| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Behavioral Ecology Vol. 15 No. 2: 278-285
Behavioral Ecology vol. 15 no. 2 © International Society for Behavioral Ecology 2004; all rights reserved
Uncertainty in territory quality affects the benefits of usurpation in a Mediterranean wrasse
Institute of Marine Sciences, University of California, Santa Cruz, 1156 High Street, Santa Cruz, CA 95064, USA
Address correspondence to S. H. Alonzo. E-mail: shalonzo{at}ucsc.edu.
Received 29 January 2002; revised 14 August 2002; accepted 1 May 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Individuals should defend sites when the expected benefits of the territory exceed the cost of defense. However, if territory qulaity is unpredictable or difficult to assess, the expected pattern of territorial behavior is less clear. In a Mediterranean wrasse, Symphodus ocellatus, mating success is skewed with 2% of nesting males getting more than 20% of the spawning success. Yet, variation in mating success is not explained by any known physical characteristic of males or their territories. Instead, females prefer nests with a recent history of mating success because males are less likely to desert the offspring she leaves behind. Thus, territory quality is transient and determined by interactions between the sexes. I measured the frequency of territorial takeovers and the uncertainty in mating success among days at a nest. Observations indicated that S. ocellatus males usurped their neighbor's successful nests when males were unsuccessful and larger than their successful neighbor. Sites that achieved mating success had a significantly higher probability (0.84) of remaining sucessful between consecutive days than unsuccessful territories had of becoming successful (0.30). Unsuccessful males obtained higher and more certain fitness returns if they usurped a successful neighbor's territory. Interactions within and between the sexes drive uncertainty in success, which influences territorial behavior in this species.
Key words: territoriality, stochasticity, labridae, sexual conflict, assessment.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Individuals should defend sites or take over territories when the expected benefits of the territory exceed the costs of defense or usurpation (see Brown, 1964
; Calsbeek et al., 2002; Carpenter and MacMillen, 1976; Davies, 1978; Davies and Houston, 1984; Gill and Wolf, 1975; Noble, 1939; Piper et al., 2000; Ruxton et al., 1999; Schoener, 1983, 1987; Stamps and Krishnan, 2001; Switzer, 1993). Theoretical predictions have focused on cases in which resources are fixed or animals have good information about the costs and benefits of resource defense (see Carpenter and MacMillen, 1976; Davies, 1978; Davies and Houston, 1984; Gill and Wolf, 1975; Schoener, 1983, 1987; but see Ruxton et al., 1999; Stamps and Krishnan, 2001; Switzer, 1993). However, studies have shown that resources at a site can vary, information about resources may be incomplete, and the costs and benefits of resource defense are not always predictable (see Bollman et al., 1997; Nager et al., 2000; Orians and Wittenberger, 1991; Ruxton et al., 1999; Stamps, 1994; but see Rendon et al., 2001). This unpredictability can greatly influence territorial behavior.
When animals defend stable and predictable resources, a positive correlation is expected between the territory defender's quality (resource holding potential [RHP]) and the quality of the resource (Maynard Smith and Parker, 1976). Although examples exist in which this pattern has been found (see Alcock, 1987; Alcock et al., 1977; Andersson, 1994; Bart and Earnst, 1999; Brown, 1964; Campanella and Wolf, 1974; Jones, 1981; Petrie, 1984; Searcy and Kasukawa, 1989; Severinghaus et al., 1981; Shutler and Weatherhead, 1991; Wells, 1977), in other cases the relationship has not been supported (see Beletsky and Orians, 1993; Cordera et al., 1996; Pribil, 1998). These results could arise from unpredictable variation in the critical resource, making it difficult for high-quality individuals to defend the best territories. For example, in yellow-headed blackbirds, the food resources important to offspring growth and survival are not known until after territories are established (Orians and Wittenberger, 1991). As a result, females cannot predict the territory that will have the highest-quality food (Nager et al., 2000; Orians and Wittenberger, 1991). In this case, no correlation between individual and territory quality would be expected because territory quality is unpredictable. In other systems, the quality of the territory will vary owing to interactions between individuals. This will be especially common in cases in which territories are defended for reproduction. For example, the success of a male's reproductive territory may depend not only on resources intrinsic to the site but also on female choice among territories. If female choice among territories varies, a strong relationship may not exist between territory or resource quality and male RHP because the "quality" is dynamic rather than fixed.
