The production of F₁ interspecific hybrid cotton (Gossypium hirsutum x G. barbadense) seed was studied using several A line (female parent) X B line (male parent) row ratios in a field experiment in 1983. Honey bee (Apis mellifera L.) foraging density, flowering rates, pollen/stigma counts and seed yield data were obtained. Also, the percentage of hybrid progeny was determined on seed from rows representative of 2A X 2B, 4A X 2B, 6A X 2B and 10A X 2B planting ratios. Pollen deposition on A-line stigmas adjacent to Pima pollen sources in the 2A X 2B area averaged only 19.5% of that deposited on A-line stigmas adjacent to the upland pollen source. Honey bee visitation was nearly equal (often exceeding 1%) in both areas. Also, pollen dispersal was generally low (averaging 50 pollen grains or less) except when wild bee visitation was about 1%. When pollen/stigma counts were very low (averaging 1-11 grains/stigma), no correlation with distance from pollen source could be documented, but seed yields, seeds/boll, and percent hybrids all were negatively correlated with distance from pollen source. Harvestable F₁ hybrid seed (kg/h) was highest at the 4A X 2B planting ratios with yields equal to 63% (63% = 1 , 2 3 7 kg/ha) of the maximum upland A-line yield obtained in this experiment. The results suggest that at least 72 m of isolation distance is required when attempting to obtain a specific Pima X upland hybrid when upland pollen is available in the area.