Premium
Relationships among Flowering Phenology, Insect Visitors, and Seed‐Set of Individuals: Experimental Studies on Four Co‐occurring Species of Goldenrod (Solidago: Compositae)
Author(s) -
Gross Ronald S.,
Werner Patricia A.
Publication year - 1983
Publication title -
ecological monographs
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 4.254
H-Index - 156
eISSN - 1557-7015
pISSN - 0012-9615
DOI - 10.2307/1942589
Subject(s) - biology , solidago canadensis , pollinator , phenology , pollination , abundance (ecology) , botany , pollen , raceme , hand pollination , insect , ecology , invasive species , inflorescence
Four co—occurring species of goldenrod bloom at different times with varying degrees of overlap; in order of peak flowering they are Solidago juncea, S. graminifolia, S. canadensis, and S. nemoralis. All four species are self—incompatible and require an insect vector for successful seed—set. First, the relationship between flowering time and seed—set of individual plants of these four species was determined. All four species of Solidago had significant differences in both the percentage of filled seeds and the total seed—set of clones that flowered at different times. Early—flowering clones had lower seed—set than did late—flowering clones in S. canadensis, S. graminifolia, and S. nemoralis. In contrast, early—flowering ramets of S. juncea had significantly greater seed—set than did late—flowering ramets. Secondly, the underlying factors limiting seed—set were investigated by observational and experimental techniques, in order to lend insight into the processes which can select for flowering time in natural plant populations. The abundance of pollinators on goldenrods and other plant species and the abundance of flower predators were monitored over the season. Experimental hand—pollinations were performed on individual plants of each species over the entire flowering season, to determine if seed—set was limited by the amount of pollen reaching stigmas or by factors intrinsic to individual plants which flowered at different times. Apis mellifera, the introduced honeybee, is the major pollinator of goldenrods in this system. The major factor influencing the relative abundances of honeybees on these four species of goldenrod seemed to be overlap among the flowering periods of the goldenrods and those of several introduced plant species. Apis began visiting goldenrods at a point when the abundance of flowers in the weedy flora had declined greatly. The flowering period of S. juncea overlapped almost entirely with the flowering periods of several species in the weedy summer flora. This appears to explain the lack of Apis visits to S. juncea, which was visited only by small, native bees and beetles. However, the results from experimental pollinations suggest that seed production of S. juncea was at its potential maximum throughout the flowering season. In early—flowering clones of S. graminifolia, the natural seed—set was significantly lower (by 57%) than the maximum potential seed—set determined from experimental pollinations, but in late—flowering clones there was no differences between actual and potential seed—set. Of the reduction from potential seed—set in early—flowering clones, the low frequency of honeybee visits was estimated to account for most of the loss, and flower predation by the blister beetles, Epicauta pennsylvanica, accounted for the remainder. In S. canadensis, the seed—set of both hand—pollinated and control flowers was greater in late—relative to early—flowering clones of S. canadensis (i.e., late—flowering clones had greater physiological potential for seed production). The differences in the seed—set of clones flowering at were different times were due to physiological or microenvironmental differences among clones. There was also some degree of pollen limitation of seed—set (17—34% of the potential maximum seed—set) at all times. Again, pollen limitation of seed—set was due mostly to pollinators, as opposed to flower predators. In S. nemoralis, the late—flowering clones had significantly higher seed—set than early—flowering clones, but maximum potential seed—set was not determined for this species. Such differences in seed—set could select for flowering time, given the right combination of heritability and morphological or life history tradeoffs. The selection would not necessarily lead to further divergence in flowering phenology among these congeners, but rather for later flowering in S. graminifolia, S. canadensis, and S. nemoralis and for earlier flowering in S. juncea.