Biol. Bull.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Proestou, D. A.
Right arrow Articles by Twombly, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Proestou, D. A.
Right arrow Articles by Twombly, S.
Related Collections
Right arrow Ecology
Right arrow Molluscs
Right arrow Population Biology
Right arrow Reproduction
Biol. Bull. 214: 194-202. (April 2008)
© 2008 Marine Biological Laboratory

Patterns of Male Reproductive Success in Crepidula fornicata Provide New Insight for Sex Allocation and Optimal Sex Change

Dina A. Proestou1,*, Marian R. Goldsmith and Saran Twombly2

University of Rhode Island, Department of Biological Sciences, Kingston, Rhode Island 02881

* To whom correspondence should be addressed. E-mail: dpro1791{at}postoffice.uri.edu


   Abstract
TOP
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Literature Cited
 
The size-advantage model and sex-allocation theory are frequently invoked to explain the evolution and maintenance of sequential hermaphroditism in many taxa. A test of current theory requires quantitative estimates of reproductive success and knowledge of the relationship between reproduction and size for each gender. Reproductive success can be difficult to measure. In species where polyandry occurs, it can be quantified only by determining paternity of offspring. We employed microsatellite loci to establish paternity for 12 families of Crepidula fornicata, where a family is defined as a single female, her brood, and the males stacked on top of her. Genetic data were analyzed and paternity was assigned to a single potential father for more than 83% of the offspring tested. Estimates of reproductive success revealed that one male within the family fathered the majority of offspring and that he was usually the largest male and the one closest to the brooding female. The dominant male's success also tended to decrease as the number of mature males within the family increased. Our results suggest that sperm competition could be a driving force in determining male reproductive success and the timing of sex change in C. fornicata.


   Introduction
TOP
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Literature Cited
 
Sequential hermaphroditism is a rare reproductive strategy, yet it has evolved independently in several taxa (Ghiselin, 1969; Warner, 1975; Charnov, 1982; Collin, 1995). The size-advantage hypothesis proposes that sex change should occur if an individual produces more offspring as one sex when small and as the opposite sex when large (Ghiselin, 1969; Warner, 1975; Iwasa, 1991). Coupled with sex-allocation theory, the size-advantage hypothesis has provided a framework for studying sex change in several teleost fishes (Hoffman et al., 1985; Cowen, 1990; Rogers, 2003; Munoz and Warner, 2004), shrimp (Charnov, 1982; Charnov and Skúladóttir, 2000; Chiba et al., 2003), annelids (Berglund, 1991), and molluscs (Wright, 1988; Collin, 1995; Chen and Soong, 2000). Sex-allocation theory analyzes the relationship between the amount of resources allocated to male and female function and male and female fitness to explain the occurrence of sequential hermaphroditism and the direction and timing of the change (Charnov, 1979, 1982; Campbell, 1998).

A number of models have been developed in an attempt to determine the time (size) of sex change in sequential hermaphrodites (Charnov, 1982; Iwasa, 1991; Collin, 1995; Charnov and Skúladóttir, 2000; Munoz and Warner, 2003). Most imply that the life-history transition is an evolutionary stable strategy (ESS) and predict that the optimal time (size) of sex change yields maximum lifetime reproductive success. Charnov's original ESS model suggests that the optimal time of sex change (t*) depends on the existing sex ratio in the population and the relative fitness of each sex at that time. It predicts a critical age (size) above which all individuals should be the second sex and below which all individuals should be the first sex; however, it assumes a constant environment and similar growth and mortality rates for both sexes (Charnov, 1982; Iwasa, 1991). In some cases, these assumptions may not hold true. Iwasa (1991) was able to show that differential growth and mortality rates, as well as costs associated with changing sex, also influence the time of sex change.

Moreover, in many species of fish and molluscs, the timing of sex change is labile and appears to be determined, at least in part, by social interactions (Hoagland, 1978; Warner, 1988; Warner and Swearer, 1991; Warner et al., 1996). Social interactions can mediate the reproductive gain of individuals and confound the predictions of traditional theory. For example, Munoz and Warner (2003, 2004) observed that, upon removal of the dominant male, the largest female in a harem of the protogynous (female first) bucktooth parrotfish, Sparisoma radians, deferred sex change to smaller females when her reproductive output was greater than the combined output of all other females in the mating group (size-fecundity skew) and elevated competition intensity from bachelor males prevailed.

