The Society for Integrative and Comparative Biology
Patterns of Parental Investment and Sexual Selection in Teleost Fishes: Do They Support Bateman's Principles?1
1 SUNY at Buffalo, Department of Biological Sciences, Buffalo, New York 14260
 |
SYNOPSIS |
|---|
TOP
 SYNOPSIS INTRODUCTIONSUBSTRATE BROODING FISHESFISHES WITH INTERNAL...CONCLUSIONSReferences |
|---|
Bateman demonstrated differences in variance for fertility and mating success between the sexes, with males usually having a greater variance than females. Thus in general, male reproductive success increases with number of mates acquired. These results have been referred to as "Bateman's principles" and taken together with other parameters (e.g., relative parental investment) have been proposed to estimate a component of sexual selection. For this review I examine patterns of parental care and sexual selection in teleost fishes (substrate brooding and with internal fertilization). I present data for the pumpkinseed sunfish Lepomis gibbosus, in which I estimated cost of paternal care and compared direct measures of the intensity of selection on possible sexually selected traits to measures of sexual selection based on Bateman's principles.
Despite high levels of paternal care in substrate brooding fishes, sexual selection tends to act more strongly on males than on females, which suggests that maternal investment is higher than paternal investment and that parental care does not limit the reproductive rate for males. In pumpkinseed sunfish, selection favors parents with high levels of defense that may exclude predators more effectively and, as suggested by Bateman's measures, alternative reproductive strategies may decrease the opportunity for sexual selection within the parental strategy. In teleost fishes with internal fertilization, patterns of parental investment and intensity of sexual selection seem to support Bateman's principles, but further studies using these systems and these measures of selection will improve the understanding of factors affecting the intensity of sexual selection and its relation to mating systems.
|
INTRODUCTION |
|---|
|
TOP
SYNOPSISINTRODUCTIONSUBSTRATE BROODING FISHESFISHES WITH INTERNAL...CONCLUSIONSReferences |
|---|
From a series of experiments with Drosophila melanogaster, Bateman (1948)
found that (1) males had greater variance in number of offspring produced (fertility) than did females, and (2) that this differential variance in fertility between the sexes was due to differential mating success (defined as number of mates): most females mated with similar number of males, while many males did not mate at all and some mated with several females. Finally, Bateman showed that (3) female reproductive success was not proportionally affected by mating success, whereas male reproductive success increased with the number of mates acquired (the Bateman gradient). Bateman (1948) considered this gradient as the cause of what he referred to as "intra-male" selection. Hence the relationship between number of mates and reproductive success may also be referred as the sexual selection gradient. These results have been generalized for most sexually reproducing organisms and are referred to as "Bateman's principles" (Arnold, 1994; Arnold and Duvall, 1994).
Bateman's principles have been used to derive statistical measures for components of sexual selection (Wade, 1979; Wade and Arnold, 1980; Arnold, 1994; Arnold and Duvall, 1994). From Bateman's first principle, the standardized variance in reproductive success (the opportunity for selection, I) provides the maximum possible strength of selection acting on a population, including both natural and sexual selection (Wade, 1979; Wade and Arnold, 1980). Based on the second principle, the standardized variance in mating success (the opportunity for sexual selection, IS) provides the maximum strength of sexual selection (Wade, 1979; Wade and Arnold, 1980). Finally, the Bateman gradient indicates how much of the total reproductive success is determined by mating success. A strong gradient would indicate that individual reproductive success is mostly determined by number of mates (strong sexual selection is possible), whereas a weak gradient indicates that reproductive success is not determined by number of mates (little or no sexual selection is possible). Bateman's results implied that the intensity of sexual (particularly intrasexual) selection was higher for males than for females, and that only males had the opportunity to increase their reproductive success by mating with several mates. Because Bateman (1948) proposed that this sex difference in the intensity of sexual selection was caused by differential investment in gamete production (anisogamy), sex differences in mating behavior could ultimately be explained by relative investment in gametes. Since parental investment may continue beyond the production of gametes, an overall measure of relative parental investment might be a better measure of sexual selection (Trivers, 1972). In general, the sex that contributes less parental investment will also have higher variance in mating success and be under stronger sexual selection. Historically, these results have been used to explain general differences in mating behavior between the sexes (i.e., choosy females and undiscriminating males, Bateman, 1948; Trivers, 1972).
Levels of parental investment affect potential rates of mating (PRM, as defined by Sutherland, 1985a) and reproduction (PRR, see Clutton-Brock and Vincent, 1991; Clutton-Brock and Parker, 1992), and ultimately the operational sex ratios (OSR, Emlen and Oring, 1977). Therefore, these parameters (PRM, PRR and OSR) have been proposed as measures of sexual selection; but because their reliability has been questioned (Sutherland, 1985a; Andersson, 1994) I do not further consider them for this review. Maternal investment is often relatively higher than paternal investment among endothermic vertebrates (although some groups, such as birds, also show pronounced paternal investment). Even among ectothermic vertebrates (where parental care is rare), high female investment is the norm for those reptiles, amphibians and non-teleost fishes that provide care (see Clutton-Brock [1991] for a review). An exception is the teleost fishes, where paternal care is the most common form of care in those families that provide it (Sargent and Gross, 1993). Despite apparent high levels of paternal investment in teleost fishes, sexual selection appears to act more strongly on males than on females (e.g., bullheads, centrarchids, cichlids, damselfishes, gobies, minnows, sculpins and sticklebacks for reviews see Turner, 1993; Dugatkin and FitzGerald, 1997). For this reason, the relationship between patterns of investment and sexual selection are particularly interesting in this group.
