Experiments were carried out to test the performance and some aspects of feeding behavior in two populations of Ceratitis capitata (a population reared in the laboratory for 16 years, i.e., approximately 160 generations, and a wild one obtained from infested coffee, Coffea arabica grains). Two types of food were used in the experiment: an artificial yeast diet used for laboratory rearing and papaya (Carica papaya), a natural host of the fly. The performance parameters tested were percent emergence, time to emergence, adult female size, and egg production during the pre-oviposition phase (first five days of adult life). The behavioral aspects tested were food preference by newly hatched larvae, induction, estimated ingestion of the two diets, whether the larvae placed on one diet stayed there or moved to the other diet, and acceptance of food for oviposition. The results indicated that the performance of the wild population was superior when the flies fed on papaya, whereas the performance of the laboratory population was similar with the two diets; the wild population showed a strong preference for papaya in all choice experiments, whereas the laboratory population showed no diet preference; the females of the wild population only oviposited on pieces of papaya that had not been peeled, and did not oviposit in the artificial diet; the females of the laboratory population oviposited indiscriminately on unpeeled and peeled papaya and on the artificial diet.