The key to evaluating existing theory on the evolution and timing of sex change in sequential hermaphrodites is quantifying reproductive success. Female reproductive success is usually reported as the number of eggs produced, while male reproductive success is represented by the number of eggs fertilized (Charnov, 1982). Because many of the sex-changing species studied thus far are polygamous and a single spawn or brood can be fertilized by more than one male, measuring male reproductive success can be difficult. One way to estimate male reproductive success is to use genetic markers to analyze paternity of a group of offspring from a known set of potential parents and determine the percentage of offspring fathered by specific males. By assessing an individual male's reproductive success, our current understanding of the timing of sex change with respect to sex allocation could be dramatically enhanced.

Crepidula fornicata (Linneaus, 1758), the Atlantic slipper snail, is an ideal species in which to study the evolution and timing of sex change and to evaluate existing theory. It is a protandrous (male first) sequential hermaphrodite, and females brood eggs that are internally fertilized by males that stack on top of them (Coe, 1936). Planktonic larvae are released and, at the end of the larval period, settle on hard substrates in response to a water-soluble chemical secreted by adults (Hoagland, 1978; McGee and Targett, 1989). Maturation from juvenile to male and the maintenance of the male state is promoted by pheromones released by females (Feral, 1978; LeGall and Streiff, 1978). In the absence of such cues, juveniles will pass through an accelerated male phase and become female to minimize the amount of time spent in reproductive isolation (Coe, 1938; Warner et al., 1996).

Although Ghiselin's size-advantage model (Ghiselin, 1969) is considered to be the best explanation for sex change in the genus Crepidula (Collin, 1995), substantial overlap between male and female size suggests that size alone cannot be the trigger (Coe, 1936, 1938). Hoagland (1978) and Warner et al. (1996) agree that the timing of sex change is labile and is greatly influenced by social interactions. In the adult form, C. fornicata is sedentary and forms stacks to mate. Usually, a few large females are on the bottom and several smaller males stack on top. The presence of more than one male in a stack results in multiple paternity (Gaffney and McGee, 1992), and the ability of more than one male to fertilize the eggs of a single brood leads to sperm competition (Birkhead, 1999). Sperm competition, the competition of ejaculates from more than one male for the fertilization of eggs (Parker, 1984), could effectively reduce the reproductive success of males in a stack and may cause individuals to change sex earlier than they would in the absence of competition.

Despite a long history of research on sequential hermaphroditism in the genus Crepidula, only one study has considered sex change in the context of sex-allocation theory. Collin (1995) observed that sex change occurs only in the bottom-most male in a stack and developed a series of models to predict the optimal size at sex change. First, she modeled a scenario in which male reproductive success is independent of size and assumes that all males in a stack share fertilizations equally. She also modeled a situation in which male reproductive success scales with size. Both models underestimated the size at which sex change occurs in natural populations, indicating that they do not account for some factor or factors influencing the process.

A large individual should possess a reproductive advantage when it functions as a female because fecundity is assumed to scale with size, the number of fertilizations a male acquires is limited by the number of eggs produced, and in most cases, males share fertilizations within a brood. What is unclear is whether large males (which are closest to the female) have a reproductive advantage over small males and whether the advantage diminishes as the number of males competing for fertilizations increases. The overall objective of this study was to establish patterns of reproductive success and determine whether individual male reproductive success is affected by size, position, and the intensity of sperm competition within a stack of C. fornicata. We used microsatellite markers to genotype offspring and parents, and paternity analysis to distinguish among potential fathers for offspring from mating stacks that consisted of a range of sex ratios. Assigning paternity allowed us to quantify reproductive success for each male sampled and to determine how fertilizations were divided among males.


   Materials and Methods
TOP
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Literature Cited
 
Study site and sampling strategy
Stacks of C. fornicata were collected from Saunderstown, Rhode Island, at the peak of the reproductive season. The number of individuals in each stack sampled varied from two to six. Sex ratios (male:female) represented among the stacks collected were 1:1, 2:1, 3:1, 4:1, 5:1, 3:2, and 4:2. These sex ratios are typical of the stacks observed at several Rhode Island sites, where most stacks consist of three or four individuals (Proestou, unpubl. data). At least two stacks from each sex ratio category were collected, and each stack met the following criteria: (1) stacks were collected from the shallow subtidal zone, (2) selected stacks were found at least 0.5 m from each other, (3) females within the stack were brooding embryos, and (4) the stack was the only one present on a substrate. In stacks containing more than one male, the female in the stack was 32 ± 1 mm, and males varied in size. This sampling strategy was devised in an attempt to avoid the insemination of females by males from neighboring stacks and also to minimize any effects of female size on brood size and male reproductive success.

