The use of interspecific hybridization in blueberry (Vaccinium L. section Cyanococcus Gray) breeding has resulted in the incorporation of novel traits from wild germplasm and the expansion of the geographic limits of highbush blueberry (V. ×corymbosum L.) production. The objectives of this study were: 1) to estimate the impact of wide hybridization on inbreeding, heterozygosity, and coancestry of the cultivated tetraploid highbush blueberry; 2) to establish the usefulness of microsatellite markers in assessing genetic relationships among southern highbush blueberry [SHB (V. ×corymbosum)] cultivars; and 3) to report on the expected genetic contribution of wild Vaccinium clones to SHB cultivars. Pedigrees of 107 highbush blueberry cultivars were used to calculate tetrasomic inbreeding coefficients (F), pedigree-based genetic distances, and expected genetic contributions of wild clones. Genotyping data from 21 single-locus microsatellite markers screened across 68 genotypes were used to calculate heterozygosity and proportion of shared alleles distances (Dsa). The results indicated that the effects of wide hybridization on genetic diversity of cultivated blueberry were lower than previously thought. First, SHB cultivars exhibited similar levels of molecular relatedness as historical northern highbush blueberry [NHB (V. ×corymbosum)] cultivars (median Dsa between pairs of cultivars equals 0.19 in both cultivar groups), despite the broader genetic base and larger pedigree distances in the former cultivar group. Second, levels of heterozygosity in modern NHB cultivars were not statistically different from those of SHB (χ2 = 4.0; df = 3; P > 0.05), despite the considerably higher levels of inbreeding in the former cultivar group (median F equal to 0.0035 and 0.0013, respectively). Furthermore, pedigree-based genetic distances were significantly correlated with Dsa (r = 0.57, P ≤ 0.0001), indicating that microsatellite markers are reliable tools in most cases to assess the genetic relationships among SHB cultivars. Finally, seven Vaccinium species constitute the current genetic base of cultivated SHB. In this article, we report on the expected genetic contributions of V. angustifolium Ait., V. constablaei Gray, V. corymbosum, V. darrowii Camp, V. elliottii Chapman, V. tenellum Ait., and V. virgatum Ait. (syn. V. ashei Reade) clones to 38 SHB cultivars.

Rabbiteye blueberry (Vaccinium ashei Reade) often exhibits problems with low fruit set. Little is known about the duration of flower receptivity in this species. The objective of this study was to investigate the effects of flower age at pollination on fruit set, seed number per fruit, and stigmatic receptivity. `Brightwell' and `Tifblue' rabbiteye blueberry plants were kept under controlled conditions in a growth chamber. Day/night temperatures during pollination were 23 °C/10 °C. Flowers were hand pollinated with self- or cross-pollen at 2-day intervals ranging from 0 to 8 days after anthesis (DAA). Flower age at pollination had a significant effect on both fruit set and seed number per fruit. Rabbiteye blueberry flowers were able to produce optimum fruit set during a period of at least five days. Fruit set was markedly reduced 6 to 8 DAA, depending on the cultivar. Flower age at pollination also had a significant effect on stigmatic receptivity, which was assessed as the number of germinated tetrads on the stigma 24 hours after pollination. Stigmas pollinated 0 DAA had a significantly lower number of germinated tetrads than those pollinated 8 DAA. Flower age at pollination and stigmatic receptivity were positively associated. To our knowledge, this is the first quantitative evidence of delayed stigma maturation in blueberry. Stigmatic receptivity and fruit set were not correlated. Overall, the data strongly suggest that stigmatic receptivity was not a limiting factor for fruit set of `Brightwell' and `Tifblue'. It is hypothesized that ovule longevity determines the duration of flower receptivity in these two rabbiteye blueberry cultivars.

Blueberries are bee-pollinated species that benefit from cross-pollination. Cross-pollination is particularly critical for optimum fruit set of rabbiteye blueberries (Vaccinium ashei Reade) because of their limited degree of self-fertility. In order to determine if the failure to set adequate commercial fruit loads is due to a lack of cross-pollination, research was needed to establish how much out-crossing rabbiteye blueberry pollinators actually do. A novel method was developed to identify pollen grains on the bodies of bumblebees by cultivar. The technique discriminates between two cultivars, based on differences in pollen diameter. Bumblebees were collected in a plot composed of blueberry plants of the cultivars Brightwell and Climax since these cultivars produce pollen of different size. Pollen loads of bumblebees contained low proportions of cross-pollen regardless of the cultivar they were visiting. Data suggest that inadequate levels of cross-pollination play a major role in low fruit set problems of rabbiteye blueberry. The composition of bees' pollen load changed with the phenology of the crop. The greatest likelihood for cross-pollination occurred around the time of maximum bloom overlap. Bumblebees foraging on `Brightwell' flowers carried more total blueberry pollen and a higher proportion of self-pollen than those visiting `Climax'. This may be due to differences in pollen release between flowers of these two cultivars.

Fruit set of rabbiteye blueberries (Vaccinium ashei Reade) can be pollen-limited under certain conditions. The objective of this study was to determine production, release, and viability of pollen, as well as pollen-ovule ratios in the rabbiteye blueberry cultivars Austin, Brightwell, Climax, and Tifblue. In vitro tetrad germination varied among genotypes, although, values were high (≥80%) in all cultivars. Pollen viability does not seem to contribute to reproductive failure in the cultivars studied. Total pollen production per flower averaged 8434 tetrads across all cultivars. On a per ovule basis, pollen production was very low relative to other xenogamous species. The low pollen-ovule ratio of rabbiteye blueberry (≈400) may be an indicator of the high efficiency of its pollen dispensing mechanism. Total pollen production varied among cultivars. Furthermore, a significant difference in pollen release was found between two cultivars with similar total pollen production per flower. The possible mechanism regulating pollen release in these cultivars is discussed.

Blueberry species (Vaccinium section Cyanococcus) benefit from cross-pollination. Outcrossing increases fruit set, berry size, and ripening rate. Although knowledge of pollen dispersal is essential for maximizing cross-pollination and achieving optimal planting designs, this process has not been quantified previously in blueberry plantings. A novel method was developed to estimate the proportion of self- and cross-pollen transported by blueberry pollinators. The proposed technique requires a consistent difference in pollen size between two cultivars to predict the composition of a pollen mixture based on frequency distributions of pollen diameter. Vaccinium ashei Reade `Brightwell' and `Climax' were chosen for this study because they produce pollen tetrads of different size. Tetrad diameter and number were analyzed with a particle counter. The technique was validated by predicting the proportion of `Brightwell' in pollen mixtures where the cultivar composition was known, and predicted and actual values were linearly correlated (r = 0.995, P < 0.0001). The technique was then applied to pollen samples extracted from the bodies of pollinators that were collected in a mixed `Brightwell' and `Climax' blueberry plot. Numbers of blueberry tetrads extracted per bumblebee (Bombus spp.) averaged 4595 and 797 in 2003 and 2004, respectively, which was considered adequate to make accurate predictions based on frequency distributions of tetrad diameter. The proportion of `Brightwell' pollen carried by bumblebees changed with the phenology of the crop following an expected pattern, indicating that the method performed well under field conditions. This technique could potentially be used to quantify the likelihood for outcrossing and establish the effect of cultivar arrangements on pollen dispersion, as well as to examine pollen collection and manipulation, and cultivar preference by bees.