Search

Procedures are described for partitioning the selfing rate into contributions resulting from the different modes of chasmogamous selfing-geitonogamous, facilitated, prior, competing, and delayed-and for examining several functional aspects of self-fertilization associated with the different modes-reproductive assurance, degree of seed and pollen discounting, and relative abilities of cross- and self-pollen to fertilize ovules. The procedures involve floral manipulations-e.g., exclusion of pollinators, emasculation, and covering of stigmas-and typically require joint estimation of the mating system parameters in manipulated and unmanipulated plants. Two methods of mating system estimation are discussed, the first in which manipulations and selfing rate estimates are made using random parent plants that are resident in the population, and the second using plants with unique genotypes introduced from a separate population. The asymptotic variances of estimators for the different modes of selfing are compared...

Many functional-ecological, morphological, and physiological-factors affect the occurrence of selffertilization. Six modes of self-pollination are distinguished. These differ in whether they utilize specialized flowers, whether they involve the transfer of pollen within or between flowers, whether they are autonomous or mediated by vectors, and their timing relative to opportunities for outcrossing. The various modes of selfing are subject to different structural constraints. Prepotency, the preferential success of cross-pollen in achieving fertilizations when it competes with self-pollen, influences the frequency of selffertilization in some species. The amount of self-fertilization may depend on environmental conditions and the vector species visiting each flower and may vary among the flowers of one plant. To gain information on the prevalence of autonomous self-pollination, 66 species for which the degrees of selfcompatibility and autofertility (seed set in isolation) have been published were surveyed...

Stigmatic exudate-eating chironomid flies (Smittia) and pollen-eating halodid beetles are the most common visitors to flowers ofPseudowintera colorata. Plants are self-incompatible. After selfing pollen tubes penetrate the nucellus; chase pollination experiments show that the incompatibility reaction occurs before fertilization. When pollen is applied to stigmas, the exudate dries rapidly and does not usually reappear. A review of anthophytes suggests that their common ancestor around 200 mya was bisexual and pollinated at least in part by nectar-seeking insects, possibly including flies. The angiosperm stigma appears to have evolved from a pollination drop mechanism via an adaxial stigmatic surface.

Temporal patterns of variability in the longevity of the male and female phases of individual flowers and in the gender expression of plants of a dichogamous New Zealand tree,Pseudowintera colorata (Winteraceae), were documented in field studies. Two measures for the duration of phases in a dichogamous flower are distinguished; the nominal phases based on morphological features of the flower, and the effective phases reflecting the duration of their functions. Flower and phase longevity and phenotypic gender varied considerably throughout the season and among individuals. Temporal variability in phenotypic gender was loosely synchronized among the 12 plants sampled. Three effects of an environmental factor (temperature) were noted. First, increased temperatures shortened the duration of the female phase but had no effect on the duration of the male phase. Second, pollination frequency was positively correlated with temperature. These results indirectly suggest that increased pollination may shorten the duration of the female phase. Third, average population maleness, measured as the proportion of open flowers in the population on a given day which were in the male phase, was positively correlated with temperature. It is postulated that temperature indirectly influences temporal patterns of gender expression in the population through its differential effects on the longevity of the male and female phases in individual flowers. A theoretical model of bet-hedging shows that, if the direction of an environmental effect on the proportions of the sexual phases is irreversible, selection favours asynchronous dichogamy and reduces the temporal variability as much as possible. If the direction of the response is reversible, heterodichogamy is favoured.

In outcrossing hermaphrodite plants, the separate functions of pollen dispersal and pollen receipt may interfere with one another so that fitness as a paternal or maternal parent is compromised (van der Pijl, 1978; Bawa and Opler, 1975; Lloyd and Yates, 1982; Lloyd and Webb, 1986; Webb and Lloyd, 1986; Bertin and Newman, 1993; Harder and Barrett, 1995). This is particularly likely in flowers in which the sex organs are close together and mature at the same time. Interference can potentially take several forms, including the obstruction by pistils of efficient pollen dispatch by pollinators, stamens restricting access by pollinators to stigmas, thus reducing pollen deposition, and the deleterious effects of self-pollination on maternal function due to stigmatic, stylar, or ovular clogging. Although there is some experimental evidence for self-pollen interference (Shore and Barrett, 1984; Barrett and Glover, 1985; Bertin and Sullivan, 1988; Palmer et al., 1989; Waser and Price, 1991; Scribailo and Barrett, 1994), the other two forms of pollen-pistil interference have seldom been investigated (see, however, Barrett and Glover, 1985; Kohn and Barrett, 1992a).

