| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Department of Biology, The Pennsylvania State University, University Park, Pennsylvania 16802 USA
Received for publication February 27, 2003. Accepted for publication May 30, 2003.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Campanula rapunculoides • Campanulaceae • inbreeding depression • mating system • New York • partial self-incompatibility • Pennsylvania • pollen performance • self-fertility
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Selfing plants contribute two haploid genomes to their selfed progeny and may contribute one haploid genome to outcrossed progeny on conspecifics (as pollen parent), while outcrossing plants contribute one haploid genome through each of their ovules (as seed parent) and one through each of the seeds they sire on conspecifics (Fisher, 1941
). The benefits of selfing are thus dependent on the extent to which selfers contribute to the outcross pollen pool (the extent of pollen discounting) and on the relative fitness of selfed vs. outcrossed progeny (the magnitude of inbreeding depression) (Barrett, 1998
). Consequently, the potential genetic transmission advantage of selfers can be independently or multiplicatively offset by a loss of fitness through pollen discounting or inbreeding depression. In general, it is thought that traits promoting self-fertilization will only invade a population when inbreeding depression is less than 0.5 (Lande and Schemske, 1985
; Holsinger, 1988
) although the time of occurrence of self- vs. cross-fertilization occurs is also important (Lloyd, 1992
).
Inbreeding depression has now been estimated for many species with a range of mating systems (Darwin, 1876
; Husband and Schemske, 1996
; Byers and Waller, 1999
). The majority of these studies compare the relative fitness of selfed progeny (Ws) to that of outcrossed progeny (Wx) through measures of seed set, survivorship to reproduction, and reproductive output through the female function. A smaller group of studies, however, have examined the consequences of inbreeding on male reproductive function. These studies have found adverse effects of inbreeding on pollen viability (Ritland and Ganders, 1987
; Krebs and Hancock, 1990
; Willis, 1993
, 1999
; Carr and Dudash, 1995
, 1997
; Johannsson et al., 1998
; Stephenson et al., 2001
), in vitro pollen tube length (Johannsson et al., 1998
; Stephenson et al., 2001
), and pollen production (Carr and Dudash, 1996
; Willis, 1999
). If inbreeding depression adversely affects pollen production and performance, then most studies have underestimated the magnitude of inbreeding depression and perhaps overestimated the transmission advantage (the degree of pollen discounting) (Stephenson et al., 2001
).
In addition to the interest in the mean level of inbreeding depression in plant populations, theoretical work (Uyenoyama and Waller, 1991a
, b
) has indicated that individual variation in selfing rates may lead to individual variation in genetic load, such that lineages that practice more self-fertilization may have lower genetic load than those that primarily outcross from the exposure of deleterious recessives to selection. If there are genetic loci that modify the selfing rate (for example, loci that influence the separation or timing of female and male reproductive fertility), then loci determining rates of self-fertilization may become associated with loci influencing genetic load (Uyenoyama et al., 1993
).
Several recent studies on plants from natural populations have attempted to detect an association between genetic load and a floral trait that affects mating history (Ashman, 1992
; Carr et al., 1997
; Mutikainen and Delph, 1998
; Chang and Rausher, 1999
; Daehler, 1999
; Takebayashi and Delph, 2000
). Most of these studies have used species that naturally vary in their degree of anther-stigma separation (herkogamy), an easily identifiable character that could influence selfing rates, to find associations between levels of inbreeding depression and degree of herkogamy. However, plants with genetically controlled partial self-incompatibility systems are also suitable subjects to investigate whether variation in individual genetic load is associated with breeding history. Partially self-incompatible (also called pseudo self-fertility; Levin, 1996
) species have genetically controlled self-incompatibility systems but harbor individuals that differ in their strength of self-incompatibility (SI) (i.e., strong and weak SI-phenotypes). Environmental conditions and/or genetic modifiers can cause this variation in the strength of SI (Levin, 1996
; Stephenson et al., 2000
). When partial SI has a heritable genetic basis, individual variation in self-fertility may result in populations of individuals with different rates of self-fertilization. If individuals differ in their ability to self-fertilize, more self-fertile families may exhibit less genetic load than more strongly self-incompatible families.
