Behavioral Ecology Advance Access originally published online on November 16, 2005
Behavioral Ecology 2006 17(1):88-96; doi:10.1093/beheco/arj004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Size, operational sex ratio, and mate-guarding success of the carrion beetle, Necrophila americana
Department of Zoology, Rudman Hall, University of New Hampshire, Durham, NH 03824, USA
Address correspondence to M.P. Scott. E-mail: mps{at}cisunix.unh.edu.
Received 31 August 2004; revised 20 September 2005; accepted 25 September 2005.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONTHE MATE-GUARDING MODELTEST OF THE MODELREFERENCES |
|---|
When male insects guard females until oviposition, the benefits from last-male sperm precedence must outweigh the costs of relinquishing additional fertilizations. The profitability of guarding is increased when males guard large, fecund females and when females are scarce because fewer fertilizations are sacrificed. However, the male reproductive success is not only determined by the profitability of guarding but also by his ability to maintain guarding. In this study, we used male carrion beetles (Necrophila americana) to examine the effects of sex ratio, male relative size, and female quality on the ability to guard. First, we present a model of mate guarding that explores factors, such as sperm precedence, sex ratio, male size, and female quality, that influence the profitability of postcopulatory riding. Our model predicts that large N. americana males should preferentially guard the largest female only when the sex ratio is male biased and sperm precedence is above 80%. In contrast, small males gain little from guarding because they are not likely to maintain it and be the last male to mate. Then, we tested these predictions by manipulating sex ratio, relative male size, and female quality. All males in equal sex ratio and large males in male-biased sex ratio guarded females significantly longer than did males in female-biased sex ratio. In male-biased sex ratio, large males guarded significantly longer and achieved more takeovers than small males. Large females were guarded longer. The success of guarding males in this beetle depends on their size relative to other males and the operational sex ratio.
Key words: carrion beetle, female quality, male size, mate guarding, Necrophila, operational sex ratio, resource-holding potential.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONTHE MATE-GUARDING MODELTEST OF THE MODELREFERENCES |
|---|
Postinsemination associations, such as mate guarding, present a conflict for male insects because they must balance the cost of forfeited matings with the benefits of last-male sperm precedence (Alcock, 1994
; Simmons, 2001). When male insects guard females until oviposition, they often fertilize the majority of their clutch (Alcock, 1994; Simmons, 2001; Simmons and Siva-Jothy, 1998). However, these guarding males relinquish time that could be spent acquiring additional fertilizations from other females. Mate guarding is promoted in a male-biased sex ratio because these forfeited matings are minimized when females are scarce (Alcock, 1994; Schofl and Taborsky, 2002). The profitability of mate guarding is also increased when the last male to mate with a female gains a significant fertilization advantage (Alcock, 1994). Although often not considered in mate-guarding studies, the size of males and females in the mating arena are also important. Large females are preferred when there is a positive relationship between female size and the number of eggs she deposits (Brown and Stanford, 1992; Larsson, 1988; Savalli and Fox, 1998; Wen, 1993). In many insects, large males with higher resource-holding potential (RHP) outcompete smaller males for access to the most valuable female and can guard her until she oviposits (Alcock, 1991; Clark, 1988; Larsson, 1988; Safryn and Scott, 2000). This study investigates how relative male size, female quality, and sex ratio (operational sex ratio [OSR]) affect the duration of mate guarding in the male carrion beetle, Necrophila americana (Coleoptera: Silphidae). Females are attracted to medium to large carcasses that are required for reproduction, so males need only find a carcass to locate receptive females. Males are often seen riding on the backs of females throughout reproductive bouts, and females do not appear to take an active role in accepting or rejecting males (Knox TT, personal observation). Aggression during takeovers is limited, and the risk of injury to males is very low (Knox TT, personal observation). Females oviposit within 12 h in the soil nearby and continue to mate and oviposit until the carcass is unusable (approximately 1 week, depending on the size of the carcass, weather, and temperature) (Knox TT, personal observation). Thus, the duration of reproductive bouts is dependent on the viability of the carcass, an unpredictable and rare resource, and this ecology allows us to consider the payoff of exceptionally long guarding episodes. We first present a model to explore mate guarding and then test predictions empirically.
|
THE MATE-GUARDING MODEL |
|---|
|
TOPABSTRACTINTRODUCTIONTHE MATE-GUARDING MODELTEST OF THE MODELREFERENCES |
|---|
The assumptions
We use the factors of sperm precedence (the proportion of offspring sired by the last male to mate, P2), sex ratio, relative male size, and female quality, to develop a model to predict the conditions for which guarding becomes profitable for males. This explicit consideration of female quality and relative male size is the focus, and major contribution, of this study. Our model calculates a guarding index (GI) that represents the relative reproductive payoff a male would receive if he guarded a female until oviposition and equates to the expected duration of guarding or riding time. When the index is positive, then the benefits of guarding outweigh the costs and a guarding strategy should be adopted. On the other hand, if the index is negative, the costs outweigh the benefits and a male should try to fertilize as many females as possible. The assumptions we make are as follows. Some parameters have been simplified.
