The Cost of Over-initiating Fruit

Abstract

Catalpa speciosa regularly aborts a sizable portion of its immature fruit crop. In order to determine the effects of wasting resources on aborted fruit, inflorescences of C. speciosa that initiated from 2-6 fruit were formed by hand-pollinating the appropriate number of flowers on each inflorescence. At the end of the growing season, I compared seed number, fruit weight, seed weight and percent seed germination from the mature fruit on inflorescences that differed with respect to the number of aborted fruits. Two-way analyses of variance reveal that neither the number of fruits that mature per inflorescence of Catalpa speciosa nor the number of fruit that abort per inflorescence affect the number of seeds per mature fruit. However, the number of mature fruit and (independently) the number of abortions per inflorescence significantly affect the weight of the mature fruit and their seeds and the percentage of the seeds that germinate. From these data I conclude that aborted fruit contain some of the resource(s) that limits fruit production and that the over-initiation of even a few juvenile fruit can reduce the number of germinable seeds produced during the current growing season. INTRODUCTION Many plant species regularly initiate more fruits than can be developed to maturity with the available resources (Lloyd, 1980; Stephenson, 1981). In these species, the juvenile fruits abort until the resource needs of the remaining fruits more closely match the parent plant's ability to provide the necessary resources for seed and fruit growth (Lloyd, 1980; Stephenson, 1981). Because resources rather than pollination usually limit seed and fruit production on these species and because resources are wasted on fruits that later abort, the consistent production of surplus juvenile fruits appears to be resource-inefficient. Presumably, if fewer fruits were initiated, then more seeds could develop to maturity. At least four hypotheses have been proposed to explain the adaptive significance of producing surplus juvenile fruits. The production of surplus juvenile fruits: (1) allows plants to take advantage of the occasional good years when resources are plentiful (Willson and Price, 1977; cf., Lack, 1954; Ricklefs, 1965, 1969; Willson, 1966; Howe, 1976); (2) functions in fruit/seed predator satiation when a portion of the fruit/seed crop is regularly damaged by predators (Janzen, 1971a,b); (3) is a by-product of selection for increased flower production in order to increase the male contribution to fitness in plants with bisexual flowers (Willson and Rathcke, 1974; Willson and Price, 1977; Charnov, 1979; Willson, 1979); (4) allows plants to regulate seed quality by selectively aborting fruits based on seed number or paternal parentage (Janzen, 1977; Charnov, 1979; Bertin, 1982; Lee and Bazzaz, 1982). These hypotheses assume that the increase in fitness derived from producing surplus fruits is greater than the decrease in fitness associated with the wastage of resources on aborted fruits. Consequently, there is interest in estimating the (in terms of fitness) of aborting juvenile fruits. One approach to this problem has been to compare the amount of resources that are contained in aborted fruits and mature fruits. These studies show that fruits abort early in their development and contain from < 1% to 10% of the biomass, total protein and inorganic nutrients of a mature fruit (see Lloyd, 1980; Stephenson, 1981; Bookman, 1983, and references therein). These studies suggest that fruit abortion is relatively inexpensive compared to a mature fruit. There are, however, two problems with interpreting these results. First, the specific nutrient(s) that limits fruit and seed production is usually unknown. Consequently, the relative cost of aborting a fruit varies greatly depending upon which resource is being compared. Second, it is unclear as to what ef-