416 VEDDER. 



bases precipitated by the phosphotungstic acid, was rendered 

 slightly acid with H2SO4. The barium sulphate was filtered off, 

 and the remaining filtrate diluted with distilled water in such 

 proportion that each cubic centimeter of this fluid represented 

 the substances obtained by this method from 1 gram of the 

 polishings. 



Four fowls were fed on polished rice with a daily addition 

 of 10 cubic centimeters of this fluid containing the substances 

 precipitated from the extract of rice polishings by phospho- 

 tungstic acid. 



One case of neuritis appeared after twenty-five days when this 

 experiment was discontinued. As a control 4 fowls were fed on 

 polished rice with a daily addition of 10 cubic centimeters of 

 untreated extract of rice polishings. These fowls remained well 

 for sixty days, when they were released. 



This failure to obtain the neuritis-preventing substance in 

 the phosphotungstic acid precipitate confirms our previous work. 

 But while we have thus far been unable to confirm Funk's state- 

 ment that this substance is precipitated by phosphotungstic acid, 

 neither do we regard our work as disproving this statement, 

 since the method we used differed slightly from that used by 

 Funk. It is plain, however, since both filtrate and precipitate 

 from phosphotungstic acid were ineffective, that either the neu- 

 ritis-preventing substance was destroyed by the phosphotungstic 

 acid, or else that it was precipitated by that reagent and was 

 subsequently destroyed by the slaked lime and baryta used. 

 In either case it is evident that this neuritis-preventing sub- 

 stance is a most delicate compound and that chemical manipu- 

 lations with this substance must take place only under certain 

 as yet undetermined conditions if they are to be successful. 

 Further experiments will be continued along this line. 



In pursuing investigations into the cause of beriberi it is by 

 no means uncommon to find instances where beriberi has de- 

 veloped in spite of the fact that the patients had received what 

 was supposed to be a tolerably well-balanced ration in addition to 

 the staple article of rice. This observation has been frequently 

 urged as an insuperable objection to the theory that beriberi is 

 caused by a rice diet. It is apparent, however, that the neuritis- 

 preventing substance is not present in all articles of food. Thus 

 in a previous paper, we showed that polyneuritis was not pre- 

 vented in fowls by giving them an extract of onions. It was 