- Fertilization success (percent paternity) is limited to two values P1 and P2, and P1 < P2.
- If a male guards, he gets a benefit of P2. The nonguarding male to mate with a female gets a fraction of P1, depending on how many males are present on the carcass. As the number of males on the carcass increases, it is more likely that the female has had previous mates and the proportion of P1 a nonguarding male receives for each female is reduced.
- Cost of guarding only includes the number of females that a male will not be able to fertilize because he is guarding an already inseminated female. Escalated fights with injury are rare in N. americana, and time to search for a carcass, although probably time consuming and metabolically expensive, is not a cost of forfeited matings because males wait for additional females at a carcass until it is not viable.
- The success of an individual male depends on the ability to resist takeover and remain riding on the female until she oviposits. A value of 1 is attached to the largest male on the carcass, and smaller males are assigned a value less than one indicating their proportional size to the largest male. If the size of a male is in the smallest 10% of the male population on a carcass, then his relative size value is 0.1.
- Larger females are more valuable resources because of the greater number of eggs they will deposit. Female size is denoted by the clutch size.
- Male competitor's decision to displace a male and guard is made before takeover. There is no relevant information conveyed during a contest. Males cannot give false information about their guarding ability (RHP).
In nature, the sex ratio may fluctuate over time when different males and females arrive at the carcass. Our model accounts for these changes by using relative male size and number of males present, along with the size and number of females present. For example, if a male arrives at a carcass and is larger than all other males present, the guarding payoff will decrease for all previous males because their relative size decreases when a larger competitor enters the arena. A male has the potential to achieve a share of P1 for all the other females present on the carcass, except for the P2 fertilization success he receives from his guarded female. Because last-male sperm precedence after three partners is unclear (Zeh J and Zeh D, 1994), we limit our model to P1 and P2.
The model
The variables used in the equation of the model are listed in Table 1. These factors are varied to calculate the GI which will predict whether a male should or should not guard. This index is calculated by subtracting the costs from the benefits of guarding while varying the factors of male size, female size, sperm precedence, and sex ratio. The payoff a male will receive if he guards a female is calculated from the following equation:
 |
|
The paternity of the last male to mate with the female before oviposition is multiplied by the clutch size of the guarded female to give the number of eggs the guarding male gains [P2C]. This number is then multiplied by the relative size of the male divided by the number of male competitors present minus one (S/M–1). This factor represents the likelihood of the male being able to ride the female until she oviposits when the sex ratio is varied between a female-biased sex ratio (3:6), an equal sex ratio (3:3), and a male-biased sex ratio (6:3).
The cost of guarding is given by the number of eggs the guarding male will not fertilize. This is calculated from the product of the paternity (P1) of a male not guarding when the female deposits her eggs, the average clutch size (c), the number of females present (F), and the male's relative size divided by the number of male competitors present (S/M–1). This factor represents the availability of females to a particular male by calculating the probability of a male of a certain size to obtain P1 with a given number of male competitors present. If the GI is positive, then the male should guard, and if the GI is negative, it is in the male's best interest to copulate with all the females and avoid guarding one specific female. If the equation is rewritten as
 |
The parameters in this equation are varied to investigate how female size would affect the payoffs for small and large males. The clutch size of the guarded female (C) is varied between 10, 7.5, and 5 eggs. In addition, we vary P2/P1 values, male size, and sex ratio to see in which scenarios guarding becomes profitable. Each female has a different value depending on the size of the male and the P2 values.
