| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
0 Department of Botany, Duke University, Durham, North Carolina 27708-0338 USA; and Department of Zoology, Duke University, Durham, North Carolina 27708-0325 USA
Received for publication January 29, 1999. Accepted for publication July 1, 1999.
ABSTRACT
A polymorphism for anthocyanin production was used as a genetic marker to document the relationship between anther–stigma separation and outcrossing rate in the predominantly self-fertilizing weed Datura stramonium. White-flowered plants that differed in anther–stigma separation were placed into populations consisting exclusively of purple-flowered plants. Self vs. outcross origin of progeny was evident in the hypocotyl color of the seedlings. Outcrossing rates measured for single flowers were significantly positively correlated with anther–stigma separation, albeit with some scatter around the regression line, especially for flowers with exserted stigmas. We also performed an 8 x 8 diallel cross to determine whether anther–stigma separation is genetically determined. Heritability in two field plots was ~0.3 and in the greenhouse was ~0.2. Maternal effects, epistasis, and dominance appeared to be relatively unimportant. Genotypes performed consistently across the three environments, although total plant size varied more than fivefold. It appears that the mixed-mating system of D. stramonium has a heritable basis and would be capable of responding to selection.
Key Words: anther–stigma separation • Datura • diallel • heritability • mating system • outcrossing • selfing • Solanaceae
The evolution of mixed-mating systems in angiosperms is an enduring problem in plant biology (e.g., Darwin, 1878
; Bodmer, 1958
; Clegg, 1980
; Barrett and Eckert, 1990
) that has attracted renewed attention from workers interested in measuring how various traits influence the degree of outcrossing in an assortment of self-compatible species (e.g., Horowitz and Harding, 1972
; Jain, 1976
; Rick, Fobes, and Holle, 1977
; Clay, 1982
; Schoen, 1982
; Marshall and Abbott, 1984
; Wyatt, 1984
; Thomson and Stratton, 1985
; Lyons and Antonovics, 1991
; Dole, 1992
). Recent investigations of mating system traits have also focused on the genetics that underlie these characters (e.g., Ennos, 1981
; Shore and Barrett, 1990
; Holtsford and Ellstrand, 1992
; Carr and Fenster, 1994
; Fenster and Barrett, 1994
). Such information is important for understanding the evolution of selfing in different taxa and is especially needed to assess the potential for evolutionary changes in the level of outcrossing within species that already possess a mixed-mating system.
The latter issue is of particular interest for two reasons. First, inbred and outcrossed progeny are likely to differ genetically and hence may differ in fitness. As a result, a plant's mating system may affect the persistence of a population or of particular genotypes within a population. Secondly, it remains unclear under what conditions a mixed-mating system itself can persist and at what relative levels of partial selfing or outcrossing. For example, models based on inbreeding depression generally predict evolution of either complete selfing or complete outcrossing (e.g., Fisher, 1941
; Lande and Schemske, 1985
; Charlesworth, Morgan, and Charlesworth, 1990
), while other models, which include such factors as pollen discounting (Holsinger, 1991
), genetic associations between fitness alleles and mating-system alleles (Uyenoyama and Waller, 1991
), overdominance (Holsinger, 1988
; Charlesworth and Charlesworth, 1990
), or biparental inbreeding (Uyenoyama, 1986
) specify various theoretical conditions for stable intermediate rates of selfing and outcrossing.
This paper is the second in a series considering the evolution of a mixed-mating system in the large annual weed, Datura stramonium, which is distinctive because it has a very high selfing rate, >95% (Motten and Antonovics, 1992
), despite bearing showy flowers attractive to hawk moths. Moreover, the intensive inbreeding has apparently not eliminated the potential for outcrossing, and there is significant variation in outcrossing rate among individual plants (0–19%; Motten and Antonovics, 1992
). The factor most strongly contributing to this variation is the position of the stigma, i.e., whether it overlaps or protrudes above the anthers (Motten and Antonovics, 1992
).
Several factors make D. stramonium particularly well suited for a study of the role of stigma position on outcrossing rates. First, the location of the stigma varies continuously over a wide range, from 4 mm below to 12 mm above the anthers (personal observations; see Fig. 1). Secondly, because flowers are borne singly, outcrossing rates are not affected by inflorescence characteristics and floral display size (Horowitz and Harding, 1972
; Rick, Holle, and Thorpe, 1978
; Motten and Antonovics, 1992
). In addition, the effect of stigma position on outcrossing is not confounded by the influence of other floral traits such as dichogamy (Breese, 1959
; Vasek, 1965
; Schoen, 1982
; Wyatt, 1984
; Holtsford and Ellstrand, 1992
) or stigma curling (Dole, 1992
), and varies independently of flower size (Rick, Holle, and Thorp, 1978
; Carr and Fenster, 1994
). And lastly, precise outcrossing estimates can be obtained from individual flowers because the capsules produce hundreds of seeds whose genotypes can be readily assessed by a color marker expressed in the seedlings' hypocotyls (Motten and Antonovics, 1992
).
