Behavioral Ecology Vol. 13 No. 3: 419-426
© 2002 International Society for Behavioral Ecology
Body size and sex allocation in simultaneously hermaphroditic animals
a Section of Ecology, Behavior, and Evolution, Division of Biology, University of California, San Diego, CA 92093-0116, USA b Laboratory of Ornithology, Cornell University, Ithaca, NY 14850, USA c Department of Biology, University of New Mexico, Albuquerque, NM 87131-1091, USA
Address correspondence to L. Angeloni, who is now at the Department of Zoology, Birge Hall, University of Wisconsin-Madison, Madison, WI 53706, USA. E-mail: langeloni{at}facstaff.wisc.edu .
Received 1 January 2001; revised 26 July 2001; accepted 19 August 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONDISCUSSIONREFERENCES |
|---|
Sex allocation theory predicts that hermaphrodites optimally divide resources between male and female function, and this strategy may vary with a wide range of individual and population traits. We present a model of reproductive strategies for simultaneously hermaphroditic animals that incorporates multiple sex allocation tactics depending on the traits of an individual, its current mate, and the population at large. We examined the effect of resource variation on sex allocation in a sperm-storing population with two body-size classes. We found that an individual's sex allocation depends on its own size relative to its mate; when stored sperm is displaced exponentially, small animals (with fewer resources) invest a greater proportion of resources in male function than do large animals, and animals of either size invest more in male function when mating with a large mate than with a small mate. Optimal sex allocation depends on the size distribution in the population, the disparity of resources between size classes, the cost of filling a sperm storage organ, and the shape of the sperm displacement function. A function with S-shaped returns to sperm transfer results in a contrary finding: large animals invest more resources in male function than small animals when the cost of filling a sperm storage organ is high. Under many conditions, pronounced sex allocation differences between individuals result in similar proportions of sperm displaced and paternity gained. Thus, variation in individual traits within a population leads to multiple sex-allocation strategies that can produce equal outcomes in the game of sperm competition.
Key words: body size, mating strategies, resources, sex allocation, simultaneous hermaphrodites, sperm competition.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONDISCUSSIONREFERENCES |
|---|
Optimal partitioning of limited resources among life-history components is a fundamental determinant of life-time reproductive success and may require an assessment of internal and environmental conditions (Charlesworth, 1980
;
Maynard Smith, 1982;
Stearns, 1976). For
simultaneous hermaphrodites, the division of reproductive resources into male
and female function is an important strategy for optimizing fitness
(Charnov, 1982). To partition
resources efficiently, simultaneous hermaphrodites should adjust the number of
sperm transferred and eggs produced depending on relative traits of the sperm
donor and sperm recipient. A few empirical studies have shown that
hermaphrodites do flexibly adjust mating strategies depending on relative
behaviors or traits of the current mate
(DeWitt, 1996;
Peters and Michiels, 1996;
Tomiyama, 1996;
Vreys and Michiels, 1998;
Yusa, 1996). Recent models of
sex allocation have begun to elucidate the theoretical basis for
hermaphroditic mating strategies (e.g.,
Charnov, 1996;
Greeff and Michiels, 1999;
Pen and Weissing, 1999;
Petersen and Fischer, 1996).
The present study builds on previous models and is the first to examine how
individual variation in a trait, reproductive resource, affects sex allocation
in sperm-storing hermaphrodites.
Simultaneous hermaphroditism is widespread throughout the animal kingdom
and is the primary mode of reproduction among several invertebrate groups
including flatworms, arrow worms, sea slugs, land snails, barnacles,
earthworms, and leeches (Ghiselin,
1969; Michiels,
1998). Many simultaneous hermaphrodites have internal
fertilization and long-term sperm storage in an organ that can hold a limited
amount of sperm (Baur, 1998;
Michiels, 1998). Therefore,
they can evolve sperm competition strategies similar to those studied in
dioecious species (Baur, 1998;
Charnov, 1996;
Michiels, 1998). Assuming a
trade-off between resources (or energy) invested in sperm and eggs, any amount
of resource devoted to sperm competition reduces that available for female
function. Hence, there should be an optimal strategy for reproductive
investment that is a product of the combined forces of sex allocation and
sperm competition (Charnov,
1996). An individual's optimal strategy is likely to depend on the
total amount of reproductive resources available to it and how that compares
to the resources of potential mates.
In an earlier model, Charnov
(1996) provided the first
predictions of sex allocation for sperm-storing simultaneous hermaphrodites
that mate multiply. When there are diminishing returns to sperm transfer in
species with limited sperm-holding capacity, less than half of the
reproductive resources for one mating event should be invested in male
function. An increase in the reproductive resources available to individuals,
an increase in the efficiency of converting resources into sperm, or a
decrease in the size of the sperm storage organ all result in greater
potential to displace sperm, which in turn leads to selection for reduced
investment in male function in a population. Therefore, the sex allocation
strategy in a given population is highly dependent on the average amount of
resource available for reproduction relative to the average sperm-holding
capacity (Charnov, 1996).
Individual variation in traits and circumstances may lead to the adoption
of diverse strategies within populations rather than a single common strategy.