Before performing paternity analysis, we determined the gender-specific relationships between reproductive success and body size in the absence of competition, using the stacks containing two individuals. The length of the shell was used to measure size. We determined the effect of female size on brood size by examining females of various sizes (17–42 mm) that were mated with a single male of uniform size (~22 mm), and we analyzed the effect of male shell length on brood size by examining males of various sizes (11–32 mm) that were mated with females of uniform size (~30 mm). We assumed that all embryos brooded by the female were fertilized by the male stacked on top of her. C. fornicata broods are packaged in capsules containing 50 to 450 embryos, and each brood contained between 15 and 70 capsules (Proestou, unpubl. data). The total number of embryos brooded by females was estimated by counting the number of capsules within the brood and the total number of embryos in 10 capsules and then multiplying the average number of embryos per capsule by the total number of capsules. The relationship we detected between female size and brood size corresponded well with that of a previous study in which brood sizes were inferred from dry weight measurements (Collin, 1995), thereby validating our quantification method.

All other stacks were dismantled, and we recorded the position, size (in mm), and sex of each individual. Sex was determined by examining gross morphology, and an individual with a penis longer than the right tentacle was classified as a male (Coe, 1936). A piece of foot tissue from each adult was removed and stored in a 1.5-ml microcentrifuge tube at –80 °C. The masses of embryos were also removed from beneath the females, and 20 to 30 embryos that had reached 400 µm in size were isolated from each and stored individually at –80 °C in 0.5-ml microcentrifuge tubes. Because it is not known whether embryos within a capsule are fertilized by a single male or a number of males, we sampled two to three embryos from 10 different egg capsules to ensure an accurate measure of individual male reproductive success.

We used 10 stacks and 12 families in the paternity analyses. A family consisted of a single female and all of the males stacked on top of her. When a stack contained more than one female, it also contained more than one family.

DNA extraction, PCR amplification, and analysis of experimental samples
Adult tissue samples were pulverized on dry ice with a mortar and pestle, and DNA was extracted according to a method specifically designed for molluscs (Winnepenninckx et al., 1993). The extraction of DNA from embryos involved adding 10 µl of an extraction buffer consisting of 50 mmol l–1 KCl, 10 mmol l–1 Tris-HCl (pH 8.3), 2.5 mmol l–1 MgCl2, 0.01% gelatin, 0.9% Tween 20, and 10 mg/ml Proteinase K to each microcentrifuge tube containing an embryo. DNA was then extracted by incubating tubes for 1 h at 65 °C, followed by an incubation at 94 °C for 15 min (Simpson et al., 1999).

DNA extracted from adults and embryos of each family served as the source of nuclear DNA for PCR amplification. Five polymorphic microsatellite loci—Cf8, Cf34, Cf10a, Cf13a, and Cf20a—were utilized in this study. All were quite variable, with 22 to 45 alleles per locus and observed heterozygosities ranging from 0.333 to 0.841. Additional information regarding these loci, including GenBank accession numbers, repeat motifs, primer sequences, and optimal amplification conditions, can be found in Proestou (2006). At least three microsatellite loci were amplified for each family. Amplification products were separated by electrophoresis in 6% polyacrylamide sequencing gels and were visualized with the Silver Sequence DNA Sequencing System (Promega). Gels were then scored to reveal the genotype of each individual at each locus.

Paternity analysis
Paternity analyses were conducted for stacks containing three or more individuals by treating each brood of offspring as a population. We determined the paternity of an offspring in a given family using two methods: exclusion and maximum likelihood estimation. For each method, paternity was assigned by including only the males present within a stack, and all males sampled, as candidate fathers. Considering all males sampled in the analyses allowed us to assess the ability of males from outside the mating stack to achieve paternity.

Exclusion is conceptually simple and uses genotypic incompatibilities between the males and offspring in a population to exclude some males from the pool of potential fathers. If all but one male is excluded, then paternity can be assigned with 100% confidence. Exclusion is most effective when only a few males are considered as potential parents and the maternal parent is known (Meagher, 1986). The limitation of the exclusion method is that it is often not possible to exclude all potential fathers due to overlapping alleles among fathers, genotyping errors, null alleles, and mutations (Jones and Arden, 2003).

When exclusion does not reveal paternity with 100% confidence, other methods can be employed. The maximum likelihood method, derived by Thompson (1976) and applied by Meagher (1986), considers the probability of the offspring genotype given the known maternal and alleged paternal genotypes to distinguish the most-likely father. Likelihood of paternity by an alleged father can be meaningful only if it is compared to the likelihood of paternity by a random male. The two likelihoods can be written as a likelihood odds ratio:

Formula
where go, gm, and ga are the genotypes of the offspring, mother, and alleged father, respectively. For detailed information on how the probabilities are calculated, see Kalinowski et al. (2007). The natural logarithm of this likelihood odds ratio is the LOD score, which serves as a measure of the likelihood that a given male has fathered a specific offspring. The LOD scores of all candidate fathers are compared, and the male with the highest LOD score is assigned paternity.

