| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
2Department of Biological Sciences, University of Pittsburgh, Pittsburgh, Pennsylvania 15260 USA; 3Pymatuning Laboratory of Ecology, Linesville, Pennsylvania 16424 USA
Received for publication September 19, 2003. Accepted for publication February 17, 2004.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: floral traits • Impatiens capensis • mating system • plant–pollinator interactions • pollinator fauna • pollinator visitation • transgenerational effects • vegetative herbivory
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Understanding the causes and consequences of shifts in mating system is of primary importance to evolutionary biologists because changes in mating patterns can have profound effects on the reproductive fitness of individuals (Charlesworth and Charlesworth, 1987
), the genetic variation within populations (Hamrick and Godt, 1990
), and speciation (Barrett, 1990
). Recently much attention has been paid to how stochasticity in pollinator availability affects plant mating system (e.g., Eckert and Schaefer, 1998
; Vogler and Kalisz, 2001
; Culley, 2002
) and how herbivory influences plant–pollinator interactions and plant fitness (e.g., Schemske and Horvitz, 1988
; Strauss et al., 1996
; Krupnick et al., 1999
; Mothershead and Marquis, 2000
; Adler et al., 2001
). However, there is a void in our current understanding of how vegetative herbivory (hereafter, herbivory) influences mating system.
Herbivory can affect plant mating system at least two ways. First, by reducing plant resources, herbivory may have direct consequences on mating system. Resource limitation caused by herbivory can affect flower production (e.g., Quesada et al., 1995
; Lehtila and Strauss, 1997
; Mothershead and Marquis, 2000
), flowering phenology (Juenger and Bergelson, 1997
; Agrawal et al., 1999
), and seed mass and number (e.g., Stephenson, 1981
; Koptur et al., 1996
; Agrawal, 2001
). Herbivory can influence traits that reflect the quality of offspring, including progeny size, reproduction, and herbivore resistance; thus, it can have transgenerational consequences (Agrawal, 2001
). Despite the evidence that maternal herbivory generally can affect offspring traits, there is little information on whether such stress can differentially affect the production and/or vigor of selfed vs. outcrossed progeny. In one of the only studies to investigate the direct effect of leaf damage on mating system, plants with greater fungal damage had reduced production of selfed relative to outcrossed progeny, and the former had lower survival (Levri and Real, 1998
). In addition, in plants that produce heteromorphic flowers—large, biotically pollinated (chasmogamous, CH) and small, self-pollinated (cleistogamous, CL) flowers—herbivory may directly alter the mating system by changing the relative production of these flower types. Specifically, if herbivory reduces resources, we expect a decline in the more costly CH flowers, resulting in a shift in the mating system toward selfing.
Second, herbivory can affect mating system via its effects on floral display and subsequent pollinator visitation. For instance, leaf damage can reduce the number of simultaneously open flowers on a plant (Strauss et al., 1996
; Strauss et al., 1999
; Elle and Hare, 2002
) and, thus, decrease the potential for pollinators to affect geitonogamy (selfing among flowers on a plant) (Harder and Barrett, 1995
). Herbivory can also reduce flower morphology and reward, which in turn may reduce pollinator visitation (Strauss et al., 1996
; Mothershead and Marquis, 2000
) and may increase autogamy in plants capable of this mode of selfing. Moreover, because floral morphology largely determines the composition of the pollinating fauna (Baker and Hurd, 1968
), herbivory-induced changes in floral phenotype could influence the composition of the pollinator pool. However, because we have only begun to understand how herbivory alters floral traits and the abundance and composition of the pollinator fauna, we are ill-equipped to address its impact on mating system.
Because the direct and indirect effects of herbivory may have opposing influences on mating system (i.e., they may increase and decrease the selfing rate) they need to be addressed in concert. In this study, we addressed these issues in Impatiens capensis, which produces heteromorphic flowers. We conducted a survey of vegetative herbivory and mating system in natural I. capensis populations and performed a controlled field experiment in which we artificially damaged leaf tissue to address the following questions: (1) Does leaf damage lead to greater production of CL relative to CH flowers and seeds in I. capensis? (2) Does leaf damage decrease floral display size and, thus, the potential for geitonogamous selfing in CH flowers? (3) Does leaf damage alter the size or shape of CH flowers? (4) Does leaf damage alter the abundance or composition of insects visiting CH flowers? (5) Does leaf damage affect the quality of CL or CH progeny or their contribution to total plant fitness?