Actual mating strategies do vary between individuals within populations (e.g.,
Angeloni and Bradbury, 1999;
DeWitt, 1996;
Otsuka et al., 1980;
Petersen, 1995;
Petersen and Fischer, 1996;
Schärer et al., 2001;
St. Mary, 1994;
Tomiyama, 1996;
van Duivenboden and ter Maat,
1985; Yusa, 1996),
and this variation is not currently explained by models that predict a single
population strategy. In this study we considered the importance of resource
variation within a population and its effect on individual sex allocation. In
natural populations individual resources can vary with body size. Many of the
invertebrate hermaphrodites listed above exhibit indeterminate growth and/or
asynchronous breeding or settlement. This creates natural populations of
sexually mature individuals with considerable heterogeneity in body size
(e.g., Angeloni et al., 1999;
Audesirk, 1979;
Carefoot, 1987;
Crozier, 1918;
Trowbridge, 1993;
Vreys and Michiels, 1997).
Body size affects sex allocation of sequential hermaphrodites through the
timing of sex change (Ghiselin,
1969; Warner,
1975), and there is growing evidence that body size affects
reproductive strategies of simultaneous hermaphrodites (e.g.,
Angeloni and Bradbury, 1999;
Charnov, 1987;
Crozier, 1918;
DeWitt, 1996;
Otsuka et al., 1980;
Peters and Michiels, 1996;
Petersen, 1995;
Petersen and Fischer, 1996;
Schärer et al., 2001;
St. Mary, 1994;
Tomiyama, 1996;
Vreys and Michiels, 1997;
Yusa, 1996). Body size can
reflect an individual's total resource level, which is likely to affect two
key determinants of sex allocation: the energy available for gamete production
and the energy put into the construction of the sperm storage organ, and thus
the number of sperm it can hold. Because both of these factors are likely to
vary within populations and increase with body size, there is a clear need to
demonstrate the effects of size variation on sex allocation strategies in
sperm-storing simultaneous hermaphrodites. We offer a model of a population of
individuals with two resource levels which, for simplicity, we will refer to
as size classes (small and large) as a first step toward understanding the
effect of within-population variation on sex allocation.
Adding body size to Charnov's model
Charnov's (1996) model
describes a population where R units of resource are available to
each individual for reproduction per mating event, and r is the
proportion of R allocated to sperm production, leaving (1 -
r) for egg production. The constant c converts resource into
sperm number so that the resources allocated to male function, rR,
are used to produce and transfer rRc sperm to a mate. As sperm flows
into the sperm stores, a constant mixing and flushing of new and old sperm
results in exponential sperm displacement
(Parker and Simmons, 1994).
Upon completion of sperm transfer, the donor's sperm now make up a proportion,
= 1 - exp (-r
), of the total number of sperm (µ) that
can be stored (Figure 1a);
= Rc/µ and is the amount of sperm that can be produced (if
all resource were invested in sperm) relative to the amount of sperm that can
be stored.
|
For simplicity we retain many of the assumptions of Charnov's
(1996) model: animals mate
frequently at regular intervals with reciprocal insemination and follow each
mating event with the production of eggs. A sperm donor fertilizes a
proportion of those eggs based on the proportion of its own sperm within the
storage organ; thus, we assume that there is no sperm layering or selective
sperm use. The resources expended in one mating event in no way affect the
amount of resource available for the next mating event. There is no mortality
and there are no virgins. The assumption that we have not retained is that all
animals are identical. Other studies have extended Charnov's
(1996) model to explore the
effects of different assumptions (see Discussion;
Greeff and Michiels, 1999;
Pen and Weissing, 1999).
In Charnov's (1996) model a
rare mutant allocating 
of its
resources to sperm, in a population of individuals that allocate r to
sperm, gains fitness through female function proportional to (1 -
) R. The mutant gains
fitness through male function from the mate's first egg clutch and reduced
paternity in subsequent clutches as other nonmutants displace the mutant's
sperm. This male fitness is proportional to an infinite series, (1 -
r) (R) [ (R)]
{1 + [1 - (rR)] + [1 - (rR)]2 +...}
which sums to:
 |
Because the proportionality constant is the same for male and female
function, total fitness gained by the mutant in one mating event is simply the
addition of this equation to that for female function, (1 -
) R.
The evolutionarily stable allocation (r*) was found by
analytically solving for the strategy that maximizes fitness, preventing a
rare mutant from invading the population with a different strategy
(Maynard Smith, 1982);
r* decreased from 0.5 with an increase in the ratio of
potential sperm production to the amount of sperm that can be stored (
= cR/µ) (Charnov,
1996; Figure
1b).
