Behavioral Ecology Vol. 11 No. 2: 132-141
© 2000 International Society for Behavioral Ecology
Male mating strategies and the mating system of great-tailed grackles
a Department of Ecology and Evolutionary Biology, Rice University, MS 170, 1600 Main, Houston, TX 77005-1892, USA b Museum of Vertebrate Zoology, 3101 Valley Life Sciences Bldg., University of California, Berkeley, CA 94720-3160, USA c 1820 Lubbock Street, Houston TX 77007, USA d Department of Biology, University of Miami, PO Box 249118, Coral Gables, FL 33124-0421, USA
Address correspondence to K. Johnson at the New Mexico Natural Heritage Program, Biology Department, University of New Mexico, Albuquerque, NM 87131-1091, USA. E-mail: kjohnson{at}unm.edu .
Received 15 April 1999; accepted 18 July 1999.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Great-tailed grackles (Quiscalus mexicanus) are sexually dimorphic, dichromatic, colonially nesting blackbirds. In this study, males pursued three basic types of conditional mating strategies, each of which employed a different set of mating tactics. Territorial males defended one or more trees in which several females nested. They achieved reproductive success by siring the offspring of their social mates and through extrapair fertilization. Resident males lived in the colony but did not defend territories or have social mates. Transient males passed through the colony, staying no more than a few days, and probably visited more than one colony. Residents appeared to queue for access to territories, but transients did not. Residents and transients gained all paternity through extrapair fertilizations and provided no parental care. Territorial males sired the majority of offspring, but residents and transients also sired small numbers of nestlings. Territorial males were larger and had longer tails than nonterritorial males. The number of social mates was related to body size, and males that sired nestlings were heavier and had longer tails than males with no genetic reproductive success. Males that gained paternity through extrapair fertilization were heavier and had longer tails than males that did not. The mating system of great-tailed grackles can best be categorized as "non-faithful-female frank polygyny."
Key words: conditional mating tactics, great-tailed grackles, mating systems, male mating strategies, Quiscalus mexicanus, sexual selection.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The many recent studies reporting extrapair mating by male and female birds have revealed that traditional views of avian mating systems are inadequate (Johnson and Burley, 1997
;
Parker and Burley, 1997). New
conceptual views of mating systems, however, are slow in coming (but see
chapters in Parker and Burley,
1997). It is now apparent that mating systems can best be
classified and their evolution best studied from a perspective that includes
the individual mating tactics used by both sexes
(Burley and Parker, 1997) and
the relative success rates of all phenotypic classes in a population. A
complete characterization of an avian mating system must also include social
and genetic aspects of the mating system, sexual selection dynamics, and
allocation of reproductive effort to mating versus parenting
(Johnson and Burley, 1997). In
this paper we adopt the approach suggested by Johnson and Burley
(1997) in examining male
strategies of reproduction in a polygynous icterid, the great-tailed grackle
(Quisaclus mexicanus). A future paper will focus on female strategies
(Johnson et al., in preparation).
In socially polygynous species with high sexual dimorphism/dichromatism,
sexual selection on males should be intense
(Andersson, 1994;
Arnold and Duval, 1994;
Trivers, 1972). In polygynous
species in which female choice is a major determinant of male reproductive
success, as in most birds (Johnson and
Burley, 1997), a small number of attractive males should obtain
most matings via a mating strategy that is conventional for, or is thought to
typify, males of that species. This strategy may contain several discrete
mating tactics (Johnson and Burley,
1997) for obtaining, for example, territories, social mates, or
extrapair copulation (EPC) partners. Less attractive males cannot compete
successfully for mates via the conventional strategy; instead, they may adopt
other conditional or alternative strategies, which consist of different suites
of behavioral tactics. Conditional strategies differ from alternative
strategies. With conditional strategies, an individual's suite of tactics
changes with circumstances (Dawkins,
1980, McRae, 1998;
Thornhill, 1981), whereas with
alternative strategies, an individual's strategy is set for life (Gross,
1982,
1985;
Thornhill and Alcock, 1983).
In birds, alternative strategies are rare. Conditional strategies are thought
to be more common, but mating system studies tend to focus on conventional
reproductive strategies (Johnson and
Burley, 1997), which may underrepresent the dynamic complexity of
real mating systems.
In conditional strategies, the benefit-to-cost ratio of each strategy depends on an individual's circumstance, broadly defined. In addition to changes in physical condition, social variables influence circumstance. In particular, there may be a frequency-dependent component to the profitability of a given strategy. Thus, as the fraction of individuals adopting the strategy increases, there may be increasing benefit to adopting a less common conditional strategy. Finally, even the most attractive males may deviate from the conventional strategy by adopting unusual tactics, if those tactics increase their mating success without substantial cost.
In this paper we report the results of a 4-year investigation of the mating system of great-tailed grackles (Quiscalus mexicanus). Great-tailed grackles are sexually dimorphic, dichromatic, colonially nesting blackbirds. Male weights range from 200 to 250 g; females typically weigh between 100 and 120 g. Male plumage is black with blue and purple iridescence, and males have a large, keel-shaped tail. Females are chocolate brown and have a smaller keel-shaped tail.
The mating system of great-tailed grackles has long been described as
socially polygynous (Kok,
1972; Selander and Giller,
1961). More recent phylogenetic studies support the hypothesis
that sexual dimorphism in grackle body and tail size evolved as a result of
sexual selection
(Björklund,
1991; Webster,
1992). Here we identify three reproductive strategies of adult
male grackles and the phenotypic traits associated with each strategy. We
document transitions of individual males between strategies. We also report on
the differential genetic reproductive success associated with each strategy.
