GENDER VARIATION IN HERMAPHRODITE PLANTS - EVIDENCE FROM 5 SPECIES OFALPINE RANUNCULUS

Ovule and seed estimates of gender were calculated for a total of 230 plants in populations of the Australian alpine perennial herbs Ranuncu lus muelleri, R. dissectifolius, R. graniticola, R. millanii and R. ni phophilus. The distribution of gender using the two estimates differed . The distribution of gender based on the seed estimates is likely to more accurately reflect the functional gender of the plants, as the am ount of seed produced by a plant is the result of all the factors that affected ovule production, as well as any environmental effects that are particular to seed set. Based on the ovule estimates of gender, pl ants were phenotypically as well as morphologically hermaphrodite, wit h ovule estimates of gender centred around 0.5 for all five species. H owever, seed estimates of gender indicate that three species (Ranuncul us muelleri, R. dissectifolius and the clonal R. millanii) do not cons ist of populations of entirely functional hermaphrodites. Rather the d istribution of gender in populations of these three species was irregu larly bimodal with plants exhibiting a range of genders from non-seedi ng to functionally hermaphrodite plants. Ranunculus niphophilus popula tions had an amodal distribution of gender using seed based estimates with plants exhibiting a range of genders from nearly non-seeding to v ery female. Only in R. graniticola did seed based estimates of gender indicate that plants were functionally as well as phenotypically herma phrodite. Plant size was correlated with several aspects of gender. In the four non-clonal species (R. muelleri, R. dissectifolius, R. grani ticola and R. niphophilus), vegetative plants were smaller than flower ing plants. In R. muelleri and R. dissectifolius, smaller plants in th e population were unisexual-male. In R. muelleri, size was correlated with gender in bisexual plants, with larger plants being more female. In the clonal R. millanii, non-flowering clumps were less dense than f lowering clumps. In this species, the main factor affecting gender was nor size/density, but rain, with flooding of the depressions in which it occurs in the second season resulting in large numbers of function ally unisexual-male clumps. Gender was not closely correlated between seasons, although in some populations individuals exhibited some consi stency.