| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Botany, University of Washington, Box 355325, Seattle, Washington 98195-5325
Received for publication June 9, 1998. Accepted for publication December 11, 1998.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Linanthus • parviflorus • Polemoniaceae • pollen limitation • reproductive assurance • self-incompatibility • wind pollination
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Primack and Inouye, 1993
; Parker, 1997
), and low visitation can limit reproduction in animal-pollinated species, despite considerable investment in floral displays. In a recent survey of the literature, female reproductive success was found to be significantly pollen limited in at least some years or sites in 62% of the 268 species studied (Burd, 1994
).
Though outcrossing predominates in the angiosperms, self-fertilization also is common and has evolved from outcrossing in numerous plant taxa (Stebbins, 1974
). Selection for reproductive assurance is thought to be one of the major factors involved in the evolutionary transformation from animal pollination to self-fertilization (Stebbins, 1957
). When pollinators are scarce or absent, plants that can self-fertilize autonomously (without the service of a pollinator) will produce more seeds than obligately outcrossing plants. If this increase in seed set outweighs any cost to offspring quality due to inbreeding depression, mutations that confer selfing will be favored. Selection for reproductive assurance is expected to be especially important for annual plant species that have only a single opportunity for reproduction (Stebbins, 1950
; Baker, 1959
). The observed association between selfing and the annual life history may reflect the importance of reproductive assurance for annual species (Lloyd, 1980
; Barrett and Eckert, 1990
).
Self-fertilization is thought to represent an evolutionary solution to pollen limitation of reproduction in many animal-pollinated plant lineages (reviewed in Barrett, 1988
; Wyatt, 1988
). Wind pollination might be viewed as a potential alternative solution, providing reproductive assurance in the face of variation in pollinator abundance. A few flowering plant families are predominantly wind pollinated (e.g., Poaceae, Cyperaceae, and Plantaginaceae) and are characterized by a suite of distinctive floral traits adapted for this mode of pollination (Whitehead, 1969
; Faegri and van der Pijl, 1979
), but in addition to the discrete wind and animal pollination syndromes, a number of studies suggest that some plant taxa may undergo a combination of wind and insect pollination (Pojar, 1973
; Lock and Hall, 1982
, Stelleman, 1984
; Gomez and Zamora, 1996
). An association between wind pollination and colonization of areas with low insect abundance has been noted in two primarily insect-pollinated plant taxa (Berry and Calvo, 1989
; Gomez and Zamora, 1996
) and supports the hypothesis that wind pollination may provide reproductive assurance. The frequency of these mixed pollination systems in the angiosperms and the contribution of wind pollination to fitness in these species remains largely unknown.
Here I present evidence for wind pollination in Linanthus parviflorus (Polemoniaceae), a predominantly insect-pollinated, self-incompatible annual and suggest that wind pollination provides reproductive assurance in this species. The observation of extremely low pollinator visitation in some years in populations of L. parviflorus (Goodwillie, 1997b
) suggests that pollen limitation is expected to be a critical problem for this obligately outcrossing species. Phylogenetic analysis of ITS (internal transcribed spacer) rDNA sequence data indicates that selfing has evolved several times in related Linanthus species (Goodwillie, in press
), and selection for reproductive assurance is likely to have been an important factor in those evolutionary transformations. Because wind pollination in L. parviflorus can provide seed set in the absence of pollinators, it may represent an alternative means of reproductive assurance and allow for the persistence of L. parviflorus populations when pollinators are scarce.
This study provides several lines of evidence for wind pollination and its contribution to reproduction in L. parviflorus. The extent of wind pollination was experimentally tested in field manipulations replicated at two sites, and airborne L. parviflorus pollen was shown to be present in populations of the species. In addition, upper and lower plots in two populations of L. parviflorus, representing contrasting degrees of wind exposure, were compared to explore the relationship between the extent of pollen limitation and the abundance of airborne pollen, providing supporting evidence that wind pollination confers reproductive assurance.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Plants are 10–20 cm tall, with a terminal headlike inflorescence. The pink or white five-petalled flowers have a slender corolla tube, 
2.5 cm in length. Flowers are visited nearly exclusively by long-tongued beeflies (e.g., Bombylius spp.) that hover and probe the long corolla tube for nectar (Grant and Grant, 1965
). Populations typically are large, with densities as high as 2400 plants/m2. The flowering period extends from early April to early June, and seeds mature within 30 d. Self-fertilization is prevented in L. parviflorus by a highly effective sporophytic self-incompatibility system (Goodwillie, 1997a
).
