| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Biology, Marsh Life Science Building, University of Vermont, Burlington, Vermont 05405, USA
Received for publication January 5, 1999. Accepted for publication June 3, 1999.
ABSTRACT
I investigated the mating systems and phenotypic variation of two sympatric spring ephemerals, Trillium erectum and T. grandiflorum (Liliaceae), and phenotypic selection acting through female reproductive success for 11 morphological characters in five sympatric populations of the two species. I examined the degree of self-compatibility, pollinator-visitation rates, and pollen limitation of fruit and seed production in both species. Both Trillium species were self-compatible, but outcrossed flowers produced more successful fruits and seeds than self-pollinated flowers. Pollinator-visitation rates to the two species were low compared to other insect-pollinated spring ephemerals. In addition, both T. erectum and T. grandiflorum experienced pollen limitation in fruit and/or seed production; however, levels of fecundity in both species may be influenced by resource availability as well. I found significant phenotypic variation in 11 morphological characters within and among the five study populations. The sizes of all morphological characters were positively correlated. In general, larger T. erectum and T. grandiflorum produced more seeds. Phenotypic selection analysis revealed that direct and indirect selection acted on the size of morphological characters for both species. But there was no detectable selection acting on plant shape. This study reveals that variation in plant size exists within and among populations of both species, and this variation is associated with variance in female reproductive success. Spatial and temporal variation in pollinator and/or resource abundance may play a role in the phenotypic variation exhibited by both Trillium species.
Key Words: Liliaceae • mating system • phenotypic selection • pollen limitation • self-compatibility • spring ephemeral • Trillium erectum • Trillium grandiflorum.
The evolution of plant traits is influenced by biotic interactions, such as attraction of pollinators (for reviews see Grant and Grant, 1965
; Faegri and van der Pijl, 1978
; Real, 1983
; Waser, 1983
; Proctor, Yeo, and Lack, 1996
) and avoidance of herbivores and seed predators (e.g., Beattie, Breedlove, and Ehrlich, 1973
; Zimmerman, 1980
; Augspurger, 1981
; Hainsworth, Wolf, and Mercier, 1984
; Brody, 1992
), and by abiotic factors, such as nutrient and water availability (e.g., Clausen, 1951
; Stewart and Schoen, 1987
; Herrera, 1993
). To date, most studies of phenotypic selection on plant traits have concentrated on summer annual and perennial herbaceous species (but see O'Connell and Johnston, 1998
; Totland et al., 1998
). However, our current understanding of phenotypic selection is deficient in studies involving selection in spring-ephemeral, or vernal, species. The lack of attention to vernal species may be due, in part, to the uncertainties of their natural histories. Here I focus on phenotypic variation, mating system biology, and phenotypic selection on plant traits in two vernal species, Trillium erectum and T. grandiflorum.
Trillium blanket forest floors soon after snowmelt in eastern North America. Phenotypic diversity in qualitative and quantitative plant characters across the Trillium genus has been attributed to their specific pollinators. For example, the small, red, nectarless flowers of T. erectum are associated with dipteran pollinators (Robertson, 1896
; Weed, 1900
; Davis, 1981
), while the larger, white, nectar-producing flowers of T. grandiflorum are associated with hymenopteran pollinators (Carter, 1892; Robertson, 1896). However, the amount of intraspecific phenotypic variation in quantitative characters within Trillium species has been debated. Studies of phenotypic variation have revealed populations with relatively low phenotypic variability (T. erectum and T. grandiflorum; Serota and Smith, 1967
) and ones with high variability (T. erectum; Ringius and Chmielewski, 1987
). It is unknown how phenotypic variation relates to pollination success and plant fitness in T. erectum and T. grandiflorum (for the relationship between individual biomass and plant reproductive characters see Davis, 1981
; Kawano, Ohara, and Utech, 1986
; Ohara, 1989
).
In Trillium, as with many spring ephemerals, studies combining phenotypic variation, pollination success, and plant fitness are complicated by the uncertainties of mating system characteristics. For example, the breeding system of T. grandiflorum is thought to shift from predominantly outcrossed and insect-pollinated in southern North America to predominantly autogamous in northeastern populations (Fukuda and Grant, 1980
). The shift is thought to result in decreased phenotypic and genotypic variation in populations along a northern transect (Fukuda and Grant, 1980
). However, other studies have refuted this hypothesis (Broyles, Sherman-Broyles, and Rogati, 1997
). Clearly, the mating system of plants will have a strong effect on phenotypic and genotypic variation and selection of plant traits (for reviews, see Lloyd and Barrett, 1996
; Proctor, Yeo, and Lack, 1996
).
Missing from much of the debate on Trillium are detailed ecological studies encompassing information on mating system characteristics, the amount of phenotypic variation in plant characters within and among populations, and the role mating system and phenotypic variation play in reproductive success. In this study, I investigated the mating system and phenotypic variation of T. erectum and T. grandiflorum and phenotypic selection acting through female plant function for 11 morphological characters in five sympatric populations of the two species. I first investigated the mating system biology of T. erectum and T. grandiflorum, including the degree of self-compatibility, the identity of pollinator visitors, pollinator-visitation rates, and the degree to which fruit and seed set was pollen limited. I then asked the following questions. Do morphological characters vary within and among populations of the two Trillium species? Do both species experience phenotypic selection acting through female plant function? And if so, what is the strength of selection acting on each character?
MATERIALS AND METHODS
Study system
Trillium erectum L. and T. grandiflorum (Michx.) Salisb. (Liliaceae) are long-lived spring ephemerals that occur in the understory of deciduous woodlands throughout eastern North America. Mature plants of both species consist of one or more stems arising from a tuber-like rhizome. Each stem bears a whorl of three leaves and a single terminal flower with three petals. Each flower has two whorls of three stamens, one whorl opposite the petals and one opposite the sepals. The red, nectarless flowers of T. erectum produce a fetid odor that attracts dipterans. In contrast, the white, protandrous flowers of T. grandiflorum produce no discernible odor. Insects, including Hymenoptera and Diptera, are attracted to the flowers by nectar secreted from septal glands located between the ovary and stamen filaments. Insect visitors often collect pollen and/or nectar from T. grandiflorum.
The reproductive biology of both Trillium species is unclear. Both have been described as apomictic (Jeffrey and Haertl, 1939
; Fryxell, 1957
), self-compatible (Carter, 1892
; Ohara, 1989
), and predominantly outcrossing (Broyles, Sherman-Broyles, and Rogati, 1997
).
