Pollination of Blueberry (vaccinium Ashei) by Honey Bees (apis Mellifera) and Nectar-thieving Carpenter Bees (xylocopa Virginica)

In a 2000 field study in which mature plants, potted pollenizers, and 0-12800 honey bees were tented together, honey bees were shown to effectively pollinate rabbiteye blueberry, variety “Climax.” There was a consistent and significant increase in fruit-set and number of seeds as bee numbers increased within the range of 0-6400 bees. In conference, the authors will present results of a 2001 field study, in progress at time of writing, which tests the pollination efficacy and interactions of honey bees and the ubiquitous nectar thief, Xylocopa virginica. It is expected that the data will shed light on the costs to pollination, if any, incurred from primary nectar thievery by Xylocopa and secondary thievery by Apis. Introduction Rabbiteye blueberry Vaccinium ashei is grown commercially in the southeastern USA. A crop native to this region, it is visited by an assortment of native bee pollinators as well as the exotic western honey bee, Apis mellifera. The most common bee visitors in order of their relative abundance are honey bees, bumble bee queens (Bombus spp., primarily B. impatiens), bumble bee workers, carpenter bees (Xylocopa spp., primarily X. virginica), and the Southeastern blueberry bee (Habropoda laboriosa) (Delaplane 1995). There is evidence that the species vary in their efficacy as pollinators of blueberry. Cane & Payne (1990) determined from monitoring tagged flowers that single-bee visits by Bombus spp. and H. laboriosa are the most likely to result in fruit-set. The best explanation for this is the ability of both bees to sonicate the blueberry flower, thus facilitating the release and transfer of pollen. The honey bee lacks this specialized behavior, and accordingly Cane & Payne noted that the probability of fruit-set with single visits by A. mellifera was only about 1%. Another relevant factor is the ubiquitous presence of carpenter bees in blueberry orchards in the region. These bees are common in blueberry but they engage without exception in nectar thievery (Delaplane 1995) by puncturing holes in the bases of corollae to suck out nectar, completely bypassing the exposed stigmata. The problem is compounded by the fact that honey bees rapidly learn to visit the puncture holes made by Xylocopa spp. and thus become secondary nectar thieves. It is thought that these types of flower visits contribute little to pollinating the crop and may in fact be counterproductive (Delaplane & Mayer 2000). Although Bombus spp. and H. laboriosa are efficient pollinators, native, and naturally-occurring in the region, their numbers are variable and unpredictable. All too often the bee fauna of a given orchard are dominated by A. mellifera and Xylocopa spp., most of whom are engaged in nectar thievery. It is thus of practical importance to better understand the pollination efficacy of honey bees in rabbiteye blueberry as well as the effects of nectar thievery on pollination. Materials & Methods Pollination efficacy of honey bees The pollination efficacy study in 2000 involved tenting mature plants of rabbiteye blueberry (variety “Climax”) with a potted pollenizer (var. “Premier”) and varying numbers of honey bees. Each tent (1.8 x 1.8 x 1.8 m) contained two mature plants and two potted pollenizers. The tenting material was light beige-colored Lumite (Bioquip Corporation) and impassable to all but the smallest insects. There were eight experimental plots as follows: plants tented with either 0, 400, 800, 1600, 3200, 6400, or 12800 honey bees, and one open plot. Each plot (tent) with honey bees contained a Langstroth-style hive sized either as a standard hive body (12800 bees) or a four-frame nucleus hive. In order to eliminate excessive population changes in the bee colonies, the colonies were kept broodless but socially stable by providing each synthetic queen mandibular pheromone (Winston & Slessor 1993) in lieu of a queen. Because the experiment was conducted in a permanent orchard interplanted with “Climax” and “Premier,” the open plot was not provided with potted pollenizers of “Premier.” After bloom the bees were removed and the tenting material replaced with poultry netting to minimize the effects of shade and to discourage unauthorized picking of fruit. In spring prior to bloom, forty fruit buds in each plot were labeled and the number of unopen florets recorded for each. At harvest the following parameters were recorded for each labeled fruit bud: fruit-set ((no. ripe fruit  no. unopen florets) 100), weight per berry (g), number of mature seeds per berry, and percent sucrose content of juice. Differences among the eight experimental conditions were analyzed in a completely randomized design analysis of variance using the ANOVA procedure (SAS Institute). Means were separated with Duncan’s test. Interactions of honey bees and carpenter bees This study involved tenting blueberry plants with various combinations of honey bees and carpenter bees. The work was conducted in 2001 under experimental conditions similar to that described above except that potted pollenizers were either “Tifblue” or “Brightwell.” In order to observe effects of learned secondary thieving behavior in Apis, we employed a switchback design in which X. virginica were either added or removed from some tents during each