Most flowering plants function as both male and females, through the production of either pollen or seeds. Theory suggests that female reproductive success is usually limited by resources available for seed production. The male function in plants is limited by the ability of the plant to disperse pollen. This leads to the prediction that the attractiveness of flowers to pollinators has relatively limited effect on the production of seed, but a large effect on the dispersal of pollen, and hence the fitness of the plant as a male. Despite this predicted importance of male fitness to the evolution of floral attraction, the distribution of paternal reproductive success in natural populations has been exceedingly difficult to measure. DNA "fingerprinting" can be used to determine paternity unambiguously in plant populations, and therefore to estimate paternal reproductive success. Methods will be developed to measure paternal and maternal fitness in the prairie plant Heliopsis helianthoides in a collaborative study that has both intensive field and laboratory components. The three objectives are: 1) to document variation among H. helianthoides for hypervariable loci; 2) to determine the mode of inheritance for these loci; 3) to use seedlings of known percentage to determine if the hypervariable loci can be used to assign paternity in a wild population of H. helianthoides. This work will contribute the basic data necessary to apply the technique of DNA "fingerprinting" to the evaluation of current hypotheses on the evolution of floral structures.