Trillium erectum and T. grandiflorum accumulate photosynthate energy into their underground storage organs. The photosynthate is then translocated into fertilized ovules from the storage organs. Because there is no direct translocation of photosynthate energy from leaves to fertilized ovules, there is a time lag in energy translocation that may allow for the possibility of resource limitation of fruit and seed production (S. Kawano, personal communication).
Flowers from both species produce a single fruit with three locules. Fruits produce seeds bearing elaiosomes that are dispersed by a variety of ants (Gates, 1941
; Ohara and Higashi, 1987
; Gunther and Lanza, 1989
).
I studied five sympatrically flowering populations of T. erectum and T. grandiflorum in Chittenden County, Vermont. All populations were located around Indian Brooke Reservoir in mixed deciduous and coniferous forests. Flowers of both species bloomed concurrently from mid-April to late-May in 1997 and 1998, when this work was conducted. Dominant herbaceous perennials associated with the Trillium populations included dutchman's breeches [Dicentra cucullaria (Papaveraceae, subfamily Fumarioideae)], jack-in-the-pulpit [Arisaema triphyllum (Araceae)], and trout lily [Erythronium americanum (Liliaceae)].
Field methods
Mating system
To test the self-compatibility of T. erectum and T. grandiflorum, I randomly assigned similar-sized plants of both species to groups of three. I bagged all plants in each triplet prior to flower opening with mesh bagging made of bridal veil. In 1997, I bagged three triplets of T. erectum and nine triplets of T. grandiflorum, and in 1998, I bagged ten triplets of each species. One member of each triplet was pollinated with self pollen once stigmas became receptive (on the first day of flower opening; R. E. Irwin, unpublished data on gynoecial maturity). The second member of each triplet was emasculated and pollinated with outcrossed pollen from two pollen donors growing at least 5 m away. The third member of each triplet was emasculated prior to flower opening and left unpollinated to ensure that bagged flowers did not receive extraneous pollen and to test whether either species was apomictic. I compared the number of successfully initiated fruits, the number of seeds per fruit, and the seed : ovule ratio (number of mature seeds produced divided by the total number of ovules per fruit; measured in 1998 only) between selfed and outcrossed treatments using chi-square and one-way ANOVA analyses. I did not include bagged, unpollinated flowers in these analyses because they did not produce seed-bearing fruits (see Results).
To determine the types of potential pollinators visiting T. erectum and T. grandiflorum, I applied tanglefoot to flowers and collected mired insects. I identified the insects to order. Under a dissecting microscope, I determined whether insects carried Trillium pollen on their bodies. Because some insect visitors may be large enough to avoid being caught in tanglefoot, I also performed 10 h of pollinator observations in May 1997 and recorded the different orders of insects visiting each species.
In addition, I investigated pollinator-visitation rates to each plant species. In May 1997, I applied tanglefoot to six newly opened flowers of each species once a week for 4 wk. I returned to flowers 24 h after tanglefoot application and counted the number of insects mired in the tanglefoot that were known carriers of Trillium pollen (determined above). This method of estimating pollinator-visitation rates has been used in other studies of T. erectum (Davis, 1981
). I calculated the rate of pollinator visitation to each plant as the number of insects mired in the tanglefoot divided by 24 h. I used a one-way ANOVA to examine variation in visitation rates among the four sampling dates.
To test whether fruit and seed production were pollen limited, I performed hand-pollinations in 1997 and 1998. I chose nine pairs of T. erectum and 30 pairs of T. grandiflorum in 1997, and 30 pairs of each species in 1998, which I matched on the basis of overall plant size. I hand-pollinated one member of each pair with pollen from at least two pollen donors once stigmas were receptive. The other member of each pair was an unmanipulated, open-pollinated control. I compared the number of successfully initiated fruits, the number of seeds per fruit, and the seed : ovule ratios (1998 only) between hand- and open-pollinated flowers using chi-square and one-way ANOVA analyses.
Phenotypic variation and plant reproductive success
In April 1998, I randomly chose 15–20 plants of each species prior to flowering at five sites around Indian Brooke Reservoir (90 T. erectum and 100 T. grandiflorum total). The five sites were separated from each other by at least 500 m, and each site contained more than 100 individuals of each species. I measured 11 morphological variables on each plant on the first day of flowering. I measured petal length and width, sepal length and width, stamen length (anther and filament), carpel length (stigma, style, and ovary), and stem diameter to the nearest 0.01 mm, using hand-held digital calipers. The uppermost petal, sepal, and stamen, and the leaf clockwise relative to the uppermost petal were chosen for measurement. Leaf length and width, stem height, and pedicel length were measured to the nearest 0.1 cm. Stem height was measured as the vertical distance from the ground to the base of the whorled leaves, and pedicel length was the distance from the leaves to the sepals and petals. Because the whorled leaves, petals, sepals, and stamens of T. erectum are approximately symmetrical, measuring one of each per plant should provide a good indication of size for each character for both species in general (Ringius and Chmielewski, 1987
).
Once fruits were mature, I collected all fruit capsules and counted the number of mature and aborted seeds per fruit. For each successful fruit, I determined the number of seeds produced and the seed : ovule ratio. For each plant, I converted seed set and the seed : ovule ratio to relative fitness estimates by dividing by the mean values over all plants.
Analysis of phenotypic variation
To examine patterns of phenotypic variation within T. erectum and T. grandiflorum, I also calculated coefficients of variation for each plant character in each site. I calculated Pearson product-moment correlation coefficients for all pairs of characters within sites and over all sites. I included relative fitness components in the correlation analysis to determine how plant characters were related to plant fitness.
I analyzed among-site variation in plant characters using a MANOVA with site as a fixed effect and the 11 morphological characters as dependent variables. Subsequent one-way ANOVA and Tukey-Kramer HSD tests identified significant differences among sites for each of the morphological characters. By using a multivariate ANOVA and subsequent univariate tests, I reduced the probability of inflating the Type I error rate (Rencher, 1995
).