We present the results against a conceptual background that includes sexual
selection, allocation of reproductive effort to parenting versus mating, and
the presence of female mating tactics.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The study site is a traditional nesting colony of great-tailed grackles on the Rice University campus, Houston, Texas, USA. The colony site is approximately 2 ha in area and contains two buildings, grassy lawns, and many trees, especially live oak (Quercus virginiana), water oak (Q. nigra), loblolly pine (Pinus taeda), and longleaf pine (P. elliottii). From 1993 to 1996, females constructed nests in 29 of these trees.
We captured adult and juvenile grackles in two modified Australian crow traps placed within the colony boundaries. Each morning during the breeding season (March through July), the traps were baited with dry cat food and white bread. We took the following data on each captured bird: age, sex, weight, left and right tarsus length, left and right wing length (1995 and 1996 only), and left and right central rectrix length. For the analyses, we averaged left and right measurements, except for rectrices, for which we took the longer measure. Each bird was uniquely banded with a U.S. Fish and Wildlife Service numbered band and three anodized aluminum color bands. We took blood by brachial or jugular venipuncture.
Great-tailed grackles construct their nests up to 20 m above the ground in the tops of tall trees. We used a snorkel lift to reach nests. We checked nests two or three times per week, recording the number of eggs or nestlings and the estimated age of nestlings. When nestlings were about 12-13 days old, we removed them from the nests for banding with colored plastic and numbered aluminum bands. We drew approximately 20 µl of blood from each nestling and froze the blood-buffer mixture at -70°C. We returned nestlings to nests as quickly as possibly after processing. Tissue from dead nestlings and unhatched eggs was frozen at -70°C.
Each day during the breeding season we walked transects through the colony site, stopping for 5-10 min at each male's territory to record male identity, location, and behaviors. For each alarm calling or mobbing event observed during daily observations or nest checks, we recorded the identity of the male, location, nest or fledgling defended, and stimulus (squirrel, snorkel lift, investigator on foot, etc.). We assigned territory ownership based on the daily behavior data. Behavior of transient and nonterritorial resident males was also recorded. Territorial males were defined as males observed on a territory at least 10 times over a 2-week period. In 1994-1995 all territorial males were observed more than 50 times during a breeding season.
During the observation periods we noted the identity of females constructing nests and the locations of their nests. We checked the behavior of all nesting females each day to determine the stage of the nesting cycle. Social pairings were assigned based on the location of females' nests. For the single tree defended by several males, we assigned a social mate only if a particular male defended the female during nest construction and defended the nest and fledglings from potential predators. The genetic results presented are for a sample of families in which the mother was captured, color banded, and bled.
A partial genomic library for great-tailed grackles was made in Lambda Zap
Express (Stratagene, La Jolla, California), and primers were developed from
the sequences of positive clones (see
Hughes et al., 1998, for
primer sequences). We extracted DNA from either 10 µl of blood-buffer mix
(n = 171) or from powdered tissue samples taken from unhatched eggs
or dead nestlings (n = 10;
Strassmann et al., 1996). DNA
was amplified by polymerase chain reaction, from seven primer sites.
Radioactively labeled microsatellite regions were electrophoresed on
denaturing polyacrylamide gels and visualized using autoradiography
(Strassmann et al., 1996).
To assign paternity, we first determined the maternally inherited allele at each locus by matching nestling alleles with maternal alleles. Because no nestling failed to match its social mother at all loci (i.e., we detected no intraspecific brood parasitism), nestling alleles that did not match the mother had to have come from the genetic father. We therefore constructed a paternal profile from the nestling alleles at all loci that did not match the mother. That profile was then matched to the sire at all loci. Paternity was not assigned if more than one male matched a particular nestling's paternity profile, or if a single male matched a nestling at fewer than all seven loci.
Allele frequencies were computed using Relatedness 4.2
(Queller and Goodnight, 1989).
We used allele frequencies from adult males to compute exclusion
probabilities. Paternal exclusion probabilities were computed for individual
loci and all loci combined using the methods of Jamieson
(1965), Chakraborty and Schull
(1976), and Primmer et al.,
(1995).
Territorial males spent several years in the colony. Carry-over rates were 0.67 (1993-1994), 0.67 (1994-1995), and 0.875 (1995-1996). Single-year analyses would thus be non-independent. We include 1-year analyses for the years 1993-1995 for purposes of illustrating between-year variation (see Table 2); however, variation in significance of results is affected by the number of males we were able to recapture and measure each year. Single-year tests for number of nests started and number of successful nests are Spearman rank correlations. Single-year tests for territoriality and genetic reproductive success are Mann-Whitney U tests. Data collection was terminated less than halfway through the 1996 season; sample sizes of males measured that year were too small for analysis.
|
The largest numbers of males were measured in 1994 and 1995, and no territorial male was measured in 1993 or 1996 that was not also measured in 1994 and/or 1995. Therefore, to increase sample sizes and avoid the problem of non-independence between years, we combined the male phenotypic measures for 1994 and 1995 for the logistic regression analyses of territory acquisition and genetic offspring production. These analyses include all territorial males measured during the study, along with all nonterritorial males measured in the same season. Male data from the 2 years were also combined for the extrapair fertilization (EPF) analyses. A few males were captured in both years; we averaged trait values for those males such that each male was only included once. Data from all four years were included in the parental effort (PE) and paternity confidence analyses.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Paternity analysis
We genotyped a total of 120 nestlings (at most 49% of all nestlings hatched), 48 adult males, and 28 adult females at 7 microsatellite loci. We were able to assign paternity to 106 nestlings (88% of genotyped nestlings). Three nestlings (2.5%) matched more than one male. These could probably be assigned if more microsatellite loci were available.
We were unable to assign paternity to 11 nestlings (9%). These 11 were all genotyped but failed to match any male in our sample. Four matched no male at more than four loci, and five matched no male at more than five loci. A single-locus mutation thus cannot account for the failure of these nine nestlings to match a male. Of the two nestlings that matched males at six loci, one nestling matched three males at all except one locus, and one nestling matched two males at all except one locus and one male at all except another locus. Thus, even if we assumed a mutation occurred in the case of these two nestlings, we would still be unable to assign paternity to a single male. Therefore, we assume these nestlings were sired by transients.
