Behavioral Ecology Vol. 12 No. 2: 237-245
© 2001 International Society for Behavioral Ecology
Components of fertilization success in the bluehead wrasse, Thalassoma bifasciatum
a College of the Atlantic, 105 Eden St., Bar Harbor, ME 04609, USA b Department of Ecology, Evolution, and Marine Biology, University of California, Santa Barbara, CA 93016, USA c Department of Biology, Eastern Michigan University, Ypsilanti MI 48197, USA d Department of Biology, University of Padova, Viale Trieste 75, 35121, Padova, Italy
Address correspondence to C.W. Petersen. E-mail: chrisp@ecology.coa.edu. D.Y. Shapiro is now at Pfizer Global Research and Development, 2800 Plymouth Road, Ann Arbor, MI 48105, USA. A. Marconato is now deceased.
Received 7 August 1999; revised 25 August 2000; accepted 9 September 2000.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
In many species of marine organisms, males and females release gametes directly into the water column. Although free-spawning marine invertebrates appear to have highly variable fertilization success, in tropical reef fishes the average fertilization success is quite high, typically over 90%; nevertheless, substantial variation has been reported, and fertilization has a direct effect on fitness. We investigated the factors affecting fertilization success in natural spawnings of the bluehead wrasse, Thalassoma bifasciatum. During a two-year study at a site in St. Croix, we found extensive and predictable variation in fertilization success in pair spawns of this reef fish. Fertilization success averaged 95%, but was affected by the amount of sperm released, the water velocity at a site, the mating success of the male, and the size of the female. As sperm released in a spawn increases, and as water velocity at a site decreases, sperm concentrations should remain higher in the vicinity of eggs for a longer period of time, and both of these factors are correlated with increasing fertilization success. The recent history of individuals with partners or sites did not affect the fertilization success of their spawn. In an evolutionary context, the real and predictable variance in fertilization success in this species may influence the mating choices of males and females. However, there is currently no evidence that females use differences in fertilization success among males or sites in their reproductive decisions.
Key words: bluehead wrasse, Coral reef fish, mate choice, fertilization, reproductive success, spawning, sperm limitation, Thalassoma bifasciatum.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The release of gametes into the water column with subsequent external fertilization is exhibited by a wide variety of marine organisms including algae, invertebrates and fish (Brawley and Ladd, 1992; Giese and Kanatani, 1987
;
Levitan, 1996a;
Levitan and Petersen, 1995).
Studies of fertilization success (the percentage of eggs fertilized) in these
species have indicated that fertilization success is typically less than 100%
with considerable intraspecific variation
(Coma and Lasker, 1997;
Kiflawi et al., 1998;
Lasker et al., 1996;
Lasker and Kapela, 1997;
Levitan, 1996a;
Levitan and Petersen, 1995;
Marconato et al., 1995,
1997;
Marconato and Shapiro, 1996;
Serrão et
al., 1996; Yund,
2000). Incomplete fertilization of a batch of eggs and variation
in fertilization success both have important implications for the ecology and
population biology of marine species
(Brazeau and Lasker, 1992;
Levitan, 1991;
Levitan et al., 1992;
Levitan and Petersen, 1995;
Pennington, 1985). In
addition, fertilization is a major determinant of fitness and thus has
evolutionary implications through selection on morphological, physiological,
and behavioral characteristics correlated with fertilization success
(Clifton, 1997;
Giambartolomei, 1997;
Knight, 1997; Levitan,
1996a,b,
1998;
Levitan and Petersen, 1995;
Pearson et al., 1998, 1992;
Shapiro et al., 1994;
Yund, 1998).
A wide range of factors has been shown to influence the fertilization
success of an individual spawn or fertilization event in free-spawning marine
species, including the number of sperm released in the spawn, the spawning
behavior of individuals, the physical environment where the spawning takes
place, and gamete characteristics (reviewed in
Levitan and Petersen, 1995;
see also Coma and Lasker, 1997;
Lasker and Kapela, 1997;
Levitan,
1996a,b,
1998; Marconato et al.,
1995,
1997;
Marconato and Shapiro, 1996).
Variation in fertilization success has direct effects on reproductive success
of individuals of each sex, and thus can influence a variety of reproductive
characteristics, including mate choice of both sexes, the spatial and temporal
patterns of mating, gamete allocation patterns among spawns, and gamete
characteristics (Levitan,
1993,
1996a,b,
1998;
Levitan and Petersen, 1995).
The effects of natural selection on fertilization success will depend on the
extent that individual traits can affect variation in fertilization success
among spawns, and the degree that fertilization success interacts with other
fitness components such as mating success or mate quality. It is possible that
fertilization success is relatively constant or at least unpredictable enough
among spawns to render it insignificant in current selection for reproductive
characteristics in marine fishes. Alternatively, although the spawning
behavior of marine fishes may result in a high average fertilization success
(typically greater than 90%: Kiflawi et
al., 1998; Marconato et al.,
1995,
1997;
Marconato and Shapiro, 1996;
Warner et al., 1995), there
may still be significant predictable variation among spawns that could result
in selection on male and female spawning behavior and gamete allocation.