Analysis of phenotypic selection
Regression techniques can be used to express the effects of plant characters on fitness as selection coefficients. The coefficients provide estimates of direct selection and indirect selection acting through correlated traits (Lande, 1979
; Lande and Arnold, 1983
; Arnold and Wade, 1984a, b
). They also provide estimates that can be expressed in units of standard deviation that can be compared among populations, species, and studies (Lande, 1979
; Lande and Arnold, 1983
; Arnold and Wade, 1984a, b
). Regression techniques are a tool for suggesting hypotheses about the forces of selection and can identify within-generation change as "selection" (not to be confused with between-generation changes or "evolutionary response to selection"); therefore, selection can be measured within a single generation (Fisher, 1930
; Lande and Arnold, 1983
; Arnold and Wade, 1984a, b
). Selection coefficients are concerned with changes in the mean of a phenotypic distribution and its association with a relative fitness distribution. Selection coefficients do not provide information regarding why an association exists between a character and relative fitness. If high covariance is detected between character traits and relative fitness, it does not reveal the cause and effect involved in the consequences of evolutionary change (Mitchell-Olds and Shaw, 1987
; Wade and Kalisz, 1990
).
To determine whether there was significant phenotypic selection on plant traits, I calculated opportunities for selection, selection differentials, and selection gradients for female plant function for both species. I calculated all selection coefficients for both relative seed set and relative seed : ovule ratios for T. erectum and T. grandiflorum. In all analyses for each species, I included site as a class variable to remove site-specific effects.
The opportunity for selection (I) is the variance in relative fitness. It sets an upper boundary to which a character can be shifted by selection (Arnold and Wade, 1984a
). If characters are invariant, there is no selection opportunity (I = 0). I calculated selection opportunities as the error variance in an ANOVA of relative fitness with site as a class variable.
Selection differentials (S) measure the change in the mean of a character before and after selection in a single generation. They measure both direct selection on a character and indirect selection acting on correlated characters (Arnold and Wade, 1984a
). To calculate selection differentials, I first used ANCOVAs with site as a class variable to estimate the within-site regression coefficient (bi) of relative fitness on characters, i. To compare selection differentials among characters and species, I calculated the standardized selection differential for each character as S' = bi x Vi1/2, where Vi is the variance of a character among plants nested within sites.
Selection gradients measure only the direct force of selection acting on a character. They represent the partial regression coefficients of relative fitness on traits (Lande and Arnold, 1983
). When characters are highly correlated, as in this study (see Results), selection gradients do not accurately represent direct selection acting on a character (Lande and Arnold, 1983
). To avoid problems associated with correlations among characters, I used a principal components analysis to reduce the number of dimensions in the data set (Rencher, 1995
). Principal components scores were calculated from the correlation matrix of plant characters for each species over all sites. I used the first three principal components in a multiple regression on relative plant fitness components, with site included as a class variable. Standardized selection gradients were the standardized partial regression coefficients in the models. All statistical analyses were carried out using programs in mainframe SAS (SAS, 1985a, b
; Cody and Smith, 1997
).
RESULTS
Mating system of T. erectum and T. grandiflorum
For 1997 and 1998 combined, 85% of cross-pollinated T. erectum produced seed-bearing fruits, while only 62% of self-pollinated T. erectum produced successful fruits. Cross-pollinated plants also produced more seeds (mean ± 1 SE: 50.64 ± 11.01 seeds vs. 36.38 ± 12.91 seeds) and had higher seed : ovule ratios (mean ± 1 SE: 0.67 ± 0.08 vs. 0.46 ± 0.11) than self-pollinated plants (P < 0.05 in all statistics above).
Cross-pollinated T. grandiflorum were also more successful than self-pollinated plants. For both years combined, 90% of cross-pollinated T. grandiflorum produced successful fruits, while only 47% of self-pollinated plants produced successful fruits. Cross-pollinated flowers also produced more seeds per plant (mean ± 1 SE: 18.13 ± 2.64 seeds vs. 12.22 ± 3.53 seeds) and had higher seed : ovule ratios (mean ± 1 SE: 0.66 ± 0.08 vs. 0.35 ± 0.14) than self-pollinated flowers (P < 0.05 in all statistics above). For both species, the fruits of bagged, unpollinated flowers did not produce seeds; therefore, it is unlikely that either species is apomictic.
Trillium erectum was visited primarily by dipterans (including Anthomyiidae, Sciaridae, and Sarcophagidae) and to a lesser extent by coleopterans. Trillium grandiflorum was visited primarily by hymenopterans (including Apidae in the genera Apis and Bombus) and to a lesser extent by dipterans. All insects collected on T. erectum and T. grandiflorum carried Trillium pollen. Qualitatively, the orders of insects caught in the tanglefoot matched those I observed during the 10 h of pollinator observations; therefore, tanglefoot gave an accurate estimate of the variety of insect orders visiting T. erectum and T. grandiflorum.
Over the four sampling dates, pollinator-visitation rates to T. erectum ranged from 0.06 to 1.29 visits·h-1·flower-1 with a mean (± 1 SE) of 0.51 ± 0.08 visits·h-1·flower-1. Two plants were missing when I returned after 24 h, so all subsequent analyses were conducted on 22 T. erectum. Plants varied in pollinator-visitation rates among the four sampling dates (F3, 21 = 5.82, P = 0.006). Visitation rates increased from the first sampling date (1 May 1997) to the third date (17 May 1997), but decreased on the last date (23 May 1997).
For T. grandiflorum, pollinator-visitation rates ranged from 0.03 to 0.18 visits·h-1·flower-1 with a mean (± 1 SE) of 0.08 ± 0.02 visits·h-1·flower-1. Thirteen plants were missing when I returned after 24 h, so all subsequent analyses were conducted on 11 T. grandiflorum on two sampling dates. Plants did not vary in pollinator-visitation rates between the two sampling dates (F2, 10 = 0.01, P = 0.94). On average, T. grandiflorum received significantly fewer pollinator visits than T. erectum (F1, 31 = 22.21, P < 0.0001).
The effects of supplemental pollen addition to T. erectum varied among years and depended on the response variable examined. In 1997, 89% of hand-pollinated plants produced successful fruits, while only 33% of open-pollinated plants produced fruits (
2 = 6.32, P = 0.01). But in 1998, fruit set of hand- and open-pollinated plants did not differ significantly (2 = 0.28, P = 0.59). The number of seeds produced did not vary between hand- and open-pollinated plants in 1997 (F1, 9 = 0.75, P = 0.41), but in 1998, hand-pollinated plants produced significantly more seeds than open-pollinated plants (mean ± 1 SE: 46.85 ± 3.73 vs. 19.22 ± 3.66 seeds; F1, 22 = 21.09, P < 0.0001). The seed : ovule ratios (measured in 1998 only) were greater in hand-pollinated plants than open-pollinated plants (mean ± 1 SE: 0.70 ± 0.04 vs. 0.55 ± 0.03; F1, 22 = 4.27, P = 0.05).
