12. Jahrestagung der Deutschen Gesellschaft für Tropenökologie - gtö. Ulm, 17 - 19.02.1999: Poster P-1.6


Sex-specific floral traits in Ecuadorian fly-pollinated dioecious and trioecious species of Oreopanax (Araliaceae)

Andreas Gumbert
Botan. Museum und Botan. Garten, Berlin
present address: Inst. für Neurobiologie, Freie Universität Berlin

Susanne S. Renner
Dept. of Biology, Univ. of Missouri - St. Louis, and The Missouri Botanical Garden, St. Louis

In unisexual plants, females are thought to invest more energy into their offspring than males because they produce seeds. Male plants that potentially pollinate many female flowers compete for access to females. These functional differences may require sex-specific morphological and physiological floral adaptations. We studied three fly-pollinated woody species of Oreopanax (O. confusus, O. ecuadorensis, O. seemannianus) in ecologically different environments in Ecuador in order to analyse sex-specific floral adaptations.

We found that two species had functionally strictly unisexual flowers, with plants being dioecious, while the third, O. seemannianus, had some hermaphroditic flowers in addition to unisexual ones. Because of the specific distribution of flowers and inflorescences on individuals, O. seemannianus appears to belong to a small group of angiosperm species that are trioecious, with some plants functioning as males, others as females, and still others as hermaphrodites. Male individuals did not differ from females in size but had significantly more and larger inflorescences. Flowers were generally arranged in heads sitting on branched inflorescences (unbranched in O. seemannianus). Differences in inflorescence architecture resulted in a significantly higher flower production in male plants: male individuals had more flowers per head, more heads per lateral axis within an inflorescence, and more lateral axes per inflorescence. The floral sex ratio on a population level among the three species varied from 1:9.1 to 1:58.8, and the pollen/ovule ratio reached 78 000-226 000, which is a high value for obligately outcrossed entomogamous plants. Female inflorescences were less showy, due to the smaller number of flowers and due to smaller petals in female flowers that fell off during anthesis in O. confusus and O. ecuadorensis. Female flowers secreted significantly less nectar than males, e.g. in O. confusus as little as 8.1% of the male flowers. Anthesis of male flowers lasted only 1-4 days, and in O. confusus and O. ecuadorensis the anthers fell off after only 1-2 days, during which time all pollen was removed by insects at least in the latter species. Female flower longevity was 7-9 times higher than that of males.

Longer-lived male flowers would presumably only be selectively advantageous if the period of pollen presentation were extended. This may be constrained in Araliaceae, however, because stamen number is constant in the family and because the anthers open completely along a vertical slit so that all pollen is presented simultaneously to the pollinators. Females, which may be more resource-limited, invest less into signaling structures (petals) and rewards (nectar) but may be able to compensate low visitation rates by extending the period of receptivity.