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). The CH flowers are orange and produce copious nectar that attracts various species of bees and occasionally the ruby-throated hummingbird (Rust, 1977
). These flowers are self-compatible, but strong protandry prevents autogamy. Outcrossing rates for CH flowers have been estimated to range between 0.29 and 0.71 (Waller and Knight, 1989
). In contrast, the obligately self-pollinating CL flowers have reduced petals, anthers, and sepals and lack nectaries. Cleistogamous and chasmogamous flowers are easily distinguished by their positions on the plant and pedicel structure (Schemske, 1978
).
A variety of herbivores feed on I. capensis (see Schemske, 1978
). Vegetative damage at the populations studied here was caused by chrysomelid beetles, grasshoppers, leaf miners, aphids, and occasionally by white-tailed deer (J. A. Steets, personal observation).
Survey of populations
To determine whether there is a relationship between mating system and herbivory, we surveyed 10 natural populations in Crawford and Lawrence Counties in northwestern Pennsylvania, USA. In each population, we collected one I. capensis plant at 5-m intervals along a transect, for a total of 20 plants. We measured total number of leaves and number of leaves damaged by herbivores. As in other studies (Waller, 1980
; Le Corff, 1993
; Gross et al., 1998
), we estimated total reproduction by counting the number of CL and CH pedicels. In I. capensis, flower production is a good predictor of fruit production for both flower types (CL fruits = 0.55 x CL flowers + 0.14, P < 0.0001, r2 = 0.76, N = 23; CH fruits = 0.57 x CH flowers –0.70, P < 0.0001, r2 = 0.81, N = 23), therefore, we estimated plant mating system as the proportion of CL flowers. Because the plants experienced a killing frost a few days after our collections, the estimates of CL and CH flower production represent total lifetime reproduction.
For each population, we calculated the mean proportion of CL flowers and mean proportion of leaves damaged. To explore the relationship between the proportion of CL flowers and leaf damage, we fit the data to two functions: (1) a linear function (PROC REG, SAS Institute, 2001
) and (2) an exponential function reaching an asymptotic maximum at 
1 (PROC NLIN, SAS Institute, 2001
). This latter function was chosen because the mating system metric, proportion of CL flowers, is bounded between 0 and 1.
Experimental manipulation of leaf damage
To explore the effects of leaf damage on plant mating system, pollinator visitation, and composition of pollinator fauna, we transplanted 50 seedlings from each of three wild populations in Crawford County, Pennsylvania, USA, into 11.4-cm2 pots of Fafard #4 soil (Conrad Fafard, Agawam, Massachusetts, USA). Seedlings were situated along an edge of a wooded area at the Pymatuning Laboratory of Ecology, Crawford County, Pennsylvania, USA. We matched plants from each population according to their height, preexisting CL flowers, and leaf damage. One member of each pair was randomly assigned to the damaged treatment. The other served as an undamaged control. After the initial assignment of treatments, plants were not treated as paired individuals. All plants were intermixed in a single random block and watered twice daily. Although natural herbivory was not controlled, every week plants in the damage treatment had an additional 50% of leaf area removed from all new leaves by manually clipping each leaf in half (perpendicular to the midrib) with scissors. Although this level of damage is high, it is still within the range of damage observed in wild populations (<1–75% of leaf tissue removed per plant; J. A. Steets, personal observation).
Effect of leaf damage on plant size and mating system
We measured plant size (height, stem diameter at first internode, number of leaves, nodes, and branches) and CL and CH flower production at the beginning (prior to treatment application) and end (3 d prior to a killing frost) of the experiment. In addition, 1–5 CL and CH seed pods were collected from a subset of the plants (see below: Effect of leaf damage on progeny), and the mean number of seeds within each type of pod was determined. For each flower type, we estimated seed production as the product of the mean seeds per pod and flower number. The mating system of each plant was calculated as the proportion of CL flowers and seeds.
We analyzed data on initial (plant size, number of CL and CH flowers) and final (plant size, number of CL and CH flowers, and proportion CL flowers) traits using multivariate analyses of variance (PROC GLM, MANOVA option, SAS Institute, 2001
) with treatment, population, and their interaction designated as fixed effects, followed by individual ANOVAs on each variable. Population was treated as a fixed effect in all analyses because the three populations were located very close to one another and, therefore, do not represent a random sample of all I. capensis populations. To determine if damage affected seeds per pod or proportion of CL seeds per plant, we performed ANOVAs with treatment, population, and their interaction as fixed effects. Proportion of CL seeds per plant and seed production per pod were not analyzed in the MANOVA on final vegetative and reproductive traits because we lacked seed production data for some individuals. Proportion CL flowers and seeds were arcsine transformed to improve normality (Zar, 1999
).