We used the computer program CERVUS ver. 3.0 (Marshall et al., 1998; Kalinowski et al., 2007) to assign paternity using the maximum likelihood method and to determine the level of statistical confidence for each paternity assignment. CERVUS calculates a statistic delta ({Delta}), which discriminates between non-excluded males by subtracting the LOD score of the second most-likely male from the LOD score of the most-likely male. Prior to performing the paternity analysis, CERVUS performs a simulation to assess the significance of {Delta} values. Parameters used in the simulation were (1) the total number of candidate males, (2) the proportion of candidate males sampled, and (3) the rate of typing error. Critical {Delta} values were established for both 80% and 95% confidence levels. If the {Delta} scores obtained by each father/offspring combination did not exceed the critical {Delta} for either 80% or 95% confidence, CERVUS did not assign paternity.

Statistical analyses
Relationships between male reproductive success and size, in the absence and presence of competition, were determined using linear regression analysis. The effect of male position within a stack on male reproductive success was also examined with a one-way ANOVA and ANCOVA with size as a covariate. All statistical analyses were performed using the SAS system (ver. 8.1).


   Results
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Literature Cited
 
Reproductive success and shell length
When male and female reproductive success was evaluated in the absence of competition, the number of embryos brooded scaled with shell length when females were paired with single males of uniform size. A linear regression of the data (number of eggs = –1967.144 + 650.356 x shell length) was statistically significant (P < 0.0001), with a coefficient of determination of 0.617 (n = 96). In contrast, there was no significant relationship between the number of eggs assumed to have been fertilized and male size when males of various sizes were paired with females of uniform size (r2 = 0.098, P = 0.179, n = 20). The data demonstrate that in the absence of multiple males, reproductive success as a male is determined primarily by brood size (Fig. 1).


Figure 1
View larger version (13K):
[in this window]
[in a new window]

  		Figure 1. Relationship between body size and reproductive success for (A) female and (B) male Crepidula fornicata. Female reproductive success was determined as the number of embryos brooded by individual females. Male reproductive success was quantified as the number of embryos brooded by the female with which he was paired.

 
Paternity assignment
A total of 316 offspring from 12 families were examined to determine paternity. Of the methods used, exclusion, when only males from the mating stack were included as candidate fathers in the analysis, was able to resolve the greatest number of paternity tests (Table 1). There was an overall decrease in paternity assignment when the exclusion method was applied to data sets containing all males sampled as candidate fathers; however, the magnitude of the decrease varied considerably by family. The number of paternity assignments made with at least 80% confidence via the likelihood method did not change when different candidate pools were considered (Table 1), but again, differences in paternity assignments with the two sets of potential fathers were revealed when each family was considered separately.


View this table:
[in this window]
[in a new window]

  		Table 1 Outcome of paternity analysis with exclusion and likelihood methods when only males within a stack and all males sampled were included as candidate fathers

 
Including all sampled males as candidate fathers did not significantly alter the outcome of paternity analysis. Paternity was never assigned to a sampled male outside the stack with the exclusion method, and only six such assignments were made with the likelihood method. These results suggest that, for the families examined in this study, the overwhelming majority of fertilizations occur within the mating stack. We proceeded to use the outcome of the exclusion and likelihood analyses in which all males were considered as potential fathers to establish consensus paternity assignments and calculate conservative estimates of male reproductive success.

Although paternity was assigned to sampled males outside of the mating stack less than 2% of the time, our analysis revealed that unsampled males also fathered offspring. "Extra-stack" fertilizations were detected in one of the families tested (family 52) because the genotypes of some offspring included alleles absent in the mother and candidate fathers at all three loci. This result is not surprising as female Crepidula are capable of storing sperm (Hoagland, 1978) and small male Crepidula have previously been reported to "rove’"from stack to stack (Gaffney and McGee, 1992).

General patterns of male reproductive success
We quantified reproductive success by calculating the percentage of offspring per brood that was assigned to each male. All 12 families were included in these calculations even though two families contained multiple females. The presence of more than one female/brood per family does have the potential to confound reproductive success estimates; however, our estimates did not change when the two families were removed from the analysis, primarily because there was minimal overlap in paternity assignments among the broods (47A & B, 53A & B) sharing the same pool of fathers (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2 Reproductive gain by each male in each family

 
A single male was assigned paternity unequivocally in 5 out of 12 families (26, 24, 45, 46, and 47A); however, there were a proportion of tests for which additional candidate males could not be excluded and paternity remained unassigned (Table 2). Within these families, the lone father fertilized at least 74% of the embryos genotyped, and he was always the largest male in the stack and positioned directly above the mother. When more than one male fathered the offspring in a brood (families 22, 51, 52, 48, 47B, 53A, and 53B), paternity was not shared equally, and the most successful male was not necessarily the largest or the closest to the female (Table 2).