Study sites
All experiments and observations were carried out in the spring of 1996 in L. parviflorus populations at two study sites in Napa County, California, separated by 6 km. The sites at Wantrup Reserve and Mast Hill are ecologically similar, characterized by grasslands and oak woodlands. An experiment investigating wind pollination was replicated at two plots within the Wantrup Reserve. Pollen limitation experiments were carried out in two plots at both the Wantrup Reserve and Mast Hill.
Airborne pollen counts
Pollen traps were used to test for the presence of airborne L. parviflorus pollen and to provide a relative measure of its density in all experimental plots. Glass microscope slides coated with vaseline and attached to wire stakes were placed in plots at flower level between 0900 and 1000 before flowers opened and collected between 1800 and 1900 after the flowers closed for the night. Pollen grains were stained with aniline blue in lactophenol (Kearns and Inouye, 1993
), and counts were made of the number of grains deposited in the area of two coverslips (9.68 cm2) in a day. (For simplicity, this unit of measurement will be referred to as "pollen grains per slide"). Pollen was identified by reference to stained preparations of known L. parviflorus pollen. At each experimental plot, four pollen traps were set out on 4 d during peak flowering at randomly chosen locations within the population. Sampling days were consecutive when weather permitted, but on overcast or rainy days flowers remain closed and sampling was not possible. Because of slight differences in flowering phenology, sampling dates differed among plots.
For comparison, airborne pollen was quantified using the same methods and sampling protocol in two populations of a closely related species, L. jepsonii, adjacent to the study populations of L. parviflorus at Wantrup Reserve and Mast Hill. L. jepsonii has a similar morphology and flowering phenology and is visited by the same pollinators, but is self-compatible and partially selfing.
Wind pollination experiment
The capacity for wind pollination in L. parviflorus and its effectiveness relative to insect pollination were assessed by quantifying seed set in plants enclosed in mesh tents that exclude pollinators but allow passage of airborne pollen (see below) and comparing it to seed set in open-pollinated control plants. To test for possible confounding effects of the tent unrelated to pollination, seed set in two additional treatments, pollen supplementation and pollen supplementation with tent, were compared. Plants were subjected to these four treatments throughout the entire flowering period: (1) control (open pollination), (2) tent (wind pollination), (3) pollen supplementation, and (4) pollen supplementation with tent. A comparison of seed set in control and pollen-supplemented treatments provided an indication of the extent to which reproductive success was limited by pollen deposition in the population.
In tented treatments, entire plants were enclosed in a wire-framed tent stretched with 1-mm nylon mesh. In the pollen-supplemented treatments, pollen was applied to all open flowers by brushing the stigmas with the anthers of flowers from two donor plants. Pollen supplementation treatments were carried out on alternate days. Since flowers remain open for >2 d (unpublished data) this was sufficient to ensure that all flowers were treated. The experiment was replicated at two plots at the Wantrup Reserve, separated by 500 m. At each plot, 30 groups of four adjacent plants of similar size (blocks) were marked at the start of the flowering season, and individual plants were randomly assigned to one of the four treatments. Fruits were collected as they matured, and total seeds were counted for each plant. The number of flowers produced by each plant was determined by counting calyces at the end of the season.
To test whether pollen flow was affected by the mesh tents, airborne pollen was quantified inside and outside tents. At a site adjacent to one of the experimental plots, eight pollen traps were set out on 6 d during peak flowering. On each day, four of the traps were randomly selected to be enclosed in pollinator exclusion tents. Traps were constructed and pollen grains were counted as described above.
Self-incompatibility
Linanthus parviflorus from populations in Napa County have been shown to be self-incompatible in greenhouse experiments (Goodwillie, 1997a
). To determine whether L. parviflorus is fully self-incompatible under field conditions, pollen tube growth was compared in hand self- and outcross-pollinations on plants in the wind pollination experimental plots. Pollen tube assays are thought to be the best indication of self-incompatibility, since variation in seed set can be attributed to a number of other factors such as inbreeding depression or patterns of resource allocation (Weller, Donoghue, and Charlesworth, 1995
). At each experimental plot, 30 randomly selected plants were enclosed in cardboard cylinders covered with tightly meshed fabric that excluded both wind- and insect-borne pollen but allowed light. On each plant, one flower was hand self-pollinated and one was hand cross-pollinated. Flowers were removed after they senesced, the style was fixed in 3:1 ethanol-acetic acid, and pollen tubes were inspected using the aniline blue staining method (Martin, 1959
).
