The present investigation was designed to quantitatively assess the possible influence of countercurrent exchange on passive absorption from the small intestine of the dog. Villus blood flow was measured with a modification of the microsphere method. Simultaneously, the absorption from the gut lumen of five diffusible gases (H2, He, CH4, 133Xe, and CO) was determined. Villus blood flow averaged 0.247±0.03 (SEM) ml/min per g. The observed absorption of H2, He, CH4, and 133Xe was only 16.2±1.8, 12.8±2.3, 12.0±1.8, and 15.8±1.4%, respectively, of what this villus blood flow could carry away if it reached perfect equilibrium with the luminal gases. This low absorption rate could result from diffusion limitation to absorption or countercurrent exchange. The diffusive permeability of the barrier separating the liminal gases and villus blood flow was assessed by measuring the absorption rate of CO. Because absorbed CO binds tightly to hemoglobin, it cannot exchange, and when present in low concentrations its uptake is entirely diffusion limited. Knowledge of the diffusion rate through tissue of the unbound gases relative to that of CO made it possible to calculate the degree to which each of the unbound gases should equilibrate with villus tip blood. The percentage equilibration between lumen and blood at the villus tip for H2, He, CH4, and 133Xe was 99.7, 99.9, 75.6, and 36.0%, respectively. Each of these values greatly exceeded the percentage equilibration of blood leaving the villus (calculated from the observed absorption rate and villus blood flow) and indicated an exchange of 83.8, 87.2, 84.1, and 56.1% of initially absorbed H2, He, CH4, and 133Xe. This result is in accord with theoretical calculations which suggest that countercurrent exchange should be exceedingly efficient in the dog. The striking effect of countercurrent exchange on passive absorption in the dog differs from our previous studies in the rabbit where no exchange was demonstrated. This marked species difference may result from anatomical differences in villus architecture. The dog has long, densely packed villi while the rabbit has broad, widely spaced villi. In the dog, only the villus tips may equilibrate with the lumen, hence a countercurrent gradient may be established in the villus. The entire villus of the rabbit may equlibrate with the lumen and no gradient for countercurrent exchange can therefore be established.