This paper is not intended as a comprehensive review of the relationship between parental investment and sexual selection and/or Bateman's principles in teleost fishes. Instead, I present particular examples from teleost systems where at least one of these measures is known. In the first section I consider substrate brooders, where male parental care is the norm. I concentrate on the pumpkinseed sunfish Lepomis gibbosus. I present direct measures of the intensity of selection on potentially sexually selected traits (selection differentials and gradients) in this species, and compare these direct measures to measures of sexual selection based on Bateman's principles (including the opportunity for sexual selection and Bateman's gradient). In the second section I consider fishes with internal fertilization (an important group within the teleosts) where the most complete studies come from sex-role-reversed species. Comparisons of parental care and the strength of sexual selection between sex-role-reversed and conventional sex role species would be ideal. However, more complete studies of conventional sex role species with internal fertilization are needed (see below).
|
SUBSTRATE BROODING FISHES |
|---|
|
TOP
SYNOPSISINTRODUCTIONSUBSTRATE BROODING FISHESFISHES WITH INTERNAL...CONCLUSIONSReferences |
|---|
There is some evidence that the strength of sexual selection is large on females of some substrate brooder fishes with male parental care (e.g., Sargent et al., 1986
; Forsgren et al., 2004). Specifically, conspicuous coloration occurs in courting females, but not in males or juveniles (e.g., reddish abdomen in gobies, Hidaka and Takahashi, 1987; Forsgren et al., 2004). In addition to coloration patterns, females of biparental cichlids aggressively keep rivals away from their partner (Turner, 1993). Furthermore, three-spined stickleback males choose larger females because they are more fecund (Sargent et al., 1986). These examples suggest that substrate brooding in fishes might correspond to high levels of male investment. However, there are no known examples where both costs of paternal investment and measures of sexual selection have been obtained for a species of substrate brooder fish. In this section, I present and discuss estimates of these parameters for the pumpkinseed sunfish L. gibbosus.
Paternal investment in pumpkinseed sunfish
In pumpkinseed sunfish, the parental period begins when males start defending a territory. Parental males dig a nest to which they attract females to spawn (Scott and Crossman, 1973). Female pumpkinseeds approach territories of nesting males to spawn. A female lays a small number of eggs during single dips that are repeated several times during the duration of a mating event (Scott and Crossman, 1973). Defense continues during the mating period and extends for a period of about ten days, when the free-swimming fry finally disperse from the nest (Carlander, 1977, personal observation). Guarding the territory also consists of "patrolling" the nest. In addition, pumpkinseed males fan the nest to circulate water among eggs to increase their oxygenation. Some males ("sneakers") adopt an alternative reproductive strategy in which they do not have nests of their own and instead intrude during spawnings at the nests of parental males to steal fertilizations (Gross, 1979; O.R.-C., unpublished data).
In pumpkinseed sunfish parental males lose weight between the day they receive eggs and the end of the parental care period (Rios-Cardenas and Webster, 2005). Pumpkinseed sunfish do not feed during the winter, as the stomach is shrunken and mucus filled, but an increase in water temperature initiates a feeding response (Carlander, 1977). Furthermore, growth rate (measured as weight gain) and average food intake increase with water temperature, particularly during the breeding season months (May–September; Carlander, 1977). Therefore, parental males seem to lose weight in large part because they do not leave the nest to feed during the parental care period. This reduced feeding while tending the nest is directly related to providing care to the brood. Furthermore, parental care activities certainly contribute to the deficit of energy revealed by this weigh loss. Thus, even though weight loss of juvenile or non-reproductive but mature individuals was not monitored, this weight loss suggests that parental care in this species is energetically costly. In addition, defensive behaviors have other costs (e.g., increased risks of predation and of agonistic retaliation and subsequent aggressive escalation by the intruders; see Grant [1997] for a review). Costs of paternal care will likely have an effect on future survival and/or reproductive success. Costs of parental care might be mitigated if males with eggs in their nest are more likely to receive eggs from additional females than are males without eggs, as has been found in some other fish (e.g., Unger and Sargent, 1988). However, in pumpkinseed sunfish, both observational and manipulative experiments showed that males with large clutches lose eggs over the parental care period whereas males with small clutches acquire additional eggs (Rios-Cardenas and Webster, 2005).
To determine whether male parental investment is in fact higher than that of females, ideally we should have an equivalent estimate of maternal investment. Unfortunately, as is often the case in field studies, obtaining comparable measures of parental investment for both sexes is difficult and such data are not available for pumpkinseed females. However, a previous study using substrate brooder fishes (pumpkinseed sunfish, bluegill sunfish L. macrochirus, rock bass Ambloplites rupestris, three-spine sticklebacks Gasterosteous aculeatus and mottled sculpins Cottus bairdi) with paternal care show that female future reproductive success increases with accelerating returns with body length. In contrast, male future reproductive success increases with diminishing returns (Gross and Sargent, 1985). This suggests that in these fishes, the production of eggs requires a higher investment and results in a higher cost on future reproductive success than that of sperm and paternal care. In general, most fish species with paternal care do not show any signs of sex-role reversal, but still show signs of sexual selection acting most strongly on males (Clutton-Brock and Vincent, 1991; Turner, 1993; Dugatkin and FitzGerald, 1997). For example, the pumpkinseed sunfish shows pronounced sexual size dimorphism, with males being larger than females (Scott and Crossman, 1973; O.R.-C., unpublished data). Additionally, with respect to brood age, models that optimize reproductive effort (defined as the proportion of parental resources devoted to fecundity, including mating effort, and offspring survivorship, which includes parental effort) based on the loss in expected future reproduction that is attributable to present reproduction (see Sargent and Gross [1993] for further details) predict an increase in parental care during the egg stage of the young, followed by a decrease in care after hatching (Sargent and Gross, 1993). In the pumpkinseed sunfish, patterns of male nest guarding and offspring age do not support this prediction (O.R.-C., unpublished data). Instead, both patrolling and defensive behaviors decrease within the egg stage and remain relatively low after the first day eggs are received (Rios-Cardenas, 2003). This pattern of parental care may be related to sexual selection, because it is during the first two days, when males receive eggs from females, that the mating period and the parental period overlap. Other examples of higher-than-expected level of parental care during the mating period exist in fifteen-spined sticklebacks Spinachia spinachia (Östlund and Ahnesjö, 1998) and sand gobies Pomatoschistus minutus (Pampoulie et al., 2004).
