4 PROCEEDINGS OP THE AMERICAN ACADEMY 



together for some time, the product was treated with water, and the 

 precipitate washed, dried, and anal); zed, with the following result: — 



0.1430 g. of substance gave 0.36G6 g. of CO^ and 0.0868 g. of U.f>. 



Calculated for C28H30O7. Found. 



Carbon 70.42 69.92 



Hydrogen 6.25 6.68 



Whereas before this treatment a portion of the same samjjle gave 

 carbon 67.63, hydrogen 6.36, as on page 2. This experiment proves, 

 therefore, that the nascent hydrogen given off by zinc dust and acetic 

 acid first converts the curcumin into the dihydride, from which the 

 zincic acetate afterward removes one molecule of water. The anhy- 

 dride is broken up only incompletely by the action of water at high 

 temperatures or by boiling with sodic hydrate, the product in each 

 case being a mixture of the dihydride and its anhydride. 



In order to study the oxidation of the dihydi'ide, some diethylcur- 

 cumin, prepared according to the method given in our first paper, was 

 treated with zinc dust and acetic acid. The product thus obtained 

 gave the following results on analysis, which indicate that it is a mix- 

 ture of di- and monoethylcurcumin dihydride : — 



I. 0.2130 g. of substance gave 0.5484 g. of CO^ and 0.1415 g. of 

 H.,0. Ash 0.0004. 

 11. 0.1 938 g. of substance gave 0.4986 g. of COg and 0.1336 g. of 

 H,0. 



Calculated for Found. Calculated for 



(C,H,)ChU,504. I. II. (CiU,),C,tU,,0^. 



Carbon 69.56 70.35 70.15 71.05 



Hydrogen 7.25 7.40 7.66 7.89 



Some of this substance was mixed with a saturated neutral solution 

 of potassic permanganate and allowed to stand at the ordinary tem- 

 perature. The oxidation took place very slowly, several weeks' standing 

 being necessary to complete the reaction. The principal product was 

 proved by its melting-point to be ethylvanillic acid, while the presence 

 of a small quantity of ethylvanilliu was indicated by its characteristic 

 smell. 



The most striking fact in connection with this reaction was that the 

 dihydride was oxidized with so much more difficulty than curcumin 

 itself. 



Action of JBromine on the Anhydride of Curcumin Dihydride. 

 When the hydride, dissolved in glacial acetic acid, was mixed with 

 an excess of bromine and allowed to stand over night, the liquid 



