168 PROCEEDINGS OF THE AMERICAN ACADEMY. 



color seemed to be analogous to the phenoxy compound. Here we were 

 more successful. A sample of this green product of the action of sodic 

 methylate on clilorauil was washed with equal parts of methyl alcohol 

 and water, which did not decompose it so rapidly as water alone, then 

 with methyl alcohol alone, and finally with benzol. It gave the follovy- 

 ing result on analysis, after it had been dried in vacuo : — 



0.1810 gram of the substance gave 0.0418 gram of sodic sulphate. 



Calculated for CeCIiOaCHaONa. Found. 



Sodium 7.66 7.48 



Our other analyses pointed to a similar composition for this substance 

 and related compounds, but in no other case were we so fortunate as to 

 obtain a sample free from sodic chloride. We should not venture to 

 publish this single analysis, therefore, if it were not confirmed by the 

 results given by Asti'e, who made a green compound by the action of 

 potassic ethylate in ethereal solution upon quinone, which had the for- 

 mula CgH^OaKOCoHs, and a similar substance with alcoholic potassic 

 hydrate, C6H40oKOri. Our attempts to prepare compounds like these 

 for analysis have given us a high appreciation of the experimental skill 

 which Astre has shown in purifying these substances. That his com- 

 pounds are analogous to ours we think can admit of no doubt ; not 

 only on account of the analyses, but from the method of formation and 

 the color. 



Astre, however, does not assign to his compounds the formulas given 

 above, but considers that the potassium has replaced one of the atoms 

 of hydrogen attached to the benzol ring, so that the substances con- 

 tain alcohol or water of crystallization, thus: C6H3KO2C0H5OH and 

 C6H3K02H20' His principal reason for this is based on his experiment 

 with metallic potassium and an ethereal solution of quinone. In this 

 case he found that hydrogen was given off, and a green compound 

 formed having the formula (C6H3KOOH).,C4HioO, which amounts to 

 a molecule of dipotassic quinhydrone combined with one molecule of 

 ether. The attempt to apply Astre's theory to our compound leads to 

 the inadmissible formula CeClsNaOoCHsOCl, and therefore, if his com- 

 pounds and ours are similarly constituted, his theory must be abandoned. 

 In this case it is necessary to explain his result with metallic potassium 

 and quinone in ethereal solution, and this can be done perhaps by sup- 

 posing a very minute quantity of water or alcohol to have been present 

 in the ether used. This would have given nascent hydrogen, which 

 would have formed a little quinhydrone ; this quinhydrone then, by act- 



