ANALYSIS OF SUNSCREENS 343 "equivalent to so much of" the uv absorber whose standard curve is ac- tually used for the interpretation. Figure 6 is an example chosen from our files, showing evalua- tion of different commercial products using various concentrations of screening agent. For calculation of the transmitted tanning energy, the process is ex- actly the same as described above for plotting the erythemal curve, al- though it should be noted that in the border area of middle to near ultra- violet (320 mt• and 337.5 mt•) this band of radiation causes erythemal as well as tanning effects, so that this particular area has to be evaluated for its dual value as far as human skin is concerned. In other words, to cal- culate "total erythemal ener•o3r transmitted," we temporarily ignore the fact that four 5-mt• bands actually have dual value (contributing both erythemal and tanning effects), and simply calculate as if the effect was totally erythemal. Then, when "total tanning radiation" is calculated, the same 20-mt• portion of the absorbance curve is calculated for its con- tribution to this curve. "Incident Erythemal" values are thereby conveniently grouped into one column, while the values for "Incident Tan" are grouped in another, to avoid any possible confusion (work sheets, Fig. 6, right). The tanning curves are usually less precise and therefore less reliable than transmitted erythemal curves, which are about 95% accurate. It is desirable for the analytical chemist to have available as many "standard" evaluation curves as possible (Fig. 7), not only to see how much of a particular screen is present in the sample, but also to compare its "equivalency" to other screens. Commercial suntan preparations on the market today fall into four groups, insofar as concentration is concerned: For fast tanning For normal skin For sensitive skin For extra sensitive skin transmitting approximately 15% of incident erythemal radiation transmitting from 4-8% of the incident erythemal radiation transmitting from 1-4% of the incident erythemal radiation transmitting under 1% of erythemal radia- tion (sometimes called "Total Block") As a consequence, each tested screen must be carefully diluted until one is sure (by hand scanning) that the lambda maximum is well posi-

•44 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS (•) H OMOME•ITHyI. SALICYLATE I"- AMINOBENZOATE Figure 7. Comparative evaluation curves of four most widely used screens tioned on the graph. Then one can proceed to make the spectrogram, having noted the final dilution in grams per liter. SUMMARY This procedure gives a reliable evaluation of the amount of the pro- tection given by the screen in question. It deals with variables such as: the position of the maximum absorbance of the screen, in relation to a narrow band of nonuniform erythemal radiation, varying shape of the absorbing curve, and unknown concentration of the screen. Considering that the spectrophotometric work is done on solutions diluted to contain from 4 to 5 ppm of the sunscreen, the reproducibility of approximately 95% is very good. ACKNOWLEDGMENT The author would like to express thanks to Donald E. Conner, Direc- tor of Research, Van Dyk g: Co., and to Robert L. Goldemberg, Director of Technical Services, Van Dyk • Co., for consultation advice in prepar- ing this paper. (Received November 7, 1971)