The most polymorphic locus, locus 37, had 14 alleles. The least polymorphic loci were 17 and 22, with 6 alleles (Table 1). Paternal exclusion probabilities for each locus (i.e., the probability of detecting an incorrectly assigned sire) are shown in Table 1. The combined exclusion probability for all loci was 0.99897.
|
The proportions of nesting females that were banded were 0.45 (1993), 0.33 (1994), 0.37 (1995), and 0.59 (1996). We assigned parentage only to the nestlings of females that were banded and bled. Therefore, it is possible that apparent genetic reproductive success was influenced by the settling patterns of banded females. To address this question, we looked at the proportions of social mates that were banded for each male over the 4 years. There was only one male whose social mates were all banded, and he was not a highly successful male (n = 3 nests). It is therefore impossible to include results for only males whose mates were all banded. Territories in which only one banded female nested complicate the question, because they have a proportion of banded females equal to 1.0. Eliminating these single-nest trees, a significant correlation existed between the number of banded social mates and the total number of social mates over all years (r =.95, p =.005, n = 10). Therefore, we make the assumption that the sample of banded mates for each male is random and representative of relative genetic reproductive success.
Variance in male reproductive success
The mean number of nestlings sired by the 11 successful males (20% of all
males) in this study was 9.6 ± 13.2. One male sired 48 (40%) of the
nestlings we genotyped (Figure
1), but the majority of males (80%) sired none of the nestlings
produced in the colony. It is not possible to ascertain the variance in
reproductive success for the males in this study because not all nestlings
produced in the colony were genotyped, and males could have sired nestlings
outside the study site. However, microsatellite genotyping of this sample of
120 nestlings suggests that the variance in male reproductive success was
substantial.
|
Territoriality
In spite of the clear reproductive advantage enjoyed by territorial males,
a minority of great-tailed grackle males in this study owned a territory. In
the 4 years of this study, approximately one-third of the captured males were
territorial: 1993, 0.37 (7/19); 1994, 0.28 (8/29); 1995, 0.32 (9/28); and
1996, 0.37 (7/19). These numbers overestimate the actual proportions of
territorial males, because several transient males moved onto and out of the
study site before they could be captured, whereas all breeding territorial
males were captured and banded.
Nest trees that appear suitable are extremely abundant on the Rice University campus. Of the 29 trees in which females nested, however, only 12 were used by females in multiple years. Only five new males acquired territories over the course of the study, two in 1994 and three in 1995. The new males carved territories out of a large multi-male tree heavily used by females or established new territories on the periphery of the colony. Two moved to more desirable territories in subsequent years, and one abandoned his territory after nests were depredated by squirrels. Most nonterritorial males never acquired a territory attractive to females within or adjacent to the traditional colony boundary.
A core area of approximately 1 ha contained eight heavily used nest trees. Nest trees were so closely situated that males could easily have performed EPCs almost anywhere in the colony while still being within sight or sound of their own territories. There was no apparent tendency of males to sire young in territories adjacent to their own.
The territorial reproductive strategy involves several stages: acquiring a territory, attracting social mates, siring genetic offspring, and fledging social offspring. Territorial males provide parental care in the form of predator defense, and they frequently employ EPC as a conditional mating tactic. We discuss each of these aspects of territoriality below.
Territory acquisition
In 1993, territorial males were heavier than nonterritorial males. In 1994
territorial males had significantly greater tail length and body weight than
nonterritorial males, but tarsus length did not differ between the two groups
(wing length was not measured in 1994). In 1995, territorial males had
significantly greater wing length and body weight than nonterritorial males,
but tail length and tarsus length did not differ between the two groups
(Table 2).
We combined the data for 1994 and 1995 in a logistic regression analysis to
determine which combination of the traits weight, tarsus length, and tail
length would best classify territorial versus nonterritorial males. The best
model was the two-variable model including weight and tail length (AIC
criterion = 12.22). The sensitivity of this model was 85.7%, indicating that
85.7% of territorial males were correctly classified; and the specificity was
90%, indicating that 90% of nonterritorial males were correctly classified,
for probability levels between 0.42 and 0.88. This model was significantly
better than the next best model by a chi-square test (
2 = 13,
p <.001). For the small sample of males for which we had 2
sequential years of tail measurements, tail length was similar across years
(r =.9, p =.03, n = 7). Thus, larger males with
longer tails were more likely to acquire territories.
Age also strongly influences territory acquisition. During the 4 years of the study, no second-year male ever acquired a territory in the colony. All resident males were after-second-year males (ASY), and all males to which paternity was assigned were ASY males. No male that hatched in the colony during the study gained a territory there. Although we do not have exact ages of all adult males for each year, territorial males ranged in age from at least 3 to at least 7 years of age over the 4 years of the study.
Social mate attraction
Once males acquired territories, they attempted to attract females to nest
in the trees they defended. Female nest starts were somewhat synchronous at
the beginning of the breeding season but became less synchronous as the season
wore on. For example, there were 10 nests under construction in tree C between
18 and 25 March 1995 and 2 between 18 and 25 May 1995. Similarly, tree E had
four nests during the same dates in March and two during the same period in
May. Territorial males were both simultaneously and sequentially socially
polygynous, as some females renested after failure or after success, whereas
others simply started their first nest late.
Considering only territorial males, we examined the relationship between number of nests started in a male's territory and that male's traits (Table 2). Sample sizes are small because there were few territorial males and all were not recaptured each year for measurement. No traits were significantly related to number of nests started in 1993. In 1994, the number of nests started in a male's territory was significantly correlated with his weight and tarsus length, but not with his tail length. In 1995, the number of nests started was significantly correlated with weight. When 1994 and 1995 data are combined, number of nests started was correlated with weight, but not tarsus or tail length (r =.86, p =.02, n = 8). Once males have accomplished territory ownership, large body size remains important in social mate attraction.
