| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Reproductive Biology |
Department of Biological Sciences, Idaho State University, Pocatello, Idaho 83209 USA and The Rocky Mountain Biological Laboratory, Crested Butte, Colorado 81224 USA
Received for publication March 15, 2007. Accepted for publication August 6, 2007.
ABSTRACT
Floral display size represents a tradeoff between the benefits of increased pollinator visitation and the quantity of pollen received vs. the costs of increased self-pollination and reduced pollination quality. Plants with large floral displays often are more attractive to pollinators, but pollinators visit more flowers per plant. Intraplant foraging movements should increase self-pollination through geitonogamy, lowering outcrossing rates in large plants. Local genetic structure should also increase inbreeding and decrease outcrossing estimates, if pollinators move between neighboring, related plants. These predictions were tested in a population of larkspurs (Delphinium barbeyi) in Colorado. Allozymes were used to estimate outcrossing rates of plants varying in display size. Floral displays varied widely (2–1400 flowers; 1–26 inflorescences per plant), and outcrossing rate decreased significantly with increasing display size. Large, multistalked plants self over twice as frequently as single-stalked plants (46 vs. 21%). Local population structure is significant, and biparental inbreeding depresses outcrossing in plants surrounded by genetically similar neighbors. Protandry, coupled with stereotypical bottom-up pollinator foraging, reduces self-fertilization by autogamy or geitonogamy within inflorescences. Selfing is predominantly (>60%) by geitonogamy between inflorescences in large plants. Geitonogamy may be a significant cost to plants with large floral displays if inbreeding depression and/or pollen and ovule discounting results. If so, floral display size, particularly inflorescence number, may be under contrasting selection for pollination quantity vs. quality.
Key Words: Colorado • Delphinium barbeyi • geitonogamy • genetic structure • inflorescence size and number • pollinator behavior • Ranunculaceae • selfing rate
Floral display size—the number of flowers per inflorescence, plant, or genet—presents an evolutionary dilemma (Charlesworth and Charlesworth, 1987a
; Klinkhamer and de Jong, 1993
; Finer and Morgan, 2003
). In animal-pollinated species, large floral displays may increase pollinator visitation, and thus pollen receipt and donation (Willson and Rathcke, 1974
; Klinkhamer et al., 1989
; Mitchell, 1994
; Emms et al., 1997
; Ohashi and Yahara, 2001
). However, increased attractiveness of large floral displays is often associated with an increased frequency of intraplant pollinator movements during each visit (Geber, 1985
; Hessing, 1988
; Vaughton and Ramsey, 1998
; Ohashi and Yahara, 2001
; Galloway et al., 2002
; Karron et al., 2004
; Huang et al., 2006
). Sequential moves between flowers on the same plant increases the potential for geitonogamy, the transfer of self-pollen between flowers on an individual plant. Such geitonogamous selfing can incur a number of fitness costs in both self-compatible and self-incompatible species (de Jong et al., 1993
; Klinkhamer and de Jong, 1993
; Harder and Barrett, 1995
, 1996
; Snow et al., 1996
). If selfed offspring are less fit than those resulting from outcrossing, then geitonogamy will result in inbreeding depression (Darwin, 1876
; Charlesworth and Charlesworth, 1987b
; Husband and Schemske, 1996
). Deposition of self-pollen on stigmas may interfere with outcross pollen deposition, germination, growth, and fertilization, usurping ovules that might otherwise be outcrossed (Lloyd, 1992
; de Jong et al., 1993
; Klinkhamer and de Jong, 1993
; Harder et al., 2001
). Finally, self-pollen deposited on stigmas is unavailable for male reproductive success through outcrossing, resulting in pollen discounting (Charlesworth, 1980
; Holsinger et al., 1984
; Lloyd, 1992
; Harder and Barrett, 1996
). In addition, whereas within flower selfing (autogamy) may confer ecological advantages of reproductive assurance in low or variable pollinator environments, geitonogamy, by requiring pollinators, contributes little to reproductive assurance but retains the costs of showy floral display (Lloyd, 1992
; Johnston, 1998
; Kalisz and Vogler, 2003
; Ashman et al., 2004
; Morgan and Wilson, 2005
). Thus, the evolution of floral display size depends on the tradeoffs between the benefits of attraction in terms of pollination quantity and the costs of geitonogamy, including its effects on pollination quality.
While floral display size influences pollinator behavior in ways that promote geitonogamy, quantifying the relationship between floral display, outcrossing rate, and modes of selfing has only begun. By comparing similar species (Sun and Ganders, 1988
; Williams et al., 2001
), populations within species (Brunet and Sweet, 2006a
), or groups of individuals within populations (Appendix), the effects of display size variation on mean selfing rates have been examined at the population or group level. Because large sample sizes are required to accurately estimate individual outcrossing rates (Ivey and Wyatt, 1999
; Ritland, 2002
), outcrossing rate variation among individual plants differing in display size within a population remains poorly known. To date, intrapopulation studies have generally manipulated display size in artificial arrays and used plants homozygous for allozyme markers to detect outcrossing (Crawford, 1984
; Snow et al., 1996
; Karron et al., 2004
; but see Vrieling et al., 1999
for an alternative approach). Studies examining natural populations used floral emasculations and genetic markers to dissect the components of the mating system (Schoen and Lloyd, 1992
; Leclerc-Potvin and Ritland, 1994
; Eckert, 2000
; Herlihy and Eckert, 2004
; Brunet and Sweet, 2006b
). I am aware of only one study that examined the relationship between outcrossing and floral display size in a natural population of unmanipulated plants (Brunet and Eckert, 1998
; in Aquilegia caerulea). All but one (Herlihy and Eckert, 2004
) of these previous studies have supported a negative relationship between display size and outcrossing rate (Appendix). Yet, in most of these studies, display size varied over a relatively narrow range (<20 flowers per plant open per day). Some plants produce much larger floral displays with hundreds or even thousands of flowers open simultaneously, and little is known about the fitness tradeoffs of pollination quantity vs. quality in such species (de Jong et al., 1992
; Barrett et al., 1994
; Snow et al., 1996
). In particular, plants that produce multiple flowering stalks of the same genet, including clonal species, may undergo high levels of self-fertilization (Handel, 1985
; Eckert, 2000
).
