DETERMINATION OF TRACE COMPONENTS IN DYE 553 Tri-Sil/BSA.* The solution should be stored under refrigeration in rubber- capped serum vials. Pthalic acid-diethylaminophenol standard was prepared by dissolving 50 mg of phthalic acid and 20 mg of m-diethylaminophenol in sufficient anhy- drous methanol to make 100.0 ml. This was stored in a ½ool dark place. o- (2-Hydroxy-4-diethylaminobenzoyl) benzoic acid (HDBA) was synthe- sized by the method of Ritchie et al. (4). Triethylrhodamine was also synthesized by the method of Bitehie et al. (4). Theory: C, 75.33% H, 6.32% N, 6.75%. Found: C, 74.41% H, 6.15%, N, 6.88%. Methods Determination of Phthalic Acid and Diethylaminophenol Into a silylation vial, 1.0 ml of standard solution of phthalie acid and diethylaminophenol was pipetted, then evaporated to dryness in a stream of nitrogen. Next, 100- 0.5 mg of sample was weighed into this vial. Into an- other vial, 100 + 0.5 mg of sample was weighed. Both vials were capped with septurns and 1.0 ml of silylating mixture was injected into each, followed by heating for 10 rain at 60øC. Conditions for the gas-liquid chromatography were: sample size, 4.0 /•1 helium flow rate, 50 ml/min injection port and flame ionization detector temperature, 230øC temperature, 150-250øC at 6ø/min, and held at 250øC until other volatile components were eluted (about 15 min). Determination of HDBA Two grams of D and C Bed No. 37 was dissolved in sufficient methanol to make 100.0 ml. A standard solution of HDBA containing 10.0% mg in metha- nol was prepared. Standard and sample (0.50 ml) were applied to 18-cm strips, 2.0 cm from the bottom of two LQIF tlc plates. (The bottom portion of the plates is of different composition from the rest of the plate and may readily be distinguished. ) If the band of dye is kept within 0.5 cm wide, no difficul- ties in migration are encountered. The plates were migrated 12.0 cm with 10.0% methanol in chloroform. The HDBA is detected as a dark stripe i cm above the broadest red band (Rhodamine) when viewed under filtered ultraviolet light. The HDBA was then scraped off without loss into a 100-ml beaker. The spot collector was packed with sufficient fine glass wool above and be- low the sintered glass disk to prevent loss of fines. The bulk of the silica in the beaker was next transferred to the spot collector by the use of vacuum. Methanol (25 ml) was placed into the beaker and aspirated through the silica * Cat. No. 490,107, Pierce Chemical Co., Rockford, Ill.

554 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS in the spot collector apparatus, followed by evaporation of the methanol ex- tract to less than 6 ml in a stream of nitrogen. The residue was transferred quantitatively to a 10-ml volumetric flask, diluted to volume, and read in a 1-era cell from 450 to 280 nm against a methanol blank. Absorbance was de- termined at 350 nm by the baseline correction method as shown in Fig. 1. The absorbance was taken as the distance between the peak at 350 nm and the drawn baseline at the same wavelength. 0.4 0.3 0.2 0.! I I I 310 350 390 Wavelength, nm Figure 1. Absorption spectrum of HDBA eluate 1 430 Assay of Standard Triethylrhodamine A slurry of silica gel (Grade 922, 200-325 mesh ),* in chloroform was pre- pared. This was poured into a 2.5 x 48 cm chromatographic column. The col- umn was eluted with chloroform until the silica became translucent (almost clear). Triethylrhodamine (100 mg) was put on the column in 0.5 ml of CHCi:•. The column was then eluted with 300 ml each of the following sol- vents: CHCi:,, 1% ether in CHCla, 50% acetone in CHC13, and 1% methanol in CHC13. The eluates were discarded. The column was then eluted with 33.3% methanol in CHCla. Twelve-milliliter fractions were collected. Five- microliter aliquots of each fraction were spotted on Brinkmann F254 silica gel *Daviso'n Chemical Div., W. R. Grace & Co., 101 H. Charles St., Baltimore, Md. 21203.