Effect of leaf damage on floral display and geitonogamy
We counted the number of open CH flowers per plant (i.e., CH floral display size) and noted their sexual phase on three dates during peak CH flowering. We calculated the mean daily CH floral display for each plant and analyzed this data using ANOVA with treatment, population, and their interaction as fixed effects. To determine if damage affected the potential for geitonogamous self-pollination, plants were categorized as follows: (1) no potential for geitonogamy; defined as one open CH flower or more than one open CH flower but all flowers in the same sexual phase; and (2) potential for geitonogamy; defined as more than one CH flower open and flowers in both sexual phases. Data were analyzed using log-likelihood goodness-of-fit G tests (Zar, 1999
), where the null hypothesis was equal potential for geitonogamy between damaged and undamaged plants. No heterogeneity was detected among survey dates; therefore, only the analysis of pooled data is reported.
Effect of leaf damage on floral traits
For plants that had CH flowers open on collection days, 1–4 CH flowers on the central axis were measured for nectar spur length, lateral petal length and width, flower opening length and width, porch (side) petal length, and androecium/gynoecium length. When more than one flower was measured per plant, we used the mean of these in the analysis. Because floral traits can be highly correlated, we performed a principal components analysis (PROC FACTOR, SAS Institute, 2001
) on the correlation matrix of the seven floral traits. This allowed us to create several orthogonal composite variables, the first two of which describe flower size and shape (see Results). We explored the effect of leaf damage on these using ANOVAs with treatment, population, and their interaction as fixed effects.
Effect of leaf damage on pollinator abundance and composition
We observed pollinator visitation on three days during peak CH flowering and pollinator activity (1000–1600 hours). As commonly occurs in I. capensis wild populations, flowers of neighboring plants intertwine; thus, rather than separating plants, we observed flowers in both treatments simultaneously by observing small patches composed of 10–30 flowers. During 8 h of observation (24 20-min sessions) we recorded a total of 441 visits. Visitation by bumble bees, honey bees, and miscellaneous small bees were recorded separately. We analyzed all data using log-likelihood goodness-of-fit G tests (Zar, 1999
), in which the null hypothesis was equivalent visitation to flowers of damaged and undamaged plants. No heterogeneity was detected among observation days; therefore, only the analysis of pooled data is reported. Tests of heterogeneity were performed to determine if visitation pattern differed among pollinator types (bumble bees, honey bees, and small bees).
Effect of leaf damage on progeny
To assess the potential for transgenerational effects of herbivory, we collected CL and CH seeds from damaged and undamaged plants late in the season. As a result of early season mortality among the experimental plants (22 out of 150 plants died), we were unable to collect seeds from all individuals. In total, 377 CL seeds from 100 plants and 541 CH seeds from 113 plants were collected. Seeds were stored in distilled water in cell culture trays at 4°C for approximately 4 mo to break dormancy (Leck, 1979
). Once germinated, seeds were planted in 96-well plug trays with Fafard #4 soil (Conrad Fafard) and moved to a growth chamber at 5°C with an 8-h daylength to simulate early spring germination conditions. When seedlings emerged and cotyledons reflexed, we estimated cotyledon size as the product of length and width of the largest cotyledon. Two months after the first seed germinated, we harvested, dried, and weighed aboveground biomass for one CL and one CH seedling per maternal plant. We calculated percentage germination, mean number of days to germination, and mean cotyledon size for CL and CH seedlings from each maternal plant. To determine if maternal damage affected percentage germination or dry biomass, we performed individual ANOVAs with maternal treatment, population, flower type, and their interactions as fixed effects. In the biomass analysis germination date was included as a covariate. To determine if maternal damage affected days to germination or cotyledon size, we performed a MANOVA with maternal treatment, population, flower type, and their interactions as fixed effects, followed by individual ANOVAs on each trait.