We extend the model by allowing for two resource levels:
RL for large animals with more resources and
RS for small animals with fewer resources. Large animals
occur at frequency P in the population, and small animals occur at
frequency (1 - P). We allow an individual to adjust its sex
allocation depending on its own resources and those of its mate, so there are
four sperm allocation strategies, one for each of the following possible
situations: (1) a large animal mating with another large animal
(rLL); (2) a large animal mating with a small animal
(rLS); (3) a small animal mating with a large animal
(rSL); and (4) a small animal mating with another small
animal (rSS). There are then four corresponding sperm
displacement values that are functions of the sperm allocation of the donor:
(rLL), (rLS),
(rSL), and (rSS). We first use
an exponential sperm displacement function so that the proportion of sperm in
a small recipient's organ displaced by a large donor is:
 |
 |
As before, the coefficient c converts resources into number of
sperm produced. We have added coefficient d to convert resources into
number of sperm stored, effectively allowing for variation in the size of the
sperm storage organ. We combine c/d into one variable, m,
which represents the conversion of resources to sperm production relative to
sperm storage. For most of this study we have made the resource level of large
animals twice the resource level of small animals
(RL/RS = 2). In this case and when m =
1, a large animal putting all of its resources into sperm production and none
into egg production could hypothetically produce enough sperm to fill a small
mate's storage organ two times (LS =
mRL/RS = 2), while a small animal could only
produce enough sperm to fill half of a large mate's storage organ
(SL = mRS/RL = 0.5).
The variable m is essentially the number of times an individual could
fill a sperm storage organ the size of its own if it were investing all of its
resources in sperm. Thus, in a population where m is low, it is more
costly to fill a sperm storage organ (in terms of resource expenditure) than
in a population where m is high.
Fitness through female function of a large mutant that allocates
to sperm when mating with a small
nonmutant is proportional to:
 |
LS). We add to this
function all of the paternity gained from eggs produced by the nonmutant after
subsequent mating events which displace a fraction of the mutant donor's
sperm. The paternity gained from those clutches depends on the strategy and
size (large or small) of subsequent mates, when P is the probability
of mating with a large animal, and (1 — P) is the probability
of mating with a small animal. Thus, fitness through male function of a large
mutant from a mating event with a small non-mutant is proportional to:
 |
 |
Finding optimal sex allocation required solving for the four strategies that simultaneously maximized fitness equations. As the equations were too complex to solve analytically, we used numerical methods. We calculated optimal r* values for hypothetical populations that varied in the following population parameters: the frequency of large animals (P = 0.1, 0.25, 0.5, 0.75, 0.9), the ratio of resources of a large animal to a small animal (RL/RS = 1.2, 2, 5), and the relative conversion of resources to sperm production over sperm storage (m = 0.1-50). Because the trajectories of the r* values did not change significantly beyond m = 20, we only show the data up to that point. See Table 1 for all combinations of parameters explored.
|
|
and 
are
respectively calculated as
.
|
|
|
|
|
).
Note the similarity between these results and those in
For each set of parameters, we explored the range of possible strategies (from 0 to 1). We began by dividing the potential population and mutant rvalues into 10 increments and found the four mutant
values that provided maximum fitness
for each combination of population r values. The set of mutant
values that were least deviant from
corresponding population rvalues were then the four optimal sex
allocation strategies for that iteration
(
).
For each subsequent iteration, the range explored for each r was
reduced by half, centered around the r* from the prior
iteration, and divided into 10 increments. The four optimal strategies were
calculated again, and this process was repeated until the
r* values were calculated to the nearest 0.005.
Results
The body sizes of an individual and its current mate affect its sex
allocation. We found the highest optimal allocation to sperm in small animals
mating with large animals (
) and the lowest
allocation to sperm in large animals mating with small animals
(
). Individuals paired with same-size mates
invest more evenly in sperm versus eggs
(
). Body
size has the greatest effect on sex allocation when m is low; with an
increasing m, sex allocation does vary with body size, but all four
strategies begin to converge on a relatively low investment in sperm. For
example, Figure 2a shows
optimal strategies for populations of varying m values and a given
size distribution (RL/RS = 2; P =
0.5). When a large and small animal mate with each other and m 
2, the small animal invests all of its resources in sperm and the large animal
invests all of its resources in eggs. The strategies move away from those
values as m increases (i.e., it pays for the small animal to also
produce eggs and the large animal to inseminate the small animal;
Figure 2a). When two small
animals mate, the outcome is similar to that of Charnov's
(1996) model with only one
size class; they invest slightly less than 50% of resources for the mating
event in sperm when m is low, with reduced levels of male allocation
as m increases (Figure
2a). When two large animals mate, they invest 70% of resources in
sperm at a low m, with lower levels of male allocation as m
increases. At m 
2, the strategy of two mating large animals is
the same as two mating small animals. When m is high, the cost of
filling a sperm storage organ is so low that no animal ever invests more than
half of its reproductive resources in male function, and most resources are
used for egg production (Figure
2a).
Although body size affects sex allocation strategies, it does not have a
pronounced effect on the proportion of sperm displaced or the paternity
achieved in the eggs produced by the mate
(Figure 2b). Three of the
scenarios (small mating with large, large mating with large, small mating with
small) result in almost equal displacements that increase with m. A
large animal mating with a small one transfers no sperm and thus achieves no
fitness through male function when m 2
(Figure 2b).
