Temporal changes in the pollinator fauna of tristylous Pontederia cordata, an aquatic plant

Temporal changes in the composition of the pollinator fauna of the tristylous aquatic plant Pontederia cordata were documented during the flowering seasons of 1982 and 1983. A total of 18 species of insects, mainly bumblebees, and a single species of hummingbird were observed visiting flowers during the 2-month flowering period. The abundance of pollinators within each season varied greatly, with highest densities corresponding to peak flowering in both years. Although overall abundance was similar each year, the diversity of species was greater in 1983 because of the addition of short-tongued bumblebees and the longer residency periods of other species. It is suggested that low summer rainfall in 1983 decreased flowering levels in terrestrial plant communities of the region, resulting in several bee species shifting their feeding onto P. cordata. The study demonstrates that changes in the structure of a plant's pollinator fauna occur over time and that despite the possession of a specialized pollination mechanism, tristylous P. cordata is visited by a wide range of pollinator species.

The reproductive biology of boreal forest herbs. I. Breeding systems and pollination

Detailed observations and experimental studies of the reproductive biology of 12 boreal forest herbs were conducted over a 3-year period (1978–1980) in spruce–fir forests of central New Brunswick. The species examined were Aralia nudicaulis, Chimaphila umbellata, Clintonia borealis, Cornus canadensis, Cypripedium acaule, Linnaea borealis, Maianthemum canadense, Medeola virginiana, Oxalis montana, Pyrola secunda, Trientalis borealis, and Trillium undulatum. All taxa are insectpollinated perennials and most exhibit clonal growth. Floral syndromes of the understory community are relatively unspecialized with many species possessing small white or green flowers. A total of 103 taxa of insects were collected from flowers during the 1979 season. Bombus spp. are the major pollinators of 5 of the 12 species. Syrphid flies, bee flies, and halictid and andrenid bees were also commonly observed. Controlled pollinations were undertaken to determine the breeding systems of herbs. Bagged, self-, cross- and open-pollinated treatments were used to investigate the capacity for self-pollination, compatibility status, and factors influencing fecundity in each species. A diversity of reproductive systems was revealed. Six species are completely dependent on insects for pollination, four species are weakly autogamous, one is strongly autogamous, and one appears to be apomictic. Comparisons of fruit-set and seed set from controlled self- and cross-pollinations indicate that four species are strongly self-compatible, one is dioecious, and the remainder display varying degrees of self-incompatibility. Despite this variation, outbreeding appears to be the most common reproductive mode. In several species there is evidence that low pollinator service limits fruit-set.

The reproductive biology of boreal forest herbs. II. Phenology of flowering and fruiting

The flowering and fruiting phenologies of 12 boreal forest herbs were recorded during 1979 (flowering and fruiting) and 1980 (flowering only) in spruce–fir forests of central New Brunswick. The species studied were Aralia nudicaulis, Chimaphila umbellata, Clintonia borealis, Cornus canadensis, Cypripedium acaule, Linnaea borealis, Maianthemum canadense, Medeola virginiana, Oxalis montana, Pyrola secunda, Trientalis borealis, and Trillium undulatum. Flowering in the community occurred from mid-May to the end of July. The order of flowering was maintained in the 2 years, but the degree of synchronization of inflorescences differed in several species. Fruiting in the community began in mid-July and extended beyond the end of September. The percentage of buds that ultimately bore fruit ranged from 0 (Cypripedium acaule) to 61% (Aralia nudicaulis). With the exception of Cypripedium acaule, which received little pollinator service, the self-incompatible species, Cornus canadensis, Maianthemum canadense, and Medeola virginiana, experienced the lowest levels of fruit-set. Pollen limitation and predation of developing fruit appear to be the major factors limiting percentage fruit-set in boreal forest herbs. Fruit production varied with time of flowering of inflorescences in several species, with periods of low fruit-set tending to coincide with lower densities of flowering inflorescences. Significant rates of fruit removal by herbivores occurred in all sarocochorous species. Disappearance of fruits from infructescences ranged from 31 (Medeola virginiana) to 95% (Aralia nudicaulis), with highest removal rates occurring during periods of greatest fruit availability.

Pollinator foraging behavior and pollen collection on the floral morphs of tristylous Pontederia cordata L

The foraging behavior of the pollinators of tristylous Pontederia cordata was studied to determine if differences in floral morphology would lead to preferential visitation of the floral morphs. Although nectar production is not different in the three floral morphs, differences in the production and size of pollen grains produced by the three anther levels results in the morphs offering variable amounts of resources to pollen-collecting insects. Bumblebees (Bombus spp.) and the solitary bee Melissodes apicata used P. cordata primarily as a nectar source and therefore did not seem to exhibit any morph preference. In contrast, honeybees visited flowers mainly for pollen and preferred to forage on long-level anthers of the short-and mid-styled morphs. An analysis of the composition of corbicular pollen loads indicated that, relative to the frequency of production in the population: 1) honeybees collected an excess of pollen from long-level anthers; 2) bumblebees collected the three types of pollen without any apparent preference; and 3) M. apicata preferentially collected pollen from the short-level anthers - presumably because their proboscides are modified by the presence of tiny hairs. The results suggest that P. cordata in Ontario is serviced by a diverse, unspecialized pollinator fauna which is not co-adapted to the tristylous floral polymorphism.

