EVALUATION OF ANTIPERSPIRANT ACTIVITY Figure 5. Gas supply and regulating system determination of the ratios concerned. In a sense they have been forced to do this by the time required to collect sufficient perspiration for an accurate weight to be made or sufficient data to produce the single point by averaging. The constancy of the ratios cannot be determined by this type of data treatment. The ratio line or constancy of the ratio can be determined conveniently only by the continuous recording of the ratio itself rather than its components. The procedure described in this paper continuously records the ratio changes which result directly in a ratio line rather than a single point. A consequence of this precision is the discovery of several forms of error or deviations from the ideal ratio line. The analysis of these deviations and the understanding of their operation has led to the determination and application of the necessary corrections to obtain valid results. Four types of tests are represented by the lines in Fig. 2. Line 4, passing through tlze origin, is the ideal test along which the ratios of the

756 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Figure 6. Absorption cup. The outer concentric space uses a vacuum to hold the cup to the subject. Moisture is picked up in the center section values from area A to those of area B are equal. This test, however, is quite rare. Line 5, not passing through the origin, represents the more common form of test. It may be displaced to one side or the other of the ideal line. The ratios along this line are not equal. Line 6, another form of test, has the same properties as line 5 but a starting point also above the origin. Again the ratios are not constant along this line. Line 7 is quite rare and represents the only specific case of an individual with different perspiration ratios at different rates. It can also be dis- placed from the origin. The important observation to be made here is that the perspiration ratios of most subjects change along a linear function as illustrated by lines 5 and 6. This fact has been repeatedly demonstrated by the pro- cedure to be described. The fact that the ratios along these lines are not constant is due to an additive factor, the source of which will be explained later. If this additive factor is eliminated, the lines will pass through the origin and become ideal lines. Consequently, the original assumption, that two areas on an individual would retain a constant ratio at different perspiration rates, has been proven correct. The problem has remained with the elimination of experimental errors.