IV. ESTIMATION 217 



room for improvement with regard to permanence and specificity. In its 

 more highly developed forms, the antimony trichloride reaction and its 

 associated steps of purification are complicated procedures, which give 

 fairly reliable estimations with oily solutions containing 5000 or 10,000 

 I.U. per gram or more of vitamin D. 



Another color reaction has been described by Sobel et al.^^ With a chloro- 

 form solution of dichlorohydrin and acetyl chloride, vitamin D gives a green 

 color with peak absorption at 625 m^u. The color is more stable and more 

 nearly specific than that of the antimony trichloride reaction. Rouir and 

 Pirlot,^^ applying Sobel's reaction to pharmaceuticals, stressed the impor- 

 tance of correcting for oil. CampbelP^ increased the sensitivity of the re- 

 action twentyfold, which makes it only five times less sensitive than the 

 antimony trichloride reaction. He found that by measuring the absorption 

 at 410 m/x, instead of 625 m/x, the interferences given by maize oil, fish oil, 

 ergosterol, and 7-dehydrocholesterol were minimized. The reaction deserves 

 further study with high-potency oils. 



A third promising color reaction is that of Schaltegger.^^ Carbenium salts 

 produced by treating vitamin D with perchloric acid are reacted with alde- 

 hydes, giving colors of improved specificity. Banchetti,'^ using vanillin as 

 the aldehyde in the reaction, was able to detect 50 y of vitamin D in the 

 presence of two hundred times as much of other sterols. The possibilities 

 of this reaction have not been explored. 



Investigators of color reactions should note the report of McMahon et 

 al.,^'^ in which is described the separation and identification of vitamin D 

 from other sterols by partition paper chromatography, the sterol spots 

 being developed by spraying with antimony pentachloride in chloroform. 



A chemical method for estimating vitamin D, which is not a color reac- 

 tion in the usual sense, was described by Green^^ in 1951. It is based on the 

 liberation of iodine in the reaction between vitamin D and iodine trichloride 

 in carbon tetrachloride. A comprehensive study was made of methods for 

 separating vitamin D from interfering substances by means of differential 

 solubility, digitonin precipitation, and chromatography on Floridin. Highly 

 efficient purification is claimed, and the purified product can be reacted 

 stoichiometrically with the reagent. The older techniques of condensation 

 with maleic or citraconic anhydride, and of selective ultraviolet irradiation 



15 A. E. Sobel, A. M. Mayer, and B. Kramer, Ind. Eng. Chem. Anal. Ed. 17, 160 (1945)* 



16 E. V. Rouir and G. Pirlot, Bull. soc. chim. hiol. 29, 1005 (1947). 

 " J. A. Campbell, Anal. Chem. 20, 766 (1948). 



18 H. Schaltegger, Helv. Chim. Acta 29, 285 (1946). 



i« A. Banchetti, Ann. chim. appl. 38, 394 (1948). 



20 J. M. McMahon, R. B. Davis, and G. Kalnitsky, Proc. Soc. Expll. Biol. Med. 75, 



799 (1950). 

 " J. Green, Biochem. J. 49, 36, 45, 54 (1951). 



