Behavioral Ecology Vol. 10 No. 2: 169-177
© 1999 International Society for Behavioral Ecology
A multivariate analysis of phenotype and paternity in male harbor seals, Phoca vitulina, at Sable Island, Nova Scotia
a Department of Biology, Dalhousie University, Halifax, Nova Scotia B3J 4J1, Canada b Marine Fish Division, Department of Fisheries and Oceans, Bedford Institute of Oceanography, Dartmouth, Nova Scotia B2Y 4A2, Canada
Address correspondence to D. W. Coltman, who is currently at the Institute of Cell, Animal and Population Biology, Edinburgh University, Ashworth Laboratories, West Mains Road, Edinburgh, EH9 3JT, UK. E-mail: david.coltman{at}ed.ac.uk
Received 1 December 1997; revised 3 August 1998; accepted 31 August 1998.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Understanding the links between phenotype and reproductive success is critical to the study of the evolution of mating systems and life-history patterns. We examined the relationship between phenotype and mating success of male harbor seals (Phoca vitulina concolour) at Sable Island, Canada. Cluster analyses of eight traits including mating success determined by genetic paternity analysis, haul-out behavior, body mass, and mass change identified four groups of males with distinct characteristics. The most successful males were of moderate body size, were rarely sighted alone, were associated with many different groups on shore, and were sighted on the haul-out relatively infrequently. Large males that hauled out frequently alone, previously thought to be socially dominant, were less likely to be successful. Also less successful were smaller, younger males. Contrary to our hypotheses, and unlike most terrestrially breeding pinnipeds, body size and reproductive effort were not positively associated with mating success, and some successful males appeared to spend considerable time foraging in deep water.
Key words: body size, mating success, microsatellites, phenotype, pinniped, reproductive strategy, testosterone.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
In polygynous mammals, male reproductive success is largely determined by the number of offspring sired, and thus males compete intensely for mates, whereas females invest more heavily in parental care (Trivers, 1972
). Identifying
the sources of variation in male mating success is important to our
understanding of reproductive strategies and the evolution of sexually
dimorphic traits, as phenotypic characteristics that are associated with
mating success may be under strong sexual selection. Relatively few empirical
studies, however, have identified clear links between phenotype and male
mating success in natural populations of large mammals. The studies that have
done so usually focus on the importance of social dominance and body size in
polygynous systems (e.g., Bercovitch and
Nurnberg, 1996; Clutton-Brock
et al., 1982; Cowlishaw and
Dunbar, 1991; Owen-Smith,
1993).
There are three families of pinnipeds. Of the 18 species in the family
Phocidae, 15 species mate aquatically, as does the single species in the
Odobenidae. In contrast, all 15 species in the Otariidae generally mate on
land (Boness et al., 1993).
Among pinnipeds, male reproduction is well understood in the terrestrially
breeding species. When mating occurs on land, successful males may maximize
mating success by controlling access to females, such as in gray seals,
Halichoerus grypus (Boness and
James, 1979) and northern elephant seals, Mirounga
angustistrostris (Le Boeuf,
1974). Males in other species, such as Callorhinus
ursinus (Bartholomew and Hoel,
1953), may defend territories within the breeding colony where
females aggregate. Male body size is thought to be important in these species
for several reasons. In southern elephant seals, Mirounga leonina,
body size confers a fighting advantage that enables large males to maintain
high dominance status (McCann,
1981; Modig,
1996). In other species, large size enables males to prolong their
period of tenure through the effect of size on fasting endurance
(Lindstedt and Boyce, 1985).
Length of tenure has been correlated with mating success in some species
(Arctocephalus gazella; Arnould
and Duck, 1997). In gray and northern elephant seals, body size is
also correlated with reproductive effort, which in turn is associated with
male mating success (Deutsch et al.,
1990; Tinker et al.,
1995). Consequently, pronounced size dimorphism is thought to have
evolved in terrestrially breeding pinnipeds in response to intense sexual
selection on male body size for both competitive and energetic reasons
(Bartholomew, 1970).
Phocid seals that mate aquatically are less size dimorphic than Otarriid
seals and most other polygynous mammals
(Weckerly, 1998). It has been
suggested that body size is less likely to confer a competitive advantage in
the three-dimensional aquatic environment
(Bartholomew, 1970). There are
scant empirical data on male mating success in aquatically mating phocids to
test this idea, however. As phocid seals represent more than half of extant
pinniped species, this represents a major gap in our understanding of male
reproduction in pinnipeds. On a more general level, the importance of body
size to male mating success in these taxa is of basic interest to zoologists,
as phocid seals are at the bottom end of a continuum of sexual dimorphism in
large mammals (Weckerly,
1998). Aquatically mating pinnipeds compete for mates in a medium
in which there is comparatively less potential for males to monopolize
females, or resources important to females, which predicts reduced polygyny
and potential for sexual selection (Emlen
and Oring, 1977).