Pollinator movements between plants are usually spatially restricted (Levin and Kerster, 1974
; Pyke, 1978
; Waser, 1982
), and neighboring plants are often more genetically related than random (Heywood, 1991
; Williams, 1994
; Williams et al., 2001
). Pollen transfer between such genetically related near-neighbors results in biparental inbreeding (Uyenoyama, 1986
) and will bias estimates of the selfing rate upward (Waller, 1993
; Griffin and Eckert, 2003
). Thus, autogamy, geitonogamy, and biparental inbreeding all may influence the selfing rate estimated from genetic markers (Ritland, 1984
; Schoen and Lloyd, 1992
; Eckert, 2000
; Herlihy and Eckert, 2004
). All three forms of inbreeding may affect the evolution of plant mating systems and the stability of intermediate outcrossing rates, although in different ways. For instance, autonomous selfing (autogamy without pollinator visitation) is often selectively advantageous because it confers reproductive assurance, while facilitated autogamy (requiring pollinator visitation) is almost never advantageous because of gamete discounting. Likewise, geitonogamy is disadvantageous because of severe gamete discounting and may be an unavoidable consequence of selection for outcross pollination (Lloyd, 1992
; de Jong et al., 1993
; Herlihy and Eckert, 2004
; Goodwillie et al., 2005
). Biparental inbreeding, by reducing both the genetic (transmission) cost of outcrossing and the fitness difference between self and outcross matings (lowering inbreeding depression), can either limit or facilitate selection for selfing (Lloyd, 1979
; Uyenoyama et al., 1993
; Waller, 1993
).
The potential for biparental inbreeding depends on the local density of related flowering plants, their degree of relatedness, and the relative size of their floral displays. This combination of factors determines a plant's local mating environment (LME). A neighboring plant with many flowers may contribute more pollen than a similarly related small neighbor. Moreover, plants surrounded by many related pollen sources (high floral density) should appear more selfed because of their higher proportion of biparental inbreeding. Hence, large plants in dense neighborhoods of related individuals should have the highest levels of inbreeding because of the combined effects of geitonogamy and biparental inbreeding. The LME should therefore be associated with the selfing rate of individual plants, and the proportions of uniparental vs. biparental inbreeding making up the total effective selfing rate should depend on plant size and its LME.
In an earlier study (Williams et al., 2001
), the effects of different floral display size between two sympatric larkspur species, Delphinium barbeyi and D. nuttallianum was investigated. Delphinium barbeyi has multiple inflorescences with many flowers (
= 450) per plant, while D. nuttallianum has a single flowering stalk with few flowers per plant ( = 4.7). It was found that bumblebees and to some extent hummingbirds, the two primary pollinators, visited more flowers per plant as flower number increased. On larger plants, bees visited multiple inflorescences. Delphinium barbeyi had significantly lower outcrossing rates than D. nuttallianium. Given these floral and mating system differences lead to more restricted gene flow, fine-scale genetic structure was much more pronounced in D. barbeyi. It was hypothesized that the differences in outcrossing rates seen between these species should also be found within D. barbeyi populations, among individual plants differing in flower and inflorescence number.
To investigate the effects of floral display size and the LME on outcrossing rate variation, I estimated outcrossing rates of individual D. barbeyi plants from a natural, unmanipulated population in Colorado. In particular, I tested the following predictions: (1) large plants, with more flowers or inflorescences, have lower outcrossing rates than small plants, (2) between inflorescence geitonogamy is the predominant form of self-fertilization in large, multistemmed plants, and (3) high local density of related flowering plants increases biparental inbreeding and decreases estimates of individual plant outcrossing rates.
MATERIALS AND METHODS
Species and study site
Delphinium barbeyi Huth (Ranunculaceae) is a long-lived, herbaceous perennial native to subalpine meadows in Colorado (Warnock, 1997
). Individual plants form distinct clumps, reaching diameters of up to 1 m. Plants produce 1–40+ inflorescences ( = 16), each with approximately 25 flowers. Total flower number per plant ranges from <10 to >1400 ( = 450; Williams et al., 2001
). The main pollinators are bumblebees (particularly Bombus appositus and B. flavifrons) and hummingbirds (Selasphorus platycercus, S. rufus, and Stellula calliope) (Waser, 1982
; Williams et al., 2001
). Bees and hummingbirds preferentially visit plants with larger floral displays, and visit more flowers on large vs. small plants (Williams et al., 2001
). A D. barbeyi population near the present study site had a multilocus outcrossing rate (tm) of 0.55 (±0.04 SE). In a survey of genetic structure within 12 populations using spatial autocorrelation, there was significant local genetic similarity among neighboring plants (Williams et al., 2001
).
Field work for this study was conducted from 24 July to 10 August 2002 in a population of D. barbeyi located immediately south of the Rocky Mountain Biological Laboratory (RMBL, 2900 m a.s.l.) in southwestern Colorado, USA. Physical and floristic characteristics of the site are described elsewhere (Langenheim, 1962
; Williams et al., 2001
). The population contained several hundred flowering plants of all sizes. All flowering plants (N = 144) in a 14 x 26 m area were tagged and mapped to the nearest 10 cm.
Collection of plant material and electrophoresis
For each tagged plant in the plot, the number of flowering stalks (inflorescences) per plant and the number of flowers and fruits per inflorescence were counted. Fresh leaf tissue was collected from each plant for immediate electrophoretic analysis of adult genotypes. As fruits matured, 1–6 infructescences were collected from each plant. Infructenscences were placed in paper bags and sealed for several days in plastic garbage bags with moth crystals (
-dichlorobenzene) to kill any seed predators. Fruits were then allowed to air dry in a warm greenhouse for several weeks. Bulk samples of seeds were collected from each infructescence for electrophoresis and outcrossing estimation. Seeds used for outcrossing analysis were therefore sampled from several fruits per infructescence.
Using fresh leaf tissue, I genotyped each flowering plant on the grid (N = 144) for three polymorphic enzyme loci; phosphoglucose isomerase (Pgi; three alleles); phosphoglucomutase (Pgm; three alleles), and acid phosphatase (Acp; three alleles). Electrophoretic techniques, grinding, and running buffers were as described for D. nuttallianum in Williams and Waser (1999)
. Several tissue samples from different sides of large plants (clumps) were genotyped to confirm they were a single plant. Only Pgi and Pgm had activity in seeds, so outcrossing rates were estimated based on only two polymorphic loci. Progeny arrays segregated as expected for all maternal genotypes, and all potential paternal alleles were detected in outcrossed progeny.
