394 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS water. Finally dissolve in not over 25 ml. of 1 per cent ammonia and separate by ascending chromatography on Whatman No. 2 paper using 1 per cent ammonia saturated with amyl alcohol as a developing agent.2 As the development of the chromatogram proceeds it will be seen that where the original drop was placed there will sometimes appear a brownish coloured spot which represents some of the lake colours stripped from the laking agent by the ammonia and not removed by the subsequent purifica- tion. Immediately above this will be a bluish red spot due to FD&C Red No. 3, if any be present either as such or as a lake. Immediately above this will be a less intensely blue-red spot due to D&C Red No. 21. Above this will be D&C Orange No. 5 and finally D&C Red No. 27. If rhodamine lakes are present, the dye will strip off and may easily be confused with D&C Red No. 27. However, if sufficient time be allowed, the rhodamine will finally outstrip the D&C Red No. 27 and appear as a purplish red spot toward the top of the chromatogram. This procedure will enable the analyst to determine qualitatively which of the bromos are present and relatively how much of each may be estimated by the size of the spots. If the identification of the lake colours seems necessary, they may be stripped and determined by the method of Sclar and Freeman. 8 ANALYSIS OF THE LIPSTICK BASE The total amount of base present will already have been deternfined as outlined in "Total Pigments and Bromos." Digest the base in 95 per cent alcohol by boiling gently for several minutes on the hot plate. Use approximately 200 ml. of alcohol for the base obtained from 4.00 grams of lipstick. During this digestion, the hydrocarbons will appear as a melted bubble on the bottom of the beaker, but almost everything else will dissolve. Allow to cool to room temperature, and filter into a tared beaker. Wash the original beaker and the undissolved waxes on the filter with several portions of cold alcohol. It is best to use a 4-inch funnel and a 17.50 cm. filter paper to facilitate the filtration and washing. The jelled mass of waxes on the funnel should be broken up occasionally with a glass rod during the washing to make sure that all alcohol soluble material is removed at this stage. This is a critical phase of this method, and great care must be exercised to make sure that sufficient attention be given to the washing with cold alcohol or some of the alcohol soluble material may be entrapped in the mass of wax crystals and lost. Little, if any, hard wax will be lost at this point by too much washing. Combine the washings in a tared beaker, dry to constant weight on the steam bath, weigh the residue, and record as the castor oil fraction.

A METHOD FOR THE ANALYSIS OF LIPSTICK 395 Dissolve the residue on the filter into the original tared beaker with boiling chloroform, evaporate to constant weight on the steam bath, weigh and record as the hard wax fraction. The castor oil fraction will contain all of the castor oil, about one-half of the lanolin, about 20 per cent of the candelilla wax, as well as any other alcohol soluble alcohols or esters which may be present. ANALYSIS OF THE CASTOR OIL FRACTION Saponify the castor oil fraction by boiling under a reflux condenser for three hours. Use 50 mi. of 0.50 normal KOH and do not add any additional alcohol more than sufficient to transfer the sample from the original beaker to the saponification flask. When the saponification is finished, transfer the solution to a 250 ml. separatory funnel, rinse the flask with a small amount of alcohol and then with sufficient water to bring the mixture in the funnel to slightly more than 50 per cent water. Add 5.0 ml. concentrated HC1, and extract three times with 25 mi. portions of chloroform. Use each portion of chloroform to clean out the saponification flask by boiling the solvent gently in the flask before adding it to the separ. atory funnel. Combine the chloroform extracts in another funnel, wash with 20 ml. water and add the washings to the original funnel. The aqueous portion will contain all of the glycerin or glycols which were present as esters in the original lipstick. If any methyl esters were present methyl alcohol will also be present. The unsaponifiable matter will contain all of the alcohols released by saponification or present originally, and will include about 25 per cent of the lanolin present in the original lipstick, as lanolin alcohols. Evaporate the chloroform on the steam bath. Transfer the residue to a separatory funnel with alcohol using about 50 mi. Make alkaline with KOH, add about 60 mi. of water, and extract with petroleum ether. Wash the petroleum ether extracts with water, filter through a dry filter paper, evapor- ate to constant weight and report as unsaponifiable matter. Make the remaining solution in the separatory funnel acid with HC1 and again extract with three portions of chloroform using 25 ml. for each extrac- tion. Combine these extracts, evaporate the solvent, weigh and record as fatty acids. The fatty acid fraction should contain nothing but fatty acids which may have come from castor oil, lanolin or any other esters present in the original lipstick. Make the aqueous extract containing the glycerin and glycols freed by saponification up to volume in a volumetric flask. Use an aliquot to deter- mine the glycerin and propylene glycol, again using the method of Shupe.•