No abstract

No abstract

The number of flowers produced by inflorescences of Yucca whipplei (Agavaceae) consistently exceeds the number of fruits produced by about one order of magnitude. To determine the factors responsible for low fruit set, the relation between pollinator availability, the amount of resources spent on reproduction (as indicated by inflorescence size), and the number of fruits matured was studied during 1978 and 1979 at 18 locations in chaparral, coastal sage scrub, and desert scrub communities of southern California.The following results support the conclusion that pollinators do not usually limit fruit production in Yucca whipplei. Rather, fruit production is limited by the amount of resources available to support developing fruits. (1) Fruit production is positively correlated with inflorescence size both within and between populations. The average size of inflorescence for a population is an excellent predictor of mean fruit production. Furthermore, 54% of the total variance in fruit production of individual plants can be explained by inflorescence size. (2) In contrast, although fruit production within most populations is positively correlated with an index of the number of pollinator visits to an inflorescence, the relative abundance of pollinators for a population is a poor predictor of mean fruit production, and only 9% of the total variance in fruit production can be explained by the visitation index. Furthermore, at four sites studied for two years, there was little change in average inflorescence size or fruit production from 1978 to 1979, despite large differences in relative abundance of pollinators at each of the sites. (3) Based on geographic proximity, and physiographic and vegetational similarities, study sites were grouped into regional clusters. Both inflorescence size and fruit production varied considerably between regions. Of the total variation in fruit production, 27% can be attributed to differences between regions. Most of this variation is the result of regional differences in inflorescence size, which in turn influence fruit production.Why does Yucca whipplei produce such large inflorescences if so few fruits can be supported? Two relevant hypotheses are discussed: (1) the floral display is the result of selection for pollen dissemination at the expense of fruit set; and (2) the floral display is the result of selection for a bet-hedging strategy either to increase the probability of adequate pollination when pollinators are unusually rare, or to allow individuals to support more fruits when resources are unusually abundant.

Flowering and fruiting patterns of Yucca whipplei (Agavaceae), a monocarpic perennial of the southern California chaparral, were studied to determine the role of fruit abortion in the regulation of seed yield. Data on reproductive expenditure, fruit abortion and fruit maturation were obtained for 108 flowering individuals at two sites. The proportion of flowers maturing fruits was less than 10% at both sites. Inflorescences produced an average of 2,045±111 and 1,743±77 flowers at the two sites, initiated an average of 439±48 and 304±21 fruits, and matured an average of 194±16 and 162±10 fruits, respectively. Overall, 51% of the fruits which initiated development were aborted and abscized in an early stage of development.The hypothesis that fruit abortion serves to regulate seed yeild, matching fruit production to the available resources, yields three predictions which are borne out by our findings: (1) the proportion of initiated fruits which abort is an increasing function of the number of fruits initiated, (2) the number of fruits matured is an increasing function of estimates of the amount of resources available to support fruits, and (3) on any inflorescence, fruits initiated early have a lower probability of aborting than fruits initiated late.These results suggest that low fruit set in Yucca whipplei is caused by shortage of resources to support developing fruits, rather than by a shortage of pollinators. Possible explanations of the adaptive significance of regulating maternal investment via fruit abortion are mentioned, but further experimentation is required before any conclusions can be drawn.

The adult behavior of the yucca moth, Tegeticula maculata Riley, is finely tuned to the reproductive biology of its specific host plant, Yucca whipplei Torr. The female moths oviposit in the ovaries of the yucca flowers and actively pollinate the same flowers with pollen which they have collected previously. The selective pressures imposed on the moths by 1) the plant's need for pollen transfer via an insect pollinating agent, 2) its partial self-incompatibility, and 3) its ability to regulate seed set by aborting excess fruits, have molded the pollinator's behavior in such a way that its offspring have the greatest possible chance of surviving through the early larval stages. The evolutionary responses of the pollinator include the following: 1) the female moths consistently fly to a different plant after collecting pollen, thus insuring cross-fertilization of the flowers, 2) they always pollinate after depositing the first egg in a flower, but not necessarily after subsequent ovipositions, and 3) females emerging near the end of the flowering season frequently oviposit in developing seed pods, as opposed to open flowers which are more likely to be aborted by the plants.