106 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS The results obtained •vhen alcohols were used for vehicles are shown in Fig. 14. Ethyl alcohol gave significantly higher blood levels than propylene glycol. Adding 35 per cent glycerol had no significant effect. To test the relative efficiency of water-in-oil rs. oil-in-water emulsions, our Develop- ment Laboratory at Bristol-Mye. rs prepared two emulsions, which contained. 60 per cent H20 and were identical in every detail except for the emulsifier which in one case was hydrophobic (water-in-oil) and in the other hydro- IlO 90 70 J 5O =t, 3o I•ABS. ETHANOL i i - •5%ABS. ETHANOL - / • •5%GLYCER- i i - I t / PROPYLENE I t • GLYCOL -- t -- I // i I 2 TIME (HOURS) Fig. 14.--Percutaneous absorption of 10% methyl salicylate in various alcohols. IO0 80 - 60 - 40 - 20 - o // 0 o WATER-IN-OIL - / ! l/! TIME (HOURS) Fig. 15.--Percutaneous absorption of 10% methyl salicylate in oil-in-water and water-in-oil vehicles.

THE PERCUTANEOUS ABSORPTION OF SALICYLATES 107 philic (oil-in-water). Figure 15 indicates no significant difference in the plasma levels produced with these bases. To compare all of the various types of vehicles a typical one from each group was selected and compared (Fig. 16). All of the curves are significantly different from one another. 'EFFECT OF THE SUBSTITUENT GROUP Volatile vehicles, usually ether were used when different derivatives ot Q OIL-•N-HaO U m / ,/ / • LIGHT ,' / '/ • LANOLIN I•/• .--•• GLYCOL MIX 15 3o I 2 (MIN) TIME (HOURS) Fig, 16.•Percutaneous absorption of 10• methyl salicylate in •arious type• of vehicles. 50 i i EMETHYL + E _-•N-BUTYL + E 5 THYL + E //•••--iso PROPYL+ E _ a. 40 ,,,• 50 ß J .•/ / •1 SO AMY L -I- E ! / • •.•'•.'-'AMYL + E 13:: U 20 ! // • ••BENZYL +E "'2 If! , .• •••'-GLYCO + E / o' !•?•, __= .. =_ I i _DiETHYLAMiNE.iH• j •, 0 , ' I 15 $0 I 2 $ H•O (MIN) .,-.M. ABS. ETHANOL TIME ( HOURS ) E ETHER Fig. 17.--Percutaneous absorption of different salicylate esters and salts in vola- tile vehicles.