Pollen limitation in exposed and protected plots
Additional evidence for wind pollination in L. parviflorus and its role in reproductive assurance was obtained by exploring the relationship between pollen limitation and the amount of airborne pollen in plots that differ in wind exposure. If plants in exposed areas undergo more wind pollination, they are expected to show lower pollen limitation than areas protected from wind. Populations of L. parviflorus extending from windy hilltops to lower, protected glades were studied at the Wantrup Reserve and Mast Hill.
To assess experimentally the magnitude of pollen limitation, 30–60 pairs of adjacent plants of similar size were marked in each of two plots at both sites, one in an upper, exposed area and one in a lower, more protected area. The exposed and protected plots at each site were separated by <15 m. One plant in each pair was randomly assigned to a pollen supplementation treatment, and the other served as an unmanipulated control. On pollen-supplemented plants, all flowers were hand-pollinated, as described above for the wind pollination experiment. Fruits of both hand-pollinated and control plants were collected when mature, and seeds and flowers were counted for each plant. Airborne pollen loads were quantified in the exposed and protected plots at both sites using the pollen trap technique and sampling protocol described above.
Data analysis
All statistical analyses were performed on SPSS version 7.5 (SPSS, 1997
). The most relevant variable for comparison of treatments in both the wind pollination and pollen limitation experiments was the total number of seeds per plant, because this is a direct measure of total female fitness. However, a composite variable, seeds per flower, controls for variation in flower number and therefore may be more useful for detecting differences among treatments. All analyses were carried out for both dependent variables.
Wind pollination experiment
To statistically test for the occurrence of wind pollination in experimental plants, one-tailed one-sample t tests were used to determine whether seeds per plant and seeds per flower in tented treatments were significantly greater than zero. To test for an overall effect of pollination treatments on seeds per plant and seeds per flower, two-way nested ANOVA was used with plot as a random factor, block nested within plot, and treatment as a fixed factor. Error terms for this and all other mixed-model ANOVAs in the study were calculated using a linear combination of variance components of the random effects and quadratic terms of the fixed effects (SPSS, 1997
). Comparisons of treatment pairs relevant to individual experimental questions (see Experimental design) were tested for significance using Scheffé's post hoc tests. The same analysis of variance model was used to test for differences in flower number.
Two-way ANOVA, with tent as a fixed factor and sampling date as a random factor, was used to test for an effect of the tent on pollen flow. The quantity of airborne pollen (log transformed) at the two experimental plots was compared using nested one-way ANOVA, with plot as a random factor and sampling date nested within plot. ANOVA was used to compare airborne pollen in L. jepsonii and L. parviflorus, with species as a fixed effect, site as a random effect, and sampling date nested within the interaction.
Pollen limitation in exposed and protected plots
Two-way nested ANOVA with site as a random factor, exposure (exposed vs. protected) as a fixed factor, and sampling date nested within the site by exposure interaction was used to test for differences in the mean airborne pollen count (log transformed). To test for significant pollen limitation at each plot (exposed and protected plots at Wantrup Reserve and Mast Hill), seeds per plant and seeds per flower in control and pollen supplementation treatments were compared using paired t tests, and significance levels were adjusted to hold the experimentwise Type I error rate at 0.05 using the sequential Bonferroni method (Rice, 1989
).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
10% of the overall mean for L. parviflorus populations.
Self-incompatibility
In self-pollinated flowers, pollen tubes were absent in all styles in plot 1 and in all but one style in plot 2, while dense pollen tube growth was observed in styles of all cross-pollinated flowers (plot 1: mean = 88.5 tubes, N = 30, plot 2: mean = 56.9 tubes, N = 30). These data indicate that self-fertilization in L. parviflorus at the experimental sites is prevented by a highly effective self-incompatibility system.