Effect of leaf damage on cumulative female fitness
We calculated the cumulative fitness from cleistogamy and chasmogamy as the product of flower production, seeds per fruit, percentage germination, and seedling dry biomass for each flower type. The sum of these two products reflects total cumulative plant fitness. Although these cumulative fitness functions falsely assume full fruit set, they are still useful parameters for comparing treatment effects because flower production is a good estimator of fruit production in I. capensis (see above: Survey of populations). To determine if leaf damage affected total cumulative fitness, cumulative fitness from cleistogamy or chasmogamy, or the proportion of fitness derived from CL progeny, we performed ANOVAs with treatment, population, and their interaction designated as fixed effects. The proportion of fitness from CL progeny was arcsine transformed to improve normality (Zar, 1999
).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
|
Effect of leaf damage on floral traits
Combined, the first two principal components explained 68% of the variance in floral traits among plants (Table 2). The first principal component (PC1) explained 54% of floral trait variance and had large positive associations with all flower dimensions and, thus represents overall flower size. The second principal component (PC2) explained an additional 14% of floral trait variance and had large positive correlations with spur length and androecium/gynoecium size, but negative ones with lateral petal dimensions, thus it reflects floral shape. The remaining principal components explained less than 12% each and were not investigated further. Damage significantly reduced overall flower size by 197% (F1,65 = 26.60; P < 0.0001). This reduction reflects a 10–15% reduction in most flower dimensions (data not shown). There was no significant effect of population or an interaction between population and damage with respect to PC1 (F1,65 = 2.07, 1.63; P = 0.13, 0.20, respectively). Damage, population, and their interaction did not explain a significant amount of the variation in flower shape as reflected by PC2 (F1,65 = 1.87, 1.66, 0.20; P = 0.18, 0.20, 0.82, respectively).
|
|
Effect of leaf damage on cumulative female fitness
Damage reduced cumulative female fitness by 37% (F1,65 = 5.36; P = 0.024). Further, the cumulative fitness from chasmogamy was significantly reduced by damage, whereas the fitness derived from cleistogamy was not significantly affected by leaf damage (47% reduction vs. 23% reduction; F1,85 = 7.65, F1,89 = 2.53; P = 0.007, 0.11, respectively). As a result, the overall effect of damage was to increase the proportion of fitness achieved through CL progeny (0.66 ± 0.04 vs. 0.55 ± 0.04; F1,65 = 4.23; P = 0.04).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
) or the fact that these plants were grown in pots. Nevertheless, both experiments show that plants respond to herbivory by decreasing production of outcrossing flowers.
The results of this study support other investigations that have found that mixed-mating systems can be modified by environmental factors. For instance, low soil moisture (Schemske, 1978
; Waller, 1980
; Bell and Quinn, 1987
), low light intensity (Schemske, 1978
; Waller, 1980
; Bell and Quinn, 1987
), low nutrient availability (Le Corff, 1993
), and high plant density (Lu, 2000
) all decrease CH flower production but not CL flower production. Taken together, these data indicate that not only abiotic, but also biotic stress can result in a shift in mating system toward greater selfing. Furthermore, our ongoing research suggests that this herbivory response may be quite general and that other forms of antagonism (i.e., deer grazing, intraspecific competition) can have more severe effects on mating system than insect feeding (J. A. Steets and T.-L. Ashman, unpublished data).
This study is among the first to demonstrate that leaf damage has significant effects on plant mating system (see also Levri and Real, 1998
; Elle and Hare, 2002
). Levri and Real (1998)
found that fungal leaf damage altered the mating system of Kalmia latifolia; diseased plants produced proportionally more outcrossed progeny because they set a lower proportion of selfed fruits relative to outcrossed fruits and produced selfed seedlings with lower survival than outcrossed seedlings. Our study adds to the findings of Levri and Real (1998)
in that leaf damage can also alter mating system at the level of flower production in plants with heteromorphic flowering systems. We found that damaged plants will produce the small, selfing flowers over the more expensive CH flowers. Based on the findings of Levri and Real (1998)
, selective abortion of selfed fruits may offset the increase in proportion CL flower production found here. We are currently investigating this hypothesis.
Both the within-generation (i.e., relative seed production) and between-generation (i.e., relative seedling vigor) effects of leaf damage on mating system may have consequences for demography and population genetic diversity. For example, we found that damage reduced the biomass of CL seedlings more than that of the CH seedlings. Because initial plant size is an important determinant of survival in dense stands of I. capensis (Schmitt et al., 1987
), our results suggest that the smaller, CL seedlings of damaged plants may have higher mortality than the larger CH seedlings or either seedling type from undamaged plants. Thus, a population experiencing moderate levels of herbivory may have higher genetic diversity as a result of increased selfed seedling mortality. Further, if herbivore damage changes other demographic parameters (e.g., fecundity) of selfing and outcrossing progeny, then their contribution to population growth rate may also change. To address this hypothesis, life table response experiments are needed (Caswell, 2001
).