|
; Faegri and van der Pijl, 1979
), the significance of continuous variation in stigma position has not often been studied directly and quantitatively under natural pollination conditions (Karron et al., 1997
). Many previous studies have either relied on correlations between outcrossing rate and mean stigma position across different populations (Barrett and Shore, 1987
; Dole, 1992
; Belaoussoff and Shore, 1995
) or are based on rates of autogamy in a greenhouse rather than actual rates of outcrossing in the presence of natural pollinators (Ennos, 1981
; Epperson and Clegg, 1987
; Carr and Fenster, 1994
; Robertson, Diaz, and Macnair, 1994
). In other studies, outcrossing has been assessed indirectly, by counting the pollen grains deposited by floral visitors on stigmas of different heights (Thomson and Stratton, 1985
; Murcia, 1990
). In addition, despite a growing literature on the quantitative genetics of stigma position (Ennos, 1981
; Shore and Barrett, 1990
; Holtsford and Ellstrand, 1992
; Fenster and Barrett, 1994
; Fenster and Ritland, 1994
; Robertson, Diaz, and Macnair, 1994
), the number of taxa investigated thus far is small, and none are as highly self-fertilizing as D. stramonium. Our main goal in this investigation was to determine the potential for evolution of outcrossing rate through change in stigma position. The study consisted of two parts: (1) measuring the effectiveness of stigma position in promoting outcrossing by natural pollinators and (2) analyzing the underlying genetic basis for stigma position and estimating its heritability under field conditions.
MATERIALS AND METHODS
Stigma position and outcrossing rate
We measured outcrossing rates (t) of individual flowers by taking advantage of a naturally occurring polymorphism for hypocotyl and flower color. The polymorphism is inherited in simple Mendelian fashion (Blakeslee and Avery, 1917
) with purple hypocotyls and purple-tinged flowers dominant over green hypocotyls and pure white flowers. Because of this trait, outcrossing events in white-flowered plants could be easily recognized by the presence of seedlings with purple hypocotyls, allowing us to obtain accurate estimates of t for flowers with many different stigma positions. Previous work (Motten and Antonovics, 1992
) has shown that pollinators do not discriminate between the two flower color morphs.
Measurements of t were made using potted, white-flowered plants placed into populations of exclusively purple-flowered individuals. Initial measurements were made during the summer of 1984 in a large population (>200 plants, distributed over >1000 m2) east of Infinity Road in Durham County, North Carolina, using 1–4 potted plants per night. More extensive measurements were made in 1986 in a smaller population (~50 individuals in <300 m2) at the Field Station of the Duke University Department of Zoology, also in Durham County, using 1–3 potted plants per night. In both populations potted plants were placed overnight within clumps of purple-flowered plants. Where necessary, the pots were raised on blocks so the white flowers would be at the same height as the surrounding purple flowers. On nights when more than one potted plant was used, pots were separated by at least 10 m. The potted plants had only one open flower per night and were set out during the morning of the day the flower bud was ready to open, before the anthers had started to dehisce. Plants were removed the following morning, by which time further pollination was unlikely because the style was ready to abscise from the ovary.
Anther–stigma separation of the flowers on the potted plants was determined early in the morning after pollination. Two distances were measured with calipers on each flower: (a) from the base of the ovary to the top of the anthers, and (b) from the base of the ovary to the tip of the stigma. Both distances were measured to the nearest 0.1 mm. Stigma position was then calculated by subtracting the height of the anthers from the height of the stigma.
For each flower that subsequently produced a mature capsule, seeds were germinated and scored for purple (heterozygous) seedlings. A minimum of 40 seedlings was scored per capsule, with at least 90 seedlings scored for more than half of the capsules. Because there was little or no opportunity for pollen flow between white flowers, the frequency of heterozygous seedlings from the capsule of a white-flowered, potted plant is equivalent to t.
Quantitative genetic analysis of stigma position
Material for the genetic analysis was collected from a dimorphic population in an overgrazed pasture and barnyard at the intersection of Infinity Road and Snowhill Road in Durham County. The outcrossing rate for this population in 1984 was quite low, 1.3% (Motten and Antonovics, 1992
), so it was likely that most individuals were highly homozygous. To further assure an assumption of homozygozity for the genetic analyses, flowers were bagged in the field to exclude insects and allowed to self-pollinate. Selfed seeds were collected from eight plants representative of the population, including two white-flowered and six purple-flowered individuals. Plants from these seeds were raised in the Botany Department greenhouses at Duke University and subjected to an additional one or two generations of selfing without selection to produce eight inbred lines. One large representative from each of these lines was then grown in the greenhouses of the Duke University phytotron and used in a full 8 x 8 diallel crossing design, including selfs.
Progeny from the diallel cross were raised in three different environments. During the fall of 1986 the progeny were grown in the greenhouse in 5 cm diameter pots. During the summer of 1987 progeny were planted at two field sites that corresponded to natural habitats, the Experimental Plot of the Department of Botany (hereafter referred to as the Botany Plot) and the Field Station of the Department of Zoology. These sites were chosen for their different quality for growth of D. stramonium. The Botany Plot site has a clay soil with dry, nutrient-poor regions. The Field Station site has a sandy loam soil that is rich in nutrients from previous use as a goat pasture. To accentuate the differences between the sites and promote maximum growth at the Field Station, plants at the Field Station were treated twice with Peter's 20–20–20 liquid fertilizer. Estimates of plant size at the two field sites were obtained by measuring basal stem diameters at the end of the growing season.