In this paper, we continue our exploration of the effects of inbreeding depression in strong and weak SI families using Campanula rapunculoides L., a species that is partially self-incompatible. In a previous paper, we generated four levels of inbreeding (f = 0, 0.25, 0.50, and 0.75) in six families and examined 10 vegetative and flowering traits (Vogler et al., 1999b
). We found that cumulative fitness (survivorship from fertilization to flowering x probability of flowering x relative flower production) decreased significantly with the intensity of inbreeding and that mean inbreeding depression was >0.90 after one generation of selfing. However, we also found that the decline in fitness with inbreeding was steeper for the progeny of strong SI lineages than for progeny from lineages with weak SI. This difference suggested that weak SI lineages might have purged some genetic load through selfing. In this paper, we extend the number of families examined to 11 and specifically focus on the effects of inbreeding on male fitness traits and on changes in self-fertility. We are especially interested in testing whether modifiers of self-incompatibility are coevolving with levels of genetic load to understand the potential impact of the coevolutionary dynamic on male fitness and self-fertility.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). It overwinters as a rosette and, in midsummer, each rhizomatous cluster produces one to eight flowering shoots (racemes) of 10–60 flowers that mature acropetally (bottom upwards). The blue, bell-shaped flowers are bumble bee pollinated and protandrous. Thus, at anthesis, the flowers are staminate (male phase) and the stigmatic lobes are tightly appressed. After 1–3 d during which the pollen is removed by bees, the stigmatic lobes reflex and pollen deposition can occur (female phase). Thus, at any given time during the flowering season, an inflorescence will consist of (from bottom to top) developing fruits, female-phase flowers, male-phase flowers, and flower buds. Although floral protandry and the upward foraging of bumble bees on inflorescences tend to promote outcrossing, there are opportunities for self-pollination through geitonogamous transfer of pollen between inflorescences of the same genotype and by autogamy in late female-phase flowers that have not received pollen from conspecifics (Stephenson et al., 2000
). Previous work in our laboratory has shown, for two populations, that most genotypes become significantly more self-fertile as flowers age, that there are strong genetic differences between individuals in their strength of SI, that there is heritable variation for the strength of SI (Vogler et al., 1998
; Good-Avila and Stephenson, 2002
), and that SI is gametophytically controlled (Good, 2000
). We assess the strength of self-incompatibility of each experimental plant according to a self-incompatibility index (SI index = seed set after self-pollination/seed set after cross-pollination) adopted from Zapata and Arroyo (1978)
. We do not know the natural selfing rate of individuals in natural populations, although we are currently examining this, but a natural garden experiment showed that moderately SI individuals (SI index of 0.4–0.6, self seeds/cross seeds) had selfing rates of about 20% and that this rate could be significantly changed by limiting access of plants to pollinators (Good-Avila et al., 2001
).
Production of the inbred families
The plants used in this study were taken from two natural populations, one outside State College, Pennsylvania, USA, and one outside Duanesburg, New York, USA. These 11 maternal families differed in their strength of self-incompatibility as determined by hand self-pollinations on day-4 female-phase flowers, but all families exhibit high consistency across clones and years (Good-Avila and Stephenson, 2002
). We call those plants with an SI index >0.4 in day-4 female-phase "weakly SI" and those with SI index <0.22 as "strongly SI" (modified from Zapata and Arroyo, 1978
). By this criterion, five of our maternal families are strongly SI (SI indices of 0.05, 0.11, 0.15, 0.20, 0.22) and six weakly SI (SI indices of 0.41, 0.45, 0.58, 0.85, 0.86, 0.91), and both populations were represented in the strong and weak SI families. In 1995–1996 clones of the 11 maternal families were hand selfed and outcrossed. All of the maternal families from the Pennsylvania population (three families) were crossed to one donor plant from the same population (3–5 times) but situated at least 10 m from the maternal families. The maternal families from the New York population (eight families) were crossed with two plants from the same population (but again, separated from the maternal plants) 2–4 times (a total of six cross pollinations were performed), and then all seeds were pooled. The outcrossed and selfed progeny from these crosses were grown to flowering in the following year and subjected to the following treatments. Outbred maternal plants were outcrossed, mated with a full-sibling or selfed—thereby generating levels of inbreeding in the progeny of f = 0.0, 0.25, or 0.5. Inbred maternal plants were self-fertilized, generating a fourth level of inbreeding in the progeny of f = 0.75. The calculated inbreeding coefficients are minimum estimates that will be higher if some parents themselves have an inbreeding coefficient greater than zero.
In the winter of 1997–1998, 100 seeds from each of the four inbreeding levels were germinated on moist filter paper in petri dishes and kept at room temperature for 10–18 d. From each family, 20–50 seedlings were randomly chosen and sown into flats in the greenhouse. At week 8, 396 plants, up to 10 plants per family per inbreeding level, were randomly chosen, transplanted to 4-L pots and left to flower. At the end of flowering, the flowering stalks of the plants that flowered were removed leaving only the rosette and any plants that died were eliminated from the experiment, leaving 347 plants. These plants were transferred to a 4°C coldroom for 4 mo. In the spring of 1999, these plants were returned to the greenhouse and left to flower synchronously. Of these 347 plants, 244 flowered in the summer of 1999. A mean of 8–10 progeny per family flowered from the outcross families (f = 0.0, N = 92, X = 8.4), 2–8 progeny per family flowered from sib-crosses (f = 0.25, N = 54, X = 4.9), 3–10 progeny per family flowered from selfed crosses (f = 0.5, N = 0.61, X = 5.5), and 2–8 progeny per family flowered from the twice-selfed crosses (f = 0.75, N = 37, X = 3.4).