Mating system analysis
Using bulk seed samples from 51 plants that spanned the range of display size variation, I estimated outcrossing rates for (1) the population as a whole, (2) groups of plants with similar display sizes, and (3) individual plants. From 12–40 ( = 19.7) seeds from each maternal plant had scorable enzyme activity (N = 1003 total progeny). All mating system analyses were performed using the computer program MLTR (ver. 3.1, revised September 2004; Ritland, 2002
) on progeny arrays with known maternal genotypes. Standard errors of mating system parameter estimates were calculated from the standard deviation of 1000 bootstrap estimates where progeny arrays were the unit of resampling (population and group level analyses), or 100 bootstraps per family where individuals within families were resampled (individual plant level analysis). Ninety-five percent confidence intervals (CIs) of the estimates were calculated as 1.96 times the standard error.
Population level
Population level single- (ts) and multilocus (tm) maximum likelihood outcrossing rates were estimated for the entire sample of 51 plants, regardless of size. In addition, the maternal fixation index (F) and the correlation of selfing (rs) among families and among loci were estimated using the Newton–Raphson method. Because only 51 plants from a large population of potential pollen donors were sampled, pollen and ovule allele frequencies (p) were not constrained to be equal (i.e., t, p, and F were jointly estimated). However, estimates of pollen and ovule allele frequencies did not differ significantly (Table 1) and constraining them to be equal does not alter the results.
|
). Plants were grouped in two different ways: five groups spanning the range of plant sizes (1, 2–3, 4–7, 8–11, and 12–26 inflorescences) and two groups splitting the size distribution in half (1–8 inflorescences vs. 9–26 inflorescences). Because group sizes are small (Table 3) pollen and ovule gene frequencies were constrained to be equal. This increases statistical efficiency because one less parameter is estimated. Using the alternative assumption of unequal pollen and ovule gene frequencies does not alter the results.
|
recommends using the method-of-moments estimation procedure (Cruzan and Arnold, 1994
) as being less upward biased (Ivey and Wyatt, 1999
) for individual family outcrossing estimates. Multilocus, method-of-moments family level estimates were computed in MLTR for comparison to maximum likelihood estimates. Both methods gave similar, correlated outcrossing estimates (r = 0.742). Because the standard errors for the method-of-moments estimates ( = 0.235) were larger than for the maximum likelihood estimates ( = 0.172), I report only the latter results. As can be seen, standard errors of individual plant outcrossing rate estimates can be large, so I also estimated the minimum outcrossing rate (tmin) for each plant. This was calculated as the proportion of unambiguously outcrossed seeds (containing a nonmaternal allele) in each progeny array. Although such measures underestimate the true outcrossing rate (because of biparental inbreeding and limited marker variation), a strong correlation between tm and tmin indicates that their rank orders are correct.
Biparental inbreeding
I used two methods to estimate the amount of biparental inbreeding at the population and group levels. Because multilocus estimates of outcrossing (tm) are less sensitive to biparental inbreeding than single-locus estimates (ts), their difference, tm – ts, is frequently used to estimate its magnitude (Shaw et al., 1981
; Ritland, 2002
). However, tm – ts tends to greatly underestimate the true magnitude of biparental inbreeding, especially when estimates are based on few loci (Leclerc-Potvin and Ritland, 1994
; Ritland, 2002
; Griffin and Eckert, 2003
; Herlihy and Eckert, 2004
). The correlation of selfing among loci (rs) provides an alternative measure of the degree of biparental inbreeding contributing to the overall "effective selfing" rate (true selfing plus biparental inbreeding; Ritland, 1984
, 2002
). When estimates of rs are <1.0, some loci in an individual progeny's multilocus genotype indicate outcrossing, while other loci appear selfed as the result of biparental inbreeding. When selfing rates are low (1 – t = s < 0.2), rs corresponds to the proportion of s due to true uniparental selfing, with the remainder of the selfing rate due to biparental inbreeding (Ritland, 2002
). With higher selfing rates, rs should still give a relative measure of the importance of biparental inbreeding in groups of plants of different sizes.
Analysis of spatial genetic structure and the local mating environment
Patterns of genetic similarity as a function of distance between plants were estimated using multilocus spatial autocorrelation (Smouse and Peakall, 1999
) with the program GenAlEx (Peakall and Smouse, 2001
). The multilocus genetic distance among individuals was used to estimate the genetic correlation (r) as a function of distance among all 144 mapped, flowering adults. The 95% CIs of the genetic correlations were estimated from 999 permutations of the genetic and geographic distance matrices. A correlogram was generated for genetic correlations in 2.5 m distance classes.
If local genetic structure facilitates biparental inbreeding, then detectable outcrossing should depend on a plant's relative size and genetic similarity with its neighbors. An index of the local mating environment (LME) around each plant was devised to attempt to account for these effects on individual plant outcrossing rates. For each plant (target), all other flowering plants within a 2-m radius (neighborhood) were identified. For each neighbor, the number of alleles at the Pgi and Pgm loci that differed from the target were found (0–4) and multiplied by the number of flowers on the neighboring plant. For example, if the target plant has the multilocus genotype Pgi-11, Pgm-23, then a neighbor with the genotype Pgi-22, Pgm-22 would differ by two alleles detectable as outcrosses (at Pgi) from the target. The local mating index for the target plant is calculated as the sum of the number of different alleles x flowers on all surrounding plants, divided by the total number of alleles x flowers (including those on the target plant) within its neighborhood. The LME index varies from 0 to 
1. Low values indicate that few unique alleles are found in the neighborhood, making it difficult to detect outcrossing (thus, biparental inbreeding would be high). High values of the LME index indicate that neighboring plants are genetically distinct from the target and that outcrossed pollen will be more easily detected in the target's progeny (and biparental inbreeding would be low).
Effects of floral and ecological characteristics on the mating system
Population and group level estimates of mating system parameters were tested for significance by the nonoverlap of their 95% CIs. The significant deviation of estimates from zero or one and the difference between estimates for different floral display size groups were tested with two-tailed tests. The relationships among floral characters (inflorescence and flower number), the LME index, and individual plant outcrossing rates were examined using linear regression in Statview for Windows (ver. 5.0.1, SAS Institute, Cary, North Carolina, USA). The joint effects of plant size and LME on outcrossing rate were tested with multiple regression in Statview.
RESULTS
Distribution of plant size
The size of flowering displays ranged from 1–26 inflorescences, with a median size of four flowering stalks ( = 6.1 ± 0.48 SE, N = 144). The total number of flowers per plant ranged from 2–1413, with a median of 71 ( = 169.2 ± 23.3 SE, N = 122). There was a strong correlation between flower number (Fls) and inflorescence number (Infls) per plant (r = 0.89), as well as between inflorescence number and plant diameter (r = 0.74). The 51 plants used in the mating system analysis had a size distribution similar to the population as a whole, with a slight oversampling of large plants (median Infls = 6.0, range 1–26; median Fls = 111, range 2–1382).