Plants at all three locations were grown in a completely randomized block design with one replicate of each cross per block. Five blocks were used in the greenhouse and three blocks at each of the two field sites. Heights of anthers and stigma were measured as described above for six flowers per plant in the greenhouse, five flowers per plant at the Botany Plot, and seven flowers per plant at the Field Station. Corolla length was also measured on each flower. Flowers from the same plant were collected on different nights and were chosen at random when more than one was open. Statistical descriptions and analyses of the diallel (described in the Results) are based on mean values per plant.
RESULTS
Stigma position and outcrossing rate
Outcrossing rate increased steadily with increasing projection of the stigma above the anthers (Fig. 2). This pattern was apparent in both years, with the smaller 1984 data set falling well within the range of the 1986 values. A linear regression of outcrossing rate vs. stigma position for flowers with protruding stigmas (stigma height 
0) was significant for each year (r2 = 0.31, P < 0.004 for 1984; r2 = 0.15, P < 0.01 for 1986) and for both years combined (r2 = 0.32, P < 0.0001).
|
0 mm; (b) stigma protruding above but still partially overlapping the anthers, corresponding to stigma positions >0 mm and <3.4 mm (stigmas ranged in length from 1.8–3.4 mm); and (c) stigma completely above the anthers, corresponding to stigma positions >3.5 mm. In the first class, most of the flowers (16 of 22) were totally selfed. The mean value of t was 0.01, with t > 0.05 for only one flower. In the second class, the majority of flowers (17 of 26) were at least partially outcrossed with a mean value of t = 0.053. In the third class, all but one of the 25 flowers experienced some outcrossing, with a mean value of t = 0.165. This class had the greatest range of variation in t and included the two flowers with the highest outcrossing rates, 0.48 and 0.49.
Diallel results
As expected, the size of the progeny from the diallel varied greatly depending on where they were raised (Table 1). The potted, greenhouse-grown plants had the smallest stem diameters and smaller flowers, on average only ~75% as long as those of field-grown plants. Plants from the Field Station were the largest, with an average basal diameter nearly double that of the Botany Plot plants. This difference translates into a 3–4 fold difference in plant height, flower number, and capsule number and a 5–6 fold increase in seed production (A. Motten, unpublished data). However, despite the difference in overall plant size between the two field sites, the mean corolla length did not differ significantly (F = 3.32, P = 0.14 in a nested analysis of variance).
|
|
bio model. The analysis of variance included progeny of selfed individuals and was intended to detect additive genetic variance for stigma position by measuring the maternal, paternal, and maternal x paternal components of variance. Analyses of variance were performed with and without flower size as a covariate. The quadratic analysis did not include progeny from selfs and was intended to partition the variance in stigma position into four components: (a) nuclear general effects, which correspond to additive, nuclear variation, (b) nuclear specific effects, which are equivalent to nonadditive nuclear effects, including dominance and epistasis, (c) reciprocal effects, which include maternal and cytoplasmic effects, and (d) reciprocal specific effects, which represent all other interactions of parental and offspring genomes (Antonovics and Schmitt, 1986
). At all three locations the effect of male and of female parent was highly statistically significant (Table 2), indicating significant additive genetic variance for stigma position. Although the block terms were also highly significant at the two field sites, there were no significant interactions at any of the three locations between block and male or female parent. When flower size was included as a covariate, the maternal x paternal interaction term was marginally statistically significant in the greenhouse; otherwise this interaction was not significant. The quadratic analyses confirmed the results of the ANOVAs with a highly statistically significant nuclear general effect at each location (Table 3). In addition, the quadratic analysis detected a marginally statistically significant nuclear specific effect at the Field Station site.
|
|
The full diallel design made it possible to use the graphical analysis of Mather and Jinks (1977, 1982)
to determine whether the additive-dominance genetic model was sufficient to explain the stigma position results or whether gene interaction must be invoked. For this analysis, we calculated variances and covariances from the phenotypic measures of stigma position (after correcting for environmental error), then regressed the covariance of family means with the means of their nonrecurrent parents (Wr) on the variance of family means (Vr ; see Fig. 4). The resulting line has a slope equal to one if there is no interaction between loci, assuming the parental lines are completely inbred.
|
). For all three sites Wr–Vr did not differ significantly across progeny array when compared with differences over replicate blocks (results not shown). Thus the graphical analysis supports the quadratic analysis in suggesting that epistasis does not appear to play an important role in anther–stigma separation in our population of D. stramonium.
The intercept of the Wr/Vr regression line also gives an estimate of dominance, because the contributions of sampling variance were eliminated from the estimate (Lynch and Walsh, 1998
). As evident in Fig. 4, the degree of dominance was higher at the Field Station than at the other two sites; this agrees with the results from the quadratic analysis. In addition, we used the graphical analysis to assess the relative numbers of recessive alleles carried in different inbred lines. Lines fixed for dominant alleles show low Vr and Wr, while lines with increasing numbers of recessive alleles show high Vr and Wr (Mather and Jinks, 1982
). Whether the trait tends to exhibit directional dominance is revealed by noting whether trait values increase or decrease with the sum of Vr and Wr, an index of increasing numbers of recessive genes. For our data, there is a trend for lines with large anther–stigma separation to have high Vr and Wr, suggesting that alleles for increased separation tend to be recessive. For example, line 2 in Fig. 5 had the highest mean trait value in all three environments, and it consistently had the highest values of Vr and Wr.