Traits analyzed
From the 244 plants that flowered, we collected pollen and seed set data to assess the effect of inbreeding on male fitness and variation in self-fertility. To assess male fertility, we collected one flower from each plant on the first day of the male phase and assessed pollen production and diameter using an electronic particle counter (Elzone 180 with EX 1010 software; Micromeritics, Norcross, Georgia, USA) as described by Vogler et al. (1999b)
. In addition, we determined mean in vitro pollen tube length by collecting fresh pollen from two male-phase flowers per plant and immersing the pollen from each flower in 5 mL of Brewbaker and Kwack (1963)
solution. One milliliter of this solution was placed on three petri plates containing 12% agarose dissolved in the same Brewbaker and Kwack medium (Kearns and Inouye, 1993
). The three plates were incubated at 22°C for 24 h and then the growth of the pollen tubes was halted by adding 70% ethanol. The length of 20 pollen tubes per plate was measured with the aid of an image analysis system. The mean of these 20 tubes was recorded, and then the mean of all three plates per plant was taken as the mean pollen tube length for that plant. Pollen tube length data were not collected for one of the families, family B, leaving only 236 plants for this analysis.
To assess possible changes in self-fertility under inbreeding, we calculated the mean outcrossed seed production and the mean number of seeds set in day-1 and day-4 female-phase flowers as previously described (Good-Avila and Stephenson, 2002
). Then we estimated changes in self-fertility by calculating the relative self-fertility of individuals as the mean number of selfed seeds in day-1 or day-4 flowers/mean number of outcrossed seeds. Seed set data could not be collected for a few plants, and 224 plants are included in the seed set analyses.
Data analyses
We calculated the mean and standard error of pollen production per flower, pollen diameter, in vitro pollen tube length, outcrossed seed set, day-1 selfed seed set, and day-4 selfed seed for all families at each inbreeding level. We calculated the effect of inbreeding at each level for each trait as 1 – WI/WO. Although inbreeding depression usually specifically refers to the difference in fitness between selfed and outcrossed progeny, here we use the term to describe the decline in fitness of all inbred (f = 0.25–0.75) progeny. Significant differences in mean phenotype between levels were examined using Tukey-Kramer pairwise comparisons and an overall alpha of 0.05. We also calculated the self-fertility (mean production of selfed/outcrossed seeds) in 1- and 4-day-old flowers. Lastly, we calculated the average inbreeding depression separately for weak or strong SI families at each level of inbreeding.
Fixed effect model analyses of variance were conducted on six components of reproductive fitness (mean pollen production, pollen diameter, in vitro pollen tube length, outcrossed seed set, day-1 and day-4 self-fertility) to determine if strength of maternal SI affects progeny performance (Proc GLM, SAS Institute, 1994
). Type III sums of squares were used to test the significance of maternal SI (the self-incompatibility phenotype of the family, weak or strong), the coefficient of inbreeding (f = 0.0, 0.025, 0.5, or 0.75) and the interaction of maternal SI and the inbreeding coefficient (maternal SI x f). The three seed-set characters were square-root transformed prior to analysis to meet the assumption of normality of residuals in analysis of variance. When maternal SI affected a trait, a Student's t test was performed to examine the difference between the means of weak and strong SI families. When there was a significant maternal SI x f interaction, we conducted pairwise Tukey-Kramer comparisons with an overall alpha of 0.05 to determine which comparisons was significantly different.
A second series of analyses of variance was performed on the same components of fitness to examine the response to inbreeding of families irrespective of maternal SI. This was achieved using the residuals of the regression of Y = maternal SI as the response variable for the following models, an approach that will underestimate among-family variation as compared to a mixed model that uses maternal SI as a fixed effect and maternal family as a random one. In the first model, we studied the effect of maternal family, inbreeding level (f), and their interaction (maternal family x f) on progeny vigor such that
 | (1) |
outline the following model: 
and ß are the coefficients for linear and quadratic effects for the variate fj, i, and ßi are the coefficients for linear and quadratic effects of the inbreeding-level fj on an individual maternal lineage, and Cij is the residual after accounting for differences in fj between clones. If loci causing inbreeding depression interact independently and additively, then fitness will decline linearly with increased inbreeding. If loci determining fitness have independent multiplicative effects, as we expect when there are epistatic interactions among loci, then fitness is expected to decline exponentially with inbreeding. The full model described in Eq. 2 is nonlinear, but the following model can be fitted sequentially and used to estimate only the terms , ß, and R
 | (3) |
To visually examine the interrelationship between inbreeding depression and strength of maternal SI, we plot the change in fitness over inbreeding coefficient for each trait for each maternal family identified by its strength of SI (strong or weak).
All statistical analyses were performed using JMP version 3.0.1 (SAS 1994
).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
|
The analysis of variance reveals that there were significant interactions between maternal SI and f (coefficient of inbreeding) for pollen production and pollen diameter, indicating that weak and strong SI families responded differently to inbreeding for these traits (Table 2). Tukey pairwise comparisons indicate that although weak SI families have greater mean pollen productions at low inbreeding coefficients than strong SI families (f = 0.0 and f = 0.25), the effect reverses at higher inbreeding coefficients (f = 0.5 and f = 0.75) with strong SI families showing greater mean pollen production (see also Fig. 1). For pollen diameter, the significant interaction term is caused by the aberrant mean pollen diameter for strong SI families at f = 0.5 (pollen diameter was greater for plants with f = 0.5 than for plants with f = 0.25). In general, pollen diameter was greater in the strong than weak SI families, and it tended to decline as the inbreeding coefficient increased.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Krebs and Hancock, 1990
; Willis, 1993
, 1999
; Carr and Dudash, 1995
, 1997
; Johannsson et al., 1998
; Stephenson et al., 2001
), in vitro pollen tube length (Johannsson et al., 1998
; Stephenson et al., 2001
), and pollen production (Carr and Dudash, 1996
; Willis, 1999
).
