EVIDENCE FOR SELECTIVE FRUIT PRODUCTION IN ASCLEPIAS.

Abstract

Fruit abortion has been hypothesized as a means by which plants can optimally distribute limited resources among the more economically packaged, higher quality seeds (Janzen, 1977a, 1977b; Charnov, 1979; Lloyd, 1980; Stephenson, 1981; Bertin, 1982a; Lee and Bazzaz, 1982a, 1982b). For some species, there is evidence for selective production of fruits based on paternity (Bertin, 1982a) and/or number of seeds per fruit (Heinicke, 1917; Stephenson, 1981; Bertin, 1982a, 1982b; Lee and Bazzaz, 1982a, 1982b). This study examines the potential for selective fruit production in Asclepias speciosa, a member of a genus with relatively high rates of both flower abscission (93.0-99.3%) and fruit abortion (30-95% of initiated pods) (Stephenson, 1981). Members of the genus Asclepias have several ideal characteristics for a study of selective fruit production. The umbel provides a collection of flowers sharing a common period of anthesis (flowers open within + 2 days of each other and wilt simultaneously). When identically treated, each flower within an umbel has an equal chance of forming a pod (Bookman, 1982) allowing for a direct comparison of pod production between different pollen sources or donors. There is a delay in anthesis of flowers on each successive umbel so that competition for maternal resources among pods within an umbel is much greater than that between pods of different umbels (Wyatt, 1980). This allows for replication on different umbels of the comparisons of pod production between pollen donors. Asclepias has pollen grains bound and transported in packages as pollinia. I have found that usually in A. speciosa,