Over an infinite number of matings, large animals invest more energy in female function than in male function (Figure 2c). Small animals can have either male- or female-biased allocations depending on m, but they always invest a greater proportion of resources in sperm than large animals (Figure 2c).
Changing the proportion of large and small animals in the population from
equality does not have a pronounced effect on strategies; three
(
)
remain virtually unchanged. The only effect is on
in a population with m < 2; when
mating with each other, large animals invest more in sperm if they are rare
than if they are common (Figure
3).
Increasing the disparity in resources (RL/RS
= 5) results in a greater effect of body size on allocation strategies for
many potential populations (Figure
4). When a large and small animal mate, the former produces only
eggs and the latter produces only sperm over a wider range of potential
population m values. A greater disparity in resources results in a
slightly higher when m < 6.
There is no effect on 
. Decreasing the
resource disparity (RL/RS = 1.2) affects
optimal strategies in the opposite direction: they converge more rapidly with
increasing m.
Sperm stores remain constant with body size
Sperm storage organs may not always increase with body size in
hermaphroditic populations. To consider a case in which sperm stores remain
constant across both body sizes, we modified the model so that an increase in
reproductive resources corresponds to an increase in potential sperm
production, but not in sperm storage ability. For example,
LS (potential sperm production by a large animal
divided by sperm stores of a small animal) is now represented by
cRL/µ and is equivalent to
LL.
Results
Body size still affects sex allocation when all individuals have equal
sperm stores, and strategies are similar to those previously described for
variable sperm storage. For example, Figure
5 shows strategies for populations of varying c/µ
values and a given size distribution (RL/RS =
2; P = 0.5). When compared with
Figure 2a, it is clear that the
optimal allocations are similar at low m and low c/µ
values (when it is more costly to fill a sperm storage organ). The only
exception is for , which is slightly reduced
when sperm stores are constant. When c/µ increases,
drops below levels seen before and remains
lower than the value for ; both
and move
away from their respective 0 and 1 values more rapidly and at lower
c/µ values when sperm stores are constant. The proportion of sperm
displaced, and therefore the proportion of paternity gained in the egg clutch
produced by the mate, is greater when mating with a large animal
(LL and SL) than with a
small animal (LS and
SS).
S-shaped sperm displacement
Some hermaphroditic populations may not have exponential sperm
displacement, with fitness returns diminishing over the entire range of sperm
production. There may be cases where sperm cooperate, or a minimal investment
in sperm is needed before any fitness is gained, so that returns to sperm
production increase initially (Pen and
Weissing, 1999). To investigate this possibility, we modified the
model by changing the sperm displacement function to one that is S-shaped,
with initially increasing returns to sperm investment (following
Pen and Weissing, 1999;
Figure 6). When a large animal
inseminates a small animal the proportion of sperm displaced now corresponds
to
 |
|
The form and terms (e.g., the use of the number 6) of this equation were chosen because they generate a sigmoid curve within the range of possible sperm displacement values.
Results
An S-shaped sperm displacement function complicates sex allocation
strategies. For example, Figure
7a shows strategies for populations of varying m values
and a given size distribution (RL/RS = 2;
P = 0.5). At a low m, when a large and small animal mate,
the roles are reversed from the case of exponential sperm displacement, with
the large animal investing all of its reproductive resources in sperm and the
small animal investing everything in eggs. If the cost of filling a sperm
storage organ is reduced (m > 2), the roles revert to those
previously described: The large animal invests only in eggs and the small
animal invests only in sperm, but moving away from those values at m
> 5. Two small animals should not inseminate each other unless m
> 1 (Figure 7a). All four
r* values have an intermediate peak with an increasing
m before they ultimately decrease, similar to the shape of the
allocation curve predicted with only one size class
(Figure 6b;
Pen and Weissing, 1999).
Figure 7b shows the predicted
average allocations for large and small animals over an infinite number of
mating events.
Constant strategies for large and small animals regardless of mate
size
Some hermaphrodites may not be capable of flexibly adjusting sex allocation
strategies with each mating event. Instead of shifting allocation with the
size of the current mate, an individual would continuously allocate resources
in the same way dependent only on its body size and population parameters. To
predict optimal strategies under these conditions, we modified the model so
that only two strategies are possible, one for large animals
(
)
and one for small animals
(
).