Results
Male size, female size, and P2 all have an important influence on the GI (Figure 1). When the OSR is male biased, a large male has both the most to gain and the most to lose from mate guarding, while a small male receives very little benefit from guarding. Because a small male will not be able to maintain guarding, he is not likely to receive the P2 fertilization benefit. In this case, the small male is better off trying to copulate quickly with as many unguarded females as possible. Once a female is guarded, he is not going to be able to displace the guard and inseminate the female.
|
As shown in Figure 1a–d (male-biased sex ratio), the small male's GI hovers around the break-even point under all conditions of P2 and female quality. However, the large and medium size males receive very different payoffs when P2 and female quality are varied. When P2 is 70% and 75% (Figure 1a,b), the GI for all males is negative, indicating that guarding is unprofitable when sperm precedence is low. The large male would experience the greatest cost if he decided to guard. In Figure 1c, P2 = 80% and guarding the best quality female is the only scenario that benefits males. In Figure 1d, when P2 = 85%, the GI is positive for all males but is the most advantageous for large males because they can successfully guard large females until oviposition. Even guarding a poor female becomes marginally profitable in this case.
The numbers of males and females on a carcass are also important factors that affect the profitability of mate-guarding behavior in males. Figure 2a–f illustrates the relationship between P2, male size, sex ratio, and female quality. In a male-biased sex ratio, males of all sizes should guard the best quality females on a carcass when P2 is above 75% (Figure 2a), average females when P2 is above 80% (Figure 2c), and poor females when P2 is above 85% (Figure 2e). In a female-biased sex ratio, the loss of additional matings, rather than the likelihood of being displaced, becomes the factor that reduces the profitability of guarding (GI). A nonguarding male, in a female-biased sex ratio, attains P1 for multiple females which outweighs the paternity of P2 for a single guarded female. For this reason, P2 plays an important role in calculating whether guarding is a profitable strategy. In a female-biased sex ratio, males of all sizes should only choose to guard the best female when P2 is above 85% (Figure 2b). This model predicts that average- and poor-quality females should never be guarded in a female-biased sex ratio (Figure 2d,f). These results demonstrate that larger males have more to lose when guarding a smaller female because they have the ability to takeover and inseminate all the females on the carcass. However, larger males have the most to gain from guarding a large female, while small males lose and gain little when adopting mate-guarding behavior.
|
The GIs for an equal sex ratio (3:3 males:females) are not shown but demonstrate the same patterns as those for a male-biased sex ratio. However, when the sex ratio is equal, the GIs are higher than those for male-biased populations because the cost of guarding a female is less when fewer males are present.
The predictions
Our model makes the following predictions that have been tested for other mate-guarding species and that can be experimentally tested for N. americana.
- Relatively large males should only guard the most valuable (largest) females on the carcass, which would result in size-assortative mating.
- As the OSR becomes more male biased, mate guarding becomes more profitable and should increase.
- As sperm precedence (P2) increases, mate guarding becomes more profitable and should increase. If the last male to mate with a female fertilizes more than 85% of the eggs, males should guard females regardless of the sex ratio. When the profitability of mate guarding is increased, we expect the guarding duration to increase.
- Relatively small males should not guard but should inseminate unguarded females quickly because there is very little to gain from guarding and they are less likely to maintain that guarding.
|
TEST OF THE MODEL |
|---|
|
TOPABSTRACTINTRODUCTIONTHE MATE-GUARDING MODELTEST OF THE MODELREFERENCES |
|---|
Methods
General care
Adult N. americana were collected from pitfall traps in eastern New Hampshire during the summers of 2001–2003. They were sexed and weighed, and their pronotal widths (mm) were measured. Beetles were maintained in same-sex groups in 19.5 x 13 x 11–cm plastic boxes with moist paper towels. They were fed small pieces of kidney. Beetles were stored, and experiments were conducted in a shaded shed with outdoor conditions and 18:6 h of low light:dark schedule.
Female fecundity
Nine females of varying sizes (mean pronotal width = 9.16 mm, range = 8.38–9.8 mm) were placed in separate 41.5 x 20 x 15–cm boxes containing a gray squirrel carcass (road killed) and one male. These females were allowed to breed until the carcass was unusable (a maximum of 11 days); however, only the first 7 days were used for statistical analysis. Carcass viability was variable but was never less than 7 days. At 24-h intervals, eggs were retrieved from the boxes and counted for each female, yielding an oviposition rate for females. A regression analysis was performed to examine the relationship between fecundity and pronotal width of females.
Operational sex ratio
We determined the natural density of beetles, natural sex ratio from pitfall captures, and OSR on carcasses. Pitfall traps baited with beef kidney were used to collect beetles. The traps were buried in the ground and covered with chicken wire to prevent interference from mammals. Traps were sampled every day from 1 June to 1 September for natural sex ratios.
