Developmental stability in flowers of Clarkia tembloriensis (Onagraceae)

Developmental instability of floral traits is examined in four populations of Clarkia tembloriensis (Onagraceae) with different natural outcrossing rates. Developmental instability is estimated using fluctuating asymmetry (FA) and within plant variance. The results are coupled with those from a previous study of leaf traits. In the first experiment, flowers were collected from the same growth chamber-grown plants that had been previously used to estimate leaf developmental stability in two C. tembloriensis populations. These populations differed in FA for only one floral trait, long filament length. After adjusting for organ size differences, we found floral FA values were about half those of leaves. These are the first quantitative data indicating that flowers are more developmentally stable than leaves. In a second experiment, greenhouse grown plants from two other C. tembloriensis populations (one highly outcrossing and one predominantly self-pollinating) did not differ significantly in floral FA or in within-plant variance of floral traits, though earlier studies of the same populations revealed significant differences in FA of leaf traits. In both experiments, FA values of different floral traits were uncorrelated. We attribute the lack of significant differences in floral stability between populations to the greater canalization of floral organs and to the magnification of measurement error that occurs when calculating FA. We also found that the shorter styles of selfers are the greatest difference in flower form between predominantly self-pollinating and predominantly outcrossing populations of C. tembloriensis.     

Effect of variation in herkogamy on outcrossing within a population of Gilia achilleifolia

The diverse features of floral morphology are often thought to be well-designed mechanisms to manipulate plant mating systems. We evaluated the effectiveness of one such mechanism, anther-stigma separation (herkogamy), in controlling variation in the level of outcrossing among plants in a population of Gilia achilleifolia. Variation in outcrossing rates within populations has the potential to influence the coevolution between inbreeding depression and mating system. Using four polymorphic allozymes, we compared the outcrossing-rate estimates of two groups of individuals under natural conditions: one group with low herkogamy and another with high herkogamy. The high herkogamy group had a higher outcrossing rate (0.572) than the low herkogamy group (0.335). This suggests that the within-population variation in outcrossing rate could potentially cause the previously observed association between herkogamy and inbreeding depression (Takebayashi and Delph, 2000). A previous study of floral traits among G. achilleifolia populations failed to detect a relationship between herkogamy and outcrossing rate, demonstrating that the functionality of traits may be obscured in among-population studies as a consequence of uncontrolled environmental variation. Additionally, the effect of herkogamy on outcrossing rate in delayed selfers such as G. achilleifolia may be particularly prominent when pollinator availability is low. Our population-level estimate of outcrossing rate (0.444) was somewhat lower than an estimate from the same population, 15 years prior to our study (0.75), suggesting that pollinator availability may fluctuate among years. Both within-year and among-year variation in the outcrossing rate may have a strong influence on mating-system evolution.     

Quantitative studies of the mating system in two sympatric species of Ipomoea (Convolvulaceae)

The rates of outcrossing in sympatric populations of Ipomoea purpurea and I, hederacea were estimated (using electrophoretic markers) to be 70% and 7% respectively. The difference in outcrossing rate is not apparently due to differences in pollinator service received by the species, but is associated with differences in anther-stigma distance. In I. purpurea stigmas are generally exserted and there is much genetic variation for anther-stigma distance. Variation in this character has a significant effect on the ease with which selfpollination occurs. In contrast there is no variation for the character in the I. hederacea population, the anthers being invariably held at the same level as the stigma, an arrangement promoting self-pollination.     

