54 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS During both control periods, the subjects refrained from washing their skins with the exception of the cutaneous areas of their faces and hands. On each day of the test period, the soap under test was used in a tub bath during which both the manner of use of the soap and the time of bathing were super- vised by a laboratory assistant (5). The amount of soap used in each bath was determined as the loss in dry weight of a fresh cake of soap before and after the bath. Through- out the test period, a daily tub bath with the soap under test was the only method of cleansing the cu- taneous surfaces of the subjects' bodies with the exception of their faces and hands. Samples of perspiration were col- lected at intervals of four hours after the baths on all days of each week with the exception of Satur- days and Sundays. On Mondays of the first week of the test periods, the subjects did not take baths. Results of analyses of samples of perspiration collected on these Mon- days were added to comparable results for samples of the preceding seven days in calculations of average results for the first control period. Intensities of odors which are summarized graphically in Chart IV were recorded as dilutions with air and not as pO values. Among the 15 subjects, averages of intensities of odors of six daily samples of per- spiration varied from 32 to 128 with average deviations ranging from +4 to +8. The average ofintensi- tie.• of perspiratory odors deter- mined during the first control period for any one subject was adopted as a standard and comparable intensities of odors determined during the test and second control periods were calculated as percentage fractions of this standard. One week intervened between the experiment with the plain soap and the experiment with the germicidal soap. Approximately one-half of the subjects participated in the two comparative experiments in the order stated in the preceding sen- tence but this sequence was reversed for the remaining subjects. Chart IV presents a graphic summary of the changes in the average odor-producing capacities of samples of perspiration collected from subjects during test periods, at intervals of four hours after baths with either soap (i.e., between 2 and $ p.m.), and, during the second control period, at corresponding times on each of the last four days of the second control period. Curves in Chart IV indicate pro- gressive decreases in odor-producing capacities of perspiration through- out periods of daily tub baths with either plain or germicidal soap. Comparative rates of declination of the curves give evidences of signifi- cantly more rapid reductions in potentialities for development of perspiratory odors resulting from baths with the germicidal soap than those determined following compa- rable cleansings of the subjects' skins with the product lacking bactericidal properties. It will be noted that four consecutive daily

TESTING DEODORANTS WITH CHLOROPHYLL AND DERIVATIVES 55 baths with the former soap effected reductions in perspiratory odors which were approximately equiv- alent to the final result achieved after eleven days of baths with the plain soap. Inasmuch as the soaps had shown equivalent detergent efficiencies, it appeared probable that the differ- ences between the two soaps, in respect to capacities to inhibit developments of perspiratory odors, were dependent upon variations in their skin-degerming actions. This hypothesis was confirmed by the results of tests of the comparative effects of baths with either soap upon viable, cutaneous bacteria. These experimental findings are summarized in Chart V. In view of the fact that the primary objective of this series of experiments was the evaluation of the comparative inhibitory effects of the two soaps upon the mech- anisms involved in productions of obnoxious perspiratory odors, deter- minations were made of the bac- terial populations of all samples of perspiration rather than utilizing direct methods of preparations of cultures of micro-organisms sur- viving on washed areas of skin. In fact, in experimental procedures in which extensive areas of skin are involved, washings with soap and water or samplings by means of per- spiration yield more reliable criteria of the residual bacterial flora of skin than do the direct methods of culturing cutaneous bacteria. Tech- niques adopted in preparations of either direct or indirect cultures of cutaneous bacteria have been described in earlier publications (5 and 23). In the construction of Chart V, the scheme of plotting experimental data during test and second control periods was the same as that which had been used in the preparation of Chart IV. It will be noted that, during the first four days of the test periods, no significant changes in numbers of viable micro-organisms were demon- strable in samples of perspiration collected four hours after tub baths with the plain soap. However, continuation of these daily baths through the last five of the eleven days of the test period did effect delayed but progressive declines in numbers of bacteria sampled from the skin in perspiration. The final result was an average reduction of about 50 per cent below the average for the first control period. On the other hand, baths with the germicidal soap effected prompt de- creases in bacterial populations of perspiration. In contrast with the lack of any indications of anti- bacterial actions of the plain soap, a decrease of 70 per cent in numbers of viable bacteria in perspiration was observed as the average result of the first baths with the germicidal soap. Throughout the second week of the test period, numbers of viable bacteria sampled from skin following washings with the germicidal soap, were maintained at levels varying from 74 to 85 with a mean of 80 per cent below the average for the first control period.