Results
In a population of individuals that are unable to adjust sperm and egg
investment with the size of the current mate, the optimal strategies are
similar to average strategies when allocation is flexible. For example,
Figure 8 shows sex allocation
for populations with varying m values, exponential sperm
displacement, and a given size distribution
(RL/RS = 2; P = 0.5). Note the
similarities between Figure 8
and Figure 2c. Small animals
have either male- or female-biased allocations depending on m. Large
animals always invest more in eggs than in sperm and invest a greater
proportion of resources in female function than do small animals. The
proportion of sperm displaced varies with the size of the sperm recipient, not
the size of the sperm donor. A greater proportion of sperm is displaced, and
paternity is gained, when inseminating a small animal than a large animal.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONDISCUSSIONREFERENCES |
|---|
Our model is the first to show that variation in individual traits can lead to a diversity of mating strategies within populations of sperm-storing simultaneous hermaphrodites. Previous models have predicted a single strategy for an entire population (Charnov, 1996
; Greeff and Michiels,
1999; Leonard,
1990; Pen and Weissing,
1999), ignoring differences between individuals. The body size of
an individual, the size of its current mate, and the size distribution of the
population all influence its sex allocation strategy. Body size has its
greatest effect on sex allocation when the resource disparity between size
classes is high, when sperm stores vary with body size, and when filling a
sperm storage organ is costly (i.e., m is low). The exact
relationship between body size and sex allocation depends on the form of sperm
displacement. For species with exponential sperm displacement (where returns to sperm transfer diminish with increasing investment), our model predicts that large animals allocate more energy to female function than do small animals. If an individual can adjust its investment strategy with each mating event, it should increase male allocation when mating with a large animal as compared with its optimal level when mating with a small animal. Strategies are highly dependent on the cost (in terms of resource expenditure) of filling a sperm storage organ in the population. For example, if filling a sperm storage organ is costly, a small animal mating with a large animal should put all of its resources into displacing sperm; this is because the small animal never reaches the point on the sperm displacement curve where returns to investment in sperm are so diminished that it pays to invest in eggs. This strategy of investing nothing in eggs drives the large animal to invest nothing in sperm when mating with a small mate, as it is not worthwhile to inseminate a small mate that will not produce eggs after the mating event. With a reduced cost of filling a sperm storage organ, it eventually becomes profitable for a small animal to divert some energy to egg production when mating with a large animal and for a large animal to invest a minimal amount of resource to inseminate a small mate. In general, as the cost of filling a sperm storage organ is reduced, the optimal investment in sperm is also reduced. If the cost of filling a sperm storage organ is extremely low, sex allocation strategies become less dependent on body size, and all animals converge on low sperm investment because fewer resources are needed to approach 100% paternity.
Support for the prediction that large animals invest more in female
function than small animals can be found in the few studies on sex allocation
in sperm-storing simultaneous hermaphrodites
(Angeloni and Bradbury, 1999;
DeWitt, 1996;
Otsuka et al., 1980;
Yusa, 1996). These studies
used the amount of time spent mating in each sexual role as the currency for
sex allocation. Although difficult to measure, estimates of egg and sperm
production would provide more accurate tests of the prediction that sex
allocation varies with body size.
Our model suggests that reciprocal mating (where both members of a mating
pair transfer sperm) should occur between animals of similar size, whereas
unilateral mating events (where only one individual transfers sperm) are more
likely between animals very different in size. However, when the cost of
filling a sperm storage organ is reduced, reciprocal mating should be
ubiquitous. Reciprocity is widespread in real hermaphroditic populations and
has been suggested as a way of resolving conflict between two individuals who
prefer to mate in the same sexual role
(Fischer, 1988;
Leonard, 1991;
Michiels, 1998).
Distinguishing between reciprocity due to a mutual willingness to transfer
sperm (predicted by this model) and reciprocity due to conflict over mating
roles has proven difficult empirically
(Michiels, 1998). Many
hermaphrodites also mate unilaterally
(Leonard, 1991;
Michiels, 1998), but
predictions about the relationship between this trait and body size remain
untested. Although our model with exponential sperm displacement predicts
unilateral mating events between individuals of different sizes, this is not
to be confused with the evolution of separate sexes; these animals also mate
reciprocally with same-sized mates and thus are still hermaphrodites.
A population made up almost entirely of large animals or entirely of small
animals approximates Charnov's
(1996) model with only one
size class; the strategy of the common type when mating with each other
similarly decreases from 50% allocation with decreasing cost of filling a
sperm storage organ. When large animals are rare, they invest greater amounts
in sperm when they encounter each other. This can be explained by the high
levels of sperm competition they face in the receptacle of a large animal; the
competition is greater because the population is dominated by small animals
that specialize in sperm production when mating with large animals.
A reduced resource disparity between large and small animals and a constant size of sperm stores both slightly reduce the effect of body size on sex allocation over a wide range of population conditions. However, the overall effect of body size is consistent, suggesting that our findings may be applicable to real populations with mating pairs that vary in size and resource disparity and with different allometric relationships between body size and internal organs. Constant sperm stores result in a reduced investment in sperm between mating large animals because they are less difficult to fill up (more like small animals).
The sperm displacement function can affect sex allocation dramatically. In
a population with an S-shaped displacement curve, returns to sperm transfer
are initially increasing and then eventually diminishing as the sperm storage
organ is filled. This kind of displacement is likely to occur in species that
have sperm cooperation or that require an initial minimum investment in sperm
or semen before any paternity is gained
(Pen and Weissing, 1999). In
this case small animals are females, investing only in eggs when it is
difficult to transfer enough sperm to take advantage of the steepest segment
of the fitness-gain curve. Conversely, large animals (which are capable of
producing the amount of sperm needed for optimal returns to sperm transfer)
should invest everything in male function when mating with small animals that
are only investing in female function. Determining the sperm displacement
function of a hermaphrodite is crucial for predicting sex allocation. This
function has not been measured in real hermaphrodites, but could be
accomplished in controlled experiments to estimate paternity after varying
amounts of sperm transfer. At this time it is unclear whether an S-shaped
function, an exponential function, or some other function best approximates
sperm displacement in real hermaphrodites.