|
; see Table 4) does not lend itself to easy calculation of standard errors, the consistency between the two field sites suggests that the estimates are robust. The greenhouse estimate is substantially lower, despite the lower environmental variation shown by the plants in this setting.
|
DISCUSSION
Effect of stigma position on outcrossing rate
The results shown in Fig. 2 quantify the previously reported effect of stigma position on outcrossing rate in Datura stramonium (Motten and Antonovics, 1992
). Flowers with stigmas below or completely overlapped by the anthers show little or no outcrossing. Successful self-pollination is virtually guaranteed in these flowers because pollen is shed onto an already receptive stigmatic surface before the corolla opens (Motten and Antonovics, 1992
). The prevalence of flowers of this type is responsible for the very low outcrossing rate observed in populations of D. stramonium. However, flowers with stigmas that are completely above the anthers (separation >4 mm) consistently produce at least some outcrossed progeny, and on average the outcrossing rate increases with greater stigma exsertion. Flowers with protruding stigmas thus retain the potential for significant amounts of outcrossing, and continuous variation in stigma position in such flowers is the most direct influence on the extent of outcrossing.
Because our results are based on observations of single flowers under natural pollination conditions, they explicitly support the evidence for a positive association of anther–stigma separation and outcrossing reported in a variety of other diverse taxa studied using controlled pollinations or measurements of population or family means, including Ipomoea purpurea (Epperson and Clegg, 1987
), Eichhornia paniculata, (Glover and Barrett, 1986
; Fenster and Barrett, 1994
), and Turnera ulmifolia (Shore and Barrett, 1990
; Belaoussoff and Shore, 1995
). Our results also support a role for stigma exsertion for species in which outcrossing rate is influenced by other floral traits, such as Nicotiana rustica (Breese, 1959
), Lycopersicon pimpinellifolium (Rick, Holle, and Thorp, 1978
), Mimulus guttatus (Ritland and Ritland, 1989
), and Clarkia tembloriensis (Holtsford and Elstrand, 1992
). In addition, our single-flower data from Datura validate the work of Thomson and Stratton (1985)
on Erythronium grandiflorum and Murcia (1990) on Ipomoea trichocarpa, who inferred a potential for higher outcrossing rate with greater stigma exsertion based on measurements of the proportion of non-self pollen deposited by individual floral visitors.
In comparing our results with those of other studies, several features of Fig. 2 stand out. One is the nonlinearity of Fig. 2 that results from nearly complete selfing in flowers with stigmas <1 mm above the anthers. This pattern emphasizes that stigma position in D. stramonium is closely associated not only with the potential for outcrossing but also with the capacity for autogamy. In this respect Datura contrasts with Gilia achilleifolia, in which stigma exsertion is strongly correlated with autogamous fruit set but not with outcrossing rate (Schoen, 1982
), and with the Mimulus guttatus complex, in which anther–stigma separation is significantly correlated with outcrossing (Ritland and Ritland, 1989
) but only weakly so with autogamy (Carr and Fenster, 1994
; Robertson, Diaz, and Macnair, 1994
). The nonlinearity also suggests that mean values of stigma position may not accurately predict the actual rate of outcrossing, either for individual plants that produce only an occasional flower with a fully exserted stigma or for populations in which most outcrossing occurs as a result of a few individuals bearing flowers with strongly exserted stigmas.
With regard to the role of stigma position in promoting outcrossing, a second distinctive feature of Fig. 2 is the considerable scatter in the data. This variability is due in part to our measuring t for individual flowers rather than whole plants or populations, even though we deliberately attempted to minimize ecological factors likely to affect outcrossing rate, such as plant density and spatial pattern, floral abundance, and geitonogamy (Motten and Antonovics, 1992
; Karron et al., 1995
).
A potentially important factor contributing to the variability among individual flowers may be pollinator behavior, as suggested in other studies by direct observations of floral visitors. Thomson and Stratton (1985)
and Murcia (1990)
noted that nectar-foraging bumble bees differed as much as 2- to 3-fold in the amount of selfed pollen that an individual deposited on exserted stigmas of the same length. Greater variability is probable in our system because the flowers were open-pollinated, and different flowers may have received different numbers of visits or been visited by different pollinators. While the most common visitors were small, pollen-collecting bees that make extensive contact with the anthers and stigma, D. stramonium also attracts hawk moths, which probe for nectar with long thin tongues while hovering above or just inside the flower (Motten and Antonovics, 1992
). In a comparison between bumble bees and hawk moths visiting Ipomoea trichocarpa, whose flowers are morphologically similar to those of D. stramonium, Murcia (1990)
noted that the moths deposited a lower proportion of selfed pollen.
A third distinctive trend in Fig. 2 is the surprisingly modest increase in outcrossing rate with increasing stigma exsertion. Only rarely were outcrossing rates for individual flowers as high as 0.5, and the predicted value for a stigma height 10 mm above the anthers was only 0.23. This gradual increase in outcrossing rate is similar to the pattern reported for a range of populations of Turnera sampled by Belaoussoff and Shore (1995)
, but it is considerably less steep than was observed for a single, variable population of Turnera by Barrett and Shore (1987)
or for Ipomoea purpurea (Ennos, 1981
; Epperson and Clegg, 1987
) and Eichhornia paniculata (Glover and Barrett, 1986
). Moreover, a much stronger effect of stigma exsertion than we observed is also implied by the measurements of stigma deposits of self and non-self pollen made by both Thomson and Stratton (1985)
and Murcia (1990)
. The considerable degree of pollinator-mediated selfing emphasizes the importance of measuring actual outcrossing rates rather than potential for autogamy when using stigma exsertion to infer breeding system characteristics (Ennos, 1981
; Armbruster, 1988
; Dole, 1992
).
