Variance in Reproductive Characteristics in Four Morning Glory Species (\u3cem\u3eIpomoea\u3c/em\u3e, Convolvulaceae)

A major goal of evolutionary ecology is to understand the factors that determine the amount of phenotypic variation within natural populations. I have adopted a comparative approach and examined patterns of flower size variation in four species of the genus Ipomoea (I. trichocarpa, I. hederacea, I. coccinea and I. quantoclit) The study species are herbaceous, annual plants (Convolvulaceae) that are commonly found in fields, roadsides and waste places from Florida to east Texas and southeast North Carolina. The unifying theme of my research is to examine the causes and consequences of flower size variation in the study species in Bulloch County, Georgia. The objectives were: 1) to quantify the amount ofvariation in flower size, fruit mass and number of seeds per fruit within each species and 2) to determine if patterns of biomass allocation to floral parts are related to changes in flower size and 3) to determine if flower size influences fruit set and 3) to determine the levels of automatic self-fertilization (autogamy) in natural populations and in the greenhouse. Comparing corolla diameters from two populations sampled in Bulloch County, Georgia during the 1996 and 1998 flowering seasons determined flower size variation. Flower size variation was significantly variable both spatially and temporally within and among natural populations. There was more variation in flower size in I. trichocarpa and I. hederacea, which are more generalists in their pollinator fauna, on the other hand I. coccinea and /. quamoclit, have a more specialized pollinator fauna and had significantly more variation in flower size. Fruit mass was also significantly different within and among the four species. Seed number, however, was relatively invariant within species, but significantly different among the four species. The proportion of flower biomass allocated to attraction (corolla) and reproductive organs (i.e. reproductive allocation) varied significantly within and among the four morning glory species. Petal mass correlated positively with corolla diameter. The percent of reproductive allocation was negatively correlated with petal mass. Reproductive allocation, for each species, represented a significant portion of the total biomass. To determine if flower size influences pollinator attraction, I used three treatments of varying corolla reduction in each species. I found no significant treatment effect on fruit set. In fact fruit set was as high in all species whether or not they had been subjected to treatment. The results may be due to the fact that Ipomoea are able to set fruit without pollinators (i. e. autogamy). Pollinator exclusion experiments conducted in the field and greenhouse revealed that fruit set ranged from 32-98 percent in the absence of pollinators. The results of this study indicate that Ipomoea are autogamous despite producing large and showy floral displays. Flower size variabihty is likely a result ofgenetic and environmental factors