VIII, B, 3 Vedder and Williams: Vitamines in Rice 185 



possible, of course, that these different chemical compounds may 

 all contain Funk's base as a nucleus. This question can be 

 solved only by obtaining complete information as to the consti- 

 tution both of Funk's base and of these several compounds. We 

 had hoped to begin this v^ork, but have been greatly hampered 

 owing to the extremely minute amounts of Funk's base which 

 so far we have been able to obtain. This leads us naturally 

 to the consideration of the second point outlined above ; namely, 

 is Funk's base completely recovered by Funk's method? 



From what has already been said with regard to the action 

 of alkalies on the protective substances, it appears probable that 

 Funk's base is not completely recovered by his method, but that 

 the greater part of it is destroyed when the precipitated phos- 

 photungstates are broken up by barium hydroxide, which is a 

 powerful alkaline reagent. This seems the more probable from 

 several observations that we have made. In the first place, 

 the curative action of the simple hydrolyzed extract is very 

 powerful, the extract from 200 grams of polishings being 

 usually sufficient to produce a complete cure. It is evident from 

 experiment 46 that precipitation with phosphotungstic acid does 

 not injure the protective substance; but, when the phosphotung- 

 states are broken up by barium and the decomposed product 

 used for curative experiments, it is found that a quantity from 

 twenty-five to fifty times greater than the original 200 grams 

 of polishings is required to cure a fowl. Manifestly the greater 

 part of the curative substance is lost in the manipulation, and 

 it can only be the result of the use of the alkaline barium hydrox- 

 ide. In the second place, during this decomposition of the phos- 

 photungstates by barium a strong methylamine smell is given 

 off as was previously noticed by Funk. In our experiments we 

 have observed that the loss of curative properties appears to be 

 proportional to the strength of this methylamine odor given off, 

 and we deduce from this fact that the methylamine odor indi- 

 cates the amount of decomposition of nitrogenous bases to which 

 class the protective substances belong. 



In the course of our attempts to discover a better method for 

 the isolation of this base, we tried a method used by Suzuki, 

 Shimamura, and Odake.(9) This method consisted in precipi- 

 tating the unhydrolyzed extract by tannic acid, treating this 

 precipitate with 3 per cent sulphuric acid, and subsequently 

 removing the excess of sulphuric acid and tannic acid by barium 

 hydroxide. By following this method exactly we succeeded in 

 curing fowls by the extract so obtained, and therefore confirmed 

 the above authors' observation. We also found that if the 



