1. The percutaneous absorption of a novel oxazolino corticosteroid, 14C-9alpha-fluoro-11beta,21-dihydroxy-3,20-dioxopregna-1,4-dieno[17alpha,16alpha-d]-2'-methyloxazolidine-21-acetate (L 6400) and 3H-triamcinolone acetonide (TA) was compared in rats, pigs and humans after application of a cream formulation to the skin and occlusion. 2. Within 96 h, up to 4.0% of the applied dose of L 6400 (permeability constant Kp 0.28) and up to 8.7% of TA (Kp 0.42) were percutaneously absorbed in rats. 3. Microhistoautoradiography showed that most of the applied radioactivity remained on the skin of rats. Radioactivity in the skin was mainly associated with the epidermis. 4. Within 120 h, up to 1.5% of the applied dose of L 6400 (Kp 0.16) and up to 0.5% of TA (Kp 0.07) were percutaneously absorbed in pigs. Plasma concentrations of radioactivity were maximal 48 h after application of 3H-TA whereas they were undetected after application of 14C-L 6400, because of the relatively lower specific activity of this 14C-steroid. 5. Both corticosteroids were allowed to remain in contact with human skin for 5 h and during that time, up to 1.1% of the applied dose of L 6400 (Kp 0.82) and up to 0.9% of TA (Kp 0.94) were absorbed. Results in four human subjects varied twenty-eightfold for L 6400 and fivefold for TA. Plasma concentrations of radioactivity were undetected after application of L 6400. In one subject treated with 3H-TA they were maximal after 2 h but were undetected in another subject and were found only in early samples from two others. 6. The results suggested that both corticosteroids were poorly absorbed through normal skin in rats, pigs and humans.