Integrative and Comparative Biology Advance Access originally published online on May 10, 2006
Integrative and Comparative Biology 2006 46(4):439-448; doi:10.1093/icb/icj045
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sexual selection and reproductive success in hermaphroditic seabasses
College of the Atlantic 105 Eden Street Bar Harbor, ME 04609, USA
Correspondence: 1E-mail: chrisp{at}coa.edu
 |
Synopsis |
|---|
Top
 Synopsis IntroductionThe reproductive biology of...Is tit-for-tat a useful...Sexual selection, evolutionary...References |
|---|
Mating behavior in simultaneously hermaphroditic seabasses has been often cited as an example of cooperation among unrelated conspecifics. The predominant mating behavior in this group involves egg trading, where individuals reciprocally fertilize parcels of eggs from a partner. Egg trading has been suggested as a good example of a tit-for-tat cooperative mating strategy. Although simultaneous hermaphroditic fishes are often held up as strong examples of cooperation in mating behavior, a closer examination reveals significant sexual selection and sexual conflict between male and female roles among individuals. In the 7 species where data exist, there is a significant increase in male reproductive success with individual size, and in all but 1 species success through male function increases faster than reproductive success through female function. Despite this male-size advantage in simultaneous hermaphrodites, most species maintain their hermaphroditism for their entire life, and the increased male allocation while engaging in biased forms of reciprocation appear to increase the evolutionary stability of hermaphroditism in these species. Thus, egg-trading behavior is probably more complicated than was initially recognized, with individuals releasing different numbers of eggs in spawns, spawning at different rates as males and females, and partitioning male effort between pair and alternative mating tactics. The departures from equal reciprocity can probably be best understood by including aspects of traditional mating-system theory, with individuals increasing male mating success through a variety of behavioral tactics.
|
Introduction |
|---|
|
Top
SynopsisIntroductionThe reproductive biology of...Is tit-for-tat a useful...Sexual selection, evolutionary...References |
|---|
The initial theoretical and empirical work on simultaneous hermaphrodites emphasized identifying patterns and processes in a population that could result in simultaneous hermaphroditism being an evolutionarily stable strategy (ESS) (Charnov 1979
, 1982). Charnov's seminal work (1979, 1982) noted that if individuals had the potential to obtain higher reproductive success by specializing as males, it could result in the breakdown of simultaneous hermaphroditism as an ESS. For hermaphroditism to be stable, individuals are not expected to obtain higher fitness as males than as hermaphrodites. The earliest development of this idea predated Charnov (1979) and was the low-density model of Ghiselin (1969), which suggested that male reproductive success was limited by low encounter rates with conspecifics.
The first detailed study of the mating system in an outcrossing hermaphroditic animal came from Fischer's (1980, 1981) studies of the Black Hamlet (Hypoplectrus nigricans). In this species, individual hermaphrodites pair up in the late afternoon, and following a long courtship, take turns releasing small parcels of eggs for the other individual to externally fertilize, a behavior he called egg trading. Fischer proposed that fish released eggs in order to be able to fertilize their partner's eggs. In the absence of reciprocation individuals would delay releasing more eggs (Fischer 1980). Fischer proposed that the male role was the preferred spawning role in this species, but that the long courtship period, parceling of a daily clutch of eggs, and the conditional reciprocity in egg trading kept individuals from successfully specializing as males. These factors allowed simultaneous hermaphroditism to be an evolutionarily stable sex allocation pattern in this species.
On the heels of the work by Fischer (1980, 1981), Axelrod and Hamilton (1981) applied the Prisoner's Dilemma game theory to model the evolution of cooperative behavior. They found that in an iterated version of the game, a strategy that combined initial cooperation, and what they called being "nice, forgiving, but provokable," was able to do well against a wide range of strategies. These traits were best embodied in a simple strategy called tit-for-tat, which always started cooperatively and then repeated the behavior of its partner from the previous round. This model was embedded in a 2-person symmetric game, so both members of the pair had the same strategy set, which biologically meant that both individuals not only had the same possible behaviors, but also that the fitness pay-offs for those behaviors were the same for both members of the pair. As a possible example of an organism employing tit-for-tat, Axelrod and Hamilton (1981) used Fischer's recent article (1980) on egg trading in black hamlets.
The theoretical work on hermaphroditism by Charnov (1979) and cooperation by Axelrod and Hamilton (1981) assumed that all individuals in a population were equivalent, either by using the model of a symmetric game or by assuming all individuals had the same total amount of resources to allocate to reproduction. This fits well with the empirical results from Fischer (1980), where he found little evidence of size-related differences in behavior among individuals. This is in sharp contrast to both theory and empirical work on sequentially hermaphroditic organisms, where changes in allocation are thought to stem from changes in reproductive success via each sex as individuals change in size, age, or resources devoted to reproduction (for example, Warner 1975).
As the literature on hermaphroditic seabasses expanded beyond the initial work on H. nigricans, parallels with the body of work on sequential hermaphrodites became clear. Hermaphroditic seabasses also exhibit size-related changes in reproductive tactics and mating success (Fischer and Petersen 1986; Petersen and Fischer 1986; Petersen 1987; Oliver 1997), or sex allocation (Hastings and Petersen 1986; Petersen 1990b; Petersen and Fischer 1996). However, all of the studied species in this clade (including the genera Diplectrum, Serraniculus, Serranus, and Hypoplectrus) maintain functional hermaphroditic individuals among the majority of the mature individuals in the population, in spite of large variation on returns on male function among individuals within the population.
