| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Department of Botany, University of Toronto, 25 Willcocks St., Toronto, Ontario, Canada, M5S 3B2
Received for publication March 8, 2001. Accepted for publication August 2, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: delayed selfing • enantiostyly • experimental pollinations • floral design • Haemodoraceae • outcrossing rates • spatial analysis • style-morph ratios • Wachendorfia
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Coyne, 1987
; Tuinstra, de Jong, and Scharloo, 1990
). Despite this, conspicuous directional asymmetries in morphology are known from a wide range of organisms (reviewed in Palmer, 1996
). However, the co-occurrence of both asymmetric forms in a population is much less common, although such variation has been reported from various animals, including snails (Johnson, 1982
; Asami et al., 1998
), cichlid fish (Hori, 1993
), and birds (Benkman, 1996
). Directional asymmetry polymorphisms can be used as model systems for asking evolutionary questions concerning their origin and maintenance within populations. The phenotypes (hereafter morphs) are easily recognizable in populations, are often under simple genetic control, and equilibrium morph ratios implicate frequency-dependent selection as the primary mechanism maintaining these asymmetries (Johnson, 1982
; Hori, 1993
; Benkman, 1996
; Asami et al., 1998
). Determining the reproductive systems of species with asymmetric polymorphisms is critical for understanding their evolution and maintenance in nature. This is particularly important in plants in which hermaphroditism and clonal propagation can interfere with outcrossing opportunities (Barrett and Harder, 1996
), thus reducing the intensity of frequency-dependent selection.
Enantiostyly is a plant sexual polymorphism in which the style is positioned to the left (left-styled) or to the right (right-styled) side of the flower. The polymorphism is widely distributed occurring in at least a dozen animal-pollinated angiosperm families in both monocotyledons and dicotyledons (Barrett, Baker, and Jesson, 2000
; Barrett, Jesson, and Baker, 2000
). Enantiostyly has usually been interpreted as a floral design that promotes insect-mediated cross-pollination in a manner functionally analogous to heterostyly (Todd, 1882
; Wilson, 1887
; Robertson, 1890
; Iyengar, 1923
; Ornduff, 1974
; Ornduff and Dulberger, 1978
; Fenster, 1995
). However, the hypothesis that the floral polymorphism promotes proficient cross-pollination between flower types has not been examined experimentally in any detail (although see Bowers, 1975
). Moreover, the occurrence of two distinct forms of the polymorphism (monomorphic and dimorphic enantiostyly—Barrett, Baker, and Jesson, 2000
; Barrett, Jesson, and Baker, 2000
) complicates this simple adaptive explanation.
In monomorphic enantiostyly, both left- and right-styled flowers occur on the same plant, resulting in a form of within-organism antisymmetry (see Palmer, 1996
). In contrast, dimorphic enantiostyly is a genetic polymorphism analogous to directional asymmetry, with populations composed of left- and right-styled plants. The two forms of enantiostyly differ markedly in their abundance and phylogenetic distribution among angiosperm families. Monomorphic enantiostyly is widely distributed, whereas dimorphic enantiostyly is only known from a few species in three monocotyledonous families (for Haemodoraceae; Wilson, 1887
; Ornduff and Dulberger, 1978
; for Pontederiaceae; Barrett, Baker, and Jesson, 2000
; Barrett, Jesson, and Baker, 2000
; and unpublished data; for Tecophilaeaceae; Dulberger and Ornduff, 1980
). Phylogenetic evidence suggests that dimorphic enantiostyly is most likely derived from monomorphic enantiostyly (reviewed in Barrett, Baker, and Jesson, 2000
; Barrett, Jesson, and Baker, 2000
). The rarity of dimorphic enantiostyly implies that there must be strong constraints on its evolutionary origins.
Wachendorfia Burm. (Haemodoraceae) is a small genus of four species of insect-pollinated geophytes native to the Cape Province of South Africa (Helme and Linder, 1992
; Fig. 1). The first report of enantiostyly in Wachendorfia involved W. paniculata (Wilson, 1887
), a common species of open sites in fynbos vegetation. Subsequent research on this species reported 1 : 1 morph ratios from four populations (Ornduff, 1974
) and weakly developed self-incompatibility in which intermorph crosses produced more seeds than self- or intramorph pollinations (Ornduff and Dulberger, 1978
). There have been no detailed studies of the reproductive ecology of the remaining species of Wachendorfia, although both Ornduff (1974)
and Helme and Linder (1992)
mentioned that all species in the genus possess enantiostyly and that the floral morphology of W. brachyandra suggests some degree of autogamy.