In a Mediterranean wrasse, Symphodus ocellatus, males defend sites against conspecific males and predators (Lejeune, 1985; Taborsky et al., 1987; Ylieff et al., 1997, 1998). Males go through a series of nest cycles in which they construct nests out of algae (1–4 days), spawn with females (2–4 days), and then usually fan and defend eggs (2–7 days) before leaving the nest site and starting another nest cycle (Lejeune, 1985; Taborsky et al., 1987). Females visit nests to spawn and appear to choose among nests (Alonzo and Warner, 2000a; Taborsky et al., 1987; Wernerus, 1985, 1988; Wernerus et al., 1989), but these territories do not offer any food resources to the female or her offspring. Instead, for females, the critical resource at these territories is male parental care. Males that have experienced high mating rates tend to remain with their nests through the parental care phase of the nest cycle, whereas males with low mating success often desert the nest before the eggs can develop (Taborsky et al., 1987). For males, territory quality is determined by their ability to attract females to their nest. Yet, females are only attracted to nests in which other females are spawning. Therefore, the critical resources affecting perceived quality of the territory are dynamic depending on interactions within and between the sexes.
In this species, alternative male strategies exist with large colorful males defending nests and small drab males sneaking spawns (Taborsky et al., 1987). Growth data from otoliths indicate that sneakers and nesting males represent separate life histories (Alonzo et al., 2000). Females appear to avoid spawning with small sneaker males. However, females usually spawn at nests with sneakers present because nests with high mating rates lead to nest fidelity by males and also attract sneakers (Alonzo and Warner, 2000a; van den Berghe et al., 1989).
In S. ocellatus, the distribution of mating success among nesting males is not explained by any measured behavioral or physical characteristics of the nest or nesting male, individual male identity, or nest site (Lejeune, 1985; Wernerus, 1985, 1988; Wernerus et al., 1987, 1989), and past research indicates that success is unpredictable before the nest cycle (Wernerus 1985, 1988; Wernerus et al., 1987, 1989). In contrast, because females are attracted to successful nests, less uncertainty may exist during a cycle. If territory quality is predictable after but not before the establishment of territories, an unsuccessful male could increase his mating success by taking over the nest of a nearby successful male, and nest takeovers should be common. By using a combination of field observations and experiments, I examined the predictability of nest success during the nest cycle, the conditions under which nest takeovers occur, and the benefits of usurpation by nesting males in S. ocellatus.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The study site and species
Research was conducted May–June 1997 under natural conditions near the University of Liège (Belgium) Marine Laboratory, La Station de Recherches Sous-Marine et Océanographique (STARESO), located near Calvi, Corsica, France. I made all observations on rocky substrate along 200 m of shore in water less than 15 m deep by using scuba during approximately 100 h underwater. In 1997, before the beginning of the reproductive season, I caught and individually marked 128 nesting males by using a subcutaneous injection of Alcian blue, a vital stain, and measured their standard length to the nearest 0.1 mm. This marking method is less invasive than a tattoo, and marked individuals return within minutes to their nests. Nesting males have a distinct color pattern and are larger than all other individuals in the population, which easily distinguishes them from females and sneaker males (Lejeune, 1985
; Taborsky et al., 1987). Females can be distinguished from sneaker males by the absence of a brightly colored spot on their opercules, by the shape of their abdomen and genital papillae, and by the absence of aggression toward them by the nesting male (Lejeune, 1985; Taborsky et al., 1987). Nests occur within meters of one another, and both females and sneakers move freely between nests. Experiments have shown that sneakers and females respond to changes at the nest and redistribute themselves among nests within 30 min (Alonzo and Warner, 1999; Alonzo and Warner, 2000a). In contrast, nesting males cannot move freely between nests because of the territorial behavior of other nesting males and the risk of losing their current nest.
Measures of mating success
Finding a suitable measure of male mating success was critical to understanding territorial behavior in this species. Spawning success would seem to be the most direct measure of success, yet spawns were clumped in time. Given that observations had to be conducted by using scuba, this limited the amount of time I could spend observing each nest. I therefore considered using other measures of spawning success. My criteria for picking a suitable measure was that it showed high repeatability between observations of the same male within a short period of time and, therefore, low within-male variance compared with between-male variance. I used data I collected in earlier studies to test for a suitable measure. First, I analyzed data from 30-min observations of 28 nests taken in May–June 1996 (Alonzo and Warner, 1999). Each 30-min session was separated into 30 consecutive 1-min observations for each individual nest. In each minute, I tallied the total spawns and spawns that occurred without any sneakers (pair spawns). I also recorded the number of males with the sneaker phenotype present in the vicinity of the nest (within 3 m of the nest and oriented toward the nest) at the start of each 1-min interval. These data were analyzed by using ANOVA. I found that the among-male variance in total spawns (among-male sum of squares = 222.13) was much less than the within-male variance (within-male sum of squares = 1163.47). The same pattern existed for variation within- and between-males in pair-spawns (among-male sum of squares = 71.50, within-male sum of squares = 377.00). Thus, 1-min observations create very large sampling error on the direct measures of mating success. By contrast, variation among males in the number of sneakers at the nest was greater than the variation within males among observations (among-male sum of squares = 2907.96, within-male sum of squares = 1947.70). As a result, within-male variation in spawning was greater than within-male variation in the number of sneakers present at the nest. The repeatability of the number of sneakers at the nest (0.59) is four times greater than the repeatability of either the total number of spawn at the nest (0.14) or the number of pair spawns at the nest (0.14). This suggests that the number of sneakers might be a more reliable measure of mating success than is the number of spawns given the constraints on observation time.