of three successive weeks (7 days each). Additionally, either one or two Xylocopa were inserted; males exclusively were captured from nature and used. For tents receiving Xylocopa, small blocks of wood with a 1 cm-diameter tunnel were inserted to provide nesting sites for the bees. For each tent receiving Apis, one nucleus colony with ca. 3200 honey bees was provided. It is convenient to explain the treatments in table I. table I. Experimental plots (tents) and their treatment assignments. tent no. Xylocopa used week 1 week 2 week 3 1 1 Apis only Apis + Xylocopa Apis only 2 2 Apis only Apis + Xylocopa Apis only 3 1 Apis + Xylocopa Apis only Apis + Xylocopa 4 2 Apis + Xylocopa Apis only Apis + Xylocopa 5 na Apis only 6 2 Xylocopa only 7 2 Apis + Xylocopa 8 na all bees excluded 9 na open plot In spring prior to bloom, forty fruit buds in each plot (80 in the open plot) were labeled and the number of unopen florets recorded for each. At harvest the same parameters as described above were recorded for each labeled fruit bud in addition to speed of ripening which was calculated as the percentage fruit ripe at one arbitrarily-chosen date. Results & Discussion Pollination efficacy of honey bees The results are shown in table II. There was a consistent and significant increase in fruit-set and number of seeds as bee numbers increased within the range of 0-6400 bees. At the highest bee density (12800) there was a significant decrease in both dependent characters; we believe that this is best explained by aberrant foraging behavior by bees under what were excessively crowded conditions. Sucrose content of juice appeared to vary negatively with fruit-set, with higher sucrose values in those tents with the lowest bee densities and fruit-set. Apparently those plants with fewer developing berries invested each with a comparatively higher fraction of available carbohydrate reserves. This is reverse to the trend noted by Eischen & Underwood (1991) who found in cantaloupe a simultaneous increase in sucrose content and fruit-set. There were no treatment effects for average weight of berry. table II. Means  standard errors of characters measured to determine efficacy of honey bees as pollinators of rabbiteye blueberry, variety “Climax.” Column means with the same letter are not different at the α = 0.05 level. Numbers in parentheses = n. initial bee number fruit-set (%) berry weight (g) seeds per berry sucrose content (%) open plot 39.1  7.3 (22) cd 0.75  0.06 (16) 21.2  3.2 (16) a 11.6  0.4 (15) b 0 31.1  4.8 (31) cd 0.7  0.06 (22) 0.4  0.3 (22) d 15.7  0.5 (22) a 400 26.3  4.5 (32) d 0.9  0.06 (20) 1.6  0.8 (20) d 16.3  0.6 (20) a 800 32.8  4.9 (38) cd 0.8  0.06 (27) 1.7  0.4 (27) d 15.7  0.5 (27) a 1600 55.5  5.5 (32) b 0.8  0.04 (31) 11.8  0.9 (31) b 11.4  0.3 (31) b 3200 74.9  5.3 (26) a 0.8  0.04 (28) 11.9  1.1 (28) b 11.7  0.3 (28) b 6400 80.0  4.1 (34) a 0.7  0.02 (33) 13.8  1.0 (33) b 12.2  0.4 (33) b 12800 45.1  6.7 (29) bc 0.7  0.04 (26) 6.6  0.8 (26) c 12.4  0.5 (25) b These results demonstrate that honey bees are capable of pollinating rabbiteye blueberry, variety “Climax.” Our results are consistent with the late research of Sampson & Cane (2000) on “Climax,” but counter to the results of Cane & Payne (1990) on “Tifblue.” Apparently, the efficacy of honey bees on rabbiteye blueberry is variety-dependent. Interactions of honey bees and carpenter beesAt the time of this writing (15 July 2001), data collection is not complete. Results and discussion will bepresented at Apimondia in Durban. It is expected that the data will shed light on the costs to pollination,if any, incurred from primary nectar thievery by Xylocopa and secondary thievery by Apis. ReferencesCane, J.H. & Payne, J.A. 1990. Native bee pollinates rabbiteye blueberry. Alabama AgriculturalExperiment Station 37, 4 Delaplane, K.S. 1995. Bee foragers and their pollen loads in south Georgia rabbiteye blueberry.American Bee Journal 135, 825-826 Delaplane, K.S. & Mayer, D.F. 2000. Crop pollination by bees. CABI Publishing, Oxon, UnitedKingdom Eischen, F.A. & Underwood, B.A. 1991. Cantaloupe pollination trials in the lower Rio Grande valley.American Bee Journal 131, 775 Sampson, B. & Cane, J.H. 2000. Pollination efficiencies of three bee (Hymenoptera: Apoidea) speciesvisiting rabbiteye blueberry. Journal of Economic Entomology 93, 1726-1731 Winston, M.L. & Slessor, K.N. 1993. Applications of queen honey bee mandibular pheromone for bee-keeping and crop pollination. Bee World 74, 111-128 POLLINATION OF BLUEBERRY (VACCINIUM ASHEI) BYHONEY BEES (APIS MELLIFERA) AND NECTAR-THIEVINGCARPENTER BEES (XYLOCOPA VIRGINICA) Keith S. Delaplane and Selim Dedej Department of Entomology, University of Georgia, Athens, Georgia 30602 USAE-mail: correspondence, [email protected]: (706) 542-1765, Fax (706) 542-3872Website: www.ent.uga.edu/bees CV of presenting author, Keith S. Delaplane, Professor Professional biographyDr. Keith S. Delaplane is Professor of Entomology at the University of Georgia, USA, where heconducts research, teaches, and advises graduate students in honey bee biology and crop pollination. Hewas keynote speaker for the plenary session on varroa mites in Apimondia Vancouver in 1999. His latestaccomplishment in pollination is the 2000 publication of his book, Crop Pollination by Bees, co-authored with Daniel F. Mayer and published in the United Kingdom by CAB International.