|
; Barrett, Baker, and Jesson, 2000
; Barrett, Jesson, and Baker, 2000
). However, reductions in disassortative mating through either selfing or a reliance on clonal propagation are likely to result in deviations from 1 : 1, potentially leading to floral monomorphism and destabilization of the polymorphism. These processes are well known in heterostylous plants, resulting in biased morph ratios and in some cases the breakdown of heterostyly (reviewed in Barrett, 1992
). The major goal of this study was to determine whether similar evolutionary processes might occur in populations of Wachendorfia species. Here we report the first comparative data on the reproductive ecology of an enantiostylous plant group. In this study we employed field surveys, experimental pollinations, and electrophoretic analyses of mating patterns in Wachendorfia species to address the following specific questions: (1) Do the four species differ in aspects of floral biology that are likely to influence mating patterns and hence morph frequencies? (2) What is the relation between floral features predicted to influence selfing and the outcrossing rates of populations? (3) What are the ratios of left- and right-styled plants in natural populations of each species and are differences in their frequency and distribution associated with the reproductive systems of populations? We conclude by discussing the various reproductive factors likely to influence the maintenance of enantiostyly and consider evidence for the evolutionary breakdown of the polymorphism.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). In the remaining largely terrestrial species, only limited clonal propagation occurs through corm splitting. Wachendorfia paniculata is the most abundant of the four species, ranging from Nieuwoudtville to Port Elizabeth in the Cape Province. Wachendorfia thyrsiflora has a similar range, whereas W. parviflora is confined to the western Cape, and the rarest species, W. brachyandra, is restricted to the extreme southwestern Cape (Helme and Linder, 1992
).
Wachendorfia paniculata is a highly variable taxon previously divided into two species and three varieties (Barker, 1950
). More recently, Helme and Linder (1992)
distinguished three distinct forms of W. paniculata. Form 1 is the most common type occurring throughout the species' range. Form 2 is restricted to mountains in northern areas of the western Cape. Form 3 is found along the coast of the Cape Peninsula and southern Cape, especially in exposed, often windy areas. These three forms differ morphologically from one another: Form 1 is the most variable, Form 2 is frequently very tall, with lax panicles and long-thin leaves, and in Form 3, plants are of dwarf stature with very narrow hairy leaves. Forms 1 and 2 both exhibit large flowers with significant spatial separation between anthers and stigma (herkogamy), whereas Form 3 has smaller flowers with less herkogamy.
To determine reproductive attributes of the four species, we measured flower diameter, stigma-anther separation, daily flower number, total flower number, fruit set, and seed set in several populations of each species. Sample sizes and the populations sampled are presented in RESULTS. Pollinators were also recorded whenever they were observed, and their foraging behavior noted.
Compatibility relationships
We performed controlled pollinations in natural populations of W. brachyandra (Oranjekloof) and W. paniculata Form 1 (Vredehoek and Camps Bay) in 1997, and W. thyrsiflora (Oranjekloof) in 1998 to determine the compatibility status of species. Sample sizes for plants and flowers are presented in RESULTS. Three treatments were applied to each style morph: self-pollination, intramorph pollination, and intermorph pollination. All three treatments were applied to flowers on a given plant each day. Cross-pollinations used a single pollen donor for each flower that was pollinated. Prior to pollination, we bagged plants with fine curtain mesh to exclude pollinators and emasculated flowers early each morning, prior to anther dehiscence. Following pollination, flowers were rebagged for a 3–5 d period. Fruits were collected 4 wk later, and while the seeds had not completely matured, it was possible to record whether a capsule was formed (fruit set) and the number of seeds present in each capsule (seed set).
We performed general linear models on seed and fruit set data in S-Plus (Mathsoft, 1997
); capsules that produced no seeds were excluded from the analysis of seed set. Due to violation of assumptions of normality, we analyzed data for W. paniculata seed set using general linear models (GLM) with an inverse link function (Venables and Ripley, 1994
). In W. brachyandra, we performed a logistic GLM, coding seed set as a binary response; capsules with either one or two seeds were coded as 0, and capsules with three seeds were coded as 1. For W. thyrsiflora, we log-transformed the number of seeds and analyzed the data using a Gaussian distribution. Transformation of data did not influence the qualitative results. For all species except W. brachyandra, which exhibited near maximal fruit set, we analyzed fruit set data using logistic regression. We performed pairwise contrasts comparing self- and outcross- and comparing intramorph and intermorph pollinations to examine the effect of treatment on mean seed and fruit set.