Social offspring production
Having attracted a number of females to his territory, a male great-tailed
grackle's reproductive success then hinges on the number of his offspring
fledged. For the subset of territorial males, the number of successful nests
produced on a territory was significantly correlated with weight and tarsus
length in 1994, weight in 1995, and no traits in 1993
(Table 2). When 1994 and 1995
males are combined, number of successful nests was not correlated with tarsus
length but was correlated with weight (r =.84, p =.03,
n = 8). Although it appears that larger males have yet another
advantage here, the apparent advantage is likely due to the strong correlation
between the number of nest starts and the number of successful nests. A
partial correlation between weight and number of successful nests, removing
the effects of nests started, was not significant (r =.08, n
= 8). Variance in the ability to acquire mates appears to be the primary
influence on social offspring production.
Genetic offspring sired
Territorial males sired 84% (100/120) of the nestlings in this study, an
average of 8.3 nestlings per territorial male. Twenty-seven (23% of all or 27%
of the 100 offspring of territorial males) of these were produced by EPFs
within the colony, and 73 (61% of 120 or 73% of 100) were sired by a male that
was territorial in the tree where the nest was located
(Figure 2). These 73 can be
subdivided into 46 (38% of all or 63% of 73) sired by the owner of a
single-male tree, and 27 (23% or 37% of 73) sired by a male in a multi-male
tree (Figure 2).
|
Two trees (13%) were multi-male trees, 11 (69%) were defended by a single male, and 3 (19%) had a single male in 1 year and multiple males in another year. Thus, the proportions of nestlings shown in Figure 2 were not representative of the available trees, with proportionally fewer nestlings sired in single-male trees. This occurs because tree C, a traditional multi-male tree, also had more nests started than any tree in 1993-1995. One male was never able to defend all of this popular tree.
In 1993, males that sired offspring were heavier and had marginally longer
tarsi than males with no paternity. In 1994, males that sired nestlings were
heavier and had longer tails than males with no paternity. In 1995, those that
sired nestlings were heavier than those with no paternity
(Table 2). When 1994 and 1995
data were combined for the logistic regression, the best model was the
two-variable model including weight and tail length. This model was
significantly better than the next best model by a chi-square test
(2 = 13, p <.001). Sensitivity for this analysis
was 85.7% and specificity was 90%, for probabilities of 0.42 to 0.88. (The
results for this analysis are identical to those for the territoriality
analysis above because the set of measured territorial males was identical to
the set of measured males that sired nestlings.)
Genetic reproductive success is thus related to male size and tail length; however, this may simply occur because large males with long tails acquire territories. Looking at only the 1994-1995 males that sired offspring, the number of genetic offspring sired was still significantly correlated with weight (r =.9, p =.01, n = 9), but not with tarsus (r =.3, p =.5, n = 10) or tail length (r =.3, p =.5, n = 8).
Parental effort
Although number of mates had an overwhelming influence on the reproductive
success of the male grackles in this study, males do provide some parental
care in the form of nest and fledgling defense. We examined two measures of
male PE, alarm calling and mobbing. Both behaviors were directed at squirrels,
the only significant nest predator on the Rice University campus, and
investigators, as we approached nests in the snorkel lift or as we walked in a
male's territory. Both male and female grackles learned to recognize
individual investigators and selectively called at us and mobbed us.
Only 2 of 419 mobbing events were performed by nonterritorial males, one a resident and one a transient. Of 605 alarm calling events, only 1 was performed by a nonterritorial resident. Fewer than one-third of the males in the population each year were territorial, and although nonterritorial males sired 5% of all nestlings in this study, more than 99.7% of all nestling defense was performed by territorial males. Defense behavior is thus a reproductive tactic of territorial males.
Of 419 mobbing events, 87% were performed by a territory holder whose nests or fledglings were threatened. Similarly, 85% of 605 alarm calling events were performed by a territorial male whose nests or fledglings were threatened. Looking only at mobbing events performed for nests in which the mobbing male had paternity, we found that 97% (n = 109) were on that male's territory. Of 154 calls in defense of nests with paternity, 97% were again in the male's territory. Thus, males concentrate their defense efforts on their own territories and thereby defend primarily their own nestlings.
Considering all mobbing events, males mobbed significantly more times for fledglings they had sired or nests in which they had paternity than for unrelated offspring (paired Wilcoxon, Z = -2.24, p =.025, n = 8; Figure 3). Males also alarm called more often for their own offspring (paired Wilcoxon, Z = -2.5, p =.01, n = 10; Figure 3).
|
Paternity
More than half of the males that sired offspring in this study also lost
potential paternity when their social mates produced offspring sired by other
males. Six of the nine males that sired offspring on their own territories
lost paternity to EPFs (Figure
1). The proportion of social nestlings lost to EPFs varied from 0%
to 100% (Figure 1).
The most successful males were not necessarily the males with the lowest proportions of losses to EPFs. There was no relationship between the number of genetic offspring gained via EPF and the number of social nestlings lost to EPF (r =.2, p =.6, n = 12). Nor was there a relationship between genetic offspring gained via EPF and number of nestlings sired from a male's own territory (r =.4, p =.2, n = 12) or the total number of nestlings sired (r =.5, p =.09, n = 12). Likewise, there was no relationship between the number of nestlings lost to EPF and the total number sired (r = -.3, p =.2, n = 12) or the number of nestlings sired from a male's own territory (r = -.4, p =.1, n = 12). However, only 12 males sired offspring or lost offspring to EPF (Figure 1), so statistical power is not high.
