JACKSON AND TORREY. — -AN ORGANIC OXIDE. 147 



ing to another than the a series shows these reactions, and this, so far 

 as it goes, throws the balance in favor of adopting the ortho attachment 

 of the oxygen (Formula III.) rather than the meta attachment (Formula 

 IV.) in our oxide. 



The decision between Formula III. (with the ortho attachment) and 

 Formula IV. (with the meta attachment) could not be founded on any 

 study of the derivatives of the oxide, as both formulas would give the 

 same products when the oxygen bond was opened. We accordingly tried 

 to solve the problem by the study of analogous substances, as follows : 

 we exposed to the action of boiling sulphuric acid resorcine as au 

 example of a substance with meta hydroxyls, and pyi'ocatechine (or 

 some of its derivatives) as a substance with ortho hydroxyls, in the hope 

 of obtaining an oxide similar to the one we have been studying, but 

 in neither case was an oxide of this class formed. Resorcine under the 

 action of boiling dilute sulphuric acid gave a body, which we recognized 

 by its properties as resorcine ether formed by the union of two mole- 

 cules of resorcine. The experiments with pyrocatechine and its deriva- 

 tives, when the substance underwent any change, gave colored products, 

 which were probably formed from a quinone. These experiments, there- 

 fore, threw no light upon the question. 



From the discussion given above it appears that we cannot decide 

 with certainty whether the oxygen in our oxide is in the ortho or meta 

 position ; we are of the opinion, however, that it has the ortho attach- 

 ment, not only from the analogy between our oxide and ethylene oxide 

 pointed out above, but also from the fact that the formation of anhy- 

 drides and similar couplings by two radicals in the meta position to each 

 other are unknown, whereas these actions are among the most familiar 

 characteristics of the ortho series. 



We also describe tetrabromguaiacol, melting at 162° to 163°, and 

 tribromveratrol, melting at 86° to 87°, which were made in connec- 

 tion with our work on the action of sulphuric acid on derivatives of 

 pyrocatechine. 



When we first undertook this work, we devoted a great deal of time 

 to the preparation of mixed acetals like the dichlordiethoxyquinonedi- 

 methyldibeuzoylacetal, as we hoped by the saponification of these sub- 

 stances to be able to determine which of the foi>mulas of the oxide was 

 correct ; but these experiments led to no result, as the action of sodic 

 methylate on dichlordiethoxy quinone, or of sodic ethylate on dichlordi- 

 methoxyquinoue, produced mixtures from which the desired compounds 

 could not be isolated. This is not surprising, as we found subsequently 