To determine the OSR on carcasses, nine medium-sized carcasses (raccoon, beaver, or fox) were placed on the ground in eastern New Hampshire woods and were covered by a wooden frame with chicken wire as the top surface. This gave beetles access to the carcass and surrounding oviposition sites in the soil while excluding large carrion-feeding vertebrates. All N. americana individuals were removed from the carcasses and sexed at 2-, 4-, and 6-day intervals (three replicates each). Each carcass was only sampled once to avoid influencing future sex ratio counts. Means ± standard errors are presented throughout.
Effects of sex ratio and size on the guarding duration
Mate-guarding experiments with varied sex ratios were conducted during the summers of 2002 and 2003. A clear plastic box (41.5 x 20 x 15 cm) filled with 9 cm of soil and a gray squirrel carcass was used as a breeding resource. Males and females were individually marked. Male and female beetles were placed simultaneously in the box and observed by X-10 surveillance video during daylight hours. The total observation time for each replicate was approximately 7500 min over a 7-day period, but males and females were not always visible. Females were in the soil during oviposition bouts that lasted approximately 2–4 h, and both males and females were often feeding under or inside the carcass.
This experiment consisted of three sex ratio treatments with six individuals present in each box: male biased (4:2), female biased (2:4), and equal (3:3). In male-biased sex ratio, four males with an approximate 0.3-mm difference in the pronotal width were selected to give a variety of male sizes ranging from very small (8.40 mm) to very large (9.90 mm). One large female and one small female were used in the male-biased sex ratio to test whether female size had an effect on the riding duration when male-male competition was high. In female-biased sex ratio, four females with an approximate 0.3-mm difference in the pronotal width were chosen to give a variety of female sizes ranging from very small (8.40 mm) to very large (9.95 mm). Males in these replicates were approximately the same size (mean = 9.16 mm). In equal sex ratio, female size was varied similar to the female-biased sex ratio replicates, but participating males were approximately the same size.
There were three replicates of each treatment in which the total riding time (min) and total number of takeovers were determined for each male and female present. Means ± standard errors are presented. Each replicate was monitored until reproduction ceased, which varied from 7 to 9 days. The first 7 days of each replicate were used for statistical analysis. ANOVA (SYSTAT 9) was used to compare total riding time and total takeovers to sex ratio, female size, and male size. Riding time (min) and number of takeovers were dependent variables, while sex ratio, female size, and male size were categorized as independent variables. Tukey's comparisons were used to identify significant differences between sex ratio treatments for total riding time and total takeovers. Regression analysis and analysis of covariance (ANCOVA), within each sex ratio treatment, were used to identify significant trends between pronotal widths of males and females for total riding time and total number of takeovers.
Male mate choice
Three males were placed in a 41.5 x 20 x 15–cm box, containing a gray squirrel carcass, and given a choice of three females with which to mate (N = 10). Each box contained one large (pronotal width > 9.10 mm), one medium (pronotal width = 8.61–9.10 mm), and one small (pronotal width < 8.60 mm) female and one large (pronotal width > 9.40 mm), one medium (pronotal width = 8.91–9.40 mm), and one small (pronotal width < 8.90 mm) male. These boxes were observed for the first 15 min and monitored for 2 min every 15 min for 2 h to note any changes in pairing. Male preferences were analyzed with a chi-square test for independence.
Results
Female fecundity
Oviposition began within 12 h. Bouts were difficult to observe because females were buried in the dirt while depositing eggs. However, females absent for greater than 2 h were presumed to be ovipositing. Females oviposited approximately once a day with an average of three eggs per clutch. There was a highly significant correlation between the pronotal width (mm) of females and the number of eggs deposited in 7 days (Figure 3; F1,7 = 117.57, r = .97, p << .001, R2 = .94).
|
Operational sex ratio
The natural sex ratio collected in pitfall traps was 1.0:0.9 (males:females), N = 273; the average sex ratio on a carcass was 2.1:1.0 and did not change significantly among days 2, 4, and 6. However, the number of individuals present increased over time. Carcasses sampled on day 2 had 3.3 ± 0.9 individuals present (2.0:1.3), those sampled on day 4 had 7.3 ± 0.9 individuals present (5.0:2.3), and those sampled on day 6 had 6.6 ± 0.3 individuals present (4.6:2.0).