This article reports the results of a study on sperm allocation and fertilization success in the bluehead wrasse, Thalasoma bifasciatum, a tropical reef fish. We examine the role of eight factors affecting fertilization success of natural spawns of the bluehead wrasse: number of sperm, water velocity, male mating success acting via an effect on sperm number or independent of sperm number, female and male size, and familiarity of either sex with the spawning partner or the spawning site. We used data from spawns in situ under both natural and manipulated conditions. Our goal is to provide a comprehensive review of the factors that do and do not affect fertilization success in pair spawns in this species, and to examine these results in light of theoretical predictions concerning fertilization success and mate choice.
Biology of Thalassoma bifasciatum
The bluehead wrasse is a common inhabitant of shallow coral reefs
throughout the Caribbean. Size and color are sexually dimorphic; females and
smaller males have a yellow and black initial-phase (IP) coloration, while
larger individuals are brightly colored terminal-phase (TP) males. Some of
these TP males are the result of sex change by females
(Warner et al., 1975).
Spawning occurs year-round during a limited period of approximately 2 h in
the afternoon (Warner and Robertson,
1978). In patch-reef environments, females typically migrate from
their upcurrent feeding areas to downcurrent segments of the reef to spawn.
Preferred spawning sites consist of a subset of upward projections on the
downcurrent periphery of the reef (Warner,
1988,
1990a). During the spawning
period these sites are either actively defended by single large TP males or
patrolled by large numbers of IP males.
A pair-spawn consists of a female and terminal-phase male both rushing
upward 0.3-1.5 m toward the surface and releasing gametes at the apex of a
spawning rush. Occasionally IP males join the pair just as they are spawning
and presumably release sperm in a behavior called streaking
(Warner and Robertson, 1978);
these spawns were rare at our field site and were excluded from the data
analysis.
Bluehead wrasse also engage in a behavior known as group-spawning, where a
female mates with typically 5-20 IP males in a spawning rush. When both group-
and pair-spawning occur on the same reef, they tend to occur at different
locations (Petersen et al.,
1992; Warner and Hoffman,
1980a; Warner et al.,
1975). A comparison of pair- and group-spawns from this research
site is reported elsewhere (Marconato et
al., 1997), and here we report only on pair-spawning.
A female spawns on 2 out of 3 days on average, and releases her daily egg
production in one, or rarely, two spawns
(Schultz and Warner, 1991;
Warner, Petersen, and Shapiro, unpublished data). The transparent eggs are
planktonic, approximately 600 µm in diameter, and positively buoyant once
fertilized. Hatching occurs after 20-24 h at 25-28°C. There is no parental
care.
Past work on the fertilization dynamics of T. bifasciatum has revealed the following patterns:
- More sperm released in a spawn results in higher fertilization
(Warner et al., 1995). The
curve that best fit the data is a Michaelis-Menton relation (proportion of
eggs fertilized = ([0.994xsperm]/[0.117 + sperm],
Figure 1). The effects of sperm
release were not statistically examined in combination with any other
independent variables (such as mating success, water velocity, see below).
- Sperm release differs among males
(Shapiro et al., 1994), and
males with high mating success release fewer sperm and achieve lower
fertilization success (Warner et al.,
1995). Group-spawns, in which there is approximately 10 times as
much sperm released as in pair-spawns
(Shapiro et al., 1994),
achieve equal or slightly higher fertilization success than pair spawns
(Marconato et al., 1997, but
see Petersen et al.,
1992).
- Males release more sperm when spawning with larger females
(Shapiro et al., 1994). The
effect of this pattern of sperm release on fertilization success was not
reported.
- Fertilization success is negatively correlated with a rank estimate of
water movement, with higher fertilization success occurring on days of less
water movement (Petersen et al.,
1992). However, this result is suspect for two reasons: first, the
technique used by Petersen et al.
(1992), who reported mean
fertilization rates of 75%, underestimated fertilization success by damaging
eggs, and it is unknown if this egg damage was correlated with water movement;
and second, overall water movement was estimated for the reef, not for
individual spawning sites.
|
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
General
Data were collected during May-September 1993-1994 at seven patch reefs within Tague Bay off the northeast coast of St. Croix, Virgin Islands (see Warner et al., 1995
for a
description of the site). To obtain data on mating success, observers
maintained focal observations on individual territorial TP males for the
entire daily reproductive period. Mating success data were taken for an
average of 5 days for each male (N = 37 males, range 2-15 days per
male).
Collection of eggs and sperm
To measure sperm output and fertilization success, we followed the methods
described in Shapiro et al.
(1994). Within a few seconds
of spawning, gametes were captured in large plastic bags of known volume
(approximately 50 L) by snorkelers. Estimates of the amount of sperm released
per spawn were obtained by collecting a sample of water (500 ml) from the bag,
adding several drops of rose bengal dye to stain the sperm, and preserving the
sample 20 min later with formalin. Samples were returned to the lab, filtered,
and stained sperm counted at 400x magnification within known areas at 20
locations on the filter paper. The total number of sperm released in the spawn
was calculated assuming all sperm from the spawn had been collected in the
bag.
Fertilization of eggs was estimated by collecting a second sample
(approximately 2-4 1) of water from the collection bag, staining eggs with
rose bengal dye 2-4 h after spawning and scoring them for development. We did
not use any sample in which fewer than 20 eggs were present. All eggs that
were not developing were counted as unfertilized, following the
recommendations of Marconato et al.
(1997).
