To determine the effects of pollen source on blueberry production, we made a partial diallel set of crosses involving seven rabbiteye (Vaccinium ashei Reade) and seven southern highbush (SH; V. corymbosum L.) parents. Pollination of rabbiteye blueberry flowers with SH pollen reduced fruit set, seeds per berry, and berry weight and increased fruit development period (FDP) compared to pollination with rabbiteye pollen. Pollination of SH flowers with rabbiteye pollen resulted in about the same fruit set and FDP but fewer seeds per berry and slightly lower berry weight compared to intraspecific pollination. Self-pollination significantly decreased the number of seeds per berry and berry weight and increased FDP in SH. Pollination of rabbiteye and SH flowers with mixed pollen produced the same results as intraspecific pollination. Using `Tifblue' and `Baldwin' (rabbiteye) as the pollen parent significantly increased FDP in rabbiteye blueberry. Using `Georgiagem' and `Cape Fear' as pollen parents produced the longest FDP, and using `O'Neal' and `Gulfcoast' produced the shortest FDP in SH blueberry. The heaviest berries were produced by using `Blue Ridge', `O'Neal', and `Gulfcoast' (SH) as pollen parents on SH females. These results suggest that xenia possibly could be used to increase yield and reduce FDP in blueberry.