Outcrossing rates
Population level
Pollen and ovule allele frequencies did not differ significantly from each other or from the allele frequencies of the 144 sampled adult plants in the population (based on overlapping CIs, not shown; Table 1). The multilocus outcrossing rate (tm) for the population was 0.685 (±0.036 SE), indicating a mixed mating system (Table 2). The correlation of selfing among families (rs-families = 0.197) was low, indicating high variance in outcrossing rates among plants. Although the single-locus outcrossing estimates for Pgi and Pgm differed substantially, the average of single locus estimates (ts = 0.661 ± 0.034) was only slightly lower than tm (Tables 1 and 2). Two measures of biparental inbreeding gave differing results (Table 2). The difference between tm – ts was low (0.024 ± 0.010), suggesting little mating among relatives. In contrast, the low correlation of selfing among loci (rs-loci = 0.507 ± 0.173) suggests that as much as half of the apparent selfing occurring in the population is the result of biparental inbreeding.
|
= 0.635 ± 0.033). The estimated maternal fixation index (F) was 0.203 (±0.009). Estimates of tm slightly exceeded unity (1.025–1.200) for four families with doubly heterozygous maternal genotypes. These high "out-of-bound" estimates were truncated to 1.0, and truncated tm estimates for individual plants averaged 0.629 (±0.031). Minimum outcrossing estimates (tmin) averaged only 0.383 (±0.028). There was a moderately strong correlation between tm and tmin (r = 0.766), indicating that the rank order of outcrossing rates among plants is a good estimate of their true relative magnitudes.
Relationship of floral display size with outcrossing rates and biparental inbreeding
Group level
When plant size is used as a categorical variable, small plants (1–8 inflorescences) outcross at a significantly higher rate (tm = 0.763 ± 0.039) than large plants (>8 inflorescences, tm = 0.542 ± 0.046; Table 3A). There is a trend for larger plants to self at higher rates when plants are grouped into five categories (Table 3B). However, smaller group sample size and higher standard errors lead to overlap of the 95% CIs among all groups. Estimates of tmin also suggest that small plants outcross more frequently than larger plants, but the differences among groups are not significant (Table 3).
Both tm – ts and the correlation of selfing among loci (rs) indicate that significant biparental inbreeding occurs (Table 3). Biparental inbreeding appears to contribute more strongly to the selfing rate of small than large plants. For the two-group comparison, only the estimates for small plants differ significantly from the null expectation of no biparental inbreeding (tm – ts > 0; rs < 1.0). The trend for decreasing biparental inbreeding with display is not significant in the five-group comparison. The two estimation methods indicate that very different magnitudes of biparental inbreeding contribute to effective selfing (1 – ts = 0.283) in the population. The difference tm – ts accounts for only 16% (0.045/0.283) of the effective selfing rate, while the correlation of selfing method estimates that biparental inbreeding contributes up to 86% (1 – rs = 0.863) of effective selfing in small plants.
Individual plant level
There was a significant negative association between individual plant size and outcrossing rate, albeit with considerable scatter in the data. The regression of tm on inflorescence number is significant, although the correlation is fairly weak (tm = 0.714 – 0.011 x Infls, P = 0.026, r = 0.312; Fig. 1A). The negative relationship between tm and total flower number per plant was only marginally significant (tm = 0.68 – 0.0002 x Fls, P = 0.073, r = 0.253; Fig. 1B). Floral display size was not as good a predictor of the minimum outcrossing rate (tmin = 0.444 – 0.008 x Infls, P = 0.074, r = 0.252; tmin = 0.417 – 0.0001 x Fls, P = 0.145, r = 0.207).
|
|
= 0.282 ± 0.025; Fig. 3) and varied from 0.0–0.87. Low LME indicates that most plants are surrounded by genetically similar neighbors, supporting the spatial autocorrelation results. There was only a marginally significant positive association between individual plant outcrossing rates (tm) and LME (tm = 0.538 + 0.323 x LME, P = 0.069, r = 0.257; Fig. 3A). However, LME was strongly associated with minimum outcrossing estimates per plant (tmin = 0.176 + 0.735 x LME, P < 0.0001, r = 0.660; Fig. 3B).
|
DISCUSSION
Large floral display size is predicted to increase levels of self-fertilization as the result of more intraplant pollinator movements and higher levels of geitonogamy (Geber, 1985
; de Jong et al., 1993
; Snow et al., 1996
). This prediction was strongly supported in a population of D. barbeyi, where plants with large floral displays outcross significantly less frequently than plants with small displays. Local genetic structure also facilitates mating among relatives, further lowering the outcrossing rate. Three functional modes of inbreeding—autogamy, geitonogamy (Lloyd and Schoen, 1992
), and biparental inbreeding (Uyenoyama, 1986
)—all appear to contribute to the total effective selfing rate (Ritland, 1984
), and their relative contributions vary with floral display size and local genetic structure. However, none of these factors explain more than 14% of the variation in outcrossing rates among plants. In the following sections, I examine the functional explanations for these patterns and their potential evolutionary significance.
Mechanism of selfing
Floral display size (number of flowers and inflorescences) varies widely among D. barbeyi plants, and individuals with large floral displays outcross at significantly lower rates than do those with small displays. On average, plants with a single inflorescence produce 79% outcrossed seeds, while the largest plants with >8 flowering stalks outcross only 54% of the time. Inflorescence number was a better predictor of individual plant outcrossing rate than flower number. This is expected if selfing occurs predominantly by geitonogamy between flowers on different inflorescences rather than among flowers within an inflorescence or within individual flowers (autogamy). Eckert's (2000) investigation of modes of self-fertilization in Decodon verticillatus is the only other study to specifically document the effect of multiple flowering branches on the mating system. He found that multi-branched plants selfed at higher frequencies than plants with a single inflorescence, and that the number of flowers on other branches was negatively correlated to outcrossing rates (Appendix). Geitonogamy, particularly between-branch crossing, was the predominant (82%) mode of selfing (Eckert, 2000
). Reducing the synchrony of flowering among different branches of multi-stemmed plants may limit this form of geitonogamy but at the potential cost of reduced pollinator attraction.
