SUNLIGHT AND PRETREATED HAIR 85 10 4 8 6 ! /. /. / / /m dyed ß ß .$ permed , ß ß ß bleached unfreofed Irrodiofion lime in weeks Figure 4. Shift in color on the blue yellow axis of untreated, permed, dyed, and bleached human hair after irradiation with VIS. change in tensile strength caused by cosmetic treatments is detectable in permed hair. Since during the permanent-wave process a complete reoxidation of the reductively cleaved disulfide cross-links is not achieved, decreasing values in tensile strength support Table II Modification of the Breaking Tenacity of Untreated and Cosmetically Treated Human Hair After Irradiation With Individual Parts of Sunlight for Four Weeks Untreated Permed Bleached Dyed cN/tex V cN/tex V cN/tex V cN/tex V Non-irradiated 13.0 9.3 10.6' 10.5 13.3 13.5 12.8 10.1 UV-B 12.9 6.8 9.6 8.2 10.2' 10.1 11.2' 8.4 UV-A 8.6* 9.6 7.7* 7.1 5.1' 17.0 8.1' 7.2 VIS 10.1' 11.3 10.4' 13.7 9.9* 18.0 11.2' 8.6 IR 12.1 6.5 10.2 5.1 11.3' 13.9 11.7 11.9 V = coefficient of variation in %. * = statistically certain differences between the mean values and the initial value of the non-irradiated sample, with 95% certainty. * = statistically certain differences between the mean values and the initial value of the sample not cosmetically treated, with 95% certainty.

86 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS the indications made by Robbins (9) about a correlation between tensile strength and the number of cleaved disulfide cross-links (Table II, line 1: non-irradiated, untreated -- 13.0 cN, non-irradiated, permed = 10.6 cN). Additional irradiation, especially with UV-A, leads to a loss in tensile strength in the case of untreated as well as perreed, bleached, or oxidatively dyed hair. Furthermore, a minor effect of visible light on untreated and treated hair is observed (loss in tensile strength by 2.9 cN/tex). The extent of photochemically caused losses in tensile strength is highest for bleached hair thus, it is reduced by UV-A by ca. 62%, and by UV-B and VIS by ca. 25% in comparison to untreated hair. Regarding the influence of dyeing or permanent waving on photooxidative reduction of tensile strength, only a trend towards modification is, in comparison with untreated hair, detectable, but it cannot be mea- sured with statistical significance. TENSILE ELONGATION The permanent-wave process causes a trend towards reduction of the elongation prop- erties from 46.5% for untreated hair to 44% for treated hair (Table III, line 1: non- irradiated). The chemical bleaching process decreases the elongation of the hair fiber significantly from 46.5 to 42.4%. According to Alexander et aL, this effect is put down to an oxidative destruction of cystine (10). The oxidized cystine residues are more readily accessible to water molecules. An increase in water absorption follows a plasticizing effect on the fiber, which results in an increased elongation. These effects are comparable to the decrease in stiffness of swollen fibers (11). UV-B irradiation reduces the elongation of permed hair significantly, while that of untreated, bleached or dyed hair, however, shows tendencies (Table III, line 2: UV-B). UV-A irradiation reduces the elongation properties of untreated as well as of cosmeti- cally treated hair the most, compared with the other ranges of sunlight (Table III, line 3: UV-A). The extent of the photochemically caused loss in elongation by VIS is somewhat less and Table III Modification of the Breaking Elongation of Untreated and Cosmetically Treated Human Hair After Irradiation With Individual Parts of Sunlight for Four Weeks Untreated Perreed Bleached Dyed Din % V Din % V D in % V Din% V Non-irradiated 46.5 5.0 44.0 4.3 42.4* 3.2 43.9 6.3 UV-B 42.8 4.8 41.0' 10.1 41.4 10.1 40.8 10.1 UV-A 38.5* 11.7 39.0* 9.1 36.0* 7.6 38.5* 7.7 VIS 38.7* 24.8 43.6 13.6 36.6* 10.7 39.4* 9.3 IR 45.9 6.0 45.4 4.4 44.1 5.6 40.8 7.4 D = breaking elongation. V = coefficient of variation in %. * = statistically certain differences between the mean values and the initial value of the non-irradiated sample, with 95% certainty. * = statistically certain differences between the mean values and the initial value of the sample not cosmetically treated, with 95% certainty.