Models for the evolution of a mixture of cleistogamous (closed, autogamous) flowers and chasmogamous (open) flowers are described. The ‘basic’ model takes into account features associated with cleistogamous self-pollination, including the greater economy and certainty of cleistogamous fertilization and the inability of cleistogamous flowers to contribute pollen to the outcrossed pollen pool. Complete cleistogamous selfing is favoured when allocation to maternal function, fertilization rate, and viability of progeny are sufficiently greater for the cleistogamous component, and when the resources spent on ancillary structures in cleistogamous flowers, cleistogamous seed costs, and inbreeding depression are low. The result is discussed with respect to the cost of sex argument and relevant ecological data. Suggestions for the apparent rarity of cleistogamy are presented. The ‘complex habitat’ model extends the basic model to situations in which the success of reproduction by cleistogamy or chasmogamy varies according to the environment of the parent. In this situation, reproduction by both cleistogamy and chasomogamy is sometimes selected. A ‘near and far dispersal’ model addresses the question of the evolution of dual modes of dispersal, which occur in some cleistogamous and non-cleistogamous plants. A dual mode of dispersal may evolve if a narrowly dispersed seed type is more successful in establishing at the sites located within its dispersal range compared with a second, more widely dispersed seed type which experiences less sib competition. The prediction is discussed with respect to data from amphicarpic plants.

A model is presented which expresses the seed production of dioecious and gynodioecious animal-pollinated Angiosperms in terms of the relative seed-fecundity of the sexes, the number of pollinator visits to each flower and the sex ratio. The model predicts that the maximum seed set occurs when females predominate, providing the pollinators visit each flower more than once and the seed set of males is not high. There is strong evidence that a marked preponderance of females in species of four genera is due to more frequent fertilisation by female-determining pollen nuclei than by male-determining nuclei. Two hypotheses have been proposed to explain this differential fertilisation. Several objections are raised to the hypothesis of Lewis (1942), Mulcahy (1967) and Kaplan (1972) that female-predominant sex ratios have been selected because they maximise the total seed production of populations. It is considered that the hypothesis of Smith (1963) that the differential fertilisation is a consequence of the genetic differentiation of sex chromosomes offers a more likely explanation of female-predominant sex ratios.

The gender of plants of populations of two New Zealand montane shrub species during two flowering seasons is described: Discaria toumatou has hermaphrodite flowers and Leptospermum scoparium is andromonoecious. Individuals in such species that are monomorphic for their sexual type are often considered to behave equally as pollen and ovule parents in sexual reproduction. Counts of the numbers of pollen- and fruit-bearing flowers on a sample of plants demonstrate, however, that in one season individuals within each population span the range from exclusively pollen- to predominantly ovule-contributors in functional gender (the proportions of genes that are likely to be transmitted through pollen and ovules, based on parental investments). The distribution of gender within a population changes somewhat between years but is basically consistent. Gender in successive years is not correlated among individuals within a population, indicating that the environment influences femaleness. Femaleness is not c...

Gender (maleness or femaleness) is a quantitative phenomenon in plants. For accurate descriptions of gender, numerical estimates of the relative capabilities of plants as pollen and ovule parents have advantages over the verbal morphological descriptions of gender that are traditionally used. Full descriptions require the distribution of gender among subsets of a sexual class (e.g., among individual plants) to be indicated as well as the average condition. Estimates of gender can be based on the pollen and ovule or seed production of each subset independently (phenotypic gender), or else the pollen and ovule or seed production of the whole population can be taken into account to assess the functional gender of any subset. The functional gender of a plant estimates the proportions of its genes which are transmitted through pollen (its maleness) or through ovules (its femaleness). Prospective estimates of functional gender are based on the relative maternal and paternal investments before anthesis ...