This is, however, a counter-intuitive measure of mating success, as sneakers surely reduce fertilization success per spawn of the nest-holding male. Nevertheless, the number of sneakers is positively related to spawning success. This finding comes from several sources. First, past studies have shown that number of sneakers present at a nest, and the total number of spawns at the nest are significantly and positively correlated (Lejeune, 1985; r =.91, p value not available; Wernerus et al., 1989 r =.37 N = 39 p =.02). To confirm this for my study, I examined 10-min observations of 88 randomly selected nests in May and June of 1994 (Alonzo SH, unpublished data). I counted the number of sneakers at each nest every minute over the 10-min period and averaged that value. I also recorded the total number of spawns and the number of spawns not joined by sneakers (pair spawns) in each observation. Although the total number of spawns at the nest may not indicate the mating success of nesting males because of the effect of shared paternity with sneakers, the number of pair-spawns is a measure of known fertilizations. In fish species with external fertilization, it is unlikely that sperm from previous spawning events will fertilize eggs in later spawning events, making each spawning event independent.
I found a significant and positive relationship between the number of sneakers present at the nest and total spawns (r2 =.16, ß = 0.92, F = 16.81, p <.001) as well as with pair spawns (i.e., just the nesting male and a female mating, r2 =.07, ß = 0.44, F = 6.33, p =.01). A two-parameter Ricker equation (a logistic function with diminishing returns, y = 
x/[1 + ßx]) also produced a strong positive relationship between number of spawns and the presence of sneakers at the nest (r2 =.25, = 1.92, ß = 4.20, F = 28.7, p <.001) as well as between pair spawns and sneakers at the nest (r2 =.27, = 1.84, ß = 4.12, F = 32.3, p <.001). Although the relationship appears to be nonlinear, the number of sneakers present at the nest always corresponds with an increase in the mating success of nesting males. Furthermore, nests with sneakers and nests without sneakers have significant and large differences in total spawns (nests with sneakers: 10.58 [SE ± 1.66], n = 38; nests without: 2.60, n = 50 [SE ± 0.71], t = 4.41, p <.001) and pair spawns (nests with sneakers: 7.21 [SE ± 1.23], n = 38; nests without: 2.40 [SE ± 0.71], N = 50, t = 3.48, p <.001). Thus, the presence or absence of sneakers is a good indicator of being successful or unsuccessful in general. In S. ocellatus, although relative paternity (the percentage of eggs fertilized) will invariably decrease as the number of sneakers present at the nest increases, the total paternity (the number of eggs fertilized) nesting males obtain under natural conditions is associated with more sneakers present at the nest because sneakers are attracted to a higher frequency of spawning. Furthermore, the slope of the regression of spawns joined by sneakers on total spawns was 0.30 (r2 =.6, p <.001), indicating that sneakers do not overwhelm nesting males as the spawning rate increases.
In summary, the number of sneakers associated with a nest is a proxy of mating success and is a more reliable measure of the success a male may obtain given the constraints I faced in observation time per nest. Thus, I used number of sneakers at a nest as a measure of nest success in all subsequent analyses.
Nest surveys
In 1997, nests were surveyed daily to determine the occurrence of nest takeovers and the uncertainty in mating success at a nest between days. Nest cycles are not synchronous, and males always leave their nest at the end of the nest cycle and usually build a new nest on a nearby site (Wernerus et al., 1989). I surveyed the entire research area daily and placed numbered markers by all nests. I removed the markers only after the nest had been deserted for three consecutive days. I observed all nests within the study area daily for 1 min to determine the activity of the nesting male (building, spawning, or fanning), the individual mark of the nesting male, and the number of sneakers at the nest. This species is reproductively active during daylight, and past studies have not found any variation in mating rate with time of day (Lejeune, 1985). Nest surveys were conducted during the same period in the morning daily for consistency. I designated males as building when they carried and put algae into their nest. Males in the building phase of the nest cycle do not mate. In contrast, males in the spawning phase of the nest cycle actively court and spawn with females. Males in the fanning phase of the nest cycle use their pectoral fins to fan the surface of the nest, do not add algae to the nest, and do not spawn with visiting females. I determined the number of sneakers present at the nest by counting the number of sneakers oriented toward and within 3 m of the nest. Nests were surveyed daily unless weather conditions did not permit diving (only 3 of 53 days). The number of nests within a 10-m radius was measured after the nest cycle had been completed or the nest was deserted. From the survey data and these measurements, it was possible to determine the average number of active nests found within 10 m during the spawning period of each nest. This measure was used to indicate local density and the number of successful neighbors within 10 m for each nest. Nests can be less than 1 meter to 10s of meters apart, and individual nests may have from zero to eight neighbors in a 10-m radius.
