VITAMIN B (B,) 37 



ing these crystals and repeating the precipitation with silver nitrate followed by 

 ammoniacal silver nitrate, a voluminous silver precipitate was again formed, but 

 free from the crystallizable inactive fraction. This substance, the yield of which 

 in the dry state was 0.7 gram from 300 grams of activated fuller's earth, when 

 administered in doses of 0.004 gram on alternate days to pigeons on a polished 

 rice diet prevented the development of polyneuritic symptoms. In view of recent 

 work, it may be of significance that this substance did not prevent a slight loss 

 in weight as did the activated solid or crude extract. 



Williams (1921) reaffirmed much of his earlier work upon the 

 antineuritic properties of substances of known structure and continued 

 the development of his theory that antineuritic activity may be con- 

 nected with isomerism of the type above described. He reported that 

 /?-hydroxypyridine and ^-methylpyridine gave some evidence of anti- 

 neuritic property in curative tests on polyneuritic pigeons and tri- 

 methyluracil and 4-phenylisocytosine in protective tests. In his opinion 

 protective experiments are much more convincing than curative. 



The following year Seidell (1922) announced that when a con- 

 centrated extract prepared from yeast by the method of Osborne and 

 Wakeman (described beyond) was used in place of autolyzed yeast in 

 the preparation of activated fuller's earth the resulting product con- 

 tained less nitrogen and about twice as much of the active antineuritic 

 material as the product made in the original way. This new method 

 of preparing the activated solid was used by Seidell (1922b) in a 

 further study of the various silver fractions obtained by his previous 

 method. In this and later work the test for activity of the various 

 fractions consisted (Seidell, 1922a) in preparing a solution of definite 

 concentration, adsorbing it on such an amount of fuller's earth that 

 each 0.1 gram corresponded to a definite number of milligrams of 

 the solids present, and using this preparation as the sole source of 

 antineuritic vitamin for pigeons of known weight on an exclusive pol- 

 ished rice diet. The unit of activity was defined as "that quantity of 

 vitamin, given to a normal pigeon on each alternate day, which is just 

 sufficient to replace the vitamin deficiency of an exclusive diet of pol- 

 ished rice." 



The activity of the silver nitrate precipitate was much higher than 

 previously found. This was attributed to the presence of adenine (also 

 precipitable by silver nitrate) in the extracts prepared from autolyzed 

 yeast and its absence in the extracts obtained by the new method. 

 Only a little over half of the total active vitamin, however, was pres- 

 ent in the two silver precipitates. The loss of active material by this 

 method was thought to be due to the rather high solubility of the 

 silver-vitamin complex. 



The fractions obtained by silver precipitation were quite stable 