Dichogamy is the separation of the presentation of pollen and stigmas in time within a plant. It is a common but neglected feature of outcrossing angiosperms. Dichogamy has been almost universally interpreted as an outcrossing mechanism, but many dichogamous species are also self-incompatible (and sometimes also herkogamous and/or with unisexual flowers). In outcrossing species, there is almost invariably a clash between selection to place pollen and stigmas in similar positions for effective pollination and selection to keep the androecia and gynoecia apart to avoid interference between them. We suggest that the separation of pollen and stigmas acts in general to reduce this self-interference and it often also reduces self-fertilisation. Mechanisms preventing self-fertilisation primarily increase maternal fitness, whereas mechanisms avoiding self-interference primarily promote paternal fitness. Five independent ways of subdividing dichogamy are recognised: protandry or protogyny; intrafloral or ...

The effective gender of a plant is defined as the proportion of the plant's genes that are transmitted through pollen (its maleness) or through ovules (its femaleness). Formulae are derived that enable the average effective gender of the morphs of a heterostylous population and the gender of individual plants to be estimated. Estimates require a knowledge of the morph ratio and the seed set that results in each morph from self-fertilization and from legitimate and illegitimate cross-fertilizations. If no illegitimate fertilizations occur, the average gender of the morphs can be estimated from their seed production alone; in this situation the average femaleness of one morph is exactly equal to the average maleness of the other morph. The average gender of long- and short-styled morphs is calculated for populations of fiveCordia species (Boraginaceae) from published data ofOpler & al. (1975). In two species, both morphs transmit their genes equally through pollen and ovules. In the other species, the long-styled morph acts predominantly (or in one species exclusively) as an ovule parent and the short-styled morph succeeds predominantly (or exclusively) as a pollen parent.

Sexually dimorphic seed plants have partial or complete separation of ovule and pollen functions into two separate sexes, here uniformly called male and female. Secondary sex characters (differences between the sexes in structures other than the androecia and gynoecia) of such populations are reviewed. In a number of perennial species, males exceed females in vigor, growth rate or vegetative reproduction. In several monocarpic or short-lived polycarpic species, including hemp, spinach and species ofSilene, females are larger than males. In asparagus, males exceed females in total growth, but individual shoots of females are larger. In some long-lived species, the greater survival rate of males than of females contributes to a predominance of males, but in species ofSilene andRumex acetosa males have a higher mortality rate than females. Males and females sometimes have different microdistributions, and inMercurialis perennis andRumex acetosella this appears to be associated with different environmental optima of the sexes. Differences between males and females in habit, leaf characters and minor morphological features are described. In a considerable range of species, male inflorescences emerge or bolt earlier and males begin flowering sooner than females. Several authors have suggested that females spend proportionately more of their resources on sexual reproduction (have a higher reproductive effort) than do males. The only available data, those forRumex acetosella, support this. There are more reports of males flowering earlier in life and more frequently, having more numerous inflorescences and more flowers per inflorescence, than of females exceeding males in these features. Differences between individual male and female flowers are reviewed. Secondary sex characters may be incidental consequences of the developmental or genetic basis of sex determination and not adaptive themselves, or they may be selected as optimal strategies for the sexes, or both kinds of explanation may simultaneously be applied. We propose that many of the observed differences between the sexes are of direct selective value in relation to the distinct roles of males and females in sexual reproduction. In particular, the benefits to males of increasing pollen production and to females of increasing ovule production differ in such a way that the optimal reproductive effort will often be less for males than for females; this appears to explain a number of secondary sex characters.

