2O8 JOURNAL OF THE SOCIETY OF COSMoeTIC CHEMISTS and the animals placed in separate metabolic cages. The animals were killed at various times up to 96 h after application and excreta, blood, skins and protective patches were monitored for tritium. Rats on all treatments were fed on pellets of Spital diet (BOCM/Silcock, Process Development Department, Bromborough, Cheshire) and given water ad libitum. Analysis of biological samples for tritium Urines were made up to 50 ml and 2.0 ml aliquots were counted in 18.0 ml of Triton X-100: Toluene (1: 2, v/v) liquid scintillator containing 5.0 g PPO and 0.2 g POPOP/1. The samples were thoroughly mixed and counted in a Packard 4322 liquid scintillation spectrometer. A channels ratio technique was used to determine the counting efficiency and [1,2-alii n-hexadecane was used as an internal standard. All other aqueous samples (skin rinsings, standards and washings) were monitored in a similar manner. Faeces were freeze-dried and the sublimate was monitored for tritium. Aliquots of up to 350 mg of the residue were combusted in a Packard 305 sample oxidizer to determine the tritium content as tritiated water. Re- coveties of greater than 98•o were recorded from standards with counting efficiencies of up to 25•o. Full depth, 1 cm diameter punch autopsies of frozen skin were monitored for tritium either after combustion in a similar way to the faecal samples or after solubilizing in Soluene (Packard Instruments Ltd) at 40øC overnight and neutralizing with an excess of solid CO,,.. The protective patch was soaked in 50 ml ethanol at room temperature overnight and further extracted with 50 ml ethanol. The extracts were combined and aliquots were counted to determine the tritium content of the patch. Recovery of radio-activity by this method was better than 99•o. Blood tritium levels were determined by cornbusting up to 0.5 ml aliquots of blood in the Packard 305 sample oxidizer. RESULTS Turnover of subcutaneously injected Triclosan The recoveries of tritium in the urine and faeces of female rats during 4 days are presented in Table I, from which it can be seen that 89.2•o of the dose was recovered, some 33•o being present in the urine. The faeces

PERCUTANEOUS ABSORPTION OF TRICLOSAN Table I. Turnover of Triclosan injected subcutaneously 209 Time Tritium recovery (dpm x 10 4) (days) Urine Faeces Total l 17.487 •:5.079 23.210-26.112 40.697:3_8.722 (4) 2 3.5894-1.303 12.1554-4.011 15.7444-6.041 (3) 3 1.126 4- 0.840 2.289 • 1.037 3.415 4-1.149 (2) 4 0.857 4-0.611 1.170 4-0.610 2.027 4-1.021 (2) Total 23.059 38.824 61.883 Four female Colworth-Wistar rats (120 g) were injected with 0.5 ml of a 50% aqueous polyethylene glycol 400 solution containing 69.453 x 10 • dpm [aH] Triclosan. The rats were placed into individual metabolism cages and the excreta collected separately every 24 h for 4 days. The tritium content of the faeces was determined by combustion and scintilla- tion counting and that of the urine by counting directly. always contained the greater proportion of radio-activity throughout the 4 days. Using these data, a semi-logarithmic plot of total excreted tritium against time indicated a biological half-life of approximately 14 h. In the first two days 81.3•o of the dose was excreted, and this figure is used to correct the recoveries in 2 days from rats treated topically with [all] Triclosan in shampoo or aerosol deodorant. The level of tritium in blood of sub- cutaneous-injected rats ranged from the equivalent of 0.2 [tg ml 4 at 0.5 h to a maximum of 5.2 [tg ml 4, at 6 h after injection and decreased over 24 and 48 h to approximately 0.3 [xg ml 4 at 96 h. Application of [all] Triclosan in shampoo Some 855/0 of the [all] Triclosan applied to rat skin at various concentra- tions in the shampoo base was recovered in the rinse water and the patch, with additional smaller amounts remaining in the treated skin (Table II). The amount of [all] Triclosan residing in skin, at 48 h after the shampoo, was proportional to the concentration of [all] Triclosan in the shampoo applied to the skin. The extent of penetration of [all] Triclosan through rat skin washed with shampoo was calculated from the amount of tritium in the excreta during the 2 days after treatment. Penetration of [all] Triclosan through rat skin increased in proportion to the concentration in the shampoo (Table III) but was independent of the duration of contact of the shampoo with the skin for periods up to 20 min before rinsing (Table IV). The concentration