Determining the occurrence of nest takeovers
Nests are abandoned at the end of the nest cycle or if they are unsuccessful (Taborsky et al., 1987). These nests are usually left unattended by other males after desertion. Therefore, I inferred that a nest had been taken over when the male changed between consecutive days. Aggressive interactions between nesting males are commonly observed (Taborsky et al., 1987), and abandonment of successful nests is rare (Taborsky et al., 1987). Takeovers could only be recognized if at least one of the males was marked. When a marked male took over another male's territory, I compared the number of sneakers present on consecutive days at the deserted nest with the nest taken over. I also compared the number of sneakers at the deserted nest with the nest taken over on the day before the takeover. I also compared the standard length of the original male and the usurping male. Standard length and the number of sneakers before and after takeovers were compared using paired t tests because differences did not deviate significantly from normality according to a Shapiro-Wilks test (Shapiro and Wilk, 1965; Zar, 1996). Because I expected larger males to take over the nests of smaller males and all males to take over nests with more sneakers, these tests were directional with 
= 0.1 and 
= 0.4 and = 0.05 (Rice and Gaines, 1994). Finally, the number of sneakers present at the nest on the day before and after the takeover were compared by using a sign rank test (because the data could not be normalized using any of the standard transformations) to determine any effect of a change in male identity on mating success (as measured by the number of sneakers present at the nest). By using logistic regression and forward model selection (Tabachnik and Fidell, 2001), I examined the effect of the maximum number of sneakers present during the cycle, local density of nests during the spawning phase of the nest cycle, number of unsuccessful neighbors during the spawning phase of the cycle, size of the original male, male identity, day in the season, and length of the nest cycle on the probability of a successful takeover occurring.
Calculating uncertainty in success within the nest cycle
Because of the skewed distribution of mating success observed in this species (Alonzo and Warner, 2000a), males tend to fall into two main categories: Males at nests with no sneakers and low to no mating success (unsuccessful males) and males at nests with sneakers and higher mating success (successful males). I determined the number of between-day transitions from being successful or unsuccessful (as determined by the presence or absence of sneakers) within the spawning phase of a single nest cycle. By using the survey data, I determined the number of the four possible between-day transitions in success. Whenever multiple transitions were observed for a nest, one transition was chosen at random to create a set of independent observations. I then examined whether some transitions were significantly more probable than others by using a two-by-two 
2 test for independence on this subsample (Siegel and Castellan 1988). I also examined the effect of day in the cycle on these transition probabilities by calculating the daily probability of each transition for the first 5 days of spawning and testing for an effect of day in the nest cycle. By using a randomization test (1000 iterations; Good, 2000), I calculated the 95% lower and upper confidence limits for the prediction that the frequency of each of the four transitions is independent of day in the nest cycle. I also examined the predictions that the conditional transition probabilities (e.g., unsuccessful tomorrow given successful today) and the frequency of successful and unsuccessful nests were independent of day in the nest cycle by using similar randomization tests.
Removal experiment
Although past research indicated that male identity does not explain any of the skew in mating success, I conducted one experiment to control for the effect of male identity on expected success following a takeover. In 1997, I experimentally removed eight males from successful nests and compared the success of the original and replacement male. Before removals, I observed these nests for 10 min and determined the average number of sneakers at the nest (as defined above), the number of females visiting (within 10 cm of the nest), and the total number of spawns. After the first observation, the nesting male was caught by using a small aquarium net and held in a live bait bucket. I observed the nest a second time 30 min after the original male's removal. I compared each of the observation variables by using a paired t test because the differences did not deviate significantly from Normal according to a Shapiro-Wilks test (Shapiro and Wilk, 1965; Zar, 1996). All tests were two-tailed. The minimum detectable difference of the t test was also calculated for each variable (with = 0.05 and 1 - ß = 0.90; Zar, 1996) to examine the ability of the small sample to detect any differences in number of spawns, number of female visits, and the number of sneakers present at the nest.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Takeovers of nest sites
