SUPERMAN, a regulator of floral homeotic genes in Arabidopsis.

We describe a locus, SUPERMAN, mutations in which result in extra stamens developing at the expense of the central carpels in the Arabidopsis thaliana flower. The development of superman flowers, from initial primordium to mature flower, is described by scanning electron microscopy. The development of doubly and triply mutant strains, constructed with superman alleles and previously identified homeotic mutations that cause alterations in floral organ identity, is also described. Essentially additive phenotypes are observed in superman agamous and superman apetala2 double mutants. The epistatic relationships observed between either apetala3 or pistillata and superman alleles suggest that the SUPERMAN gene product could be a regulator of these floral homeotic genes. To test this, the expression patterns of AGAMOUS and APETALA3 were examined in superman flowers. In wild-type flowers, APETALA3 expression is restricted to the second and third whorls where it is required for the specification of petals and stamens. In contrast, in superman flowers, APETALA3 expression expands to include most of the cells that would normally constitute the fourth whorl. This ectopic APETALA3 expression is proposed to be one of the causes of the development of the extra stamens in superman flowers. The spatial pattern of AGAMOUS expression remains unaltered in superman flowers as compared to wild-type flowers. Taken together these data indicate that one of the functions of the wild-type SUPERMAN gene product is to negatively regulate APETALA3 in the fourth whorl of the flower. In addition, superman mutants exhibit a loss of determinacy of the floral meristem, an effect that appears to be mediated by the APETALA3 and PISTILLATA gene products.     

Evaluating pollen flow indicators for an insect-pollinated plant species

For insect-pollinated plant species, reproductive success and genetic exchange via the transfer of pollen between flowers depends (i.a.) on the efficiency, abundance and behaviour of floral visitors. These in turn are expected to respond to plant population size and flower density. High floral densities for example usually attract large numbers of pollinators that visit more flowers per plant or patch, which increases pollen deposition at short distances. Thus, population characteristics might serve as indicators for pollen dispersal patterns and help to identify suitable habitat size and quality for conservation measures. To test this hypothesis, we observed floral visitors of a generalist, entomophilous species, Comarum palustre, and compared their abundance and visitation rates in populations of different sizes and flower densities. At the same time, we mimicked pollen flow using fluorescent dye. In the large and dense populations, pollinator abundance and visitation rates were high and dye was dispersed to the edges of the populations (up to 200 m). In the medium-sized population with high flower density, insect abundance and visitation rates were unexpectedly low and dye dispersal declined very quickly. On the contrary, in the smallest population with scattered flowers, especially bumble bee abundance was similar to the large populations and dye dispersal mirrored this high bumble bee activity. Thus, our results indicate that in smaller habitat fragments, the mere size of a population might be insufficient to suggest pollen flow for a plant species. Instead, the abundance of its major pollinators should be considered.