Four assays were conducted to assess the attraction and feeding responses of oriental fruit fly (Bactrocera dorsalis (Hendel)) and melon fly (Bactrocera cucurbitae Coquillett) to formulations of baits. In all assays except the toxicology trial, only blank formulations of bait (i.e., no toxicant) were used. In separate choice attraction assays for B. dorsalis and B. cucurbitae, significantly more flies arrived at stations with bait than water, but no differences existed among blank baits of GF-120 Fruit Fly Bait, GF-120 NF Naturalyte Fruit Fly Bait, Provesta ® 621 autolyzed yeast extract, and Mazoferm ® E802. In a second attraction assay with treatment surface area kept constant, bait volume significantly regressed to numbers of B. dorsalis alighting on the bait, with less bait resulting in higher numbers of responders. In feeding assays for both species, the time spent feeding and time spent on a leaf were significantly effected by bait. For B. dorsalis the longest feeding times resulted with Provesta 621, with significantly less feeding on the other baits, and with all baits significantly higher than water. The longest feeding times for B. cucurbitae resulted with Mazoferm E802 and Provesta 621, which were significantly higher than both GF-120 baits, and all baits except GF-120 NF were significantly higher than water. In separate toxicology assays for each species, significantly higher mortality resulted from bait formulations containing the toxicant spinosad compared with blank baits, but no differences existed between GF-120 and GF-120 NF formulations.