As mentioned before, there is no equivalent estimate of maternal investment to compare with the known estimate of paternal investment for the pumpkinseed sunfish. However, Bateman's gradient has been recognized as a fundamental aspect in the process of sexual selection (Arnold and Duval, 1994) and has been favored as a measure of sexual selection (Jones et al., 2000). Therefore, to better understand the effects of sexual selection on the pumpkinseed sunfish mating system, I analyzed Bateman's principles, and direct measures of the intensity of selection (using selection gradients and differentials) on apparently sexually selected traits (male size and male parental behavior).
Direct measures of the intensity of selection and Bateman's principles in pumpkinseed sunfish
Material and methods
I conducted this study in Lincoln Pond, situated in The Edmund Niles Huyck Preserve, southwest of Albany, New York during the 1999–2001 breeding seasons. This study was part of a multi-year investigation of the mating and paternal care patterns. Therefore, I did not conduct an extensive (but intrusive) sampling of this pumpkinseed sunfish population that would have allowed me a more equivalent sampling of both sexes (since parental males are territorial, while females individually visit territories only momentarily to lay eggs, the probabilities of capturing females were extremely low when compared to males). The methods used to collect samples and data on measures of body size (weight, standard length, depth and width), parental behaviors (frequency of fanning, patrolling and defensive behavior) and paternity of pumpkinseed sunfish have been described elsewhere (Rios-Cardenas and Webster, 2005). In brief, I used microsatellite analyses (at five loci) to obtain genotypes of 35 parental males (putative fathers at a nest), a sample of each parental male's brood (on average 49 young per nest), and 139 other adults in the population (including 21 females). I used these data to estimate the proportion of offspring within a clutch that were sired by the parental male (following Neff et al., 2000) and their reproductive success.
For all nests, I also used parental male genotypes and progeny arrays (excluding any young that could have not been sired by their putative father) to reconstruct maternal genotypes and the minimum number of mothers contributing young at each nest. For this analysis I used the program Gerud 1.0 (Jones, 2001). I initiated the analysis of every nest with an exhaustive search. However, because of constraints on computation time, for those nests that showed a partial result of at least five mothers (six cases), I switched to a targeted search (with an initial search genotype number of 10 and a later search genotype number of 99). Finally, I used this minimum number of mothers at each nest to estimate mating success of parental males (defined here as the minimum number of females that produced young with the parental male).
Whenever a progeny array was consistent with multiple combinations of maternal genotypes, I ranked the solutions by likelihood using information on both Mendelian segregation and allele frequency data (see Jones, 2001). Because I only captured five females that laid eggs in sampled nests, I relied on the results provided by Gerud 1.0 (the estimated mother genotypes from single solution arrays, the most likely combination of maternal genotypes from multiple solution arrays, and the number of young compatible with each estimated genotype, Jones, 2001) to estimate the number of nests in which a female produced young. These estimates of reproductive and mating success should be considered with caution since they are derived from estimated genotypes and not actual genotypes, and thus this method may have underestimated the number of females laying eggs in more than one nest. There are no quantitative descriptions of female multiple matings (at different nests). However, when these events occur, females deposit most of their eggs at one nest (Gross, 1980). This has two consequences: (1) the likelihood of finding the same female's eggs in multiple nests is low; and (2) the underestimation of the number of females laying in multiple nests will mainly affect the estimate of the opportunity for sexual selection, but the estimation of the opportunity for selection and, more important, Bateman's gradient will be affected to a lesser degree.
Using the estimates of both mating and reproductive success, and following Lande and Arnold (1983), I calculated selection differentials and gradients (which measure the strength of selection on particular traits) for male size and behavioral (parental care) traits. Briefly, I transformed absolute fitnesses (based on mating and reproductive success) to relative fitnesses. I used the cube root of body weight to make its dimensionality the same as that of other body measures. I transformed all phenotypic measurements to natural logarithms and then standardized them to have a variance of one by expressing them in terms of their standard deviations. Selection differentials are given by the covariance between the trait and fitness. I calculated selection gradients using multiple regressions of all size and behavioral traits independently.
For the analysis of behavioral traits I included parental behaviors observed only during the first two days of the parental care period. With respect to reproductive success, selection on guarding behaviors could result from intrasexual selection (male–male competition to increase paternity) or natural selection (to increase survival of the brood). To distinguish between these alternatives for guarding behaviors that showed significant levels of selection with respect to reproductive success, I analyzed the relationships between guarding and both proportion of paternity and number of young produced. Intrasexual selection predicts a positive relationship between guarding behaviors and proportion of paternity. Natural selection predicts no such difference, but does predict a positive relationship between guarding behaviors and number of total young.