We statistically removed the effects of sperm number on fertilization
success using both a parametric and a nonparametric technique. The first
method tested the residuals from the regression of sperm number on
fertilization success for effects of other independent variables in a
parametric statistical technique (multiple regression or ANCOVA). The minimum
sperm value used in this method was truncated at 1.5 million per spawn,
because previous work had shown an increasingly non-linear response and
increasing variability of sperm amount on fertilization success below this
point, which would violate the assumptions of the tests
(Warner et al., 1995). The
mean residual for a male was defined as the parametric fertilization
effectiveness (parametric FE), with a minimum of five residuals required
to include a male in an analysis. Positive values represent higher than
expected fertilization success for the amount of sperm released, while
negative values of parametric FE represent lower than expected fertilization
success for the amount of sperm released
(Figure 1).
Because it was impossible to entirely remove the heteroscedasticity of the data set, we also used a nonparametric technique to examine the effects of factors other than sperm number on fertilization success. This technique had the advantage of fewer assumptions and used a larger range of sperm values (it included spawns with less than 1.5 million sperm), but potentially had lower statistical power. For most analyses, both techniques were used and are presented in the results. In some cases sample sizes were substantially reduced using one method, either because a large number of spawns had less than 1.5 million sperm or because samples per group were too low to calculate an accurate estimate using nonparametric techniques. In these cases, only one method is used and reported.
The second technique used a variable called nonparametric fertilization effectiveness (nonparametric FE), which compared fertilization success for classes of spawns with similar sperm release. After initially subdividing the range of sperm-release values, the effect of higher sperm release on fertilization success was tested for each category; if sperm number had an effect on fertilization success using a median test, the category was divided again and tested until, within each category, there was no effect of increasing sperm number on fertilization success. This led to the formation of 14 categories, some of which spanned a relatively small range of sperm numbers where the fertilization curve was especially steep (Figure 1, lower line). Each spawn was scored as above, at, or below the median fertilization success for all spawns within a category of sperm-release. Due to small sample sizes, spawns with less than 0.25 x 106 sperm and those with more than 20.0 x 106 sperm were excluded from the analysis.
After classifying every spawn as below, equal, or above the median fertilization success for its category of sperm release, a nonparametric FE score, which expresses the proportion of spawns above the median fertilization success, can be recorded for each category of spawns (rare points on the median were counted as half above, half below). Nonparametric FE values can range from 0 to 1. A set of spawns with average fertilization would have a nonparametric FE of 0.5. A set of spawns with a value above 0.5 indicates higher than expected fertilization success based on sperm alone, and a value below 0.5 indicates lower than expected fertilization success. Individual males, females, or sites can be assigned a nonparametric FE value. Nonparametric FE was calculated for all males or sites with at least 10 spawns in which estimates of both sperm release and fertilization success were obtained; these values were then used as individual data points within an analysis.
Water velocity and water mixing
To estimate water movement during the spawning period at a spawning site,
in 1994 current meters (S4® InterOcean) were mounted in the water column
at a typical location of gamete release for 4 days. Meters were deployed
before spawning began and left until spawning was completed. The current
velocity, current direction, and depth of the current meter were measured
twice and averaged each s, for a min once every 15 min for the entire duration
of the spawning period (typically 1.5-2.5 h). From these data we calculated
mean current velocity for each spawning period and the variation in current
velocity and direction at a site. For the subset of these data that were
collected on days where we also collected data on fertilization, we examined
variation in FE among days within a site.
Characteristics of the male and female
All males in the study were captured and their standard length measured at
the beginning of observations. In 1992 and 1993 404 females were captured,
measured, and tagged with a unique sequence of colored beads using small tags
inserted into their dorsal musculature. Untagged females were estimated to lie
within three standard length size classes, small (< 55 mm), medium (55-65
mm), and large (> 65 mm). For the analysis of the effect of female size on
fertilization success, we tested for differences among size classes for both
estimates of FE. In addition, we calculated a fertilization curve (proportion
of eggs fertilized versus number of sperm released in a spawn) for each female
size class. Following Warner et al.
(1995), we fit a
Michaelis-Menton function using the NON-LIN analysis in Systat, assuming an
asymptotic fertilization success of 100% for all size classes.
Female and male bluehead wrasse show strong mating-site fidelity (Warner,
1985,
1986,
1987,
1990a), and increased site or
mate familiarity could lead to better coordination of the spawning pair and
increase fertilization success. To examine this idea, we classified males and
females as familiar with each other and with a site for the subset of spawns
where male and female identity were known. A mating pair was considered
familiar if they had mated at least two times in the past, or if no other
mating partners had been observed previously for the female. A site or male
was considered unfamiliar for the first 2 days an individual was observed to
spawn at a new site; afterward the site was considered familiar.
Manipulation of mating success of individual males
We removed at least 50% of the females on a reef on two reefs, reducing the
average male mating success by at least half. Spawns were collected before and
after the manipulation for four TP males.
The male mating success of TP males on a reef was increased by removing the majority of IP males on a reef, changing a previously successful group-spawning site to a pair-spawning site. Females that had previously spawned at these sites either continued to spawn at the same site with a newly established territorial TP male, or changed their usual spawning site to one of the pre-existing pair spawning sites. This manipulation was carried out on four reefs and spawns were collected before and after the manipulation for eight males.