Wind pollination
The number of seeds per plant and seeds per flower in wind-pollinated treatments at each plot was significantly greater than zero (seeds per plant, plot 1: t = 7.606, df = 29, P < 0.0001, plot 2: t = 6.535, df = 27, P < 0.0001; seeds per flower, plot 1: t = 9.679, df = 29, P < 0.0001, plot 2: t = 9.060, df = 27, P < 0.0001). The number of flowers per plant differed significantly between experimental plots, but treatments were not significantly different for flower number (Table 1), suggesting that there was no bias in the assignment of plants to treatments. There was no significant interaction between plot and treatment effects for either seed set variable. Treatments had a highly significant effect on both the number of seeds per plant and seeds per flower. In post hoc tests, a comparison between pollen supplementation and pollen supplementation with tent treatments showed no significant difference in the number of seeds per flower or seeds per plant, although for both variables the means for tented plants was higher (Fig. 1). Pollen-supplemented plants produced significantly more seeds per flower than control plants, indicating pollen limitation of reproduction; the difference between these treatments in seeds per plant was marginally significant (Fig. 1). Wind-pollinated plants (tent treatment) produced significantly fewer seeds per flower than control plants, while the difference in seeds per plant was not significant (Fig. 1).
|
|
Pollen limitation at exposed and protected plots
Airborne pollen counts
Mean pollen counts ranged widely among the plots, from 161.2 to 497.9 pollen grains per slide (Fig. 2). Airborne pollen counts were not significantly different between Wantrup Reserve and Mast Hill, but the effect of wind exposure approached significance (P = 0.106) and there was significant variation among sampling dates (Table 2).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
At the experimental plots, insect and wind pollination together (control treatments) produced 70–80% of the number of seeds produced with pollen supplementation. Wind pollination alone produced substantial seed set, 50% of seed set with pollen supplementation. Mean airborne pollen counts at the wind experiment plots were less than half of that at the exposed plot at Mast Hill in the pollen limitation study, suggesting that wind pollination at some sites may be considerably more extensive than the results for the wind pollination experiment indicate. The lack of pollen tube growth after hand self-pollination of flowers in wind exclusion tubes indicates that seed set in tented plants was not due to self-fertilization. Furthermore, negative results for self-pollinated flowers in the wind exclusion tubes eliminate an alternative explanation, that pollination in tented plants was effected by crawling insects.
Reproductive assurance
In the pollen limitation experiment, plants in upper, exposed plots where airborne pollen loads were high were not significantly pollen limited, while those in both of the lower, protected plots were found to be significantly pollen limited for seed number per flower. Though these data do not provide direct evidence of wind pollination, they support the experimental finding of wind pollination in L. parviflorus. The observed differences in the magnitude of pollen limitation could also be explained by differences in pollinator visitation rates at exposed and protected plots. Data are not available to fully eliminate this possibility, but exposed and protected plots at each site were part of the same continuous population, separated by <15 m, and pollinator visitation is unlikely to, and has not been observed to, differ at this scale. Furthermore, adjacent exposed and protected plots were not distinguished by any obvious ecological differences that might account for differences in pollen limitation.
The suggested relationship between airborne pollen and pollen limitation provides supporting evidence that wind pollination is occurring and leads to further conclusions as well. First, it suggests that wind pollination provides reproductive assurance at some sites, and second, it indicates that wind pollination may vary widely among L. parviflorus sites. Significant pollen limitation was found at both protected plots, and marginally significant pollen limitation was found in the wind pollination experimental plots, indicating that neither insect visitation nor wind pollination was sufficient to effect full seed set at these sites. The lack of pollen limitation and higher pollen counts at exposed plots suggest that wind pollination provides full reproductive assurance. Congruent with these findings is the fact that open, exposed habitats are more typical of L. parviflorus, at least in the study area; however, the demonstration of wind pollination at the experimental plots indicates that, even at sites with relatively lower airborne pollen counts, wind pollination can contribute substantially to reproduction in the absence of pollinators.
Wind pollination may provide partial or full reproductive assurance in years with low pollinator abundance. In five years of study of L. parviflorus populations in Napa County, I observed great variation in pollinator visitation rate. In 1995, a year with extensive flooding in the early spring, beeflies were virtually absent. If crashes in pollinator populations are frequent, selection for reproductive assurance is likely to be strong and may have played a role in the evolutionary transition from obligate, relatively specialized insect pollination to self-fertilization that has occurred repeatedly in related species. Selfing has evolved from self-incompatibility at least three times within this section of Linanthus (Goodwillie, in press
), and the capacity for autonomous self-fertilization ensures that reproduction will occur even when pollinators are absent or scarce. Evidence presented here suggests that wind pollination serves a similar function in L. parviflorus. Wind pollination was found to provide substantial reproductive assurance and may allow for the persistence of populations of a self-incompatible annual under conditions of low pollinator visitation. Of the self-incompatible species in Linanthus section Leptosiphon, L. parviflorus is the most widespread and common, occurring in large populations throughout much of California (Hickman, 1993
). Furthermore, wind pollination in L. parviflorus may be hypothesized to contribute to the maintenance of self-incompatibility in the species; if wind pollination provides seed set in the absence of pollinators, it may reduce the selective advantage of selfing genotypes if they arise under pollinator-limited conditions.
Wind pollination and floral characters
If wind pollination contributes to the reproductive success of L. parviflorus, can we identify traits that represent adaptations for wind pollination? Many features of L. parviflorus flowers, like those of other species in the section, are clearly adapted for pollination by insects: showy corolla lobes, nectar-guide markings, and a long, nectar-containing corolla tube that receives the beefly's proboscis. However, a distinctive feature of L. parviflorus flowers, reflexed corolla lobes, may be associated with wind pollination. In other species in the section, the corolla lobes are held at right angles to the corolla tube and are cupped slightly upward (e.g., L. bicolor, L. jepsonii, L. acicularis), or form a trumpet-like arrangement, with the lobes at a greater than 90° angle from the tube (e.g., L. montanus, L. androsaceous, L. nudatus). Linanthus parviflorus is unique in that the corolla lobes are often reflexed, bent back to form an acute angle to the corolla tube. The reflexing increases with flower age, in some plants to an angle of <15°. An intriguing hypothesis is that reflexed corolla lobe position in L. parviflorus represents an adaptation that promotes wind pollination by removing structures that impede air flow and exposing the stigma and anthers to airborne pollen. The highly reduced corollas that characterize typical wind-pollinated taxa are thought to serve the same function (Whitehead, 1969
). Along with increased corolla lobe reflexing, stigma lobes elongate throughout the life of a flower of L. parviflorus, approximately doubling in length in 3 d in an unfertilized flower. Upon fertilization, stigma elongation stops and flower senescence occurs, usually within a day (personal observation). This phenomenon occurs to some degree in related self-compatible species, but is greatly exaggerated in L. parviflorus. Stigma elongation may be viewed as a possible adaptation to wind pollination as well, in that continual elongation of an unpollinated stigma may increase its chances for intercepting airborne pollen. Long, plumose stigmas are typical of many exclusively wind-pollinated groups, such as the grasses (Whitehead, 1969
; Faegri and van der Pijl, 1979
).
The Polemoniaceae are well known for extraordinary diversity of pollination syndromes and mating systems (Grant and Grant, 1965
), and this family has played an important part in the study of plant mating system evolution. Floral adaptations associated with animal pollination and autogamy have been studied in several species in the Polemoniaceae (Levin, 1972
; Schoen, 1982
; Galen and Stanton, 1989
; Campbell et al., 1991
). This study provides the first report of wind pollination in this family. Whether the phenomenon is unique to L. parviflorus or common to other species or genera in the Polemoniaceae is unknown. Indeed, the frequency of wind pollination in primarily insect-pollinated taxa has yet to be determined for higher plants. Evidence for mixed insect and wind pollination has been shown for a variety plant taxa. In Urginea maritima (Liliaceae), wind-, insect- and self-pollination all occur (Dafni and Dukas, 1986
). This mediterranean species produces nectar-bearing flowers during late summer, a season that is characterized by low insect abundance. Wind pollination is thought to result primarily in self-fertilization, which cannot occur in this species without a pollen vector and has been hypothesized to represent a means for reproductive assurance in an environment harsh to insects. Wind pollination has been demonstrated in populations of Hormathophylla spinosa (Cruciferae), a primarily ant-pollinated self-incompatible species and is thought to allow colonization and persistence of populations in high montane environments with low insect abundance (Gomez and Zamora, 1996
). Selection for reproductive assurance is hypothesized to have played a role in the evolution of wind pollination in high elevation species of Espeletia (Asteraceae), a genus of primarily insect-pollinated, exclusively self-incompatible species (Berry and Calvo, 1989
). Contributions of insect pollination to primarily wind-pollinated species have also been reported (Clifford, 1962
; Pojar, 1973
; Adams, Perkins, and Estes, 1981
; Lock and Hall, 1982
; Stelleman, 1984
). Plantago lanceolata (Plantaginaceae), typically regarded as a wind-pollinated species, has been shown to be partially insect pollinated at some sites (Clifford, 1962
; Stelleman, 1984
), with populations at those sites characterized by adhesive pollen hypothesized to facilitate transfer by insects (Stelleman, 1984
). Similarly, a grass species, Paspalum dilatatum, with conspicuous purple anthers and stigmas and relatively large, adhesive pollen grains, was shown to be pollinated by halictid bees in addition to wind (Adams, Perkins, and Estes, 1981
). Interestingly, in several of these studies (Pojar, 1973
; Lock and Hall, 1982
; Stelleman, 1984
), the evolution of traits promoting insect pollination was attributed to selection for reproductive assurance in environments in which air movement of pollen is limited, such as in the forest understory or wet environments.
Mixed wind and insect pollination has received relatively little empirical study, and the small number of existing reports may not reflect its frequency among angiosperm taxa. Wind pollination has undoubtedly remained unnoticed in many insect-pollinated taxa. The recognition of wind pollination as a selective factor may contribute to our understanding of the evolution of floral traits in some plant species. Alternatively, wind pollination in some insect-pollinated taxa may represent an ecological consequence of harsh abiotic conditions, rather than an adaptation (Gomez and Zamora, 1996
). But whether or not selection for this mode of pollination leads to adaptive character evolution, even a moderate degree of wind pollination may be important in the evolutionary trajectory of plant species and may have implications for the evolution of plant mating systems.
|
FOOTNOTES |
|---|
2 Current address: Department of Forest Sciences, University of British Columbia, 2424 Main Mall, Vancouver, BC, Canada V6T 1Z4.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Baker, H. G. 1959 Reproductive methods as factors in speciation in flowering plants. Cold Spring Harbor Symposia on Quantitative Biology 24: 177–190.[ISI][Medline]
Barrett, S. C. H. 1988 The evolution, maintenance, and loss of self-incompatibility systems. In J. Lovett Doust and L. Lovett Doust [eds.], Plant reproductive ecology, 98–124. Oxford University Press, New York, NY.
———, and C. G. Eckert. 1990 Variation and evolution of plant mating systems. In S. Kawano [ed.], Biological approaches and evolutionary trends in plants, 229–254. Academic Press, New York, NY.
Berry, P. E., and R. N. Calvo. 1989 Wind pollination, self-incompatibility, and altitudinal shifts in pollination systems in the high Andean genus Espeletia (Asteraceae). American Journal of Botany 76: 1602–1614.[CrossRef][ISI]
Burd, M. 1994 Bateman's principle and plant reproduction: the role of pollen limitation in fruit and seed set. Botanical Review 60: 83–139.[CrossRef][ISI]
Campbell, D. R., N. M. Waser, M. V. Price, E. A. Lynch, and R. Mitchell. 1991 Components of phenotypic selection: pollen export and flower corolla width in Ipomopsis aggregata. Evolution 45: 1458–1467.
Clifford, H. T. 1962 Insect pollination of Plantago lanceolata L. Nature 193: 196.
Dafni, A., and R. Dukas. 1986 Insect and wind pollination in Urginea maritima (Liliaceae). Plant Systematics and Evolution 154: 1–10.[CrossRef][ISI]
Faegri, K., and L. van der Pijl. 1979 The principles of pollination ecology, 3rd ed. Pergamon Press, Oxford.
Galen, C., and M. L. Stanton. 1989 Bumble bee pollination and floral morphology: factors influencing pollen dispersal in the alpine sky pilot, Polemonium viscosum (Polemoniaceae). American Journal of Botany 76: 419–426.[CrossRef][ISI]
Gomez, J. M., and R. Zamora. 1996 Wind pollination in high-mountain populations of Hormathophylla spinosa (Cruciferae). American Journal of Botany 83: 580–585.[CrossRef][ISI]
Goodwillie, C. 1997a The genetic control of self-incompatibility in Linanthus parviflorus (Polemoniaceae). Heredity 79: 424–432.[CrossRef][ISI]
———. 1997b Self-incompatibility and mating system evolution in Linanthus section Leptosiphon (Polemoniaceae). Ph.D. dissertation, University of Washington, Seattle, WA.
———. In press Multiple origins of self-compatibility in Linanthus section Leptosiphon (Polemoniaceae): phylogenetic evidence from ITS sequence data. Evolution.
Grant, V., and K. A. Grant. 1965 Flower pollination in the phlox family. Columbia University Press, New York, NY.
Hickman, J. C. 1993 [ed.]. The Jepson manual: higher plants of California. University of California Press, Berkeley, CA.
Horvitz, C. C., and D. W. Schemske. 1990 Spatiotemporal variation in insect mutualisms of a Neotropical herb. Ecology 71: 1085–1097.[CrossRef][ISI]
Kearns, C. A., and D. W. Inouye. 1993 Techniques for pollination biologists. University of Colorado Press, Niwot, CO.
Levin, D. A. 1972 Low frequency disadvantage in the exploitation of pollinators by corolla variants in Phlox. American Naturalist 104: 445–467.
Lloyd, D. G. 1980 Demographic factors and mating patterns in angiosperms. In O. T. Solbrig [ed.], Demography and evolution in plant populations, 67–88. Blackwell, Oxford, UK.
Lock, J. M., and J. B. Hall. 1982 Floral biology of Mallotus oppositifolius (Euphorbiaceae). Biotropica 14: 153–155.
Martin, F. W. 1959 Staining and observing pollen tubes in the style by means of fluorescence. Stain Technology 34: 125–128.[ISI][Medline]
Parker, I. M. 1997 Pollinator limitation of Cytisus scoparius (Scotch broom), an invasive exotic shrub. Ecology 78: 1457–1470.[ISI]
Pojar, J. 1973 Pollination of typically anemophilous salt marsh plants by bumble bees, Bombus terricola occidentalis Grne. American Midland Naturalist 89: 448–451.[CrossRef][ISI]
Primack, R. B., and D. W. Inouye. 1993 Factors affecting pollinator visitation rates: a geographical comparison. Current Science 65: 257–262.
Rice, W. 1989 Analyzing tables of statistical tests. Evolution 43: 223–225.[CrossRef][ISI]
Schoen, D. J. 1982 The breeding system of Gilia achilleifolia: variation in floral characteristics and outcrossing rate. Evolution 36: 352–360.[CrossRef][ISI]
SPSS. 1997 Advanced statistics 7.5. SPSS Inc., Chicago, IL.
Stebbins, G. L. 1950 Variation and evolution in plants. Columbia University Press, New York, NY.
———. 1957 Self-fertilization and population variability in higher plants. American Naturalist 41: 337–354.[CrossRef]
———. 1974 Flowering plants: evolution above the species level. Belknap Press, Cambridge, MA.
Stelleman, P. 1984 Reflections on the transition from wind pollination to ambophily. Acta Botanica Neerlandica 33: 497–508.[ISI]
Weller, S. G., M. J. Donoghue, and D. Charlesworth. 1995 The evolution of self-incompatibility in flowering plants: a phylogenetic approach. In P. C. Hoch and A. G. Stephenson [eds.], Experimental and molecular approaches to plant biosystematics, 355–382. Missouri Botanical Garden, St. Louis, MO.
Whitehead, D. R. 1969 Wind pollination in the angiosperms: evolutionary and environmental considerations. Evolution 23: 28–35.[CrossRef][ISI]
Wyatt, R. 1988 Phylogenetic aspects of evolution of self-pollination. In L. D. Gottlieb and S. K. Jain [eds.], Plant evolutionary biology, 109–131. Chapman and Hall, New York, NY.
This article has been cited by other articles:
|
|
 |
|
  R. Qu, X. Li, Y. Luo, M. Dong, H. Xu, X. Chen, and A. Dafni Wind-Dragged Corolla Enhances Self-Pollination: A New Mechanism of Delayed Self-Pollination Ann. Bot., November 1, 2007; 100(6): 1155 - 1164. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Mazer, H. Paz, and M. D. Bell Life history, floral development, and mating system in Clarkia xantiana (Onagraceae): do floral and whole-plant rates of development evolve independently? Am. J. Botany, December 1, 2004; 91(12): 2041 - 2050. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. R. NEAL and G. J. ANDERSON Does the 'Old Bag' Make a Good 'Wind Bag'?: Comparison of Four Fabrics Commonly Used as Exclusion Bags in Studies of Pollination and Reproductive Biology Ann. Bot., May 1, 2004; 93(5): 603 - 607. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Totland and M. Sottocornola Pollen limitation of reproductive success in two sympatric alpine willows (Salicaceae) with contrasting pollination strategies Am. J. Botany, June 1, 2001; 88(6): 1011 - 1015. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