Finally, I estimated the opportunity for selection (I), sexual selection (IS), and the Bateman gradient (ßss) for female and male pumpkinseed sunfish. I is the variance in reproductive success divided by its mean reproductive success squared (Wade [1979]; mating success in the case of IS, Wade and Arnold [1980]). I estimated ßss by the weighted least-square regression of reproductive success on mating success (Arnold and Duvall, 1994; Arnold, 1994). When estimating mating and reproductive success from paternity analysis, nonmating individuals are not accounted (because they do not produce offspring at all). This poses a problem since excluding nonmating individuals increases estimates of average reproductive success and reduces the variance, causing and underestimation of the potential strength of sexual selection (Wade and Shuster, 2004). On the other hand, individuals who are incapable of mating should not be included in the analysis since they would cause the opposite effect, an overestimation of the potential strength of sexual selection (Arnold and Duval, 1994). For this reason and for males only, I estimated I, IS and ßss using data from the paternity analysis (only mating males), but also considering nonmating males.
The sample of successful individuals from which paternity analyses were performed represents about 12% of the parental males in the population that obtained eggs over the three breeding seasons. During this same period, I observed 134 individuals with territories who failed to mate (the nest was abandoned before obtaining eggs). Therefore, I only included 16 nonmating males (12% of the 134 observed over three years) for the analysis considering these types of males. However, some of these nonmating males could have had successful attempts later in the season, thus including them would cause an overestimation of Bateman's principles. As an alternative, I also estimated the opportunity of sexual selection of males as measured by harem size (H + (k – 1), Wade and Shuster, 2004). Where H is the mean number of mates per mating males, and k is the ratio of the variance in harem size (Vharem) and H.
Results
Analyses of selection differentials and selection gradients for body size of pumpkinseed males using reproductive and mating success as measures of fitness, showed no significant selection on any of the body measures (Table 1). Thus male size did not seem to affect the number of young produced or the number mates attracted by parental males.
|
Analyses of parental behaviors showed that for most behaviors, there was no significant selection with respect to reproductive success or mating success, the one exception was the level of defense with respect to reproductive success where the selection differential was significantly different from zero and the selection gradient was marginally significant (Table 2). These results suggest that males that show higher levels of defense during the first two days of the parental care period are able to produce more of their own young. Defense behaviors during the first two days of the parental care period were not significantly related to proportion of paternity (Linear regression: R = 0.035, F1,21 = 0.025, N = 23, P = 0.874), but there was a significant and positive relationship between those defensive behaviors and number of total young produced (Linear regression: R = 0.493, F1,21 = 6.724, N = 23, P < 0.05), thus males that showed high levels of defense were also the ones that produced more total young.
|
On average, females produced 207 young and laid eggs in a single nest, while males produced 517 young and received eggs from 3.6 females. Based on the analyses that only included mating males, the opportunity for selection was higher for females than it was for males (Table 3). In addition, the opportunity for sexual selection was low and very similar between females and males and represented only 9% and 20% of the opportunity for selection for females and males respectively (Table 3). In both sexes the Bateman's gradient was not significantly different from zero (Table 3; Fig. 1A). Based on the analyses that included nonmating males, as expected, both the opportunity for selection and the opportunity for sexual selection were higher than when nonmating males were excluded from the analysis. This time, the opportunity for sexual selection represented 52% of the opportunity for selection (Table 3). Bateman's gradient was significantly different from zero when nonmating males were included (Table 3; Fig. 1B). As long as more than 7% of the 134 unmated individuals observed during the three years (about 9 individuals) were capable of mating, but failed to do so because they lost in reproductive competition (legitimate nonmating individuals), Bateman's gradient would be significant (data not shown). Finally, the opportunity of sexual selection as measured by Wade and Shuster's (2004)
harem size was: (H + [k – 1] = 3.63 + [0.37 – 1] = 3).
|
|
Discussion
The energetic costs of paternal care suggest that male parental investment is high for pumpkinseed sunfish. There is no equivalent measure of the cost for females. However, body size reduction of females appear to have higher costs on their fertility than on males (Gross and Sargent, 1985
). This may result in higher relative parental investment by females than males despite the lack of maternal care. Moreover, males often provided care to broods from multiple females simultaneously (average of 3.6 females contributing to a nest), whereas most females deposited eggs in a single nest. For those individuals that breed multiple times in a season this may further result in a faster succession of clutches for males than females. Thus, pumpkinseed sunfish males seem to have a higher potential reproductive rate than do females (consequently, the operational sex ratio may be male-biased). My results support the idea that paternal care in substrate brooding fishes represents low levels of reproductive investment since it does not seem to affect the ability to produce more offspring. Low costs of parental care have been suggested as the reason for the evolution of male parental care in this group of fishes (Gross and Sargent, 1985).
Analyses of selection differentials and gradients did not show significant selection acting on male body size or most paternal behaviors. Given that alternative reproductive strategies exist in this species (Gross, 1979; O.R.-C., unpublished data) and the large differences in size between parental and sneaker males (O.R.-C., unpublished data), variation in body size among parentals is not likely to affect the efficiency of defense against sneakers. Thus, large parental males did not necessarily produce more young. Lack of a strong correlation between size and reproductive success has also been attributed to males that act like "sneaker" males in Atlantic salmon Salmo salar (Garant et al., 2001).
In addition, because males with large body size or high levels of defense did not attract more females to their nest, pumpkinseed sunfish females do not seem to be choosing among successful parental males (those that were able to obtain some eggs) based on these traits. However, there was significant selection for males that showed high levels of defense during the first two days of the parental care period, when males receive eggs. Because levels of defense during this period were not related to paternity, but were related to total brood production, natural selection seems to favor more defensive males that may guard their nest against predators more efficiently. This result might also arise if females laid more eggs for more defensive males. This does not appear to be the case, as levels of defense during the first two days of the parental care period do not improve the probability of receiving more eggs (Rios-Cardenas and Webster, 2005).
