THE unsaponifiable matter of halibut-liver oil, after removal of the major part of the sterols, is a viscid orange-brown oil containing 60-70 % vitamin A, which on fractional distillation yields a pale yellow oil [Heilbron et al., 1932]. If the vitamin concentrate, whether purified by distillation or not, be dissolved in light petroleum and passed through a column packed with adsorbent (A1203 or Ca(OH)2) a red zone appears. Recovery of the adsorbed material by elution of the adsorbent with methyl alcohol shows that the colour is much more intense than that of the original concentrate. In view of the important biological relation between the red pigment carotene and vitamin A, this phenomenon has been studied in detail. At the same time, it was convenient to examine the reported separation of vitamin A into cx-and fl-forms [Karrer, Walker et at., 1933; Karrer and Morf, 1933;Euler et al., 1934]. As the main distinctions between these two forms and the other fractions obtained depend upon the spectroscopic -examination of the colours they produce with antimony trichloride, a brief r6sum6 of the earlier spectroscopic data on this reaction is necessary.The blue colour characteristic of vitamin A [Carr and Price, 1926] has been shown to exhibit two absorption maxima at 620 and 583 mu in concentrates, and 606 and 572 mp, in oils [Drummond and Morton, 1929;.Measurements of the intensities of the two absorption bands in the antimony trichloride blue solution for numerous liver oils and concentrates [Giulam and Morton, 1931] showed that the two bands varied so much in relative intensity as to make it seem probable that they were derived from two separate, but related, chromogenic substances. This was the more probable when it was found [Heilbron et al., 1931] that in oils which showed the 572 m,u band more strongly than the 606 m,u band, the intensity at 606 m,u could be increased by oxidation with a little ozone or hydrogen peroxide. Further, the 606 m,u band could be completely suppressed by adding a few drops of antimony trichloride before the main bulk of the reagent.Certain indole-like substances likely to be present in liver extracts were found to inhibit the formation of the blue colour . Thus, a liver oil giving a blue colour with antimony trichloride showing a strong 606 m,u band and a weak satellite at 572 mp,, when treated with a trace of 7-methylindole before addition of the antimony trichloride, gave a violet colour showing only the 572 mp, band, the 606 mp, band having been completely suppressed. These workers 1933] regarded this evidence as supporting the theory of two separate chromogens responsible for the two 1 Keddey Fletcher-Warr Student of the University of London.