In this study, we found that two generations of selfing decreased pollen production per flower by 63%, pollen diameter by 16% (a 40% decrease in pollen volume), and pollen tube growth in vitro by 39%. Moreover, in a previous study (Vogler et al., 1999b
), we found that flower number per plant decreased by more than 40% after two generations of selfing. Consequently, f = 0.75 plants produced only 22% of the pollen of an outcrossed plant because they produced both fewer flowers and fewer pollen grains per flower. If smaller plants with fewer flowers are not as attractive to pollinators, as reported for other species (Stephenson, 1979
; Devlin et al., 1992
), then pollen donation to conspecifics under field conditions may be even more reduced than that predicted by pollen production per se.
Moreover, the germination and initial growth rate of pollen tubes are known to be strongly dependent upon resources such as starches, lipids, and phytate stored in the pollen (Stephenson et al., 1994
; Delph et al., 1997
). These resources are secreted by the tapetal "nurse" cells of the anther during pollen development and are metabolized upon germination and growth of the pollen tubes (see reviews by Stephenson et al., 1994
; Hormaza and Herrero, 1996
). Recently, several studies that have experimentally varied environmental conditions that affect the ability of a plant to provision its developing pollen (e.g., soil nutrients, herbivory) have shown positive correlations among resource availability, pollen grain size, the chemical composition of the pollen, in vitro pollen performance, and the ability to sire seeds under conditions of pollen competition (see reviews by Stephenson et al., 1994
; Delph et al., 1997
). In this study, we find that both pollen diameter (volume) and in vitro pollen performance are affected by inbreeding. We suspect that the differences in the size and in vitro performance of pollen from inbred and outbred plants are due to differences in the ability of inbred and outbred plants to provision their developing pollen grains. Consequently, inbred plants not only produce fewer pollen grains that can be dispersed to conspecifics but those that are deposited onto a stigma of a conspecifics may be less likely to achieve fertilization, at least under competitive conditions.
Because species with any form of SI are strong outcrossers, it is reasonable to expect that they should have severe inbreeding depression because species with SI undergo negative assortative mating, a condition which maximizes heterozygosity which, in turn, is thought to maintain high genetic load (Hartl and Clark, 1989
). This expectation is largely supported by the few studies of inbreeding depression in predominantly SI species (Eckert and Barrett, 1994
; Luijten et al., 1996
; Chen et al., 1997
; Daehler, 1999
; Vogler et al., 1999b
; Borba et al., 2001
) although Cheptou et al. (2000)
found only moderately high levels of inbreeding depression in a partially SI species. Similarly, there are theoretical reasons to expect that, within a species, lineages that differ in traits that affect the selfing rate (e.g., stigma-anther separation, strength of SI) will differ in the amount of inbreeding depression they experience upon selfing (Uyenoyama et al., 1993
). The reasons for this are two-fold: first, traits that promote selfing are more likely to spread in families with low levels of inbreeding depression, and second, families with a history of inbreeding are more likely to have purged at least some of their deleterious recessives. Of the several studies that have sought to find associations between lineages that possess traits expected to alter the selfing rates, three have found positive associations between the trait and inbreeding depression (Chang and Rausher, 1999
; Daehler, 1999
; Takebayashi and Delph, 2000
), while three have not (Ashman, 1992
; Carr et al., 1997
; Mutikainen and Delph, 1998
). Such mixed findings are not unexpected, however, because as Schultz and Willis (1995)
demonstrated, background levels of mutation may be sufficiently high to overwhelm our ability to detect associations between mating history and genetic load.
In C. rapunculoides, we have now accumulated considerable evidence that variation in the strength of SI among individuals may be of evolutionary importance. For example, we have shown that the strength of SI has a heritable genetic basis (Good-Avila and Stephenson, 2002
), that strong and weak SI families differ in inbreeding depression for vegetative traits (Vogler et al., 1999b
) and show differential performance as pollen and seed parents (Good-Avila and Stephenson, 2003
), and here, we have shown that strong and weak families have differential reproductive output through both male and female function. The strong influence of variation in self-fertility on reproductive success has certainly strengthened our ability to detect associations between genetic load and modifiers of SI. Schultz and Willis (1995)
cite three criterion that should increase the ability to detect associations between genetic load and traits that influence selfing rates and, interestingly, C. rapunculoides appears to satisfy each one. Firstly, rates of self-fertilization in C. rapunculoides are probably highly variable in natural populations. Although most plants are strongly SI (SI index <0.2), about 15% of the plants from two natural populations are completely self-compatible. Secondly, highly deleterious mutations probably contribute strongly to inbreeding depression as evidenced by the high levels of inbreeding depression in early-expressed traits (Vogler et al., 1999b
). Lastly, selfing rates are certainly much less than 1 in this species (probably in the range of 0.1 to 0.3). Therefore, mean levels of genetic load and among-family variation in load may be high enough that the association between genetic load and modifiers of the selfing rate could be detected.