Measures of Bateman's principles based on the analyses with only mating males did not show a high potential for selection for parental males relative to that of females. The opportunity for selection was almost twice as large for females as for males. The opportunity for sexual selection was low and similar for both sexes and on males only represented 20% of the opportunity for selection. Bateman's gradient was not significant for either sex. However, when nonmating males were considered in the estimation of Bateman's principles the opportunity for sexual selection represented the majority of the opportunity for selection and Bateman's gradient was significantly different from zero. From 134 individuals that apparently failed to mate, it is very likely that at least 9 of these individuals are legitimate nonmating males (see above). Thus, a significant level of the opportunity for sexual selection appears to be real and not an overestimation caused by the inclusion of nonmating individuals. Furthermore, a value of harem size of 3 when k = 0.37 suggests that a good part of the variance in male reproductive success comes from the difference between males wining and losing in reproductive competition (Wade and Shuster, 2004).
Together these results suggest that sexual selection is strong on males, but it is mainly responsible for sorting out those males that are unable to maintain a territory and/or are incapable to attract females to it, from those that can. However, among successful parental males the opportunity for sexual selection is lessen and having larger bodies or attracting additional females does not significantly increase their reproductive success. It is possible that a threshold body size exists that defines unsuccessful from successful parental males. The inclusion of body sizes for legitimate nonmating individuals in the analyses of selection differentials and gradients may have shown significant selection on this trait. The threshold in body size may explain the size dimorphism between females and parental males; a mechanism like this might have originally favored and may be maintaining the alternative life histories that exist in the pumpkinseed sunfish. Disruptive selection should be favoring delayed maturation of parental males, and those unsuccessful males may be just matured (small) parentals who have very low reproductive success (if any at all) in their first reproductive season, and need to grow a little bit more to become successful in the following year.
The discrepancy between the direct measure of selection on defense behavior and Bateman's measures of the potential for selection among successful parental males might be related to the behaviors of females and sneaker males. Because females lay eggs in multiple nests (a maximum number of 3 nests may be an underestimation), females partition their eggs into various clutches lowering the number offspring per mating for parental males. As suggested by others (Shuster and Wade, 2003), multiple mating by females may reduce the variance in male reproductive success among parental males and also increase the chances of sperm competition. As a result, the opportunity for selection would appear to be low.
In bluegill sunfish, females lay more eggs per dip when sneakers are present (Fu et al., 2001). This in part gives an advantage to sneakers that end up fertilizing more eggs than parentals per dip (i.e., during sperm competition). If this is also true for pumpkinseed sunfish, parental males may mate many times, but they might lose in sperm competition, reducing the number of young they produce. This would further decrease the strength of selection on parental males (e.g., Lorch, 2002). Even if pumpkinseed sunfish sneakers do not win at sperm competition, the low levels of the opportunity for sexual selection found in this study agree with theoretical predictions. When males with otherwise very low mating success adopt alternative reproductive strategies, a negative covariance between social reproductive success (offspring produced per nest) and cuckoldry success (young produced in nests of other males) is produced, particularly in highly polygynous species like pumpkinseed sunfish, where all cuckoldry is done by sneaker (non-nesting) males (Webster et al., 1995; Jones et al., 2001).
The use of paternity analyses to conduct studies of Bateman's principles in wild populations of substrate brooding fishes has certain limitations: not all individuals (females and males) contributing to a clutch are sampled, thus only estimated genotypes of other possible parents can be used. In this study, the use of estimated mother genotypes may have underestimated mainly the number of mates acquired by females (see above). These limitations also prevent the estimation of the variance in reproductive and mating success for sneaker males and for males in overall (both strategies). In unmanipulated wild populations it is also difficult to distinguish between legitimate nonmating individuals from those who are incapable to mate. Fortunately, one of the advantages of working with substrate brooders fishes is that most species can be bred in the lab. Thus, researchers considering the study of Bateman's principles in fish may want to complement field studies with breeding experiments under controlled conditions, this would take care of most of the limitations of field studies.
Salmons are also substrate brooders but females are the nesting sex and males compete for access to them. For both sexes of Atlantic salmon there is a correlation between mating success and reproductive success. However, body size does not affect reproductive success (Garant et al., 2001). In contrast, in Pacific salmon Oncorhynchus kisutch there is selection on body size for both sexes, but sexual selection is stronger for males (Fleming and Gross, 1994). Clearly, there is much that needs to be done to have a complete understanding of the relationship between the strength of sexual selection and mating systems; however, substrate brooder fishes are a promising group where this type of studies may provide rewarding results.
|
FISHES WITH INTERNAL FERTILIZATION |
|---|
|
TOP
SYNOPSISINTRODUCTIONSUBSTRATE BROODING FISHESFISHES WITH INTERNAL...CONCLUSIONSReferences |
|---|
Sex-role-reversed species
Sex-role reversed systems refer to those species where female reproduction is limited by access to parental males, thus females compete for males. In some extreme cases represented by species of the fish family Syngnathidae (e.g., pipefishes), males in fact become pregnant (Rosenqvist, 1990
). Aggressive interactions between females for access to males (female intra-sexual selection) are known in pipefish, with subordinate females sometimes being inhibited from developing brightly colored skin that is attractive to males (Rosenqvist, 1990).