Daily variation in sex-organ position
Preliminary observations indicated that stigma-anther separation in Wachendorfia flowers changed during anthesis, which has potential implications for self-pollination. To quantify these morphological changes, we measured the distance from the stigma to all anthers (three) in a flower in W. brachyandra (Oranjekloof), W. paniculata Form 1 (Silvermine), and W. thyrsiflora (Oranjekloof). During the course of a day, we measured one flower on each of eight plants per species at 2-h intervals. We present data for stigma-anther separation for a single anther only. The anther chosen was the closest anther to the stigma. Changes in the distance between the anther and the stigma were analyzed using linear mixed-effects models (Mathsoft, 1997
). Time was treated as a random effect and species as a fixed effect. We used log-likelihood ratios to test for the importance of the fixed effects in the final model.
Measurement of outcrossing rates
To determine multilocus outcrossing rates (tm) in the four species, we performed horizontal starch gel electrophoresis on open-pollinated seeds from two populations each of Wachendorfia brachyandra [Oranjekloof (OK) and Kommetjie (Kom)], W. paniculata Form 1 [Camps Bay (CB) and Kirstenbosch (Kir)], W. paniculata Form 3 [Red Hill (RH), Scarborough (Scar)] and W. parviflora (Red Hill and Kirstenbosch), and one population of W. thyrsiflora (Oranjekloof). We scored 20–28 families sampled from each population for allozyme variability. Eighteen enzyme systems were originally screened for polymorphism, following the methods of Glover and Barrett (1987)
. Three polymorphic enzyme systems (Pgi, Pgm-1, and Pgm-2) were reliably resolved in W. brachyandra, W. parviflora, and W. thyrsiflora, and five enzyme systems were resolved in W. paniculata (Pgi, Pgm-1, Pgm-2, Pgd, and Adh) on a histidine-citrate buffer system (pH = 6.2). Ritland's (1990)
MLTR program was used to estimate tm for each population. Standard errors were calculated as the standard deviation of 1000 bootstraps with the family as the unit of resampling. Differences in outcrossing rate between populations were assessed using pairwise comparisons of bootstrap estimates (Eckert and Barrett, 1994
). Two populations were considered to differ significantly in outcrossing rate if 100(1 – 
PC/2) of the differences between randomly paired bootstrap estimates lay either all above or all below zero (where PC is the Type I error rate). When examining differences in the estimate of outcrossing rate between the four populations of W. paniculata, the experimental error rate (EW) was held at EW = 0.05 using Sidak's correction for nonorthogonal contrasts.
Population surveys
To determine the relative frequencies of left- and right-styled plants in populations of each species of Wachendorfia, we surveyed 31, 14, 5, and 4 populations of W. paniculata, W. thyrsiflora, W. brachyandra, and W. parviflora. These sample sizes reflect the differing abundance of the four species. Only populations of Forms 1 and 3 of W. paniculata were sampled. Pooled goodness-of-fit G tests were conducted to determine whether morph ratios of all populations in a species differed from 1 : 1. G tests of heterogeneity were also performed to examine heterogeneity in morph ratios between populations. Data on style-morph ratios, population sizes, and localities are available on request from L. K. Jesson.
We investigated spatial structuring of morphs in W. brachyandra, W. paniculata, and W. thyrsiflora in either 3 x 3 m or 6 x 6 m plots by mapping the location and morph of each flowering plant. Plot size was dependent on the density of plants in a population. Two populations of W. paniculata Form 1 were mapped: one in the Silvermine Reserve on the Cape Peninsula (6 x 6 m plot) and the other in natural vegetation at Kirstenbosch Botanic Gardens in Capetown (3 x 3 m). Plots for W. brachyandra (6 x 6 m) and W. thyrsiflora (3 x 3 m) were located at Kommetjie and Oranjekloof, respectively. We performed a log-linear analysis to test for independence between the morph of each focal plant and the morph of its nearest neighbor, as well as the distance to the nearest neighbor (Pielou, 1961
).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
There was striking variation among the four species in mean flower diameter. Two species were characterized by large flowers (W. paniculata Form 1 = 38.08 mm, SE = 0.49, N = 88; W. thyrsiflora = 32.02 mm, SE = 0.35, N = 40), while the other two possessed much smaller flowers (W. brachyandra = 19.68 mm, SE = 1.02, N = 40; W. parviflora = 18.31 mm, SE = 0.86, N = 57). Comparisons of mean daily and total flower number per inflorescence also revealed large differences among the four species (Table 1). Wachendorfia brachyandra, W. paniculata, and W. parviflora all had daily flower numbers ranging between 1 and 11 flowers, whereas W. thyrsiflora had a much higher daily flower number, ranging between 1 and 90 flowers. Total flower number of W. thyrsiflora was also considerably higher than the other three species. There was considerable variation between species in natural fruit set, ranging from 45% in W. paniculata at Kirstenbosch to 84% in W. thyrsiflora at Orangekloof. Ovule number in Wachendorfia species is fixed at three (Helme and Linder, 1994
). The mean number of seeds per fruit was similar among the four species, ranging from 2.28 in W. paniculata to 2.6 in W. parviflora (Table 1).
|
|
|
|
|
|
EW = 0.05, PC = 0.017). The two populations of W. parviflora had significantly different outcrossing estimates (proportion of pairwise differences in bootstrap estimates lying above zero = 1), from mixed mating (tm = 0.60) to predominant selfing (tm = 0.07). The single population of W. thyrsiflora examined also exhibited a mixed mating system (tm = 0.76).