Indirect effects of leaf damage
Leaf damage indirectly affected some traits associated with mating system (CH floral traits and pollinator visitation), but not others (CH floral display, potential for geitonogamy, and composition of the pollinator fauna). Our findings add to the growing body of work demonstrating that herbivory decreases flower production, floral attraction, pollinator visitation, and female reproductive success (e.g., Strauss et al., 1996
; Mothershead and Marquis, 2000
). These effects might be expected to result in changes in the outcrossing rate of CH flowers because pollinator type, fit, and efficiency are believed to affect pollen removal and deposition. However, a test of this idea requires direct estimation of CH flower outcrossing rate in damaged and undamaged plants. Such work is currently underway in our laboratory.
In the only other study to investigate the effect of herbivory on mating system, Elle and Hare (2002)
reported that leaf damage reduced floral display and selfing rate of Datura wrightii, the latter of which was likely caused by a decrease in geitonogamy. Conversely, we found that damage had no significant effect on CH floral display size and potential for geitonogamy in I. capensis. The contrasting findings between our study and those of Elle and Hare (2002)
are likely due to the extreme differences in floral display size of the study organisms. Datura wrightii can produce over 100 flowers in a single night (Elle et al., 1999
), whereas in the present study, I. capensis produced under five CH flowers per day because resources were limiting in pots. However, in natural populations of I. capensis, removal of natural insect herbivores results in a large increase in CH floral display (J. A. Steets and T.-L. Ashman, unpublished data), suggesting that geitonogamy may contribute to mating system in wild populations of this species.
If we wish to understand how leaf damage affects a whole plant mating system, we need to consider its effects on the proportional production of CL and CH flowers, fruits, and seeds as well as on CH outcrossing rate. Based on the mean CL and CH flower and seed production (Table 1) and an average estimate of 0.5 for CH outcrossing rate taken from the literature (Waller and Knight, 1989
), we estimate that undamaged plants have an overall plant outcrossing rate of 0.26, whereas damaged plants have an outcrossing rate of 0.20 in the absence of any indirect effects. However, if the smaller CH floral display size, lower potential for geitonogamy, and greater proportion of small bee pollinators translate into reduced actual geitonogamy in damaged plants, then the outcrossing rate of their CH flowers could increase. Given the above estimates, the outcrossing rate of CH flowers on damaged plants would have to increase from 0.50 to 0.78 to oppose the direct effects of herbivory. Therefore, the indirect effects of herbivory on outcrossing rate of CH flowers would have to be substantial to counter the direct effects of herbivory that favor selfing.
Consequences for mating system evolution
Whereas the majority of mating system evolution models have incorporated factors such as inbreeding depression and population structure to predict equilibrium levels of selfing (e.g., Lande and Schemske, 1985
; Holsinger, 1991
), variation in the pollination environment may also influence mating system evolution. For example, in an unpredictable pollination environment selection favors the ability for plants to both self and outcross (Schoen and Brown, 1991
; Schoen et al., 1996
). Although often overlooked, herbivores and other antagonists (e.g., competitors) are likely to serve as selective agents in the evolution of mating systems. In a model by Schoen and Lloyd (1984)
, a heteromorphic flowering system will be maintained when individuals are able to produce the appropriate flower type in response to heterogeneity in the parental environment that is an indicator of variation in pollinator activity (i.e., individuals produce CL flowers under conditions of low pollinator activity and CH flowers when pollinator activity is high). Our data provide support by showing that leaf damage both reduces pollinator activity and increases proportional CL flower production. In contrast, in a low herbivory environment, pollinator activity is more dependable and plants reproduce more via CH flowers. Our results combined with those from other studies (Levri and Real, 1998
; Elle and Hare, 2002
) indicate that herbivores are likely to have a role in shaping the evolution of the mating system. However, further work is needed to determine whether plasticity to herbivory with respect to mating system is adaptive and whether heterogeneity in the herbivore environment maintains it.