Anther-stigma separation is associated with inbreeding depression in Datura stramonium,a predominantly self-fertilizing annual

Abstract.— Genetically based variation in outcrossing rate generates lineages within populations that differ in their history of inbreeding. According to some models, mating-system modifiers in such populations will demonstrate both linkage and identity disequilibrium with fitness loci, resulting in lineage-specific inbreeding depression. Other models assert that differences among families in levels of inbreeding depression are mainly attributable to random accumulation of genetic load, unrelated to variation at mating-system loci. We measured female reproductive success of selfed and outcrossed progeny from naturally occurring lineages of Datura stramonium , a predominantly self-fertilizing annual weed that has heritable variation in stigma-anther separation, a trait that influences selfing rates. Progeny from inbred lineages (as identified by high degree of anther-stigma overlap) showed equal levels of seed production, regardless of cross type. Progeny from mixed lineages (as identified by relatively high separation between anthers and stigma) showed moderate levels of inbreeding depression. We found a significant correlation between anther-stigma separation and relative fitness of selfed and outcrossed progeny, suggesting that family-level inbreeding depression may be related to differences among lineages in inbreeding history in this population. Negative inbreeding depression in putatively inbred lineages may be due in part to additive effects or to epistatic interactions among loci.     

Variation in outcrossing rate and population genetic structure of Clarkia tembloriensis (Onagraceae)

Summary Outcrossing rate estimates for eight accessions of Clarkia tembloriensis indicate that this annual plant species has a wide interpopulational range of outcrossing rate ( ). Populations' t estimates were significantly correlated with observed heterozygosity and mean number of alleles per locus. Estimated fixation indices, , for most populations were very close to their expected values, F_eq, for a given Nei's gene diversity statistics showed that the group of outcrossing populations have more total genetic variation and less differentiation among populations than does the group of selfing populations. These results indicate that the breeding system of C. tembloriensis has had a strong influence on the amount and distribution of genetic variation within and among its populations.     

DEVELOPMENTAL STABILITY IN LEAVES OF CLARKIA TEMBLORIENSIS (ONAGRACEAE) AS RELATED TO POPULATION OUTCROSSING RATES AND HETEROZYGOSITY

Four natural populations of Clarkia tembloriensis, whose levels of heterozygosity and rates of outcrossing were previously found to be correlated, are examined for developmental instability in their leaves. From the northern end of the species range, we compare a predominantly selfing population (t? = 0.26) with a more outcrossed population (t? = 0.84), which is genetically similar. From the southern end of the range, we compare a highly selfing population (t? = 0.03) with a more outcrossed population (t? = 0.58). We measured developmental stability in the populations using two measures of within-plant variation in leaf length as well as calculations of fluctuating asymmetry (FA) for several leaf traits. Growth-chamber experiments show that selfing populations are significantly more variable in leaf length than more outcrossed populations. Developmental instability can contribute to this difference in population-level variance. Plants from more homozygous populations tend to have greater within-plant variance over developmentally comparable nodes than plants from more heterozygous populations, but the difference is not significant. At the upper nodes of the plant, mature leaf length declines steadily with plant age, allowing for a regression of leaf length on node. On average, the plants from more homozygous populations showed higher variance about the regression (MSE) and lower R² values, suggesting that the decline in leaf length with plant age is less stable in plants from selfing populations than in plants from outcrossing populations. Fluctuating asymmetry (FA) was calculated for four traits within single leaves at up to five nodes per plant. At the early nodes of the plant where leaf arrangement is opposite, FA was also calculated for the same traits between opposite leaves at a node. Fluctuating asymmetry is significantly greater in the southern selfing population than in the neighboring outcrossed population. Northern populations do not differ in FA. Fluctuating asymmetry can vary significantly between nodes. The FA values of different leaf traits were not correlated. We show that developmental stability can be measured in plants using FA and within-plant variance. Our data suggest that large differences in breeding system are associated with differences in stability, with more inbred populations being the least stable.     

