1030 MINNESOTA BOTANICAL STUDIES. 



purified by treating with tartaric acid and sodium carbonate, 

 extracting with ether and finally treating with benzoline. A 

 substance was thus obtained which corresponded with 

 Couerbe's veratrine. Analyses gave numbers for the formula, 



C 37 H 53 N 0„. 



On a saponification with alcoholic sodium hydroxide the base 

 breaks up into veratric acid (dimethylprotocatechuic acid) 

 C9 H10 4 , and a base resembling closely cevine, Cog H 45 N 8 . 

 C37 H 53 N O n + H 2 O = C 9 H 10 4 + C 28 H 45 ~N 8 . 



In their examination of crystallized veratrine Wright & Luff 

 showed that it was a distinct alkaloid, that it probably did not 

 exist in an isomeric form; that it yielded cevadic acid and a new 

 base which they named cevine, by treating with alcoholic pot- 

 assium hydroxide or heating in a closed tube at 200° C. By a 

 series of analyses they arrived at the formula 



C 32 H 49 N 9 



and represented the saponification by the following reaction: 



C 32 H 49 N 9 + H 2 O = C 5 H 8 2 -h C 27 H 43 N 8 . 



Although the base cevine was not studied, yet the derivatives. 

 of cevadine itself were made and analyzed, which leaves no 

 chance for doubt as to the correct conclusion. 



It therefore appears that the veratrine of Merck and Schmidt 

 and Koppen was not veratrine at all but cevadine which upon 

 saponification yields cevine and cevadic acid. A comparison 

 of analyses will serve to show their identity. 



Calculated for Merck. Schmidt & Koppen. Wright & Luff. 

 C:, 2 H 49 N O fl 



C=64.97 64.81 64.63 64.72 



H= 8.29 8.71 ■ 8.62 8.57 



N= 2.37 5.50 2.66 2.31 



Goldingoldsalt21.ua 21.01 21.09 21.04 



The above numbers are the mean of several analyses with 

 the exception of the nitrogen determination by Merck which 

 is undoubtedly the result of an error. 



The structure of cevadine is still a mystery. It is known, 

 however, that there is one hydroxyl group present, and from 

 the succeeding experiments one methoxyl group. 



Later Wright 16 made a careful estimation of all the alkaloids 

 present in both Veratrum album and Veratrum viride. The re- 

 sults showed that the total amount of alkaloidal matter in V. 

 viride was only about one-fifth of that present in V. album. 



(16) Journ. Chem. Soc. 35: 421. 