|
FOOTNOTES |
|---|
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Agrawal A. A. 2001 Transgenerational consequences of plant responses to herbivory: an adaptive maternal effect?. American Naturalist 157: 555-569[CrossRef][ISI]
Agrawal A. A. S. Y. Strauss M. J. Stout 1999 Costs of induced responses and tolerance to herbivory in male and female fitness components of wild radish. Evolution 53: 1093-1104[CrossRef][ISI]
Baker H. G. P. D. Hurd 1968 Intrafloral ecology. Annual Review of Entomology 13: 385-414[CrossRef][ISI]
Baldwin I. T. 1990 Herbivory simulations in ecological research. Trends in Ecology and Evolution 5: 91-93
Barrett S. C. H. 1990 Mating system evolution and speciation in heterostylous plants. In D. Otte and J. A. Endler [eds.], Speciation and its consequences, 257–238. Sinauer, Sunderland, Massachusetts, USA
Barrett S. C. H. C. G. Eckert 1990 Variation and evolution of mating systems in seed plants. In S. Kawano [ed.], Biological approaches and evolutionary trends in plants, 229–254. Academic Press, London, UK
Bell T. J. J. A. Quinn 1987 Effects of soil moisture and light intensity on the chasmogamous and cleistogamous components of reproductive effort of Dichanthelium clandestinum populations. Canadian Journal of Botany 65: 2243-2249[Medline]
Caswell H. 2001 Matrix population models: construction, analysis, and interpretation. Sinauer Associates, Sunderland, Massachusetts, USA
Charlesworth D. B. Charlesworth 1987 Inbreeding depression and its evolutionary consequences. Annual Review of Ecology and Systematics 18: 237-268[CrossRef][ISI]
Culley T. M. 2002 Reproductive biology and delayed selfing in Viola pubescens (Violaceae): an understory herb with chasmogamous and cleistogamous flowers. International Journal of Plant Science 163: 113-122[CrossRef]
Eckert C. G. A. Schaefer 1998 Does self-pollination provide reproductive assurance in Aquilegia canadensis (Ranunculaceae)?. American Journal of Botany 85: 919-924[Abstract]
Elle E. J. D. Hare 2002 Environmentally induced variation in floral traits affects the mating system in Datura wrightii. Functional Ecology 16: 79-88[CrossRef][ISI]
Elle E. N. M. van Dam J. D. Hare 1999 Cost of glandular trichomes, a ‘resistance’ character in Datura wrightii Regel (Solanaceae). Evolution 53: 22-35
Gross J. B. C. Husband S. C. Stewart 1998 Phenotypic selection in a natural population of Impatiens pallida Nutt. (Balsaminaceae). Journal of Evolutionary Biology 11: 589-609[CrossRef][ISI]
Hamrick J. L. M. J. Godt 1990 Allozyme diversity in plant species. In A. H. D. Brown, M. T. Clegg, A. L. Kahler, and B. S. Weir [eds.], Plant populations, genetics, breeding, and genetic resources, 43–63. Sinauer, Sunderland, Massachusetts, USA
Harder L. D. S. C. H. Barrett 1995 Mating cost of large floral displays in hermaphrodite plants. Nature 373: 512-515[CrossRef]
Holsinger K. E. 1991 Mass-action models of plant mating systems: the evolutionary stability of mixed mating systems. American Naturalist 138: 606-622[CrossRef][ISI]
Juenger T. J. Bergelson 1997 Pollen and resource limitation of compensation to herbivory in scarlet gilia, Ipomopsis aggregata. Ecology 78: 1684-1695[ISI]
Koptur S. C. L. Smith J. H. Lawton 1996 Effects of artificial defoliation on reproductive allocation in the common vetch, Vicia sativa (Fabaceae: Papilionoideae). American Journal of Botany 83: 886-889[CrossRef][ISI]
Krupnick G. A. A. E. Weis D. R. Campbell 1999 The consequence of floral herbivory for pollinator service to Isomeris arborea. Ecology 80: 125-134[ISI]
Lande R. D. W. Schemske 1985 The evolution of self-fertilization and inbreeding depression in plants. I. Genetic models. Evolution 39: 24-40[CrossRef][ISI]
Leck M. A. 1979 Germination behavior of Impatiens capensis Meerb. (Balsaminaceae). Bartonia 46: 1-11
Le Corff J. 1993 Effects of light and nutrient availability on chasmogamy in an understory tropical herb, Calathea micans (Marantaceae). American Journal of Botany 80: 1392-1399[CrossRef][ISI]