Our model shows that an individual's sex allocation strategy can depend on
its own size as well as the size of its mate. Because an individual's own size
relative to the population may be constant over a period of time that spans
several mating events, sex allocation traits that are less flexible (e.g.,
structural and gonadal allocation to male vs. female function) should reflect
its average optimal allocation [for large animals:
; for small
animals: 
].
The sizes of an individual's mates over that same time interval are likely to
vary; hence, more flexible sex allocation traits (e.g., sperm and egg
production for a particular mating event) should reflect the relative sizes of
both the individual and its current mate
().
In species that are not capable of estimating the body size of a mate or do
not have the flexibility to adjust egg and sperm production for each mating
event, an individual might instead invest in a constant sex allocation
strategy dependent only on its own body size
(
). This
strategy is similar to the average allocation over several mating events of an
animal that is capable of making adjustments. It remains to be determined
whether hermaphroditic animals are capable of making size assessments when
mating. Courtship periods that involve tactile, visual, or chemical contact
may facilitate this kind of size assessment (e.g.,
Vreys and Michiels, 1997), and
if there is a mechanism available, selection should bring it to fixation.
Although this model predicts that sex allocation varies dramatically with
body size, under many conditions this leads to sperm displacement and
paternity patterns that are invariant with body size. Individuals that begin
with different total resource levels adjust the proportion of those resources
invested in sperm, evening out sperm competition abilities. This prediction
may prove important to biologists currently working on measuring sperm
precedence patterns in hermaphrodites, particularly those who have not found
an effect of body size on paternity (Baur,
1994).
A useful extension of this model would allow for individual strategies to
change dynamically over time as total resource levels change, either due to
growth or resource expenditure in previous mating events. Two other studies
have modeled different extensions of Charnov's
(1996) model. As confirmed in
this study, Pen and Weissing
(1999) showed that alternate
sperm displacement functions have dramatic effects on sex allocation. They
also found that an effect of clutch size on offspring survival results in any
number of different outcomes depending on the shape of the egg survival
function. A study by Greeff and Michiels
(1999) demonstrated that
reducing the number of lifetime matings from infinity reduces male allocation,
while sperm digestion increases male allocation and can lead to conditional
reciprocity with higher levels of sperm donation serving as a nuptial gift.
Ideally, a complex model would combine all of these extensions of Charnov's
original formulation to predict sex allocation in a population with a
continuous body size distribution.
Many studies have shown that sex allocation depends on body size and/or
mating group size across a range of animal and plant taxa
(Angeloni and Bradbury, 1999;
Charnov, 1980,
1982,
1987;
DeWitt, 1996;
Klinkhamer et al., 1997;
Otsuka et al., 1980;
Petersen and Fischer, 1996;
Raimondi and Martin, 1991;
Schärer et al., 2001;
Trouvé et al., 1999;
Yusa, 1996). Sex allocation
shifts toward female function with an increase in body size in organisms very
different from the model we describe: barnacles that lack long-term sperm
storage (Charnov, 1980,
1987), fish that lack internal
fertilization (Petersen and Fischer,
1996), and plants (Klinkhamer
et al., 1997). These common trends clearly result from different
specific mechanisms not included in this model, yet all are similar in that
they can be explained by the shape of fitness gain curves
(Charnov, 1982). For example,
in animal-pollinated plants, male gain curves may decelerate due to pollinator
saturation of pollen or high levels of local mate competition and geitonogamy
(pollination from a flower on the same plant), resulting in increased
allocation to female function with size and resource level
(Klinkhamer et al., 1997). In
contrast, increasing fitness returns to male investment in wind-pollinated
plants contributes to greater male allocation with size
(Klinkhamer et al., 1997). Our
findings thus add to a growing understanding of the importance of fitness-gain
curves in determining sex allocation of simultaneous hermaphrodites
(Charnov, 1982).
Similarities between dioecious species, sequential hermaphrodites, and
simultaneous hermaphrodites are proving greater than differences in many ways.
Our study emphasizes the importance of fitness-gain curves for simultaneous
hermaphrodites, and they play an equally important role in the timing of sex
change in sequential hermaphrodites and offspring sex ratio of dioecious
animals (Charnov, 1982). Body
size affects sex allocation of both simultaneous and sequential hermaphrodites
(Ghiselin, 1969;
Warner, 1975) and is an
important factor in the sexually selected mating strategies of dioecious
animals (Andersson, 1994).
Although sexual selection research has focused on dioecious species,
hermaphrodites face similar selection pressures, and both groups must
optimally allocate resources to sperm competition
(Charnov, 1996;
Galvani and Johnstone, 1998;
Michiels, 1998). Future work
on hermaphrodites will likely reveal a similar diversity of sexually selected
strategies to those of animals with separate sexes, and perhaps insight can be
gained by considering the parallels between these groups.