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
examined pollination in monomorphically enantiostylous Solanum rostratum and found that pollen from flowers of one style type was more likely to be deposited on flowers of the opposite form. However, because flowers with right- and left-bending styles occur on the same plant in S. rostratum, this condition is not a true genetic polymorphism. The results of our study demonstrate that all four species of Wachendorfia exhibit dimorphic enantiostyly; however, the expression of the polymorphism varies significantly both among species and among populations. This variation appears to be largely governed by differences in the reproductive systems of populations. We now discuss how variation in the mating patterns and degree of sexuality of populations can influence morph ratios and the implications that this variation has for the functioning and evolutionary maintenance of the polymorphism.
Variation in reproductive systems
Our field observations, as well as earlier taxonomic descriptions of Wachendorfia species, suggested that their mating systems might range from largely outcrossing in W. paniculata and W. thyrsiflora to at least partially selfing in W. brachyandra and W. parviflora. Measurements of floral traits tended to support these inferences. Both W. paniculata and W. thyrsiflora have large, showy flowers and inflorescences with large floral displays. In contrast, W. brachyandra and W. parviflora have considerably smaller, less colorful flowers with much reduced daily and total floral displays. Fruit set tended to be higher in populations of W. brachyandra and W. parviflora, a frequent condition in species with a high degree of self-pollination. Lower investment in floral structures that attract pollinators and a greater investment in seeds are common in selfing species (Cruden and Lyon, 1985
; Lloyd, 1987
). These findings suggested at least a partial reliance on selfing in W. brachyandra and W. parviflora.
Our compatibility data supported the prediction that W. brachyandra was at least partially selfing. Controlled crosses revealed that W. brachyandra is completely self-compatible, whereas W. paniculata and W. thyrsiflora display some self-sterility. Unfortunately, because of the early and limited duration of flowering in W. parviflora, we were not able to conduct controlled pollinations on this species, but its small flower size and high selfing rates indicate that, like W. brachyandra, the species is almost certainly highly self-compatible. While there were no differences in fruit set between the different types of pollinations in W. paniculata and W. thyrsiflora, selfing reduced the number of seeds set per fruit. Earlier pollination studies of W. paniculata (Ornduff and Dulberger, 1978
), as well as other members of the Haemodoraceae (Hopper, 1980
), have also shown reduced seed set after selfing. It is not known whether reduced fertility following self-pollination results from self-incompatibility and/or from the selective abortion of selfed embryos due to inbreeding depression (Barrett, 1988
; Charlesworth, 1989
).
Delayed selfing is a floral strategy that optimizes outcrossing in situations with high pollinator visitation but gives reproductive assurance in environments with infrequent pollinator visitation (Lloyd, 1992
). Our measurements of changes in stigma-anther separation during anthesis demonstrated that W. brachyandra exhibits delayed selfing. At the end of the day, the closest anther was in contact with the stigma, resulting in the autonomous self-pollination of flowers. Estimates of outcrossing rate in the two populations of W. brachyandra examined were tm = 0.39 and tm = 0.58, values consistent with a partial reliance on selfing. Since the longevity of Wachendorfia flowers is normally 1 d, populations are vulnerable to pollen limitation if weather conditions in early spring restrict pollinator activity. Indeed, there is good evidence for widespread pollen limitation among insect-pollinated plants in the Cape region (Johnson and Bond, 1997
). Under these circumstances, delayed selfing may be particularly important in overcoming the constraints imposed by short floral longevity.
Contact between stigmas and anthers was also observed in populations of W. parviflora, and outcrossing rates ranged from tm = 0.61 at Kirstenbosch to tm = 0.07 at Red Hill. Flowers of W. parviflora at Red Hill were often not open by mid-afternoon, and the nearest anther was observed to touch the stigma in 25% of the flowers measured. It is possible that the very high selfing rate of this population may be due to poor pollinator service (no pollinators were observed) and the sparse distribution of plants at this particular site. Clearly, given the large difference in outcrossing rate between the two populations of W. parviflora investigated, more work on this species is required. Wachendorfia paniculata and W. thyrsiflora also exhibited a decrease in stigma-anther separation during anthesis; however, the large initial distance between the sex organs (because of their larger flower sizes) prevents contact between stigmas and anthers. It seems likely that in these species, any selfing that does occur results from geitonogamy enhanced by their larger display sizes.