It is not clear why outcrossing rates are so low even in flowers with exserted stigmas. One possibility is that this trait acting alone may be insufficient to maintain a predominantly outcrossing mixed-mating system. A congener of D. stramonium with larger flowers and typically much more exserted stigmas, D. meteloides (A. Motten, personal observations) still has an outcrossing rate of only 0.31 (Snow and Dunford, 1961
). In contrast, in other genera with mixed-mating systems and outcrossing rates higher than in Datura, stigma position is often combined with additional mechanisms that reduce selfing, such as dichogamy (Breese, 1959
; Schoen, 1982
; Holtsford and Ellstrand, 1992
), marked variation in floral size (Rick, Fobes, and Holle, 1977
; Carr and Fenster, 1994
; Robertson, Diaz, and Macnair, 1994
), sensitive stigmas (Ritland and Ritland, 1989
), and pollen–style interactions or cryptic incompatibility (Epperson and Clegg, 1987
; Cruzan, 1989
). Because Datura is an annual weed that lives in disturbed habitats often in small populations, selection is likely to favor traits that promote reproductive assurance (Antonovics, 1968
). Variation in stigma position may represent a simple mechanism for retaining the ability to self-fertilize while also maintaining the capacity to produce at least some outcrossed progeny.
Genetics of stigma position in Datura
The results from the diallel cross show that anther–stigma separation in Datura stramonium has a heritable basis. Because this trait is the primary determinant of outcrossing rate, we conclude that the mating system of D. stramonium retains the potential for evolutionary change. Genetic variation in stigma position was characterized largely by additive effects; a fit of our diallel results to the bio model in the quadratic analysis of Cockerham and Weir (1977
; Table 4) showed that maternal effects are not important, and our graphical analysis following Mather and Jinks (1982
; Fig. 4) failed to find evidence of epistatic effects. In both analyses, dominance was expressed as a trend, with a marginally significant effect in one of the three environments studied. These results are consistent with the results reported by Fenster and Ritland (1994)
for the Mimulus guttatus species complex, where traits influencing mating system evolution are dominated by additive effects, but differ from the situation in Turnera ulmifolia, where dominance and epistasis appear to be important (Shore and Barrett, 1990
).
The heritability estimates are also robust with regard to environmental effects. Despite considerable difference in plant size between the two field sites, the heritability estimates were nearly identical, ~30% in each case. These estimates were also higher than those observed for greenhouse-grown plants, contrary to our expectations that greater environmentally induced variation might lead to lower heritability under field conditions. Although the environmental variance was lower in the greenhouse, so was the estimate of heritability because the genetic additive variance was so much lower than in the field (Table 4). One reason for this smaller additive variance may be that D. stramonium fares poorly in small pots, and as suggested by Fig. 3, the added stress in the greenhouse environment reduced the range of expression of genetically based traits. The relative consistency in the heritability estimates for plants of different sizes also indicates that response to selection is unlikely to be impeded by environmental influences affecting plant quality. In this respect, D. stramonium differs from species like Impatiens capensis, in which individuals grown in shade produce more cleistogamous flowers than ones in the sun (Waller, 1980
), or other species in which outcrossing depends on plant size by altering the probability of geitonogamy (de Jong, Waser, and Klinkhamer, 1993
).
Our finding of a heritable basis for anther–stigma separation accords well with reports from other species. Heritability values of ~0.5 have been found for this trait in Mimulus guttatus (Carr and Fenster, 1994
; Robertson, Diaz, and Macnair, 1994
) and Ipomoea purpurea (Ennos, 1981
). In other species where anther–stigma separation does not lead to variation in outcrossing rates, reported heritabilities of style length are astonishingly high: 0.76–0.94 in Saxifraga granulata (Andersson, 1996
), 0.9 in wild radish (Conner and Via, 1993
), and 0.4–0.8 in Ipo-mopsis aggregata (Campbell, 1996
). Heritability of anther height in these species tends to be lower but still significant (Conner and Via, 1993
; Andersson, 1996
; Campbell, 1996
). This difference in heritability is not due simply to anther–stigma separation's being a composite character; style length in Mimulus guttatus, for example, has the same degree of heritability as does anther–stigma separation (Robertson, Diaz, and Macnair, 1994
).
Given that the anther–stigma separation trait in D. stramonium is heritable, and that selfing can be assured by overlapping anthers and stigmas, the question still remains why low levels of outcrossing persist, especially since our results indicate that variation in this trait is maintained in a population despite extensive selfing. One possibility is that anther–stigma separation is maintained by advantages other than increased outcrossing rate, such as promoting pollen export or otherwise reducing interference between male and female function (Webb and Lloyd, 1986
). Another possibility is that within a range of values, a mixed-mating system in this species may be evolutionarily stable. Mixed-mating systems are generally thought to be a transitory condition to fixed states of complete selfing or outcrossing, although a number of circumstances have been suggested that could account for stable mixed mating in particular cases (see Lande and Schemske, 1985
, and review by Uyenoyama, Holsinger, and Waller, 1993
). Our heritability estimates suggest that it is worthwhile to examine factors that may contribute to maintaining variation in stigma position and hence a mixed-mating system. In particular, the heritabilities are useful for predicting response to selection and to determine whether possible advantages to outcrossing can exert sufficient selective pressure to balance the inherent advantage of selfing (Fisher, 1941
) and the benefits for a colonizing annual of reproductive assurance (Antonovics, 1968
). Inbreeding depression may be a selective factor maintaining stigma position in D. stramonium because it varies between low values and the critical value of one half, depending on environmental conditions (A. Motten, unpublished data).