Only a single study of sex-role reversed fish (Jones et al., 2000) has determined the Bateman gradient for both sexes. As expected from the high levels of paternal investment in sex-role-reversed species, males of the broad-nosed pipefish (Syngnathus typhle) tend to discriminate among mates, while females compete for access to mates more intensely than do males (Berglund and Rosenqvist, 1993). Thus, sexual selection is expected to be stronger in females than in males. When Jones et al. (2000) analyzed the relationship between mating success and reproductive success (Bateman's gradient), they found a significant Bateman gradient for females, which in addition was significantly steeper than that of males. Studies of the strength of sexual selection in sex-role-reversed species are important because of the specific predictions derived from these systems. Since the direction of sexual selection was in fact reversed for the broad-nosed pipefish, this study represents the first empirical test of the possible generality of the Bateman gradient as a measure of the potential for sexual selection (but see above).
Conventional sex-roles
Most teleost fishes with internal fertilization belong to the Cyprinodontiformes, where the best-studied group is the livebearers (family Poeciliidae). In contrast to sex-role-reversed species, the females in this group become pregnant and provide all parental care. Thus, in this group we would expect maternal investment to be higher than paternal investment and hence sexual selection to be stronger in males than in females. In many livebearers, males are more colorful than females and the most brightly colored males obtain the most matings, presumably through female choice (for reviews see Turner, 1993; Dugatkin and FitzGerald, 1997; Houde, 1997). Females also choose males based on body size (e.g., Reynolds and Gross, 1992) and/or the presence of an ornament (e.g., Basolo, 1998), a sexual signal (e.g., Morris, 1995), parasitic load (which is correlated to coloration and courtship display rates, Houde [1997]), or they may copy other female's choices (see Dugatkin and FitzGerald [1997] for a review). However, males may also exercise mate choice for large (fecund) or virgin (or recently spent) females (see Turner [1993] for review).
No study exists on this group that estimates the potential for sexual selection using Bateman's principles. However, a study with Xiphophorus maculatus analyzed the relationship between mating and reproductive success and found what seems to be a high potential for sexual selection for females! (See Fig. 1 in Borowsky and Kallman [1976]). This might imply that the direction of sexual selection is reversed (as in the sex-role-reversed fishes) or that sexual selection might be strong for both sexes. Unfortunately, these same data do not exist for males and thus more studies are needed before we can tell more about the actual patterns of the intensity of sexual selection for this group.
Discussion
In teleost fishes with internal fertilization, patterns of parental investment and the intensity of sexual selection seem to support Bateman's principles. In conventional sex-role fishes, general patterns of investment and sexual selection might be congruent with Bateman's principles, and in sex-role reversed species the intensity and direction of sexual selection are inverted. However, these conclusions are based on a very small number of studies and more studies that specifically measure parental investment and the intensity of selection for both sexes are needed.
Further studies using species with internal fertilization may prove particularly rewarding for several reasons. First, internal fertilization ensures that the parentage of at least one of the parents is certain, facilitating parentage assignment. In addition, the size of most live-bearing fish makes them more manageable for controlled breeding experiments. Furthermore, alternative reproductive strategies are common among livebearers (e.g., Houde, 1997). Future studies could estimate parental investment for males and females, the intensity of selection of sexually selected traits, and measures of the potential for selection based on Bateman's principles for females and all types of males participating in alternative strategies. A study with these characteristics will provide information on the factors affecting the intensity of selection in this group of fishes, and a better understanding of the relationship between sexual selection and mating systems.
|
CONCLUSIONS |
|---|
|
TOP
SYNOPSISINTRODUCTIONSUBSTRATE BROODING FISHESFISHES WITH INTERNAL...CONCLUSIONSReferences |
|---|
Despite high incidence of paternal care in substrate brooding teleost fishes, sexual selection tends to act more strongly on males than on females. This suggests that maternal investment is higher than paternal investment and that this form of parental care allows a higher potential reproductive rate for males and a male-biased operational sex ratio. However, further studies should test these predictions for specific systems.
In pumpkinseed sunfish, selection differential and selection gradient for high levels of defense suggest that natural selection is acting on parental males. Parentals that have high levels of defense during the period when they receive eggs may be able to exclude predators more effectively, thus producing more young than other less aggressive parentals. The low levels of the potential for selection suggested by Bateman's measures of successful parental males provide empirical evidence that particularities of the mating system (i.e., alternative reproductive strategies) may "erode" (Shuster and Wade, 2003) or decrease the opportunity for selection (Jones et al., 2001). When considering nonmating males, high levels of the opportunity for selection (specially sexual selection) suggest that sexual selection determines the success of mating males over nonmating males and that this might be based on body size. However, further studies should test this prediction.
As suggested by the studies presented here, to fully understand the effects of sexual selection on animal mating systems we need to obtain estimates of the intensity of sexual selection. Predicting intensity of sexual selection based solely on measures of parental investment may be as limited (e.g., Wilson et al., 2003) as using only measures of gamete investment (see Wade and Shuster, 2002). On the other hand, measures of the intensity of sexual selection based on Bateman's principles may have flaws (Sutherland, 1985a, b; Grafen, 1987; Sutherland, 1987) and in systems with alternative reproductive strategies such mating strategies affect the opportunity for sexual selection as defined by these measures. However, even when direct measures of the intensity of selection on sexually selected traits and Bateman's measures disagree, their study provides valuable information on the specific factors affecting the intensity of sexual selection in the mating system under study. This suggests to me we should not disregard Bateman's principles just yet.
|
ACKNOWLEDGMENTS |
|---|
I would like to thank Zuleyma Tang-Martinez for organizing and inviting me to this symposium. Mike Webster, Molly Morris and Kevin De Queiroz provided comments that improved this paper. I thank The Edmund Niles Huyck Preserve and Biological Field Station and Molly Morris for logistic support provided at different stages of this study in the field and in the lab respectively. Michelle Rainka provided invaluable help in the field and in the lab. I am grateful to Samuel Holzman and Victor Callirgos for assistance in the field, and Nancy Urban, Dmitriy Akselrod and Christina Costa for assistance in the lab. I was supported by an IIE/Fulbright/Garcia-Robles/CONACYT fellowship for a portion of this study. This study was supported by grants from the Graduate Group in Evolutionary Biology and Ecology at SUNY Buffalo, Graduate and Post Graduate Research Grants from the E. N. Huyck Preserve and a Doctoral Dissertation Improvement grant from the NSF (IBN-0073536). The NSF sponsored this symposium.
|
FOOTNOTES |
|---|
1 From the Symposium Bateman's Principle: Is It Time for a Reevaluation? presented at the Annual Meeting of the Society for Integrative and Comparative Biology, 5–9 January 2004, at New Orleans, Louisiana.