The effect of male size and sex ratio on the guarding duration of males
Large males, in male-biased sex ratio, guarded females significantly longer (1560 ± 48 min; ANCOVA, F1,8 = 22.42, p = .001) and performed more takeovers (7.8 ± 0.31; ANCOVA, F1,8 = 24.55, p < .001) than small males (riding time = 40 ± 7 min and takeovers = 1.2 ± 0.7) (Figure 4a,b). At equal sex ratio, although males were of similar size, the largest male of each group guarded significantly longer than the smaller males (Figure 4c; ANCOVA, F1,5 = 13.32, p = .02).
|
Large males in male-biased sex ratio and all males in equal sex ratio were predicted to monopolize females, and there was no significant difference in the guarding duration (1560 ± 48 versus 1533 ± 49 min, Tukey's p = .90). However, both guarded significantly longer than males in female-biased sex ratio in which guarding was predicted to be less profitable (1163 ± 88 min, Tukey's p < .001, ANOVA, F2,18 = 23.65). Males in female-biased sex ratio, however, performed significantly more takeovers (4.8 ± 0.6) than males in equal sex ratio (1.2 ± 0.5; ANOVA, F1,13 = 42.41, p < .001).
The effect of female quality and sex ratio on the guarding duration for females
At all three sex ratios (Figure 5), larger females were guarded for significantly longer durations than smaller females (male-biased sex ratio ANCOVA, F1,2 = 988.86, p = .001; equal sex ratio ANCOVA, F1,5 = 23.35, p = .005; and female-biased sex ratio ANCOVA, F1,8 = 50.13, p < .001). Larger females experienced significantly more takeovers than smaller females only in female-biased sex ratio (ANCOVA, F1,8 = 45.42, p < .001).
|
Females experienced significantly lower guarding durations in female-biased sex ratio (574 ± 106 min) than in male-biased (1599 ± 44 min) or in equal sex ratio (1532 ± 39 min; ANOVA, F2,22 = 42.98, p < .001). Females experienced significantly more takeovers in male-biased sex ratio (9.8 ± 0.3) than in equal (1.1 ± 0.3; Tukey's p < .001) or female-biased (2.4 ± 0.5; Tukey's p < .001, ANOVA, F2,24 =100.11) sex ratio. When females returned from an oviposition bout, the previous guard often resumed riding behavior.
Assortative mating
Size-assortative mating was demonstrated when the largest male guarded the largest female in 9 out of 10 trials, the medium male guarded the medium female in 8 out of 10 trials, and the smallest male guarded the smallest female in 9 out of 10 trials (
2 = 32.4, p < .001).
Predicted versus observed guarding duration
Our model predicts a linear relationship because it assumes that all males would have some success at guarding. The factor S/(M–1) in the model represents the probability that a male can guard a female until she oviposits. However, the experimentally derived riding times demonstrated that the large males were more effective at excluding smaller males than the model predicted (Figure 6), and the probability for a given male, in fact, is either one or zero depending on the OSR and his size relative to other males. The empirical data for riding time demonstrated a step function in all male-biased sex ratio replicates because the two largest males monopolized the two females present. In these carrion beetles, even a small difference in size appears to predict relative RHP and provides accurate information to competitors. At equal sex ratio when male size differed little, there is agreement between the predictions of the model and the empirical data. The predicted linear relation of relative male size and guarding time can be seen in Figure 4c.
|
Estimating P2
In addition to male size, the model also considered last-male sperm precedence as a factor in the profitability of mate-guarding behavior. By utilizing empirically determined values for clutch size of the females present, relative sizes of the males present in the experiment, and the OSR from the actual male-biased sex ratio replicates conducted in this study, we solved for the P2 value that would make guarding profitable (Figure 7). In male-biased sex ratio, guarding a large female is profitable when P2 is greater than 60% for males of all sizes (Figure 7a). However, guarding a smaller female is only profitable if P2 is greater than 75% (Figure 7b). In equal sex ratio, it is profitable to guard large females when P2 is greater than 65% and small females when P2 is greater than 80%. In female-biased sex ratio, it is only profitable to guard if P2 is greater than 70% for large females and greater than 85% for small females. These values for P2 are well within the sperm precedence values determined for many insects (60–80% in most insects, Alcock, 1994
; Simmons, 2001; Simmons and Siva-Jothy, 1998). Last-male sperm precedence in experimental double matings for other silphid beetles is 91% (Nicrophorus orbicollis, Trumbo and Fiore, 1991) and 92% (Nicrophorus vespilloides, Müller and Eggert, 1989).