Summary Four geographic races of Leavenworthia crassa and one race of L. alabamica are described as self-incompatible, since the average frequencies of fruit set and of pollen tube growth into the style are significantly lower after self-pollination than after cross-pollination. The frequency of successful self-pollination in individual plants of these races varies from zero to 100%, and is positively correlated with the average number of seeds in self-pollinated siliquas. Eleven races of L. crassa and three races of L. alabaimca are adapted to self-pollination and are described as self-compatible, although self-pollinations succeed 2% less frequently than cross-pollinations. Self-pollinated siliquas of these races contain an average of 12% fewer seeds than cross-pollinated siliquas; this is probably an expression of cryptic self-incompatibility and not the result of the exposure of seed lethals on inbreeding. There is, therefore, no absolute distinction between self-incompatibility and self-compatibility in Leavemcorthia. The growth of pollen tubes is compared in successful and unsuccessful pollinations of the same type, in self- and cross-pollinations of the same individuals, and in self-incompatible and self-compatible races. Pollen grains react with the stigma even in unsuccessful pollinations and some pollen grains are inhibited in successful pollinations. In both self-incompatible and self-compatible races, the average number of emptied pollen grains on the stigmata of successful pollinations (an estimate of the number of pollen tubes in the styles) is similar after self- and cross-pollinations. But the average number of emptied pollen grains is lower in self-incompatible races than in self-compatible races. The results are interpreted in terms of the response of pollen grains to an essential growth substance supplied by the stigma.

Six plants of a self-incompatible population ofLeavenworthia crassa were grown from seed collected in nature and cross-pollinated in all combinations. The incompatibility relationships between sibs were determined in eleven of the F1 families. A one-locus sporophytic incompatibility system was established. None of the parents was homozygous at the S locus. At least five, and possibly all six, of the parents did not share an S allele. Only one pair of alleles was shown to have different interactions in the pollen and stigmata. The identity and expression of the S alleles were determined in six families. Eight pairs of alleles were independently expressed in both the pollen grains and the stigmata. Sixteen pairs of alleles showed dominance of one allele over the other in the pollen grains or the stigmata or both.

Summary Five classes of sex expression are distinguished in Cotula on the basis of the degree of separation of male and female gametes on different flowers and plants. These classes are divided into sexually monomorphic and dimorphic systems. The order–monocliny, gynomonoecy, monoecy, subdioecy, dioecy–places the five classes in a linear sequence of increasing dependence for fertilization on the transfer of pollen between flowers and plants. Four of the five major taxonomic groups in Cotula and five species have more than one class of sex expression. The four monoecious and gynomonoecious species tested are self-compatible. The morphology of hermaphrodite, male and female florets are compared. Seven additional characters are associated with sex expression in New Zealand species. In general, the extent of protogyny, brightness of flowers, degree of retraction of stigmas and anthers, floret length, head diameter, peduncle length and the number of pollen grains increase from gynomonoecious through monoecious to dimorphic (dioecious or subdioecious) populations. But within each kind of sex expression there is considerable variation in these ancillary characters. A MAS (measurements associated with sex expression) index is calculated from the four quantitative characters for each New Zealand species (or part-species where there are two kinds of sex expression in a species). The MAS index reflects the expenditure of energy on adaptations to cross-pollination and provides an approximate measure of the relative frequency of outbreeding. The MAS indices of the species and part-species indicate a continuous array in the expression of adaptations to cross-pollination and the frequency of outbreeding. The different kinds of sex expression occupy characteristic but overlapping segments of this array and ought to be considered as convenient classes which only specify the breeding system approximately.

Gynodioecy is defined as the class of sexual dimorphism in which females are constant (functionally unisexual) and males are inconstant (bisexual). Quantitative measures describe the sexual performance of males in gynodioecious populations. The relative (male/female) ovule contribution compares the ovule contributions of the sexes to the next generation. Male constancy and inconstancy, the proportions of male genes transmitted via pollen and ovules, are direct measure of maleness and femaleness. Equations permitting the values of the sex parameters to be estimated are derived. The values obtained for twelve species show that gynodioecy is not a single breeding system, but merges into hermaphroditism and dioecy at opposite extremes. The distribution of inconstancies among individual males is described for six species. In most species, seed counts overestimate the ovule contributions of males.