Morphological and growth rate differences among outcrossing and self-pollinating races of Arenaria uniflora (Caryophyllaceae)

Populations of Arenaria uniflora exhibit intraspecific variation in floral size and degree of protandry in association with the evolution of self-pollination. Heterochrony, or a simple change in the absolute timing or rate of developmental events, is proposed as the evolutionary mechanism underlying the origin of the small, self-pollinating flowers from their large, outcrossing progenitors. Inflorescence growth in two autogamous populations and their related outcrossing progenitors was studied to provide the temporal data necessary for testing the hypothesis of heterochrony. All four races showed significant variation in the growth and mature length of inflorescence organs. Inflorescences of selfing races were smaller, and had slower relative growth rates and a two-fold increase in the plastochron relative to outcrossing populations. The large-flowered races were both significantly protandrous. A more detailed growth analysis of flower development in two races indicated that the selfing flowers develop at a slower rate and for a longer duration relative to outcrossing flowers. The implications of these temporal changes in floral ontogeny for the heterochronic origin of self-pollinating floral forms are considered.     

Heritability of stigma position and the effect of stigma-anther separation on outcrossing in a predominantly self-fertilizing weed, Datura stramonium (Solanaceae)

A polymorphism for anthocyanin production was used as a genetic marker to document the relationship between anther-stigma separation and outcrossing rate in the predominantly self-fertilizing weed Datura stramonium. White-flowered plants that differed in anther-stigma separation were placed into populations consisting exclusively of purple-flowered plants. Self vs. outcross origin of progeny was evident in the hypocotyl color of the seedlings. Outcrossing rates measured for single flowers were significantly positively correlated with anther-stigma separation, albeit with some scatter around the regression line, especially for flowers with exserted stigmas. We also performed an 8 × 8 diallel cross to determine whether anther-stigma separation is genetically determined. Heritability in two field plots was ～0.3 and in the greenhouse was ～0.2. Maternal effects, epistasis, and dominance appeared to be relatively unimportant. Genotypes performed consistently across the three environments, although total plant size varied more than fivefold. It appears that the mixed-mating system of D. stramonium has a heritable basis and would be capable of responding to selection.     

Ecological context of the evolution of self-pollination in Clarkia xantiana: population size, plant communities, and reproductive assurance

The repeated evolutionary transition from outcrossing to self-pollination in flowering plants has been suggested to occur because selfing provides reproductive assurance. Reports from biogeographical and ecological surveys indicate that selfing taxa are often associated with stressful and ephemeral environments, situations in which plant abundance is low (e.g., Baker's law) and with novel plant communities, however experimental tests of ecological hypotheses are few. In this study, we examined the ecological context of selection on mating system traits (herkogamy and protandry) in a California annual, Clarkia xantiana, where natural selfing populations differ from outcrossing populations in that they are often of small size or low density and occur mainly outside the range of pollinator-sharing congeners. We constructed artificial populations of plants with broad genetic variation in floral traits and manipulated two ecological factors, plant population size, and the presence versus absence of pollinator-sharing congeners, in the center of the geographic range of outcrossing populations. We found evidence for context-dependent selection on herkogamy and protandry via female fitness in which reduced traits, which promote autonomous selfing, were favored in small populations isolated from congeners whereas selection was comparatively weak in large populations or when congeners were present. In small, isolated populations, the fertility of plants with low herkogamy or protandry was elevated by 66% and 58%, respectively, compared to those with high herkogamy or protandry. The presence of pollinator-sharing congeners augmented bee visitation rates to C. xantiana flowers by 47% for all bees and by 93% for pollen specialists. By facilitating pollinator visitation, congeners mitigated selection on mating system traits in small populations, where outcross mating success is often low (the Allee effect). We also found support for the hypothesis that pollinator availability directly influenced variation in the strength of selection on herkogamy among populations. The striking parallels between our experimental results and patterns of variation in ecological factors across the geographic range of outcrossing and selfing populations suggest that reproductive assurance may play a central role in directing mating system evolution in C. xantiana.     