Statistical analysis
For all parametric statistics, data were first checked for normality using
a Lilliefor's test and transformed when necessary. Transformations were
required for fertilization success (angular transformation), for water
velocity and sperm number per spawn (ln transform). Non-parametric statistics
follow Siegel and Castellan
(1988). Statistical analyses
were performed using the statistical software Systat version 7.0 for
Windows.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The mean fertilization success over all spawns (excluding experimental manipulations) was 95.3% (N = 810). Individual male fertilization success averaged 95% and ranged from 83-99.6% (untransformed data, N = 30 males, s = 3.8%, median 96.1%; each male datum based on 5-51 samples per male). There was a positive correlation between the average number of sperm released by a male and his mean fertilization success (Spearman rank correlation coefficient, rs =.61, N = 30 males, p <.001).
Fertilization success was positively correlated with the number of sperm
released per spawn using the subset of data in the parametric analysis
(F1,529 = 10.3, N = 531, p =.001). In
addition, both the site of spawning (F24,499 = 2.8,
N = 526, p <.001) and the size of the female
(F2,499 = 3.4, N = 526, p =.034) had
significant effects on fertilization success, after the effects of sperm
release had been removed. A similar result was obtained from the nonparametric
FE estimate, with both site (
2 = 47.3, N =
20 sites, df = 19, p <.001) and female size (see below) showing
significant effects. In both analyses, date nested within a site had no effect
on FE. These analyses suggested that some site-specific factor influenced
fertilization success, either an aspect of the male or his spawning behavior,
or some aspect of water flow at the site. We examine each of these in turn
below.
Male mating success
The mean mating success for a male was 31.8 pair spawnings per day
(N = 37 males, s = 21.5, range 2-100). Mating success varied
among males even within individual patch reefs, with six of eight reefs having
coefficients of variation greater than 50% (three to five males per reef).
These estimates of variation in mating success are probably conservative,
since we likely missed some males with very low mating success on these
reefs.
There was a negative correlation between the mating success of a male and
his mean fertilization success (Figure
2, rs = -.56, N = 30, p
<.005, also demonstrated in an overlapping data set in
Warner et al., 1995). One
clear cause of this pattern is a negative correlation between sperm release
per spawn and mean mating success for a male (rs = -.52,
N = 30, p <.005).
|
Using the parametric estimate of FE, there was still a negative correlation with mating success after the effects of sperm number had been statistically removed (r = -.68, N = 25 males, p <.001; Figure 3A). This effect was most strongly influenced by the very low parametric FE of males with very high mating success. The correlation is weaker but still significant after removing the two males with the highest mating success (r = -.43, N = 23 males, p =.04). The nonparametric estimate of FE gave a similar result, exhibiting a negative correlation with male mating success (Figure 3B, r = -.47, p =.026, N = 22 males). However, in this case the relationship disappeared when the two males with the highest mating success were removed (r = -.11, p >.5, N = 20 males), reinforcing the importance of highly variable mating success to detect effects on FE.
|
The results of field manipulations in which we altered male mating success also indicate a negative relationship between male mating success and FE. In twelve cases where male mating success was manipulated, there was a negative correlation between the change in mating success and the change in nonparametric FE, with increasing mating success correlated with decreasing fertilization effectiveness (Figure 4, r = -.69, N = 12, p =.014).
|
The reduced fertilization effectiveness of males with high mating success is not related to a decrease in FE later in the spawning period due to fatigue or other factors. Using days with at least 50 spawns, no male showed a significant relationship between fertilization effectiveness (parametric FE) and the order of spawning (six males, N = 10-30 spawns per male, r varied from -.38 to.31, p >.1 in all cases).
Effect of site: hydrodynamics
Water velocity measured by the current meters varied considerably among
sites, with average estimates ranging from 2.6-22.1 cm/sec. Mean water
velocity was negatively correlated with fertilization success of the male at
the site (r = -.47, N = 29 males, p =.01). This
effect appeared to be separate from the effect of sperm released per spawn,
because water velocity was also negatively correlated with fertilization
effectiveness, examined either using the parametric estimate (r =
-.43, p <.05, N = 22 males;
Figure 5A) or the nonparametric
FE estimate (r = -.62, p <.01, N = 21 males;
Figure 5B). In a multiple
regression, using the parametric FE for each male, both water velocity
(t = -2.61, p =.017) and mating success (t = -3.74,
p =.001) were strongly negatively correlated with this estimate of FE
(total regression R2 =.61, N = 22 males). Using
the nonparametric measure of fertilization effectiveness, mating success
remained strongly significant (t = -4.01, p <.001) while
water velocity became slightly non-significant (t = -1.86, p
=.079).
|
The correlations between water velocity and FE are also observed among the six sites where we have estimates of parametric FE and current meter data on at least two separate days. There was a nearly significant effect of site (F5,64 = 2.28, p =.06) and a significant negative effect of the day's average water velocity on fertilization effectiveness (F1,64 = 5.23, p =.03; ANCOVA, after removing non-significant interaction term, R2 =.28) for the six sites with this level of resolution.
None of the water velocity or mixing characteristics at a site covaried with male mating success in this study. Analyses of water velocity and fertilization success parameters yielded correlation coefficients that were all < 0.2 (p >.20 for all tests, N ranges from 22-40). The estimate of water mixing, estimated as the variance in the current meter readings, were not correlated with any of the fertilization success parameters in this study (rs <.14 for all measurements, p >.20, N = 27).