The flowers of D. barbeyi, like its sympatric congener D. nuttallianium, are protandrous—anthers mature before stigmas—and flowers mature from the base of the inflorescence upward (Pyke, 1978
; Williams et al., 2001
). Bumblebees tend to forage from the bottom to the top of vertical, spikelike inflorescences (Pyke, 1978
; Waddington and Heinrich, 1979
; Best and Bierzychudek, 1982
), thus visiting female phase flowers first. These coupled patterns of floral development and pollinator foraging behavior in plants with vertical inflorescences should minimize both autogamy and within branch geitonogamy (Darwin, 1877
; Pyke, 1978
; Harder et al., 2000
; Williams et al., 2001
). Indeed, the high outcrossing rates in D. barbeyi plants with a small, single inflorescence (79%), that have no opportunity for between-branch pollinator movement, resemble those found in its single-stalked congener D. nuttallianum (88–97%; Williams et al., 2001
) further suggesting that between-branch geitonogamy is the main mode of selfing in multi-stemmed plants.
Inter- and intraplant variation in floral reward and development may contribute to the relatively weak association (r = 0.31) between display size and outcrossing rate in this population of D. barbeyi. Variance in flower size among plants (Ishii and Harder, 2006
) and in nectar reward among flowers on a plant (Bernaskie and Cartar, 2004
) affect intraplant pollinator movement and geitonogamy. Variation in the degree protandry may also affect individual flower and whole inflorescence geitonogamy and outcrossing rates (Harder et al., 2000
; Galloway et al., 2002
; Routley and Husband, 2003
). Upward foraging bumblebees strongly increased selfing in terminal flowers of nonprotandrous Eichhornia paniculata inflorescences (Barrett et al., 1994
; Harder and Barrett, 1995
; Harder et al., 2000
). If the degree of protandry differs among individual Delphinium plants, then variation in autogamy and within-branch geitonogamy may contribute to the high variation in outcrossing rates among plants that cannot be explained by differences in display size.
Magnitude of display size effect on selfing
The range of floral display size among D. barbeyi plants is much greater than in previously studied species. However, the range of outcrossing rate variation is only slightly greater than that found in species with much smaller displays (Appendix). Nevertheless, species with a small range of display sizes also tended to have a smaller range of outcrossing rates (Appendix). Outcrossing rate estimates for the 51 individual D. barbeyi plants spanned a much wider range than group estimates, varying from 0.16–1.00 (Fig. 1). Other studies that have examined continuous variation in the relationship between floral display size and individual plant outcrossing rates found a similarly wide range of outcrossing rates among plants and fruits (Crawford, 1984
; Brunet and Eckert, 1998
; Vrieling et al., 1999
; Karron et al., 2004
).
Inflorescence number in D. barbeyi has an influence on outcrossing rate variation similar to that of flower number in single-stalked plants of other species (Appendix). However, both inflorescence and total flower number were rather weak predictors (r2 = 0.10 and 0.06, respectively) of individual plant outcrossing rates. A potential cause of the low predictive power of these variables is that they are measures of total, rather than daily, display size (de Jong et al., 1992
; Barrett et al., 1994
). Both the number of open flowers and the number of inflorescences with open flowers vary throughout the flowering season. Inflorescences do not mature synchronously (personal observation), and daily flower number is only a fraction of total flower number (Williams et al., 2001
). Karron et al. (2004)
have shown that outcrossing rates vary widely among fruits on the same plant and among plants with the same floral display size. The stochastic nature of pollinator visitation patterns (Mitchell et al., 2004
, 2005
) and pollen carryover (Barrett et al., 1994
), combined with daily variation in floral display, likely contribute to the low explanatory power of total display size or inflorescence number for outcrossing estimates derived from bulk seed samples used in this study. Despite these limitations of the experimental design, the correlations of outcrossing rate to floral display size are still comparable to those in other studies that used daily rather than total flower number (r2 = 0.001–0.34; Brunet and Eckert, 1998
; Vrieling et al., 1999
; Eckert, 2000
; Herlihy and Eckert, 2004
).
Local population structure and biparental inbreeding
Another factor potentially contributing to variation in outcrossing rates among plants is biparental inbreeding. Several lines of evidence suggest that biparental inbreeding is an important component of the effective selfing rate in D. barbeyi. First, significant local population structure was detected by multilocus spatial autocorrelation in the smallest distance class (2.5-m radius) in this population (see also Williams et al., 2001
). The additional, marginally significant peaks at greater distances may be due to a few patches of genetically similar or dissimilar individuals within the sampled 14 x 26 m plot, because this plot size is small relative to the 8-m mean patch diameter of genetically related plants. Stochastic variation among loci, when autocorrelation is averaged over few loci, may also lead to the nonmonotonic decline in genetic similarity seen in the correlogram (Slatkin and Arter, 1991
; Williams, 1994
).
Second, the local mating environment (LME) significantly affected outcrossing rates of individual plants. Plants surrounded by a high floral density of genetically related individuals (low LME) had lower outcrossing rates than those with genetically dissimilar neighbors (high LME). This was particularly true for minimum estimates of outcrossing (tmin), which should be more sensitive to biparental inbreeding than multilocus estimators (Shaw et al., 1981
; Ritland, 2002
; Griffin and Eckert, 2003
). Eckert (2000)
found a similar relationship in a clonal plant in that ramets with a higher density of clonal neighbors had higher selfing rates.
Third, single inflorescence plants had the highest outcrossing rates (0.787 ± 0.075), but these estimates were still significantly less than complete outcrossing. Single-stalked plants, which had a mean flower number of only 12.7, should experience little or no geitongamy because of the typical bottom-up pollinator foraging behavior (discussed earlier). Selfing in these plants should occur because of autogamy and/or biparental inbreeding. Estimates of autonomous selfing rates are only 9% based on seed set in pollinator exclosures (Williams et al., 2001
). Assuming that additional facilitated selfing in these protandrous flowers is negligible, then over half of the effective selfing rate of 21% in single-inflorescence plants may be due to biparental inbreeding.
Finally, the two estimators of biparental inbreeding, although differing widely in magnitude, suggest that mating between relatives occurs and is a more important component of effective selfing in small than large plants (Table 3). Larger plants, which experience more intraplant pollinator movements, should receive proportionally less outcross pollen and effectively self more through geitonogamy than biparental inbreeding. Small plants with little geitonogamy should receive proportionally more outcross pollen. Effective selfing and biparental inbreeding therefore should be more strongly influenced by LME in small plants. However, large plants with high rates of geitonogamy may have reduced male siring success because of pollen discounting (Harder and Barrett, 1995
), making them less effective pollen donors and less likely to contribute to biparental inbreeding. This may act to counter some of the predicted effects of LME on biparental inbreeding and hence help explain the weak relationship between LME and outcrossing.