2 Present address of Oscar Rios is Ohio University, Department of Biological Sciences, Irvine Hall, Athens, Ohio 45701; E-mail: rios-car{at}ohio.edu
|
References |
|---|
|
TOP
SYNOPSISINTRODUCTIONSUBSTRATE BROODING FISHESFISHES WITH INTERNAL...CONCLUSIONSReferences |
|---|
Andersson, M. 1994. Sexual selection: Monographs in behavior and ecology. Princeton University Press, Princeton.
Arnold, S. J. 1994. Bateman's principles and the measurement of sexual selection in plants and animals. Am. Nat, 144:S126-S149.[CrossRef][Web of Science]
Arnold, S. J., and D. Duvall. 1994. Animal mating systems: A synthesis based on selection theory. Am. Nat, 143:317-348.[CrossRef][Web of Science]
Basolo, A. L. 1998. Shift in investment between sexually selected traits: Tarnishing of the silver spoon. Anim. Behav, 55:665-671.[CrossRef][Web of Science][Medline]
Bateman, A. J. 1948. Intra-sexual selection in Drosophila. Heredity, 2:349-368.[Web of Science][Medline]
Berglund, A., and G. Rosenqvist. 1993. Selective males and ardent females in pipefishes. Behav. Ecol. Sociobiol, 32:331-336.[Web of Science]
Borowsky, R., and K. D. Kallman. 1976. Patterns of mating in natural populations of Xiphophorus (Pisces: Poeciliidae). I: X. maculatus from Belize and Mexico. Evolution, 30:693-706.[CrossRef][Web of Science]
Carlander, K. D. 1977. Life history data on centrarchid fishes of the United States and Canada: Handbook of freshwater fishery biology, Vol. 2. Iowa State University Press, Ames.
Clutton-Brock, T. H. 1991. The evolution of parental care: Monographs in behavior and ecology. Princeton University Press, Princeton.
Clutton-Brock, T. H., and G. A. Parker. 1992. Potential reproductive rates and the operation of sexual selection. Q. Rev. Biol, 67:437-456.[CrossRef]
Clutton-Brock, T. H., and A. C. J. Vincent. 1991. Sexual selection and the potential reproductive rates of males and females. Nature, 351:58-60.[CrossRef][Medline]
Dugatkin, L. A., and G. J. FitzGerald. 1997. Sexual selection. In J.-G. J. Godin (ed.), Behavioural ecology of teleost fishes, pp. 266–291. Oxford University Press, New York.
Emlen, S. T., and L. W. Oring. 1977. Ecology, sexual selection, and the evolution of animal mating systems. Science, 197:215-223.
Fleming, I. A., and M. R. Gross. 1994. Breeding competition in a Pacific salmon (coho: Oncorhynchus kisutch): Measures of natural and sexual selection. Evolution, 48:637-657.[CrossRef][Web of Science]
Forsgren, E., T. Amundsen, Ê. A. Borg, and J. Bjelvenmark. 2004. Unusually dynamic sex roles in a fish. Nature, 429:551-554.[CrossRef][Medline]
Fu, P., B. D. Neff, and M. R. Gross. 2001. Tactic-specific success in sperm competition. Proc. R. Soc. London B, 268:1105-1112.[Medline]
Garant, D., J. J. Dodson, and L. Bernatchez. 2001. A genetic evaluation of mating system and determinants of individual reproductive success in Atlantic salmon Salmo salar L). J. Hered, 92:137-145.
Grafen, A. 1987. Measuring sexual selection: Why bother? In J. W. Bradbury and M. B. Andersson (eds.), Sexual selection: Testing the alternatives, pp. 221–233. John Wiley & Sons Limited, Chichester.
Grant, J. W. A. 1997. Territoriality. In J.-G. J. Godin (ed.), Behavioural ecology of teleost fishes, pp. 81–103. Oxford University Press, New York.
Gross, M. R. 1979. Cuckoldry in sunfishes (Lepomis: Centrarchidae). Canad. J. Zool, 57:1507-1509.
Gross, M. R. 1980. Sexual selection and the evolution of reproductive strategies in sunfishes (Lepomis: Centrarchidae). Ph.D. Diss., Univ. of Utah, Utah.
Gross, M. R., and R. C. Sargent. 1985. The evolution of male and female parental care in fishes. Amer. Zool, 25:807-822.
Hidaka, T., and S. Takahashi. 1987. Reproductive strategy and interspecific competition in the lake-living gobiid fish isaza, Chaenogobius isaza. J. Ethol, 5:185-196.[CrossRef]
Houde, A. E. 1997. Sex, color, and mate choice in guppies: Monographs in behavior and ecology. Princeton University Press, Princeton.
Jones, A. G. 2001. GERUD 1.0: A computer program for the reconstruction of parental genotypes from progeny arrays using multilocus DNA data. Mol. Ecol. Notes, 1:215-218.
