The major avenues of extrapulmonary oxygen uptake were determined on submerged western painted turtles (Chrysemys picta bellii) at 10 degrees C by selectively blocking one or more potential pathways for exchange. Previous work indicated that the skin, the cloaca, and the buccopharyngeal cavity can all contribute significantly in various species of turtles. O(2) uptake was calculated from the rate of fall in water P(O(2)) in a closed chamber. Two series of experiments were conducted: in Series 1, each of the potential avenues was mechanically blocked either singly or in combination; in Series 2, active cloacal and buccal pumping were prevented pharmacologically using the paralytic agent rocuronium. In addition in Series 2, N(2)-breathing preceded submergence in some animals and in one set of Series 2 experiments arterial blood was sampled and analyzed for pH, lactate, P(O(2)), and P(CO(2)). Results in both Series 1 and Series 2 revealed that prevention of cloacal and/or buccopharyngeal exchange did not significantly affect total O(2) uptake. Interfering with skin diffusion in Series 1, however, significantly reduced O(2) uptake by 50%. N(2)-breathing prior to submergence in Series 2 did not affect O(2) uptake in paralyzed turtles but significantly increased uptake in unparalyzed turtles without catheters. Blood analysis revealed that all submerged turtles developed lactic acidosis, but the rate of rise in lactate was significantly lower in paralyzed animals. We conclude that passive diffusion through the integument is the principal avenue of aquatic O(2) uptake in this species.