A total of 128 males were marked before the reproductive season, and 59 of those males were seen defending a territory within the research area. On average 62% (SE ± 0.023) of males defending nests on any day within the study area were marked. Two hundred forty-nine nests were observed throughout the 2-month period (Figure 1). In this time, I observed 31 successful takeovers at 30 nests: eight takeovers in which an unmarked male was taken over by a marked male, six when a marked male was replaced by an unmarked male, and 17 transitions between marked males. In 14 of the 25 cases in which a marked male took over a new nest, the same male had been observed at a different nest at the study site on the previous day. Of these 14 males taking over nests, 12 out of 14 of them were originally at nests with zero sneakers present, and the remaining two were at nests with one sneaker present. The number of sneakers at the nest taken over was significantly higher than at the deserted nest (deserted nest: 0.148 SE ± 0.10; nest taken over: 4.43 SE ± 0.89; Shapiro-Wilks: W13 = 0.93, p =.27; paired t test: t13 = 4.77, N = 14, p <.001), which implies the usurping male gained by taking over the site. Similarly, the number of sneakers at the nest taken over on the day before the takeover (with the original male) was significantly higher than at the usurping male's original nest on the same day (deserted nest: 0.148 SE ± 0.10; nest taken over: 5.30 ± SE 1.33; Shapiro-Wilks: W13 = 0.92, p =.20; paired t test: t13 = 3.78, N = 14, p =.001). Males that successfully usurped another male were significantly larger than was the original male (original male: 75 mm ± SE 0.92; male taking over: 77.7 mm ± SE 0.71; Shapiro-Wilks: W17 = 0.97 p =.67; paired t test: t17 = 2.00, N = 17, p =.03). The probability of a takeover occurring was increased by day in the season, length of the nest cycle, and the maximum number of sneakers found at the nest (best-fit model predicting the probability of a takeover: N = 204 nests; maximum number of sneakers: ß = 0.086,
2 = 2.10, p =.148; day in the season (that nest building began): ß = 0.087 2 = 9.32 p =.002; length of the nest cycle: ß = 0.157, 2 = 7.73 p =.005; intercept = -3241, 2 = 9.34 p =.002). Local density, number of successful neighbors during the spawning period, and original male size or identity did not explain any variation in the probability of a nest takeover. Nests that experienced takeovers had longer spawning phases overall (without takeovers, 4.44 ± 0.21 days; with takeovers, 5.72 ± 0.53; two sample t test: t = -3.19, N = 246, p >.0016) but did not have longer spawning phases than other successful nests (i.e., nests that attracted sneakers) that did not experience a takeover (without takeovers, 5.70 ± 0.22; with takeover, 5.71 ± 0.53; two sample t test: t = -1.27, N = 178, p =.20).
|
Transition probabilities
I counted the total number of daily transitions between success states (determined by the absence or presence of sneakers) during the spawning phase of the nest cycle for all nests. In the randomly selected subsample, 199 independent nest transitions were observed. To determine the uncertainty involved, I calculated the following conditional probabilities: given that a nest was unsuccessful on day 1, the probability it was also unsuccessful on day 2 was 65/93. The probability of going from unsuccessful to successful across days was 28/93. When successful, the probability of remaining successful was 89/106. The probability of a transition from being successful to unsuccessful was 17/106. These transition probabilities give the uncertainty for a male associated with each state (Figure 2). Future success was not independent of present success state (Figure 2). Transition probabilities between success states varied across the nest cycle (Table 1). There was a statistically significant increase over the nest cycle in the frequency of successful to successful transitions compared with the other three possible transitions (Table 1 and Figure 3). However, this increase in frequency was driven mainly by an increase in the frequency of successful nests later in the nest cycle (because males desert unsuccessful nests) rather than an increase in the conditional probability of remaining successful once already successful (Table 1 and Figure 3).
|
|
|
Removal experiment
In every case, a new male took over the nest within 10 min of a nest owner's removal. This new male actively courted and spawned with females. I found no significant differences in the number of spawns, the number of female visits, or the number of sneakers present at the nest between the first and second observation even though the identity of the nesting male changed between these two observations (Table 2). Despite a small sample size, it would have been possible to detect small differences in the number of female visits and the number of sneakers present at the nest (Table 2). By contrast, the minimum detectable difference in mating success was 10 spawns per 10 min. However, successful nests experience similar levels of variation in mating success even without a change in male identity (Alonzo and Warner 1999
). Consistent with the nest survey results, marked males who took over nests were usually (four out of six cases) males that had nests near the experimental nest. These males also tended to be unsuccessful (i.e., no sneakers observed in five out of six cases).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Nest takeovers are a regular occurrence in S. ocellatus (at least 12% of all nests and a 40% chance per nesting male because males defend about four nests per season). Usurping males moved from nests without sneakers to those with sneakers and therefore probably improved their success. Unsuccessful males usually took over their successful neighbor's nests, and the males that managed to take over nests were larger than the original male. This pattern indicates that, although size is not a very good predictor of overall male mating success in this species (Wernerus et al., 1989
), relative RHP may play a role in takeover ability. Males may be opportunists, taking over nearby nests that are successful if their successful neighbor is smaller. Furthermore, the unpredictability of mating success at a site before the establishment of territories followed by a relatively high certainty of success within a nest cycle (Figure 2) can explain the occurrence of takeovers in this species. This pattern, however, depends on the fact that female behavior is not affected by the identity of the male on the nest. Thus, the relative certainty of success within the nest cycle and uncertainty between nest cycles is a direct result of females choosing nests for parental care certainty rather than individual male or nest characteristics.