|
Discussion
Testing the model
The following assumptions for the model were supported by this study: large females are more fecund than small females and small males do not succeed in guarding females when competition is intense (male-biased sex ratio). Empirical results support the model predictions; large males guarded longer than smaller males, large females were guarded longer than small ones, and large males also monopolized large females. When male size was varied in male-biased sex ratio replicates, the largest males also performed the most takeovers. When guarding was considered to be the most profitable according to the model (male-biased sex ratio and guarding the largest female), the highest guarding durations were observed for the largest males present in the replicates. Although large males compared well to the predicted profitability of the model, small males underperformed. This inconsistency may be due to an overestimation in the model of the ability of small males to maintain guarding, especially at low densities.
Males guarded females in female-biased sex ratio, albeit for shorter durations, when our model predicted that guarding would not be profitable. They should instead attempt to inseminate all females present. Takeovers were fairly frequent in this treatment which may demonstrate that males were attempting to inseminate all females but were riding females in the interim. Furthermore, the observed OSR on carcasses was always male biased, so presumably, beetles do not face this condition in nature.
Our model also overpredicted the P2 threshold to promote guarding. In our model we took an average clutch size to be 7.5 eggs and treated mate guarding between each oviposition bout independently. We found, however, that partnerships were consistent for the duration of the experiment, and the payoff for guarding was, in fact, much higher. Multiple clutches, rather than a single clutch, lowered the P2 threshold that made guarding profitable (see also Carroll, 1993).
Other models have been developed to predict mate guarding in various invertebrate species. For the most part, they consider similar factors as our model, and predictions from these models are in general agreement with those made by ours. Differences occur in the assumption of remating by females and, if females do remate, if P2 is empirically derived or estimated by the model. A model of mate guarding for the milkweed beetle (Dickinson, 1995) analyzes the reproductive success of guarding and nonguarding males using latency to remating and simulated levels of P2. Dickinson predicts from her empirical data that guarding reduces the probability of female remating and therefore increases the reproductive fitness of guarding males for all P2 values above 60%. Carroll (1993) measured P2 of soapberry bugs at 60–70%. In both studies, mates met in large aggregations. Although our model predicts that males should guard only when P2 > 80%, carrion beetle density on carcasses is low; a greater number of male competitors would decrease our predicted P2. Carroll's model, like ours, predicts that guarding should increase in male-biased populations, and it becomes more profitable as more eggs are oviposited.
Models that consider species of singly mating females (Grafen and Ridley, 1983; Härdling et al., 2004) also predict, like ours, that guarding will increase in male-biased populations and that large males will be choosy. Härdling et al. (2004) clarifies that high-quality males will only be choosier than lower quality males when takeovers are relatively easy. This would be the case when RHP was greater in large males than small males, as is the case for carrion beetles.
The roles of size and sex ratio in guarding success
In male-biased sex ratio, when male-male competition was high, riding times and takeovers were high but only for large males, as predicted. In equal sex ratio, riding time was high but takeovers were low, and in female-biased sex ratio, riding times were lower but more takeovers occurred. In all sex ratio treatments, males preferred larger, more fecund females, and smaller females in each replicate were guarded for shorter durations than were larger females. Large females also experienced more takeovers in male-biased and female-biased sex ratios. These results indicate that large males are monopolizing large females and successfully guarding females in high male-male competition, therefore presumably earning them much greater paternity and reproductive success than smaller males.
Replicates at equal sex ratio and female-biased sex ratio had some unexpected results. Male-male competition is lower in both than in a male-biased sex ratio; males in equal sex ratio did perform fewer takeovers than males in male-biased sex ratio. However, riding times might have been lower in equal sex ratio but were not. Males in equal sex ratio tended to find a female and guard her for the duration of the experiment, and because other males present were similarly occupied, there was little switching of partners. In male-biased sex ratio, females would often return from an oviposition bout, be discovered by a smaller male first, and the former guard (usually larger) would quickly displace the usurper. This scenario accounts for the greater number of takeovers observed for large males in male-biased sex ratio. In female-biased sex ratio, there were significantly more takeovers than in equal sex ratio because males had the opportunity to switch around because they were guarding for shorter durations.