The focus of this article is the behavioral ecology of these simultaneously hermaphroditic fishes, and in particular the varying degrees of social dominance and reciprocity among individuals within species. Thefirst section reviews the data on the reproductive biology of the simultaneously hermaphroditic seabasses. The emphasis of the analysis is to determine the extent to which mating and reproductive success vary with individual size among the 7 species of hermaphroditic seabasses that have been best studied to date. The second part of the article then examines how these data challenge the use of simple symmetric models like tit-for-tat. The last section of the article asks what factors help to stabilize simultaneous hermaphroditism as a sex allocation strategy in these species.
|
The reproductive biology of hermaphroditic seabasses |
|---|
|
Top
SynopsisIntroductionThe reproductive biology of...Is tit-for-tat a useful...Sexual selection, evolutionary...References |
|---|
The seabasses in the subfamily Serraninae, the "dwarf seabassses," are relatively small fishes that are typically associated with coral rubble and areas along the margins of coral and rocky reefs in the Caribbean and eastern and western Atlantic. They spawn in the late afternoon on a daily basis in the tropics, where individuals release pelagic eggs (<1 mm in diameter) that are externally fertilized and begin a planktonic part of the life cycle. There is no parental care and selfing does not appear to occur in the wild.
In their early review of simultaneously hermaphroditic seabass mating systems, Fischer and Petersen (1987) divided hermaphroditic seabasses into 2 groups, those in which conditional reciprocity is the norm through egg trading and those where the largest individuals are socially dominant males. Two species, H. nigricans and Serranus tortugarum, parcel a daily clutch and exhibit egg trading. A third species, Serranus tigrinus, form long-term monogamous pairs. Although they do not parcel clutches on a daily basis, they continue to reciprocally spawn over a period of days, making this pattern very similar to reciprocity in egg traders. However, in 2 other species, Serranus baldwini and Serranus psittacinus (referred to as Serranus fasciatus in earlier articles), large individuals lose ovarian tissue, become functional males, and defend harems of smaller hermaphroditic individuals (Fischer and Petersen 1986; Hastings and Petersen 1986; Petersen and Fischer 1986; Petersen 1987, 1990a,1990a). Adapting ideas from mating-system theory in separate-sexed species, and especially those synthesized by Emlen and Oring (1977), Fischer and Petersen (1987) suggested that the higher density and spatial predictability of hermaphrodites during spawning in S. baldwini and S. psittacinus led to mate monopolization, and the ability of large individuals to successfully specialize as males.
In support of this hypothesis, Petersen (1990a) described how the mating system of S. psittacinus changed with density in ways consistent with varying opportunities for mate monopolization and sexual selection. This system appears to change from long-term monogamy at low density, to harem polygyny with large individuals specializing as males at moderate density to a pattern of complex harems at high densities (Petersen 1990a). In complex harems, the largest individuals are males, and they spawn with large hermaphrodites inside their home range, who may then spawn in the male role with several smaller hermaphrodites that reside inside of their home range. Spatially, the pattern is small harems controlled by larger hermaphrodites within larger harems controlled by the largest individuals that are male (Petersen 1990a).
The social systems of the 2 most recently studied species, Serranus tabacarius (Petersen 1995) and Serranus subligarius (Oliver 1997) exhibit aspects of both mating-system types described by Fischer and Petersen (1987). Spawning occurs within the daily home ranges of individuals in both S. tabacarius and S. subligarius. Fischer and Petersen (1987) showed that species that spawned within their home ranges varied from long-term monogamy at low densities to harem polygyny at high densities. However, both S. subligarius and S. tabacarius are egg-trading species, with individuals having multiple partners over a spawning period.
The division of the hermaphroditic seabasses into 2 broad groups based on mating system ignores complexities, especially in the reciprocating species, that may be important in the evolution of both mating systems and sex allocation patterns in these species. The current literature suggests several situations in which there are size-specific changes in mating and reproduction in egg-trading species, yet these patterns are typically ignored in the simplified theoretical approaches that have been applied to egg-trading seabasses (Fischer 1988; Petersen 1991).
In field studies done by several different researchers, data have been collected on hermaphroditic mating systems, including data on reproductive success and behavior and how they vary among individuals within a population. Patterns of individual fecundity, mating success, and social behavior and how these factors change with individual size within a species are summarized from the current literature on the 7 species where adequate field data existed. An eighth species, Serranus scriba, is known to be an egg parceler and to have streaking behavior (Lejeune and others 1980; Havelange and Voss 1993), but quantitative details, concerning individual size, are not available, and this species was not included in the analysis. A summary of the reproductive biology of each included species is given in Table 1, and the references are given in the text.
|
Gender
Gender data exist from both behavioral and histological analysis of all 7 species; all have individuals that are functional simultaneous hermaphrodites that do release sperm and eggs over a single daily spawning period. The densities of populations are those reported from the studies; there can clearly be large variance in density within a species. These data are probably biased toward the upper end of density range, as studies are more likely to be done in areas where individuals are abundant. In addition to simultaneous hermaphrodites, pure males exist in the haremic species. The density of the haremic species is bracketed by the density of the egg-trading species, verifying that density is not the sole determinant of mating system in this group.
