Fruit set on seven self-fertile (SF) sweet cherry genotypes from the Cornell Sweet Cherry Breeding Program was investigated. The S-alleles of all tested genotypes, based on controlled pollination and PCR-based S-allele typing analysis were determined to be: S1S4’ for Lapins, and S3S4’ for Stella, Starkrimson, NY 13688, NY 13696, NY 13788 and NY 13791. Higher fruit set of open-pollinated self-fertile genotypes indicated that the additional pollen donor was probably advantageous for fruit set in self-fertiles in the preliminary field study. However, no statistically significant differences were found between fully and self- (half compatible) pollination in caged trees in subsequent years’ studies. This inconsistency may have been due to shading by cages (for self-pollination only) or to differences in bee activity in caged trees as compared with open-pollinations made during preliminary studies. Additionally, the variation in fruit setting potential was dependent on genotype. Spring frost injury reduced fruit setting potential in self-fertile sweet cherries but pollen quality was not associated with variable fruit set. The results suggest that self-pollination would produce optimal fruit set in self-fertile sweet cherry cultivars regardless of partial pollen incompatibility. However, a more detailed study including fruit setting factors other than spring frost injury, and varying setting capacity depending on genotypes, would be required for a more complete determination of the commercial utility of growing self-fertile sweet cherries in solid blocks. Key words: Prunus avium L, sweet cherry, pollination, fruit set