THE EVOLUTION OF SELF-POLLINATION IN GRANITE OUTCROP SPECIES OF ARENARIA (CARYOPHYLLACEAE). IV. CORRELATED CHANGES IN THE GYNOECIUM

Intraspecific variation in gynoecial characters was analyzed by SEM for nine populations of Arenaria uniflora, a winter annual endemic to rock outcrops in the southeastern United States. Style lengths in self-pollinating populations, formerly called A. alabamensis, were significantly shorter than in outcrossing populations, and style length showed a very strong correlation (r = 0.97) with outcrossing potential as assessed by degree of protandry. Selfers were also characterized by the production of fewer but longer stigmatic papillae. Papillae extended to the base of the style and diverged from all sides. Outcrossers had longer styles with dense but short papillae, mostly restricted to the upper surface of the style and lacking at its base. Styles also were observed to curl downward in outcrossers. These fine-scale changes appear to enhance the effectiveness of plants as either outcross-pollen or self-pollen receivers.     

VARIATION AND EVOLUTION OF BREEDING SYSTEMS IN THE TURNERA ULMIFOLIA L. COMPLEX (TURNERACEAE)

The evolutionary and functional relationships among breeding systems and floral morphology were investigated in the Turnera ulmifolia complex. Predictions of a model of breeding system evolution among distylous and homostylous varieties were tested. Chromosome counts of 73 accessions revealed an association between breeding system and chromosome number. Diploid and tetraploid populations of five taxonomic varieties are distylous and self-incompatible, whereas hexaploid populations of three varieties are homostylous and self-compatible. The latter occur at different margins of the geographical range of the complex. Crossing studies and analyses of pollen and ovule fertility in F₁'s revealed that the three homostylous varieties are intersterile. To test the prediction that, homostylous varieties are long homostyles that have originated by crossing over within the distyly supergene, a crossing program was undertaken among distylous and homostylous plants. Residual incompatibility was observed in styles and pollen of each homostylous variety with patterns consistent with predictions of the cross-over model. The intersterility of hexaploid varieties suggests that long homostyly has arisen on at least three occasions in the complex by recombination within the supergene controlling distyly. Deviation from expected compatibility behavior occurs in populations of var. angustifolia that have the longest styles. These phenotypes displayed the greatest separation between anthers and stigmas (herkogamy) and set little seed in crosses with long- or short-styled plants. This suggests that they are derived from long homostyles with shorter length styles. It is proposed that selection for increased outcrossing has favored the evolution of herkogamy in long homostyles. Estimates of outcrossing rate in a distylous population using allozyme markers confirmed that dimorphic incompatibility enforces complete outcrossing. Significant genetic variation for floral traits likely to influence the mating system, such as stigma-anther separation, occurs within and among homostylous populations of var. angustifolia on Jamaica. Estimates of the mating system of families from a population with varying degrees of stigma-anther separation, using five isozyme loci, were heterogeneous and ranged from t = 0.04–0.79. Families exhibiting the largest mean stigma-anther separation have higher outcrossing rates than those with little separation.     

Herkogamy and mating patterns in the self-compatible daffodil Narcissus longispathus

BACKGROUND AND AIMS: Floral design in self-compatible plants can influence mating patterns. This study investigated Narcissus longispathus, a self-compatible bee-pollinated species with wide variation in anther-stigma separation (herkogamy), to determine the relationship between variation in this floral trait and the relative amounts of cross- and self-fertilization. METHODS: Anther-stigma separation was measured in the field in six populations of N. longispathus from south-eastern Spain. Variation in herkogamy during the life of individual flowers was also quantified. Multilocus outcrossing rates were estimated from plants differing in herkogamy using allozyme markers. KEY RESULTS: Anther-stigma separation varied considerably among flowers within the six populations studied (range = 1-10 mm). This variation was nearly one order of magnitude larger than the slight, statistically non-significant developmental variation during the lifespan of individual flowers. Estimates of multilocus outcrossing rate for different herkogamy classes (t(m) range = 0.49-0.76) failed to reveal a monotonic increase with increasing herkogamy. CONCLUSIONS: It is suggested that the lack of a positive relationship between herkogamy and outcrossing rate, a result that has not been previously documented for other species, could be mostly related to details of the foraging behaviour of pollinators.     