The probability of a successful takeover also increased with time in the season and the length of the nest cycle. A few possible explanations exist for a temporal increase in the frequency of takeovers. First, the direct costs or benefits of a takeover may change over the season. For example, the probability of a successful takeover or the skew in mating success could increase during the season. Second, age data from otoliths indicated that all nesting males were 2 years old, and that 3-year-old males were not found in this population (Alonzo et al. 2000). Males may attempt takeovers with a greater probability late in the season as their opportunities for future success decrease. The increased probability of a takeover with longer nest cycles could occur even if all nests are exposed to the same daily risk of takeover. Because males desert unsuccessful nests it is more probable, however, that longer nest cycles are associated with takeovers simply because successful nests are not deserted.
Past research indicates that female behavior drives the skew in mating success, and females are in conflict with nesting males over parental care (Alonzo and Warner, 2000a). Females and sneakers, and sneakers and nesting males are also in conflict because both females and nesting males attempt to avoid spawning with or at the same time as sneaker males (Alonzo and Warner, 1999, 2000b; van den Berghe et al., 1989). The results presented here indicate that nesting males will also be in conflict with other nesting males, owing to competition for successful nest sites after the establishment of territories.
In the current study, I used the presence of sneakers at a nest as a proxy for the expected mating success of nesting males. A positive relationship between the number of sneakers at the nest and pair spawning exists and the number of sneakers at the nest is more repeatable than any direct measure of spawning for short observations. It is also possible that conspecifics may use sneakers as an indicator of nest success. Sneakers can be easily observed from meters away, whereas spawning females can only be seen when very close to a nest. Nesting males could use the presence of sneakers as a cue of a neighbor's success without risking aggression or leaving one's current nest. If sneakers are the cue used to indicate success to neighboring males, the presence of sneakers at the nest may carry the additional cost of allowing successful males to be easily targeted for territorial takeovers.
In this species, it is the pattern of female choice that allows nesting males to take over successful nests and have those nests remain successful. If females preferred physical characteristics of the male instead of transitory mating situations, takeovers would not be beneficial. Therefore, the territorial behavior of males is indirectly influenced by female choice. In this species, female choice generates uncertainty in mating success early in the nest cycle and certainty in success later in the nest cycle. Furthermore, female choice is a direct result of conflict between females and nesting males over nest desertion. Subsequently, understanding patterns of male territorial behavior requires an understanding of interactions within and between the sexes and how they influence the predictability of territory quality.
Variability and uncertainty in territory quality will exist in many species, and patterns of uncertainty should influence territorial behavior. For example, in another Mediterranean wrasse (S. tinca) the benefits of parental care (and thus territory defense) vary over the reproductive season (Luttbeg and Warner, 1999; Warner et al., 1995). Therefore, the perceived quality of a territory will vary through time for males and females. In this species, large "pirate" males occasionally take over successful territories (van den Berghe, 1988), and the frequency of males defending territories varies over the season with territory quality (Luttbeg and Warner, 1999; Warner et al., 1995). In the pied flycatcher, females have incomplete information about the presence of other females on the territory, and whether females join and defend territories should depend on their certainty of success at that territory (Alatola and Lundberg, 1984). In yellow-headed blackbirds, reproduction must occur while the abundance of food resources important to offspring growth and survival are still uncertain (Orians and Wittenberger, 1991). Thus, we would not expect takeovers to occur in this species. Food resources are similarly unpredictable in some great tit populations (Nager et al., 2000), and male competition among sites may be independent of the quality of a territory for raising offspring because of this uncertainty. Territory quality is the outcome of a variety of social interactions and variable environmental conditions. As a result, the benefits of territory defense will often vary, and this variation may or may not be predictable. Understanding patterns of territorial defense will require that we consider both how territory quality varies as well as whether the quality of a site is predictable.
|
ACKNOWLEDGEMENTS |
|---|
I thank all of the staff of STARESO for their endless help in making this research possible and enjoyable. I also thank Jamie Alonzo, Ryan Calsbeek, Marc Mangel, and Katriona Shea for helpful discussions. I thank David Westneat and two anonymous reviewers for their comments on an earlier version of this manuscript. This research was supported by NSF grant IBN-87-00948, an Animal Behavior Society Research Grant to SHA, and a NSF Postdoctoral Fellowship, and a National Science Foundation grant IBN-0110506 to S.H.A.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Alatola RV, Lundberg A, 1984. Polyterritorial polygyny in the pied flycatcher Ficedula hypoleuca: evidence for the deception hypothesis. Ann Zool 21:257-271.
Alcock J, 1987. The effects of experimental manipulation of resources in the behavior of two calopterygid damselflies that exhibit resource defense polygyny. Can J Zool 65:2475-2482.
