DETERMINATION OF CHLORINATED PHENOLS • o,: o , , • • i , 40 3 8 3 6 34 32 30 WAVELENGTH K ½•'l• I Figure 2 Spectra of N/10 methanolic HC1 extracts of powdered deodorants. The spectra shown in Fig. 2 clearly showed that under the conditions employed for the analysis, the acidity was sufficient to prevent formation of a phenate ion. Under other conditions the varying degree of alkalinity in a product would have produced variable amounts of phenate ion and hence non-reproducible analytical results. The use of an acidic solvent prevented the gelatinous precipitation of aluminium hydroxide, the only insoluble material being talc which was removed by centrifuging. The completeness of extraction of the phenols by the described procedure was shown by the absence of absorptions in the wavelength range 32.0 to 40 Kcm -• for a fourth extract. The use of a tangent base-line to determine the corrected absorbances at 33.8 and 35.8 Kcm -• allowed for extraneous background absorptions due to other components of the product. This can be seen by comparing Figs. 7 and 2. For a product which has been shown to contain a chloroxylenol, but

JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS the identity of which is unknown, the value of x E35.8 can be taken as 100 to determine the concentration of the chloroxylenol. This procedure will only introduce a serious error if the component happens to be 4-chloro- 2:5-dimethyl or 6-chloro-3:4-dimethylphenol. (Received: 29th June 7965) REFERENCES (1) Coggeshall, N. D. and Glesser, A. S. J. Am. Chem. Soc. 71 3150 (1949). (2) Clements, J. E. and Newburger, S. H. J. Assoc. O•c. Agr. Chemists 87 190 (1954). (3) Elvidge, D. A. and Peutrell, B. J. Pharm. Pharmacol. 18 Suppl. lilt (1961).