THE ANTIBERIBERI VITAMINE 173 



polishings with alcohol in a reflux condenser for three hours, using 

 fresh solvent, till the extraction was complete. The alcoholic 

 extracts were combined and concentrated. The residue was diluted 

 with water, sulphuric acid added (till a 3 per cent solution resulted) 

 and precipitated with 30 per cent phosphotungstic acid. The solution 

 obtained by decomposing the phosphotungstic acid precipitate was 

 very active and was called "crude oryzanin I." The yield was 1.2 

 grams from 300 grams of fat-free rice polishings. This fraction cured 

 pigeons promptly in amounts of 3 to 4 centigrams, while the filtrate 

 of the phosphotungstic acid precipitate was entirely inactive. The 

 active fraction was soluble in water and alcohol, was strongly acid 

 and gave a marked Millon and diazo reaction. On the addition of 

 lead acetate, a slight precipitate was obtained, increasing on the 

 addition of ammonia. The substance was partially precipitated 

 by sublimate, mercuric acetate, and mercuric nitrate. An attempt 

 to separate this active fraction resulted in a complete loss of 

 activity. This was never the case in our experiments; the vitamine 

 was very resistant to acids. 



The hydrolysis was accomplished by heating for two hours with 

 3 per cent hydrochloric acid. On cooling, yellowish brown crystals 

 separated out, which were recrystallized from hot alcohol. In 

 this way two substances were obtained, the first of which was less 

 soluble than the other. Both were difficultly soluble in water and 

 gave an acid reaction; they were soluble in alkali, and could be 

 reprecipitated on the addition of acid. On analysis, the first product 

 gave the formula Ci8H 16 N 2 9 and was called a-acid; the second prod- 

 uct gave the formula Ci H 8 NO 4 and was called /3-acid; both showed 

 the Millon and diazo reactions. Unfortunately, the melting point and 

 the investigation of the derivatives was not recorded in the report 

 of the Japanese investigators. From the hydrolyzate, choline and 

 nicotinic acid (m.p.214C.) could be isolated as picrates; glucose 

 was also found. One hundred parts of crude oryzanin gave 10 parts 

 of a- and /3-acids, 30 parts of choline and nicotinic acid, and 23 

 parts of glucose. One gram of crude oryzanin yielded 0.044 gram 

 nitrogen, of which 0.035 gram could be precipitated by phospho- 

 tungstic acid, of 0.009 gram was in the form of a- and /3-acids. 



As this composition shows, the investigation of rice polishings by 

 the Japanese workers yielded fundamentally different results from 

 ours. The crude oryzanin I. was then subjected to further 