In this study, we set out to identify the phenotypic traits associated with
male mating success in harbor seals (Phoca vitulina concolour). The
harbor seal is a relatively small-bodied, slightly dimorphic phocid species
(males are approximately 7% longer and 20% heavier than postpartum females
(Bowen WD, unpublished data; McLaren,
1993) that breeds on isolated islands or beaches along the eastern
coast of North America (Boulva and
McLaren, 1979). At Sable Island, females come ashore from the
middle of May to early June and wean their pups after an average of 24 days of
lactation (Muelbert and Bowen,
1993). Females commence regular foraging trips in mid-lactation
(Boness et al., 1994;
Thompson et al., 1994) as they
become energetically depleted (Bowen et
al., 1992). Mating occurs exclusively at sea and is believed to
occur in late lactation or shortly following weaning; however, copulations
have rarely been witnessed (Thompson,
1988). Males compete intensely for mates: evidence of fighting,
lacerations about the head and hindflippers, is seen exclusively during the
breeding season (Boulva and McLaren,
1979; Godsell,
1988; Thompson,
1988; Walker and Bowen,
1993). At Sable Island, Coltman et al.
(1998b) measured male mating
success by paternity analysis using microsatellites and found a low variance
in the distribution of paternities. They suggested that the low level of
polygyny seen in harbor seals was probably due to the inability of males to
monopolize females or predict their movements at sea. Previous research on
male reproductive behavior has suggested that social dominance and body size
may be important characteristics.
At Miquelon, Perry (1993)
found three males that defended aquatic territories through which females had
to pass to gain access to the sea sired at least one pup, suggesting that
mating may occur near shore with socially dominant males. At Sable Island,
Walker and Bowen (1993)
identified a group of males that were heavily wounded from fighting with other
males, usually hauled out alone, and exhibited fidelity to a single location
on the beach. They suggested that these males were socially dominant, an idea
also suggested by Boulva and McLaren
(1979), and that they may have
been reproductively successful by concentrating their aquatic activity in
nearby home ranges arrayed in parallel to the beach. By doing so, males were
thought to intercept females as they departed and returned from foraging trips
in late lactation. Patterns of onshore behavior may therefore provide clues as
to which males are socially dominant and/or reproductively successful.
Variation in circulating levels of serum testosterone may also provide
information on individual reproductive and dominance status. Elevated levels
of testosterone induce seasonal changes in reproductive physiology and
behavior (Crews and Moore,
1986; Lincoln,
1989), but circulating levels may also be a consequence of
behavior, particularly aggression
(Harding, 1981;
Wingfield et al., 1990). In
some species, it is thought that high serum testosterone levels may be
maintained in response to winning social "challenges" or other
aggressive behaviors (e.g., Harding,
1981; Wingfield et al.,
1990). Other studies have demonstrated positive correlations
between serum testosterone levels and social dominance in field studies of
mammals (e.g., Petaurus breviceps:
Stoddart et al., 1994; Mus
musculus: Zielinski and Vandenbergh,
1993). Bartsh et al.
(1992) reported that dominant
male Weddell seals maintained higher testosterone levels through the breeding
season than defeated or noncompetitive males. We hypothesized that
reproductively successful male harbor seals were socially dominant and
therefore had higher levels of serum testosterone than other males.
Finally, we sought to investigate the influence of body size on male mating
success from an energetic standpoint. Because harbor seals are slightly size
dimorphic, body size could influence male mating success. In all animals there
is a trade-off between feeding (or the maintenance of body condition) and
reproduction when such activities are mutually exclusive
(Lindstrom, 1998;
Roff, 1992), which may
ultimately influence mating success. For example, Berkovitch and Nurnberg
(1996) found that paternity in
Macaca mulatta was modulated by body condition because males with
more body fat fed less and invested more effort toward reproduction. We
previously hypothesized that body size and the quantity of stored energy may
influence male mating success by constraining male reproductive effort rather
than by conferring a direct competitive benefit
(Coltman et al., 1997,
1998a). Relatively large
males, or males in better condition at the beginning of the breeding season,
may have an advantage because they can afford to spend more time making
shallow dives (Coltman et al.,
1997), which are commonly associated with reproductive behavior
(Boness et al., unpublished data). Most harbor seals are unable to fast for
the duration of the breeding season
(Coltman et al., 1998a) and
exhibit bouts of diving behavior consistent with offshore feeding trips during
the breeding season (Coltman et al.,
1997). Thus larger males may feed less, be able to afford to
expend more total energy on reproduction, and therefore may have higher
encounter rates with potentially receptive females.
In this study, we used multivariate statistical techniques to investigate
the relationships between paternity determined genetically
(Coltman et al., 1998b) and
phenotypic traits of male harbor seals. We hypothesized a mating advantage for
large males on energetic grounds and hypothesized that successful males
exhibit behavioral and endocrine patterns which have previously been thought
to reflect social dominance.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Study site and data collection
The study was conducted during the breeding seasons (mid-May to the end of June) of 1993 and 1994 on the north beach of Sable Island (43°55' N; 60°00' W), a partially vegetated sandbar 160 km east of Nova Scotia. A total of 385 and 270 harbor seal females gave birth on both the north and south beaches of Sable Island in 1993 and 1994, respectively. Roughly 70% of the pups were born on the north beach in each year. Males and females show strong fidelity to their breeding beach both within (Godsell, 1988
) and between
seasons (Bowen WD, unpublished data).
We captured males in nets as described in Bowen et al.