Female size
There was a significant difference among female size categories in
parametric FE, with large females having lower fertilization success for the
same amount of sperm released compared with that of small females (Tukey
post-hoc comparison, p <.05). Similarly, in the nonparametric FE
analysis, large females had a lower FE than either medium or small females
(Table 1). Spawns involving
large females had significantly higher half-saturation constants of their
fertilization curves than small females
(Table 1). A larger
half-saturation constant indicates a higher number of sperm is required to
fertilize 50% of the eggs in a spawn.
|
Even though males released more sperm when spawning with large females (two-way ANOVA, female size F2,757 = 7.23, p =.001), fertilization success declined with female size (ANOVA, F2,775 = 2.97, p =.05), with a significant decrease between small and large females (p =.037, Tukey pairwise comparison; back-transformed mean fertilization success and approximate standard errors: large females 96.3% ± 0.6%, for small females 97.9% ± 0.35%).
Male size
There was no correlation between the size of a male and either measure of
fertilization effectiveness (p >.1 for all correlations).
Recent spawning history
There was no effect of female familiarity with site
(F1,116 = 0.002, p =.97) or male
(F1,116 = 0.37, p =.54) using the parametric FE
estimate (ANOVA's, N = 118 spawns). We obtained similar results using
the nonparametric fertilization effectiveness as the dependent variable: there
was no effect of either female familiarity with site (Fisher's Exact Test,
p =.4, N = 166 spawns) or the male (Fisher's Exact Test,
p =.51, N = 166).
There was no evidence that increased familiarity with a site or with a female increased fertilization effectiveness for a male. In ten cases where males spawned at two sites, FE was lower at the less used site in four of ten cases and higher in the other six.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Factors affecting fertilization success
The results provide a comprehensive picture of the factors affecting fertilization success in bluehead wrasse. Four factors affected fertilization success: the amount of sperm released, the mating success of a male, the water velocity at a spawning site, and the size of the female. Several potentially important factors were found not to affect fertilization success: size of the male and the familiarity of the female and male with the site or with each other. When combined with differences in fertilization success between group and pair spawnings (Marconato et al., 1997
), these data show that physical, behavioral, and demographic
variables all contribute to determining fertilization success in a spawn
(Figure 6). Each of these
factors is discussed individually below.
|
Male mating success. Mating success has a major impact on fertilization success in this species, especially at the extreme ranges of male mating success. The causes of this effect can be divided into two areas: the correlation of mating success with sperm release per spawn, and a second cause independent of the amount of sperm release.
The negative correlation between male mating success and fertilization
success in this species was reported previously
(Warner et al., 1995) and was
assumed to be due to a third variable, the number of sperm released per spawn,
that was positively correlated with fertilization success and negatively
correlated with mating success. Although sperm numbers are clearly important
(Figures 1 and
2), the current analysis
revealed a persistent pattern after the effect of sperm number had been
removed (Figure 3).
We lack sufficient information to conclude why mating success affects fertilization success after the effects of sperm number have been removed. Data from videotaping of individual spawns showed a significant correlation between some male spawning behavior and male mating success, and suggests that mating success might affect mating behavior or vice versa (Petersen, Warner, and Shapiro, unpublished data). The changes in spawning behavior that may occur with changes in mating success offer a challenging avenue for future research, especially since these behaviors may have a significant negative effect on fitness. However, more work is needed before a causal link can be inferred between mating behavior and fertilization effectiveness.
Males with higher mating success may also have lower fertilization effectiveness because their sperm are less effective at fertilizing eggs. We have no data to address this idea, but the change in FE when we experimentally altered fertilization success of individual males suggests that this is unlikely. If there is a difference in sperm quality, it would need to be closely linked with very recent male mating success.
Water velocity at a spawning site. Increasing water velocity is
expected to be correlated with enhanced water mixing at a site, and probably
decreases fertilization success in bluehead wrasse by diluting gametes
immediately after spawning. This pattern of decreasing fertilization with
increased water velocity has been observed in many marine invertebrates under
both artificially induced spawnings
(Levitan, 1998;
Levitan et al., 1992;
Levitan and Young, 1995;
Pennington, 1985) and under
natural fertilizations (Coma and Lasker,
1997; Lasker and Kapela,
1997), although large differences in fertilization success can
occur among samples collected minutes or meters from each other, presumably
because of spatial and temporal heterogeneity of flow
(Coma and Lasker, 1997;
Lasker and Kapela, 1997). In
this study we detected negative effects of increasing water velocity on
fertilization success both between and within sites.
Female size. The difference in fertilization effectiveness with
female size could have several causes that are not mutually exclusive. Larger
females have higher average fecundities
(Schultz and Warner, 1991;
Shapiro et al., 1994), and if
the ratio of sperm to eggs is important in determining fertilization success
(Shapiro and Giraldeau, 1996),
then this increase in egg number with female size will reduce fertilization
effectiveness. Larger females also probably create more water mixing at the
moment of spawning than smaller females, leading to larger gamete clouds. If
sperm concentration within the cloud is important in determining fertilization
success, which seems probable, then larger gamete clouds would have lower
fertilization effectiveness as measured in this study.
