Branch, Yukon Territorial Government, by the Yukon Department of Renewable Resources, and by the Champagne-Aishihik Band. In the field, we were assisted by R. Cohen, J. Terbruggen, M. Plantinga, K. Barabas, C. Wieczorek, and E. McNamara. Financial support was provided by a National Science Foundation Predoctoral Fellowship awarded to E.A.L. and by research grants from the American Museum of Natural History, Sigma Xi, the American Society of Mammalogists, and the Museum of Zoology, Department of Biology, and Rackham School of Graduate Studies at The University of Michigan. Previous versions of the manuscript benefited greatly from comments by J. Murie, D. Westneat, and an anonymous reviewer.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Altmann JA, 1974. Observational study of behavior: sampling methods. Behaviour 48: 227-265.
Arnold W, 1990. The evolution of marmot sociality. I. Why disperse late? Behav Ecol Sociobiol 27: 229-237.
Birkhead TR, Møller AP, 1992. Sperm competition in birds: evolutionary causes and consequences. London: Academic Press.
Boellstorff DE, Owings DH, Penedo MCT, Hersek MJ, 1994. Reproductive behaviour and multiple paternity of California ground squirrels. Anim Behav 47: 1057-1064.
Clutton-Brock TH, 1989. Mammalian mating systems. Proc R Soc Lond B 236: 339-372.[Medline]
Currie DR, Krupa AP, Burke T, Thompson DBA, 1999. The effect of experimental male removals on extrapair paternity in the wheatear Oenanthe oenanthe. Anim Behav 57: 145-152.[Web of Science][Medline]
Davies NB, 1991. Mating systems. In: Behavioural ecology: an evolutionary approach (Krebs JR, Davies NB, eds). Oxford: Blackwell Scientific; 263-294.
Davis LS, Murie JO, 1985. Male territoriality and the mating system of Richardson's ground squirrels (Spermophilus richardsonii). J Mammal 66: 268-279.
Dobson FS, 1984. Environmental influences on sciurid mating systems. In: The biology of ground-dwelling squirrels (Murie JO, Michener GR, eds). Lincoln: University of Nebraska Press; 227-249.
Emlen ST, Oring LW, 1977. Ecology, sexual selection,
and the evolution of mating systems. Science
197: 215-223.
Foltz DW, Schwagmeyer PL, 1989. Sperm competition in the thirteen-lined ground squirrel: differential fertilization success under field conditions. Am Nat 133: 257-265.
Gibbs HL, Weatherhead PJ, Boag PT, White BN, Tabak LM, Hoysak DJ,
1990. Realized reproductive success of polygynous red-winged
blackbirds revealed by DNA markers. Science
250: 1394-1397.
Gil D, Graves JA, Slater PJB, 1999. Seasonal patterns of singing in the willow warbler: evidence against the fertility announcement hypothesis. Anim Behav 58: 995-1000.[Web of Science][Medline]
Ginsberg JR, Huck UW, 1989. Sperm competition in mammals. Trends Ecol Evol 4: 74-79.
Goodall J, 1986. The chimpanzees of Gombe: patterns of behavior. Cambridge, Massachusetts: Belknap Press.
Goossens B, Graziani L, Waits LP, Farand E, Magnolon S, Coulon J, Bel M-C, Taberlet P, Allaine D, 1998. Extra-pair paternity in the monogamous Alpine marmot revealed by nuclear DNA microsatellite analysis. Behav Ecol Sociobiol 43: 281-288.
Hanken J, Sherman PW, 1981. Multiple paternity in
Belding's ground squirrel litters. Science
212: 351-353.
Hoogland JL, 1995. The black-tailed prairie dog: social life of a burrowing mammal. Chicago: University of Chicago Press.
Hoogland JL, Foltz DW, 1982. Variation in male and female reproductive success in a harem-polygynous mammal, the black-tailed prairie dog (Sciuridae: Cynomys ludovicianus). Behav Ecol Sociobiol 11: 155-163.
Hundertmark KJ, 1997. Home range, dispersal and migration. In: Ecology and management of the North American moose (Franzmann AW, Schwartz CC, eds). Washington, DC: Smithsonian Institution Press; 303-336.
