92 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS

HUMAN SKIN SURFACE LIPIDS midpoint in a numerical series--is not affected by exceptionally high or low values. As is apparent from Graph I,t as well as from Table 1 the medians of the acid number for symmetrical sites are virtually identical. The wide ranges tabulated for each of the test areas are due to the fact that for any given site the values differed considerably among the subjects. The values are, moreover, by no means equal for our different test areas. The lipids from the three sites of the forehead, for example, have a lower acid number than those collected from mid upper chest, forearms or soles. For the soles the values are higher than for any of the other areas tested. Furthermore, a tendency is apparent for the values of some of' the areas to rise during the summer season. Only for the mid upper back an appreciable trend in the opposite direction seems to be apparent, but to date the number of subjects and assays is smaller for this site than for any other. t The Graphs I and II and the results presented in Tables 1 and 2 were submitted for pub- lication also to the Editors of "Der Hautarzt." GRAPH REGIONAL DISTRIBUTION AND SEASONAL VARIATION OF SPREADING INDEX cm2/mcgm (MEDIAN VALUES) 2.4 2.0 1..4 • •• 1.9 2.3 2.6 2 2 * * * ') 2.2 2 8 UPPER •IGURES: M•di•s of $pr•din• Index for LE•: Medians for - •pril T•mp. Nor. LOWER (o•) _ Median: 41 R•n•: 4 - 74 RIGHT: Medians 2.93'44.0 for May - •t. LOWER Temp. (øF) - Median: 69 Range: 33 - 90