Female choice and fertilization success
None of the preference patterns exhibited by females correspond to the
hypothesis that they are selecting spawning sites based on maximizing
fertilization success. On the contrary, sites with males mating at higher
frequencies have lower fertilization success
(Warner et al., 1995), and at
least part of the pattern is due to the number of mates independent of the
number of sperm released per spawn (Figure
3). If females were selecting for maximizing fertilization
success, we would expect a distribution of females that gave approximately
equal fertilization success at each site. Instead, fertilization success
varied substantially among sites on a reef. The tremendous variability in
mating success among territorial males within a patch reef (e.g., Warner
1987,
1990a,b;
Warner and Schultz, 1992;
Warner et al., 1995, this
study), with the most successful individuals spawning 100 times per day
(Warner et al., 1995, this
study), is a major cause of the observed variability in fertilization success
in this species.
Mating success was not correlated with water velocity at a site within a
reef, despite a negative correlation between water velocity and fertilization
success. Females do not appear to be avoiding locations with higher water
mixing that decrease fertilization success. In fact, Hensley et al.
(1994) presented evidence that
bluehead wrasse spawning sites had higher water velocity than non-spawning
sites, and Shapiro et al.
(1997) presented evidence that
water velocities during the afternoon spawning period were higher than in the
morning. Both of these patterns are the opposite of what would be predicted if
spawning sites and times were chosen based on fertilization success alone.
The equivalence of fertilization effectiveness whether females chose
familiar or unfamiliar (which may only be less familiar) spawning partners or
sites suggests that the strong site and mate fidelity in this species (Warner,
1985,
1986,
1987,
1990a) is not due to a
fertilization benefit from mating with specific partners. In fact, this
fidelity may simply represent highly conservative traditional behavior whose
effect lies in mortality avoidance (Warner,
1988,
1990a,
1998).
Size-related patterns in fertilization success among females indicated that
females might benefit from changing from pair-spawns to group-spawns as they
grow larger. There is a significant decrease in fertilization success with
female size in pair spawns, with larger females having the lowest
fertilization success. In group-spawns fertilization success tends to be equal
or higher than in pair spawns (Marconato
et al., 1997), and there is no difference among females of
different size. This result suggests that the largest cost for spawning at
pair-spawning sites instead of group-spawning sites should exist for large
females. In this study, there was no difference in the distribution of female
size classes between pair- and group-spawns on a reef (unpublished data), but
in previous studies in Panama the trend is for large females to avoid
group-spawns (Warner, 1985),
the opposite to that predicted by fertilization success considerations.
Why are females not choosing situations that yield increased
fertilization success?
All else equal, we expect females to choose spawning situations that give
them the highest fitness per egg released. If differences exist in
fertilization success among sites or males, and these differences can be
assessed by females, then females could use this information in mate choice.
Unless there are offsetting fitness benefits for choosing situations with
lower fertilization success, then we expect females to prefer situations that
give higher fertilization success.
The range of fertilization success among males on a patch reef, which averages 5%, represents a small but real fertilization success difference among sites available to an individual female. Females cannot directly assess the fertilization success that their eggs achieve, but they may be able to gain information about fertilization success if they can assess either male mating success or water velocity at a site. However, there was no evidence that females used these factors to choose spawning sites or mates.
Warner et al. (1995)
outlined some of the potential compensatory benefits that might exist to
explain this pattern of female choice; lower fertilization success could be
correlated with higher genetic quality of males, increased female survivorship
or increased offspring survivorship. Alternatively, females may be constrained
in their information gathering because of tradition or mortality risks. There
is currently no evidence to suggest the occurrence of any of these processes
that might confer compensating benefits.
Different sites within reefs probably have similar long-term dispersal
capabilities (Appeldoorn et al.,
1994). Among these sites net larval export for a female will be
determined by her fecundity, her fertilization success, and the ability of
those young to avoid early predation. Current velocity may have opposite
effects on fertilization success and survivorship of young, increasing the
chance of zygotes successfully emigrating from the reef where egg predation is
probably higher (Hammer et al.,
1988; Hensley et al.,
1994; Randall and Randall,
1963), while decreasing sperm concentration and the probability of
eggs being fertilized. The antagonistic effect of these two factors on female
fitness cannot be compared since we lack any quantitative data on egg
predation rates in general, and do not know how this relationship changes with
location of the site and water velocity at the site
(Hensley et al., 1994).
The persistent intraspecific variation in fertilization success in the
bluehead wrasse suggests that the timing and location of spawning may be
controlled largely by factors other than fertilization success, which may
either be selective factors acting on offspring and adult survival, or
non-selective factors such as the inability of a species with a widely
dispersing larval phase to adapt to the selective regime in one particular
habitat (Warner, 1997).
Bluehead wrasse are found in a variety of reef environments other than patch
reefs, and in these other environments the benefits and costs of spawning in
certain places may be quite different. Females have been observed migrating
long distances downcurrent to mass group spawning sites in fore-reef
environments (Warner, 1995),
migrating to spawning areas bearing no simple relation to current direction in
large patch-reef systems (Hensley et al.,
1994), or spawning within all-purpose home ranges without
migrating in shallow back-reef habitats
(Fitch and Shapiro, 1990).