Components of reproductive success in hermaphroditic seabasses
Fecundity and individual size
Within a species, fish typically vary widely in the size of reproductively active individuals, and both fecundity and fighting ability increase with size. In fishes with planktonic eggs and no parental care, female reproductive success should be tied largely to egg production, given the high fertilization rates typical of coral reef fishes (Marconato and others 1995, 1997; Kiflawi and others 1998; Petersen and others 2001; C. W. Petersen unpublished data). Thus, hermaphroditic individuals have the potential to increase female reproductive success as they grow. However, because reproductive effort can be reallocated between male and female function, larger individuals may not have higher fecundity, depending on the total reproductive investment of the individual and how it is partitioned between the 2 sexual roles.
Individual fecundity, an estimate of female reproductive success, was estimated as either the number of eggs released in a day, determined by stripping females (Fischer 1986; Petersen 1990a) or from the size of the ovarian portion of the ovotestis (Petersen 1991, S. tabacarius; Petersen 1995; S. tortugarum; Petersen and Fischer 1996). In 6 cases where both types of fecundity estimates existed, they showed the same trends with individual size with, 1 exception.
Increased fecundity with adult size was shown by 4 of 6 species of Serranus (Table 1). These differences can be dramatic; in S. tortugarum there is nearly a 10-fold difference in fecundity between the smallest and largest size class of individuals (Petersen and Fischer 1996). Hermaphrodites in S. psittacinus show an increase in both egg number and ovarian tissue with size (Petersen 1987, 1990a), while there was no pattern of egg number or ovarian tissue amount in the other harem polygynous species, S. baldwini, among hermaphroditic individuals. In the sixth species, S. subligarius, ovary weight increased with size (Oliver 1997), but there was a surprising lack of a correlation between daily egg production and fecundity, despite large variation in adult size (Cheek 1998). H. nigricans has a slight but non-significant increase in egg number (Fischer 1986) or size of ovarian part of the ovotestis (C. W. Petersen, unpublished data). This lack of a correlation might be due to the lower variance in adult size in this species; H. nigricans has a lower coefficient of variation in adult size than the other egg-trading species (Fischer 1986).
Patterns of egg release in egg-trading species
There are no field data on the number of eggs released per parcel by egg-trading fish for any of the species, although H. nigricans exhibited a decrease in eggs per parcel in successive spawnings within a bout in an aquarium study (Fischer and Hardison 1987). The estimated fecundity of hermaphrodites increases with individual size at a higher rate than the female spawning rate of individuals in S. subligarius and S. tortugarum; thus, larger individuals release more eggs per spawn in these species. In contrast, female spawning rate and female fecundity scale similarly with size in H. nigricans and S. tabacarius.
There is a higher propensity for smaller individuals to initiate spawns as a female, in all 4 egg-trading species, especially in S. subligarius (Oliver 1997), and there are at least 2 explanations for this pattern. It might be related to the higher reward if they can successfully reciprocate with a larger fish with a larger egg parcel. However, the pattern of small individuals spawning as females with larger hermaphrodites might also be due to the ability of larger individuals to be socially dominant. In S. tabacarius, for example, there is no correlation between individual size and the number of eggs released per spawn (Petersen 1995), and social dominance by larger individuals is the more likely explanation for smaller individuals initiating spawns as females.
Male reproductive success
There are 3 ways that an individual can increase its male reproductive success through its pair-spawning behavior. It can mate with more individuals, it can mate with more fecund individuals, or it can bias its mating by spawning disproportionately in the male role. All 3 of these appear to occur among the 7 serranines considered in this review.
Large individuals have more mating partners over a spawning period in 4 of the 7 species studied, including 2 harem polygynous and 2 egg-trading species (S. baldwini, Petersen and Fischer 1986; S. psittacinus, Petersen 1987, 1990a; S. tabacarius, Petersen 1995; S. subligarius, Oliver 1997). There is no stated correlation in Hypoplectrus, and the pattern appears to be slight or non-existent in S. tigrinus, where most individuals only spawn with their long-term partner (Pressley 1981). There is no pattern of number of partners and mate size in S. tortugarum (Petersen and Fischer 1996).
Size-assortative mating increases both male and female reproductive success of individuals in egg-trading species, while in a haremic species it increases the male reproductive success of large individuals. Size-assortative mating occurs in all of the reciprocating species where there is a positive relationship between adult size and fecundity (S. tabacarius, Petersen 1995; S. subligarius, Oliver 1997; S. tigrinus, Pressley 1981; S. tortugarum, Petersen and Fischer 1996), but not H. nigricans where no size–fecundity relationship exists (Fischer 1980, 1986). In addition, in the complex harems of S. psittacinus there is size-assortative mating among hermaphrodites (Petersen 1990a).