Reproductive systems are influenced both directly and indirectly by the amount of clonal reproduction in populations (Handel, 1985
). In sexually polymorphic species, clonality affects the spatial patterning of morphs with implications for both mating patterns and fertility (Wyatt and Hellwig, 1979
; Barrett and Thomson, 1982
; Eckert, 2000
; Weller, Keeler, and Thomson, 2000
). Large clone size, especially with an abundance of infloresences, may impede cross-pollination resulting in high levels of geitonogamous selfing (Eckert, 2000
). Among the four Wachendorfia species, only W. thyrsiflora exhibits large clone sizes, and the moderate levels of selfing (24%) estimated at Oranjekloof probably result mainly from geitonogamous pollen transfers both within and between inflorescences of clones in this population. Analysis of the spatial distribution of inflorescences in this population illustrates the considerable opportunities that likely exist for inter-inflorescence geitonogamy in W. thyrsiflora (Fig. 7).
Patterns of morph-ratio variation
Equal morph ratios in plant species with polymorphic sexual systems are usually interpreted as resulting from negative frequency-dependent selection operating during the mating cycle (Fisher, 1941
; Charlesworth and Charlesworth, 1979a, b
; Charnov, 1982
; Lloyd and Webb, 1992
). For example, in distylous and tristylous populations, equal morph ratios can be explained by disassortative mating due to the higher proficiency of pollen transfer between rather than within morphs and/or the presence of heteromorphic incompatibility preventing selfing and intramorph mating (reviewed in Barrett, 1992
). However, unequal morph ratios and stylar monomorphism in heterostylous populations can arise for a variety of reasons, including founder effects and drift, clonal propagation, and, in species without heteromorphic incompatibility, morph-specific differences in selfing or assortative mating (e.g., Eckert and Barrett, 1992
; Husband and Barrett, 1992
; Barrett et al., 1997
). By analogy with heterostylous species, two features of the reproductive systems of Wachendorfia species are likely to be particularly important in influencing population morph ratios. These are the absence of a strong self-incompatibility system that prevents selfing and intramorph mating and, in W. thyrsiflora, the occurrence of extensive clonal propagation.
Wachendorfia species exhibit population morph ratios ranging from 1 : 1 to those that are strongly biased or monomorphic for a single style morph (Fig. 6). In all 31 populations of Form 1 of W. paniculata, the two style morphs were not significantly different from the expected 1 : 1 ratio. This stability in morph ratios suggests that most matings in these populations result from outcrossing, with equivalent levels of disassortative mating in the style morphs. Outcrossing rates in the two populations of Form 1 of W. paniculata were 0.86 and 0.98 (Fig. 5) but we have no information on the actual levels of disassortative mating in these populations. Morph-specific markers would enable estimation of this particular mating parameter.
The precise mechanisms maintaining 1 : 1 morph ratios in W. paniculata remain to be determined. While there is some evidence that intramorph crosses have lower fertility than intermorph crosses (Ornduff and Dulberger, 1978
; Table 3 [Fruitset, Vredehoek population]) this effect is relatively weak and not consistent among populations. Thus, it seems unlikely that intramorph incompatibility is the primary mechanism maintaining disassortative mating in populations of W. paniculata. A more attractive hypothesis is that the morphology of enantiostylous flowers promotes more proficient pollen transfer between rather than within the style morphs because of segregated pollen deposition on different sides of the bodies of pollinators (Bowers, 1975
). Field experiments with pollen dyes have provided some evidence that intermorph pollen transfer is more frequent than intramorph transfer in Wachendorfia paniculata (L. K. Jesson and S. C. H. Barrett, unpublished data). However, further field studies on the mechanics of the pollination process, as have been conducted in heterostylous species (reviewed in Lloyd and Webb, 1992
), are needed to confirm that pollinators promote disassortative pollen transfer.
Morph ratios in Form 3 of W. paniculata were strikingly different from Form 1, with biased ratios evident in five of the seven populations sampled (Fig. 6). Flowers of Form 3 are significantly smaller with less stigma-anther separation than Form 1 (Fig. 1; L. K. Jesson and S. C. H. Barrett, personal observation) and it seems probable that the biased morph ratios in Form 3 are associated with increased opportunities for selfing. Selfing is likely to be associated with unequal morph frequencies because it reduces the intensity of disassortative mating. Mean population sizes in Form 3 were significantly smaller than Form 1 and thus the joint effects of stochastic forces and selfing may explain the biased morph ratios. The smaller flowers associated with Form 3 may also lead to a less precise placement of pollen on a pollinator's body, lowering the amount of disassortive mating. However, it is important to note that while outcrossing rates in the two populations of Form 3 that we examined were lower than Form 1, they were not significantly different.