The effects of supplemental pollen also varied in T. grandiflorum. Hand- and open-pollinated plants did not differ in the number of successful fruits produced in either year of the study (1997: 2 = 0.50, P = 0.48; 1998: 2 = 0.28, P = 0.59). But supplemental pollen did increase seed production in hand-pollinated plants in 1997 (mean ± 1 SE: 22.38 ± 2.11 vs. 10.37 ± 2.22 seeds; F1, 11 = 6.70, P = 0.03) and in 1998 (mean ± 1 SE: 17.58 ± 1.21 vs. 11.68 ± 1.24 seeds; F1, 21 = 12.55, P = 0.002). However, the seed : ovule ratios (measured in 1998 only) did not vary between hand- and open-pollinated plants (F1, 21 = 0.51, P = 0.48) because fruits that produced more seeds also had more total ovules.
Phenotypic variation and female reproductive success in T. erectum and T. grandiflorum
Correlation coefficients varied among the five sites (range: for T. erectum, 15.7–22.6%; T. grandiflorum, 16.3–23.2%) and among the 11 morphological characters (range: for T. erectum, 14.8–21.3%; for T. grandiflorum, 16.1–24.5%). Trillium erectum in site 4 and T. grandiflorum in site 5 exhibited the most variation in morphological characters. For both species, mean coefficients of variation for floral characters (T. erectum: 17.3%; T. grandiflorum: 19.8%) were similar to those for vegetative characters (T. erectum: 18.9%; T. grandiflorum: 20.0%).
All of the morphological characters were positively correlated for T. erectum and for T. grandiflorum (Table 1). Moreover, relative seed set and relative seed : ovule ratios were positively correlated with all morphological characters for T. erectum (Table 1). Therefore, larger T. erectum had higher female-fitness components. For T. grandiflorum, all morphological characters were positively correlated with relative seed set. However, relative seed : ovule ratios were only weakly correlated with stamen length and were uncorrelated with all other morphological characters (Table 1).
|
2 = 6.47, P = 0.17; T. grandiflorum: 2 = 5.64, P = 0.23). Percentage fruit set per site ranged from 54.5 to 91.7% for T. erectum and 68.7 to 94.4% for T. grandiflorum.
|
Standardized selection differentials indicated that the number of seeds produced by T. erectum and T. grandiflorum were higher in plants with larger morphological characters (Table 3). Therefore, larger plants produced more seeds. However, seed : ovule ratios experienced no detectable phenotypic selection (Table 3).
|
). The partial regressions of relative seed production on the principal components showed strong direct selection acting on plant size but not on plant shape (Table 5). Relative seed : ovule ratios experienced no direct phenotypic selection in either species (Table 5).
|
|
Mating system of T. erectum and T. grandiflorum
Both T. erectum and T. grandiflorum were self-compatible in populations around the Indian Brooke Reservoir, but outcrossed plants produced more successful fruits and seeds than self-pollinated plants. These results contradict previous findings showing that neither species is fully self-compatible (e.g., Broyles, Sherman-Broyles, and Rogati, 1997
) but support other studies finding that both species are fully self-compatible (Fukuda and Grant, 1980
; Ohara, 1989
). Variability in the strength of self-compatibility suggests that T. erectum and T. grandiflorum probably possess quantitative variation for self-compatibility and/or different amounts of embryonic abortion of selfed zygotes as a result of inbreeding depression (Galen, Plowright, and Thomson, 1985
; Seavey and Bawa, 1986
; Barrett, 1987
). In addition, variability in self-compatibility may be associated with the geographic location of populations studied (e.g., Fukuda and Grant, 1980
).
Natural pollinator-visitation rates to T. erectum and T. grandiflorum were low compared to other insect-pollinated spring ephemerals (e.g., T. grandiflorum: 0.08 visits/h; T. erectum: 0.51 visits/h; vs. Claytonia virginica: 0.84 visits/h; Dentaria laciniata: 1.38 visits/h; Erythronium albidum: 4.26 visits/h; Schemske et al., 1978
). In light of low visitation rates to T. erectum and T. grandiflorum, it was not surprising that both species experienced pollen limitation in one or more years of the study. It is interesting to note that supplemental hand-pollinations affected female reproductive success in different ways in different years. For example, T. erectum seed production was equivalent between hand- and naturally pollinated flowers in 1997 when a significant difference in fruit production existed. Conversely, fruit set was equivalent in hand- and naturally pollinated flowers in 1998 when the greatest differences in seed production were observed. In general, even with application of excess pollen to stigmas, many ovules failed to set seed in both species. This suggests that the provision of nutrient and water resources, in addition to pollen provisions (both quality and quantity), may be important to fruit and seed production (e.g., Haig and Westoby, 1988
; Campbell and Halama, 1993
). Both species may differ in space and time in the importance of pollen vs. resource limitation, and pollen and resource limitation may affect different aspects of plant-reproductive output (e.g., Campbell and Halama, 1993
). Low pollinator-visitation rates and some pollen limitation of fruit and seed set have been found in other studies of T. grandiflorum (Lubbers and Lechowicz, 1989
). Both T. erectum and T. grandiflorum are visited by a number of different dipteran, hymenopteran, and coleopteran pollinators. It is unknown whether these pollinators differ in their ability to effect pollination in Trillium.
One caveat regarding two-season studies of pollen limitation is that they do not demonstrate that lifetime seed production is limited by pollinators only. In accord with resource limitation, seed production may have costs expressed in decreased survival and future fecundity (e.g., Calvo and Horovitz, 1990
). In Trillium, once individuals attain the fertile stage, they cannot stay in this same growth phase continuously. Continual fruit and seed production reduces the probability that Trillium will flower in subsequent years (Ohara and Kawano, 1986
).
Trillium apparently do not take full advantage of self-compatibility. Despite pollen limitation, seeds are more likely to be produced from cross- than from self-fertilization (Broyles, Sherman-Broyles, and Rogati, 1997
), and seeds produced from selfing may be at a selective disadvantage during germination, establishment, and juvenile stages (Davis, 1981
; Kawano, Ohara, and Utech, 1986
; Hanzawa and Kalisz, 1993
). However, because insect-pollinator populations often fluctuate enormously from one year to the next (e.g., Schemske et al., 1978
), there may be inadequate numbers of insects for full pollination of the species in some years. Autogamy can occur in flowers of T. grandiflorum when elongated stigmas reflex and contact the anthers (Fukuda and Grant, 1980
); however, autogamy in T. erectum is improbable because stigmas rarely come in contact with dehiscing anthers (S. Broyles, personal communication). Nonetheless, the ability of Trillium to self-fertilize via autogamy or via within-flower pollen transfer by an insect, despite the selective disadvantage, may serve as a bet-hedging mechanism that may increase seed production in years when pollinators are scarce.
