VITAMIN B (Bx) 61 



in amounts of about 0.1 milligram, containing 0.015 milligram of nitro- 

 gen. The yield, however, was very small, amounting to from 3 to 5 per 

 cent of the original Osborne-Wakeman fraction. Attempts were made 

 to increase the yield by removing the salts of phosphoric acid from the 

 lead-barium product before the silica gel adsorption, either by precipi- 

 tating the active material with absolute alcohol from a sulfuric acid 

 (pH 2) extract of the precipitate or by dissolving it in a minimal 

 amount of hydrochloric acid of specific gravity 1.19 and adding to the 

 solution enough 98 per cent alcohol to make the alcohol concentration 

 70 per cent. The materials thus prepared were active in daily amounts 

 of 2.5 and 1.25 milligrams respectively, but subsequent adsorption on 

 silica gel increased the activity to a much less extent than was the case 

 with the phosphorus-containing fraction, the minimum effective dose 

 of the purified fraction after adsorption being 0.6 milligram and of the 

 phosphorus-containing fraction 0.1 milligram daily. This was thought 

 to suggest the possibility that the phosphorus-containing impurity is 

 essential for selective adsorption by silica. 



A further modification of their method was described the following 

 year by Levene and van der Hoeven (1926) essentially as follows: 



The Osborne-Wakeman fraction was extracted repeatedly with 

 absolute alcohol until a dry nonhygroscopic powder was obtained. This 

 was dissolved in water and adjusted to pH 4, at which reaction some 

 inactive material precipitated out on standing. The remaining filtrate 

 was active in amounts furnishing 0.5 milligram nitrogen per day, as 

 compared with 2 milligrams for the original Osborne and Wakeman 

 fraction. This was deaminized with barium nitrite and sulfuric acid, 

 the strongly acid solution after concentration under diminished pres- 

 sure was precipitated by adding alcohol to 80 per cent concentration, 

 and the process repeated with the filtrate from the first precipitation. 

 This material was active in amounts containing about 0.15 milligram 

 nitrogen per day. The final stage consisted in the adsorption of the 

 active material from the preceding step by silica gel at pH 5, followed 

 by two extractions at pH 3, and one at pH 9.5, neutralization and con- 

 centration of the extraction product, filtration through collodion mem- 

 brane, further concentration, and precipitation with dry acetone. The 

 activity of the final fraction was reported to be such that in one case 

 normal growth for a 14-day period was secured with 0.15 milligram 

 substance per day. When observations were extended to longer periods, 

 the material was usually found ineffective after the first week. 



"The final product obtained is still a complex mixture, containing 

 complex carbohydrates and nitrogenous substances and esters of phos- 