Alcock J, Jones CE, Buchmann SL, 1977. Male mating strategies in the bee, Centris pallida Fox (Hymenoptera: Anthophoridae). Am Nat 111:145-155.[CrossRef]
Alonzo SH, Taborsky M, Wirtz P, 2000. Male alternative reproductive behaviours in a Mediterranean wrasse, Symphodus ocellatus: evidence from otoliths for multiple life-history pathways. Evol Ecol Res 2:997-1007.
Alonzo SH, Warner RR, 1999. A tradeoff generated by sexual conflict: Mediterranean wrasse males refuse present mates to increase future success. Behav Ecol 10:105-111.
Alonzo SH, Warner RR, 2000a. Dynamic games and field experiments examining intra- and inter-sexual conflict: explaining counter-intuitive mating behavior in a Mediterranean wrasse, Symphodus ocellatus. Behav Ecol 11:56-70.
Alonzo SH, Warner RR, 2000b. Sperm allocation in a Mediterranean wrasse: intrasexual sperm competition drives intersexual conflict. Am Nat 156:266-275.[CrossRef]
Andersson MB, 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Bart J, Earnst S, 1999. Relative importance of male and territory quality in pairing success of male rock ptarmigan. Behav Ecol Sociobiol 45:355-359.[CrossRef]
Beletsky LD, Orians GH, 1993. Factors affecting which male red-winged blackbirds acquire territories. Condor 95:782-791.
Bollman K, Reyer HU, Brodmann PA, 1997. Territory quality and reproductive success: can water pipits, Anthus spinoletta, assess the relationship reliably? Ardea 85:83-98.
Brown JL, 1964. The evolution of diversity in avian territorial systems. Wilson Bull 76:160-169.
Calsbeek R, Alonzo SH, Zamudio K, Sinervo B., 2002. Sexual selection and alternative mating behaviours generate demographic stochasticity in small populations. Proc R Soc Lond B 269:157-164.[Medline]
Campanella PJ, Wolf LL, 1974. Temporal leks as a mating system in a temperate zone dragonfly (Odonata: Anisoptera). I Plathemis lydia (Drury). Behaviour 51:49-87.[Web of Science]
Carpenter FL, MacMillen FE, 1976. Threshold model of feeding territoriality and test with a Hawaiian honey creeper. Science 194:639-642.
Cordera C, Macias R, Jimenez G, 1996. The number of copulations of territorial males of the butterfly, Callophrys xami (Lycaenidae). J Res Lepidoptera 35:78-89.
Davies NB, 1978. Ecological questions about territorial behaviour. In: Behavioural ecology, 1st ed (Krebs JR, Davies NB, eds). Oxford: Blackwell; 317–350.
Davies NB, Houston AI, 1984. Territory economics. In: Behavioural ecology, 2nd ed (Krebs JR, Davies NB, eds). Oxford: Blackwell; 148–169.
Gill FB, Wolf LL, 1975. Economics of feeding territoriality in the golden-winged sunbird. Ecology 56:333-345.[CrossRef][Web of Science]
Good PI, 2000. Permutation tests: a practical guide to resampling methods for testing hypotheses. New York: Springer.
Jones GP, 1981. Spawning-site choice by female Pseudolabrus celidotus (Pisces: Labridae) and its influence on the mating system. Behav Ecol Sociobiol 8:129-142.[CrossRef]
Lejeune P, 1985. Etude écoéthologique des comportements reproducteur et sociaux des Labridae méditerranéens des genres Symphodus (Rafinesque 1810) et Coris (Lacepede 1802). Cah Ethol Appl 5:1-208.
Luttbeg B, Warner RR., 1999. Reproductive decision-making by female peacock wrasses: flexible versus fixed behavioral rules in variable environments. Behav Ecol 6:666-674.
Maynard Smith J, Parker GA, 1976. The logic of asymmetric contests. Anim Behav 24:159-174.[CrossRef]
Nager RG, Keller LF, van Noordwijk AJ, 2000. Understanding natural selection on traits that are influenced by environmental conditions. In: Adaptive genetic variation in the wild (Mousseau TA, Sinervo B, Endler J, eds). New York: Oxford University Press; 95–115.
Noble GJ, 1939. The role of dominance in the social life of birds. Auk 56:263-273.
