772 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS ß Oand'•ruff Non-d•ndruff Figure 2. ' Comparison of effects of Selsun in dandruff and nondandruff subjects. count is the control value The Conti group, while below the normal value, was not as low as the terminal value in the nondandruff group. Thymidine Radioautography The scalps of nondandruff subjects were shampooed with Selsun every 4 days for 6 times. The labeling index was determined before and after treatment for both the epidermis and the external root sheath of the upper keratinizing portion of the follicular epithelium, the infundikulum. Figure 3 demonstrates that Selsun reduced the labeling index of both epithelia a little more than 50%. It is noteworthy that the values for the epidermis and external root sheaths were almost exactly alike both before and after treatment. DISCUSSION These findings are informative with regard to the fundamental aber- ration in dandruff as well as the mode of action of antidandruff medica-

SELENIUM SULFIDE ON EPIDERMAL TURNOVER 773 Epidermis Ext. RootSheath Epidermis Ext. RootSheath Pre-P• Pre-• Pos1-P• Post-i• Figure 3. Effect of shampooing every 4 days for 6 times on the turnover of the epidermal and follicular epithelium, as expressed by the Ha-thymidine labeling index. Shown are the means and one standard deviation from the means tions. They support the thesis that excess scaling reflects increased epidermopoesis resulting in an overproduction of horny cells. The de- fect would appear to be a functional one in the rate-regulating control of mitosis. By light microscopy there are no inflammatory dermal changes nor even histologic alterations in the Malpighian layer. The increased flow of cells to the surface ends in a defective horny layer in which the cornified cells become improperly patterned and separated by fissures and crevices. The corneocyte count necessarily reflects the proliferative activity of the germinative cells of the epidermis. Because it is so convenient and exempts biopsy, one can tolerate some inherent variability mainly due to loss of an unknown quantity of scales during the 4-day period there is also a sampling error because the quantity produced is not the same all over the scalp. The dandruff process is often patchy. Counts of about 1,000,000 cells/cm •' or more are characteristic of heavy dandruff. Regular shampooing every 4 days with a nonmedicated shampoo may tend to lower this somewhat but a steep decline never occurs. With an agent as active as Selsun, however, the count declined