When the findings of this study are combined with a previous study (Vogler et al., 1999b
) that found that cumulative inbreeding depression exceeded 0.90 for early traits related to growth, survivorship, and flower production, it is evident that selfing poses serious problems for growth, survivorship, and reproduction in C. rapunculoides. Given the existence of a prezygotic self-incompatibility system (Good, 2000
), a flowering phenology that promotes outcrossing (see Materials and Methods) and the tendency of outcross pollen to outcompete self-pollen even when simultaneously deposited onto the old flowers on plants with weak SI (Vogler and Stephenson, 2001
), it is evident that plants do not reproduce frequently through self-fertilization. As such, it is reasonable to ask how we can account for the maintenance of variation in the strength of SI (degree of self-fertility) in C. rapunculoides. One explanation is that variation in the strength of SI has been maintained by reproductive assurance. If selfing regularly occurs only after all opportunities for outcrossing have passed and if investments in selfed seeds do not reduce future reproductive success, then there would be no pollen discounting associated with selfing and the production of selfed seed would be better than no seed production (Holsinger, 1988
; Schoen and Brown, 1991
; Becerra and Lloyd, 1992
; Lloyd, 1992
). Occasional self-fertilization could be especially selected in colonizing species, such as C. rapunculoides, because population sizes are frequently small. We have several lines of evidence that self-fertilization is most likely to be successful in certain genotypes in older, unpollinated flowers. Our greenhouse studies have revealed that SI breaks down (self-fertility increases) with floral age (Stephenson et al., 1992
; Vogler et al., 1998
, 1999a
; Good-Avila and Stephenson, 2002
), that self-fertility of old flowers increases when prior fruit set on an inflorescence is low (Vogler et al., 1998
), and that rates of self-fertilization increase in older unpollinated flowers under field conditions when pollinator access is limited (Good-Avila et al., 2001
). Together, these experiments indicate that self-fertilization would not appreciably decrease opportunities for pollen dispersal and cross pollen deposition onto stigmas in natural populations and that self-fertilization could provide reproductive assurance. However, given that pollen production significantly declines in inbred progeny, selfed progeny would be unlikely to reap a benefit from reproductive assurance because of their reduced fitness and the cost of pollen discounting.
The maintenance of variation in the degree of self-fertility may also be related to the genetic control of SI in C. rapunculoides. Latta and Ritland (1993)
showed that when the selfing rate is controlled by a single locus, mixed mating is difficult to evolve and maintain. However, when the selfing rate is controlled by several loci of small effect, evolutionarily stable mixed mating can occur under a broader range of conditions. Given that the strength of SI in C. rapunculoides is modified by primarily additive effects of alleles at several loci (Good-Avila and Stephenson, 2002
), it is conceivable that site-specific differences and year-to-year variations in population size and pollinator availability can influence selection on the loci that modify the strength of SI. Small population sizes (few S alleles) and low pollinator availability would select for greater self-fertility (Good-Avila et al., 2001
), while the high levels of inbreeding depression reported here would select for stronger SI in large populations with abundant pollinators.
Furthermore, selection for or against self-fertility will also be affected by the genetic basis of inbreeding depression. Here, we find evidence of significant linear effects of the response to inbreeding for pollen production, pollen diameter, in vitro tube length, and outcrossed seed set. These results are in accord with a growing number of studies in plants that show that additive effects of independent loci appear to be the main cause of inbreeding depression (Willis, 1993
; Johnston and Schoen, 1996
; Dudash and Carr, 1998
; Koelewijn, 1998
; Fishman, 2001
). As theoretical work by Damgaard (2000)
has shown, the slow decline of inbreeding depression with increased inbreeding would help maintain mixed mating systems because selfing lineages continue to pay a price for self-fertilization. We also find significant quadratic effects for the response to inbreeding for pollen diameter and in vitro pollen tube length, but in neither case was the quadratic fit better than the linear one. From an inspection of Fig. 1 some families appear to have a quadratic change in fitness, while others a linear one. This difference means that independent family estimates of the linear and quadratic effects are more appropriate (Bulmer, 1980
), but the sample sizes in this experiment were deemed insufficient at the highest inbreeding coefficients to warrant this attention. Although linear and quadratic effects of inbreeding are conceptually different, distinguishing between them can be difficult because, at low to moderate inbreeding levels, they predict a similar response (Hedrick and Kalinowsky, 2000
). Overall, our data suggest that while linear effects appear to be strong determinants of the mean level of inbreeding depression, epistatic and purging effects should not be ruled out as influencing the evolution of inbreeding in species.
