The disposition of radiotocopherol after intraperitoneal administration of 14C-labelled-DL-alpha-tocopherol (1 microCi/kg bw, dose 1) and 3H-labelled-DL-alpha-tocopherol (4 microCi/kg bw dose 4) and of D-alpha-tocopherol, after intraperitoneal administration of DL-alpha-tocopherol acetate (100 mg/kg bw) was investigated in sheep. Plasma samples were taken at regular intervals after dosing and assayed for radioactivity and D-alpha-tocopherol. Plasma profiles were modelized using a compartmental approach and the input entry rate (for radioactivity) was identified using a numerical deconvolution method. Based on plasma specific activity (dose 1 and dose 4) and D-alpha-tocopherol plasma concentration (unlabelled alpha-tocopherol acetate), half-lives were not significantly different (99.6 +/- 28.3 h, 121.3 +/- 38.5 h and 75.0 +/- 49.75 h after dose 1, dose 4 and unlabelled alpha-tocopherol respectively). The times to reach maximal plasma concentrations were similar for the 3 test articles (mean values ranging from 21.5 to 26.7 h). The only significant difference was observed for the apparent volume of distribution (with respect to bioavailability) which was much larger for the unlabelled test article. Deconvolution study of plasma specific activity showed: i), that the maximal input rate was reached only after a short delay (3 to 12 h); ii), that half of the activity was absorbed after a delay of 38.13 +/- 16.76 h (dose 1) and 44.3 +/- 6.50 h (dose 4); and iii), that 90% of the activity was absorbed after 151.2 +/- 27.3 h (dose 1) and 162.3 +/- 11.2 (dose 4). It is concluded that the intraperitoneal route is of interest for alpha-tocopherol administration, but that more information is required to determine the exact process of absorption.