Patients with generalized peroxisomal disorders, rhizomelic chondrodysplasia punctata, and Refsum disease are all unable to alpha-oxidize 3,7,11,15-tetramethylhexadecanoic (phytanic) acid. The exact cause of the oxidation defect in these patients is not well characterized, in part because there is only limited knowledge of the biochemical pathway. In 1969, the alpha-oxidation of phytanic acid was reported [Tsai, S.-C., Avigan, J. & Steinberg, D. (1969) Studies on the alpha-oxidation of phytanic acid by rat liver mitochondria, J. Biol. Chem. 244, 2682-2692] to involve the formation of an alpha-hydroxyphytanic acid intermediate prior to removal of the alpha carbon. Subsequently, most researchers have had difficulty detecting this intermediate. In the present study, cofactors known to form hydroxy intermediates by both monooxygenase and dioxygenase reaction mechanisms were incubated with purified rat liver peroxisomes and either [2,3-3H]phytanic acid or [1-14C]phytanic acid. Reaction products were separated by reverse-phase HPLC. A single reaction product, identified as alpha-hydroxyphytanoyl-CoA rather than the free fatty acid, was detected when 2-oxoglutarate/Fe+2/ascorbate, cofactors associated with a dioxygenase reaction mechanism, were present. Concomitant with alpha-hydroxyphytanoyl-CoA production, there was an increased accumulation of formate and CO2. This increase in alpha-oxidation products is evidence that alpha-hydroxyphytanoyl-CoA is a true pathway intermediate and that the entire pathway functions in peroxisomes. In contrast, alpha-hydroxyphytanoyl-CoA was not formed in any quantity in mitochondria. These studies suggest that the alpha-hydroxylation step of phytanic acid oxidation, which has been shown to be defective in Refsum disease, is located in peroxisomes.