The effect of anatomic site on the in vivo relationship between the total penetration of four compounds and the amount of the compounds present in the stratum corneum at the end of application was studied in humans. For each anatomic site, 1,000 nmol of 14C-radiolabeled benzoic acid, benzoic acid sodium salt, caffeine, or acetylsalicylic acid was applied to 1-cm2 area of skin of male Caucasian patients aged 28 +/- 2 years (groups of 6-8). For each molecule and each site, a first application on the right-hand side of the body allowed total absorption to be determined by measuring the amount excreted in the urine. A second application, performed 48 h later on the contralateral site, enabled the total amount of substance present in the stratum corneum at the end of application (30 min) to be assessed after cellophane-tape stripping of the treated area. The results showed that skin permeability varied substantially, depending both on the physicochemical nature of the molecule and on the anatomical location. In general, the rank order in skin permeability of the studied areas appears to be as follows: arm less than or equal to abdomen less than postauricular less than forehead. Whatever the compound applied, the forehead was approximately 2 times as permeable as the arm or abdomen. Independent of the origin of the differences in permeability observed among sites, there exists a linear correlation (r = 0.97, p less than 0.001) between the amounts of substance present in the stratum corneum at the end of application (30 min) and the total amounts which penetrated within a 4-d period.(ABSTRACT TRUNCATED AT 250 WORDS)