(1992) near the beginning of
the breeding season (from 20 May to 1 June), at which time we started to
collect data on social behavior. In both 1993 and 1994, more than 90% of all
births on the north beach occurred between 20 May and 4 June. Seals were
individually marked with fluorescent paint on the lower back to facilitate
rapid identification during daily beach surveys, and two numbered rototags
(Dalton, Henley on Thames) were applied to the hindflippers for permanent
identification. We attempted to recapture each male near the end of the
breeding season (range 20 June—5 July). At each capture males were
weighed with a 200 ± 0.5 kg Salter spring-balance. A dorsal length
measurement (i.e., a straight line from tip of the nose to the tip of the
tail; McLaren, 1993) was taken
on the subset of males used in concurrent studies of diving behavior
(Coltman et al., 1997) and
energetics (Coltman et al.,
1998a). We sedated males (approximately 0.2 mg diazepam per kg
body mass) so that a reliable length measurement could be obtained. Body size
measurements were also taken from a selection of sedated, known-age males
handled during the course of other research to estimate size—age
relationships.
We conducted surveys of the north beach daily at approximately 1700 h to
locate marked males. Locations of hauledout males were recorded relative to
numbered stakes placed at 0.5-km intervals along the north beach, and social
context was noted (either sighted alone or in a group of seals). At Sable
Island, most harbor seals are distributed in groups of 1—31 individuals
along the 25-km section of north beach study area. For the purpose of this
study, we defined a group as a cluster of animals separated from the next
cluster by 100 m or more. Groups are typically separated by distances of 0.5
km or more and tended to form at predictable beach locations within seasons,
although the membership of these groups did change
(Schaeff et al., 1999). At
each capture and during daily beach surveys, we also recorded the presence of
fresh wounds on the neck and hindflippers of each male. We stopped collecting
observational data during the first week of July.
All males were included in a study of paternity conducted using
microsatellites (Coltman et al.,
1998a). A likelihood approach was used to assign the paternity of
all pups born on the north beach in the following year to candidate males
using a simulation program to estimate the degree of statistical confidence in
paternity assignment (Marshall et al.,
1998). We used data from paternities assigned at 50% and 80%
confidence in this study.
Testosterone assay
We took blood samples for the testosterone assay from a subset of the
males. At each capture, 10 ml of whole blood was taken from the extradural
vein as quickly as possible, usually within 5 min of capture, to avoid
possible antagonistic effects of handling stress on circulating hormone
levels. Blood samples were stored without preservative on ice in a cooler
until they were processed at the end of the day. Serum was separated by
centrifugation of whole blood for 30 min at 3500 rpm and stored at
—20°C until assayed.
Assays were performed within 4 months of collection in 1993 and 1994. Total serum testosterone was determined using a commercially available radioimmunoassay system (Testosterone/dihydrotestosterone 3H Assay System, Amersham) following the manufacturer's instructions. According to the manufacturer, the only steroid having significant cross-reactivity to the antiserum is dihydrotestosterone (40—50%), hence the measurement of circulating androgen presented here represents total testosterone plus 40—50% of circulating dihydrotestosterone. In most other mammals, levels of circulating dihydrotestosterone are low relative to testosterone (Testosterone/dihydrotestosterone 3H Assay System manual, Amersham), but they have never been measured separately in a pinniped. We thus refer to the measured hormone levels simply as testosterone for simplicity.
Steroid hormones were separated by extracting serum diluted 1:4 in distilled water with diethyl ether. The mean extraction recovery of testosterone, estimated from adding serial dilutions of 3H-labeled testosterone standard to pooled harbor seal serum, was 92.7% ± 17.9% (n = 12). Recoveries indicated that testosterone could be measured accurately in harbor seal serum using this system [testosterone = 1.3(3H-labeled added standard) — 0.12; r2 = 0.93, n = 12, p <.001). The manufacturer reports the sensitivity of the assay to be approximately 3 pg/tube. Intra- and interassay coefficients of variation were estimated to be less than 10% and 17%, (n = 10 and 6, respectively). For statistical comparisons, serum testosterone measurements were blocked into four equal time periods according the date they were taken, and differences between groups of seals and over time were tested using repeated-measures ANOVA. If a male had more than one testosterone measurement within a particular time period, the average serum testosterone concentration was used in the ANOVA model, and males lacking an observation in any time period were omitted from the statistical analysis.
Phenotypic traits and data analysis
The following eight traits describing social behavior, movement, body size
and energetics, and mating success were used in a multivariate analysis to
classify males into groups of similar phenotypes. There were:
- Social behavior:
- Proportion of times sighted alone,
- Presence or absence of wounds indicating inter-male fighting,
- Number of different groups in which observed.
Movement:
- 4. Index of haul-out site fidelity (calculated as the standard deviation of
haul-out site locations),
5. Proportion of days sighted during daily beach surveys.
Body size and energetics:
- 6. Initial body mass (estimated mass on 20 May, calculated by correcting
the first mass measurement by the daily rate of mass change over the interval
between 20 May and the date of the actual first mass measurement),
7. Daily rate of mass change (final minus initial body mass, divided by the interval between measurements in days expressed as a percentage of initial body mass).
Mating success:
- 8. Paternity of one or more pups born in the following season at 80%
confidence.
We used cluster analysis to identify groups of males with similar phenotype using these eight traits. Before cluster analysis, continuous variables were tested for normality, transformed if necessary, and all data were standardized to a mean of 0 and standard deviation of 1 to ensure measurement on a similar scale and equal weighting. A hierarchical agglomerative cluster analysis of Euclidean distances was performed using Ward's linkage to initially identify data clusters. We then these clusters as the initial data partition for a second cluster analysis by the iterative k-means algorithm to determine final groupings. Inspection of the final data groups then identified traits that were associated with paternity. These traits were used as covariates in a logistic regression model to identify predictors of paternity.