Sperm competition is also expected in the non-faithful-female frank
polygyny mating system (Johnson and
Burley, 1997). Although we did not investigate sperm competition
directly, 33% of 36 clutches having 2 or more nestlings were sired by multiple
males, suggesting that sperm competition occurs in this species.
Transiency
Of the nonterritorial males, most were transients seen for only a few days
(Figure 4). The number of
nestlings sired by known transient males was 2% (n = 2) of all
nestlings genotyped. The 34 banded transients thus sired an average of 0.058
nestlings per male. If we assume the 11 nestlings we were unable to match to a
sire were sired by transient males, the number of nestlings sired by
transients rises to 13, or 11% all nestlings genotyped. It is possible that
these 11 nestlings were not assignable for other reasons such as errors in
genotyping. We were able to genotype all 11, however, which strongly suggests
that the actual sire was not among the males sampled (and see "Paternity
analysis" above). Transiency combined with EPC thus appears to be a
tactic that is occasionally productive for males that are unable to acquire a
good territory.
|
Nonterritorial residency
Four males did not neatly fit either the territorial or transient male
category. These males, which we call residents, were observed at least 20
times in the colony during a season, but were not assignable to a particular
territory. Of these four males, one was a transient in 1 other year of the
study and a resident in 1 other year, two were territorial in 2 other years,
and one was territorial in 3 other years. Thus, nonterritorial residency
appears to be a tactic used by males aspiring to territorial status, while
also attempting to gain EPFs with socially mated females. Three of the four
resident males sired a total of four nestlings (3% of all nestlings) as
nonterritorial residents, for an average of one nestling per resident
male.
Extrapair copulation
Extrapair copulation is a mating tactic employed by all three of the above
categories of male grackles. Territorial males engaged in EPCs with the social
mates of other territorial males (Figure
1), and resident and transient males also successfully sired
offspring of colony females. Including the 11 nestlings that did not match a
male in our sample, 44 nestlings (37% of all nestlings) were sired through
EPFs, 16 (13% of all nestlings) by nonterritorial males, and 27 (23% of all
nestlings) by territorial males (Figure
2). One territorial male (GRGo) sired 34% of all extrapair
nestlings.
It is difficult to draw conclusions regarding the effect of male traits on EPF when one-third of all extrapair nestlings were sired by one male. The number of males siring nestlings through EPF was too small in any one year to allow analysis of a relationship between number of extrapair nestlings and male traits. We therefore combined the EPF results for the 2 years in which we measured the largest number of males, 1994 and 1995, and performed Mann-Whitney U tests (Table 3). The traits of males that sired nestlings through EPF were compared to those of all other males because any male can potentially sire offspring through EPF. Trait values for the 2 years were averaged. Males that sired nestlings through EPF in 1994-1995 were significantly heavier and had marginally longer tails than those with no extrapair nestlings (Table 3). Again, most of the males that gained EPFs were territorial.
|
We compared the traits of males that lost paternity to EPF to those that did not lose paternity to EPF over the 4 years. For this analysis, we compared the subset of all males that either sired nestlings or lost paternity to EPF to those that did not lose paternity to EPF because only males with social nestlings can lose paternity to EPF. Contrary to the other analyses, the males that lost paternity to EPF were heavier and had longer tarsi than the reproducing males that did not lose paternity to EPF (Table 3). This result occurs because the majority of males that lost paternity to EPF also gained paternity through EPF (Figure 1; i.e., only larger males had paternity to lose).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The mating system of the great-tailed grackles we studied can best be described as non-faithful-female frank polygyny, a mating system in which an appreciable fraction of males in the population has more than one social mate, some males have a single mate, and the majority have no social mates at all (Johnson and Burley, 1997
).
Females typically have one social mate at a time, but many rear some young
sired by EPF. In this grackle colony, nonterritorial males also sired
offspring through EPF.
Faced with varying probabilities of success, the male great-tailed grackles
in this colony used three different mating strategies: territoriality,
transiency, and nonterritorial residency. Although we have shown that
territoriality is not the most common reproductive strategy adopted by male
great-tailed grackles, it has been considered typical
(Kok, 1972;
Selander and Giller 1961), and
we therefore define it as the conventional strategy. Transiency and residency
are therefore conditional strategies because they are adopted by males that
lack the traits necessary to successfully compete for mates via the
territorial option. All three types of males used conditional mating tactics,
but the suites of tactics differed among strategy types
(Table 4).
|
Territoriality
The most successful males in this study were those able to acquire a
desirable nesting territory and attract large numbers of nesting females.
Territorial males sired an average of 8.3 nestlings per male, compared to an
average of one per resident male and 0.058 per transient male. Although
females were commonly nonfaithful, the males with the largest numbers of
social mates were still the most genetically successful. These males expended
diffuse mating effort (mating effort that cannot be partitioned among
individual mates; Johnson and Burley,
1997) in the form of competition for territories, dominance
behaviors, fights, territorial song, large body size, and tail ornaments.
Focused mating effort (focused on an individual female) occurred as
solicitation and display to individual females and guarding of social mates
during nest construction. (Adaptations that enhance sperm competition, if
present, would also represent focused mating effort.)
Because males adopting the conventional strategy also vary in
attractiveness and mating opportunity, they are expected to adopt conditional
mating tactics such as EPC and sperm competition
(Johnson and Burley, 1997).
Grackles generally conform to this expectation. Territorial males frequently
engaged in EPC. The relatively high losses to EPF by otherwise successful
males is contrary to the results of some other studies
(Kempenaers et al., 1997;
Ketterson et al., 1997;
Perreault et al., 1997), and
we did not find a relationship between EPF gains off the territory and EPF
losses on the territory, as found for red-winged blackbirds (Agelaius
phoeniceus; Gray, 1997).