There seems little doubt that the unequal morph ratios and stylar monomorphism in W. brachyandra and W. parviflora, respectively, are associated with the higher selfing rates that characterize these taxa. Species capable of selfing are especially prone to founder events, as single individuals have the capacity to found colonies. These processes probably account for the occurrence of monomorphic populations in W. parviflora (Fig. 6), as well as the strong spatial structuring of morphs within populations of W. brachyandra. Founder events are also important in species that rely heavily on clonal propagation, and the biased morph ratios and stylar monomorphism typical of W. thyrsiflora populations in the Cape Peninsula and southern Cape have most likely arisen in a similar manner. Restrictions on seedling establishment in the aquatic conditions that often characterize clonal populations can result in striking founder events and very long delays before polymorphic equilibria are reached (Eckert and Barrett, 1995
).
The two disjunct populations of W. thyrsiflora that we sampled in the eastern Cape exhibited equal morph ratios. These populations were some 800 km from the remaining populations and the ecological conditions in which they occurred were quite different. Unlike populations from the Cape Peninsula and southern Cape, which grow in river valleys in standing water, the populations from the eastern Cape are more terrestrial in nature occurring along roadsides on damp soil. Under these conditions, it seems probable that sexual recruitment occurs regularly, allowing equilibrium morph ratios to be reached more rapidly. Similar patterns of style-morph frequency variation associated with differences in sexual recruitment among geographical areas also occur in several heterostylous aquatics (Barrett, 1980
; Eckert and Barrett, 1992, 1993
). These cases provide strong evidence for the important role that ecology can play in governing morph ratios in species with polymorphic sexual systems.
Evolutionary maintenance of enantiostyly
The range of variation in style-morph ratios in Wachendorfia suggests that the polymorphism may be difficult to maintain under certain conditions. In monomorphic populations or those with strongly biased morph ratios, selection to maintain the morphology associated with enantiostyly is likely to be relaxed, particularly under conditions that favor increased selfing. Reduced stigma-anther separation, flower size, and floral display in W. brachyandra and W. parviflora may represent a derived condition associated with the evolution of selfing in these species. A phylogenetic analysis of Wachendorfia based on morphological traits found that these taxa were the most derived in the genus (Helme and Linder, 1992
), although molecular data would be necessary to confirm this interpretation because of the difficulty that the selfing syndrome presents for phylogenetic reconstruction (see Graham and Barrett, 1995
).
Breakdown of enantiostyly in Wachendorfia may represent a parallel condition to the well-known evolution of homostyly from heterostyly in many heterostylous families (reviewed in Charlesworth and Charlesworth, 1979a
; Barrett, 1992
). In both cases, outcrossing is more likely to be replaced by selfing under conditions in which disassortative pollen transfer is no longer mediated by pollinators. Elsewhere, among dimorphically enantiostylous species, populations of Heteranthera multiflora (Pontederiaceae) are commonly monomorphic (L. K. Jesson, unpublished data), as are populations of Barbaretta aura (Haemodoraceae, Ornduff, 1974
; S. Johnson, University of Natal, personal communication), and it may be that the loss of effective pollinator service makes dimorphic enantiostyly a difficult floral polymorphism to maintain. Nevertheless, 1 : 1 morph ratios are a general feature of W. paniculata Form 1, indicating that stable morph ratios can be maintained under some circumstances. This stability is likely enhanced by the large population sizes and widespread distribution of this form, which reduce the influence of stochastic forces. However, regular pollinator service, frequent sexual recruitment, and the predominantly outcrossing mating system of this form appear to be the key ingredients in maintaining dimorphic enantiostyly.
|
FOOTNOTES |
|---|
2 Author for reprint requests (phone: 416-978-4151; FAX: 416-978-5878; ljesson{at}botany.utoronto.ca
).