Social dominance of larger individuals can restrict male mating opportunities for smaller individuals in most seabass species. In addition to extensive data on social dominance patterns in the harem polygynous species (Petersen and Fischer 1986; Petersen 1987, 1990a), there is evidence for social dominance affecting mating patterns in several reciprocating species. In S. subligarius, male-role hermaphrodites are aggressive toward potential streakers (see below for a further description of this alternative male mating tactic) (Oliver 1997). In S. tigrinus, a member of a pair may direct aggression toward a solitary individual attempting to pair spawn or streak with its partner (Pressley 1981). In S. tabacarius, larger individuals increase aggression toward smaller conspecifics during the spawning period (Petersen 1995). There is no such evidence in S. tortugarum or H. nigricans.
In 6 of the 7 species, the ratio of male:female spawns and an individual's male mating success increases with individual size (Table 1). In the harem polygynous species, large individuals clearly mate disproportionately as males, to the extent that the largest individuals typically lose their ovarian tissue (Hastings and Petersen 1986; Petersen and Fischer 1986; Petersen 1987). However, there are several trends in reciprocating species that have led to a positive correlation between the proportion of male matings and individual size. Larger individuals are more likely to spawn first in the male role and less likely to reciprocate as females in a spawning sequence in S. tabacarius (Petersen 1995) and S. subligarius (Oliver 1997). In H. nigricans, most spawns occur with a primary mating partner [defined by Fischer (1980) as the fish that a focal individual mated with most often in a spawning period] where no size bias in male spawning exists. However, in matings with secondary partners, the larger individual of the pair is less likely to reciprocate and spawns more frequently in the male role (Fischer 1980). A more extreme version of this pattern is exhibited in S. tigrinus, where larger paired individuals mate as males with smaller non-paired individuals, who mate only as females (Pressley 1981; E. Fischer unpublished data). Overall, in 4 of the 5 reciprocating species (H. nigricans, S. tabacarius, S. subligarius, S. tigrinus) larger individuals spawn disproportionately in the male role.
For all of the species studied, there are at least some situations in which larger individuals obtain disproportionately higher male reproductive success (estimated number of eggs fertilized) in the population. Although this was clearly true for the harem polygynous species, it is clear from this summary that all of the species also show some form of male-size advantage in mating success from pair spawning.
In addition to pair spawning, individuals can increase male mating success by joining pairs as a streaker, releasing sperm in competition with the male-role pair-spawning individual. This alternative mating tactic is more commonly adopted by smaller individuals in the population among many separate-sexed species, including fishes (Taborsky 1994). Streaking by small individuals has the potential to weaken or reverse the trend of increased male reproductive success with size. Incidence of streaking varies from never being observed in any Hypoplectrus species, to being observed once in S. tabacarius (Petersen 1995), to much higher frequencies in S. tortugarum (Fischer 1984), S. subligarius (Oliver 1997), and S. psittacinus (Petersen 1987, 1990a, 1990b) (Table 1). Streaking is a more common behavior in smaller individuals in S. subligarius (Oliver 1997) and in smaller individuals in some populations of S. psittacinus (Petersen 1990a), but is independent of size in S. tortugarum (Petersen and Fischer 1996). There is no size-related pattern for the pair spawners involved in streaked spawns, with an exception. In S. subligarius, streaking rates are higher when the female-role fish is larger (Oliver 1997).
In S. tortugarum, the high rate of streaking by all individuals in a population disproportionately increases male reproductive success in small individuals, as they are on average spawning with larger, more fecund individuals than themselves. In S. subligarius, the high level of streaking of small individuals increases their male reproductive success, but the much higher rate of male-role pair-spawning by large individuals leads to an overall bias toward male reproductive success for larger individuals (Oliver 1997).
These patterns of male reproductive success lead to higher ratios of male:female reproductive success with size in all species except S. tortugarum, where high levels of streaking disproportionately increase the reproductive success of smaller males (Petersen and Fischer 1996). In S. tortugarum, size-assortative pairing and the larger size of clutches of larger individuals still results in larger individuals obtaining higher male reproductive success, but the gain in female reproductive success with size is even greater, leading to disproportionate female reproductive success by larger individuals in this species.
|
Is tit-for-tat a useful model for egg trading? |
|---|
|
Top
SynopsisIntroductionThe reproductive biology of...Is tit-for-tat a useful...Sexual selection, evolutionary...References |
|---|
The explanation first hypothesized by Fischer in 1980, that egg trading evolved as a way for hermaphrodites to use eggs to obtain matings as males, appears to be on very firm ground (see Landolfa 2002
, for an alternative view). Egg trading or some form of reciprocal spawning with individuals releasing parcels of eggs has been shown to occur in several other species of seabasses as well as in polychaete worms in the genus Ophryotrocha (Sella 1988; Sella and Lorenzi 2000). Fischer's (1980) hypothesized reason for the stability of simultaneous hermaphroditism in H. nigricans, the inability of individuals to successfully fertilize more eggs by specializing as a male, also appears to operate in most of the other hermaphroditic seabasses. Egg trading has been observed in many species of hermaphroditic seabasses, and delaying the continuation of mating in the absence of reciprocity has been found in a second species (S. tabacarius, Petersen 1995). The long courtship preceding spawning has not been found in other species to the extent that it exists in H. nigricans, but this behavior should not be as important in maintaining egg trading as an evolutionarily stable strategy in species where eggs are divided into more parcels per day, which occurs in S. tortugarum (Fischer 1984, Petersen and Fischer 1996) and S. tabacarius (Petersen 1995).
