BANDROWSKI'S BASE 145 tographed (as quickly as possible after completion of the extraction) on Uniplate Silica Gel G plates of 0.250 mm thickness.* The plate was spotted with a 10X pipette (more than one charge was used if the extract was dilute) while confining the spot to a diameter of 3-4 min. After evaporation of the solvent, the plate was developed with a mixture of CHClo/MeOH (9:1 by volume) for a distance of 10 cm. Fresh solution was needed for each plate. As little as 2.5 X 10 -5 mg of BB was visible without any further development or visualization other spots were made visible by iodine fumes. Because of slight variations of Rf values in this system, samples of authentic BB (and occasionally of PPD) were included on each plate. The usual Rf value of BB in this system was 0.25, and the spot appeared as a blue or violet spot without need for any visualization. The PPD spot appeared at Rf 0.60 and turned brown after exposure to iodine fumes. RESULTS AND DISCUSSION Solvent Extraction--Only trace amounts of BB were found in those chromatograms which were made with freshly prepared methanol, ethanol, and pyridine extracts of dyed hair. BB seemed to be a negligi- ble component of the color its amount did not seem to be commensurate with the considerable loss of color of the hair in these treatments. During extractions, the hair changed from a deep brown-black to a medium red, while the extracts were generally reddish. It is estimated that only about one-half of the color was removed from hair by extraction with solvents. This portion was removed fairly quickly, but the re- mainder of the dye resisted extraction. Extraction with pyridine ap- peared to remove more dye than other solvents and yielded an addi- tional red spot of Rt 0.7. Carbon tetrachloride removed no visible dyes or BB from hair. On the other hand, dimethyl formamide (DMF) completely decolorized dyed hair. The extract was faint yellow-red, but the chromatogram showed no BB. When BB was maintained in DMF at 60øC for two days, it was completely destroyed. Finally, dimethyl sulfoxide (DMSO) also completely decolorized hair to yield an almost black extract. This extract showed no BB by TLC. As in the case of DMF, BB was de- stroyed when it was heated in DMSO at 60 øC for two days. The failure to find BB in appreciable quantities in the chromato- grams of the extracts of dyed hair was not due to the technique. In fact, solutions of BB (of about the same color intensity as the extracts) in any * Analtech, Inc., Wilmington, Del.

146 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS of these extractants readily yielded the expected spot of BB on chromato- grams. The major spot on chromatographic plates was a red-brown spot which did not move from the origin. No attempt was made to resolve this spot further or to identify it chemically. The spot has the same ap- pearance and characteristics as a spot formed from a solution of pure BB which has been heated for prolonged periods with the solvents used for extraction. One might infer from this finding that BB is formed on the hair and that it is changed to a colored material during attempts to ex- tract it. Although this interpretation may appear attractive, it must be rejected. In a lengthy series of boiling tests in methanol, it was shown that BB survived the extraction treatments fairly well.* Even the weak spot of BB (which is routinely found in extracts from dyed hair) is not lost after long periods (up to four days) of boiling of methanol and ethanol extracts. t The reactivity of BB in the boiling tests was not af- fected by the presence of hair which, conceivably, could accelerate the decomposition of BB. In agreement with Cox (13), it is, therefore, con- cluded that a small amount of BB is formed on the fiber and not readily removed during rinsing and shampooing. Alternately, it might be pos- sible that the faint BB spot (Rf 0.25) in the hair extracts is due to trace amounts of PPD (which are always found in chromatograms of these ex- tracts) which was oxidized during chromatographing. Actually, the BB spot appears whenever (even pure) PPD is chromatographed. Decomposition of ttair--A final attempt was made to remove BB from hair by decomposition with 0.1/V NaOH. In experiments with un- extracted and solvent extracted hair, no filterable amounts of BB could be removed but this was not surprising in view of the small amounts of BB and other colorant which were actually produced under the described dyeing conditions. The thin-layer chromatogram of the digestion fluid gave a faint test for BB only in samples which had not been previously solvent-extracted. Alkaline treatment of dyed hair immediately freed color, but this color contained only trace quantities of BB. In separate experiments, it was established that BB is sufficiently soluble in dilute alkali to yield a satisfactory thin-layer chromatogram. It was also established that it is relatively stable in dilute sodium hydroxide. This series of experiments was repeated several times on hair which had not * When a solution of 0.5 mg of BB in 100 ml of methanol was refluxed for 16 hours, BB could still be detected chromatographically in an aliquot of the solution equivalent to 2.5-5,0 X 10 -'• mg of BB. If pyridine was used instead of methanol, the lower limit of detection was 5.0-10 X 1O-5 mg of BB, } Admittedly, the spot of BB will disappear completely if boiling is continued long enough.