764 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS swelling of the tissues in quarternary ammonium cationic agents such as Cetab, which are known to interact with proteins and are also irritant in vivo (196). The value of the collagen swelling technique in vitro as an indication of the irritancy of a particular substance to the skin in vivo is open to specula- tion. It would appear, however, that the technique may provide some evidence of the ability of a particular substance to interact and denature fibrous proteins. Since skin irritation would appear to result from denatura- tion of the fibrous protein-keratin, it is probable that the collagen swelling technique may prove useful in predicting strong skin irritants. Skin reflectance Absorption studies of skin irradiated with ir light, have indicated the possibility that spectroscopic methods may be of value in detecting the presence of foreign substances present on the surface of the skin, and on the general condition of the keratin layer (197). Absorption peaks at particular wavelengths have been correlated with the presence of moisture, carbonyl groups, polypeptide amides and absorptions due to CH2 and CHa groups (198, 199). The absorption peaks have been shown to vary according to the composition of the stratum corneum, and the presence of sweat, sebum or materials of extraneous origin. For example, after washing and drying the skin thoroughly, peaks of absorption corresponding to water were diminished. In psoriatic skins, water peaks were diminished as would be expected from regions where sweating is impaired. The technique is considered to be attractive since it may be readily carried out in vivo and imposes no serious artefacts. However, further studies would appear to be warranted to establish the value of the technique in detecting epidermal damage which is not visible to the naked eye and cannot be ascertained using simpler techniques employing less sophisticated appara- tus. At present, the interpretation of changes in the absorption properties of the skin in terms of tissue damage is not clear and caution would seem to be necessary in interpreting the results of skin reflectance studies. SUMMARY AND GENERAL CONCLUSIONS Investigations of the skin irritancy of substances intended for inclusion in cosmetics, should be carried out firstly in animals and subsequently in man. The tests should be designed to take into account the type or'

APPRAISAL OF METHODS FOR DETECTING PRIMARY SKIN IRRITANTS 765 substance under investigation and the nature of the cosmetic in which it will be included. No single animal species is entirely reliable for predicting the irritancy of all substances to man but a retrospective survey of the literature indicates that the rabbit has a skin which is sensitive to most substances likely to irritate human skin. However, other species should be used in addition to the rabbit in routine studies for primary irritants. Although no single schedule of testing is completely reliable for detecting irritants, the tests most frequently employed in animal and human investiga- tions are similar in detail. They are essentially of two types. The first is the patch test, which is a biological test designed to show whether a substance is irritant and under what conditions. This knowledge may then serve as a basis for conducting secondary tests to ascertain the mechanism by which the substance exerts its irritant action. The patch test as originally described, specified that substances should be tested only under occlusive conditions, with the skin reactions being subjectively evaluated. To be of greater value, tests should be carried out using both open and occlusive conditions and areas of intact and abraded skin. To overcome the difficulty of detecting minor degrees of damage not visible to the naked eye, histology and histochemistry have an obvious advantage, particularly in animal tests. However, in human studies, where this type of investigation is not usually practicable, an examination of successive layers of keratinized cells removed by adhesive tape stripping, is considered to be informative. The value of methods for exaggerating the sensitivity of animal skins by chemical and mechanical means and other modifications of the patch test have been discussed in detail. Following the initial animal tests and the subsequent trials with human volunteers, both with individual chemicals and definitive cosmetic formula- tions as they are intended for marketing, 'in-use' tests with panels of volunteers are essential. These tests cannot be standardized, but should be realistic for the particular cosmetic. The test should take the form of applications of the cosmetic to the skin under its exact recommended con- ditions of use, for a given period. Adverse reactions should be recorded as and when they occur and the results analysed statistically. A number of secondary tests for investigating the action of topically applied substances on the character of the skin, its biochemistry, electrical conductivity, mechanical properties and chemical constitution have been discussed. Also reference has been made to tests designed to investigate the functional capacity of the skin appendages. The conclusion reached