Axelrod and Hamilton (1981) outlined how egg trading could fit into a type of tit-for-tat solution for an iterated Prisoner's Dilemma game, and Fischer(1988) expanded upon this idea. However, 2 general critiques challenge the tit-for-tat model in understanding egg trading (reviewed in Brembs 1996; Dugatkin 1997): (1) the model is not a close enough fit for the biology of egg-trading hermaphrodites (Friedman and Hammerstein 1991; Connor 1992), and (2) the species do not suffer costs for not reciprocating, and thus the system more closely resembles by-product mutualism than cooperation (Connor 1992).
There are several ways in which the behavior of egg traders does not fit the Prisoner's Dilemma game. Individual decisions to release eggs are sequential, not simultaneous, and thus a partner has information on whether an individual cooperated by releasing eggs before its decision to release eggs. Boyd (1988) has argued that a sequential version of the Prisoner's Dilemma game maintains the qualitative properties of the version used by Axelrod and Hamilton (1981). In Boyd's model, however, individuals are equivalent and each member of a pair is equally likely to act first. These conditions are violated in several of the species reviewed here, most strongly among the reciprocating species S. subligarius and S. tabacarius.
Another way seabass egg trading differs from the Prisoner's Dilemma game is that defection by individuals is not limited only to deciding not to release eggs, but an individual may also leave the pair or decide not to form a pair. This gives individuals multiple ways to fail to reciprocate. A likely outcome is that the probability of continued interactions with the same partner [w in the model of Axelrod and Hamilton (1981)] will depend on the behavior or potentially the size of the partner, something that was not incorporated into the original game. If we assume that defection by an individual occurs when it leaves after a male-role spawn, then defection is biased toward larger individuals in S. tabacarius (Petersen 1995) and S. subligarius (Oliver 1997), but not in H. nigricans (Fischer 1980) or S. tortugarum (Petersen and Fischer 1996).
Enquist and Leimar (1993) claimed that this type of dynamic could not be understood using a repeated Prisoner's Dilemma game, and instead modeled the consequences of allowing individuals to vary time in a pair (coalition time) and time to form a new pair (search time) on the stability of cooperation. Their conclusions, however, were similar to those of Fischer (1988), in that egg trading and long courtship were examples of social control that prevented a free rider strategy (in this case, non-cooperating males) from invading the population.
Friedman and Hammerstein (1991) have also attempted to incorporate partner desertion and acquisition into a model of reciprocity. Their model emphasized the sequence of spawning events in H. nigricans, with a long courtship preceding any potential for egg trading and a limited time to spawn as its major components. They concluded that these aspects of the mating system were effective enforcement mechanisms against the evolution of cheaters that spawned as males and did not reciprocate.
Perhaps the most problematic aspect of using the Prisoner's Dilemma model are the assumptions of equivalent players and symmetric pay-offs. This has allowed theoreticians to ignore the role of variance in individual size within a population in shaping both the overall mating system of a population and the factors affecting the stability of egg trading and hermaphroditism. It is interesting to note that the model was first developed for H. nigricans, the species with the least variance in adult size and fecundity (Fischer 1986). Even though this species shows the least asymmetry, there are still changes in mating behavior and reproductive success with individual size (Fischer 1980).
In addition to problems in the underlying assumptions of symmetry in the model, there are also logistical challenges to rigorously testing some of the assumptions in natural populations. For example, we do not know if parcel size is predictable within egg traders, or if individuals are able to adjust their behavior in response to variation in parcel size among partners. If an individual releases more eggs per spawning bout, we might expect the partner to spawn more as a female to obtain these particularly fecund egg parcels. In the traditional model, this would be considered a breach of cooperation, and not fit the tit-for-tat model. An area for future research will be determining what constitutes a lack of cooperation in these reciprocating systems when resources are asymmetrically distributed. It will be difficult to evaluate the dynamics of clutch release in the field. Successful techniques have been used to estimate fecundity in tropical reef fishes that release pelagic eggs (Shapiro and others 1994; Marconato and others 1995, 1997; Warner and others 1995), and perhaps these techniques can be adapted to the smaller clutches of hermaphroditic seabasses. Aquarium studies have been used to examine simple patterns of egg release in both S. tortugarum (Fischer and Hardison 1987) and S. subligarius (Cheek 1998), and more complex experiments manipulating either numbers or sizes of mates may offer the best chance of examining patterns and conditional strategies in egg release in reciprocating species.
Finally, Connor (1992) has pointed out that in egg trading the parceling of a clutch is a mechanism to prevent cheating by altering the costs and benefits of continuing to spawn; individuals that stay and continue to spawn may never incur a net cost during the interaction so that the behavior does not fit a tit-for-tat model. Connor suggests that by not incorporating desertion into the tit-for-tat game, the game ignores the possibility that parceling has evolved to lower the benefit of not cooperating when a partner cooperates [T in the model of Axelrod and Hamilton (1981)] to a level below the reward for cooperating (R), which violates the assumption that T > R in the game.