Finally, our data suggest several things about changes in self-fertility with inbreeding. Selfed seed set and self-fertility tended to increase with inbreeding in strong SI families (Table 1), but strong SI families suffered greater inbreeding depression than weak SI families for outcrossed seed set such that their mean self-fertility increased nonsignificantly with inbreeding (ANOVA results). Self-fertility in weak SI families increased slightly in day-1 flowers and decreased slightly in day-4 flowers, but individual families differed greatly in their response to inbreeding. This suggests that self-fertility can respond in various ways to inbreeding, concordant with other studies that have found both increases (Robacker and Ascher, 1978
; Dana and Ascher, 1985
; Levin, 1995
) and decreases (Wilsie, 1958
; Krebs and Hancock, 1990
) in self-fertility in response to inbreeding in partially SI species. The tendency for weak SI families to decline in self-fertility with inbreeding, combined with high overall levels of inbreeding depression, show that it is unlikely the species could become highly self-fertile or even practice repeated generations of self-fertilization. However, the increase in selfed seed set and self-fertility in inbred strong SI families, the wide variation in self-fertilities expressed at the highest inbreeding coefficient (see Fig. 1e and f), and the difference in genetic load between weak and strong SI families suggest that variation in self-fertility may play a role in mating system evolution. We are currently examining the role of population structure in selection for variable self-fertility by calculating the outcrossing rate in natural populations that vary in size, isolation, and age. This study will enable us to further evaluate the factors that are maintaining polymorphism in levels of self-fertility in C. rapunculoides.
|
FOOTNOTES |
|---|
2 Present address: Department of Biology, Acadia University, Wolfville, Nova Scotia, Canada B0P 1X0 (telephone 902-585-1798; Fax 902-585-1059; e-mail: sara.good-avila{at}acadiau.ca
)
3 Present address: Department of Biology, State University of New York (SUNY), Oneonta, New York 13820 USA
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Barrett S. C. H. 1998 The evolution of mating strategies in flowering plants. Trends in Plant Science 3: 335-341[CrossRef][ISI]
Becerra J. X. D. G. Lloyd 1992 Competition-dependent abscission of self-pollinated flowers of Phormium tenax (Agavaceae): a second action of self-incompatibility at the whole flower level?. Evolution 46: 458-469[CrossRef][ISI]
Borba E. L. J. Semir G. J. Shepherd 2001 Self-incompatibility, inbreeding depression and crossing potential in five Brazilian Pleurothallis (Orchidaceae) species. Annals of Botany London 88: 89-99[CrossRef]
Brewbaker J. L. B. H. Kwack 1963 The essential role of calcium ion in pollen germination and pollen tube growth. American Journal of Botany 50: 859-865[CrossRef][ISI]
Bulmer M. G. 1980 The mathematical theory of quantitative genetics. Oxford University Press, Oxford, UK
Byers D. L. D. M. Waller 1999 Do plant populations purge their genetic load? Effects of population size and mating history on inbreeding depression. Annual Review of Ecology and Systematics 30: 479-513
Carr D. M. Dudash 1995 Inbreeding depression under a competitive regime in Mimulus guttatus: consequences for potentital male and female function. Heredity 75: 437-445[ISI]
Carr D. M. Dudash 1996 Inbreeding depression in two species of Mimulus with contrasting mating systems. American Journal of Botany 83: 586-593[CrossRef][ISI]
Carr D. E. M. R. Dudash 1997 The effects of five generations of enforced selfing on potential male and female function in Mimulus guttatus. Evolution 51: 1797-1807[CrossRef][ISI]
Carr D. E. C. B. Fenster M. R. Dudash 1997 The relationship between mating-system characters and inbreeding depression in Mimulus guttatus. Evolution 51: 363-372[CrossRef][ISI]
Chang S.-M. M. D. Rausher 1999 The role of inbreeding depression in maintaining the mixed mating system of the common morning glory, Ipomoea pupurea. Evolution 53: 1366-1376[CrossRef][ISI]
Chen L. H. M. Stace S. H. James. 1997 Self-incompatibility, seed abortion and clonality in the breeding systems of several Western Australian Drosera species (Droseraceae). Australian Journal of Botany 1: 191-201
Cheptou P. O. A. Berger A. Blanchard C. Collin J. Escarre 2000 The effect of drought stress on inbreeding depression in four populations of the Mediterranean outcrossing plant Crepis sancta (Asteraceae). Heredity 85: 294-302
Daehler C. C. 1999 Inbreeding depression in smooth cordgrass Spartina alterniflora (Poaceae) invading San Francisco Bay. American Journal of Botany 86: 131-139
