52 The Philippine Journal of Science 1913 



solving phenoquinone crystals in alcohol. These two solutions 

 are in fact identical. Since the curve obtained with tenth molar 

 solutions so closely resembles the quinone curve, and for other 

 reasons, it is evident that the actual amount of phenoquinone in 

 solution at this concentration is very small. 



If the above equilibrium reaction is correctly represented, an 

 increase in the concentration of phenol should result in a greater 

 concentration of phenoquinone, and in phenol as a solvent the 

 dissociation of phenoquinone should be at a minimum. The 

 curves obtained with an excess of 8 equivalents of phenol and 

 in pure phenol show this assumption to be justified. The most 

 striking characteristic of these curves is the change from a 

 definite absorption band shown in quinone to a mere step off 

 in phenoquinone in phenol and also the greatly reduced concen- 

 tration at which the incipient band appears. From this it is 

 evident that the red color of phenoquinone is largely due to 

 general absorption, and it is probable that the isorropesis present 

 in phenoquinone is quite different from that in quinone. This 

 difference, it seems to us, is due to a loss of the free affinities of 

 the two carbonyl groups, and may be best expressed by the 

 oxonium structure 



H OC 6 H 5 



\ / 

 



II 

 /\ 



II II 



V 



II 

 . 



/ \ 



C 6 H 5 H 



This representation accounts for the instability of the com- 

 pound, the dissociation in solution, and the formation of its very 

 unstable salts better than the formulas of Jackson and Oenslager * 

 and of Willstatter and Piccard. 5 



The conditions existing in dianilinoquinone, dianilinoquinone- 

 anil, and azophenine are quite different. In all of these com- 

 pounds the quinone absorption band is well marked, and in 



4 Ber. d. deutschen chem. Ges. (1895), 28, 1614; and Am. Chem. Journ. 

 (1896), 18, 1. 



'Ber. d. deutschen chem. Ges. (1908), 41, 1458. 