Given all of these potential problems, it appears that a symmetric tit-for-tat model that ignores desertion, although it has been useful to date in helping to understand reciprocating behavior, needs to be replaced by a more realistic model or set of models that address the more detailed questions that have arisen from the field studies of hermaphroditic seabasses. The generality of a symmetric game theory model fails to capture much of the richness of the behavioral interactions in these species, and would misinterpret lack of perfect reciprocal spawning for desertion when there are alternative explanations. Given that various modeling attempts have been able to capture the major robust conclusions of the Prisoner's Dilemma model for hermaphroditic seabasses, these alternatives to studying egg-trading behavior appear to offer the best avenue for continued theoretical growth in this mating system.
|
Sexual selection, evolutionary stability, and mating-system evolution in simultaneous hermaphrodites |
|---|
|
Top
SynopsisIntroductionThe reproductive biology of...Is tit-for-tat a useful...Sexual selection, evolutionary...References |
|---|
The mating systems in hermaphroditic seabasses show consistent trends toward biases in male mating and reproductive success with larger size over a wide range of densities. There is evidence for both increasing reproductive success with individual size in all species studied and an increase in the proportion of the male component of reproductive success with increasing size in 6 of the 7 species.
The only exception to this pattern is S. tortugarum, where high levels of sperm competition, especially from small male-role streakers, causes the female component of reproductive success to increase faster with individual size. All of the observations to date are consistent with the hypothesis that for externally fertilizing hermaphrodites the male role is preferred in spawning (Charnov 1979; Fischer 1980, 1987; Fischer and Petersen 1987; Leonard 1993).
Most discussions on the evolutionary stability of simultaneous hermaphroditism compare a simultaneous hermaphrodite phenotype with a male phenotype that "cheats" by refusing to reciprocate in egg-trading species (Fischer 1980, 1988; Fischer and Petersen 1987; Leonard 1990, 1993). An important question is how our knowledge of size-related trends in social dominance and reproductive success in these hermaphroditic fishes affects our understanding of the stability of hermaphroditism as an evolutionarily stable sex allocation strategy in these species.
All else being equal, we would expect asymmetries among individuals to tend to destabilize egg trading as a form of conditional reciprocity in simultaneous hermaphrodites. As asymmetries increase between 2 individuals, the individual with fewer resources (for example, eggs) to exchange is in a weaker bartering position, and at some point reciprocity does not benefit the larger individual. As variance in adult size increases within a population, behavioral dominance of subordinates should also be more likely. Behavioral dominance of hermaphrodites by large males occurs in both S. baldwini and S. pssitacinus, where large individuals specialize as males and obtain high male reproductive success by defending harems of hermaphrodites. However, despite high densities, predictable mates in time and space, and large variance in adult size in other species of Serranus, reciprocation among simultaneous hermaphrodites occurs in all other Serranus spp. studied to date.
An explanation for the lack of large males in other species of Serranus may be that the increased reproductive success of large individuals within reciprocating species makes it more difficult for a large male phenotype to successfully invade a population of egg traders. Larger hermaphrodites, by achieving disproportionately higher male reproductive success (in all 5 species) and female reproductive success (in 3 of 5 species) (Table 1), obtain higher fitness than average hermaphrodites in the population. Thus, the replacement of large hermaphrodite phenotypes by large male phenotypes, although possible, is less likely due to the asymmetries in reproductive success that large hermaphrodites obtain in reciprocating species.
As an illustration of the role of asymmetries in stabilizing hermaphroditism, using information from S. tabacarius (Petersen 1995; Table 1), it is possible to examine how asymmetries in reproductive success can alter the conditions required for a large male to achieve higher current reproductive success than an equal-sized hermaphrodite. In the simplest case, a symmetric population with all individuals of equal size and equal reproductive success, a male would have to obtain a reproductive success equal or greater than twice the female reproductive success of a hermaphrodite. In the real size-structured population, if matings were symmetric, the male would have to achieve at least twice the reproductive success of a large hermaphrodite. This would require mating with significantly more than 2 average hermaphrodites, and create a more stringent condition for the evolution of large males in the population. However, in S. tabacarius, large hermaphrodites obtain higher male reproductive success than female reproductive success, because larger hermaphrodites spawn with more, larger partners, and spawn as a male more often than as a female (Petersen 1995; Table 1). This creates an even more stringent condition for the evolution of males in this population of egg-trading hermaphrodites. By exhibiting some asymmetries in their egg-trading behavior, individuals in this species actually stabilize hermaphroditism against invasion by large male phenotypes by enabling large individuals to obtain relatively high reproductive success within the egg-trading mating system.
Another aspect of mating behavior in these species that should reduce the success of a large male phenotype in reciprocating species is the occurrence of sperm competition and alternative male mating tactics. In species, such as S. tortugarum, where sperm competition is quite high (Fischer 1984; Petersen and Fischer 1996; Table 1), large individuals that attempt to specialize as males using dominant male mating tactics will have their male reproductive success discounted by the degree of sperm competition in their spawns. Sperm competition, combined with a large increase in fecundity with size among individuals in a population, appears to decrease the likelihood of large individuals changing to males in the protogynous hermaphrodite Sparisoma radians (Munoz and Warner 2003, 2004).
