Captive breeding of endangered species is commonly proposed as a means of conserving biodiversity. The suggestion is that captive populations can be built up to provide individuals to reinforce or re-establish wild populations. However, there is evidence to suggest that captive-bred animals lack the skills necessary for survival in their natural habitat. This research was designed to assess whether a group of captive-bred Varecia variegata variegata exhibit such behavioral deficiencies in relation to feeding behavior compared to two wild groups. A further aim was to determine which of four methods of food presentation and two captive environments were most effective in encouraging the exhibition of natural feeding behavior. An identical focal animal, instantaneous time-sampling technique was used to collect data from the wild groups in the Betampona Reserve, Madagascar, and the captive group at Chester, UK. Results from Betampona confirm the highly frugivorous and arboreal nature of V. v. variegata. Under all captive conditions, except the rooftop feed in the cage environment, the captive V. v. variegata spent significantly less time feeding than their wild conspecifics. Suspensory feeding postures are an important adaptation enabling wild V. v. variegata to harvest fruit in the small-branch setting. Similar frequencies of use of such postures to those observed at Betampona were most effectively encouraged among captive V. v. variegata in the cage environment by the rooftop food presentation method and on a naturalistic island exhibit by the suspended method of food presentation. Differences in feeding behavior between the wild and captive V. v. variegata can be explained in terms of structural differences between their environments. As naturalistic captive environments allow lemurs to experience the challenges associated with feeding in the wild, it is strongly recommended that zoos endeavor to provide them with such conditions. Zoo Biol 17: 379-392, 1998.