226 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Table V Fiber Diameter Subject Range of Diameter Ratio _Average Diameter Standard Deviation Fiber Diameter Maximum/Minimum •m 1 82.9 4.8 88.9 72.4 1.23 2 79.6 8.6 92.4 58.9 1.57 3 59.6 5.9 68.8 45.3 1.52 4 70.1 4.4 79.4 63.2 1.26 5 96.5 7.4 111.3 86.6 1.29 6 58.9 7.8 75.1 44.0 1.71 7 68.8 8.9 85.5 52.4 1.63 8 61.4 10.4 77.0 45.6 1.69 9 80.6 6.9 96.5 66.7 1.45 10 75.2 5.2 86.0 61.7 1,39 11 67.4 10.0 82.3 42.4 1.94 12 84.9 6.9 96.9 73.8 1.31 13 71.4 7.2 90.8 58.7 1.55 14 69.4 7.8 86.9 53.0 1.64 15 75.2 8.8 89.8 51.1 1.76 16 70.8 10.0 99.8 53,3 1.87 17 68.4 5.4 80.5 57.2 1.41 18 71.6 5.5 81.2 54.0 1.50 19 68.6 8.1 78.2 37.6 2.08 20 67.0 9.3 83.0 52.4 1.58 Average a 73.4 Average b 71.1 Average c 86.3 aAverage for the 20 subjects. hAverage for the 17 Caucasian subjects. CAverage for the 3 Oriental subjects. other in any combination of these factors. Again, the extreme individual variability .i!.,.: needs to be pointed out. Subject 8 had nearly exclusively simple hairs, while for others more than half the fibers were compounded. ß FIBER DIAMETER Most literature references provide only mean diameters for a given population. cording to these, Caucasian hair diameters average less than 60/.•m (8), more than 60? •m (9), about 77 •m (6), and 79 •m (7). For Oriental hair, the averages range from about 85 to 100 •m (6, 7, 9). Barman and coworkers (10, 11, 12, 13, 14, 15, 16) segJf•i'i mented the hair population by size, but only by crude categories: about 100, about and 25/.•m. Our interest centered more on the variations among individuals and among !:.i! fibers on the same head. The results, concerning the calculated "idealized" diameters, are given in Table V. The mean diameter for the Caucasian group was 71 /xm, which is within the range of earlier data. This is well differentiated from the 86/xm average value of the Orientals.

FEMALE SCALP 227 The average diameters within the Caucasian group range from just under 60 to 85/am in a natural distribution pattern. The correlation between mean diameter and age is only 0.45 for the whole group that is, of no significance. With the exclusion of the Orientals, the correlation approaches zero. Therefore, the age, at least between the midtwenties and early forties, is not a governing factor for hair diameter within a somewhat varied population. Nor was the diameter found to correlate with the number of fibers, sites, or any specific group and compound size. The variation in fiber size on individual heads is shown in Table V. The standard deviation ranges from 5 to 17 per cent, but independently from any other characteristics. The range of absolute diameter of individual fibers is rather wide from 37 to 111/am, and it would certainly increase with larger data bases. Nonetheless, the findings of Barman and coworkers (10, 11, 12, 13, 14, 15, 16) of 10 to 15 per cent fiber content in the 25/am range cannot be substan- tiated with the present results, as we found only one fiber below 40/am. The occur- rence of fibers below 40/am diameter is very low in commercially available human hair as well when determined both by weight and microscopic methods. Even in wool, a 25 /•m fiber is classified as medium fine. The range of diameters on individual panelists, as expressed by the thick-thin ratio, is more restricted. On some subjects the fibers are closely homogeneous with a ratio of less than 1.3. Only in one case did the value exceed 2.0. This thick-thin ratio shows a weak positive correlation (0.49) with fiber population density and similarly weak, but negative correlation with mean fiber diameter and age of the subjects. However, the exclusion of the Orientals from the data base eliminates all significance from these already weak correlations. VISUAL APPEARANCE OF THE HAIR The hair of the panelists differed in a number of characteristics--length, curl, style, cosmetic modifications, and product use--in addition to the measured fiber population density and diameter values. Due to this large number of variables, strict conclusions cannot be drawn concerning the visual bulkiness of hair as a function of the measured parameters on the basis of 20 panelists. Nonetheless, the indications were clear that the fiber diameter had a stronger influence on the visible bulk and elevation of the hair over the head than the population density. Even the curl level and hair setting seemed to be more important factors than the fiber number. Only when a number of these characteristics--low density and diameter with straight configuration--happened to combine, was the total hair mass judged to be very small and weak by visual methods, as in the case of subject 3. When the hair was long--waist length--and very straight, as for Subject 19, a somewhat similar judgment was arrived at, in spite of the very high fiber number and average diameter. On the other hand, wavy hair--not necessarily natural wave--with average to high diameter can show up with high bulk even at low to very low fiber densities, as were seen on panelists 1, 2, 4, and 5. Subject 12 showed the highest bulkiness with naturally strongly waved thick hair, even though the density was only average and the hair reached below shoulder level. CONCLUSIONS Based on the population density, the number of fibers on a full adult head of hair may have to be revised upward by a considerable amount to ].50,000 to 200,000. The indi-