All of the data collected in hermaphroditic seabasses support the assumption, first articulated for simultaneous hermaphrodites by Charnov (1979), as in most separate-sexed species, that female function in hermaphrodites is typically limited by resources and male function is limited by access to mates (Bateman 1948). The data on mating patterns in these hermaphroditic seabasses also support the hypotheses of Fischer (1980) and Leonard (1993) that sexual selection is stronger on male than female function in these species, and that individuals preferentially spawn in the male role. Although sexual selection does appear to be stronger in the haremic species, where male reproductive success varies the most, there is substantial sexual conflict in egg-trading species, resulting in several behavioral tactics that enhance male mating success in some individuals. These tactics include size-assortative spawning, asymmetric spawning reciprocation, aggressive behavior directed toward potential mate competitors, and employing alternative male mating behaviors. These conflicts will be better understood when we increase the sophistication of our current theoretical work and, similar to previous work in sex-changing species, incorporate individual asymmetries into models of hermaphrodite sex allocation and behavioral decisions.
|
Acknowledgements |
|---|
I thank Janet Leonard for organizing the symposium "Sexual selection and mating systems in hermaphrodites." I also thank Eric Fischer and Ann Oliver Cheek for providing a productive and collegial attitude while sharing their work and insights. H. Hess, J. Edwards, and an anonymous reviewer provided helpful comments that greatly improved the clarity of the manuscript.
|
Footnotes |
|---|
From the symposium "Sexual Selection and Mating Systems in Hermaphrodites" presented at the annual meeting of the Society for Integrative and Comparative Biology, January 4–8, 2005, at San Diego, California.
|
References |
|---|
|
Top
SynopsisIntroductionThe reproductive biology of...Is tit-for-tat a useful...Sexual selection, evolutionary...References |
|---|
Axelrod, R and WD Hamilton. 1981. The evolution of cooperation. Science 211:1390–6.
Bateman, AJ. 1948. Intra-sexual selection in Drosophila. Heredity 2:349–68.[Web of Science][Medline]
Boyd, R. 1988. Is the repeated prisoner's dilemma a good model of reciprocal altruism? Ethol Sociobiol 9:211–22.
Brembs, B. 1996. Chaos, cheating and cooperation: potential solutions to the Prisoner's Dilemma. Oikos 76:14–24.
Charnov, EL. 1979. Simultaneous hermaphroditism and sexual selection. Proc Natl Acad Sci USA 76:2480–4.
Charnov, EL. 1982. The theory of sex allocation. Princeton Princeton University Press.
Cheek, AO. 1998. Ovulation does not constrain egg parcel size in the simultaneous hermaphrodite Serranus subligarius. Env Biol Fish 52:435–42.[CrossRef]
Connor, RC. 1992. Egg-trading in simultaneous hermaphrodites: an alternative to Tit-for-Tat. J Evol Biol 5:523–8.[CrossRef]
Dugatkin, LA. 1997. Cooperation among animals. New York: Oxford University Press.
Emlen, ST and LW Oring. 1977. Ecology, sexual selection, and the evolution of mating systems. Science 197:215–23.
Enquist, M and O Leimar. 1993. The evolution of cooperation in mobile organisms. Anim Behav 45:747–57.[CrossRef]
Fischer, EA. 1980. The relationship between mating system and simultaneous hermaphroditism in the coral reef fish, Hypoplectrus nigricans (Serranidae). Anim Behav 28:620–33.[CrossRef]
Fischer, EA. 1981. Sexual allocation in a simultaneously hermaphroditic coral reef fish. Amer Nat 117:64–82.[CrossRef][Web of Science]
Fischer, EA. 1984. Egg trading in the chalk bass, Serranus tortugarum, a simultaneous hermaphrodite. Z Tierpsychol 66:143–51.
Fischer, EA. 1986. Mating systems of simultaneously hermaphroditic serranid fishes. In Uyeno, T, AraiR , TaniuchiT , MatsuuraK (Eds.). Indo-Pacific Fish Biology: Proceedings of the Second International Conference on Indo-Pacific Fishes Tokyo Ichthyological Society of Japan pp. 776–84.
Fischer, EA. 1987. Mating behavior in the black hamlet—gamete trading or egg trading? Env Biol Fish 18:143–8.
Fischer, EA. 1988. Simultaneous hermaphroditism, tit-for-tat, and the evolutionary stability of social systems. Ethol Sociobiol 9:119–36.
Fischer, EA and PD Hardison. 1987. The timing of spawning and egg production as constraints on male mating success in a simultaneously hermaphroditic fish. Env Biol Fish 20:301–10.
Fischer, EA and CW Petersen. 1986. Social behavior of males and simultaneous hermaphrodites in the lantern bass. Ethology 73:235–46.
Fischer, EA and CW Petersen. 1987. The evolution of sexual patterns in the seabasses. Bioscience 37:482–9.[CrossRef]
Friedman, JW and P Hammerstein. 1991. To trade, or not to trade; that is the question. In Seldon, R (Ed.). Game equilibrium models New York Springer-Verlag pp. 257–75.
Ghiselin, MT. 1969. The evolution of hermaphroditism among animals. Q Rev Biol 44:189–208.[CrossRef][Medline]
Hastings, PA and CW Petersen. 1986. A novel sexual pattern in serranid fishes: simultaneous hermaphrodites and secondary males in Serranus fasciatus. Env Biol Fish 15:59–68.