Damgaard C. 2000 Fixation of advantageous alleles in partially self-fertilizing populations: the effect of different selection modes. Genetics 154: 813-821
Dana M. N. P. D. Ascher 1985 Pseudo-self compatibiltiy (PSC) in Petunia itegrifolia. Journal of Heredity 76: 468-470
Darwin C. 1876 The effects of cross and self fertilisation in the vegetable kingdom. Murray, London, U.K
Delph L. F. M. Johannsson A. Stephenson 1997 How environmental factors affect pollen performance: ecological and evolutionary perspectives. Ecology 78: 1632-1639[CrossRef][ISI]
Devlin B. J. Clegg N. C. Ellstrand 1992 The effect of flower production on male reproductive success in wild radish populations. Evolution 46: 103-142
Dudash M. D. Carr 1998 Genetics underlying inbreeding depression in Mimulus with contrasting mating systems. Nature 393: 692-684[CrossRef][Medline]
Eckert C. G. S. C. H. Barrett 1994 Inbreeding depression in partially self-fertilizing Decodon verticillatus (Lythraceae): population-genetic and experimental analyses. Evolution 48: 952-964
Fisher R. A. 1941 Average excess and average effect of a gene substitution. Annals of Eugenetics 11: 53-63
Fishman L. 2001 Inbreeding depression in two populations of Arenaria uniflora (Caryophyllaceae) with contrasting mating systems. Heredity 86: 177-183[ISI][Medline]
Good S. V. 2000 Partial self-incompatibility in Campanula rapunculoides L. (Campanulaceae): genetic and molecular bases, parental effects and the influence of pollinator availability on the selfing rate. Ph.D. dissertation, The Pennsylvania State University, University Park, Pennsylvania, USA
Good-Avila S. V. F. Frey A. G. Stephenson 2001 The effect of partial self-incompatibility on the breeding system of Campanula rapunculoides L. (Campanulaceae) under conditions of natural pollination. International Journal of Plant Sciences 162: 1081-1087[CrossRef]
Good-Avila S. V. A. Stephenson 2002 The inheritance of modifiers conferring self-fertility in Campanula rapunculoides L. (Campanulaceae). Evolution 56: 263-272[CrossRef][ISI][Medline]
Good-Avila S. V. A. G. Stephenson 2003 Parental effects in a partially self-incompatible herb: influence of variation in the strength of self-incompatibility on seed set and progeny performance. American Naturalist 161: 615-630[CrossRef][Medline]
Hartl D. L. A. G. Clark 1989 Principles of populations genetics, 1st ed. Sinauer, Sunderland, Massachusetts, USA
Hedrick P. W. S. T. Kalinowsky 2000 Inbreeding depression in conservation biology. Annual Review of Ecology and Systematics 31: 139-162[CrossRef][ISI]
Holsinger K. E. 1988 Inbreeding depression doesn't matter: the genetic basis of mating-system evolution. Evolution 42: 1235-1244[CrossRef][ISI]
Hormaza J. L. M. Herrero 1996 Dynamics of pollen tube growth under different competition regimes. Sexual Plant Reproduction 9: 153-160[ISI]
Husband B. C. D. W. Schemske 1996 Evolution of the magnitude and timing of inbreeding depression in plants. Evolution 50: 54-70
Johannsson M. H. M. J. Gates A. G. Stephenson 1998 Inbreeding depression affects pollen performance in Cucurbita texana. Journal of Evolutionary Biology 11: 579-588[CrossRef][ISI]
Johnston M. O. D. J. Schoen 1996 Correlated evolution of self-fertilization and inbreeding depression: an experimental study of nine populations of Amsinckia (Boraginaceae). Evolution 50: 1478-1491[CrossRef][ISI]
Kearns C. A. D. W. Inouye 1993 Techniques for pollination biologists. University Press of Colorado, Niwot, Colorado, USA
Koelewijn H. P. 1998 Effects of different levels of inbreeding on progeny fitness in Plantago coronopus. Evolution 52: 692-702[CrossRef][ISI]
Krebs S. L. J. F. Hancock 1990 Early-acting inbreeding depression and reproductive success in the highbush blueberry, Vaccinium corymbosum L. Theoretical and Applied Genetics 79: 825-832
Lande R. D. W. Schemske 1985 The evolution of self-fertilization and inbreeding depression in plants. 1. Genetic models. Evolution 39: 24-40[CrossRef][ISI]
Latta R. K. Ritland 1993 Models for the evolution of selfing under alternative modes of inheritance. Heredity 71: 1-10[ISI]
Levin D. A. 1995 Effect of inbreeding on autogamy in Phlox. Heredity 74: 108-113[ISI]
Levin D. A. 1996 The evolutionary significance of pseudo self-fertility. American Naturalist 148: 321-332[CrossRef][ISI]
Lloyd D. G. 1992 Self- and cross-fertilization in plants II: theoretical considerations. International Journal of Plant Science 153: 370-380[CrossRef]
Luijten S. H. J. G. B. Oostermeijer C. Van Leeuwen Nico C. M. Den Nijs Hans 1996 Reproductive success and clonal genetic structure of the rare Arnica montana (Compositae) in The Netherlands. Plant Systematics and Evolution 201: 15-30[CrossRef][ISI]