Variation in fecundity and other reproductive traits in freshwater mussels

1. Life histories of the highly diverse and endangered North American freshwater mussel fauna are poorly known. We investigated reproductive traits of eight riverine mussel species in Alabama and Mississippi, U.S.A.: Amblema plicata, Elliptio arca, Fusconaia cerina, Lampsilis ornata, Obliquaria reflexa, Pleurobema decisum, Quadrula asperata and Q. pustulosa, and compare our results with existing life history information for other species. 2. These eight species had reproductive traits characteristic of large, outcrossing populations: hermaphrodites were rare, we found no evidence of protandry, and sex ratios were even or slightly male‐biased. 3. Age at sexual maturity varied among species, ranging from <1 to 2 years for L. ornata to 3–9 years for Q. asperata. In all species, most mature females participated in reproduction and fertilisation success was high. 5. Fecundity was related positively to both length and age, but length was the best predictor. In six species, fecundity increased exponentially with increasing size; in two species the rate of increase in fecundity declined in larger animals. In four species, fecundity declined in older animals. These latter results indicate weak reproductive senescence; however, in all species, older individuals continued to produce large numbers of offspring. Mean annual fecundity differed widely among species ranging from 9647 to 325 709. Within‐species differences in fecundity were found among rivers and among populations within a river. 6. The wide variation in reproductive traits among species indicates the existence of widely divergent life history strategies in freshwater mussels.     

BREEDING SYSTEM EVOLUTION IN LEAVENWORTHIA: BREEDING SYSTEM VARIATION AND REPRODUCTIVE SUCCESS IN NATURAL POPULATIONS OF LEAVENWORTHIA CRASSA (CRUCIFERAE)

Populations of Leavenworthia crassa (Cruciferae) studied for 3 years exhibited among- and within-population genetic variation for a suite of floral and reproductive traits (flower width, petal length, anther position, ability to set seeds in the absence of pollinators, time to first flowering) associated with breeding system. We used electrophoretic markers to show that a population with small, monomorphically colored flowers with introrse anthers had a significantly lower outcrossing rate (t = 0.03) than did a population with larger, polymorphically colored flowers with extrorse anthers (t = 0.33). In the more-outcrossing population the correlation between higher maternal plant outcrossing rate and the suite of six traits approached significance (P < 0.067), with greater petal size, greater flower width, and reduced ability to set seeds in the absence of pollinators contributing significantly. Plants in selfing populations had a generally higher reproductive success, with a higher number of flowers per plant, a smaller proportion of unfertilized ovules, a smaller proportion of fertilized ovules aborted, a higher rate of fruit set, and overall a larger number of seeds matured than did plants from the more outcrossing populations. Pollen limitation did not appear to account for lower reproductive success in outcrossing populations. Resource limitation did not differ substantially between populations. However, within-ovary patterns of fertilization, abortion, and seed weight were significantly less random in outcrossing populations than in selfing populations, suggesting that differential gamete and embryo success may be responsible for lower reproductive success in outcrossing populations.     

