304 JOURNAL OF COSMETIC SCIENCE First scan - - •' - K I -o- Ca I Cu I Second scan -+- K II -,- Ca II = Cu II K, Ca 3500 t 12 Cu ppm ppm 3000 - 10 2500 - 1500 ' I - " 0 ..... I ! i J 0 0 5 10 15 20 25 30 position, no Figure 4. Two longitudinal scans from the hair root towards the tip in 0.2-ram increments. Note the substantial variations in K and Ca content above 2 mm from the root (position 10). Cu shows an overall more stable profile, with small variations around l0 pg/g. An example of the influence the hair fiber cross section is given in Figure 3. The data are presented as peak areas normalized to the intensity of the scattered beam, which in turn is proportional to the mass of the analyzed part of the fiber. Ca, ppm 1200 1000 800 600 400 200 -Ca I - - •- - Ca II ---m -Ca III Znl , -Znll - * -Znlll o Measurement, no 5 10 Zn, ppm 700 600 500 400 300 200 100 0 15 Figure 5. The long-term reproducibility test over a period of seven weeks, with analyses in approximately the same region of the hair fiber twice a week, i.e., 14 sets of analyses. Data for Ca and Zn are presented.

ELEMENTAL DISTRIBUTION IN HAIR 305 The variation of elemental content along a fiber for K, Ca, and Cu is presented in Figure 4, which contains both longitudinal scans. (Note the separate scaling for Cu.) The long-term reproducibility test showed a certain variation for all elements and hair fibers. In Figure 5 the results for Ca and Zn are given. Ca shows a variability of about 200 mg/g, whereas Zn varied between about 30 mg/g and 100 mg/g for the three hairs studied. A total of 250 hair fibers from 63 persons were analyzed for the presence of the elements S, Cl, K, Ca, Fe, Cu and Zn. Three or six hairs from each person were analyzed. There is a variation in elemental content between individual hair fibers. In Table I data have been calculated for the whole group. The results from the same population have also been computed from the empirical median value of each element and person (Table II). Age and sex have not been taken into account. For elements present in low concentrations, the statistical variation will influence the data distribution to an extent that is compa- rable to the concentration. DISCUSSION Present-day hair research is mainly concentrated on hair formation and growth, using the tools of modern molecular biology, e.g., immunology, gene analysis, etc. It has been shown that the distribution of particular elements, mobile and bound ions, not to mention the metal ions crucial for enzyme function, can be indicators of disturbances due to diseases specific for the integument (12,19) or as part of a more generalized distur- bance in organs of ectodermal origin, e.g., genetic disorders such as Menke's disease (4). Among scientists, interest is rarely focused on the presence of elements as a reflection of the physiological and pathophysiological events occurring in the hair follicle (10). The reason for this lack of interest appears to be an obvious one: the •, sit• determination of elements in the living tissue of a hair follicle has been hindered by our lack of micro- techniques allowing such studies without gross interference with the living cells. Hence, the studies that have been pursued were based on inert preparation techniques such as freeze fixation or analyses on the already consolidated hair fiber. It has been shown that the human scalp hair fiber is fully consolidated one millimeter above the midbulb level (cf. ref. 8). Table I Elemental Content of Hair Fibers in a Normal Caucasian Population (n = 63), Given in •g/g Element Median Mean +/- 1S.D. Minimum Maximum Unit S 3.6 3.6 0.6 1.2 4.8 % C1 * * 0.2 * 3.6 % K * 108 276 * 2182 I•g/g Ca 196 208 117 * 629 l•g/g Fe 4 6 7 * 41 t•g/g Cu 5 5 4 * 32 •g/g Zn 182 184 23 96 259 •g/g * Data below the detection limit. The data were calculated from individual hair fiber analyses. Empirical median values are expected to give a representation of baseline data, which is less influenced by extreme (high) values.