In addition to inbreeding depression, other aspects of genetic architecture may also influence the persistence of mixed-mating in D. stramonium. For example, the trend towards recessivity in increased stigma exsertion could help to promote mixed-mating (Latta and Ritland, 1993
). Carriers of this allele in the homozygous state would tend to outcross at a relatively high rate, generating progeny heterozygous for the allele. These will be likely to self, and about one-fourth of their progeny would again bear the allele in a homozygous state, creating a cycle of alternating outcrossing rates. The importance of this process in natural populations is not clear because dominance variance occurred as a trend in only one of the field sites, but it is more likely to generate stable mixed-mating than if stigma exsertion were promoted by a dominant allele. In that case, lineages would breed true for mating behavior, and consistent selection (if present) would favor one strategy over the other.
Species with mixed-mating systems present special opportunities for understanding and predicting mating system evolution, especially when, as is the case with D. stramonium, the species is already highly selfing. The situation with D. stramonium is further complicated by the indirect and variable correlation between outcrossing rate and the morphological character that controls it, stigma–anther separation. As a result, modeling the evolution of this trait is more difficult than for simple incompatibility loci. Our finding that deviation from complete selfing in D. stramonium is not just environmental noise but founded on additive genetic variance provides the basis for ongoing efforts to measure the strength of inbreeding depression as a selective force maintaining outcrossing. It also provides the context for artificial selection experiments currently in progress to verify predictions based on the heritability estimates and to measure the effect on response to selection of lineage-specific genetic structure.
FOOTNOTES
1 The authors thank Janis Antonovics for guidance in planning this work and helpful comments on the manuscript and the family of W. C. MacFarlands for permission to work with populations of Datura on their dairy farm. Financial support was provided by NSF grant BSR-8407960 to AFM; JLS was supported by NSF grant DEB-9707684 to M. K. Uyenoyama. 
2 Author for correspondence (afmotten{at}acpub.duke.edu)
.
LITERATURE CITED
Andersson, S. 1996 Floral variation in Saxifraga granulata: phenotypic selection, quantitative genetics and predicted response to selection. Heredity 77: 213–223.
Antonovics, J. 1968 Evolution in closely adjacent plant populations. V. Evolution of self-fertility. Heredity 23: 219–238.
———, and J. Schmitt. 1986 Paternal and maternal effects on propagule size in Anthoxanthum odoratum. Oecologia (Berlin): 69: 277–282.
Armbruster, W. S. 1988 Multilevel comparative analysis of the morphology, function, and evolution of Dalechampia blossoms. Ecology 69: 1746–1761. [CrossRef][ISI]
Barrett, S. C. H., and C. G. Eckert. 1990 Variation and evolution of mating systems in seed plants. In S. Kawano [ed.], Biological approaches and evolutionary trends in plants, 229–254. Academic Press, New York, New York, USA.
———, and J. S. Shore. 1987 Variation and evolution of breeding systems in the Turnera ulmifolia L. complex (Turneraceae). Evolution 41: 340–354. [CrossRef][ISI]
Belaoussoff, S., and J. S. Shore. 1995 Floral correlates and fitness consequences of mating-system variation in Turnera ulmifolia. Evolution 49: 545–556.
Blakeslee, A. F., and B. T. Avery. 1917 Adzuki beans and jimson weeds. Journal of Heredity 8: 125–131.
Bodmer, W. F. 1958 Natural crossing between homostyle plants of Primula vulgaris. Heredity 12: 363–370.
Breese, E. L. 1959 Selection for differing degrees of out-breeding in Nicotiana rustica. Annals of Botany (London) 23: 331–344 New Series.
Campbell, D. R. 1996 Evolution of floral traits in a hermaphroditic plant: field measurements of heritabilities and genetic correlations. Evolution 50: 1442–1453. [CrossRef][ISI]
Carr, D. E., and C. B. Fenster. 1994 Levels of genetic variation and covariation for Mimulus (Scrophulariaceae) floral traits. Heredity 72: 606–618. [ISI]
Charlesworth, D., and B. Charlesworth. 1990 Inbreeding depression with heterozygote advantage and its effect on selection for modifiers changing the outcrossing rate. Evolution 44: 870–888. [CrossRef][ISI]
———, M. T. Morgan, and B. Charlesworth. 1990 Inbreeding depression, genetic load, and the evolution of outcrossing rates in a multilocus system with no linkage. Evolution 44: 1469–1489. [CrossRef][ISI]
Clay, K. 1982 Environmental and genetic determinants of cleistogamy in a natural population of the grass Danthonia spicata. Evolution 36: 734–741.