However, five of the six most successful males lost paternity to EPFs. Some
paternity uncertainty appears to be a fact of life for male grackles,
particularly males that have large numbers of females nesting on their
territories.
Although the rate of mixed paternity within clutches was substantial, the
grackles did not copulate frequently, compared to other bird species in which
sperm competition is thought to occur
(Birkhead and Møller,
1992). We observed only 30 copulations over the 4-year duration of
this study. Sperm depletion and/or female control over copulation may preclude
the use of frequent copulation to overcome sperm competition
(Johnson and Burley, 1997).
Therefore we expect that grackles deal with sperm competition by other means
such as relatively large testes or perhaps various sperm morphologies.
Female choice appeared to be the primary determinant of a male grackle's paternity confidence. Males competed for access to fertile females at nest sites, and territorial males guarded females constructing nests on their territories, but a male with several fertile mates could not possibly follow them all to distant sites where they collect nesting materials and forage. Males with few fertile mates may follow their mates somewhat more frequently, but we did not observe clear cases of consortship by any male. Consistent with the importance of female choice, we observed no clear instances of forced EPC, although we cannot rule out the possibility that the potential for injury prevents females from resisting.
Whether or not male birds adjust PE based on paternity confidence is a
controversial question (Kempenaers and
Sheldon, 1998; Lifjeld et al.,
1998; Wagner et al.,
1998), and there have been results that support both sides (e.g.,
Gori et al., 1996;
Lifjeld et al., 1998;
Sheldon et al., 1997;
Stutchbury et al., 1994;
others reviewed in Johnson and Burley,
1997). In great-tailed grackles, the majority of PE is allocated
to genetic offspring, although it is unclear whether individual males actually
adjust PE to match paternity. Mobbing and calling distances may be adjusted
based on paternity confidence, but our data did not allow a test of that
hypothesis.
Male grackles appear to have a simple decision rule regarding nestling defense: to defend nests and nestlings hatched in their territories. In spite of the occurrence of extrapair paternity, this simple rule is effective in most instances. However, two types of "errors" do occur.
First, males typically fail to protect offspring they have sired in another male's territory (i.e., when they follow the territory defense rule). Given that EPF is a secondary tactic for a successful male with considerable paternity in his own territory, it probably does not pay such a male to protect extrapair nestlings at the cost of leaving his own tree undefended against predators or open to males seeking EPCs with his mates. The extrapair nestlings will also be defended by the social father in the process of defending his own nestlings in the same tree.
Rarely, males break the territory-only offspring-defense rule to defend offspring sired through EPF. We observed three instances of the most successful male alarm calling when his nestlings sired by EPF were threatened. This tactic provides the greatest potential benefit for males with large numbers of EPFs in which they have confidence of their paternity. We did not observe any male using the more risky behavior, mobbing, in defense of his offspring sired through EPF.
The second type of defense error occurs when a male defends nestlings he did not sire because he is the victim of EPF (Figure 1). It is possible that a male knows he is not the father of the nestlings most directly threatened by the predator, but he may have paternity in other nests in the same tree. We gave the male the benefit of the doubt in this analysis and assumed that, if he had paternity in any nest in the tree being threatened, he was defending that nest.
The second error type is committed by males whose mates are nonfaithful.
These apparently less attractive males may intentionally provide PE to attract
or retain social mates (Burley,
1988). In this case, apparent PE would actually function as mating
effort (Johnson and Burley,
1997). In this study, one male was territorial in all years,
attracted several social mates each year, and consistently provided PE. One
female exploited this male's PE by nesting repeatedly in his territory, but
70% of his social nestlings were sired by another male
(Figure 1). In summary, we
assume that most apparent errors of defense were actually based on tactical
decisions, particularly when a male had either gained or lost paternity to
EPC.
Residency and transiency
Transiency and nonterritorial residency are examples of conditional
strategies used by males unable to defend a high-quality territory. Males
adopting these two strategies were smaller than territorial males. They did
sire offspring, although at lower rates than the most successful territorial
males. However, it is important to consider variation in reproductive success
within strategic classes. For example, transient males that sired one
offspring gained more paternity for much less mating effort than unsuccessful
territorial males (Figure
1).
Nonterritorial resident grackles may engage in a type of queuing behavior
that gives them access to territorial vacancies as they open (McDonald,
1988a,
b;
Poston, 1997). A closely
related species, the boat-tailed grackle (Quiscalus major), queues
for access to nesting colonies of females, but the social mating system of
that species has been characterized as female-defense polygyny, not
territoriality (Post, 1992;
Poston, 1997). Three of the
four residents (75%) were territorial in other years of the study. Two became
territorial in the year after gaining residency. One lost his territory for a
year and regained it the following year. The fourth moved from transiency to
residency. Like transiency, the tactic of residency is most successfully used
by the higher quality, nonterritorial males, those that are unwilling or (more
likely) unable to retain a productive territory every year. In addition to
facilitating queuing, residency allows males more frequent access to females
for EPC, and residents have the opportunity for females to become acquainted
with them as potential EPC partners or future social mates. It is also
possible that residents deceive females into perceiving that they are
territory owners. Residents successfully practice EPC (three of four residents
sired offspring). Success rates were low relative to those of the most
successful territorial males, but high relative to those of transients, and
variation in reproductive success among residents was low compared to
variation among territorial males. Because they sire nestlings of females
mated to territorial males and typically gain only partial paternity in a
clutch, we assume sperm competition occurs. Of 1024 acts of defense, only one
mobbing act and one alarm calling act were performed by a resident male.
Residency thus appears to be a distinct conditional strategy centered around
EPC, queuing, and possibly deception. Residency is not simply territoriality
poorly executed.