Similar work on other species will be needed to determine the generality of
these results to other marine fishes, where similar spawning behaviors that
should result in high sperm concentrations in the vicinity of eggs may also
maintain high fertilization success.
|
ACKNOWLEDGEMENTS |
|---|
We thank D. Bailey, M. Berard, D. Fitch, C. Gerstner, F. Giambartolomei, J. Hagopian, T. Knight, O. Kopecny, M. Obedzinski, M. Peharda, M. Rasotto, M. Sheehy, S. Smith, E. Stephens, D. Sward, M. Talkovic, V. Vredenburg, J. Wilson, C. Winter, and L. Wooninck for excellent assistance in the field. H.C. Hess, M. Hixon, R. Ydenberg, and two anonymous reviewers commented on the manuscript. This work was supported by the National Science Foundation (OCE 92-01320).
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Appeldoorn RS, Hensley DA, Kioroglou S, Sanderson BS, 1994. Egg dispersal in a Caribbean coral reef fish, Thalassoma bifasciatum. II. Dispersal off the reef platform. Bull Mar Sci 54: 271-280.
Brawley SH, Johnson LE, 1992. Gametogenesis, gametes, and zygotes: an ecological perspective on sexual reproduction in the algae. Brit Phycol J 27: 233-252.
Brazeau DA, Lasker HR, 1992. Reproductive success in the Caribbean octocoral Briareum asbestinum. Mar Biol 114: 157-163.
Clifton KE, 1997. Mass spawning by green algae on
coral reefs. Science 275:
1116-1118.
Coma R, Lasker HR, 1997. Small-scale heterogeneity of fertilization success in a broadcast spawning octocoral. J Exp Mar Biol Ecol 214: 107-120.
Fitch WTS, Shapiro DY, 1990. Spatial dispersion and non-migratory spawning in the bluehead wrasse, Thalassoma bifasciatum. Ethology 85: 199-211.
Giambartolomei FM, 1997. Sexual pattern and gonadal development in the Caribbean parrotfish Sparisoma radians (Osteichthyes, Scaridae) (MSc. thesis). Ypsilanti: Eastern Michigan University.
Giese AC, Kanatani H, 1987. Maturation and spawning. In: Reproduction of marine invertebrates (Giese AC, Pearse JS, Pearse VB, eds). Palo Alto/Pacific Grove, California: Blackwell Scientific/Boxwood Press; 251-329.
Hamner WM, Jones MS, Carleton JH, Hauri IR, Williams DMcB, 1988. Zooplankton, planktivorous fish, and water currents on a windward reef face: Great Barrier Reef, Australia. Bull Mar Sci 42: 459-479.
Hensley DA, Appeldoorn RS, Shapiro DY, Turingan RG, 1994. Egg dispersal in a Caribbean coral reef fish, Thalassoma bifasciatum. I. Dispersal over the reef platform. Bull Mar Sci 54: 256-270.
Kiflawi M, Mazeroll AI, Goulet D, 1998. Does mass spawning enhance fertilization in coral reef fish? A case study of the brown surgeon-fish. Mar Ecol Prog Ser 172: 107-114.
Knight TA, 1997. Morphology of the urogenital region of the diandric, protogynous wrasse, Thalassoma bifasciatum (MSc. thesis). Ypsilanti: Eastern Michigan University.
Lasker HR, Brazeau DA, Calderon J, Coffroth MA, Coma R, Kim K, 1996. In situ rates of fertilization among broadcast spawning gorgonian corals. Biol Bull 190: 45-55.[Abstract]
Lasker HR, Kapela WJ Jr, 1997. Heterogenous water flow and its effects on the mixing and transport of gametes. Proc Int Coral Reef Symp 2: 1109-1114.
Levitan DR, 1991. Influence of body size and population density on fertilization success and reproductive output in a free-spawning invertebrate. Biol Bull 181: 261-268.[Abstract]
Levitan DR, 1993. The importance of sperm limitation to the evolution of egg size in marine invertebrates. Am Nat 141: 517-536.
Levitan DR, 1995. The ecology of fertilization in free-spawning invertebrates. In: Ecology of marine invertebrate larvae (McEdward L, ed). Boca Raton, Florida: CRC Press: 123-156.
Levitan DR, 1996a. Effects of gamete traits on fertilization in the sea and the evolution of sexual dimorphism. Nature 382: 153-155.
Levitan DR, 1996b. Predicting optimal and unique egg sizes in free-spawning marine invertebrates. Am Nat 148: 174-188.[Web of Science]
Levitan DR, 1998. Does Bateman's principle apply to broadcast-spawning organisms? Egg traits influence in situ fertilization rates among congeneric sea urchins. Evolution 52: 1043-1056.
Levitan DR, Petersen CW, 1995. Sperm competition in the sea. Trends Ecol Evol 10: 228-231.
Levitan DR, Sewell MA, Chia F-S, 1992. How distribution and abundance influence fertilization success in the sea urchin Strongylocentrotus franciscanus. Ecology 73: 248-254.[Web of Science]
Levitan DR, Young CM, 1995. Reproductive success in large populations: empirical measures and theoretical predictions of fertilization in the sea biscuit Clypeaster rosaceus. J Exp Mar Biol Ecol 190: 221-241.
Marconato A, Shapiro DY, 1996. Sperm allocation, sperm production and fertilization rates in the bucktooth parrotfish. Anim Behav 52: 971-980.
Marconato A, Shapiro DY, Petersen CW, Warner RR, Yoshikawa T, 1997. Methodological analysis of fertilization rate in the bluehead wrasse, Thalassoma bifasciatum: pair versus group spawns. Mar Ecol Prog Ser 161: 61-70.
