SUMMARY.—The vasoconstrictor test of McKenzie and Stoughton (1962) has been adapted for studying the release of betamethasone 17‐valerate and 21‐desoxy betamethasone 17‐propionate from ointments. Vasoconstriction on the skin of volunteers was assessed quantally by noting presence or absence of vasoconstriction or quantitatively by measuring diameters of vasoconstricted areas. The quantitative method employs corticosteroid doses higher than those used in the quantal method of McKenzie and Stoughton. The vasoconstrictor potencies of 3 betamethasone derivatives have been compared using the quantitative method in order to confirm its validity. The quantitative method was used for comparing various ointment bases with betamethasone 17‐valerate or 21‐desoxy betamethasone 17‐propionate. High relative potencies were obtained by the inclusion of propylene glycol in the ointments. Ointments containing betamethasone 17‐valerate and varying amounts of hydrogenated lanolin were compared.The quantitative and quantal methods showed that relative vasoconstrictor potency increased or diminished respectively with increased concentration of hydrogenated lanolin. The effect of hydrogenated lanolin in vehicles on the release of betamethasone 17‐valerate was studied using twoin vitromodel systems and the results are discussed in relation to those of thei