Clegg, M. T. 1980 Measuring plant mating systems. BioScience 30: 814–818. [CrossRef][ISI]
Cockerham, C. C., and B. S. Weir. 1977 Quadratic analyses of reciprocal crosses. Biometrics 33: 187–203. [CrossRef][ISI][Medline]
Conner, J., and S. Via. 1993 Patterns of phenotypic and genetic correlations among morphological and life-history traits in wild radish, Raphanus raphanistrum. Evolution 47: 704–711.
Cruzan, M. B. 1989 Pollen tube attrition in Erythronium grandiflorum. American Journal of Botany 76: 562–570.
Darwin, C. 1878 The effects of cross and self fertilisation in the vegetable kingdom, 2nd ed. J. Murray, London, UK.
deJong, T. J., N. M. Waser, and P. G. L. Klinkhamer. 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]
Ennos, R. A. 1981 Quantitative studies of the mating system in two sympatric species of Ipomoea (Convolvulaceae). Genetica 57: 93–98.
Epperson, B. K., and M. T. Clegg. 1987 First-pollination primacy and pollen selection in the morning glory, Ipomoea purpurea. Heredity 58: 5–14.
Faegri, K., and L. van der Pijl. 1979 The principles of pollination ecology, 3rd ed. Pergamon, Oxford, UK.
Fenster, C. B., and S. C. H. Barrett. 1994 Inheritance of mating-system modifier genes in Eichhornia paniculata (Pontederiaceae). Heredity 72: 433–445. [ISI]
———, and K. Ritland. 1994 Quantitative genetics of mating system divergence in the yellow monkeyflower species complex. Heredity 73: 422–435. [ISI]
Fisher, R. A. 1941 Average excess and average effect of a gene substitution. Annals of Eugenics 11: 53–63. [ISI]
Glover, D. E., and S. C. H. Barrett. 1986 Variation in the mating system of Eichhornia paniculata (Spreng.) Solms. (Pontederiaceae). Evolution 40: 1122–1131. [CrossRef][ISI]
Holsinger, K. E. 1988 Inbreeding depression doesn't matter: the genetic basis of mating-system evolution. Evolution 42: 1235–1244. [CrossRef][ISI]
———. 1991 Inbreeding depression and the evolution of plant mating systems. Trends in Ecology and Evolution 6: 307–308. [CrossRef]
Holtsford, T. P., and N. C. Ellstrand. 1992 Genetic and environmental variation in floral traits affecting outcrossing rate in Clarkia tembloriensis (Onagraceae). Evolution 46: 216–225. [CrossRef][ISI]
Horowitz, A., and J. Harding. 1972 Genetics of Lupinus. V. Intraspecific variability for reproductive traits in Lupinus nanus. Botanical Gazette 133: 155–165.
Jain, S. K. 1976 The evolution of inbreeding in plants. Annual Review of Ecology and Systematics 7: 469–495.
Karron, J. D., R. T. Jackson, N. N. Thumser, and S. L. Schlicht. 1997 Outcrossing rates of individual Mimulus ringens genets are correlated with anther–stigma separation. Heredity 79: 365–370. [CrossRef][ISI]
———, N. N. Thumser, R. Tucker, and A. J. Hessenauer. 1995 The influence of population density on outcrossing rates in Mimulus ringens. Heredity 75: 175–180.
Kerner von Marilaun, A. 1896 The natural history of plants, vol. 2, part 1. Translated and edited by F. W. Oliver. H. Holt, New York, New York, USA.
Lande, R., and D. W. Schemske. 1985 The evolution of self-fertilization and inbreeding depression in plants. I. Genetic models. Evolution 39: 24–40. [CrossRef][ISI]
Latta, R., and K. Ritland. 1993 Models for the evolution of selfing under alternative modes of inheritance. Heredity 71: 1–10. [ISI]
Lynch, M., and B. Walsh. 1998 Genetics and analysis of quantitative traits. Sinauer, Sunderland, Massachusetts, USA.
Lyons, E. E., and J. Antonovics. 1991 Breeding system evolution in Leavenworthia: breeding system variation and reproductive success in natural populations of Leavenworthia crassa (Cruciferae). American Journal of Botany 78: 270–287. [CrossRef][ISI]
Marshall, D. F., and R. J. Abbott. 1984 Polymorphism for outcrossing frequency at the ray floret locus in Senecio vulgaris L. III. Causes. Heredity 53: 145–149. [ISI]
Mather, K., and J. L. Jinks. 1977 Introduction to biometrical genetics. Cornell University Press, Ithaca, New York, USA.
———, and ———. 1982 Biometrical genetics: the study of continuous variation, 3rd ed. Chapman and Hall, London, UK.
Motten, A. F., and J. Antonovics. 1992 Determinants of outcrossing rate in a predominantly self-fertilizing weed, Datura stramonium (Solanaceae). American Journal of Botany 79: 419–427. [CrossRef][ISI]
Murcia, C. 1990 Effect of floral morphology and temperature on pollen receipt and removal in Ipomoea trichocarpa. Ecology 71: 1098–1109.