Contributions of autogamy, geitonogamy, and biparental inbreeding to selfing
The proportion of effective selfing due to geitonogamy in plants with different floral display sizes can be estimated by assuming that single-stalked plants do not self by geitonogamy (or do so at a very low rate), and that biparental inbreeding remains constant or decreases with increasing floral display size. As shown earlier, both of these assumptions are reasonable for D. barbeyi. The geitonogamy fraction of total selfing for each floral display size class can be derived (Table 4) using the methods of Schoen and Lloyd (1992)
, where selfing rates (s = 1 – tm) are calculated from the outcrossing rates of plants in different inflorescence size classes (Table 3). Because the total effective selfing rate (sn) increases with floral display size, the fraction due to geitonogamy (%gn) increases from zero in single-inflorescence plants to over 60% of the total inbreeding in the largest size class. Biparental inbreeding should contribute to effective selfing in both autogamous and geitonogamous fractions. However, because the magnitude of biparental inbreeding appears to decrease with increasing display size (see earlier), the proportion of selfing due to geitonogamy in large plants may be an underestimate.
|
; Herlihy and Eckert, 2004
) and more variable genetic markers (Ritland, 2002
) to better tease apart the relative contributions of autogamy, geitonogamy, and biparental inbreeding and their associations with floral display in Delphinium.
Evolutionary implications
Geitonogamy may be an unavoidable consequence of selection for outcrossing in plants with large floral displays. As such, geitonogamy may help explain the prevalence of mixed mating systems (Lloyd, 1992
; de Jong et al., 1993
; Harder and Barrett, 1996
; Eckert and Herlihy, 2004
; Goodwillie et al., 2005
; Brunet and Sweet, 2006b). The magnitudes of the potential fitness costs associated with high levels of inter-inflorescence geitonogamy in D. barbeyi (inbreeding depression, pollen and ovule discounting) and how they are balanced by the reproductive benefits of large floral display (pollinator attraction, pollen receipt and removal) are currently unknown. If geitonogamous selfing is costly, then selection should act to limit maximum display size and/or reduce the costs of geitonogamy by other means (de Jong et al., 1992
). Evidence suggests that a history of inbreeding has affected the magnitude and timing of inbreeding depression in D. barbeyi. Early-acting inbreeding depression appears low, because selfing does not reduce seed set compared to outcrossing. However, adult plants are much less inbred than predicted from the mating system, suggesting that later acting inbreeding depression occurs (Williams et al., 2001
). A comparison of whole plant seed set, siring success, and seedling recruitment across a range of floral display sizes, as well as a better understanding of the magnitude and timing of inbreeding depression, will be necessary to evaluate the fitness costs and benefits of floral display in D. barbeyi and other species with multiple inflorescences.
|
1 The author thanks R. Smith, D. Woods, E. Olson, and C. Cortes Baez for help in the field; J. Wheeler for help in the lab; and J. Tuttle for allowing work on his property. P. Murphy, R. Smith, M. Johnston, S. Weller, and an anonymous reviewer provided helpful comments on the manuscript. The Faculty Research Committee, University Research Committee, and NIH Bioinformatics Research Infrastructure Network at Idaho State University provided funding. This paper is dedicated to the memory of William A. Calder III, whose life and work inspired several generations of field biologists at the RMBL. 
LITERATURE CITED
Ashman T.-L. Knight T. M. Steets J. A. Amarasekare P. Burd M. Campbell D. R. Dudash M. R. Johnston M. O. Mazer S. J. Mitchell R. J. Morgan M. T. Wilson W. G.. 2004. Pollen limitation of plant reproduction: ecological and evolutionary causes and consequences. Ecology 85: 2408-2421..[CrossRef][Web of Science]
Barrett S. C. H. Harder L. D. Cole W. W.. 1994. Effects of flower number and position on self-fertilization in experimental populations of Eichhornia paniculata (Pontederiaceae). Functional Ecology 8: 526-535..[CrossRef]
Bernaskie J. M. Cartar R. V.. 2004. Variation in rate of nectar production depends on floral display size: a pollinator manipulation hypothesis. Functional Ecology 18: 125-129..[CrossRef]
Best L. S. Bierzychudek P.. 1982. Pollinator foraging on foxglove (Digitalis purpurea): a test of a new model. Evolution 36: 70-79..[CrossRef][Web of Science]
Brunet J. Eckert C. G.. 1998. Effects of floral morphology and display on outcrossing in blue columbine, Aquilegia caerulea (Ranunculaceae). Functional Ecology 12: 596-606..[CrossRef]
Brunet J. Sweet H. R.. 2006a. Impact of insect pollinator group and floral display size on outcrossing rate. Evolution 60: 234-246..[Web of Science][Medline]
Brunet J. Sweet H. R.. 2006b. The maintenance of selfing in a population of the Rocky Mountain columbine. International Journal of Plant Sciences 167: 213-219..[CrossRef][Web of Science]
Charlesworth B.. 1980. The cost of sex in relation to mating system. Journal of Theoretical Biology 84: 655-671..[Web of Science][Medline]
Charlesworth D. Charlesworth B.. 1987a. The effect of investment in attractive structures on allocation to male and female functions in plants. Evolution 41: 948-968..[CrossRef][Web of Science]
Charlesworth D. Charlesworth B.. 1987b. Inbreeding depression and its evolutionary consequences. Annual Review of Ecology and Systematics 18: 237-268..[CrossRef][Web of Science]
Crawford T. J.. 1984. What is a population?. In B. Shorrocks, [ed.], Evolutionary ecology, 135-173. Blackwell, London, UK..
Cruzan M. B. Arnold M. L.. 1994. Assortative mating and natural selection in hybridizing irises: effects of phenology and floral densities on family outcrossing rates. Heredity 72: 95-105..[Web of Science]
Darwin C. R.. 1876. The effects of cross- and self-fertilization in the vegetable kingdom. Murray, London, UK..
Darwin C. R.. 1877. The different forms of flowers on plants of the same species. Murray, London, UK..
