FECUNDITY IN DISTYLOUS AND SELF-INCOMPATIBLE HOMOSTYLOUS PLANTS OF MITCHELLA REPENS (RUBIACEAE)

Breeding system and floral morphometry of distylous Psychotria L. species in the Atlantic rain forest, SE Brazil

General patterns of floral morphology and incompatibility mechanisms have been described for many distylous plants. The absence of these patterns in typically distylous groups, as observed especially in tropical environments, is interpreted as atypical distyly, or as a new reproductive strategy derived from it. Data are presented here on the morphological and compatibility relations between floral morphs of four Psychotria dimorphic species in the Atlantic rain forest in SE Brazil: Psychotria jasminoides, P. birotula, P. mapourioides, and P. pubigera. When significant differences were found, floral parts were larger in thrum flowers. Results of controlled crosses showed that most incompatible pollen tubes were arrested in the stigma, and only in a low proportion in the upper parts of the style. We conclude that, at the study site, the majority and most important morphological and mating features of typical distyly seem to be conserved in P. jasminoides and P. mapourioides, which presented reciprocal herkogamy, self and intramorph incompatibility, and a balanced morph ratio in the population. Typical distyly in P. birotula is supported by floral morphology, pollen tube data and morph ratio and, in P. pubigera, only by floral morphology and pollen tube data.

Frequency-dependent variation in reproductive success in Narcissus: implications for the maintenance of stigma-height dimorphism

Negative frequency-dependent selection is a major selective force maintaining sexual polymorphisms. However, empirical demonstrations of frequency-dependent reproductive success are rare, particularly in plants. We investigate this problem by manipulating the frequencies of style morphs in a natural population of Narcissus assoanus, a self-incompatible herb with style-length dimorphism and intra-morph compatibility. We predicted that the reproductive success of morphs would vary negatively with their frequency because of the effects of morph-specific differences in sex-organ position on patterns of pollen transfer. This prediction was generally supported. The fruit and seed set of the two morphs did not differ significantly in plots with 1 : 1 morph ratios. However, short-styled plants produced significantly fewer seeds than long-styled plants in monomorphic plots, and significantly more seeds than long-styled plants in plots with 'long-biased' morph ratios. These patterns indicate that in the absence of physiological barriers to intra-morph mating, negative frequency-dependent selection contributes to the maintenance of stylar polymorphism through inter-morph pollen transfer. Our experimental results also provide insights into the mechanisms governing the biased style-morph ratios in populations of Narcissus species.

Fitness constraints on flower number, seed number and seed size in the dimorphic species Primula farinosa L. and Armeria maritima (Miller) Willd

Using two dimorphic species with diallelic incompatibility, Primula farinosa L. and Armeria maritima (Miller) Willd., we tested the hypothesis that an association of seedling performance with seed size, and an inverse association of seed size with seed number, might, contrary to most fitness models, select for plants which set relatively few seeds. In P. farinosa we discovered positive relationships between seed number per capsule and capsule number per inflorescence. As capsules are formed in most flowers, scapes with few flowers set the fewest seeds per capsule. This suggests that conspicuousness to pollinators and/or plant resource play an important role in reproductive success in this species. Seed size and seedling performance were also positively associated. For P. farinosa, which has multi-seeded fruits, we discovered an inverse association between seed size and seed number for capsules containing more than 65 seeds. We suggest that stabilizing selection may occur for flower number per inflorescence in this species, as inflorescences with 8-11 capsules tend to set many seeds which may be inefficiently small. This tends to confirm our hypothesis. However, for A. maritima there is a positive association between seed size and seed number per inflorescence. For such plants with single-seeded fruits there should be directional selection for large flower number pet-inflorescence, a trend which may have led to the evolution of the capitulum.