In angiosperms, gender is a quantitative phenomenon which can be measured on a continuous scale between the strictly male and strictly female extremes. Measurements of gender are of two kinds. Estimates of the "phenotypic gender" of a plant are based on the reproductive characteristics of that plant alone-counts of the percentage of seminiferous or polliniferous flowers for example. The "functional gender" of a plant takes into account the production of gametes and seeds by other individuals in the population and estimates the proportions of a plant's genes that are transmitted through its ovules, its femaleness (Gi), or through pollen, its maleness (Ai = 1 Gi) (Lloyd, 1980). Both phenotypic gender and functional gender can be estimated either from counts of pollen and ovule production (gamete estimates) or from counts of pollen and seed production (seed estimates). Such estimates of functional gender measure the prospective success of plants as pollen and seed parents based on the patterns of parental investment; hence they record parental strategies rather than the eventual success of a plant in leaving adult descendants. The techniques for describing the gender of plants which are proposed elsewhere (Lloyd, 1980) are applied here to plants of four natural (self-maintained) populations, one from each of four angiosperm species. The four populations illustrate the use of the techniques in providing accurate descriptions of plant sex and in documenting evolutionary changes in sex conditions. The four populations show the origin (populations that are monomorphic in gender) and intermediate stages (dimorphic subdioecious populations) in two important pathways leading to dioecy in angiosperms. Two species of Umbelliferae illustrate the "gynodioecy pathway" to dioecy from species with hermaphrodite flowers. Two Cotula (Compositae) populations illustrate the "paradioecy pathway" from monoecy to dioecy. The plants of all four species are perennial, and the gender estimates reported here represent the sexual reproduction of one season only. The estimates are based on counts of the numbers of flowers bearing pollen and of the number of flowers with either ovules or seeds. In the absence of information on variation in the number of pollen grains in polliniferous flowers and in the number and average fitness of seeds in fruit (one or two in the Umbelliferae but always one in Cotula), it is assumed that all polliniferous flowers of a population, and separately all ovuleor seed-bearing flowers, are equally likely to contribute to the next generation. Details of the reproductive behavior of the species and of the methods used are described elsewhere (Lloyd, 1972a, b, unpubl.; C. J. Webb, 1979). The species are considered in order of increasing complexity of the patterns of gender distribution.

Many genera of Angiosperms contain closely related self-incompatible (si) and self-compatible (sc) taxa. In attempting to cross such taxa, it has often been found that the pollen of the self-incompatible plants grows successfully through the styles of the self-compatible plants, but the pollen tubes of the self-compatible plants are inhibited in the styles of the self-incompatible plants, in a manner similar to the self-incompatibility inhibition (Harrison and Darby, 1955; Lewis and Crowe, 1958). Several hypotheses have been proposed to account for the occurrence of "unilateral incompatibility" in some si x sc crosses and its absence from others. Lewis and Crowe (1958) pointed out that the exceptional self-compatible species whose pollen is not inhibited in self-incompatible styles are only recently derived from selfincompatible ancestors. They proposed that self-compatibility and unilateral incompatibility are due to changes in pollenstyle interactions resulting from mutations at the S locus. Anderson and de Winton (1931) and Pandey (1964) observed that pollen tubes of the self-compatible species Nicotiana langsdorfii were inhibited in styles of the self-incompatible species N. alata containing certain S alleles, but were not inhibited in styles containing other alleles. Pandey postulated that the two types of S alleles differ by a mutation at a sub-locus of the S locus. In contrast, Martin (1964) Grun and Aubertin (1966) and others, have found that the inheritance of unilateral incompatibility in their material is complex and conditioned by a

Phenotypic models of selection are used to predict the ESS (evolutionarily stable strategy) allocation to pollen, seeds and pollination mechanisms (corolla, etc.) as a function of the frequency of self-fertilization in plants with different modes of self-fertilization. The results show that: (1) As autogamous self-fertilization increases, the ESS allocation to seeds increases and the allocations to pollen and pollination mechanisms simultaneously decrease. (2) Whether or not there is any allocation to pollination mechanisms with 100% selfing depends on whether self-pollination requires the activity of pollinators or occurs autonomously. (3) The best measure of the importance of self-fertilization is the proportion of adults derived from selfing. (4) If any autogamy occurs, the allocations to pollen and seeds are not equivalent to those to male and female functions. (5) Self-fertilization is not a special case of local mate competition. (6) Geitonogamous self-fertilization requires the same structures as outcrossing and has little effect on allocations. The models can explain the allocation patterns and high fruit set commonly associated with self-fertilization and they suggest further types of data that should be collected. Key-words: Allocations, self-fertilization, sexual strategies, plants, pollen, seeds, pollination