Mutikainen P. L. F. Delph 1998 Inbreeding depression in a gynodioecious Lobelia siphilitica: among family differences override between-morph differences. Evolution 52: 1572-1582[CrossRef][ISI]
Ritland K. F. R. Ganders 1987 Crossability in Mimulus guttatus in relation to components of gene fixation. Evolution 41: 771-786
Robacker C. D. P. D. Ascher 1978 Restoration of pseudo-self-compatibility (PSC) in derivative of a high-PSC x No-PSC cross in Nemesia strumosa Benth. Theoretical and Applied Genetics 53: 135-141[CrossRef][ISI]
Rosatti T. J. 1986 The genera of Sphenocleaceae and Campanulaceae in the southeastern United States. Journal of the Arnold Arboretum 67: 1-64[ISI]
SAS, 1994 JMP user's guide. SAS Institute, Cary, North Carolina, USA
Schoen D. J. A. D. H. Brown 1991 Whole and part flower self-pollination in Glycine clandestina and G. argyrea and the evolution of autogamy. Evolution 45: 1651-1664[CrossRef][ISI]
Schultz S. T. J. H. Willis 1995 Individual variation in inbreeding depression: the roles of inbreeding history and mutation. Genetics 141: 1209-1223[Abstract]
Stephenson A. 1979 An evolutionary examination of the floral display of Catalpa speciosa (Bignoniaceae). Evolution 33: 1200-1209[CrossRef][ISI]
Stephenson A. G. C. W. Erickson T.-C. Lau M. Quesada J. A. Winsor 1994 Effects of growing conditions on the male gametophyte. In A. G. Stephenson and T.-H. Kao [eds.], Pollen–pistil interactions and pollen tube growth, 220–229. Current topics in plant physiology, vol. 12. American Society of Plant Physiologists, Rockville, Maryland, USA
Stephenson A. G. S. V. Good D. W. Vogler 2000 Interrelationships among inbreeding depression, plasticity in the self-incompatibility system, and the breeding system of Campanula rapunculoides L. (Campanulaceae). Annals of Botany 85: 211-220
Stephenson A. G. C. N. Nayes M. H. Johannsson J. A. Winsor 2001 The performance of microgametophytes is affected by inbreeding depression and hybrid vigor in the sporophytic generation. Sexual Plant Reproduction 14: 77-83
Stephenson A. G. J. A. Winsor T. E. Richardson A. Singh T.-H. Kao 1992 Effects of style age on the performance of self and cross pollen in Campanula rapunculoides. In E. Ottaviano, D. L. Mulcahy, M. Sari Gorla, and G. B. Mulcahy [eds.], Angiosperm pollen and ovules, 117–121. Springer-Verlag, New York, New York, USA
Takebayashi N. L. F. Delph 2000 An association between a floral trait and inbreeding depression. Evolution 54: 840-846[ISI][Medline]
Uyenoyama M. K. K. E. Holsinger D. M. Waller 1993 Ecological and genetic factors directing the evolution of self-fertilization. Oxford Surveys in Evolutionary Biology 9: 327-381
Uyenoyama M. K. D. M. Waller 1991a Coevolution of self-fertilization and inbreeding depression I. Mutation-selection balance at one and two loci. Theoretical Population Biology 40: 14-46[CrossRef][ISI][Medline]
Uyenoyama M. K. D. M. Waller 1991b Coevolution of self-fertilization and inbreeding depression. II. Symmetric overdominance in viability. Theoretical Population Biology 40: 47-77[CrossRef][ISI][Medline]
Vogler D. W. C. Das A. G. Stephenson 1998 Phenotypic plasticity in the expression of self-incompatibility in Campanula rapunculoides. Heredity 81: 546-555[CrossRef][ISI]
Vogler D. W. K. Filmore A. G. Stephenson 1999b Inbreeding depression in Campanula rapunculoides L. I. A comparison of inbreeding depression in plants derived from strong and weak self-incompatibility phenotypes. Journal of Evolutionary Biology 12: 483-494
Vogler D. W. S. Peretz A. G. Stephenson 1999a Floral plasticity in an iteroparous plant: the interactive effects of genotype, environment, and ontogeny in Campanula rapunculoides (Campanulaceae). American Journal of Botany 86: 484-494
Vogler D. W. A. Stephenson 2001 The potential for a mixed mating system Campanula. International Journal of Plant Sciences 162: 801-805[CrossRef]
Willis J. H. 1993 Partial self-fertilization and inbreeding depression in two populations of Mimulus guttatus. Heredity 71: 145-154
Willis J. H. 1999 The contribution of male-sterility mutations to inbreeding depression in Mimulus guttatus. Heredity 83: 337-346
Wilsie C. P. 1958 Effect of inbreeding on fertility and vigor in alfalfa. Agronomy Journal 50: 182-185
Wu H. X. A. C. Matheson D. Spencer 1998 Inbreeding in Pinus radiata. I. The effect of inbreeding on growth, survival, and variance. Theoretical and Applied Genetics 97: 1256-1268[CrossRef][ISI]
Zapata T. R. M. T. K. Arroyo 1978 Plant reproductive ecology of a secondary deciduous tropical forest in Venezuela. Biotropica 10: 221-230
| ||||||||||||||||||||||||||
I, and in both strong and weak SI families, 