In addition to the eight characteristics used in the cluster analysis, we
analyzed additional phenotypic data collected from subsets of study males.
These data included diving behavior from time-depth recorders (TDR; n
= 18; Coltman et al., 1997),
energetics from the analysis of stable isotope dilution (n = 22;
Coltman et al., 1998a),
circulating levels of testosterone (n = 58 measurements from 12 males
in 1993 and 96 measurements from 29 males in 1994), and body length/age
(n = 41). These data were not used in the cluster analysis due to the
reduced sample; thus group differences in age, diving behavior, energetics,
and testosterone were analyzed post-hoc. We also included data on paternities
assigned at 50% statistical confidence
(Coltman et al., 1998b) for
post-hoc comparison, however we used the more conservative (80% confidence)
data for cluster classification. We compared average individual
heterozygosity, calculated as the proportion of heterozygous loci, between
groups to ascertain if paternity assignments were biased toward less
heterozygous individuals.
All data were checked for normality and transformed if necessary before parametric statistical analyses. The significance of all statistical tests was set at.05. Statistical analyses were performed in MINITAB Release 10 for Windows (Minitab Inc.) or SPSS for Windows version 6.13 (SPSS Inc.).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Cluster analysis using Ward's linkage identified 4 major groups of individuals among the 68 males studied (39 in 1993 and 29 in 1994; Figure 1). Using these groups as the initial data partition, k-means clustering affirmed these group compositions with the exception of one individual in group C that was moved to group D. The defining characteristics of these groups are shown in Table 1.
|
|
Group A males most resemble those hypothesized to have been successful in a
similar multivariate analysis by Walker and Bowen
(1993); they were most
frequently sighted alone, showed evidence of fighting, exhibited the highest
degree of fidelity to their haul-out location, and associated with few groups.
These were also the largest males with the highest mean rate of mass loss.
Because mass loss is correlated to initial body mass in breeding male harbor
seals (Coltman et al., 1997),
we also compared rates of mass loss by ANCOVA to account for the influence of
initial body mass. Differences among groups in rates of mass loss were not
statistically significant when differences in initial body mass were accounted
for (F3,63 = 2.56; p =.06), thus group B males
lost mass at a similar rate relative to their initial body size as other
males. Like males of group A, males of group B also exhibited haul-out site
fidelity, showed evidence of fighting, and associated with few groups, yet
were sighted alone less frequently than group A males and were lighter.
Neither group A nor group B included a male with paternity at 80%
confidence.
Group C consisted of males that behaved quite differently from A or B males: they tended to be sighted more frequently, showed less fidelity to a single haul-out location, were rarely sighted alone, and associated with more groups. These males were the lightest, and they tended to be shorter. We estimated the age of 37 study males from length measurements (Table 1) by the relation between age and length shown in Figure 2. Tagging information provided the ages of another four study males. The median estimated age of group C (9.0 years, n = 7) was less than all other groups combined (median = 12.5 years, n = 34; Mann Whitney U test; U = 54.5; p <.05). Fewer group C males showed evidence of fighting than males in all other groups (X2 = 4.8, df = 1; p =.029). As with groups A and B, no males in this group were assigned the paternity of a pup born in the following season at 80% confidence.
|
Group D contained all males that had paternity at 80% confidence. Other characteristics of group D included low frequency of sightings, rarely sighted alone, associated with a moderate number of groups, and absence of strong haul-out site fidelity. Most group D males showed evidence of fighting and were of moderate body size. They were also of similar estimated age (median = 11.5 years; n = 3) as males of group A (14.0 years; n = 22) and B (12.0 years; n = 9). Group D males were assigned significantly more paternities than males of other groups if the confidence level for paternity assignment was relaxed to 50% (Table 1). However, males of group D did not differ significantly in average heterozygosity from other groups (Table 1).
There were significant differences among groups in total body energy content measured at the beginning of the breeding season (F3,19 = 6.70; p <.005). Group A males had significantly more total body energy than males from groups B and C (Figure 3). ANCOVA indicated that group differences in total body energy content could be explained by variation in body mass or general body size rather than variation in body composition (ANCOVA initial mass as covariate, mass effect: F1,18 = 48.7; p <.001, group effect: F3,18 = 1.30; p >.05). Group A males catabolized significantly more total body energy per day (measured as the average daily loss in total body energy) during the breeding season than other males (Figure 4; F3,19 = 6.08; p <.005). Group D males had the highest rate of food energy intake (Figure 4; F3,13 = 4.47; p <.05). Combining energy catabolized from body stores and energy derived from feeding, there was no significant difference between group means in the daily rate of total energy expenditure (Figure 4; F3,13 = 1.73; p >.05). There was also no difference among groups in mass-specific total energy expenditure (group A: 3.70 ± 0.36, n = 6; group B: 3.68 ± 0.21, n = 4; group C: 3.29 ± 0.33, n = 5; group D: 3.60 ± 0.21 W/kg, n = 2; F3,13 = 0.35; p >.05).
|
|
Neither time spent hauled out on shore
(Figure 5; following arcsine
square-root transformation, F3,15 = 0.58; p
>.05), nor time spent in shallow dives (<20 m; following arcsine
square-root transformation F3,15 = 2.77; p =.08)
derived from the analysis of TDR records
(Coltman et al., 1997)
differed significantly among groups. However, group A males spent less time
making deep dives (>20 m) than other groups (following arcsine square-root
transformation, F3,15 = 5.27; p <.05).