Havelange, S and J Voss. 1993. Le comportement reproducteur d'un poisson hermaphrodite synchrone: Serranus scriba (Linne, 1758). Revue fr Aquariol 20:1–4.
Kiflawi, M, AI Mazeroll, D Goulet. 1998. Does mass spawning enhance fertilization in coral reef fish? A case study of the brown surgeonfish. Mar Ecol Prog Ser 172:107–14.
Landolfa, MA. 2002. On the adaptive function of gamete trading in the black hamlet Hypoplectrus nigricans. Evol Ecol Res 4:1191–9.
Lejeune, P, J-M Boveroux, J Voss. 1980. Observation du comportement reproducteur de Serranus scriba Linne (Pisces, Serranidae), poisson hermaphrodite synchrone. Cybium 10:73–80.
Leonard, JL. 1990. The hermaphrodite's dilemma. J Theor Biol 147:361–72.[CrossRef][Web of Science][Medline]
Leonard, JL. 1993. Sexual conflict in simultaneous hermaphrodites: evidence form serranid fishes. Env Biol Fish 36:135–48.
Marconato, A, V Tessari, G Marin. 1995. The mating system of Xyrichthys novacula: sperm economy and fertilization success. J Fish Biol 47:292–301.[CrossRef]
Marconato, A, DY Shapiro, CW Petersen, RR Warner, T Yoshikawa. 1997. Methodological analysis of fertilization rate in the bluehead wrasse Thalassoma bifasciatum: pair versus group spawns. Mar Ecol Prog Ser 161:61–70.
Munoz, RC and RR Warner. 2003. A new version of the size-advantage hypothesis for sex change: incorporating sperm competition and size-fecundity skew. Am Nat 161:749–61.[CrossRef][Medline]
Munoz, RC and RR Warner. 2004. Testing a new version of the size-advantage hypothesis for sex change: sperm competition and size-skew effects in the bucktooth parrotfish, Sparisoma radians. Behav Ecol 15:129–36.
Oliver, AS. 1997. Size and density dependent mating tactics in the simultaneously hermaphroditic seabass Serranus subligarius (Cope, 1870). Behaviour 134:563–94.
Petersen, CW. 1987. Reproductive behaviour and gender allocation in Serranus fasciatus, a hermaphroditic reef fish. Anim Behav 35:1601–14.[CrossRef]
Petersen, CW. 1990a. The relationships among population density, individual size, mating tactics, and reproductive success in a hermaphroditic fish, Serranus fasciatus. Behaviour 113:57–80.
Petersen, CW. 1990b. Variation in reproductive success and gonadal allocation in the simultaneous hermaphrodite, Serranus fasciatus. Oecologia 83:62–7.[CrossRef]
Petersen, CW. 1991. Sex allocation in hermaphroditic sea basses. Am Nat 138:650–67.[CrossRef]
Petersen, CW. 1995. Reproductive behavior, egg trading, and correlates of male mating success in the simultaneous hermaphrodite, Serranus tabacarius. Env Biol Fish 43:351–61.[CrossRef]
Petersen, CW and EA Fischer. 1986. Mating system of the hermaphroditic coral-reef fish, Serranus baldwini. Behav Ecol Sociobiol 19:171–8.
Petersen, CW and EA Fischer. 1996. Intraspecific variation in sex allocation in a simultaneous hermaphrodite: the effect of individual size. Evolution 50:636–45.[CrossRef]
Petersen, CW, RR Warner, DY Shapiro, A Marconato. 2001. Components of fertilization success in the bluehead wrasse, Thalassoma bifasciatum. Behav Ecol 12:237–45.
Pressley, PH. 1981. Pair formation and joint territoriality in a simultaneous hermaphrodite: the coral reef fish Serranus tigrinus. Z Tierpsychol 56:33–46.
Sella, G. 1988. Reciprocation, reproductive success, and safeguards against cheating in a hermaphroditic polychaete worm, Ophryotrocha diadema Åkesson, 1976. Biol Bull 175:212–7.
Sella, G and MC Lorenzi. 2000. Partner fidelity and egg reciprocation in the simultaneously hermaphroditic polychaete worm Ophryotrocha gracilis (Huth). Behav Ecol 11:260–4.
Shapiro, DY, A Marconato, T Yoshikawa. 1994. Sperm economy in a coral reef fish. Ecology 75:1334–44.[CrossRef][Web of Science]
Taborsky, M. 1994. Sneakers, satellites, and helpers: parasitic and cooperative behavior in fish reproduction. Adv Stud Behav 23:1–100.
Warner, RR. 1975. The adaptive significance of sequential hermaphroditism in animals. Am Nat 109:61–84.[CrossRef][Web of Science]
Warner, RR, DY Shapiro, A Marconato, CW Petersen. 1995. Sexual conflict: males with the highest mating success convey the lowest fertilization benefits to females. Proc R Soc Lond B 262:135–9.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  J. L. Leonard Sexual selection: lessons from hermaphrodite mating systems Integr. Comp. Biol., August 1, 2006; 46(4): 349 - 367. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||