De Jong T. J. Klinkhamer P. G. L. Van Staalduinen M. J.. 1992. The consequences of pollination biology for selection of mass or extended blooming. Functional Ecology 6: 606-615..[CrossRef]
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..[CrossRef]
Eckert C. G.. 2000. Contributions of autogamy and geitonogamy to self-fertilization in a mass flowering, clonal plant. Ecology 81: 532-542..[CrossRef][Web of Science]
Eckert C. G. Herlihy C. R.. 2004. Using a cost-benefit approach to understand the evolution of self-fertilization in plants: the perplexing case of Aquilegia canandensis (Ranunculaceae). Plant Species Biology 19: 159-173..[CrossRef]
Emms S. K. Stratton D. A. Snow A. A.. 1997. The effect of inflorescence size on male fitness: experimental tests in the andromonoecious lily, Zigadenus paniculatus. Evolution 51: 1481-1489..[CrossRef][Web of Science]
Finer M. S. Morgan M. T.. 2003. Effects of natural rates of geitonogamy on fruit set in Asclepias speciosa (Apocynaceae): evidence favoring the plant's dilemma. American Journal of Botany 90: 1746-1750..
Galloway L. F. Cirigliano T. Gremski K.. 2002. The contribution of display size and dichogamy to potential geitonogamy in Campanula americana. International Journal of Plant Sciences 163: 133-139..[CrossRef][Web of Science]
Geber M. A.. 1985. The relationship of plant size to self-pollination in Mertensia ciliata. Ecology 66: 762-772..[CrossRef][Web of Science]
Goodwillie C. Kalisz S. Eckert C. G.. 2005. The evolutionary enigma of mixed mating systems in plants: occurrence, theoretical explanations, and empirical evidence. Annual Review of Ecology, Evolution, and Systematics 36: 47-79..[CrossRef][Web of Science]
Griffin C. A. M. Eckert C. G.. 2003. Experimental analysis of biparental inbreeding in a self-fertilizing plant. Evolution 57: 1513-1519..[Web of Science][Medline]
Handel S. N.. 1985. The intrusion of clonal growth patterns on plant breeding systems. American Naturalist 125: 367-384..[CrossRef][Web of Science]
Harder L. D. Barrett S. C. H.. 1995. Mating cost of large floral displays in hermaphroditic plants. Nature 373: 512-515..[CrossRef][Web of Science]
Harder L. D. Barrett S. C. H.. 1996. Pollen dispersal and mating patterns in animal pollinated plants. In D. G. Lloyd, S. C. H. Barrett, [eds.], Floral biology: studies on floral evolution in animal-pollinated plants, 140-190. Chapman and Hall, New York, New York, USA..
Harder L. D. Barrett S. C. H. Cole W. W.. 2000. The mating consequences of sexual segregation within inflorescences of flowering plants. Proceedings of the Royal Society of London, B, Biological Sciences 267: 315-320..
Harder L. D. Williams N. M. Jordan C. Y. Nelson W. A.. 2001. The effects of floral design and display on pollinator economics and pollen dispersal. In L. Chittka, J. D. Thomson, [eds.], Cognitive ecology of pollination: animal behavior and floral evolution, 297-317. Cambridge University Press, Cambridge, UK..
Herlihy C. R. Eckert C. G.. 2004. Experimental dissection of inbreeding and its adaptive significance in a flowering plant, Aquilegia canadensis (Ranunculaceae). Evolution 58: 2693-2703..[Web of Science][Medline]
Hessing M. B.. 1988. Geitonogamous pollination and its consequences in Geranium caespitosum. American Journal of Botany 75: 1324-1333..[CrossRef][Web of Science]
Heywood J. S.. 1991. Spatial analysis of genetic variation in plant populations. Annual Review of Ecology and Systematics 22: 335-355..[CrossRef][Web of Science]
Holsinger K. E. Feldman M. W. Christiansen F. B.. 1984. The evolution of self-fertilization in plants. American Naturalist 124: 446-453..[CrossRef][Web of Science]
Huang S.-Q. Tang L.-L. Sun J.-F. Lu Y.. 2006. Pollinator response to female and male floral display in a monoecious species and its implications for the evolution of floral dimorphism. New Phytologist 171: 417-424..[CrossRef][Web of Science][Medline]
Husband B. C. Schemske D. W.. 1996. Evolution of the magnitude and timing of inbreeding depression in plants. Evolution 50: 54-70..[CrossRef][Web of Science]
Ishii H. S. Harder L. D.. 2006. The size of individual Delphinium flowers and the opportunity for geitonogamous pollination. Functional Ecology 20: 1115-1123..[CrossRef]
Ivey C. T. Wyatt R.. 1999. Family outcrossing rates and neighborhood floral density in natural populations of swamp milkweed (Asclepias incarnata): potential statistical artifacts. Theoretical and Applied Genetics 98: 1063-1071..[CrossRef][Web of Science]
Johnston M. O.. 1998. Evolution of intermediate selfing rates in plants: pollination ecology versus deleterious mutations. Genetica 102/103: 267-278..
Kalisz S. Vogler D. W.. 2003. Benefits of autonomous selfing under unpredictable pollinator environments. Ecology 84: 2928-2942..[CrossRef][Web of Science]
Karron J. D. Mitchell R. J. Holmquist K. G. Bell J. M. Funk B.. 2004. The influence of floral display size on selfing rates in Mimulus ringens. Heredity 92: 242-248..[CrossRef][Web of Science][Medline]
Klinkhamer P. G. L. De Jong T. J.. 1993. Attractiveness to pollinators: a plant's dilemma. Oikos 66: 180-184..[CrossRef][Web of Science]
Klinkhamer P. G. L. De Jong T. J. De Bruyn G.-J.. 1989. Plant size and pollinator visitation in Cynoglossum officinale. Oikos 54: 201-204..[CrossRef][Web of Science]
Langenheim J. H.. 1962. Vegetation and environmental patterns in the Crested Butte area, Gunnison County, Colorado. Ecological Monographs 32: 249-285..[CrossRef][Web of Science]
Leclerc-Potvin C. Ritland K.. 1994. Modes of self-fertilization in Mimulus guttatus (Scrophulariaceae): a field experiment. American Journal of Botany 81: 199-205..[CrossRef][Web of Science]
Levin D. A. Kerster H. W.. 1974. Gene flow in seed plants. Evolutionary Biology 7: 139-220..
Lloyd D. G.. 1979. Some reproductive factors affecting the selection of self-fertilization in plants. American Naturalist 113: 67-79..[CrossRef][Web of Science]
Lloyd D. G.. 1992. Self- and cross-fertilization in plants. II. The selection of self-fertilization. International Journal of Plant Sciences 151: 370-380..
