VITAMIN B (B^ 59 



tempting to concentrate or isolate vitamin B, dilute alkaline solutions 

 can be used without materially affecting its activity provided that a low 

 temperature and a short time of exposure to the alkali be employed. 



Levene and van der Hoeven (1924) in a brief note in Science 

 announced that they had succeeded in preparing from the Osborne and 

 Wakeman fraction by precipitation with barium hydroxide a material 

 which was from five to ten times more active as a source of vitamin B 

 than the original fraction. An extension of this method, together with 

 others, was described later in the same year. In this paper Levene and 

 Van der Hoeven (1924a) called attention to the fact that work hitherto 

 reported on the purification of the crude vitamin had followed two 

 distinctive principles, one aiming to find a specific reagent for precipi- 

 tation or for extraction (as in the early studies on the antineuritic 

 vitamin) and the other to remove gradually the inert material and thus 

 increase progressively the potency of the material (as in the work of 

 Osborne and Wakeman). Seidell's use of Lloyd's reagent to adsorb 

 the active material from the Osborne-Wakeman fraction was considered 

 to represent further progress in the choice of methods, and was the 

 first method used in the present study. 



Acting on the theory that the adsorbing power is dependent upon 

 the degree of ionization of the active surface of the adsorbent and the 

 degree of dissociation of the substance to be adsorbed, and that there 

 must exist for each substance a certain hydrogen-ion activity at which 

 the adsorption is optimal, the adsorbing power of Lloyd's reagent was 

 tested at different hydrogen-ion activities. The optimal adsorption of 

 the vitamin was found to take place at pH 4, while the optimal extrac- 

 tion from the adsorbed material was at pH 9. By this means a fraction 

 was obtained which was from two to three times as active as the orig- 

 inal material. The use of kaolin as adsorbent gave unsatisfactory re- 

 sults. With silica gel better results were obtained than with any other 

 adsorbent. The optimal hydrogen-ion activity for adsorption with silica 

 gel was at pH 5, and for subsequent extraction was at pH 9, although 

 a considerable portion was extracted by acids at pH 3. 



By treating with silica gel the solution containing the active material 

 obtained by the barium process noted in the preliminary report, and 

 subsequently extracting at pH 9. a material was obtained which was 

 from 200 to 400 times as potent as dried yeast, but at the expense of 

 considerable loss of the original material. A better method consisted 

 in suspending the original barium precipitate in water and passing a 

 stream of carbon dioxide through it. Lender these conditions most of the 

 potent material remained in the precipitate, while the inert material 



