766 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS 100 ß 6O Pre-treat- Treatment Period Post - treatment Period ment Period ' ' • ' •o 6'0 I o 20 30 40 DAYS Figure 1.--Average scurf weights in mg. of all nine subjects. 8000 l 7000 6000 2000 i I i 1ooo 'treatment !l I Treat .... P .... d I Perrod ! Post- treatment Period •0 20 $0 40 50 60 7'0 DAYS 5000 4000 Figure 2.--Average plate counts per mg. of scurf sample for all nine subjects.

REI.ATIONSHIP BETWEEN DANDRUFF AND MICROBIAL FLORA OF THE SCAI,P 767 medicated shampoo was used, and physical removal of scurf and bacteria resulted. However, the rapid increase in scurf and bacterial values elnpha- sizes the short term activity of non-medicated shampoos in controlling dandruff. A comparison of scurf weights and bacterial counts recorded during the treatment period indicates that, although the microbial flora can be elimi- nated very quickly, the decrease in the amount of scurf is a gradual process. Following the almost complete elimination of a microbial flora it required about three weeks before a general leveling off of scurf values was observed. It was obvious at this point that despite continued use of the antimicrobial agent it was impossible to obtain any further reduction in the weight of the scurf samples. These levels may be referred to as "threshold levels," and they differed markedly depending upon the subject. Threshold levels for each subject are listed in Table III. They represent the averages of the weights of scurf samples removed during the time when the antimicrobial agent was obviously suppressing scurf formation. In general, high threshold values were obtained in those subjects who showed high values during the pre-treatment period. The significance of such a variation among individuals (13 to 95 mg.) would seem to be that factors other than the presence of a microbial flora influence the rate and degree of epidermal desquamation of the scalp. Again referring to Figs. 1 and 2, it is apparent that, when the treatment was stopped, the bacterial counts began to rise almost immediately. After 20 days the bacterial counts for most individuals had returned to pre-treat- ment levels. There was much variation in post-treatment counts among individuals. In most instances, however, scurf weight values continued to decrease despite the rise in bacterial counts. This trend continued for about ten days before the scurf weights began to increase. Scurf weight values comparable with those of the pre-treatment period were obtained after three weeks. It appears that the elimination of the microbial flora does not result in an immediate response in scurf reduction and likewise, the rapid return of a microbial flora does not result in an immediate increase in scurf production. It may be assumed that the presence of a microbial flora promotes scurf formation but is not the physiological cause of it. Table IV shows very clearly the important influence of a scalp microbial flora upon the formation of scalp desquamative tissue. The weight of the heaviest scurf sample for each subject removed during the pre-treatment period is compared with the weight of the lightest sample removed during the treatment and post-treatment periods. SUMMARY AND CONCLUSIONS When the human scalp is given treatment with a broad spectrum anti-