Jones, A. G., G. Rosenqvist, A. Berglund, S. J. Arnold, and J. C. Avise. 2000. The bateman gradient and the cause of sexual selection in a sex-role-reversed pipefish. Proc. R. Soc. London B, 267:677-680.[Medline]
Jones, A. G., D. Walker, C. Kvarnemo, K. Lindström, and J. C. Avise. 2001. How cuckoldry can decrease the opportunity for sexual selection: Data and theory from a genetic parentage analysis of the sand goby, Pomatoschitus minutus. P. Natl. Acad. Sci. U.S.A, 98:9151-9156.
Lande, R., and S. J. Arnold. 1983. The measurement of selection on correlated characters. Evolution, 37:1210-1226.[CrossRef][Web of Science]
Lorch, P. D. 2002. Understanding reversals in the relative strenght of sexual selection on males and females: A role for sperm competition? Am. Nat, 159:645-657.[CrossRef]
Morris, M. R. 1995. Vertical bars on male Xiphophorus multilineatus: A signal that deters rival males and attracts females. Behav. Ecol, 6:274-279.
Neff, B. D., J. Repka, and M. R. Gross. 2000. Parentage analysis with incomplete sampling of candidate parents and offspring. Mol. Ecol, 9:515-528.[CrossRef][Medline]
Östlund, S., and I. Ahnesjö. 1998. Female fifteen-spined sticklebacks prefer better fathers. Anim. Behav, 56:1177-1183.[CrossRef][Web of Science][Medline]
Pampoulie, C., K. Lindström, and C. M. St. Mary. 2004. Have your cake and eat it too: Male sand gobies show more parental care in the presence of female partners. Behav. Ecol, 15:199-204.
Reynolds, J. D., and M. R. Gross. 1992. Female mate preference enhances offspring growth and reproduction in a fish, Poecilia reticulata. Proc. R. Soc. London B, 250:57-62.
Rios-Cardenas, O. 2003. The use of molecular genetic analyses to examine mating patterns a paternal care in the pumpkinseed sunfish. Ph.D. Diss., State Univ. of New York at Buffalo, Buffalo, New York.
Rios-Cardenas, O., and M. S. Webster. 2005. Paternity and paternal effort in the pumpkinseed sunfish. Behav. Ecol. (In press).
Rosenqvist, G. 1990. Male mate choice and female-female competition for mates in the pipefish Nerophis ophidion. Anim. Behav, 39:1110-1115.[CrossRef]
Sargent, R. C., and M. R. Gross. 1993. Williams' principle, an explanation of parental care in teleost fishes. In T. J. Pitcher (ed.), Behaviour of teleost fishes, pp. 275–293. Chapman and Hall, London.
Sargent, R. C., M. R. Gross, and V. D. Berghe. 1986. Male mate choice in fishes. Anim. Behav, 34:545-550.[CrossRef]
Scott, W. B., and E. J. Crossman. 1973. Freshwater fishes of Canada: Fisheries Research Board of Canada Bulletin, Vol. 184. Fisheries Research Board of Canada, Ottawa.
Shuster, S. M., and M. H. Wade. 2003. Mating systems and strategies: Monographs in behavior and ecology. Princeton University Press, Princeton.
Sutherland, W. J. 1985a. Measures of sexual selection. Oxf. Surv. Evol. Biol, 2:90-101.
Sutherland, W. J. 1985b. Chance can produce a sex difference in variance in mating success and explain Bateman's data. Anim. Behav, 33:1349-1352.[CrossRef]
Sutherland, W. J. 1987. Random and deterministic components of variance in mating success. In J. W. Bradbury and M. B. Andersson (eds.), Sexual selection: Testing the alternatives, pp. 209–219. John Wiley & Sons Limited, Chichester.
Trivers, R. L. 1972. Parental investment and sexual selection. In B. Campbell (ed.), Sexual selection and the descent of man, pp. 136–179. Aldine, Chicago.
Turner, G. F. 1993. Teleost mating behaviour. In T. J. Pitcher (ed.), Behaviour of teleost fishes, fish and fisheries, pp. 306–331. Chapman & Hall, London.
Unger, L. M., and R. C. Sargent. 1988. Allopaternal care in the fathead minnow, Pimephales promelas: Females prefer males with eggs. Behav. Ecol. Sociobiol, 23:27-32.[CrossRef][Web of Science]
Wade, M. J. 1979. Sexual selection and variance in reproductive success. Am. Nat, 114:742-746.[CrossRef][Web of Science]
Wade, M. J., and S. J. Arnold. 1980. The intensity of sexual selection in relation to male behaviour, female choice, and sperm precedence. Anim. Behav, 28:446-461.[CrossRef]
Wade, M. J., and S. M. Shuster. 2002. The evolution of parental care in the context of sexual selection: A critical reassessment of parental investment theory. Am. Nat, 160:285-292.[CrossRef][Web of Science][Medline]
Wade, M. J., and S. M. Shuster. 2004. Sexual selection: Harem size and the variance in male reproductive success. Am. Nat, 164:E83-E89.[CrossRef][Web of Science][Medline]
Webster, M. S., S. Pruett-Jones, D. F. Westneat, and S. J. Arnold. 1995. Measuring the effects of pairing success, extra-pair copulations and mate quality on the opportunity for sexual selection. Evolution, 49:1147-1157.[CrossRef][Web of Science]
Wilson, A. B., I. Ahnesjö, A. C. J. Vincent, and A. Meyer. 2003. The dynamics of male brooding, mating patterns, and sex roles in pipefishes and seahorses (family Syngnathidae). Evolution, 57:1374-1386.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||