Lehtila K. S. Y. Strauss 1997 Leaf damage by herbivores affects attractiveness to pollinators in wild radish, Raphanus raphanistrum. Oecologia 111: 396-403[CrossRef][ISI]
Levri M. A. L. A. Real 1998 The role of resources and pathogens in mediating the mating system of Kalmia latifolia. Ecology 79: 1602-1609[CrossRef][ISI]
Lu Y. 2000 Effects of density on mixed mating systems and reproduction in natural populations of Impatiens capensis. International Journal of Plant Science 161: 671-681[CrossRef]
Mothershead K. R. J. Marquis 2000 Fitness impacts of herbivory through indirect effects on plant–pollinator interactions in Oenothera macrocarpa. Ecology 81: 30-40[CrossRef][ISI]
Quesada M. K. Bollman A. G. Stephenson 1995 Leaf damage decreases pollen production and hinders pollen performance in Cucurbita texana. Ecology 76: 437-443[CrossRef][ISI]
Rust R. W. 1977 Pollination in Impatiens capensis and Impatiens pallida (Balsaminaceae). Bulletin of the Torrey Botanical Club 104: 361-367[CrossRef][ISI]
SAS Institute. 2001 SAS software, version 8.2. SAS Institute, Cary, North Carolina, USA
Schemske D. W. 1978 Evolution of reproductive characteristics in Impatiens (Balsaminaceae): the significance of cleistogamy and chasmogamy. Ecology 59: 596-613[CrossRef][ISI]
Schemske D. W. C. C. Horvitz 1988 Plant–animal interactions and fruit production in a neotropical herb: a path analysis. Ecology 69: 1128-1137[CrossRef][ISI]
Schmitt J. J. Eccleston D. W. Ehrhardt 1987 Dominance and suppression, size-dependent growth, and self-thinning in a natural Impatiens capensis population. Journal of Ecology 75: 651-665[CrossRef]
Schoen D. J. A. H. D. Brown 1991 Whole- and part-flower self-pollination in Glycine clandestina and Glycine argyrea and the evolution of autogamy. Evolution 45: 1651-1664[CrossRef][ISI]
Schoen D. J. D. G. Lloyd 1984 The selection of cleistogamy and heteromorphic diaspores. Biological Journal of the Linnean Society 23: 303-322
Schoen D. J. M. T. Morgan T. Bataillon 1996 How does self-pollination evolve? Inferences from floral ecology and molecular genetics. Philosophical Transactions of the Royal Society of London, Series B 351: 1281-1290[CrossRef]
Stephenson A. G. 1981 Flower and fruit abortion: proximate causes and ultimate functions. Annual Review of Ecology and Systematics 12: 253-279
Strauss S. Y. J. K. Conner S. L. Rush 1996 Foliar herbivory affects floral characters and plant attractiveness to pollinators: implications for male and female plant fitness. American Naturalist 147: 1098-1107[CrossRef][ISI]
Strauss S. Y. D. H. Siemens M. B. Decher T. Mitchell-Olds 1999 Ecological costs of plant resistance to herbivores in the currency of pollination. Evolution 53: 1105-1113[CrossRef][ISI]
Vogler D. S. Kalisz 2001 Sex among flowers: the distribution of plant mating systems. Evolution 55: 202-204[CrossRef][ISI][Medline]
Waller D. M. 1980 Environmental determinants of outcrossing in Impatiens capensis (Balsaminaceae). Evolution 34: 747-761[CrossRef][ISI]
Waller D. M. S. E. Knight 1989 Genetic consequences of outcrossing in the cleistogamous annual Impatiens capensis. II. Outcrossing rates and genotypic correlations. Evolution 43: 860-869[CrossRef][ISI]
Zar J. H. 1999 Biostatistical analysis, 4th ed. Prentice Hall, Upper Saddle River, New Jersey, USA
This article has been cited by other articles:
|
|
 |
|
  C. T. Ivey and D. E. Carr Effects of herbivory and inbreeding on the pollinators and mating system of Mimulus guttatus (Phrymaceae) Am. J. Botany, October 1, 2005; 92(10): 1641 - 1649. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. H. Cole and T.-L. Ashman Sexes show differential tolerance to spittlebug damage and consequences of damage for multi-species interactions Am. J. Botany, October 1, 2005; 92(10): 1708 - 1713. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