Many studies have investigated the factors of female quality and OSR on guarding duration and have found similar results when ecological conditions are similar to those of N. americana. Long guarding durations, similar to those found in N. americana, have been noted in species that use the same site for mating and oviposition because the cost of guarding is reduced when males do not have to accompany females to their oviposition sites (staphylinid beetle: Alcock, 1991; tiger beetle: Shivashankar and Pearson, 1994). The scarcity of females in low population densities may also promote long riding times because the cost of forfeited matings is reduced when males are unlikely to find additional females to inseminate (tropical bug: Carroll and Loye, 1990; snapping shrimp: Matthews, 2002). Many studies support an increase in the guarding duration in a male-biased OSR (tiger beetle: Shivashankar and Pearson, 1994; firebug: Schofl and Taborsky, 2002; water strider: Clark, 1988). If mate guarding is increased by a male-biased sex ratio, then the opposite trend is expected with female-biased sex ratios. Matthews (2002) illustrated that male Alpheus angulatus, in female-biased sex ratios, were significantly more likely to abandon recently mated females than in equal sex ratios. Male preferences for large females (blister beetle: Brown and Stanford, 1992) and size-assortative mating (blister beetle: Brown, 1990; curculionid beetle: Larsson, 1988 and Harari et al., 1999; longhorn beetles: McLain and Boromisa, 1987) have also been demonstrated in low population densities. Because large males can effectively displace smaller males (weevil: Harari et al., 2003; curculionid beetle: Larrson, 1988; snapping shrimp: Matthews, 2002), they can monopolize large females through longer copulation durations (curculionid beetle: Larsson, 1988) and achieve greater paternity.
When ecological factors, such as high population density, are present, mate-guarding experiments yield different results. Schofl and Taborsky (2002) found higher guarding durations for firebugs (Pyrrhocoris apterus) when the OSR was male biased, but it was the small males that had higher riding times. In high-density aggregations, large males have a great advantage over small males because they can takeover any female they encounter. Because small males are unlikely to find another female that is unguarded, it may be in their best interest to mount females when they locate one, especially if the riding male has an advantage of ownership in takeover attempts. It is probably advantageous for large males in high-density aggregations to abandon females more quickly and inseminate additional females because the profitability of guarding is reduced when many females are present and many matings are forfeited. The nonguarding male's share of P1 is reduced at high population density when females mate with many males but is compensated by more matings. Schofl and Taborsky (2002) also found that there was no relationship between female size and guarding duration, which may also be a consequence of high-density aggregations. When there are many individuals present and male-male competition is high, all females present are likely to have a guard and be guarded until oviposition, regardless of size.
The combination of reproductive synchrony, low density on carcasses, male-biased OSR, and carcass rarity may promote the long riding times and the size-dependent guarding success reported in our study. The lengthy guarding durations especially by large males and the lack of guarding success by small males were surprising until the ecological factors of this system were considered. Small males do attempt riding large females but are displaced by larger males. The lack of intrasexual aggression in N. americana, which is commonly noted in other mate-guarding species (Brown and Stanford, 1992), may be partially due to their low populations densities on carcasses. Males of differing relative sizes may minimize conflict because relative RHP is clear and no contests are needed. These beetles appear to be able to judge even small differences in size accurately. In our replicates when sizes were similar, slightly larger males guarded longer than slightly smaller males.
The model predicted that small males could expect at best a very marginal payoff for guarding. The reproductive benefit that guarding males receive and the lack of reproductive opportunities may explain why small males attempt guarding even though their guarding success is limited. The ecological factors of a short 3-month span to reproduce, carcasses rarity, and low-density populations make it crucial that males of all sizes invest energy and time into securing paternity, through guarding, when the opportunity presents itself.
|
ACKNOWLEDGEMENTS |
|---|
We would like to thank Jonathan Waage, Brett Gibson, S. Carmen Panaitof, Sandra Safryn, and Ryan Marsh for their helpful comments on this project. Joshua Knox provided help with construction and technical support. Two anonymous reviewers and the editor made helpful comments on the manuscript. This work was partially funded by the University of New Hampshire Graduate School to T.T.K.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONTHE MATE-GUARDING MODELTEST OF THE MODELREFERENCES |
|---|
Alcock J, 1991. Adaptive mate-guarding by males of Ontholestes cingulatus (Coleoptera: Staphylinidae). J Insect Behav 4:763–771.[CrossRef]
Alcock J, 1994. Postinsemination associations between males and females in insects: the mate-guarding hypothesis. Annu Rev Entomol 39:1–21.[CrossRef][Web of Science]
Brown WD, 1990. Size-assortative mating in the blister beetle Lytta magister (Coleoptera: Meloidae) is due to male and female preference for larger mates. Anim Behav 40:901–909.[CrossRef][Web of Science]
Brown WD, Stanford R, 1992. Male mating tactics in a blister beetle (Coleoptera: Meloidae) vary with female quality. Can J Zool 70:1652–1655.[CrossRef]
Carroll SP, 1993. Divergence in male mating tactics between two populations of the soapberry bug: I. Guarding versus nonguarding. Behav Ecol 4:156–164.