In half of the families considered (22, 45, 52, 46, 47B, 53B), more than 15% of the offspring could not be assigned to single fathers with 80% confidence (Table 2). To assess the potential impact of inconclusive paternity tests on estimates of male reproductive success, the required confidence level for paternity assignment was reduced from 80% to 65%. Paternity assignments made with 65% confidence agreed with our conservative estimates of reproductive success. For example, of the 13 offspring to which a father could not be assigned with 80% confidence in family 22, 8 were assigned to male 22-2, 3 were assigned to male 22-1, and only 2 could not be assigned a father with 65% confidence.

To investigate the influence of absolute size on reproductive success when multiple males are present in a mating stack, we regressed reproductive success against shell length for all males sampled. The relationship was significant (RS = –59.55 + 3.25 x shell length, P = 0.00013) but explained only about 30% of the variation in reproductive gain (Fig. 2). We observed a great deal of size overlap between males that did (20.9 –34.6 mm) and did not (13.5–30.6 mm) acquire fertilizations. Taken together, these results indicate that additional factors play a role in determining reproductive success.


Figure 2
View larger version (7K):
[in this window]
[in a new window]

  		Figure 2. Relationship between absolute male size (shell length) and reproductive success in the presence of competition. Reproductive success was quantified as the percentage of offspring per brood fathered by each male. Symbols indicate the position of a male within a stack.

 
Male position is one such factor. Within a mating stack, the oldest (largest) males are closest to the female and the youngest (smallest) individuals are furthest away; therefore, male position within a stack is roughly equivalent to relative size. Significant differences in mean reproductive success with respect to position were detected with a one-way ANOVA, F4, 39 = 7.68, P = 0.0001. Overall, male reproductive success decreased as relative size decreased and the distance from the female increased (Fig. 3). The effect of position remained significant even after the data were adjusted for differences among positions due to variation in size (ANCOVA, F4, 38 = 5.12, P < 0.005).


Figure 3
View larger version (6K):
[in this window]
[in a new window]

  		Figure 3. Relationship between reproductive success and male position within the mating stack. Each bar represents the average percentage of offspring fathered by males occupying a position. Position 1 is the position closest to the female and position 5 is the position farthest from the female.

 
There is a tendency for the average reproductive success of the bottom male in a stack to decrease as the number of males per stack increases. When two males were present in a stack, the bottom male fertilized 70% of the embryos; when four or more males competed for fertilizations, the percentage of embryos fertilized by the bottom male decreased to 47%. An analysis of variance failed to detect a significant effect of competition intensity. This is likely a consequence of small sample size and high variability within groups.

Finally, reproductive success estimates revealed that females store and use sperm to fertilize eggs during subsequent reproductive seasons. This was most evident in family 47A, which consisted of a mother (47-6), an additional female (47-5), and 4 males (47-4, 47-3, 47-2, and 47-1). Seventeen out of 20 embryos brooded by female 47-6 were fathered by 47-5, who was simultaneously brooding her own embryos. Although 47-5 had fathered the brood of 47-6, it was evident that she did not fertilize her own brood. The offspring of female 47-5 were fertilized by males 47-2, 47-3, and 47-4 (Table 3).


   Discussion
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Literature Cited
 
Questions surrounding the factors influencing the timing of sex change in Crepidula have been tested since the early 1900s. Through a series of laboratory experiments, Coe (1936, 1938) demonstrated that when males are isolated, sex change occurs regardless of size. In addition, he was able to show that when two to three males are cultured in the absence of a female, only the largest male changes sex. Data collected from laboratory and field studies indicate that the duration of the male phase in C. fornicata and the Eastern white slipper snail C. plana is strongly influenced by the presence of individuals in the female phase (Gould, 1952; Hoagland, 1978). Warner et al. (1996) argue that the timing of sex change in the shelf limpet C. norrisiarum is regulated by the composition of mating groups (sex ratio) and the relative size of males within a mating group. These studies have all contributed to a general understanding of sex change in the genus.