Lloyd D. G. Schoen D. J.. 1992. Self- and cross-fertilization in plants. I. Functional dimensions. International Journal of Plant Sciences 151: 358-369..
Mitchell R. J.. 1994. Effects of floral traits, pollinator visitation, and plant size on Ipomopsis aggregata fruit production. American Naturalist 143: 870-889..[CrossRef][Web of Science]
Mitchell R. J. Karron J. D. Holmquist K. G. Bell J. M.. 2004. The influence of Mimulus ringens floral display size on pollinator visitation patterns. Functional Ecology 18: 116-124..[CrossRef]
Mitchell R. J. Karron J. D. Holmquist K. G. Bell J. M.. 2005. Patterns of multiple paternity in fruits of Mimulus ringens (Phrymaceae). American Journal of Botany 92: 885-890..
Morgan M. T. Wilson W. G.. 2005. Self-fertilization and the escape from pollen limitation in variable pollination environments. Evolution 59: 1143-1148..[Web of Science][Medline]
Ohashi K. Yahara T.. 2001. Behavioral responses of pollinators to variation in floral display size and their influences on the evolution of floral traits. In L. Chittka, J. D. Thomson, [eds.], Cognitive ecology of pollination: animal behavior and floral evolution, 274-296. Cambridge University Press, Cambridge, UK..
Peakall R. Smouse P. E.. 2001. GenAlEx V5: genetic analysis in Excel. Population genetic software for teaching and research. Australian National University, Canberra, Australia. Website http://www.anu.edu.au/BoZo/GenAlEx [accessed 5 August 2007]..
Pyke G. H.. 1978. Optimal foraging in bumble bees and coevolution with their plants. Oecologia 36: 281-293..[CrossRef][Web of Science]
Ritland K.. 1984. The effective proportion of self-fertilization with consanguineous matings in inbred populations. Genetics 106: 139-152..
Ritland K.. 2002. Extensions of models for the estimation of mating systems using n independent loci. Heredity 88: 221-228..[CrossRef][Web of Science][Medline]
Routley M. B. Husband B. C.. 2003. The effect of protandry on siring success in Chamerion angustifolium (Onagraceae) with different inflorescence sizes. Evolution 57: 240-248..[Web of Science][Medline]
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][Web of Science]
Shaw D. V. Kahler A. L. Allard R. W. 1981. A multilocus estimator of mating system parameters in plant populations. Proceedings of the National Academy of Sciences, USA 78: 1298-1302..
Slatkin M. Arter H. E.. 1991. Spatial autocorrelation methods in population genetics. American Naturalist 138: 499-517..[CrossRef][Web of Science]
Smouse P. E. Peakall R.. 1999. Spatial autocorrelation analysis of individual multiallele and multilocus genetic structure. Heredity 82: 561-573..[CrossRef][Web of Science][Medline]
Snow A. A. Spira T. P. Simpson R. Klips R. A.. 1996. The ecology of geitonogamous pollination. In D. G. Lloyd, S. C. H. Barrett, [eds.], Floral biology: studies on floral evolution in animal-pollinated plants, 191-215. Chapman and Hall, New York, New York, USA..
Sun M. Ganders F. R.. 1988. Mixed mating systems in Hawaiian Bidens (Asteraceae). Evolution 42: 516-527..[CrossRef][Web of Science]
Uyenoyama M. K.. 1986. Inbreeding and the cost of meiosis: the evolution of selfing in populations practicing partial biparental inbreeding. Evolution 40: 388-404..[CrossRef][Web of Science]
Uyenoyama M. K. Holsinger K. E. Waller D. M.. 1993. Ecological and genetic factors directing the evolution of self fertilization. Oxford Surveys in Evolutionary Biology 9: 327-381..
Vaughton G. Ramsey M.. 1998. Floral display, pollinator visitation and reproductive success in the dioecious perennial herb Wurmbea dioica (Liliaceae). Oecologia 115: 93-101..[CrossRef][Web of Science]
Vrieling K. Samitou-Laprade P. Cuguen J. Van Dijk H. De Jong T. J. Klinkhamer P. G. L.. 1999. Direct and indirect estimates of the selfing rate in small and large individuals of the bumblebee pollinated Cynoglossum officinale L. (Boraginaceae). Ecology Letters 2: 331-337..[CrossRef][Web of Science]
Waddington K. D. Heinrich B.. 1979. The foraging movement of bumble-bees on vertical inflorescences: an experimental analysis. Journal of Comparative Physiology 134: 113-117..[CrossRef][Web of Science]
Waller D. M.. 1993. The statics and dynamics of mating system evolution. In N. W. Thornhill, [ed.], The natural history of inbreeding and outbreeding, 97-117. University of Chicago Press, Chicago, Illinois, USA.
Warnock M. J.. 1997. Delphinium. In Flora of North America Editorial Committee, [ed.], Flora of North America north of Mexico, vol. 3, 196-240. Oxford University Press, New York, New York, USA..
Waser N. M.. 1982. A comparison of distances flown by different visitors to flowers of the same species. Oecologia 55: 251-257..[CrossRef][Web of Science]
Williams C. F.. 1994. Genetic consequences of seed dispersal in three sympatric forest herbs. II. Microspatial genetic structure within populations. Evolution 48: 1959-1972..[CrossRef][Web of Science]
Williams C. F. Ruvinsky J. Scott P. E. Hews D. K.. 2001. Pollination, breeding system, and genetic structure in two sympatric Delphinium (Ranunculaceae) species. American Journal of Botany 88: 1623-1633..
Williams C. F. Waser N. M.. 1999. Spatial genetic structure of Delphinium nuttallianum populations: inferences about gene flow. Heredity 83: 541-550..[CrossRef][Web of Science][Medline]
Willson M. F. Rathcke B. J.. 1974. Adaptive design of the floral display in Asclepias syriaca L. American Midland Naturalist 92: 47-57..[CrossRef][Web of Science]
This article has been cited by other articles:
|
|
 |
|
  R. Zhao, H. xia, and B.-r. Lu Fine-scale genetic structure enhances biparental inbreeding by promoting mating events between more related individuals in wild soybean (Glycine soja; Fabaceae) populations Am. J. Botany, June 1, 2009; 96(6): 1138 - 1147. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. D. Karron, K. G. Holmquist, R. J. Flanagan, and R. J. Mitchell Pollinator visitation patterns strongly influence among-flower variation in selfing rate Ann. Bot., June 1, 2009; 103(9): 1379 - 1383. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
