The purpose of this study was to clarify the influence of skin thickness on the in vitro permeabilities of 3 model drugs with different physicochemical properties (nicorandil (NR), isosorbide dinitrate (ISDN) and flurbiprofen (FP)) through Sprague-Dawley rat (rat) or Yucatan micropig (YMP) skin. Intact, dermis-split, stratum corneum-stripped or stratum corneum-stripped and dermis-split rat or YMP skin (rat skin thickness: approximately 0.4, 0.9 or 1.2 mm; YMP skin thickness: approximately 0.4, 0.9, 1.8 or 2.8 mm) were set in Franz-type diffusion cells to determine the permeation rate, lag time and resistance ratio of the viable epidermis and dermis against whole skin (R(ved)/R(tot)) of the drugs. The YMP skin permeabilities of the drugs decreased with an increase in the skin thickness, and significant differences were observed in the permeation rates and lag times between intact and dermis-split (0.4 mm) YMP skins. The decreases in the permeabilities of the drugs through the YMP skin were larger than those through the rat skin. The influence of resistances of ISDN and FP through the dermis-split rat or YMP skin was greater at 0.9 mm skin thickness than 0.4 mm skin thickness. The R(ved)/R(tot) values for the YMP skins were relatively large for lipophilic drugs (ISDN and FP), and these ratios increased with an increase in the dermis thickness. These results suggest that in vitro skin permeation studies must be done using dermis-split (0.4 mm) skin with the thinnest dermis for predicting in vivo human percutaneous absorption rate.