Carroll SP, Loye JE, 1990. Male biased-sex ratios, female promiscuity, and copulatory mate guarding in an aggregating tropical bug, Dysdercus bimaculatus. J Insect Behav 3:33–48.
Clark SJ, 1988. The effects of operational sex ratio and food deprivation on copulation duration in the water strider (Gerris remigis Say). Behav Ecol Sociobiol 23:317–322.[CrossRef][Web of Science]
Dickinson JL, 1995. Trade-offs between postcopulatory riding and mate location in the blue milkweed beetle. Behav Ecol 6:280–286.
Grafen A, Ridley M, 1983. A model of mate guarding. J Theor Biol 102:549–567.[CrossRef]
Harari AR, Handler AM, Landolt PJ, 1999. Size-assortative mating, male choice and female choice in the curculinoid beetle Diaprepes abbreviatus. Anim Behav 58:1191–1200.[CrossRef][Web of Science][Medline]
Harari AR, Landolt PJ, O'Brien CW, Brockman HJ, 2003. Prolonged mate guarding and sperm competition in the weevil Diaprepes abbreviatus (L.). Behav Ecol 14:89–96.
Härdling R, Kokko H, Elwood RW, 2004. Priority versus brute force: when should males begin guarding resources? Am Nat 163:240–252.[CrossRef][Medline]
Larsson FK, 1988. Experimentally induced density-dependent mating patterns in Phyllobius maculicornis Germ. (Coleoptera: Curculionidae). Am Midl Nat 119:380–384.[CrossRef]
Matthews LM, 2002. Tests of the mate-guarding hypothesis for social monogamy; does population density, sex ratio, or female synchrony affect behavior of male snapping shrimp (Alpheus angulatus). Behav Ecol Sociobiol 51:426–432.[CrossRef][Web of Science]
McLain DK, Boromisa RD, 1987. Male choice, fighting ability, assortative mating and the intensity of sexual selection in the milkweed longhorn beetle, Tetraopes tetrapthalmus (Coleoptera: Cerambycidae). Behav Ecol Sociobiol 20:239–246.[CrossRef][Web of Science]
Müller JK, Eggert AK, 1989. Paternity assurance by "helpful" males: adaptations to sperm competition in burying beetles. Behav Ecol Sociobiol 24:245–249.[CrossRef][Web of Science]
Safryn SA, Scott MP, 2000. Sizing up the competition: do burying beetles weigh or measure their opponents? J Insect Behav 13:291–297.[CrossRef]
Savalli UM, Fox CW, 1998. Sexual selection and the fitness consequences of male body size in the seed beetle Stator limbatus. Anim Behav 55:473–483.[CrossRef][Web of Science][Medline]
Schofl G, Taborsky M, 2002. Prolonged tandem formation in firebugs (Pyrrhocoris apterus) serves mate guarding. Behav Ecol Sociobiol 52:426–433.[CrossRef][Web of Science]
Shivashankar T, Pearson D, 1994. A comparison of mate guarding among five syntopic tiger beetles species from peninsular India (Coleoptera: Cicindelidae). Biotropica 26:436–442.[CrossRef][Web of Science]
Simmons LW, 2001. Sperm competition and its evolutionary consequences in insects. Princeton: Princeton University Press.
Simmons LW, Siva-Jothy MT, 1998. Sperm competition in insects: mechanisms and the potential for selection. In: Sperm competition and sexual selection (Birkhead TR, Møller AP, eds). San Diego: Academic Press; 341–434.
Trumbo ST, Fiore AJ, 1991. A genetic marker for investigating paternity and maternity in the burying beetle Nicrophorus orbicollis (Coleoptera: Silphidae). J N Y Entomol Soc 99:637–642.
Wen YH, 1993. Sexual dimorphism and mate choice in Hyalella azteca (Amphipoda). Am Midl Nat 129:153–160.[CrossRef]
Zeh J, Zeh D, 1994. Last male sperm precedence breaks down when females mate with three males. Proc R Soc Lond B 257:287–292.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||