Jarman PJ, 1974. The social organisation of antelope in relation to their ecology. Behaviour 48: 215-266.[Web of Science]
Kaufmann JH, 1983. On the definitions and functions of dominance and territoriality. Biol Rev 58: 1-20.
Keane B, Waser PM, Creel SR, Creel NM, Elliott LF, Minchella DJ, 1994. Subordinate reproduction in dwarf mongooses. Anim Behav 47: 65-75.
Kenward RE, Hodder KH, 1996. Ranges V: an analysis system for biological location data. Wareham, Dorset, UK: Institute of Terrestrial Ecology.
Kleiman DG, 1977. Monogamy in mammals. Q Rev Biol 52: 39-69.[Medline]
Lacey EA, 1991. Reproductive and dispersal strategies of male Arctic ground squirrels (Spermophilus parryii plesius) (PhD dissertation). Ann Arbor: The University of Michigan.
Lacey EA, Wieczorek JR, Tucker PK, 1997. Male mating behaviour and patterns of sperm precedence in Arctic ground squirrels. Anim Behav 53: 767-779.
Langmore NE, 1996. Territoriality and song as flexible
paternity guards in dunnocks and alpine accentors. Behav Ecol
7: 183-188.
Ligon DJ, 1999. The evolution of avian breeding systems. Oxford: Oxford University Press.
Maher CR, Lott DF, 1995. Definitions of territoriality used in the study of variation in vertebrate spacing systems. Anim Behav 49: 1581-1597.
McLean, I. G, 1983. Paternal behaviour and killing of young in arctic ground squirrels. Anim Behav 31: 32-44.
Meier PT, 1992. Social organization of woodchucks (Marmota monax). Behav Ecol Sociobiol 31: 393-400.
Michener GR, McLean IG, 1996. Reproductive behaviour and operational sex ratio in Richardson's ground squirrels. Anim Behav 52: 743-758.
Møller AP, 1990. Changes in the size of avian breeding territories in relation to the nesting cycle. Anim Behav 40: 1070-1079.
Møller AP, Birkhead TR, 1989. Copulation behavior of mammals: evidence that sperm competition is widespread. Biol J Linn Soc 38: 119-131.
Morin PA, Wallis J, Moore JJ, Woodruff DA, 1994. Paternity exclusion in a community of wild chimpanzees using hypervariable simple sequence repeats. Mol Ecol 3: 469-478.[Medline]
Murie JO, 1995. Mating behavior of Columbian ground squirrels. I. Multiple mating by females and multiple paternity. Can J Zool 73: 1819-1826.
Murie JO, Harris MA, 1978. Territoriality and dominance in male Columbian ground squirrels (Spermophilus columbianus). Can J Zool 56: 2402-2412.
Ostfeld RS, 1985. Limiting resources and territoriality in microtine rodents. Am Nat 126: 1-15.
Rodrigues M, 1998. No relationship between territory size and the risk of cuckoldry in birds. Anim Behav 55: 915-923.[Web of Science][Medline]
Rood JP, 1983. The social system of the dwarf mongoose. In: Advances in the study of mammalian behavior (Eisenberg JF, Kleiman DG, eds). Special Publication 7 of the American Society of Mammalogists. Lawrence, KS: Allen Press; 454-488.
Schwagmeyer PL, 1990. Ground squirrel reproductive behavior and mating competition: a comparative perspective. In: Contemporary issues in comparative psychology (Dewsbury DA, ed). Sunderland, Massachusetts: Sinauer Associates; 175-196.
Schwagmeyer PL, Foltz DW, 1990. Factors affecting the outcome of sperm competition in thirteen-lined ground squirrels. Anim Behav 39: 156-162.
Sherman PW, 1989. Mate guarding as paternity insurance in Idaho ground squirrels. Nature 338: 418-420.[Medline]
Sundberg J, 1994. Paternity guarding in the yellowhammer Emberiza citrinella: a detention experiment. J Avian Biol 25: 135-141.
Tobias J, Seddon N, 2000. Territoriality as a paternity guard in the European robin, Erithacus rubecula. Anim Behav 60: 165-173.[Web of Science][Medline]
Wilkinson L, 1986. SYSTAT: The system for statistics. Evanston, IL: SYSTAT.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||