There have been several calls for additional work on hermaphroditic mating
systems (Jarne and Charlesworth,
1996; Leonard,
1991; Michiels,
1998), and despite growing research in this field, there are still
few empirical studies of sex allocation in sperm-storing hermaphrodites and no
measures of the parameters that go into the models. Our model emphasizes the
importance of measuring body size and resource distributions within
populations, the shape of sperm displacement curves, and the energetic cost of
filling sperm-storage organs. Although measuring sperm displacement and its
costs present a challenge, these estimates are needed to accurately predict
sex allocation strategies within populations.
|
ACKNOWLEDGEMENTS |
|---|
We thank R.S. Burton, K.R. Crooks, N.D. Holland, J.R. Kohn, T. Price, D.F. Westneat, and an anonymous reviewer for useful discussion and comments on the manuscript. This research was supported by funds from a Dissertation Improvement Grant from the National Science Foundation to L.A. and J.W.B. (IBN-9902307). During preparation of the manuscript L.A. was supported by a Ford Foundation Dissertation Fellowship.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONDISCUSSIONREFERENCES |
|---|
Andersson MB, 1994. Sexual selection. Princeton, New Jersey: Princeton University Press.
Angeloni L, Bradbury J, 1999. Body size influences mating strategies in a simultaneously hermaphroditic sea slug, Aplysia vaccaria. Ethol Ecol Evol 11: 187-195.
Angeloni L, Bradbury J, Chaine A, 1999. Growth, seasonality, and dispersion of a population of Aplysia vaccaria Winkler, 1955. Veliger 42: 1-9.
Audesirk TE, 1979. A field study of growth and
reproduction in Aplysia californica. Biol Bull
157: 407-421.
Baur B, 1994. Multiple paternity and individual variation in sperm precedence in the simultaneously hermaphroditic land snail Arianta arbustorum. Behav Ecol Sociobiol 35: 413-421.[Web of Science]
Baur B, 1998. Sperm competition in molluscs. In: Sperm competition and sexual selection (Birkhead TR, Møller AP, eds). London: Academic Press; 255-305.
Carefoot TH, 1987. Aplysia: its biology and ecology. Oceanogr Mar Biol Annu Rev 25: 167-284.
Charlesworth B, 1980. Evolution in age-structured populations. Cambridge: Cambridge University Press.
Charnov EL, 1980. Sex allocation and local mate competition in barnacles. Mar Biol Lett 1: 269-272.
Charnov EL, 1982. The theory of sex allocation. Princeton, New Jersey: Princeton University Press.
Charnov EL, 1987. Sexuality and hermaphroditism in barnacles: a natural selection approach. In: Barnacle biology (Southward AJ, ed). Rotterdam: Balkema; 89-103.
Charnov EL, 1996. Sperm competition and sex allocation in simultaneous hermaphrodites. Evol Ecol 10: 457-462.
Crozier WJ, 1918. Assortative mating in a nudibranch, Chromodoris zebra Heilprin. J Exp Biol 27: 247-292.
DeWitt TJ, 1996. Gender contests in a simultaneous hermaphrodite snail: a size advantage model for behaviour. Anim Behav 51: 345-351.
Fischer EA, 1988. Simultaneous hermaphroditism, tit-for-tat, and the evolutionary stability of social systems. Ethol Sociobiol 9: 119-136.
Galvani A, Johnstone R, 1998. Sperm allocation in an uncertain world. Behav Ecol Sociobiol 44: 161-168.[Web of Science]
Ghiselin MT, 1969. The evolution of hermaphroditism among animals. Q Rev Biol 44: 189-208.[Medline]
Greeff JM, Michiels NK, 1999. Sperm digestion and reciprocal sperm transfer can drive hermaphroditic sex allocation to equality. Am Nat 153: 421-430.[Web of Science]
Jarne P, Charlesworth D, 1996. Hermes meets Aphrodite: an animal perspective. Trends Ecol Evol 11: 105-107.
Klinkhamer PGL, de Jong TJ, Metz H, 1997. Sex and size in cosexual plants. Trends Ecol Evol 12: 260-265.[Web of Science]
Leonard JL, 1990. The hermaphrodite's dilemma. J Theor Biol 147: 361-372.[Web of Science][Medline]
Leonard JL, 1991. Sexual conflict and the mating systems of simultaneously hermaphroditic gastropods. Am Malacol Bull 9: 45-58.
Maynard Smith JM, 1982. Evolution and the theory of games. Cambridge: Cambridge University Press.
Michiels NK, 1998. Mating conflicts and sperm competition in simultaneous hermaphrodites. In: Sperm competition and sexual selection (Birkhead TR, Møller AP, eds). London: Academic Press; 219-254.
Otsuka C, Rouger Y, Tobach E, 1980. A possible relationship between size and reproductive behavior in a population of Aplysia punctata Cuvier, 1803. Veliger 23: 159-162.
Parker GA, Simmons LW, 1994. Evolution of phenotypic optima and copula duration in dungflies. Nature 370: 53-56.
Pen I, Weissing FJ, 1999. Sperm competition and sex allocation in simultaneous hermaphrodites: a new look at Charnov's invariance principle. Evol Ecol Res 1: 517-525.