Our data reinforce the idea that absolute male size is not an important trigger for sex change despite the predictions of the size-advantage hypothesis. Male reproductive success did not vary with size when different-sized males were paired with females of approximately the same size (Fig. 1B). We assumed that all embryos brooded by the females were fertilized by the attached male; however, we have subsequently shown that sperm storage and fertilization by males outside the mating stack do occur. Therefore, the result may be biased due to undocumented competitive effects. Moreover, when multiple males are present in a mating stack, relative rather than absolute size accounts for a majority of the variation in male reproductive success in C. fornicata. A male's position relative to the female (which is equivalent to relative size) determines his reproductive success, and the effect of position remains even after adjusting for differences in absolute size.

In addition, our data show that reproductive success through male function does not cease upon sex change. When multiple females were present in a stack, we (and others) observed that at least some of the embryos brooded by the bottom-most female were fertilized with stored sperm produced by the second female when she was still functioning as a male (Gaffney and McGee, 1992).

Gaffney and McGee (1992) also documented fertilization of embryos by males outside the mating stack in 1 out of 6 "substrate groups." The paternity assignments of the 12 Rhode Island families examined in this study accord with their finding, but indicate that the frequency of "extra stack" fertilizations is low when mating stacks are 0.5 m or more apart. Paternity of six offspring from different families was assigned to various sampled males from outside the mating stack, while the paternity of three offspring from family 52 was assigned to an unsampled male.

Obtaining quantitative estimates of male reproductive success in a variety of social contexts allowed us to highlight the potential role of sperm competition in influencing male reproductive output. Our results suggest a negative relationship between competition intensity and the reproductive success of the largest male in the stack, but the trend is weak and should be investigated further by examining both a broader range of sex ratios and a larger sample.

The general patterns of reproductive success we have reported stimulate new questions regarding the mechanisms driving sex change in C. fornicata. For example, do we see a higher average reproductive success in large males because they release more ejaculate and dampen the contribution of smaller males? Similarly, is easy access to the female seminal receptacle responsible for the reproductive dominance exhibited by males in position 1, or is it the ability to block insemination by other males? Even though large males dominate fertilizations within a stack, there is a point at which remaining a male is no longer beneficial. We have provided a preliminary indication that this could occur if competition among males leads to increased effort with diminishing returns. Thus, within the context of sex-allocation theory, sex change should occur when the negative effect of competition from other males outweighs the positive effects of being a large male. In addition, we have shown that lifetime reproductive success is maximized not only by changing sex when that balance is met but also by inseminating females at the end of the brooding season, changing sex between seasons, and essentially functioning as both sexes in the following season. These questions, and those pertaining to how males sense competition from other males, merit additional study.

One logical way to build upon the current study is to manipulate stacks experimentally in the laboratory. "Virgin" females can be obtained by isolating males for 30–50 days (Coe, 1936), and males of defined size can be stacked on top of them. Several treatments can be established and replicated. An experimental approach would eliminate confounding factors associated with female mating history and allow us to better tease apart the effects of size, position, and competition on male reproductive success. Our initial attempts to examine paternity in C. fornicata focused on experimental manipulation of stacks, and they were fraught with problems. As C. fornicata grows, its shell conforms to the shape of the substrate. When separated from the original substrate, the morphology of the shell prevents reattachment to other available substrates. The ability to attach to alternative substrates (glass plates, conspecifics that were not the original substrate) has been evaluated and found to be low (Proestou, unpubl. data). As a result of the aforementioned difficulties, we sampled natural stacks collected from the field, choosing stacks carefully to control for absolute size and potential cross-fertilization.

Because C. fornicata is a sedentary species that forms relatively permanent mating stacks consisting of few females and several males, genotyping individuals with molecular markers and performing paternity analysis were critical tools for elucidating the patterns of male reproductive success. Orton (1950) and Hoagland (1978) both observed copulations between a female and several males within a stack, but conceded that copulations do not always translate into reproductive success. Gaffney and McGee (1992) recognized this and used allozyme markers to demonstrate multiple paternity in C. fornicata. Although they were able to show that offspring are sired largely, but not exclusively, by a single male in the majority of broods (only six were included in the analysis), they were unable to establish paternity definitively due to limited polymorphism in the markers used. The use of microsatellite rather than allozyme loci greatly enhanced our ability to assign paternity and quantify reproductive success.

The departure of allele frequencies from Hardy-Weinberg expectations at all microsatellite loci used in this study indicated the presence of null alleles (Jones and Arden, 2003). In spite of null alleles, we were able to assign paternity to more than 80% of the tested offspring because the markers were highly polymorphic, the maternal parent was always known and segregation of maternal alleles could be followed, and both the exclusion method and CERVUS were able to account for genotyping errors caused by null alleles. The high percentage of offspring without an assigned father in some families (22, 45, and 53) was primarily caused by shared alleles between the offspring and more than one candidate father.