Orians GH, Wittenberger JF, 1991. Spatial and temporal scales in habitat selection. Am Nat 137:S29-S49.[CrossRef]
Petrie M, 1984. Territory size in the moorhen, Gallinula chloropus: An outcome of RHP asymmetry between neighbors. Anim Behav 26:138-147.[CrossRef]
Piper WH, Tischler KB, Klich M, 2000. Territory acquisition in loons: the importance of take-over. Anim Behav 59:385-394.[CrossRef][Web of Science][Medline]
Pribil S, 1998. Reproductive success is a misleading indicator of nest-site preferences in the red-winged Blackbird. Can J Zool 76:2227-2234.[CrossRef]
Rendon MA, Garrido A, Ramirez JM, Rendon-Martos M, Amat JA, 2001. Despotic establishment of breeding colonies of greater flamingos, Phoenicopterus ruber, in southern Spain. Behav Ecol Sociobiol 50:55-60.[CrossRef]
Rice WR, Gaines SD., 1994. "Heads I win, tails you lose": testing directional alternative hypotheses in ecological and evolutionary research. Trends Ecol Evol 9:235-237.[CrossRef]
Ruxton GD, Armstrong JD, Humphries S, 1999. Modeling territorial behaviour of animals in variable environment. Anim Behav 58:113-120.[CrossRef][Web of Science][Medline]
Schoener TW, 1983. Simple models of optimal territory size: a reconciliation. Am Nat 121:608-629.[CrossRef]
Schoener TW, 1987. Time budgets and territory size: some simultaneous optimization models for energy maximizers. Am Zool 27:259-291.
Searcy WA, Kasukawa K, 1989. Alternative models of territorial polygyny in birds. Am Nat 134:323-343.[CrossRef]
Severinghaus L, Kurtak BH, Eickwort GC, 1981. The reproductive behavior of Anthidium manicatum (Hymenoptera: Megachilidae) and the significance of size for territorial males. Behav Ecol Sociobiol 9:51-58.[CrossRef]
Shapiro SS, Wilk MB, 1965. An analysis of variance test for normality (complete samples). Biometrika 52:591-611.
Shutler D, Weatherhead PJ, 1991. Owner and floater red-winged blackbirds: determinants of success. Behav Ecol Sociobiol 28:235-241.[CrossRef]
Siegel S, Castellan NJ, 1988. Nonparametric statistics for the behavioral sciences. Boston: McGraw Hill.
Stamps J, 1994. Territorial behavior: testing the assumptions. Adv Study Behav 23:173-231.
Stamps J, Krishnan VV, 2001. How territorial animals compete for divisible space: a learning-based model with unequal competitors. Am Nat 157:154-169.[CrossRef]
Switzer PV, 1993. Site fidelity in predictable and unpredictable habitats. Evol Ecol 7:533-555.[CrossRef]
Tabachnik BG, Fidell LS, 2001. Using multivariate statistics. Boston: Allyn and Bacon.
Taborsky M, Hudde B, Wirtz P, 1987. Reproductive behavior and ecology of Symphodus (Crenilabrus) ocellatus a European wrasse with four types of male behavior. Behaviour 102:82-118.
van den Berghe E, 1988. Piracy as an alternative reproductive tactic for males. Nature 25:697-698.
van den Berghe E, Wernerus F, Warner RR, 1989. Female choice and the mating cost of peripheral males. Anim Behav 38:875-884.[CrossRef]
Warner RR, Wernerus F, Lejeune P, van den Berghe E, 1995. Dynamics of female choice for parental care in a fish species where care is facultative. Behav Ecol 6:73-81.
Wells KD, 1977. Territoriality and male mating success in the green frog (Rana clamitans). Ecology 58:750-762.[CrossRef][Web of Science]
Wernerus FM, 1985. Etude des parametres orientant le choix du partenaire au cours de la reproduction chez Symphodus melanocercus (Risso 1981) et Symphodus ocellatus (Forskal 1775), Labrides mediterraneens. (Memoire de Licence) Liege: Universite de Liege.
Wernerus FM, 1988. Etude des mécanismes sous-tendent les sytèmes d'appariement de quatre espèces de poissons labridés méditerranéens des genres Symphodus Rafinesque, 1810 et Thalassoma Linne, 1758. Cah Ethol Appl 9:117-320.
Wernerus FM, Lejeune P, van den Berghe EP, 1989. Transmission of mating success among neighbouring males in the Mediterranean labrid fish, Symphodus ocellatus. Biol Behav 14:195-206.
Wernerus FM, Michel Ch, Voss J, 1987. Introduction a l'etude de la selection sexuelle chez Symphodus ocellatus (Forskal 1775) et S. melanocercus (Risso 1810) poissons Labrides mediterraneens. Cah Ethol Appl 7:19-38.
Ylieff MY, Michel C, Voss J, 1997. Dynamics of nest defense in the Mediterranean labrid fish Symphodus ocellatus (Forskal, 1775), I: agonistic behaviour between nesting territorial male and five other labrid species. Cah Ethol 17:49-74.
Ylieff MY, Michel C, Voss J, 1998. Dynamics of nest defense in the Mediterranean labrid fish Symphodus ocellatus (Forskal, 1775), II: aggressive behaviour of nesting territorial male towards gasteropod molluscs. Cah Ethol 18:39-52.
Zar JH, 1996. Biostatistical analysis. Upper Saddle River, New Jersey: Prentice Hall.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. H. Alonzo and K. L. Heckman The unexpected but understandable dynamics of mating, paternity and paternal care in the ocellated wrasse Proc R Soc B, January 7, 2010; 277(1678): 115 - 122. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||