Visual floral guides such as colored anthers, lines, dots, and UV-absorption patterns on petals are commonly observed in insect-pollinated angiosperms. Floral guides that are known to enhance foraging efficiency of visitors on flowers thus promote return visits (foraging facilitation hypothesis, which predicts that visitors will discriminate against flowers with inferior floral guides). In this study, we experimentally examined the hypothesis that floral guides also prevent pollen-theft behavior by floral visitors (theft prevention hypothesis), which has rarely been tested. Nectarless flowers of Commelina communis have three types of brightly colored floral organs: large blue petals, rewarding yellow anthers, and nonrewarding yellow anthers. Colored floral organs were removed artificially from plants in two natural populations of C. communis. Removal of the nonrewarding yellow anthers diminished hoverings in front of flowers and tended to reduce the number of total floral visitor landings, supporting the foraging facilitation hypothesis. Additionally, removal of the rewarding yellow anther decreased the frequency of legitimate landings on flowers and the legitimate landing-to-total landing ratio, which is consistent with the theft prevention hypothesis. The nonrewarding anthers and the rewarding yellow anthers were shown to play an important role in increasing visitor landings and orienting floral visitors toward a landing point appropriate for pollination, respectively. We also showed that the absence of yellow anthers decreased both pollen dispatch from brown anthers and receipt by stigmas in C. communis. These findings support both the foraging facilitation hypothesis and the theft prevention hypothesis.

No abstract

Pollinator-mediated selection in the deceptive orchid Pogonia japonica was studied by comparing the amount of variation in and the relative size relationships between the petals, sepals, and gynostemium. The gynostemium length had a significantly lower slope in log-log regression on the indicator of the flower size for its length than did the petals or sepals. This indicates that pollinator-mediated selection may lead to low phenotypic variation of gynostemium length. Pollinia removal by pollinators and pollinia-pollinated stigma were rarely observed in the field. Flowers whose pollinia were removed had a significantly smaller gynostemium than flowers with intact pollinia. There was no significant difference in the sizes of floral organs between pollinated and unpollinated flowers. These results indicate that the size of the gynostemium in Pogonia has evolved to increase male reproductive success.