In conclusion, despite our relatively small sample sizes and our inability to manipulate stack composition experimentally, we were able to elucidate previously unknown patterns of reproductive behavior in a species that has historically been a model for sex change research. The main findings of our analysis of reproductive allocation and gender switching are that reproductive success as a male increases with size and proximity to the female, individuals can achieve reproductive success as a male even after they have changed sex, and reproductive success of the largest male tends to decrease as the number of males in a stack increases. These results indicate that competition may be an important factor determining male reproductive success and the optimal time of sex change. Only after the effects of competition have been quantified can Ghiselin's size-advantage model (Ghiselin, 1969) be refined.


   Acknowledgments
 
This work was supported by a National Science Foundation doctoral dissertation improvement grant DEB–9902242 to S. Twombly, D. Proestou, and M. Goldsmith as well as a Sigma Xi student research grant to DP. We thank R. Grosberg, B. Cameron, R. Toonen, and A. Wilson for tips on microsatellites and assistance with polyacrylamide gels and silver staining techniques; J. Chandlee for use of laboratory space and equipment; and R. Grosberg, P. Yund, and three anonymous reviewers for valuable comments on the manuscript. Any opinions, findings, conclusions, or recommendations expressed are those of the authors and do not reflect the views of the National Science Foundation.


   Footnotes
 
Received 8 May 2007; accepted 31 October 2007.

1 Current address: University of Rhode Island, Department of Fisheries, Animal and Veterinary Science, 20A Woodward Hall, Kingston, RI 02881. Back

2 Current address: National Science Foundation, Division of Environmental Biology, Arlington, VA 22230. Back


   Literature Cited
TOP
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Literature Cited
 

