158 PROCEEDINGS OF THE AMERICAN ACADEMY. 



with water. Sodic hydrate dissolves it, formiug a purple solution, from 

 which acids throw down the unaltered substance. The properties of the 

 substance, therefore, confirm the inference drawn from its preparation 

 and analysis, that this is the free oxyquinone. 



Telloiv Product of the Action of Isoamylamine on the Oxide. 



In the description of the action of isoamylamine en the oxide it was stated 

 that a yellow substance was formed at first, which was converted by heating 

 the mixture into the purple isoamylamine salt of the dichlorisoamylami- 

 doxyquinone. It would have been interesting to determine the compo- 

 sition of this intermediate product between the salt and the oxide, but 

 unfortunately it was so unstable, and its properties rendered its purifi- 

 cation so doubtful, that we can make no certain statement in regard to it. 



It was prepared by dissolving 0.5 gram of the oxide 

 C6Cl20CH30(OCOC,IQoOCH3 



in a large quantity of benzol, and adding with continual stirring 0.6 

 gram of isoamylamine. The solution immediately turned yellow, and 

 after standing for a minute or two suddenly began to coagulate, and 

 finally solidified entirely into a yellow gelatinous mass with a slight 

 greenish fluorescence. It was washed on a filter with benzol, until the 

 filtrate, which was red at first, became colorless, and was then dried 

 on a porous plate ; but, as it darkened somewhat, it was washed again 

 with chloroform and alcohol, until it was yellow throughout. After 

 drying in vacuo an analysis was made, which gave 16.66 per cent of 

 chlorine. This is not far removed from 15.93, the percentage of chlo- 

 rine in the not improbable intermediate compound 



C6Cl2CH30(OHNHoC5H„)0(OCOC6H5)OCH3, 



in which one of the acetal groups has been saponified to a quinone oxy- 

 gen. We did not attempt to get more analyses of this substance, be- 

 cause, owing to its instability and amorphous character, we did not feel 

 that much confidence could be placed in them, and therefore preferred 

 to devote our time to work which had a more immediate bearing on 

 the nature of the oxide, and which also promised to give more trust- 

 worthy results. 



Action of Isoamylamine on Chloranil Dichlordiisoamylamidoquinone, 



C„Cl2(C,HnNH),0,. 



The work described in this and the two following sections of the paper 

 was undertaken to obtain bodies closely related to our isoamylamine salt 