Rick, C. M., J. F. Fobes, and M. Holle. 1977 Genetic variation in Lycopersicon pimpinellifolium: evidence of evolutionary change in mating systems. Plant Systematics and Evolution 127: 139–170. [CrossRef][ISI]
———, M. Holle, and R. W. Thorp. 1978 Rates of cross-pollination in Lycopersicon pimpinellifolium: impact of genetic variation in floral characters. Plant Systematics and Evolution 129: 31–44. [CrossRef][ISI]
Ritland, C., and K. Ritland. 1989 Variation of sex allocation among eight taxa of the Mimulus guttatus species complex (Scrophulariaceae). American Journal of Botany 76: 1731–1739. [CrossRef][ISI]
Robertson, A. W., A. Diaz, and M. R. Macnair. 1994 The quantitative genetics of floral characters in Mimulus guttatus. Heredity 72: 300–311.
SAS. 1987 SAS/STAT guide for personal computers, version 6. SAS Institute, Cary, North Carolina, USA.
Schoen, D. J. 1982 The breeding system of Gilia achilleifolia: variation in floral characteristics and outcrossing rate. Evolution 36: 352–360. [CrossRef][ISI]
Shore, J. S., and S. C. H. Barrett. 1990 Quantitative genetics of floral characters in homostylous Turnera ulmifolia var. angustifolia Willd. (Turneraceae). Heredity 64: 105–112. [ISI]
Snow, R., and M. P. Dunford. 1961 A study of interchange heterozygosity in a population of Datura meteloides. Genetics 46: 1097–1110.
Thomson, J. D., and D. A. Stratton. 1985 Floral morphology and cross-pollination in Erythronium grandiflorum (Liliaceae). American Journal of Botany 72: 433–437. [CrossRef][ISI]
Uyenoyama, M. K. 1986 Inbreeding and the cost of meiosis: the evolution of selfing in populations practicing biparental inbreeding. Evolution 40: 388–404. [CrossRef][ISI]
———, K. E. Holsinger, and D. M. Waller. 1993 Ecological and genetic factors directing the evolution of self-fertilization. Oxford Surveys in Evolutionary Biology 9: 327–381.
———, and D. M. Waller. 1991 Coevolution of self-fertilization and inbreeding depression. II. Symmetric overdominance in viability. Theoretical Population Biology 40: 47–77. [CrossRef][ISI][Medline]
Vasek, F. C. 1965 Outcrossing in natural populations. II. Clarkia unguiculata. Evolution 19: 152–156.
Waller, D. 1980 Environmental determinants of outcrossing in Impatiens capensis (Balsaminaceae). Evolution 34: 747–761. [CrossRef][ISI]
Webb, C. J. and D. G. Lloyd. 1986 The avoidance of interference between the presentation of pollen and stigmas in angiosperms: II. Herkogamy. New Zealand Journal of Botany 24: 163–178. [ISI]
Wyatt, R. 1984 The evolution of self-pollination in granite outcrop species of Arenaria (Caryophyllaceae). I. Morphological correlates. Evolution 38: 804–816. [CrossRef][ISI]
This article has been cited by other articles:
|
|
 |
|
  A. Lankinen, W. S. Armbruster, and L. Antonsen Delayed stigma receptivity in Collinsia heterophylla (Plantaginaceae): genetic variation and adaptive significance in relation to pollen competition, delayed self-pollination, and mating-system evolution Am. J. Botany, July 1, 2007; 94(7): 1183 - 1192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Waites and J. Agren Stigma receptivity and effects of prior self-pollination on seed set in tristylous Lythrum salicaria (Lythraceae) Am. J. Botany, January 1, 2006; 93(1): 142 - 147. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Nagpal, C. M. Ellis, H. Weber, S. E. Ploense, L. S. Barkawi, T. J. Guilfoyle, G. Hagen, J. M. Alonso, J. D. Cohen, E. E. Farmer, et al. Auxin response factors ARF6 and ARF8 promote jasmonic acid production and flower maturation Development, September 15, 2005; 132(18): 4107 - 4118. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. MEDRANO, C. M. HERRERA, and S. C. H. BARRETT Herkogamy and Mating Patterns in the Self-compatible Daffodil Narcissus longispathus Ann. Bot., June 1, 2005; 95(7): 1105 - 1111. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. R. Herlihy and C. G. Eckert Evolution of self-fertilization at geographical range margins? A comparison of demographic, floral, and mating system variables in central vs. peripheral populations of Aquilegia canadensis (Ranunculaceae) Am. J. Botany, April 1, 2005; 92(4): 744 - 751. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K.-Y. Chen and S. D. Tanksley High-Resolution Mapping and Functional Analysis of se2.1: A Major Stigma Exsertion Quantitative Trait Locus Associated With the Evolution From Allogamy to Autogamy in the Genus Lycopersicon Genetics, November 1, 2004; 168(3): 1563 - 1573. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. K. Schueller Self-pollination in island and mainland populations of the introduced hummingbird-pollinated plant, Nicotiana glauca (Solanaceae) Am. J. Botany, May 1, 2004; 91(5): 672 - 681. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. S. Armbruster, C. P. H. Mulder, B. G. Baldwin, S. Kalisz, B. Wessa, and H. Nute Comparative analysis of late floral development and mating-system evolution in tribe Collinsieae (Scrophulariaceae s.l.) Am. J. Botany, January 1, 2002; 89(1): 37 - 49. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Stone Does anthocyanin affect outcrossing rates in Datura Stramonium (Solanaceae)? Am. J. Botany, March 1, 2000; 87(3): 348 - 354. [Abstract] [Full Text]
|
|
| |||||