The effect of pollination intensity and incompatible pollen on seed set in Turnera ulmifolia (Turneraceae)

Controlled pollination experiments were performed on the self-incompatible distylous herb Turnera ulmifolia L. to investigate the effects of pollination intensity and large amounts of incompatible pollen on seed set. In the first experiment, known numbers of compatible pollen grains ranging from 1 to 100 were applied to stigmas of the floral morphs. In both morphs, increasing amounts of pollen generally resulted in increased levels of seed set, although considerable variance was observed at all pollination intensities. Approximately two to seven pollen grains are required to produce a single seed and more than 95 grains are required to achieve maximum seed set in T. ulmifolia. Regression analysis of the seed set data failed to detect a difference in the response of the floral morphs to pollination intensity. In the second experiment, known proportions of compatible and incompatible pollen were applied to stigmas at various time intervals. Most treatments involving mixtures of compatible and incompatible pollen had no significant effect on seed set when compared with the controls. Clogging was only observed in the long-styled morph when one anther of compatible pollen was applied to stigmas 1.5 and 3.0 h after pollination with five anthers of incompatible pollen. The clogging of stigmas by incompatible pollen seems unlikely to have played a major role in the evolution and maintenance of distyly in Turnera ulmifolia.

Colonizing ability in the Echinochloa crus-galli complex (barnyard grass). II. Seed biology

Comparisons of some seed parameters of two taxa of barnyard grass (Echinochloa crus-galli complex) were made in an effort to understand differences in habitat preference and colonizing ability. The taxa were E. crus-galli var. crus-galli, a cosmopolitan weed of wet, disturbed ground and E. crus-galli var. oryzicola, a crop mimic restricted to rice fields. In California, where all seed collections originated, the two taxa are ecologically differentiated within the rice-field ecosystem. Echinochloa crus-galli var. crus-galli is a weed of paddy banks and shallow water around the periphery of rice fields whereas E. crus-galli var. oryzicola is found primarily within permanently flooded rice fields. A survey of seed weight in 10 populations of each of the two taxa demonstrated that seeds of E. crus-galli var. oryzicola were on the average two to three times heavier than those of E. crus-galli var. crus-galli. Differences in weight were reflected in the buoyancy characteristics of fresh seeds of the two taxa. Approximately 50% of E. crus-galli var. crus-galli seeds remained afloat after 4–5 days in water whereas during the same period over 95% of the seeds of E. crus-galli var. oryzicola had sunk. Germination tests on 9- and 18-month-old seeds of 18 populations of E. crus-galli var. crus-galli and 11 populations of E. crus-galli var. oryzicola revealed significant differences between the taxa in the number of germinating seeds. The decay of dormancy in E. crus-galli var. oryzicola was more rapid than in E. crus-galli var. crus-galli following dry storage and burial in soil. The difference results in the greater germination synchrony of the crop mimic in comparison to the widespread weed. Emergence of seedlings from saturated and flooded soils was compared in the two taxa. In all treatments E. crus-galli var. oryzicola exhibited significantly greater levels of seedling emergence. The differences may explain changes in abundance of the two taxa in California rice fields, following the introduction of permanent flooding as a weed control practice, as well as their current microdistribution within the rice-field ecosystem.

Spatial pattern, floral sex ratios, and fecundity in dioecious Aralia nudicaulis (Araliaceae)

Aralia nudicaulis L. (Araliaceae) is a perennial understory herb of the boreal forest which forms large clones by means of an extensive subterranean rhizome system. Populations are dioecious and pollinated primarily by bumble bees. To examine the relationships between the spatial pattern of sexual morphs and female fecundity, all flowering ramets were mapped in a 1-ha block of spruce–fir forest in central New Brunswick and the fecundity of females determined. Within the block, A. nudicaulis exhibited a 2.5:1 sex ratio of male/female flowering ramets. Ramets of the sexual morphs were negatively associated owing to cloning and a reduced likelihood of flowering in female ramets under dense canopy compared with males. Pattern analysis of the sex ratio of flowering ramets revealed a monotonic decline in variance with block size indicating that samples of sex ratio using small quadrats are likely to give misleading estimates for this species. The fecundity of females exhibited no systematic positional variation within the forest block. In most cases fecundity was not correlated with male and female flowering density, ramet sex ratio, and several subjectively estimated habitat variables. It is proposed that the long flight distances of pollinators, pollen carry-over, and the predominance of male flowers at the site interact to dampen out spatial effects on fecundity.