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Barker W. 1950 Haemodoraceae. In R. Adamson and T. Salter [eds.], Flora of the Cape Peninsula, 205–208. Juta, Cape Town, South Africa
Barrett S. C. H. 1980 Sexual reproduction in Eichhornia crassipes (water hyacinth). II. Seed production in natural populations. Journal of Applied Ecology 17: 113-124[CrossRef][ISI]
Barrett S. C. H. 1988 The evolution, maintenance, and loss of self-incompatibility systems. In J. Lovett Doust and L. Lovett Doust [eds.], Plant reproductive ecology: patterns and strategies, 98–124. Oxford University Press, New York, New York, USA
Barrett S. C. H. [ed.] 1992 Evolution and function of heterostyly. Springer-Verlag, Berlin, Germany
Barrett S. C. H. A. M. Baker L. K. Jesson 2000 Mating strategies in monocotyledons. In K. Wilson and D. Morrison [eds.], Systematics and evolution of monocots, vol. 2. 258–269. CSIRO Publishing, Melbourne, Australia
Barrett S. C. H. W. W. Cole J. Arroyo M. B. Cruzan D. G. Lloyd 1997 Sexual polymorphisms in Narcissus triandrus (Amaryllidaceae): is this species tristylous?. Heredity 78: 135-145[CrossRef][ISI]
Barrett S. C. H. L. D. Harder 1996 Ecology and evolution of plant mating. Trends in Ecology and Evolution 11: 73-79
Barrett S. C. H. L. K. Jesson A. M. Baker 2000 The evolution and function of stylar polymorphisms in flowering plants. Annals of Botany 85: 253-265
Barrett S. C. H. J. D. Thomson 1982 Spatial pattern, floral sex ratios, and fecundity in dioecious Aralia nudicaulis (Araliaceae). Canadian Journal of Botany 60: 1662-1670[CrossRef]
Benkman C. W. 1996 Are the ratios of bill crossing morphs in crossbills the result of frequency-dependent selection?. Evolutionary Ecology 10: 119-126
Bowers K. 1975 The pollination ecology of Solanum rostratum (Solanaceae). American Journal of Botany 62: 633-638[CrossRef][ISI]
Charlesworth B. D. Charlesworth 1979a The maintenance and breakdown of distyly. American Naturalist 111: 499-513
Charlesworth D. 1989 Evolution of low female fertility in plants: pollen limitation, resource allocation and genetic load. Trends in Ecology and Evolution 4: 289-292[CrossRef]
Charlesworth D. B. Charlesworth 1979b A model for the evolution of distyly. American Naturalist 114: 467-498[CrossRef][ISI]
Charnov E. L. 1982 The theory of sex allocation. Princeton University Press, Princeton, New Jersey, USA
Coyne J. A. 1987 Lack of response to selection for directional asymmetry in Drosophila melanogaster. Journal of Heredity 78: 119
Cruden R. D. Lyon 1985 Patterns of biomass allocation to male and female functions in plants with different mating systems. Oecologia 66: 299-306[ISI]
Dulberger R. R. Ornduff 1980 Floral morphology and reproductive biology of four species of Cyanella (Tecophilaeaceae). New Phytologist 86: 45-56[CrossRef][ISI]
Eckert C. G. 2000 Contributions of autogamy and geitonogamy to self-fertilization in a mass-flowering, clonal plant. Ecology 81: 532-542[CrossRef][ISI]
Eckert C. G. S. C. H. Barrett 1992 Stochastic loss of style morphs from populations of tristylous Lythrum salicaria and Decodon verticillatus (Lythraceae). Evolution 46: 1014-1029[CrossRef][ISI]
Eckert C. G. S. C. H. Barrett 1993 Patterns of genotypic diversity and clonal reproduction in Decodon verticillatus (Lythraceae). American Journal of Botany 80: 1175-1182[CrossRef][ISI]
Eckert C. G. S. C. H. Barrett 1994 Post-pollination mechanisms and the maintenance of outcrossing in self-compatible Decodon verticillatus (Lythraceae). Heredity 72: 396-411[ISI]
Eckert C. G. S. C. H. Barrett 1995 Style morph ratios in tristylous Decodon verticillatus (Lythraceae): selection versus historical contingency. Ecology 76: 1051-1066[CrossRef][ISI]
Fenster C. B. 1995 Mirror image flowers and their effect on outcrossing rate in Chamaecrista fasciculata (Leguminosae). American Journal of Botany 82: 46-50[CrossRef][ISI]
Fisher R. A. 1941 The theoretical consequences of polyploid inheritance for the mid style form in Lythrum salicaria. Annals of Eugenics 11: 31-38
Glover D. E. S. C. H. Barrett 1987 Genetic variation in continental and island populations of Eichhornia paniculata (Pontederiaceae). Heredity 59: 7-17[ISI]
Graham S. W. S. C. H. Barrett 1995 Phylogenetic systematics of the Pontederiales: implications for breeding-system evolution. In P. J. Rudall, P. J. Cribb, D. F. Cutler, and C. J. Humphries [eds.], Monocotyledons: systematics and evolution, 415–441. Royal Botanic Gardens, Kew, UK
Handel S. N. 1985 The intrusion of clonal growth patterns on plant breeding systems. American Naturalist 125: 367-384[CrossRef][ISI]