QUANTITATIVE VARIATION IN PHLOX: COMPARISON OF SELFING AND OUTCROSSING SPECIES

Variation of 20 quantitative characters was examined within and among 10 populations of the predominantly outcrossing Phlox drummondii and 4 populations of the predominantly selfing P. cuspidata grown in a greenhouse. Multivariate analysis of variance, considering all characters simultaneously, indicated that there were significant differences among populations in both species while analysis of individual characters demonstrated that there were significant population differences for 19 characters in P. drummondii and 13 characters in P. cuspidata. On average, 16% of the total phenotypic variation in P. drummondii occurred among populations compared to less than 4% of the total variation in P. cuspidata. In addition, P. drummondii exhibited significant differences among families within populations more frequently than P. cuspidata. Most observed variation in both species occurred within families where environmental and genetic sources of variation could not be partitioned. There was a trend for P. drummondii to have higher heritabilities than P. cuspidata for most characters even when assumptions about breeding systems were relaxed. Thus, the outbreeding species exhibited larger genetic differences among populations and among families within populations than the selfing species in the greenhouse environment. These data suggest that P. drummondii has the greater evolutionary potential of the two species and are consistent with the hypothesis that differences in population structure result from differences in the breeding systems of the two species.     

REPRODUCTIVE ASSURANCE MECHANISMS IN THREE TAXA OF THE MIMULUS GUTTATUS COMPLEX (SCROPHULARIACEAE)

The evolution of inbreeding in plants has often been attributed to selection for the ability to set seed in the absence of mates or pollinators. Mechanisms of reproductive assurance in five populations of mixed mating Mimulus guttatus, three populations of inbreeding M. platycalyx, and two populations of inbreeding M. nasutus were examined in a pollinator-free greenhouse. Reproductive assurance was manifested in all populations by autofertility, vegetative reproduction, or both. The inbreeding taxa had significantly greater levels of autofertility and less vegetative reproduction. Three modes of autofertility were identified: 1) due to corolla abscission only, occurring in three M. guttatus populations; 2) due to both corolla abscission and direct anther-stigma contact by curling of the lower stigmatic lobe into the anthers, occurring in two M. guttatus populations; and 3) direct stigma-anther contact by stigma curling alone prior to corolla abscission, found in each M. platycalyx and M. nasutus population. Stigma-anther distance and its interaction with stigma curling contributed to differences in autofertility among populations. Significant levels of intrapopulation quantitative genetic variation were found for seven of ten traits examined; average levels were similar between inbreeding and mixed mating populations. Genetic variation within populations for autofertility per se was not detected, but significant levels controlling stigma-anther distance were found in two M. guttatus populations. These results show that evolution of inbreeding by natural selection for reproductive assurance is possible in Mimulus, and illustrate the complex changes in floral dynamics and morphology it may involve.     

PHENOTYPIC PLASTICITY IN THE SAILFIN MOLLY,POECILIA LATIPINNA (PISCES: POECILIIDAE). I. FIELD EXPERIMENTS

Sailfin mollies (Poecilia latipinna) display marked interdemic variation in body size. We employed “common-garden” experiments in field enclosures to explore the potential role of environmental factors in determining the interdemic phenotypic variation in growth rate, age at maturity, and size at maturity. The largest single, consistent source of variation for all traits was family identity within populations. Environmental effects acted predominantly through family x environment interactions. There was little evidence for any intrinsic variation among populations once family heterogeneity had been accounted for. In general, when statistically significant differences existed, fish raised in a saltwater pond grew faster than their broodmates raised in a freshwater pond. Both males and females tended to mature at a smaller size and later in the freshwater pond than in the saltwater pond. The effects of the environmental conditions differed among the three years in which we performed these studies. In only one year was there a substantial difference between fish raised under the two environmental conditions. These results indicate that direct environmental effects are not strong enough to account for the differences in body size among natural populations and that intrinsic differences among natural populations are due to different frequency distributions of genotypes that are present in all populations.     

Evolutionary consequences of ecological factors: pollinator reliability predicts mating‐system traits of a perennial plant

The reproductive‐assurance hypothesis predicts that mating‐system traits will evolve towards increased autonomous self‐pollination in plant populations experiencing unreliable pollinator service. We tested this long‐standing hypothesis by assessing geographic covariation among pollinator reliability, outcrossing rates, heterozygosity and relevant floral traits across populations of Dalechampia scandens in Costa Rica. Mean outcrossing rates ranged from 0.16 to 0.49 across four populations, and covaried with the average rates of pollen arrival on stigmas, a measure of pollinator reliability. Across populations, genetically based differences in herkogamy (anther–stigma distance) were associated with variation in stigmatic pollen loads, outcrossing rates and heterozygosity. These observations are consistent with the hypothesis that, when pollinators are unreliable, floral traits promoting autonomous selfing evolve as a mechanism of reproductive assurance. Extensive covariation between floral traits and mating system among closely related populations further suggests that floral traits influencing mating systems track variation in adaptive optima generated by variation in pollinator reliability.     