|
Levels of circulating serum testosterone varied greatly with time and among individuals, generally decreasing over the breeding season (Figure 6). Mean levels between 4 and 5 ng/ml were observed early in the breeding season, falling to approximately 1 ng/ml later in the breeding season when estrous females became increasingly available. Patterns were similar among groups (Figure 6), as mean levels decreased significantly with time (repeated measures ANOVA, F3,69 = 34.01; p < 0.001). Variation between groups was not significant overall (repeated measures ANOVA, F3,24 = 1.25; p >.05), nor were the effects of age, initial body mass, nor time of day sampled (data not shown). Compared against groups B, C and D combined, mean levels of serum testosterone among group A males declined to low levels earlier (repeated-measures ANOVA, group-by-time interaction; F9,72 = 4.05; p <.01).
|
The males of group D stood apart from all other groups in their frequency of sighting and the proportion of times sighted alone, suggesting these characteristics were associated with reproductive success. A logistic regression model incorporating both of these variables as covariates successfully predicted the mating success of 89.7% of all observations [paternity at 80% confidence as response variable: model X2 = 16.6, df = 2, p <.005; constant = 1.38; ß (times sighted) = —6.12, p <.05; ß (times sighted alone, following arcsine transformation) = —6.89, p <.05]. The probability of male mating success therefore increases with decreasing frequency of sightings on shore and with decreasing frequency of hauling out alone (Figure 7).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
One potential criticism of this study is that the genetic analysis had limited power to detect paternity. Few paternities could be assigned at the 80% confidence level due to the large number of candidate males considered for each pup (paternity was assigned for 4% of all sampled pups at 80% confidence, or for 30.9% of all pups at 50% confidence; Coltman et al. 1998b
).
Therefore, although most males assigned paternities at 80% confidence were
probably successful, we cannot conclude that males of other groups did not
sire offspring. Fourteen males from groups A, B, and C would have been
assigned paternity of at least one pup if the confidence level in paternity
assignment had been relaxed to 50%. However, even at this level of confidence
in paternity assignment, males of group D would be assigned the paternity of
significantly more offspring than males from other groups
(Table 1). Thus it seems
reasonable to conclude that group D males were relatively more successful than
other males based on the genetic data. We used the 80% confidence paternity
data in the multivariate analysis because they are more conservative, given
that our study objective was to identify traits associated with success.
One potential source of bias in paternity assignment exists that also could
influence these results. Paternity assignment based on maximum likelihood is
potentially statistically biased toward individuals that have a large number
of homozygous loci (Devlin et al.,
1988). However, we found no evidence that this occurred in our
data (Coltman et al., 1998b),
and there was no significant difference among groups in genetic variation,
measured as individual heterozygosity
(Table 1).
Group A males most clearly resemble the group of males described by Walker
and Bowen (1993) which were
hypothesized to be reproductively successful. These males frequently
hauled-out independently of groups at a particular beach site and showed
evidence of fighting; however, the genetic data suggest that they were
relatively unsuccessful. Group A also contained the largest males. We have
previously hypothesized a mating advantage for relatively large males or for
those males in good condition at the beginning of the season because large
males should be able to fast longer and thus spend more time on reproduction
(Coltman et al., 1997,
1998a). Indeed, data from TDRs
indicated that group A males spent less time making deep dives
(Figure 5), which are more
likely to be associated with foraging
(Coltman et al., 1997).
However, we found no evidence that large body size and the associated
differences in aquatic activity (Coltman
et al., 1997) were positively associated with mating success. By
contrast, data from the only other aquatically mating phocid studied to date
indicated that large, territorial Weddell seal males obtained more copulations
than smaller males (Bartsh et al.,
1992). However, observed copulations may not provide a good
measure of paternity. Other studies using genetic estimates of paternity in
pinnipeds (gray seals: Amos et al.,
1993, 1995; captive
harbor seals: Harris et al.,
1991) and many other vertebrates (e.g., birds:
Birkhead et al., 1990;
primates: Berard et al., 1993;
ungulates: Pemberton et al.,
1992) have found that traits thought to be predictors of mating
success were often unreliable.
One possible interpretation of the low mating success in group A males, and
perhaps in group B males (Table
1), that consistently hauled-out at the same site is that the wide
dispersal of females along the beach at Sable Island resulted in a low
probability of encountering females. Evidence for this suggestion comes from
the observation (Coltman,
1997) that males were not more likely to mate with females located
nearby onshore. An alternative explanation is that males that haul-out alone
during the breeding season may do so as a result of defeat in intrasexual
competition. Such changes in behavior have been observed in Weddell seals,
where males that failed to defend an underwater territory received wounds and
spent the majority of the ensuing breeding season hauled out on shore
(Bartsh et al., 1992). Bartsh
et al. (1992) also found that
defeated males showed a marked decline in serum testosterone coincident with
their withdrawal from competition, possibly resulting from the stress of
defeat. All harbor seals showed a decline in serum testosterone toward the end
of the breeding season, a typical pattern among male pinnipeds
(Atkinson and Gilmartin, 1992;
Bartsh et al., 1992;
Noonan et al., 1991). However,
group A males showed the most marked decline over the first three time blocks
(Figure 6). This resembles the
pattern seen in the testosterone profiles of defeated male Weddell seals
(Bartsh et al., 1992),
suggesting that some males in this group may have been defeated in intrasexual
competition. Another possibility is that the rapid decline in these males may
result from low encounter rates with estrous females
(Harding, 1981).
