696 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS On this basis we are able to discard those experimental toiletry systems which yield visible and deleterious changes (such as disruption and gross removal of the cuticle) but nevertheless we still need to be able to assess the effects on the fibre surface of much milder and practicable treatments. The present paper highlights some new methods for the scanning electron microscope by means of which we are now able to assess finer change in surface structure with certainty. In addition the paper is con- cerned with the inherent variability in the surface architecture of human hair and how this arises. THE VARIATION IN SURFACE ARCHITECTURE ALONG THE LENGTH OF A NORMAL HUMAN HAIR To study the natural variation in the surface architecture of human hairs along their length we have carried out a detailed scanning electron microscope examination of the hair from a 21-year-old Caucasian woman, who maintains her hair at an average length of about 40 cm and who for the last few years has only applied the normal hair care treatments of brushing, combing (nylon brush and comb) and use of a sha•-T•poo contain- ing a conventional anionic detergent. Samples of hair were carefully plucked from different sites on the scalp. Segments 1 cm long were cut from each hair at intervals of 4 cm (making note of their order) and stuck down to double-sided adhesive tape on a scanning electron microscope mounting stub. The fibre speci•nens were vacuum-coated first with 10 nm carbon and then with 40 nm silver and examined by secondary electron emission in a Cambridge Stereoscan Mk H scanning electron microscope at an accelerating potential of 20 kV. Photo- graphs were obtained in the magnification range from x 200 to x 10 000. In some cases stereo-pair micrographs were obtained and these greatly facilitated the interpretation of the three-dimensional surface architecture of the fibres. In all the hairs examined there was a gradual change in surface structure from root to tip and six easily recognisable stages could be identified. These were: Stage 1 (Fig. 1). Where the hair emerges from the scalp and for a short distance the scale surfaces are relatively smooth and featureless, the scales are imbricate and their free edges are of relatively smooth contour. This appearance is perhaps that which can be accepted for hair in prime condi- tion and free from externally-promoted defects. Stage 2 (Fig. 2). The fibre surface is similar to that of stage 1 except that

Figures 1--4. Hair surface--root to tip, stage 1 (root), stage 2, stage 3 and stage 4 respectively. Facing page 696