Microvascular diameters in the hamster cheek pouch were measured with a Vickers image-shearing eyepiece, and oxygen tension (Po2) was measured amperometrically with 2-6µ microcathodes. Perivascular Po2 was dependent on both the type of microvessel observed and the composition of the solution bathing the tissue. During application of a solution with a mean Po2 of 17 mm Hg and a Pco2 of 0 mm Hg, a longitudinal gradient in perivascular Po2 was observed: Po2 decreased from 44 ± 2 (SE) mm Hg at the large arterioles to 18 ± 2 mm Hg at the capillary origin. Tissue Po2 was 10 ± 1 mm Hg. Suffusion of the cheek pouch with a solution with approximately the same Po2 and a Pco2 of 32 mm Hg resulted in an elevation of perivascular Po2 at all sites. Large arteriolar Po2 under these circumstances was 47 ± 2 mm Hg, capillary origin Po2 was 29 ± 3 mm Hg, and tissue Po2 was 17 ± 3 mm Hg. CO2 also produced vasodilation: the average vascular diameter increased 18 ± 7% when the solution Pco2 was increased from 0 to 32 mm Hg. The tissue showed evidence of regulation of tissue O2supply both with and without CO2 in the suffusion solution. The effects of CO2 on the distribution of O2 were compared with the effects of other vasodilators, and it was found that the tissue Po2was not consistently changed by the application of the vasodilators, whereas it was elevated 70% by CO2. This difference is attributed in part to the effect of CO2 on the oxyhemoglobin dissociation curve.