Phenotypic variation in T. erectum and T. grandiflorum
The high levels of phenotypic variation exhibited by T. erectum and T. grandiflorum in this study are characteristic of plants with predominantly outcrossed breeding systems (e.g., Herrera, 1996
). The levels of variation are similar to those reported in Ontario populations of T. erectum (Ringius and Chmielewski, 1987
). Along with high levels of phenotypic variation, T. erectum and T. grandiflorum show high levels of allozyme variation as well (Broyles, Sherman-Broyles, and Rogati, 1997
). The phenotypic variation in this study was attributed to both within- and among-population differences. Significant variation found both within and among populations suggests that selection pressures may be micro site and/or population specific and may affect plant characters in different ways. Phenotypic and genotypic variation both within and among populations may be affected by a number of different factors, including (and described in more detail in the section on Phenotypic Selection below): (1) variation in pollinator communities (for reviews see Faegri and van der Pijl, 1978
; Real, 1983
; Proctor, Yeo, and Lack, 1996
), (2) environmental heterogeneity (e.g., Thompson, 1983
; Heywood and Levin, 1986
; Antonovics, Clay, and Schmitt, 1987
, Platenkamp and Foin, 1990
; Schwaegerle and Levin, 1990
), and (3) temporal variation in selection pressures (e.g., Campbell, 1989
; Scheiner, 1989
; Schemske and Horovitz, 1989
; Widén, 1991
; Kelly, 1992
).
One surprising finding in this study was that coefficients of variation were similar for both floral and vegetative characters in both species. Conventional wisdom, however, predicts that most floral traits should display less variation within a species than vegetative traits (Berg, 1960
) because floral characters should strongly influence the quantity of pollen removed from and deposited in flowers and, thus, male and female function. In Trillium, however, correlation of growth of floral and vegetative characters may constrain variation to that of overall size rather than shape. Trillium erectum adjusts to higher resource levels by producing larger flowers with more ovules (Davis, 1981
). In contrast, most plants would adjust to fluctuating resources by adjusting the number of ramets or flowers produced (e.g., Stanton, Bereczky, and Hasbrouck, 1987
; Campbell and Halama, 1993
). Trillium is unique in that it is constrained to the number of flowers on each ramet (but see Davis, 1981
) but can adjust to resource levels by adjusting flower and plant size.
Phenotypic selection in T. erectum and T. grandiflorum
Variation in morphological characters was associated with female reproductive success. For both species, there was positive selection for all plant characters on seed production. I also found positive direct selection for plant size on seed production. However, plant shape did not significantly affect seed production for either species. While larger plants had a greater number of ovules fertilized, positive direct and indirect selection did not act on the percentage of ovules fertilized. Most likely, pollen and/or resource limitation of the percentage of ovules fertilized is proportional to plant size (Wolfe, 1983
). Whether many of the characters chosen to measure in this study have direct effects on female reproductive success or are merely correlated with more important factors of plant reproduction is debatable (S. Kawano, personal communication). Female reproductive success may not simply be a consequence of plant size if there are other factors responsible for final energy investment and reproductive output.
A positive effect of plant size on female reproductive success has been documented in other vernal species and summer-flowering species as well (e.g., Abrahamson and Hershey, 1977
; Barrett and Thomson, 1982
; Kawano, Hiratsuka, and Hayashi, 1982
; Wolfe, 1983
; Dudash, 1991
). In addition, plant size, but not plant shape, as a determinant of female fecundity has also been documented (e.g., Herrera, 1993
).
The standardized selection differentials and gradients reported for these two vernal species are similar to those in some summer annual and perennial species (e.g., Stewart and Schoen, 1987
; Herrera, 1993
), but not others (e.g., Campbell, 1989
; Galen, 1996
). In summer-flowering species with estimates of selection similar to those of T. erectum and T. grandiflorum, environmental heterogeneity was often linked to variation in female fecundity, whereas in summer-flowering species with estimates of selection different from those reported here, variation in pollinator attraction was often linked to variation in female fecundity. These relationships did not always hold true however (e.g., Scheiner, 1989
; Bennington and McGraw, 1995
). In addition, I found no relationship between the magnitude of selection coefficients and components of the mating system biology of plants. Clearly, the relationship between the magnitude of selection coefficients and the degree to which plants are pollen vs. resource limited deserve further attention.
Because standardized selection differentials reported in this study were high (
0.48; Table 3), it may appear that genetic variation would disappear after a few generations. However, because T. erectum and T. grandiflorum are long-lived perennials, selection in one year may be very different from selection in another year. Many studies have found significant selection coefficients in some years of study but not in others (e.g., Campbell, 1989, 1991
; Scheiner, 1989
; Schemske and Horovitz, 1989
; Widén, 1991
). In addition, many studies have found opposing selective forces on the same characters (e.g., Campbell, 1989
). In light of this evidence, it is unlikely that genetic variation would disappear after only a few generations of selection in T. erectum or T. grandiflorum.
In this study, I measured selection acting on phenotypic characters through female reproductive success. However, T. erectum and T. grandiflorum are hermaphroditic. Therefore, selection arising from phenotypic variation in plant characters may act through male function (seeds sired) as well as female function (seeds produced; for a review see Stanton et al., 1992
). There may be a close correlation between seeds produced and seeds sired (e.g., Dudash, 1991
; O'Connell and Johnston, 1998
), or male and female fitness may be negatively or only weakly correlated (e.g., Bertin, 1982
; Stanton, Snow, and Handel, 1986
; Ennos and Dodson, 1987
; Schlichting and Devlin, 1989
; Broyles and Wyatt, 1990
; Devlin and Ellstrand, 1990
). I am currently examining the effects of phenotypic variation in T. erectum and T. grandiflorum on male plant function.