Peters A, Michiels NK, 1996. Do simultaneous hermaphrodites choose their mates? Effects of body size in a planarian flatworm. Freshwater Biol 36: 623-630.
Petersen CW, 1995. Reproductive behavior, egg trading, and correlates of male mating success in the simultaneous hermaphrodite, Serranus tabacarius. Environ Biol Fishes 43: 351-361.
Petersen CW, Fischer EA, 1996. Intraspecific variation in sex allocation in a simultaneous hermaphrodite: the effect of individual size. Evolution 50: 636-645.[Web of Science]
Raimondi PT, Martin JE, 1991. Evidence that mating group size affects allocation of reproductive resources in a simultaneous hermaphrodite. Am Nat 138: 1206-1217.[Web of Science]
Schärer L, Karlsson LM, Christen M, Wedekind C, 2001. Size-dependent sex allocation in a simultaneous hermaphrodite parasite. J Evol Biol 14: 55-67.
Stearns SC, 1976. Life-history tactics: a review of the ideas. Q Rev Biol 51: 3-47.[Medline]
St. Mary CM, 1994. Sex allocation in a simultaneous
hermaphrodite, the blue-banded goby (Lythrypnus dalli): the effects
of body size and behavioral gender and the consequences for reproduction.
Behav Ecol 5:
304-313.
Tomiyama K, 1996. Mate-choice criteria in a
protandrous simultaneously hermaphroditic land snail Achatina fulica
(Férrusac) (Stylommatophora: Achatinidae). J Molluscan
Stud 62:
101-111.
Trouvé S, Jourdane J, Renaud F, Durand P, Morand S, 1999. Adaptive sex allocation in a simultaneous hermaphrodite. Evolution 53: 1599-1604.[Web of Science]
Trowbridge CD, 1993. Population structure of two common species of ascoglossan (= sacoglossan) opisthbranchs on the central coast of Oregon, USA. Veliger 36: 99-106.
van Duivenboden YA, ter Maat A, 1985. Masculinity and receptivity in the hermaphrodite pond snail, Lymnaea stagnalis. Anim Behav 33: 885-891.
Vreys C, Michiels NK, 1997. Flatworms flatten to size
up each other. Proc R Soc Lond B 264:
1559-1564.
Vreys C, Michiels NK, 1998. Sperm trading by volume in a hermaphroditic flatworm with mutual penis intromission. Anim Behav 56: 777-785.[Web of Science][Medline]
Warner RR, 1975. The adaptive significance of sequential hermaphroditism in animals. Am Nat 109: 61-82.[Web of Science]
Yusa Y, 1996. The effects of body size on mating
features in a field population of the hermaphroditic sea hare Aplysia
kurodai Baba 1937 (Gastropoda: Opisthobranchia). J Molluscan
Stud 62:
381-386.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Janicke and L. Scharer Sex allocation predicts mating rate in a simultaneous hermaphrodite Proc R Soc B, December 7, 2009; 276(1676): 4247 - 4253. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. M. Hermann, B. Genereux, and W. C. Wildering Evidence for age-dependent mating strategies in the simultaneous hermaphrodite snail, Lymnaea stagnalis (L.) J. Exp. Biol., October 1, 2009; 212(19): 3164 - 3173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. N. Ludwig and P. J. Walsh Multiple Mating, Sperm Storage, and Mating Preference in Aplysia californica Biol. Bull., December 1, 2008; 215(3): 265 - 271. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. G. Norton, A. F. Johnson, and R. L. Mueller Relative size influences gender role in the freshwater hermaphroditic snail, Helisoma trivolvis Behav. Ecol., November 1, 2008; 19(6): 1122 - 1127. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. KOSNIK, D. JABLONSKI, R. LOCKWOOD, and P. M. NOVACK-GOTTSHALL QUANTIFYING MOLLUSCAN BODY SIZE IN EVOLUTIONARY AND ECOLOGICAL ANALYSES: MAXIMIZING THE RETURN ON DATA-COLLECTION EFFORTS Palaios, December 1, 2006; 21(6): 588 - 597. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. G. NORTON and J. M. BRONSON THE RELATIONSHIP OF BODY SIZE AND GROWTH TO EGG PRODUCTION IN THE HERMAPHRODITIC FRESHWATER SNAIL, HELISOMA TRIVOLVIS J. Mollus. Stud., May 1, 2006; 72(2): 143 - 147. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Anthes and N. K. Michiels Do "sperm trading" simultaneous hermaphrodites always trade sperm? Behav. Ecol., January 1, 2005; 16(1): 188 - 195. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Ohbayashi-Hodoki, F. Ishihama, and M. Shimada Body size-dependent gender role in a simultaneous hermaphrodite freshwater snail, Physa acuta Behav. Ecol., November 1, 2004; 15(6): 976 - 981. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Angeloni, J. W. Bradbury, and R. S. Burton Multiple mating, paternity, and body size in a simultaneous hermaphrodite, Aplysia californica Behav. Ecol., July 1, 2003; 14(4): 554 - 560. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||