30 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS tion, when other surfactants (sodium lauryl sulphate, ether sulphate, isethionate and non-ionic lauryl triethoxylate) had no effect. After 3 days of application, sodium lauryl sulphate had the greatest effect upon the skin in terms of dryness, scaling and cracking of the stratum corneum erythema and oedema. Correlation of these results are discussed. INTRODUCTION The skin irritation potential of soaps and detergents (hereafter termed surfactants) is an expression of the complex interactions occurring in the various regions of the skin following its exposure to solutions of these compounds. Often, in the skin of surfactant-treated laboratory animals (generally after exaggerated application) one may observe signs of primary cutaneous irritation, a response manifested by drying, scaling or even cracking of the stratum corneum, and oedema and erythema in the dermis. In this paper the complex response of skin has been defined in terms of the various types of skin-surfactant interactions which may occur and may contribute to the overall skin response. Specific laboratory test methods have been devised to study some of these interactions. In particular, the effects upon the stratum corneum have been examined by studying the way in which a series of highly pure model surfactants may denature keratin, a phenomenon which may contribute to the observed superficial skin rough- ness. Also, the extraction of soluble compounds from the stratum corneum by washing with surfactant solutions has been measured and related to the removal of natural moisturizer from the horny layer, an effect which may give rise to a lower water-binding capacity (1) and consequently a lower flexibility (2). The rates of percutaneous absorption (skin penetration) of highly pure radioactively-labelled surfactants have been measured using guinea-pigs in vivo. Skin penetration of surfactants is a prerequisite of their causing a response in the living cells of the epidermis and underlying dermis (3). The interaction of surfactants with the living cells of the epidermis and dermis has been studied indirectly by measuring the ability of highly pure sur- factants to release histamine from rat peritoneal mast cells in vitro. Histamine release from mast cells is an initial reaction in the development of erythema (4). In each of these experimental methods on aspects of skin-surfactant interactions we have obtained evidence that the observed experimental result is related to the chemical structure of the surfactant used.

SKIN IRRITATION POTENTIAL OF SURFACTANTS 31 The overall response of rat skin in vivo to solutions of pure model sur- factants has been assessed macroscopically, and the compounds have been ranked according to their observed irritation potential. It was found that notable differences between the various experimental tests existed which did not correlate with macroscopic results. Explanations for these differences are discussed. METHODS AND RESULTS Determination of sulphydryl groups (SH) liberated from human callus by treatment with surfactant solutions If surfactants in contact with the stratum corneum can cause denatura- tion of the keratin, this may be assessed by measuring the increase in SH groups as the proteins unfold. The procedure of Harrold (5) was followed, with little modification, in which liberated SH was determined with 1-(4-p- chloromercuri- phenylazo)- naphthol- 2. Human callus (obtained from a chiropodist) was powdered and used with a series of pure surfactants all of which possessed the lauryl (12-carbon) lipophilic chain but with a variety of hydrophilic head groups. Two concentrations of each were studied (1 mta and 10 mta), the lower being at or below the determined critical micelle concentration of the surfactants (see Tables V! and VII). The results shown in Table ! revealed that all of the anionic surfactants liberated more sulphydryl than water alone, but generally only at 10 mta. The three alcohol ethoxylate non-ionic surfactants did not denature the keratin. The results obtained were very similar to those reported by Harrold in that sodium lauryl sulphate, dodecyl benzene sulphonate, soap and non-ionics (in that order) possessed decreasing abilities to liberate sulphydryl. Also, as reported by Harrold, they had very little or no effect below the critical micelle con- centration. From these data it is clear that the sulphate or sulphonate moiety of the surfactants imparts upon the molecules the ability to unfold keratin. In a second experiment we examined a series (C•o-C•0) of sodium alkyl carboxylates (soaps), alkyl isethionates and alkyl sulphates. In each instance considerably more SH was liberated from the keratin than by water. For all three groups of surfactants the C• and Cx4 chain homologues showed maximum activity, but, nevertheless, it seems that the anionic head groups are the principle cause of keratin denaturation by surfactants.