VITAMIN B (5i) 69 



fractionation it will be observed that the active material was not pre- 

 cipitated by platinic chloride, whilst in the second it was. In the first 

 method the precipitation of the active substance appeared to be de- 

 pendent on the presence of an excess of phosphotungstic acid, whereas 

 in the second the amount of this reagent required was much less. More- 

 over, charcoal adsorbed most of the active material at pH 4 and 5 in 

 our first method, whereas the curative substance in yeast extracts has 

 been found by Kinnersley and Peters to be adsorbed selectively by the 

 same reagent in the zone of neutrality, an observation which we have 

 been able to confirm. Jansen and Donath (1926) described a procedure 

 in which the activity was found to be associated with the precipitate 

 obtained by treatment of their concentrate from rice-polishings with 

 picrolonic acid, whereas in the first procedure which we adopted for 

 the concentration of wheat-embryo extract we have been able to recover 

 most of the activity in the picrolonic acid filtrate. Similar apparently 

 conflicting statements regarding the properties of vitamin Bi have ap- 

 peared repeatedly in the literature. It has been suggested that the 

 factors which various workers were attempting to isolate from different 

 sources were probably not identical. In this paper, however, it has been 

 shown that even when the same biological techniques are adopted, and 

 the same substance chosen as the raw material, the properties of the 

 active factor are profoundly modified according to the treatment under- 

 gone. It does not, therefore, appear strange that the substance should 

 exhibit even greater differences in behaviour in the hands of investi- 

 gators, starting from different raw materials and adopting different 

 biological methods." 



Williams and Eddy (1929) stated that the Jansen-Donath method 

 had been applied to the fractionation of rice polishings with greater 

 success than had been the case with yeast, but without confirmation of 

 all of the claims of Jansen and Donath. They were also inclined to 

 attribute some of the discrepancies in estimates of recoveries to differ- 

 ences in the method of physiological testing. Continued protection of 

 pigeons against polyneuritis, but at the same time loss in weight as 

 compared with growth in rats, was thought to confirm their previously 

 expressed theory of a third factor in rice polishings necessary for the 

 growth of pigeons but not of rats. In a more detailed report of these 

 studies Williams, Waterman, and Gurin (1930) state that in their 

 opinion the most important result of their work was the evidence that 

 the pure antineuritic vitamin, in small doses at least, has scarcely any 

 appreciable effect upon the weight curves of pigeons on a polished rice 

 diet. 