Helme N. H. P. Linder 1992 Morphology, evolution and taxonomy of Wachendorfia (Haemodoraceae). Bothalia 22: 59-75
Hopper S. D. 1980 A biosystematic study of the Kangaroo Paws, Anigozanthos and Macropidia (Haemodoraceae). Australian Journal of Botany 28: 659-680[CrossRef]
Hori M. 1993 Frequency-dependent natural selection in the handedness of scale-eating cichlid fish. Science 260: 216-219
Husband B. C. S. C. H. Barrett 1992 Genetic drift and the maintenance of the style length polymorphism in tristylous populations of Eichhornia paniculata (Pontederiaceae). Heredity 69: 440-449[ISI]
Iyengar M. O. 1923 On the biology of the flowers of Monochoria. Journal of the Indian Botanical Society 3: 170-177
Johnson M. S. 1982 Polymorphism for direction of coil in Partula suturalis: behavioural isolation and frequency dependent selection. Heredity 49
Johnson S. D. W. J. Bond 1997 Evidence for widespread pollen limitation of fruiting success in Cape wildflowers. Oecologia 109: 530-534[CrossRef][ISI]
Lloyd D. G. 1987 Allocations to pollen, seeds and pollination mechanisms in self-fertilizing plants. Functional Ecology 1: 83-89
Lloyd D. G. 1992 Self- and cross-fertilization in plants. II. The selection of self-fertilization. International Journal of Plant Sciences 153: 370-380[CrossRef]
Lloyd D. G. C. J. Webb 1992 The selection of heterostyly. In S. C. H. Barrett [ed.], Evolution and function of heterostyly, 180–206. Springer-Verlag, Berlin, Germany
Mathsoft. 1997 S-Plus 4. Data analysis products division. Mathsoft, Seattle, Washington, USA
Maynard Smith J. K. Sondhi 1960 The genetics of a pattern. Genetics 45: 1034-1050
Ornduff R. 1974 Heterostyly in South African flowering plants: a conspectus. Journal of South African Botany 40: 169-187
Ornduff R. R. Dulberger 1978 Floral enantiomorphy and the reproductive system of Wachendorfia paniculata (Haemodoraceae). New Phytologist 80: 427-434[CrossRef][ISI]
Palmer A. R. 1996 From symmetry to asymmetry: phylogenetic patterns of asymmetry variation in animals and their evolutionary significance. Proceedings of the National Academy of Science 93: 14279-14286
Pielou E. 1961 Segregation and symmetry in two-species populations as studied by nearest-neighbour relationships. Journal of Ecology 49: 255-269[CrossRef][ISI]
Ritland K. 1990 A series of FORTRAN computer programs for estimating plant mating systems. Journal of Heredity 81: 235-237[ISI]
Robertson C. 1890 Flowers and insects. V. Botanical Gazette 15: 199-204[CrossRef]
Todd J. E. 1882 On the flowers of Solanum rostratum and Cassia chamaecrista. American Naturalist 16: 281-287[CrossRef]
Tuinstra E. J. G. de Jong W. Scharloo 1990 Lack of response to family selection for directional asymmetry in Drosophila melanogaster: left and right are not distinguished in development. Proceedings of the Royal Society of London B 241: 146-152[Medline]
Venables W. N. B. O. Ripley 1994 Modern applied statistics with S-plus. Springer-Verlag, New York, New York, USA
Weller S. G. K. H. Keeler B. A. Thomson 2000 Clonal growth of Lithospermum caroliniense (Boraginaceae) in contrasting sand dune habitats. American Journal of Botany 87: 237-242
Wilson J. 1887 On the dimorphism of the flowers of Wachendorfia paniculata. Transactions of the Proceedings of the Botanical Society of Edinburgh 17: 73-77
Wyatt R. R. L. Hellwig 1979 Factors determining fruit set in heterostylous bluets, Houstonia caerulea (Rubiaceae). Systematic Botany 4: 103-114
This article has been cited by other articles:
|
|
 |
|
  R. F. Sage, T. L. Sage, R. W. Pearcy, and T. Borsch The taxonomic distribution of C4 photosynthesis in Amaranthaceae sensu stricto Am. J. Botany, December 1, 2007; 94(12): 1992 - 2003. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-Y. GAO, P.-Y. REN, Z.-H. YANG, and Q.-J. LI The Pollination Ecology of Paraboea rufescens (Gesneriaceae): a Buzz-pollinated Tropical Herb with Mirror-image Flowers Ann. Bot., March 1, 2006; 97(3): 371 - 376. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Pauw Inversostyly: a new stylar polymorphism in an oil-secreting plant, Hemimeris racemosa (Scrophulariaceae) Am. J. Botany, November 1, 2005; 92(11): 1878 - 1886. [Abstract] [Full Text] [PDF]
|
|
2-h intervals over the course of 1 d. The slope of the regression of the species is shown by the solid line. For comparison, the slopes of the other species are shown by the dotted (W. paniculata), dotted and dashed (W. brachyandra), or dashed line (W. thyrsiflora)