Phenotypic selection acting through female function in T. erectum and T. grandiflorum is occurring. Nonetheless, I cannot make definitive conclusions about the underlying mechanisms (Mitchell-Olds and Shaw, 1987
; Wade and Kalisz, 1990
). However, two mechanisms are likely. First, larger plants may have a fitness advantage over smaller ones because they may have more resources with which to produce seeds. For example, T. erectum respond to higher resources by producing larger flowers with more ovules (Davis, 1981
). In this study, sites in deciduous forests with high soil moisture produced larger plants that in turn produced more seeds than sites in mixed deciduous/coniferous forests. Trillium bloom before deciduous trees produce leaves. Therefore, sites in deciduous forests receive more sunlight as well, and photosynthetic input is important for vegetative growth and subsequent fruit and seed production in T. erectum and T. grandiflorum (Reader and Bricker, 1991
).
Second, larger plants may be favored over smaller ones if they attract more pollinators and donate and/or receive more pollen (e.g., Galen and Newport, 1987
; Galen, 1989
; Johnson, Delph, and Elderkin, 1995
; Andersson, 1996
; Totland et al., 1998
; Vaughton and Ramsey, 1998
). Because T. erectum and T. grandiflorum experience some pollen limitation, increased pollinator visitation and pollen receipt should translate into increased seed production. Selection acting through female-fitness components should increase as the degree of pollen limitation increases (Johnston, 1991a, b
). Preliminary data suggest that pollinators are attracted to taller T. erectum and T. grandiflorum with larger floral characters (R. E. Irwin, unpublished data); therefore, pollinator behavior may play an important role in selection of larger morphological characters in T. erectum and T. grandiflorum.
The mating system and phenotypic variation of plants can have a profound effect upon reproductive success and phenotypic selection of plant characters (for reviews see Lloyd and Barrett, 1996
; Proctor, Yeo, and Lack, 1996
). Although T. erectum and T. grandiflorum are common spring ephemerals throughout eastern North America, few studies have examined the association among the mating system, phenotypic variation, and reproductive output of these species (but see Davis, 1981
; Kawano, Ohara, and Utech, 1986
; Ohara, 1989
). In the populations I examined, I found that both species were self-compatible, had low pollinator-visitation rates, and were pollen limited in fruit and/or seed production. In addition, plant characters were positively associated with female reproductive success, and selection acted on plant size but not plant shape. Emerging from this study is a general pattern prominent in the literature, that larger plants have higher reproductive output. Future work will use manipulative experiments to examine the role of resource vs. pollen limitation in fruit and seed production in relation to plant size.
FOOTNOTES
1 The author thanks T. Irwin and H. Irwin for help in the field and A. Brody, S. Broyles, S. Kawano, R. Mitchell, and A. Seidl for comments on the manuscript. This work was funded by a research grant from the Central Vermont Audubon Society and a training grant from the National Science Foundation. 
2 E-mail: (rirwin{at}zoo.uvm.edu
); Tel: (802-656-0703); Fax (802-656-2914).
LITERATURE CITED
Abrahamson, W. G., and B. J. Hershey. 1977 Resource allocation and growth of Impatiens capensis (Balsaminaceae) in two habitats. Bulletin of the Torrey Botanical Club 104: 160–164. [CrossRef][ISI]
Andersson, S. 1996 Floral display and pollination success in Senecio jacobaea (Asteraceae): interactive effects of head and corymb size. American Journal of Botany 83: 71–75. [CrossRef][ISI]
Antonovics, J., K. Clay, and J. Schmitt. 1987 The measurement of small-scale environmental heterogeneity using clonal transplants of Anthoxanthum odoratum and Danthonia spicata. Oecologia 71: 601–607.
Arnold, S. J., and M. J. Wade. 1984a On the measurement of natural and sexual selection: theory. Evolution 38: 709–719. [CrossRef][ISI]
———. 1984b On the measurement of natural and sexual selection: applications. Evolution 38: 720–734. [CrossRef][ISI]
Augspurger, C. K. 1981 Reproductive synchrony of a tropical shrub: experimental studies on the effects of pollinators and seed predators on Hybanthus prunifolius (Violaceae). Ecology 62: 319–326. [CrossRef][ISI]
Barrett, S. C. H. 1987 The evolution, maintenance and loss of self-compatibility systems. In J. Lovett-Doust and L. Lovett-Doust [eds.], Reproductive strategies of plants. Oxford University Press, Oxford, UK.
———, and J. D. Thomson. 1982 Spatial pattern, floral sex ratios, and fecundity in dioecious Aralia nudicaulus (Araliaceae). Canadian Journal of Botany 60: 1662–1670. [CrossRef]
Beattie, A. J., D. E. Breedlove, and P. R. Ehrlich. 1973 The ecology of the pollinators and predators of Frasera speciosa. Ecology 54: 81–91.
Bennington, C. C., and J. B. McGraw. 1995 Natural selection and ecotypic differentiation in Impatiens pallida. Ecological Monographs 65: 303–323.
Berg, R. L. 1960 The ecological significance of correlation pleiades. Evolution 14: 171–180. [CrossRef][ISI]
Bertin, R. I. 1982 Paternity and fruit production in trumpet creeper (Campsis radicans). American Naturalist 119: 694–709. [CrossRef][ISI]
Brody, A. K. 1992 Oviposition choices by a pre-dispersal seed predator (Hylemya sp.). I. Correspondence with hummingbird pollinators, and the role of plant size, density and floral morphology. Oecologia 91: 56–62. [ISI]
Broyles, S. B., S. L. Sherman-Broyles, and P. Rogati. 1997 Evidence of outcrossing in Trillium erectum and Trillium grandiflorum (Liliaceae). Journal of Heredity 88: 325–329.
———, and R. Wyatt. 1990 Paternity analysis in a natural population of Asclepias exaltata: multiple paternity, functional gender, and the "pollen-donation hypothesis." Evolution 44: 1454–1468. [CrossRef][ISI]
Calvo, R. N., and C. C. Horovitz. 1990 Pollinator limitation, cost of reproduction, and fitness in plants: a transition-matrix demographic approach. American Naturalist 136: 499–516. [CrossRef][ISI]
Campbell, D. R. 1989 Measurements of selection in a hermaphroditic plant: variation in male and female pollination success. Evolution 43: 318–334. [CrossRef][ISI]
———. 1991 Effects of floral traits on sequential components of fitness in Ipomopsis aggregata. American Naturalist 137: 713–737.
———, and K. J. Halama. 1993 Resource and pollen limitations to lifetime seed production in a natural plant population. Ecology 74: 1043–1051. [CrossRef][ISI]
Carter, A. 1892 Notes on pollination. Botanical Gazette 17: 19–22. [CrossRef]
Clausen, J. 1951 Stages in the evolution of plant species. Cornell University Press, Ithaca, New York, USA.