Group C and D males hauled out with more groups and at more sites than
males of group A and B. Group C may contain younger males than other groups,
as they were lighter, shorter, and had a lower incidence of lacerations
resulting from fighting than other males. Both mass and length continue to
increase after male harbor seals become reproductively mature
(McLaren, 1993;
Figure 2); and males thought to
be subadult have previously been shown to differ in their behavior during the
breeding season, in the way group C differs from other groups
(Walker and Bowen, 1993).
Walker and Bowen (1993) found
that subadult males rarely show evidence of wounding and show less site
fidelity than adult males. Both of these observations are consistent with the
differences between group C and other males
(Table 1). Nevertheless, group
C males are likely to be reproductively mature, as they showed similar levels
of circulating testosterone as other males
(Figure 6). Lack of breeding
experience may explain the relatively low success of group C.
Why were group D males more successful? Survey data suggest that they spent
more time at sea where the chances of encountering females must be enhanced.
In this regard, they behaved like the most successful Weddell seal males
(Bartsh et al., 1992). Our data
from TDRs do not support this suggestion, as group D males spent similar time
hauled out on shore as other males (Figure
5); however, this conclusion lacks power due to the small sample
size of successful males with diving records (n = 2). Furthermore,
time of day may be more relevant than total time at sea given the diurnal
diving patterns of females (Boness et al.,
1994). Group D males showed little site fidelity and associated
with more female groups on shore compared to groups A and B, suggesting that
the more successful males may encounter more females while out at sea, on
average, than other adult males. This may enable group D males to monitor the
breeding condition of a greater number of females on shore than males that
either remain with the same group or haul-out at the same site each day.
The mass change and energetic data did not suggest a positive association
between body size and/or reproductive effort with mating success, as has been
found in some terrestrially breeding pinnipeds
(Anderson and Fedak, 1985;
Deutsch et al., 1990;
Tinker et al., 1995). The lack
of a body size effect is somewhat surprising given the importance of body size
in the other aquatically mating species that has been studied to date, namely,
the Weddell seal (Bartsh et al.,
1992). We suggest that body size is of lesser importance to mating
success in harbor seals than it is in Weddell seals primarily because neither
females nor resources important to females can be defended by harbor seal
males. In contrast, the most successful Weddell seal males defend underwater
territories and breathing holes that females require for access to the
sea.
The energetic data indicated that two group D males had a relatively high
rate of food energy intake (Figure
4) and dove deeply more often than other males
(Figure 5), suggesting that
they spent more time on offshore foraging trips. At a site with a low density
of females at the haul-out, it may be more profitable to visit the foraging
grounds used by females rather than trying to intercept them near the shore.
Telemetry data from harbor seals in the Moray Firth, Scotland, suggests that
some males adopt the strategy of staying near offshore foraging grounds,,
while others focus their activity near haul-out sites during the breeding
season (Van Parijs et al.,
1996).
Mating systems theory predicts that where groups of females are
indefensible and group memberships and/or locations are unpredictable, males
may either rove in search of females or patrol territories smaller than female
home ranges (Clutton-Brock,
1989). This situation predicts low variance in male reproductive
success (Emlen and Oring,
1977). The variation in diving behavior and movement by males
among sites where females haul out on the beach suggests that successful male
harbor seals use several tactics to obtain matings in this population. The
tactic used by a male may depend on body condition at the beginning of the
breeding season: males that have large energy reserves can afford to make
fewer foraging trips and remain in a home range near shore, concentrating
their activity in shallow diving behavior near the haul-out site. Other males
may spend more time at sea near foraging locations. The payoff of these
tactics may vary depending on the number of breeding females present at the
colony within a season and their distribution while at sea. When female
density is low near the haul-out site, it may prove less profitable to display
(e.g., vocalize) in nearshore waters than to do so at offshore areas where
females feed. During our study, the density of breeding females was
approximately twofold lower than it had been in previous years (Bowen WD,
unpublished data). Perhaps in previous years of higher female density
nearshore, males in better body condition that spent much of their time
displaying in shallow water had a disproportionately better mating success.
Over the two seasons in our study, males that were successful in one year
where not necessarily successful the next
(Coltman et al., 1998b).
However, at this point we do not know if this was because the most successful
tactic varies over time or whether males used different tactics in different
years. Longer term paternity and behavioral data would be required to test
these ideas.
In our study, the behavioral characteristics significantly associated with
mating success were being rarely sighted and rarely hauling out alone
(Figure 7), making the
successful males perhaps the least conspicuous. It could be argued that these
same characteristics make these males the most difficult to sample, in which
case our sample might be biased toward less successful males. However, we did
not find evidence for a large number of paternal half-sibs within cohorts
(Coltman et al., 1998b),
indicating that it is unlikely that a small number of unsampled males were
highly successful. Also, the number of paternities actually assigned was not
considerably different from the expected number, estimated by simulation under
the assumption that all sampled males were equally likely to be the true
father (Coltman et al.,
1998b). Therefore, although there is the potential for bias in our
sample, we found no evidence that the effects were considerable.