Modification of flower architecture during early stages in the evolution of self-fertilization

Background and Aims

The evolution of selfing from outcrossing is characterized by a series of morphological changes to flowers culminating in the selfing syndrome. However, which morphological traits initiate increased self-pollination and which are accumulated after self-fertilization establishes is poorly understood. Because the expression of floral traits may depend on the conditions experienced by an individual during flower development, investigation of changes in mating system should also account for environmental and developmental factors. Here, early stages in the evolution of self-pollination are investigated by comparing floral traits among Brazilian populations of Eichhornia paniculata (Pontederiaceae), an annual aquatic that displays variation in selfing rates associated with the breakdown of tristyly to semi-homostyly.

Methods

Thirty-one Brazilian populations under uniform glasshouse conditions were compared to investigate genetic and environmental influences on flower size and stigma–anther separation (herkogamy), two traits that commonly vary in association with transitions to selfing. Within-plant variation in herkogamy was also examined and plants grown under contrasting environmental conditions were compared to examine to what extent this trait exhibits phenotypic plasticity.

Key Results

In E. paniculata a reduction in herkogamy is the principal modification initiating the evolution of selfing. Significantly, reduced herkogamy was restricted to the mid-styled morph and occurred independently of flower size. Significant genetic variation for herkogamy was detected among populations and families, including genotypes exhibiting developmental instability of stamen position with bimodal distributions of herkogamy values. Cloned genets exposed to contrasting growth conditions demonstrated environmental control of herkogamy and genotypic differences in plasticity of this trait.

Conclusions

The ability to modify herkogamy independently of other floral traits, genetic variation in the environmental sensitivity of herkogamy, and the production of modified and unmodified flowers within some individuals, reveal the potential for dynamic control of the mating system in a species that commonly confronts heterogeneous aquatic environments.Key words: Eichhornia paniculata, expressivity, flower morphology, herkogamy, phenotypic plasticity, pleiotropy, population variation, self-fertilization, stigma–anther separation, outcrossing, tristyly     

Comparative analysis of late floral development and mating-system evolution in tribe Collinsieae (Scrophulariaceae s.l.)

Species of Collinsia and Tonella, the two sister genera of self-compatible annuals that constitute tribe Collinsieae, show extensive variation in floral size and morphology and in patterns of stamen and style elongation during the life of the flower (anthesis). We used a nuclear ribosomal ITS phylogeny, independent contrasts, and phylogenetically corrected path analysis to explore the patterns of covariance of the developmental and morphological traits potentially influencing mating system. Large-flowered taxa maintain herkogamy (spatial separation of anthers and stigmas) early in anthesis by differential elongation of staminal filaments, which positions each of the four anthers at the tip of the "keel" upon dehiscence. Small-flowered taxa do not show this pattern of filament elongation. The styles of large-flowered taxa elongate late in the 2-5 d of anthesis, resulting in late anther-stigma contact and delayed self-pollination. Anther-stigma contact and self-pollination occur early in anthesis in small-flowered species/populations. Thus, we found complex covariation of morphological and developmental traits that can be interpreted as the result of multitrait adaptation for early selfing and high levels of autogamy, delayed selfing and higher levels of outcrossing, or intermediate levels of outcrossing. Continuous variation in these traits suggests the operation of continuous variation in selective optima or the combined effects of divergent selection and phylogenetic inertia.