Cody, R. P., and J. K. Smith. 1997 Applied statistics and the SAS programming language, 4th edition. Prentice Hall, Upper Saddle River, New Jersey, USA.
Davis, M. A. 1981 The effect of pollinators, predators and energy constraints on the floral ecology and evolution of Trillium erectum. Oecologia 48: 400–406.
Devlin, B., and N. C. Ellstrand. 1990 Male and female fertility variation in wild radish, a hermaphrodite. American Naturalist 136: 87–107. [CrossRef][ISI]
Dudash, M. 1991 Plant size effects on female and male function in hermaphroditic Sabatia angularis (Gentianaceae). Ecology 73: 1004–1012.
Ennos, R. A., and R. K. Dodson. 1987 Pollen success, functional gender and assortative mating in an experimental plant population. Heredity 58: 119–126. [ISI]
Faegri, K., and L. van der Pijl. 1978 The principles of pollination ecology. Pergamon, New York, New York, USA.
Fisher, R. A. 1930 The genetical theory of natural selection. Clarendon Press, Oxford, UK.
Fryxell, P. A. 1957 Mode of reproduction in higher plants. Botanical Review 23: 135–233. [ISI]
Fukuda, I., and W. F. Grant. 1980 Chromosome variation and evolution in Trillium grandiflorum. Canadian Journal of Genetic Cytology 22: 81–91.
Galen, C. 1989 Measuring pollinator-mediated selection on morphometric floral traits: bumblebees and the alpine sky pilot, Polemonium viscosum. Evolution 43: 882–890.
———. 1996 The evolution of floral form: insights from an alpine wildflower, Polemonium viscosum (Polemoniaceae). In D. G. Lloyd and S. C. H. Barrett [eds.], Floral biology: studies on floral evolution in animal-pollinated plants, 273–291. Chapman and Hall, New York, New York, USA.
———, and M. E. A. Newport. 1987 Bumble bee behavior and selection on flower size in the sky pilot, Polemonium viscosum. Oecologia 74: 20–23.
———, R. C. Plowright, and J. D. Thomson. 1985 Floral biology and regulation of seed set and seed size in the lily Clintonia borealis (Ait.) Raf. American Journal of Botany 72: 1544–1552. [CrossRef][ISI]
Gates, B. N. 1941 Observations in 1940 on the dissemination by ants of the seeds of Trillium grandiflorum. Rhodora 43: 206–207.
Grant, V., and K. A. Grant. 1965 Flower pollination in the Phlox family. Columbia University Press, New York, New York, USA.
Gunther, R. W., and J. Lanza. 1989 Variation in attractiveness of Trillium diaspores to a seed-dispersing ant. American Midland Naturalist 122: 321–328. [CrossRef][ISI]
Haig, D., and M. Westoby. 1988 On limits to seed production. American Naturalist 131: 757–759. [CrossRef][ISI]
Hainsworth, F. R., L. L. Wolf, and T. Mercier. 1984 Pollination and predispersal seed predation: net effects on reproduction and inflorescence characteristics in Ipomopsis aggregata. Oecologia 63: 405–409.
Hanzawa, F. M., and S. Kalisz. 1993 The relationship between age, size, and reproduction in Trillium grandiflorum (Liliaceae). American Journal of Botany 80: 405–410. [CrossRef][ISI]
Herrera, C. M. 1993 Selection on floral morphology and environmental determinants of fecundity in a hawkmoth-pollinated violet. Ecological Monographs 63: 251–275. [CrossRef]
———. 1996 Floral traits and plant adaptation to insect pollinators: a devil's advocate approach. In D. G. Lloyd and S. C. H. Barrett [eds.], Floral biology: studies on floral evolution in animal-pollinated plants, 65–87. Chapman and Hall, New York, New York, USA.
Heywood, J. S., and D. A. Levin. 1986 Interactions between seed source, planting arrangement, and soil treatment in determining plant size and root allocation in Phlox drummondii. Oecologia 68: 285–290.
Jeffrey, E. C., and E. J. Haertl. 1939 Apomixis in Trillium. Cellule 48: 79–88.
Johnson, S. G., L. F. Delph, and C. L. Elderkin. 1995 The effect of petal-size manipulation on pollen removal, seed set, and insect-visitor behavior in Campanula americana. Oecologia 102: 174–179.
Johnston, M. O. 1991a Natural selection on floral traits in two species of Lobelia with different pollinators. Evolution 45: 1468–1479. [CrossRef][ISI]
———. 1991b Pollen limitation of female reproduction in Lobelia cardinalis and L. siphilitica. Ecology 72: 1500–1503.
Kawano, S., A. Hiratsuka, and K. Hayashi. 1982 Life history characteristics and survivorship of Erythronium japonicum. Oikos 38: 129–149.
———, M. Ohara, and F. H. Utech. 1986 Life history studies on the genus Trillium (Liliaceae). II. Reproductive biology and survivorship of four eastern North American species. Plant Species Biology 1: 47–58.
Kelly, C. A. 1992 Spatial and temporal variation in selection on correlated life-history traits and plant size in Chamaecrista fasciculata. Evolution 46: 1658–1673.
Lande, R. 1979 Quantitative genetic analysis of multivariate evolution, applied to brain:body size allometry. Evolution 33: 402–416. [CrossRef][ISI]
———, and S. J. Arnold. 1983 The measurement of selection on correlated characters. Evolution 37: 1210–1226. [CrossRef][ISI]
Lloyd, D. G., and S. C. H. Barrett. 1996 Floral biology: studies on floral evolution in animal-pollinated plants. Chapman and Hall, New York, New York, USA.
Lubbers, A. E., and M. J. Lechowicz. 1989 Effects of leaf removal on reproduction vs. belowground storage in Trillium grandiflorum. Ecology 70: 85–96.
Mitchell-Olds, T., and R. G. Shaw. 1987 Regression analysis of natural selection: statistical inference and biological interpretation. Evolution 41: 1149–1161. [CrossRef][ISI]
O'Connell, L. M., and M. O. Johnston. 1998 Male and female pollination success in a deceptive orchid, a selection study. Ecology 79: 1246–1260. [CrossRef][ISI]
Ohara, M. 1989 Life history evolution in the genus Trillium. Plant Species Biology 4: 1–28.