Transient males employed EPC and probably engaged in sperm competition. It seems unlikely that most transients were prospecting for territorial vacancies as they moved through the colony, because they are younger, smaller, and less ornamented than the males holding territories. Only one transient (3%) acquired a territory in another year of the study. Only a small percentage of transients (3% of 34 transient males) was identified as sires in this study. However, as many as 11 nestlings could have been sired by unidentified transient males, bringing the proportion of nestlings sired by transients to 10%. It is also possible that transients sire offspring in more than one colony. If so, this strategy could prove as successful in the short term for individual transient males as that of non-territorial residency. Siring offspring in multiple colonies does not appear to be a viable option for a resident male. Residents, however, stand a much greater chance of obtaining a territory, which should increase their long-term success. Of 1024 acts of defense, only one mobbing act was performed by a transient male. The transient strategy, like the resident strategy, appears to be based on EPC, with no PE expended. Unlike residents, transients did not stay in the colony for more than a few days, did not benefit from queuing, rarely ascended to territorial status, and did not have an opportunity to deceive females into believing they were territorial. Because the two groups used different sets of mating tactics (Table 4) and had different success rates, we distinguish between residency and transiency.
It is possible that these two strategy types represent a single nonterritorial strategy and that transients are actually residents in another colony. However, the number of transient males was about 10 times higher than the number of residents at the colony we studied. Our observations outside the colony suggest that there is a large nonresident population composed mostly of second-year males and that there are too few colonies near ours to accommodate this large number of males as residents. More important, the residents in our colony were all ASY males. This leaves only the transient strategy available for second-year males. Some less competitive ASY males also appear to adopt the transient strategy.
Five of six transitions among strategy types in this study were in the direction of more successful strategies (i.e., from transient to resident, transient to territorial, or resident to territorial). The only exception involved a male that lost his territory, adopted the residency strategy for one season, and then regained his territory in the third year. One old, long-term territorial male lost territorial status and became a resident after the study ended. Thus, changes among strategy types occurred in both directions, supporting our assertion that these strategies are in fact conditional.
An important question is, conditional on what? Age, experience, tail length, and weight are obvious grackle traits that change between years. In addition, less obvious factors such as male dominance status or health could determine shifts between conditional strategy types over time. When considering correlates of strategy change, it is important to consider each male's circumstances relative to those of other males in the population. Demographic factors such as population sex ratio and age structure should therefore have potential to impact a particular male's choice of conditional strategy.
Sexual dimorphism
The dramatic sexual dimorphism in great-tailed grackles implies a history
of sexual selection on males, and phylogenetic studies support this hypothesis
(Björklund,
1991; Webster,
1992). Our data suggest that sexual selection on male size and
tail length are ongoing. Larger males had an advantage in territory
acquisition, social mate attraction, and genetic reproductive status. Longer
tail length was also associated with territory acquisition, in combination
with size.
It is interesting that tail length was not associated with social mate attraction but that it was associated with genetic offspring production and the ability to sire offspring through EPC. This result should be viewed with caution because sample sizes for the tests of social mate attraction may have been too small to detect a relatively weak preference for tail length. However, the data suggest that females may be using different criteria in choosing their social mates and their EPC partners. Larger males clearly had the advantage in acquiring territories, and females preferred to settle on the territories of larger males, probably because larger males were able to acquire trees with the most desirable nest sites. Females, however, appeared also to consider tail length when choosing EPC partners. The tail could thus be an aesthetic ornament or an indicator of good genes.
This discrepancy between the traits of social mates and those of EPC
partners suggests that many females are precluded from acquiring a social mate
that is also an acceptable genetic mate, for example, by male selectivity or
female-female competition for nest sites. An alternative hypothesis is that
females must choose social mates that are good nest defenders and that defense
abilities of males with longer tails are impaired by the impact of long tails
on maneuverability (Selander,
1965). Female grackles, too, exhibit variation in reproductive
success, as well as primary and conditional mating tactics. Female tactics, as
well as the constraints on female choice and the dynamic interplay of male and
female tactics, will be discussed in future papers (Johnson et al., in
preparation).
|
ACKNOWLEDGEMENTS |
|---|
We thank the many Rice University undergraduate volunteers and research students and the students in the Rice University Minority High School Student Research Apprenticeship Program who assisted with field work. The Rice University Grounds Department, particularly R. Smith and J. Alejandro, made it possible to access nests. The University of New Mexico Statistics Clinic consulted on analyses. N. Burley's comments greatly improved an earlier version. Special thanks to J. Strassmann and D. Queller for allowing us into their DNA lab. All animal handling protocols were approved by the Rice University Institutional Animal Care and Use Committee. Banding was conducted under U.S. Fish and Wildlife Service banding permit PRT-22158. This research was supported by National Science Foundation grant IBN-9407310 to K.J.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Andersson M, 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Arnold SJ, Duval D, 1994. Animal mating systems: a synthesis based on selection theory. Am Nat 143: 317-348.
Birkhead TR, Møller AP, 1992. Sperm competition in birds: evolutionary causes and consequences. San Diego, California: Academic Press.
Björklund M, 1991. Evolution, phylogeny, sexual dimorphism, and mating system in the grackles (Quiscalus spp: Icterinae). Evolution 45: 608-621.
Burley N, 1988. The differential allocation hypothesis: an experimental test. Am Nat 132: 611-628.
Burley NT, Parker PG, 1997. Emerging themes and questions in the study of avian reproductive tactics. In: Avian reproductive tactics: female and male perspectives. Ornithological Monographs 49 (Parker PG, Burley NT, eds). Washington, DC: American Ornithologists' Union; 1-20.
Chakraborty R, Schull WJ, 1976. A note on the distribution of the number of exclusions to be expected in paternity testing. Am J Hum Genet 28: 615-618.
Dawkins R, 1980. Good strategy or evolutionarily stable strategy? In: Sociobiology: beyond nature/nurture? (Barlow G, Silverberg J, eds). Boulder, Colorado: Westview Press; 331-367.