In summary, the most successful male harbor seals were of moderate body
size, associated with many groups of females on shore, and showed limited
fidelity to a particular haul-out site. Some successful males spent
considerable time foraging in deeper water and thus may have encountered mates
at offshore foraging grounds in addition to those encountered while competing
for mates closer to the breeding colony. We have determined some of the
phenotypic traits associated with paternity in this aquatically mating
species. However, unmeasured factors such as previous breeding experience, the
intensity or frequency of underwater vocalizations
(Hanggi and Schusterman, 1994)
and other courtship behaviors, and the at-sea distribution patterns of males
in relation to receptive females also may be important predictors of male
mating success.
|
ACKNOWLEDGEMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
We thank Sara Iverson, Jim McMillan, Monica Muelbert, Daryl Boness, and Brian Beck for their assistance during fieldwork. We are also grateful to Gerry Forbes (AES Canada) for logistic support provided at Sable Island. Paul Thompson and one anonymous referee provided useful recommendations that greatly improved this paper. This study was supported by NSERC Operating Grants to W. D. Bowen and J. M. Wright, an NSERC postgraduate fellowship to D. W. Coltman and the Department of Fisheries and Oceans, Canada.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONACKNOWLEDGEMENTSREFERENCES |
|---|
Amos W, Twiss S, Pomeroy P, Anderson S, 1993. Male mating success and paternity in the grey seal, Halichoerus grypus: a study using DNA fingerprinting. Proc R Soc Lond B 252:199-207.[Medline]
Amos B, Twiss S, Pomeroy P, Anderson S, 1995. Evidence
for mate fidelity in the grey seal. Science
268:1897-1899.
Anderson SS, Fedak MA, 1985. Grey seal males: energetic and behavioural links between size and sexual success. Anim Behav 33:829-838.
Arnould JP, Duck CD, 1997. The cost and benefits of territorial tenure, and factors affecting mating success in male Antarctic fur seals. J Zool 241:649-664.
Atkinson S, Gilmartin WG, 1992. Seasonal testosterone pattern in Hawaiian monk seals (Monachus schauinslandi). J Reprod Fertil 96:35-39.[Abstract]
Bartsh SS, Johnston SD, Siniff DB, 1992. Territorial behaviour and breeding frequency of male Weddell seals (Leptinychotes weddelli) in relation to age, size, and concentration of serum testosterone and cortisol. Can J Zool 70:680-692.
Bartholomew GA, 1970. A model for the evolution of pinniped polygyny. Evolution 24:546-559.[ISI]
Bartholomew GA, Hoel PG, 1953. Reproductive behaviour of the Alaska fur seal, Callorhinus ursinus. J Mammal 34:417-436.
Berard JD, Nurnberg P, Epplen JT, Schmidtke J, 1993. Male rank, reproductive behavior, and reproductive success in free ranging rhesus macaques. Primates 34:481-189.
Berkovitch FB, Nurnberg P, 1996. Socioendocrine and morphological correlates of paternity in rhesus macaques (Macaca mulatta). J Reprod Fertil 107:59-68.[Abstract]
Birkhead TR, Burke T, Zann R, Hunter FM, Krupa AP,1990 . Extra-pair paternity and intraspecific brood parasitism in wild zebra finches Taeniopygia guttata, revealed by DNA fingerprinting. Behav Ecol Sociobiol 27:315-324.
Boness DJ, Bowen WD, Francis JM, 1993. Implications of DNA fingerprinting for mating systems and reproductive strategies of pinnipeds. Symp Zool Soc Lond 66:61-93.
Boness DJ, Bowen WD, Oftedal OT, 1994. Evidence of a maternal foraging cycle resembling that of otariid seals in a small phocid, the harbor seal. Behav Ecol Sociobiol 34:95-401.
Boness DJ, James H, 1979. Reproductive behaviour of the grey seal (Halichoerus grypus) on Sable Island, Nova Scotia.J Zool 188:477-500.
Boulva J, McLaren IA, 1979. Biology of the harbour seal (Phoca vitulina), in eastern Canada. Bull Fish Res Board Can 200:1-24.
Bowen WD, Oftedal OT, Boness DJ, 1992. Mass and energy transfer during lactation in a small phocid, the harbor seal (Phoca vitulina). Physiol Zool 65:844-866.
Clutton-Brock TH, 1989. Mammalian mating systems.Proc R Soc Lond B 236:339-372.[Medline]
Clutton-Brock TH, Guinness FE, Albon SD, 1982.Red deer: behavior and ecology of two sexes . Chicago: University of Chicago Press.
Coltman DW, 1997. Phenotype and paternity of male harbour seals, Phoca vitulina, at Sable Island, Nova Scotia (PhD dissertation). Halifax: Dalhousie University.
Coltman DW, Bowen WD, Boness DJ, Iverson SJ, 1997. Balancing foraging and reproduction in the male harbour seal, an aquatically mating pinniped. Anim Behav 54:663-674.[ISI][Medline]
Coltman DW, Bowen WD, Iverson SJ, Boness DJ, 1998a. The energetics of male reproduction in an aquatically mating pinniped: the harbor seal. Physiol Zool 71:387-399.[Medline]
Coltman DW, Bowen WD, Wright JM, 1998b. Male mating success in an aquatically mating pinniped, the harbour seal (Phoca vitulina) determined by microsatellite DNA markers. Mol Ecol 7:627-638.[Medline]
Cowlishaw G, Dunbar RIM, 1991. Dominance rank and mating success in male primates. Anim Behav 41:1045-1056.
Crews D, Moore MC, 1986. Evolution of mechanisms
controlling mating behaviour. Science
231:121-125.
Deutsch CJ, Haley MP, Le Boeuf BJ, 1990. Reproductive effort of male northern elephant seals: estimates from mass loss. Can J Zool 68:2580-2593.
