262 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS be a 4, on four of the subjects with the product 6, and on two of the subjects with the product an 8, as well as on the 8% HMS control (SPF = 4) (C) an SPF -- 6, 8, or 15 range-finding assay on one subject with the SPF = 6 product, on three of the subjects with the product known to be an 8, and on all five subjects with the SPF = 15 product. The observer who read the irradiated sites had no knowledge of the labelled value of the products. RESULTS Table I presents the SPF values achieved, based on the division of the subjects' protected MED value by the unprotected MED value. All products, tested blind, achieved their proper SPF values with this range-finding technique, except for the SPF = 15 product which is, however, in its own proper category. As a footnote of interest, it should be noted that the marketed product with a value SPF = 15, which did not achieve its true SPF value with only five subjects, achieved an 8.43% Standard Error of the Mean %, well above the Monograph's (1) allowable _+ 5.0%. These data reinforce the fact that individual biological variation (both inter- and intra-subject) requires that the Sun Protection Factor assay be performed on twenty subjects [See Monograph (1)]. CONCLUSION With a change of template shape and selection of subjects with suitable back area, few but sufficient additional irradiated subsites can be added to the conventional 4, 5, or 6 test subsites to achieve an efficient range-finding technique. As few as five subjects may reveal the probable expected SPF, which can be then tested for in the conventional manner. REFERENCES (1) The Food and Drug Administration, Proposed monograph for otc sunscreen drug products, 43, 166, 37206-38629 (August 1978). (2) D. S. Berger, Specification and design of solar ultraviolet simu/ators, J. of Invest. Dermatolo., 192-199 (1%9).