———, and S. Higashi. 1987 Interference by ground beetles with the dispersal by ants of seeds of Trillium species (Liliaceae). Journal of Ecology 75: 1091–1098. [CrossRef][ISI]
———, and S. Kawano. 1986 Life history studies on the genus Trillium (Liliaceae). I. Reproductive biology of four Japanese species. Plant Species Biology 1: 35–45.
Platenkamp, G. A. J., and T. C. Foin. 1990 Costs of reproduction in the pink lady's slipper orchid: a four-year experimental study. American Naturalist 136: 638–656. [CrossRef][ISI]
Proctor, M., P. Yeo, and A. Lack. 1996 The natural history of pollination. Timber Press, Portland, Oregon, USA.
Reader, R. J., and B. D. Bricker. 1991 Response of five deciduous forest herbs to partial canopy removal and patch size. American Midland Naturalist 127: 149–157. [CrossRef][ISI]
Real, L. A. 1983 Pollination biology. Academic Press, New York, New York, USA.
Rencher, A. C. 1995 Methods of multivariate analysis. John Wiley & Sons, New York, New York, USA.
Ringius, G. S., and J. G. Chmielewski. 1987 Morphological variation within and among six populations of Trillium erectum in southern Ontario. Canadian Journal of Botany 65: 2450–2457.
Robertson, C. 1896 Flowers and insects. Botanical Gazette 21: 266–274. [CrossRef]
SAS. 1985a SAS user's guide: basics, 5th ed. SAS Institute Inc., Cary, North Carolina, USA.
———. 1985b SAS user's guide: statistics, 5th ed. SAS Institute Inc., Cary, North Carolina, USA.
Scheiner, S. M. 1989 Variable selection along a successional gradient. Evolution 43: 548–562. [CrossRef][ISI]
Schemske, D. W., and C. C. Horovitz. 1989 Temporal variation in selection on a floral character. Evolution 43: 461–465. [CrossRef][ISI]
———, M. F. Willson, M. N. Melampy, L. J. Miller, L. Verner, K. M. Schemske, and L. B. Best. 1978 Flowering ecology of some spring woodland herbs. Ecology 59: 351–366. [CrossRef][ISI]
Schlichting, C. D., and B. Devlin. 1989 Male and female reproductive success in the hermaphroditic plant Phlox drummondii. American Naturalist 133: 212–227. [CrossRef]
Schwaegerle, K. E., and D. A. Levin. 1990 Environmental effects on growth and fruit production in Phlox drummondii. Journal of Ecology 78: 15–26.
Seavey, S. R., and K. S. Bawa. 1986 Late-acting self-incompatibility in angiosperms. Botanical Review 52: 195–219.
Serota, C. A., and B. W. Smith. 1967 The cyto-ecology for four species of Trillium from western North Carolina. American Journal of Botany 54: 169–181. [CrossRef][ISI]
Stanton, M. L., T. Ashman, L. F. Galloway, and H. J. Young. 1992 Estimating male fitness of plants in natural populations. In R. Wyatt [ed.], Ecology and evolution of plant reproduction: new approaches, 62–90. Chapman and Hall, New York, New York, USA.
———, J. Bereczky, and H. D. Hasbrouck. 1987 Pollination thoroughness and maternal yield regulation in wild radish, Raphanus raphanistrum (Brassicaceae). Oecologia 74: 68–76. [CrossRef][ISI]
———, A. A. Snow, and S. N. Handel. 1986 Floral evolution: attractiveness to pollinators increases male fitness. Science 232: 1625–1627.
Stewart, S. C., and D. J. Schoen. 1987 Pattern of phenotypic viability and fecundity selection in a natural population of Impatiens pallida. Evolution 41: 1290–1301.
Thompson, J. N. 1983 Partitioning of variance in demography: within-patch differences in herbivory, survival, and flowering of Lomatium farinosum (Umbelliferae). Oikos 40: 315–317. [CrossRef][ISI]
Totland, Ø., H. L. Andersen, T. Bjelland, V. Dahl, W. Eide, S. Houge, T. R. Pedersen, and E. U. Vie. 1998 Variation in pollen limitation among plants and phenotypic selection on floral traits in an early-spring flowering herb. Oikos 82: 491–501. [CrossRef][ISI]
Vaughton, G. O., and M. Ramsey. 1998 Floral display, pollinator visitation and reproductive success in the dioecious perennial herb Wurmbea dioica (Liliaceae). Oecologia 115: 93–101. [CrossRef][ISI]
Wade, M. J., and S. Kalisz. 1990 The causes of natural selection. Evolution 44:1947–1955.
Waser, N. M. 1983 The adaptive nature of floral traits: ideas and evidence. In C. E. Jones and R. J. Little [eds.], Handbook of experimental pollination biology, 277–293. Van Nostrand-Reindhold, New York, New York, USA.
Weed, C. M. 1900 Ten New England blossoms and their insect visitors. Houghton & Mifflin, Boston, Massachusetts, USA.
Widén, B. 1991 Phenotypic selection of flowering phenology in Senecio integrifolius, a perennial herb. Oikos 61: 205–215. [CrossRef][ISI]
Wolfe, L. M. 1983 The effect of plant size on reproductive characteristics in Erythronium americanum (Liliaceae). Canadian Journal of Botany 61: 3489–3493. [CrossRef]
Zimmerman, M. 1980 Reproduction in Polemonium: pre-dispersal seed predation. Ecology 61: 502–506. [CrossRef][ISI]
This article has been cited by other articles:
|
|
 |
|
  R. E. Irwin and L. S. Adler Correlations among traits associated with herbivore resistance and pollination: implications for pollination and nectar robbing in a distylous plant Am. J. Botany, January 1, 2006; 93(1): 64 - 72. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. C. Vance, P. Bernhardt, and R. M. Edens Pollination and seed production in Xerophyllum tenax (Melanthiaceae) in the Cascade Range of central Oregon Am. J. Botany, December 1, 2004; 91(12): 2060 - 2068. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Knight Effects of herbivory and its timing across populations of Trillium grandiflorum (Liliaceae) Am. J. Botany, August 1, 2003; 90(8): 1207 - 1214. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-C. Routhier and L. Lapointe Impact of tree leaf phenology on growth rates and reproduction in the spring flowering species Trillium erectum (Liliaceae) Am. J. Botany, March 1, 2002; 89(3): 500 - 505. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. A. Wilson Floral stages, ovule development, and ovule and fruit success in Iris tenax, focusing on var. gormanii, a taxon with low seed set Am. J. Botany, December 1, 2001; 88(12): 2221 - 2231. [Abstract] [Full Text] [PDF]
|
|