Gori DF, Rohwer S, Caselle J, 1996. Accepting
unrelated broods helps replacement male yellow-headed blackbirds attract
mates. Behav Ecol 7:
49-54.
Gray EM, 1997. Intraspecific variation in extra-pair behavior of red-winged blackbirds (Agelaius phoeniceus). In: Avian reproductive tactics: female and male perspectives. Ornithological Monographs 49 (Parker PG, Burley NT, eds). Washington, DC: American Ornithologists' Union; 61-80.
Gross MR, 1982. Sneakers, satellites, and parentals: polymorphic mating strategies in North American sunfishes. Z Tierpsychol. 60: 1-26.
Gross MR, 1985. Disruptive selection for alternative life histories in salmon. Nature 313: 47-48.
Hughes CR, Kavlie R, Johnson, K, 1998. Characterization of polymorphic trinucleotide microsatellite loci for the great-tailed grackle, Quiscalus mexicanus. Mol Ecol 7: 783-784.
Jamieson A, 1965. The genetics of transferrins in cattle. Heredity 20: 419-441.[ISI][Medline]
Johnson K, Burley NT, 1997. Mating tactics and mating systems of birds. In: Avian reproductive tactics: female and male perspectives. Ornithological Monographs 49 (Parker PG, Burley NT, eds). Washington, DC: American Ornithologists' Union; 21-60.
Kempenaers B, Sheldon BC, 1998. Confounded correlations: a reply to Lifjeld et al. and Wagner et al. Anim Behav 55: 241-244.
Kempenaers B, Verheyen GR, Dhondt AA, 1997. Extrapair
paternity in the blue tit (Parus caeruleus): female choice, male
characteristics, and offspring quality. Behav Ecol
8: 481-492.
Ketterson ED, Parker PG, Raouf SA, Nolan Jr. V, Ziegenfus C, Chandler Cr, 1997. The relative impact of extra-pair fertilizations on variation in male and female reproductive success in dark-eyed juncos (Junco hyemalis). In: Avian reproductive tactics: female and male perspectives. Ornithological Monographs 49 (Parker PG, Burley NT, eds). Washington, DC: American Ornithologists' Union; 81-101.
Kok OB, 1972. Breeding success and territorial behavior of male boat-tailed grackles. Auk 89: 529-540.
Lifjeld JT, Anthonisen K, Blomqvist D, Johnsen A, Krokene C, Rigstad K, 1998. Studying the influence of paternity on paternal effort: a comment on Kempenaers and Sheldon. Anim Behav 55: 235-238.[ISI][Medline]
McDonald DB, 1988a. Cooperation under sexual selection: age-graded changes in a lekking bird. Am Nat 134: 709-730.
McDonald DB, 1988b. Correlates of male mating success in a lekking bird with male-male cooperation. Anim Behav 37: 1007-1022.
McRae SB, 1998. Relative reproductive success of
female moorhens using conditional strategies of brood parasitism and parental
care. Behav Ecol 9:
93-100.
Parker PG, Burley NT (eds), 1997. Avian reproductive tactics: female and male perspectives. Ornithological Monographs 49. Washington, DC: American Ornithologists' Union.
Perreault S, Lemon RE, Kuhnlein U, 1997. Patterns and
correlates of extrapair paternity in American redstarts (Setophaga
ruticilla). Behav Ecol 8:
612-621.
Post W, 1992. Dominance and mating success in male boat-tailed grackles. Anim Behav 44: 917-929.
Poston JP, 1997. Dominance, access to colonies, and queues for mating opportunities by male boat-tailed grackles. Behav Ecol Sociobiol 41: 89-98.
Primmer CR, Møller AP, Ellegren H, 1995. Resolving genetic relationships with microsatellite markers: a parentage testing system for the swallow Hirundo rustica. Mol Ecol 4: 493-498.[Medline]
Queller DC, Goodnight KF, 1989. Estimation of genetic relatedness using allozyme data. Evolution 43: 258-275.[ISI]
Selander RK, 1965. On mating systems and sexual selection. Am Nat 49: 129-141.
Selander RK, Giller DR, 1961. Analysis of sympatry of great-tailed and boat-tailed grackles. Condor 63: 29-86.
Sheldon BC,
Räsänen K, Dias
PC, 1997. Certainty of paternity and paternal effort in the
collared flycatcher. Behav Ecol 8:
421-428.
Strassmann JE, Solis CR, Peters, JM, Queller DC, 1996. Strategies for finding and using highly polymorphic DNA microsatellite loci for studies of genetic relatedness and pedigrees. In: Molecular zoology: advances, strategies, and protocols (Ferraris JD, Palumbi, SR, eds). New York: Wiley-Liss; 163-180.
Stutchbury BJ, Rhymer JM, Morton EM, 1994. Extrapair
paternity in hooded warblers. Behav Ecol
5: 384-392.
Thornhill R, 1981. Panorpa (Mecoptera: Panorpidae) scorpionflies: systems for understanding resource defense polygyny and alternative male reproductive efforts. Annu Rev Syst Ecol 12: 355-386.[ISI]
Thornhill RT, Alcock J, 1983. The evolution of insect mating systems. Cambridge: Harvard University Press.
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.
Wagner RH, Schug MD, Morton ES, 1998. Studying paternity and paternal care: the value of negative results. Anim Behav 55: 239-240.
Webster MS, 1992. Sexual dimorphism, mating system, and body size in New World blackbirds (Icterinae). Evolution 46: 1621-1641.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. Meunier, S. A. West, and M. Chapuisat Split sex ratios in the social Hymenoptera: a meta-analysis Behav. Ecol., March 1, 2008; 19(2): 382 - 390. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. J. Brockmann An experimental approach to altering mating tactics in male horseshoe crabs (Limulus polyphemus) Behav. Ecol., March 1, 2002; 13(2): 232 - 238. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||