Marconato A, Tessari V, Marin G, 1995. The mating system of Xyrichthys novacula: sperm economy and fertilization success. J Fish Biol 47: 292-301.
Pearson GA, Serrão EA, Brawley SH, 1998. Control of gamete release in fucoid algae: sensing hydrodynamic conditions via carbon acquisition. Ecology 79: 1725-1739.
Pennington JT, 1985. The ecology of fertilization of
echinoid eggs: the consequences of sperm dilution, adult aggregation, and
synchronous spawning. Biol Bull 169:
417-430.
Petersen CW, Warner RR, Cohen S, Hess HC, Sewell AT, 1992. Variable pelagic fertilization success: implications for mate choice and spatial patterns of mating. Ecology 73: 391-401.
Randall JE, Randall HA, 1963. The spawning and early development of the Atlantic parrot fish Sparisoma rubripinne, with notes on other scarid and labrid fishes. Zoologica (NY) 48: 49-59.
Schultz ET, Warner RR, 1991. Phenotypic plasticity in life-history traits of female Thalassoma bifasciatum (Pisces: Labridae). 2. Correlation of fecundity and growth rate in comparative studies. Environ Biol Fishes 30: 333-344.
Serrão EA, Pearson G, Kautsky L,
Brawley SH, 1996. Successful external fertilization in turbulent
environments. Proc Natl Acad Sci USA 93:
5286-5290.
Shapiro DY, Appeldoorn RS, Hensley DA, Ray M, 1997. Water flow and spawning time in a coral reef fish. Proc Int Coral Reef Symp 2: 1121-1126.
Shapiro DY, Giraldeau L-A, 1996. Mating tactics in
external fertilizers when sperm is limited. Behav Ecol
7: 19-23.
Shapiro DY, Marconato A, Yoshikawa T, 1994. Sperm economy in a coral reef fish. Ecology 75: 1334-1344.[Web of Science]
Siegel S, Castellan NJ Jr, 1988. Non-parametric statistics for the behavioral sciences, 2nd ed. New York: McGraw Hill.
Warner RR, 1985. Alternative mating behaviors in a coral reef fish: a life history analysis. Proc Int Cor R Congr 4: 145-150.
Warner RR, 1986. The environmental correlates of female infidelity in a coral reef fish. In: Indo-Pacific fish biology: proceedings of the second international conference on Indo-Pacific fishes (Uyeno T, Arai R, Taniuchi T, Matsuura K, eds). Tokyo: Ichthyological Society of Japan; 803-810.
Warner RR, 1987. Female choice of sites versus mates in a coral reef fish, Thalassoma bifasciatum. Anim Behav 35: 1470-1478.
Warner RR, 1988. Traditionality of mating-site preferences in a coral reef fish. Nature 335: 719-721.
Warner RR, 1990a. Male versus female influences on mating-site determination in a coral reef fish. Anim Behav 39: 540-548.
Warner RR, 1990b. Resource assessment versus traditionality in mating site determination. Am Nat 135: 205-217.
Warner RR, 1995. Large mating aggregations and daily long-distance spawning migrations in the bluehead wrasse, Thalassoma bifasciatum. Environ Biol Fishes 44: 337-345.
Warner RR, 1997. Evolutionary ecology: how to reconcile pelagic dispersal with local adaptation. Coral Reefs 16S: s115-128.
Warner RR, 1998. The role of extreme iteroparity and risk-avoidance in the evolution of mating systems. J Fish Biol 53(suppl. A): 82-93.
Warner RR, Hoffman SG, 1980a. Local population size as a determinant of mating system and sexual composition in two tropical marine fishes (Thalassoma spp.). Evolution 34: 508-518.
Warner RR, Hoffman SG, 1980b. Population density and the economics of territorial defense in a coral reef fish. Ecology 61: 772-780.
Warner RR, Robertson DR, 1978. Sexual patterns in the labroid fishes of the western Caribbean. I. The wrasses (Labridae). Smith Contrib Zool 254: 1-27.
Warner RR, Robertson DR, Leigh EG Jr, 1975. Sex change
and sexual selection. Science 190:
633-638.
Warner RR, Schultz ET, 1992. Sexual selection and male characteristics in the bluehead wrasse, Thalassoma bifasciatum: mating site acquisition, mating site defense, and female choice. Evolution 46: 1421-1442.
Warner RR, Shapiro DY, Marconato A, Petersen CW, 1995. Sexual conflict: males with highest mating success convey the lowest fertilization benefits to females. Proc R Soc Lond B 262: 135-139.[Medline]
Yund PO, 1998. The effect of sperm competition on male gain curves in a colonial marine invertebrate. Ecology 79: 328-339.
Yund PO, 2000. How severe is sperm competition in natural populations of marine free-spawners? Trends Ecol Evol 15: 10-13.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. M. J. Skinner and P. J. Watt Strategic egg allocation in the zebra fish, Danio rerio Behav. Ecol., September 1, 2007; 18(5): 905 - 909. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. W. Petersen Sexual selection and reproductive success in hermaphroditic seabasses Integr. Comp. Biol., August 1, 2006; 46(4): 439 - 448. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Smith and M. Reichard Females solicit sneakers to improve fertilization success in the bitterling fish (Rhodeus sericeus) Proc R Soc B, August 22, 2005; 272(1573): 1683 - 1688. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||