Devlin B, Roeder K, Ellstrand NC, 1988. Fractional paternity assignment: theoretical development and comparison to other methods.Theor Appl Genet 76:369-380.[ISI]
Emlen ST, Oring LW, 1977. Ecology, sexual selection
and the evolution of mating systems. Science
197:215-223.
Godsell J, 1988. Herd formation and haul-out behaviour in harbour seals (Phoca vitulina). J Zool 215:83-98.
Hanggi EB, Schusterman RJ, 1994. Underwater acoustic displays and individual variation in male harbour seals, Phoca vitulina. Anim Behav 48:1275-1283.
Harding CF, 1981. Social modulation of circulating hormone levels in the male. Am Zool 24:223-231.
Harris AS, Young JS, Wright JM, 1991. DNA fingerprinting of harbour seals (Phoca vitulina concolour): male mating behaviour may not be a reliable indicator of reproductive success.Can J Zool 69:1862-1866.
Le Boeuf BJ, 1974. Male-male competition and reproductive success in elephant seals. Am Zool 14:163-176.
Lincoln GA, 1989. Seasonal aspects of testicular function. In: The testis (Burger H, deKrester D, eds). New York: Raven Press; 255-302.
Lindstedt SL, Boyce MS, 1985. Seasonality, fasting endurance, and body size in mammals. Am Nat 125:873-878.[ISI]
Lindstrom K, 1998. Energetic constraints on mating
performance in the sand goby. Behav Ecol
9:297-300.
Marshall TC, Slate J, Kruuk LEB, Pemberton JM, 1998. Statistical confidence for likelihood-based paternity inference in natural populations. Mol Ecol 7:639-655.[Medline]
McCann TS, 1981. Aggression and sexual activity of male Southern elephant seals, Mirounga leonina. J Zool 195:295-310.
McLaren IA, 1993. Growth in pinnipeds. Biol Rev 68:1-79.
Modig AO, 1996. Effects of body size and harem size on male reproduction in the southern elephant seal. Anim Behav 51:1295-1306.
Muelbert MM, Bowen WD, 1993. Duration of lactation and postweaning changes in mass and body composition of harbour seal, Phoca vitulina, pups. Can J Zool 71:1405-1414.
Noonan LM, Ronald K, Raeside J, 1991. Plasma testosterone concentrations of captive male hooded seals (Cystophora cristata). Can J Zool 69:2279-2282.
Owen-Smith N, 1993. Age, size, dominance and reproduction among male kudus—mating enhancement by attrition of rivals.Behav Ecol Sociobiol 32:177-184.
Pemberton JM, Albon SD, Guinness FE, Clutton-Brock JH, Dover GA,1992
. Behavioral estimates of male mating success tested by DNA
fingerprinting in a polygynous mammal. Behav Ecol
3:66-75.
Perry EA, 1993. Aquatic territory defence by male harbour seals (Phoca vitulina) at Miquelon: relationship between active defence and male reproductive success (PhD dissertation). St. John's: Memorial University of Newfoundland.
Roff DA, 1992. The evolution of life histories. New York: Chapman and Hall.
Schaeff CS, Boness DJ, Bowen WD, 1999. Female distribution, genetic relatedness, and fostering behaviour in harbour seals (Phoca vitulina). Anim Behav (in press).
Stoddart DM, Bradley AJ, Mallick J, 1994. Plasma testosterone concentration, body weight, social dominance and scent-marking in male marsupial sugar gliders (Petaurus breviceps; Marsupialia: Petauridae). J Zool 232:595-601.
Thompson PM, 1988. Timing of mating in the common seal (Phoca vitulina). Mammal Rev 18:105-112.
Thompson PM, Miller D, Cooper R, Hammond PS, 1994. Changes in the distribution and activity of female harbour seals during the breeding season: implications for their lactation strategy and mating patterns. J Anim Ecol 63:24-30.
Tinker MT, Kovacs KM, Hammill MO, 1995. The reproductive behavior and energetics of male grey seals (Halichoerus grypus) breeding on a land-fast ice substrate. Behav Ecol Sociobiol 36:159-170.
Trivers RL, 1972. Parental investment and sexual selection. In: Sexual selection and the descent of man, 1871-1971 (Campbell B, ed). Chicago: Aldine.136 -179.
Van Parijs SM, Thompson PM, Tollit DJ, MacKay A, 1996. Changes in the distribution and activity of male harbour seals during the breeding season. Anim Behav 54:35-43.
Walker BG, Bowen WD, 1993. Behavioural differences among adult male harbour seals during the breeding season may provide evidence of reproductive strategies. Can J Zool 71:1585-1591.
Weckerly FW, 1998. Sexual-size dimorphism: influence of mass and mating systems in the most dimorphic mammals. J Mammal 79:33-52.
Wingfield JC, Hegner RE, Dufty AM, Ball GF, 1990. The "challenge hypothesis": theoretical implications for patterns of testosterone secretion, mating systems, and breeding strategies. Am Nat 136:829-846.
Zielinski WJ, Vandenbergh JG, 1993. Testosterone and competitive ability in male house mice, Mus musculus: laboratory and field studies. Anim Behav 45:873-891.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. I. Schulte-Hostedde, J. S. Millar, and H. L. Gibbs Sexual selection and mating patterns in a mammal with female-biased sexual size dimorphism Behav. Ecol., March 1, 2004; 15(2): 351 - 356. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||