    Berglund, A. 1991 To change or not to change sex: a comparison between two Ophryotrocha species (Polychaeta). Evol. Ecol. 5:128–135.
    Birkhead, T. R. 1999. Distinguished sperm in competition. Nature 400:406–407.[Medline]
    Campbell, D. R. 1998. Variation in lifetime male fitness in Ipomopsis aggregata: tests of Sex Allocation Theory. Am. Nat. 152:338–353.[Web of Science]
    Charnov, E. L. 1979. The genetical evolution of patterns of sexuality: Darwinian fitness. Am. Nat. 113:464–480.
    Charnov, E. L. 1982. The Theory of Sex Allocation. Princeton University Press, Princeton, NJ.
    Charnov, E. L., and U. Skúladóttir. 2000. Dimensionless invariants for the optimal size (age) of sex change. Evol. Ecol. Res. 2:1067–1071.
    Chen, M., and K. Soong. 2000. Sex change in the hat snail, Cayptraea morbida (Reeve) (Gastropoda: Calyptraeidae): an analysis of substratum, size, and reproductive characteristics. Veliger 43:210–217.
    Chiba, S., S. Goshima, and Y. Shinomiya. 2003. Male-male competition selects for delayed sex change in the protandrous shrimp Pandalus latirostris. Mar. Biol. 142:1153–1157.
    Coe, W. R. 1936. Sexual phases in Crepidula. J. Exp. Zool. 72:445–477.
    Coe, W. R. 1938. Conditions influencing change of sex in mollusks of the genus Crepidula. J. Exp. Zool. 77:401–423.[Web of Science]
    Collin, R. 1995. Sex, size, and position: a test of models predicting size at sex change in the protandrous gastropod Crepidula fornicata. Am. Nat. 146:815–831.[Web of Science]
    Cowan, R. K. 1990. Sex change and life history patterns of the Labrid, Semicossyphus pulcher, across an environmental gradient. Copeia 3:787–795.
    Feral, C. 1978. Presence of morphogenic and retrogressive factors of the penis in a gonochoristic prosobranch mollusc Ocenebra erinacea-L. C. R. Hebd. Seances Acad. Sci. 287:1235–1237.
    Gaffney, P. M., and B. McGee. 1992. Multiple paternity in Crepidula fornicata (Linnaeus). Veliger 35:12–15.
    Ghiselin, M. T. 1969. The evolution of hermaphroditism among animals. Q. Rev. Biol. 44:189–208.[Medline]
    Gould, H. N. 1952. Studies on sex in the hermaphrodite mollusk Crepidula plana. 4. Internal and external factors influencing growth and sex development. J. Exp. Zool. 119:93–163.[Web of Science]
    Hoagland, K. E. 1978. Protandry and the evolution of environmentally-mediated sex change: a study of the Mollusca. Malacologia 17:365–391.[Web of Science]
    Hoffman, S. G., M. P. Schildhauer, and R. R. Warner. 1985. The costs of changing sex under contest competition for mates. Evolution 39:915–927.[Web of Science]
    Iwasa, Y. 1991. Sex change evolution and cost of reproduction. Behav. Ecol. 2:56–68.[Abstract/Free Full Text]
    Jones, A. G., and W. R. Arden. 2003. Methods of parentage analysis in natural populations. Mol. Ecol. 12:2511–2523.[Medline]
    Kalinowski, S. T., M. L. Taper, and T. C. Marshall. 2007. Revising how the computer program CERVUS accommodates genotyping error increases success in paternity assignment. Mol. Ecol. 16:1099–1106.[Medline]
    LeGall, S., and W. Streiff. 1978. Control of the morphogenetic pedal factor of the penis by cerebro pleural ganglia in Crepidula fornicata-Phil. (Protandre Hermaphrodite Mollusc). C. R. Hebd. Seances Acad. Sci. 287:1305–1307.
    Marshall, T. C., J. Slate, L. E. B. Kruuk, et al. 1998. Statistical confidence for likelihood-based paternity inference in natural populations. Mol. Ecol. 7:639–655.[Medline]
    McGee, B. L., and N. M. Targett. 1989. Larval habitat selection in Crepidula (L.) and its effect on adult distribution patterns. J. Exp. Mar. Biol. Ecol. 131:195–214.[Web of Science]
    Meagher, T. R. 1986. Analysis of paternity within a natural population of Chamaelirium luteum. 1. Identification of most-likely male parents. Am. Nat. 128:199–215.[Web of Science]
    Munoz, R. C., and R. R. Warner. 2003. A new version of the size-advantage hypothesis for sex change: incorporating sperm competition and size-fecundity skew. Am. Nat. 161:749–761.[Web of Science][Medline]
    Munoz, R. C., and R. R. 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. 15:129–135.[Abstract/Free Full Text]
    Orton, J. H. 1950. Recent breeding phenomena in the American slipper-limpet, Crepidula fornicata. Nature 169:279–280.
    Parker, G. A. 1984. Sperm competition and the evolution of animal mating strategies. Pp. 2–60 in Sperm Competition and the Evolution of Animal Mating Systems, R. L. Smith, ed. Academic Press, New York.
    Proestou, D. A. 2006. Isolation and characterization of microsatellite markers in the Atlantic slipper shell Crepidula fornicata for use in paternity analysis. Mol. Ecol. Notes 6:437–439.[Web of Science]
    Rogers, L. 2003. Odds-playing and the timing of sex change in uncertain environments: you bet your wrasse. Behav. Ecol. 14:447–450.[Free Full Text]
    Simpson, R. J., C. S. Wilding, and J. Grahame. 1999. Simple protocol for extracting nuclear DNA from single embryos of a marine snail. BioTechniques 26:1050–1052.[Web of Science][Medline]
    Thompson, E. A. 1976. Inference of genealogical structure. Soc. Sci. Inform. 15:477–526.
    Warner, R. R. 1975. The adaptive significance of sequential hermaphroditism in animals. Am. Nat. 109:61–82.[Web of Science]
    Warner, R. R. 1988. Sex change and the size-advantage model. Trends Ecol. Evol. 3:133–136.
    Warner, R. R., and S. E. Swearer. 1991. Social control of sex change in the bluehead wrasse, Thalassoma bifasciatum (Pisces: Labridae). Biol. Bull. 181:199–204.[Abstract]
    Warner, R. R., D. L. Fitch, and J. D. Standish. 1996. Social control of sex change in the shelf limpet, Crepidula norrisiarum: size specific responses to local group composition. J. Exp. Mar. Biol. Ecol. 204:155–167.[Web of Science]
    Winnepenninckx, B., T. Backeljau, and R. Wachter. 1993. Extraction of high molecular weight DNA from molluscs. Trends Genet. 9:407.[Web of Science][Medline]
    Wright, W. G. 1988. Sex change in the Mollusca. Trends Ecol. Evol. 3:137–140.



This article has been cited by other articles:


Home page
 J HeredHome page
S. Le Cam, J. A. Pechenik, M. Cagnon, and F. Viard
Fast versus Slow Larval Growth in an Invasive Marine Mollusc: Does Paternity Matter?
J. Hered., July 1, 2009; 100(4): 455 - 464.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Proestou, D. A.
Right arrow Articles by Twombly, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Proestou, D. A.
Right arrow Articles by Twombly, S.
Related Collections
Right arrow Ecology
Right arrow Molluscs
Right arrow Population Biology
Right arrow Reproduction

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS