| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
2Archbold Biological Station, P.O. Box 2057, Lake Placid, Florida 33862 USA; 3Department of Biology, Indiana University, Bloomington, Indiana 47405 USA
Received for publication March 18, 2005. Accepted for publication October 27, 2005.
ABSTRACT
Hypothetically, a species with both cleistogamous (CL) flowers and delayed selfing chasmogamous (CH) flowers should display high levels of reproductive assurance because, over time, obligate selfing by CL flowers should reduce inbreeding depression and delayed selfing in CH flowers should compensate for the absence of outcross pollen. We used pollinator-exclusion experiments to investigate reproductive assurance in the CH flowers of Polygala lewtonii, an herb with a mixed mating system. We followed CH flowers from bud-break to flower/fruit abscission to quantify fruit initiation and maturation and rates of floral development. We also evaluated the efficacy of the selfing mechanism, conducted pollinator watches to assess the likelihood of pollinator limitation, and performed regression analysis to determine the effect of flower position on fruit production. Pollinator exclusion significantly reduced fruit initiation and maturation. Investigation of floral development demonstrated that the selfing mechanism is largely dysfunctional in CH flowers, indicating the failure of reproductive assurance. Low observed rates of insect visitation appear to contradict high rates of CH fruit production in open-pollinated plants, particularly given the rarity of delayed selfing. In both treatments, flower position significantly affected fruit initiation, suggesting a role for resource limitation in both pollinator-excluded and open-pollinated flowers.
Key Words: autogamy • delayed selfing • pollinator exclusion • Polygala lewtonii • Polygalaceae • reproductive assurance • selfing
At least a third of all flowering plants maintain mixed mating systems that permit both self- and cross-fertilization (Lande and Schemske, 1985
; Schemske and Lande, 1985
; Barrett and Eckert, 1990
; Barrett et al., 1996
; Richards, 1997
; Vogler and Kalisz, 2001
). Each mode of sexual reproduction entails advantages and disadvantages that have been the subject of investigation ever since Darwin (1876
, 1877
). While many plant species have developed traits that prevent or limit self-fertilization (dioecy, self-incompatibility, herkogamy, dichogamy, heterostyly), many others have developed traits that permit, promote, or ensure it (hermaphroditism, self-compatibility, cleistogamy, autogamy). Obligate outcrossing species may suffer partial or complete sexual reproductive failure due to scarcity or absence of mates or pollinators. Selfing may provide reproductive assurance (Darwin, 1876
; Lloyd, 1980, 1992; Holsinger, 1996
) by guaranteeing sexual reproduction in the absence of outcrossing. In addition, by avoiding the "cost of meiosis," selfing has an "automatic selection advantage" (Jain, 1976
) because selfed offspring contain two copies of maternal genes compared to only one for outcrossed offspring. The main disadvantage of selfing is inbreeding depression, due primarily to the expression of deleterious recessive alleles (Lloyd, 1980; Barrett and Harder, 1996
; Holsinger, 1996
; Johnston, 1998
).
Species with mixed mating systems may have chasmogamous (CH = open-pollinated) flowers only or a combination of CH and cleistogamous (CL = obligately selfing) flowers (Cruden and Lyon, 1989
; Richards, 1997
). At least 287 species in 56 angiosperm families have both CH and CL flowers (Lord, 1981
). In mixed mating system species that lack CL flowers, CH flowers generally have morphological or phenological adaptations to promote or at least allow selfing (Lloyd, 1979
, 1992
; Richards, 1997
). Selfing can occur within (autogamy) or between (geitonogamy) CH flowers on a plant, and autogamy may be either autonomous (no insect required) or facilitated (insect required) (Lloyd, 1979
, 1992
). In addition, autonomous crosses may precede flower opening, co-occur with insect-vectored cross-pollinations, or occur as a fail-safe mechanism to ensure fertilization in the absence of insect pollination ("delayed selfing" sensu Lloyd, 1992
).
Although the ability to spontaneously self-fertilize when mates or pollinators are in short supply may provide reproductive assurance (Darwin, 1877
; Lloyd, 1980; Holsinger, 1996
), its value is undercut if pollen or ovules that would otherwise have been involved in outcrosses are preempted (referred to respectively as pollen and seed discounting; Lloyd, 1992
; Holsinger, 1996
; Herlihy and Eckert, 2002
), or if selfed progeny are significantly inferior to outcross progeny due to inbreeding depression (Charlesworth and Charlesworth, 1987
; Holsinger, 1991
, 1996
; Uyenoyama et al., 1993
; Barrett and Harder, 1996
; Herlihy and Eckert, 2002
). Thus, geitonogamy, selfing between flowers on the same plant, provides no reproductive assurance because it involves both pollen and seed discounting (de Jong et al., 1993
). Facilitated autogamy, insect-mediated selfing within a flower, usually does not provide reproductive assurance for the same reason (Schoen and Lloyd, 1992
; Eckert, 2000
). In contrast, delayed spontaneous selfing (autonomous autogamy) generally provides a large measure of reproductive assurance because it occurs only after flowers have had the opportunity for outcrossing (Lloyd and Schoen, 1992
; Barrett and Harder, 1996
; Johnston, 1998
).
Most studies of reproductive assurance have relied on species that produce only CH flowers, in which selfing is facultative and perhaps rare (Motten, 1982
; Dole, 1992
; Klips and Snow, 1997
; Pannell and Barrett, 1998
; Eckert and Schaefer, 1998
; Mallick, 2001
; Fausto et al., 2001
; Kalisz and Vogler, 2003
). In such species, inbreeding depression may be a frequent problem, thus undercutting the advantages of selfing (Herlihy and Eckert, 2002
). In contrast, in species with both CL and CH flowers, inbreeding depression may be attenuated due to purging of deleterious recessives (Schoen and Lloyd, 1984
; Charlesworth and Charlesworth, 1987
; Barrett and Charlesworth, 1991
; Barrett and Harder, 1996
; Husband and Schemske, 1996
; Byers and Waller, 1999
; Keller and Waller, 2002
). In fact, recent studies of plants with CH/CL flowers have generally found low levels of inbreeding depression (Schmitt and Ehrhardt, 1990
; Culley, 2000
, 2002
). However, recent studies have also shown that even species that routinely self can maintain high levels of inbreeding depression (Lande et al., 1994
; Eckert and Barrett, 1994
; Dudash et al., 1997
; Byers and Waller, 1999
).
Polygala lewtonii Small (Polygalaceae), a federally endangered (USFWS, 1999
), short-lived, perennial herb endemic to xeric uplands maintained by frequent fire in central Florida (Menges and Weekley, 2003
), has both CL and CH flowers (James, 1957
; C. Weekley, unpublished report). Polygala lewtonii is one of about three dozen amphicarpic species (Cheplick, 1987
) in which CL flowers occur on underground stems. CH flowers, which occur on aboveground terminal racemes, are small and pea-like with stigmas and anthers in close juxtaposition. As in other members of its genus, the CH flowers of P. lewtonii possess a bi-lobed stigma with one sterile and one fertile lobe (Venkatesh, 1956
; Miller, 1971
; Brantjes, 1982
; Zomlefer, 1989
). Typically in Florida species, pollen is deposited on the sterile lobe and transferred to the fertile lobe as the flower ages if it has not been picked up by visiting insects (Zomlefer, 1989
). This mechanism putatively ensures autonomous selfing in the absence of insect pollination. Because autonomous selfing occurs only after flowers have been exposed to insects for several days, the mechanism provides for delayed selfing (sensu Lloyd, 1992
). We have previously documented high fruit set in open-pollinated CH flowers (C. Weekley, unpublished report; Menges and Weekley, 2003
), but we do not know what proportion of open-pollinated fruits are the products of outcrossing, geitonogamy or some form of autogamy.
In this study, we focused on the CH flowers of P. lewtonii and the efficacy of pollination with and without insect vectors. We tested the hypothesis that delayed selfing in CH flowers provides reproductive assurance. Our expectation was that fruit set in pollinator-excluded CH flowers would approximate fruit set in CH flowers given access to pollinators. Because the small size and intricate arrangement of floral structures in P. lewtonii precluded floral manipulations (e.g., emasculations, hand pollinations) usually employed in mating system studies (Schoen and Lloyd, 1992
), we relied on pollinator-exclusion experiments to estimate rates of spontaneous selfing (autonomous autogamy) in P. lewtonii. Our study had five objectives: (1) to compare percentage fruit initiation and fruit maturation in open-pollinated vs. pollinator-excluded flowers, (2) to compare rates of floral development in the two treatments, (3) to document the mechanics of self-pollination, (4) to investigate the effect of flower position on fruit initiation and fruit maturation in the contrasting treatments, and (5) to quantify insect visitation rates to open-pollinated flowers.
MATERIALS AND METHODS
Study species
Polygala lewtonii is a multi-stemmed herb narrowly endemic to yellow sand, xeric upland habitats on the Mount Dora and Lake Wales Ridges of central peninsular Florida (USFWS, 1999
). Adults grow to 20 cm in height with few to many (3–30) erect, decumbent, or prostrate stems. Populations typically occupy open areas in longleaf pine-wiregrass (Pinus palustris Mill-Aristida stricta Michx. var. beyrichiana Trin. & Rupt.) sandhills historically characterized by frequent (2–5 yr) lightening-ignited fires (Myers, 1990
; Menges, 1999
). Polygala lewtonii plants are killed by fire, but postfire seedling recruitment from a long-lived dormant seedbank often results in order-of-magnitude increases in aboveground population size (Menges and Weekley, 2002
). Aboveground populations decline with time-since-fire, but even in long unburned sites small populations are maintained by periodic seedling recruitment, particularly during wet winters (e.g., El Niño years).
Bright pink to purple CH flowers occur on many-flowered terminal racemes and are visited by a variety of insects. White CL flowers occur on few-flowered underground rhizomes. In addition, solitary, pale pink CL flowers may occur in the axils of aboveground leaves, although these are extremely rare. CH flowers are zygomorphic and bisexual with the ovary superior and two-loculed, with one ovule in each locule (Zomlefer, 1989
). Two of the five sepals are petal-like in color and size, mimicking the wings of papilionoid legumes, and persist after the corolla, stamens and stigma/style abscise. The corolla is reduced to three petals; the upper petals are fused, and the lower petal is a keel with a fringe of fingerlike projections. The fruit is a dehiscent, two-seeded capsule. Seeds bear stiff hairs and two aril-like outgrowths at the micropyle that attract ants (Zomlefer, 1989
; Menges and Weekley, 2002
).
Like other members of its genus, P. lewtonii has a sexual morphology that apparently ensures delayed selfing (Venkatesh, 1956
; Miller, 1971
; Brantjes, 1982
; Zomlefer, 1989
). In Florida species, flowers have a bi-lobed pistil consisting of an apical sterile lobe with a tuft of hairs and a sticky fertile lobe (Fig. 1; Zomlefer, 1989
). The androecium comprises eight stamens whose fused filaments form a tube partly enclosing the ovary and the style, with the bi-lobed stigma nested among the anthers. During flower opening, pollen is deposited on the sterile lobe. Insect visitors land on the fringed keel petal to access nectar secreted at the base of the flower, picking up pollen from the sterile lobe in the process. Insect visitors may also deposit either outcross pollen or self-pollen on the fertile lobe, thus effecting cross-pollination, facilitated autogamy (if the pollen is from the same flower), or geitonogamy (if the pollen is from another flower on the same plant). As the flower ages, the two stigmatic lobes are thought to fold together, thereby effecting spontaneous self-pollination (Zomlefer, 1989
). The functioning of this mechanism in P. lewtonii has not been previously investigated.
|
Pollinator-exclusion experiments and floral development observations
In March 2004, we selected 20 pairs of P. lewtonii plants at the Carter Creek site. Paired plants were within 0.5 m of each other and thus occupied a shared microhabitat. One member of each pair was enclosed in a Delnet heat-sealed, non-woven pollination bag (Applied Extrusion Technologies, Middleton, Delaware, USA) to exclude pollinators (pollinator-exclusion treatment) and the other was left unbagged (open-pollinated treatment). We chose Delnet pollination bags because they are made of a fine-mesh, polyethelene material with high ratings for air permeability and are designed to minimize alteration of the microenvironment. Bags were supported by bamboo skewers and held in place by sand piled around the base of the bag.
On each plant, we marked three bud-bearing stems with color-coded thread and counted the number of flower buds; altogether, 1645 buds were included in the experimental setup. However, we excluded one pollinator-excluded plant on which only a single flower opened and its paired open-pollinated plant, reducing to 38 the number of paired plants included in the analyses below. An additional 134 buds were either lost to herbivores or failed to develop. Thus, we actually observed 1468 developing flowers over the course of the experiment (794 in the open-pollinated treatment and 674 in the pollinator-excluded treatment). There was no significant difference in the mean number of flowers per plant (paired t test: t = 1.333, df = 18, P = 0.199) or per raceme (t = 1.738, df = 56, P = 0.088) in the two treatments. However, plants in the pollinator-excluded treatment had a significantly higher (Z = –2.797, P = 0.005) percentage of open flowers than plants in the open-pollinated treatment.
In P. lewtonii, blooming occurs acropetalously (from the base to the tip of the raceme), thereby facilitating tracking of flower and fruit development. On each marked raceme, we numbered enlarging buds sequentially from the base of the raceme to its apex. Conspicuous abscission scars marked the location of abscised flowers or fruits. For 59 days, we followed developing flowers daily from bud enlargement through flower opening to flower or fruit abscission. We recognized five stages of flower development: (1) swollen bud, (2) sepals expanded, (3) flower open, (4) flower closed, and (5) developing fruit. Flowers usually opened between 0900 and 1000 hours and closed between 1600 and 1700 hours on successive days. Flowers may open and close daily for several days, but apparently once a flower is pollinated it ceases to re-open. To evaluate rates of floral development, we compared the number of days that flowers in the two treatments spent in each of the five stages. We used survival analysis to compare rates of fruit abscission and also used the resulting survival curves to distinguish between fruit abortion and fruit maturation.
We defined fruit initiation as the point at which the petals, stamens, style, and stigmas abscised, leaving the two petaloid sepals between which the enlarging ovary/developing fruit could be seen. Fruits were defined as mature if they could be detached from the raceme with a slight touch and were fully enlarged or if abscission occurred more than 22 d after initiation. We arrived at the 22-d threshold by calculating the mean time to abscission of all developing fruits and by looking for breakpoints in the survival curves.
Microphotographic documentation of floral development
In 2003, using the same Carter Creek population as in the pollinator-exclusion experiments, we chose a group of P. lewtonii plants to provide flowers for microscopic observation and documentation of floral development. Five plants were bagged with Delnet mesh pollination bags to exclude pollinators, and five others were left unbagged to permit insect pollination. On each plant, 3–5 flowering stems were marked with color-coded thread. We collected 10–20 flowers of known age daily from bagged and unbagged plants. Flowers were placed in plastic petri dishes for transport to the lab and were generally examined within 2–3 h of collection. Altogether we dissected and examined approximately 300 P. lewtonii flowers, ranging from 1 to 15 days in age. For comparison of floral functioning, in 2004, we collected flowers from three widespread congeners, P. nana, P. rugellii, and P. setacea, which, in central Florida, occupy mesic to wet flatwoods habitats. Flowers of all four species were dissected in the lab using an Olympus SZX12 (Olympus America, Melville, New York, USA) dissecting stereomicroscope and photographed using an Optronics 1-CCD digital video camera (Optronics, Goleta, California, USA) and Image-Pro Plus (version 4.1) image analysis software (Media Cybernetics, Silver Spring, Maryland, USA).
Effect of flower position on fruit initiation and maturation
In our monitoring scheme, flower number was an index of flower position along the raceme. Thus, in following flower development, we were also able to evaluate the effect of flower position on fruit initiation and maturation.
Insect visitation to open-pollinated plants
During the spring 2004 flowering season, we conducted 25 30-min pollinator watches in four patches of P. lewtonii at Carter Creek, each patch with at least seven flowering plants. We focused each watch on a single focal plant within the patch and recorded data on the number of insect visitors, the number of visits to the patch and to the focal plant, and the number of flowers visited. At the beginning of each watch, we recorded the temperature and noted whether the sky was clear, partly cloudy, or overcast.
Statistical methods
For analysis of data on fruit initiation, fruit maturation, and rate of floral development, mean values for each plant in the two treatments constituted the response variable. These data were generally nonnormal and could not be transformed to normality. Therefore, we used the nonparametric Wilcoxon signed rank test to evaluate the significance of differences between the two treatments. Survival analysis of rates of fruit abscission was based on survival curves and on the Wilcoxon (Gehan) statistic. We used regression analysis to evaluate the effect of flower position on fruit initiation and maturation. All statistical tests were performed in SPSS (version 11.5; SPSS, 2002
).
RESULTS
Fruit initiation and maturation of pollinator-excluded vs. open-pollinated flowers
Fruit initiation, fruit maturation as a percentage of flowers available, and fruit maturation as a percentage of fruits initiated were all significantly lower in pollinator-excluded flowers than in open-pollinated flowers (Fig. 2). Overall, the 794 open-pollinated flowers initiated 656 fruits (82.6%), while the 674 pollinator-excluded flowers initiated only 69 fruits (10.2%). On a per-plant basis, percentage fruit initiation was significantly greater for open-pollinated than for pollinator-excluded plants (Z = –3.823, P < 0.0001). In the open-pollinated treatment, the mean percentage initiation (±SD) was 81.5 ± 8.8% and ranged from 66.7 to 100%. In the pollinator-exclusion treatment, the mean percentage initiation was 12.2 ± 14.4%; three pollinator-excluded plants initiated no fruits and for the 16 that did so, percentage initiation ranged from 2.6 to 50.0%.
|
Open-pollinated and pollinator-excluded flowers also differed significantly in the percentage of fruit initiations resulting in mature fruits. Globally, 71.2% of open-pollinated flowers that initiated fruits successfully matured them (467/656), while only 27.5% of pollinator-excluded flowers (19/69) did so. On a per-plant basis, open-pollinated plants matured 67.7 ± 21.0% of fruits initiated, while pollinator-excluded plants matured only 26.6 ± 36.6%, and again the difference was significant (Z = –3.332, P = 0.001). With one exception, all open-pollinated plants produced mature fruits, with the percentage yield for the other 18 plants ranging from 33.3 to 93.8%. Ten pollinator-excluded plants produced no mature fruits, with percentage yield in the other nine pollinator-excluded plants ranging from 7.7 to 100%.
Rate of floral development in pollinator-excluded vs. open-pollinated flowers
There were no differences between treatments in the number of days that a flower spent in stages 1 (bud enlarged) or 2 (sepals expanded). While stages 3 (flower open) and 4 (flower closed) were significantly more rapid in the open-pollinated treatment, stage 5 (developing fruit) was significantly more protracted in the open-pollinated flowers (Fig. 3). Pollinator-excluded flowers opened on twice as many days as open-pollinated flowers (4.4 ± 0.7 vs. 2.2 ± 0.4 days), and the difference was significant (Z = –3.823, P < 0.0001). Pollinator-excluded flowers spent about 2.5 times as long in the closed flower stage as open-pollinated flowers (13.7 ± 4.9 vs. 5.7 ±1.2 d; Z = –3.823, P < 0.0001), probably reflecting a longer time to petal abscission in unpollinated flowers. However, fruits from open-pollinated plants abscised significantly later than their pollinator-excluded counterparts (22.2 ± 4.7 vs. 10.4 ±10.1 d; Z = –3.219, P = 0.001).
|
|
|
|
|
Fruit initiation and maturation in open-pollinated and pollinator-excluded flowers
Denying the chasmogamous flowers of Polygala lewtonii access to insects results in significantly lower fruit initiation, fruit maturation as a percentage of flowers available for pollination, and fruit maturation as a percentage of fruits initiated. Drastically reduced fruit initiation by pollinator-excluded flowers (10.2 vs. 82.6% for open-pollinated flowers) implicates dysfunction or inefficiency of the mechanism that putatively ensures delayed selfing in the absence of insect pollinators. Lower fruit production in the pollinator-excluded treatment (27.5 vs. 71.2%) suggests inbreeding depression in autonomously selfed flowers, although other factors may be involved. Thus, reproductive assurance (Lloyd, 1992
; Holsinger, 1996
; Kalisz and Vogler, 2003
; Kalisz et al., 2004
), the ability of a self-compatible species to compensate for pollen- or pollinator-limitation by spontaneously selfing, is severely limited in the chasmogamous flowers of P. lewtonii and fruit production is minimal in the absence of insect visitors.
The failure of approximately 16% of pollinator-excluded P. lewtonii plants (3 of 19) to initiate fruits indicates a total dysfunction of the selfing mechanism in the bagged flowers on these plants, but fruit initiation as high as 50% in 13 other plants (range 2.6–50.0%) suggests that a modicum of functionality may be maintained in some individuals. The degree of dysfunction may vary by genotype, plant age, or some other life history parameter. The poor reproductive performance of bagged flowers raises the concern that these results are an artifact of the bagging treatment itself. Depending on the materials used, pollination bags may alter the temperature, relative humidity, or solar radiation experienced by bagged flowers (Kearns and Inouye, 1993
). We chose Delnet polyethelene bags because their high air permeability is designed to minimize alteration of the microenvironment. In our own field tests, we found that the temperature inside bags differed by less than one degree over a 24-h period from temperatures outside bags and that even short-term differences of up to 5°C fell within in the range of temperatures normally experienced by P. lewtonii during its flowering season. Similarly, the ranges of relative humidity and solar radiation inside the bags are unlikely to differ significantly from the ambient conditions under which P. lewtonii normally flowers. In addition, we found no evidence of the closing together of stigmas in either bagged or unbagged flowers. We conclude, therefore, that the reproductive failure observed in P. lewtonii in this study is not an artifact of the bagging treatment.
In contrast to the high rate of reproductive failure in the pollinator-excluded treatment, in open-pollinated plants a minimum of 66.7% of flowers initiated fruits and in 11 of 19 plants (57.9%) fruit initiation was greater than 80%. Fruit initiation in the open-pollinated treatment may have been due to facilitated (insect-mediated) selfing, geitonogamy (within-plant pollen transfers), or outcrossing, as well as spontaneous selfing via the closing together of the stigmatic lobes. Although pollinator-exclusion experiments provide an estimate of spontaneous selfing (autonomous autogamy) rates, they unfortunately do not allow for partitioning the modes of fruit initiation in open-pollinated flowers. Sire-exclusion analysis using molecular markers such as microsatellites would provide an assessment of the degree of selfing within a population with a mixed mating system, but cannot distinguish between geitonogamy and various modes of autogamy (Schoen and Lloyd, 1992
). Both geitonogamy and insect-mediated autogamy are likely in P. lewtonii. Neither mode of self-fertilization contributes to reproductive assurance since both co-opt pollen and ovules otherwise available for outcrossing (Schoen and Lloyd, 1992
; de Jong et al., 1993; Eckert, 2000
).
Fruit maturation as a percentage of fruits initiated in the open-pollinated treatment was 2.6 times greater than in the pollinator-excluded treatment. Only one open-pollinated plant failed to produce mature fruits, and 16 of 18 plants had fruit yields greater than 60%. In contrast, seven of the 16 pollinator-excluded plants (43.7%) that initiated fruits failed to mature them. The high rate of fruit abortion in the pollinator-excluded treatment is undoubtedly due primarily to inbreeding depression. This is a surprising result because of the expectation that in a mixed mating system species, routine selfing by CL flowers would purge most deleterious lethal alleles (Schoen and Lloyd, 1984
; Charlesworth and Charlesworth, 1987
; Barrett and Charlesworth, 1991
; Barrett and Harder, 1996
; Husband and Schemske, 1996
; Byers and Waller, 1999
; Keller and Waller, 2002
). Although some studies have found that outcrossed progeny from mixed mating system (CH/CL) species have a fitness advantage over selfed progeny (Schemske, 1978
; Waller, 1984
; Mitchell-Olds and Waller, 1985
), many other studies have shown that selfed progeny are as least as fit as outcrossed progeny over a range of life history stages (Wilken, 1982
; Cheplick and Quinn, 1982
, 1983
; Clay 1983
, Clay and Antonovics, 1985
; Antlfinger 1986
; Gara and Muenchow, 1990
; Le Corff, 1996
; Berg and Redbo-Torstensson, 1999
, 2000
). However, few studies have investigated early acting inbreeding depression by quantifying fruit maturation rates as a percentage of fruit initiations (Husband and Schemske, 1996
).
Recent studies of other mixed mating system species have found evidence for (e.g., Motten, 1982
; Piper et al., 1986
; Dole, 1992
; Klips and Snow, 1997
; Fausto et al., 2001
; Culley, 2002
; Elle and Carney, 2003
; Kalisz and Vogler, 2003
; Kalisz et al., 2004
) and against (e.g., Eckert and Schaefer, 1998
; Herlihy and Eckert, 2002
) the reproductive assurance hypothesis. However, to the best of our knowledge, only one previous researcher (Culley, 2002
) has investigated reproductive assurance in the CH flowers of a species with both CH and CL flowers. Culley (2002)
found that both the mixed mating (CH/CL) system and delayed selfing in the CH flowers of Viola pubescens contributed to the reproductive output of the species across heterogeneous environments in which both pollinator service and resources are potentially limiting.
In the case of Polygala lewtonii, no assessment of the contribution of CL flowers to its reproductive output has been undertaken. However, our results demonstrate that little reproductive assurance is provided by the CH flowers due to the inefficiency of the delayed selfing mechanism and high rates of fruit abortion due to inbreeding depression and perhaps other causes.
Rates of floral development
The two most important differences in floral development between pollinator-excluded and open-pollinated flowers were the number of days that flowers in the two treatments were open and the number of days from fruit initiation to fruit abscission. Pollinator-excluded flowers opened twice as often as open-pollinated flowers (approximately 4 vs. approximately 2 days), suggesting that open-pollinated flowers cease to open once they have been pollinated by visiting insects and that unpollinated flowers have a mean life span of approximately 4 days. The median time to abscission of fruits in the pollinator-excluded treatment was only about a third of that for open-pollinated plants (7.63 vs. 26.85 days), indicating the much higher rates of abortion in obligately-selfed fruits than in open-pollinated fruits. We interpret the higher abortion rates of pollinator-excluded flowers as evidence of early acting inbreeding depression, although resource limitation may also play a role (discussed later).
Mechanics of delayed selfing
Dissections of P. lewtonii flowers of known age, documented by microphotographs, provided no evidence that the two lobes of the bi-lobed stigma close together to effect spontaneous selfing. In contrast, using the same procedures for collecting and examining flowers, we readily documented the closing together of the stigmatic lobes in three other Polygala species. Thus, the mechanism putatively ensuring delayed autonomous selfing in P. lewtonii appears to be largely dysfunctional. However, approximately 10% of pollinator-excluded flowers did initiate fruits, suggesting that the mechanism is not totally dysfunctional. While the selfing mechanism has been documented in several species of Polygala (Venkatesh, 1956
; Miller, 1971
; Brantjes, 1982
; Zomlefer, 1989
), we are not aware of any studies which quantify its efficacy or which investigate the level of reproductive assurance provided by delayed selfing in any of the approximately 600 Polygala species known worldwide.
Flower position and resource limitation
Resource limitation is a potential source of flower and fruit abortion (Stephenson, 1981
) in Polygala lewtonii. In both the pollinator-excluded and the open-pollinated treatments, fruit initiation was significantly higher at the base of the flowering raceme than at its apex. In open-pollinated plants this could reflect pollinator foraging behavior, but because the pattern is the same for pollinator-excluded plants, resource limitation may also play a role. Fruit maturation was also significantly higher at the base of the flowering raceme than at its apex for open-pollinated plants, again suggesting resource limitation. The weak correlation between flower position and fruit maturation in the pollinator-excluded treatment reflects the over-all low percentage of fruit initiation in flowers denied access to pollinators rather than a meaningful difference in patterns of fruit maturation.
In addition, differential resource allocation is also a possible explanation for differences in rates of fruit maturation between pollinator-excluded and open-pollinated P. lewtonii plants. The protracted floral phenology of pollinator-excluded flowers compared to open-pollinated flowers may contribute to lower rates of fruit maturation because of the higher energetic costs of repetitively opening flowers and perhaps replenishing nectar rewards. However, it seems unlikely that the cost of maintaining unfertilized flowers would be greater than the cost of fruit development.
Pollinator limitation
It is difficult to reconcile our observations of low insect visitation rates in P. lewtonii with the relatively high rates of fruit initiation and maturation. Within patches of plants containing dozens of flowering racemes, the observed rate of insect visitation (number of flowers visited per unit time) was only 0.3 flowers per hour, which seems inadequate to pollinate the hundreds of flowers opening over a span of several days. However, given high levels of fruit abortion in pollinator-excluded plants (suggesting that self-pollinations often result in abortion) and high levels of fruit maturation in open-pollinated plants, it seems that many plants must be adequately serviced; in this study, over 80% of open-pollinated flowers initiated fruits and two-thirds of initiated fruits developed to maturity.
Conclusion
Due to the inadequacy of the delayed selfing mechanism and inbreeding depression in the relatively small number of fruits initiated in the pollinator-excluded CH flowers of P. lewtonii, these flowers provide only minimal reproductive assurance. However, underground CL flowers, not investigated here, may provide an additional component of reproductive assurance even in the face of substantial inbreeding depression. Assessment of the role of cleistogamy in the overall reproductive output of Polygala lewtonii awaits the development of a protocol for monitoring the development of underground CL flowers and fruits.
FOOTNOTES
The authors thank J. Atwell, S. Neimeister, M. Rickey, M. Trager, C. Valligny, and R. Zinthefer for field and lab assistance; P. Quintana-Ascencio and M. Rickey for statistical advice; and A. Faivre, E. Menges, and two anonymous reviewers for helpful comments on earlier versions of the manuscript. The project was supported by a grant from the Florida Division of Plant Industry and by Archbold Biological Station. 
4 Author for correspondence (e-mail: cweekley{at}archbold-station.org
)
LITERATURE CITED
Antlfinger A. E 1986 Field germination and seedling growth of CH and CL progeny on Impatiens capensis (Balsaminaceae). American Journal of Botany 73: 1267-1273[CrossRef][ISI]
Barrett S. C. H Charlesworth D 1991 Effects of a change in the level of inbreeding on the genetic load. Nature 352: 522-524[CrossRef][Medline]
Barrett S. C. H Eckert C. G 1990 Variation and evolution of mating systems in seed plants. In S. Kawano [ed.] Biological approaches and evolutionary trends in plants 231-254 Academic Press, London, UK
Barrett S. C. H Harder L. D 1996 Ecology and evolution of plant mating. Trends in Ecology and Evolution 11: 73-79
Barrett S. C. H Harder L. D Worley A. C 1996 The comparative biology of pollination and mating in flowering plants. Philosophical Transactions of the Royal Society of London 351: 1271-1280[CrossRef]
Berg H Redbo-Torstensson P 1999 Offspring performance in three cleistogamous Viola species. Plant Ecology 145: 49-58[CrossRef][ISI]
Berg H Redbo-Torstensson P 2000 Offspring performance in Oxalis acetosella, a cleistogamous perennial herb. Plant Biology 2: 638-645[CrossRef][ISI]
Brantjes N. B. M 1982 Pollen placement and reproductive isolation between two Brazilian Polygala species (Polygalaceae). Plant Systematics and Evolution 141: 41-52[CrossRef][ISI]
Byers D. L Waller D. M 1999 Do plant populations purge their genetic load? Effects of population size and mating history on inbreeding depression. Annual Review of Ecology and Systematics 30: 479-513[CrossRef][ISI]
Charlesworth D Charlesworth B 1987 Inbreeding depression and its evolutionary consequences. Annual Review of Ecology and Systematics 18: 237-268[CrossRef][ISI]
Cheplick G. P 1987 The ecology of amphicarpic plants. Trends in Ecology and Evolution 2: 97-101
Cheplick G. P Quinn J. A 1982 Amphicarpum purshii and the "pessimistic strategy" in amphicarpic annuals with subterranean fruits. Oecologia 52: 327-332[CrossRef][ISI]
Cheplick G. P Quinn J. A 1983 The shift in aerial/subterranean fruit ratio in Amphicarpum purshii, causes and significance. Oecologia 57: 374-379[CrossRef][ISI]
Clay K 1983 The differential establishment of seedlings from chasmogamous and cleistogamous flowers in natural populations of the grass Danthonia spicata, (L.) Beauv. Oecologia 57: 183-188[CrossRef][ISI]
Clay K Antonovics J 1985 Demographic genetics of the grass Danthonia spicata: success of progeny from chasmogamous and cleistogamous flowers. Evolution 39: 205-210[CrossRef][ISI]
Cruden R. W Lyon D. L 1989 Facultative xenogamy. In J. H. Bock, and Y. B. Linhart [eds.] The evolutionary ecology of plants 171-207 Westview Press, Boulder, Colorado, USA
Culley T. M 2000 Inbreeding depression and floral type fitness differences in Viola canadensis (Violaceae), a species with chasmogamous and cleistogamous flowers. Canadian Journal of Botany 78: 1420-1429[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 Sciences 16: 113-122
Darwin C 1876 The effects of cross and self fertilization in the vegetable kingdom John Murray London, England
Darwin C 1877 The different forms of flowers on plants of the same species John Murray London, England
de Jong T. J Waser N. M Klinkhamer P. G. L 1993 Geitonogamy; the neglected side of selfing. Trends in Ecology and Evolution 8: 321-325
Dole J. A 1992 Reproductive assurance mechanisms in three taxa of the Mimulus guttatus complex (Scrophulariaceae). American Journal of Botany 79: 650-659[CrossRef][ISI]
Dudash M. R Carr D. E Fenster C. B 1997 Five generations of enforced selfing and outcrossing in Mimulus guttatus: inbreeding depression variation at the population and family level. Evolution 51: 54-65[CrossRef][ISI]
Eckert C. G 2000 Contributions of autogamy and geitonogamy to self-fertilization in a mass-flowering, clonal plant. Ecology 81: 532-542[CrossRef][ISI]
Eckert C. G Barrett S. C. H 1994 Inbreeding depression in partially self-fertilizing Decodon verticillatus (Lythraceae): population-genetic and experimental analyses. Evolution 48: 952-964[CrossRef][ISI]
Eckert C. G Schaefer A 1998 Does self-pollination provide reproductive assurance in Aquilegia canadensis (Ranunculaceae). American Journal of Botany 85: 919-924[Abstract]
Elle E Carney R 2003 Reproductive assurance varies with flower size in Collinsia parviflora (Scrophulariaceae). American Journal of Botany 90: 888-896
Fausto J. A Eckhart V. M Geber M. A 2001 Reproductive assurance and the evolutionary ecology of self-pollination in Clarkia xantiana (Onagraceae). American Journal of Botany 88: 1794-1800
Gara B Muenchow G 1990 Chasmogamy/cleistogamy in Triodanis perfoliata (Campanulaceae): some CH/CL comparisons in fitness parameters. American Journal of Botany 77: 1-6
Herlihy C. R Eckert C. G 2002 Genetic cost of reproductive assurance in a self-fertilizing plant. Nature 416: 320-323[CrossRef][Medline]
Holsinger K. E 1991 Inbreeding depression and the evolution of plant mating systems. Trends in Evolution and Ecology 6: 307-308[CrossRef]
Holsinger K. E 1996 Pollination biology and the evolution of mating systems in flowering plants. Evolutionary Biology 29: 107-149[ISI]
Husband B. C Schemske D. W 1996 Evolution of the magnitude and timing of inbreeding depression in plants. Evolution 50: 54-70
Jain S. K 1976 The evolution of inbreeding in plants. Annual Review of Ecology and Systematics 7: 469-495
James C. W 1957 Note on the cleistogamous species of Polygala in the southeastern United States. Rhodora 59: 51-56
Johnston M. O 1998 Evolution of intermediate selfing rates in plants: pollination ecology versus deleterious mutations. Genetica 102: /103 267-278[CrossRef]
Kalisz S Vogler D. W 2003 Benefits of autonomous selfing under unpredictable pollinator environments. Ecology 84: 2928-2942[CrossRef][ISI]
Kalisz S Vogler D. W Hanley K. M 2004 Context-dependent autonomous self-fertilization yields reproductive assurance and mixed mating. Nature 430: 884-887[CrossRef][Medline]
Kearns C. A Inouye D. W 1993 Techniques for pollination biologists University Press of Colorado Niwot, Colorado, USA
Keller L. F Waller D. M 2002 Inbreeding effects in wild populations. Trends in Ecology and Evolution 17: 230-240[CrossRef]
Klips R. A Snow A. A 1997 Delayed autonomous self-pollination in Hibiscus laevis (Malvaceae). American Journal of Botany 84: 48-53[Abstract]
Lande R Schemske D. W 1985 The evolution of self-fertilization and inbreeding depression in plants. I. Genetic models. Evolution 39: 24-40[CrossRef][ISI]
Lande R Schemske D. W Schultz S. T 1994 High inbreeding depression, selective interference among loci, and the threshold selfing rate for purging recessive lethal mutations. Evolution 48: 965-978[CrossRef][ISI]
Le Corff J 1996 Establishment of chasmogamous and cleistogamous seedlings of an ant-dispersed understory herb, Calathea micans (Marantaceae). American Journal of Botany 83: 155-161
Lloyd D. G 1979 Some reproductive factors affecting the selection of self-fertilization in plants. American Naturalist 113: 67-79[CrossRef][ISI]
Lloyd D. G 1980 Demographic factors and mating patterns in angiosperms. In O. T. Solbrig [ed.] Demography and evolution in plant populations 67-88 University of California Press, Berkeley, California, USA
Lloyd D. G 1992 Self- and cross-fertilization in plants. II. The selection of self-fertilization. International Journal of Plant Sciences 153: 370-380[CrossRef]
Lloyd D. G Schoen D. J 1992 Self- and cross-fertilization in plants. I. Functional dimensions. International Journal of Plant Sciences 153: 358-369[CrossRef]
Lord E. M 1981 Cleistogamy: a tool for the study of floral morphogenesis, function and evolution. Botanical Review 47: 421-449[ISI]
Mallick S. A 2001 Facultative dichogamy and reproductive assurance in partially protandrous plants. Oikos 95: 533-536[CrossRef][ISI]
Menges E. S 1999 Ecology and conservation of Florida scrub. In R. C. Anderson, J. S. Fralish, and J. Baskin [eds.] The savanna, barren, and rock outcrop communities of North America 7-22 Cambridge University Press, Cambridge, UK
Menges E. S Weekley C. W 2002 Demographic research on four state-listed Lake Wales Ridge endemic plants Final report to Endangered and Threatened Plant Conservation Grants Program, Division of Plant Industry Gainesville, Florida, USA
Menges E. S Weekley C. W 2003 Continuation of demographic research on four state-listed Lake Wales Ridge endemic plants Final report to Endangered and Threatened Plant Conservation Grants Program, Division of Plant Industry Gainesville, Florida, USA
Miller N. G 1971 The Polygalaceae in the Southeastern United States. Journal of the Arnold Arboretum 52: 267-284[ISI]
Mitchell-Olds T Waller D. M 1985 Relative performance of selfed and outcrossed progeny in Impatiens capensis. Evolution 39: 533-544[CrossRef][ISI]
Motten A. F 1982 Autogamy and competition for pollinators in Hepatica americana (Ranunculaceae). American Journal of Botany 69: 1296-1305[CrossRef][ISI]
Myers R. L 1990 Scrub and high pine. In R. L. Myers, and J. J. Ewel [eds.] Ecosystems of Florida 150-193 University of Central Florida Press, Orlando, Florida, USA
Pannell J. R Barrett S. C. H 1998 Baker's Law revisited: reproductive assurance in a metapopulation. Evolution 52: 657-668[CrossRef][ISI]
Piper J. G Charlesworth B Charlesworth D 1986 Breeding system and evolution in Primula vulgaris and the role of reproductive assurance. Heredity 56: 207-217[ISI]
Richards A. J 1997 Plant breeding systems Chapman and Hall London, UK
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 Lande R 1985 The evolution of self-fertilization and inbreeding depression in plants. II. Empirical observations. Evolution 39: 41-52[CrossRef][ISI]
Schmitt J Ehrhardt D. W 1990 Enhancement of inbreeding depression by dominance and suppression in Impatiens capensis. Evolution 44: 269-278[CrossRef][ISI]
Schoen D. J Lloyd D. G 1984 The selection of cleistogamy and heteromorphic diaspores. Biological Journal of the Linnaean Society 23: 303-322
Schoen D. J Lloyd D. G 1992 Self- and cross-fertilization in plants. III. Methods for studying modes and functional aspects of self-fertilization. International Journal of Plant Sciences 153: 381-393[CrossRef][ISI]
SPSS. 2002 SPSS, version 11.5 SPSS Chicago, Illinois, USA
Stephenson A. G 1981 Flower and fruit abortion: proximate causes and ultimate functions. Annual Review of Ecology and Systematics 12: 23-279
U.S. Fish And Wildlife Service. 1999 Multi-species recovery plan for the threatened and endangered species of South Florida U.S. Fish and Wildlife Service Vero Beach, Florida, USA
Uyenoyama M. K Holsinger K. E Waller D. M 1993 Ecological and genetic factors directing the evolution of self-fertilization. In D. Futuyma, and J. Antonovics [eds.] Oxford surveys in evolutionary biology 327-381 Oxford University Press, London, UK
Venkatesh C. S 1956 The special mode of dehiscence of anthers of Polygala and its significance in autogamy. Bulletin of the Torrey Botanical Club 83: 19-26[CrossRef]
Vogler D. W Kalisz S 2001 Sex among the flowers: the distribution of plant mating systems. Evolution 55: 202-204[CrossRef][ISI][Medline]
Waller D. M 1984 Differences in fitness between seedlings derived from cleistogamous and chasmogamous flowers in Impatiens capensis. Evolution 38: 427-440
Wilken D. H 1982 The balance between chasmogamy and cleistogamy in Collomia grandiflora (Polemoniaceae). American Journal of Botany 69: 1326-1333[CrossRef][ISI]
Zomlefer W. B 1989 Flowering plants of Florida: a guide to common families Biological Illustrations, Inc Gainesville, Florida, USA
This article has been cited by other articles:
|
|
 |
|
  Flowering Newsletter bibliography for 2006 J. Exp. Bot., April 20, 2007; (2007) erm028v2. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||
