374 GIBBS AND PRATT. 
entirely the compound under investigation into the new form. 
Very conclusive results, which were more or less anticipated from 
a careful perusal of Baly’s work, were obtained. 
Four plates, somewhat typical of the many we have made, are 
reproduced full size at the end of this article in order to show 
the general character of the photographs from which our con- 
clusions are drawn. Much detail present on the negative is 
lost in the reproductions which are therefore of little value for 
purposes of plotting curves. In the text figures the curves are 
plotted with the logarithms of the dilutions as ordinates and the 
wave lengths as abscissas. We employed the wave lengths rather 
than the oscillation frequencies for the reason that the region 
of the spectrum with which we are perhaps most concerned is 
thus more extended in the charts. 
P-nitrophenol, by this method of evestieation is shown to 
have an absorption band, as described by Baly, Edwards, and 
Stewart,’ heading at about 300 pp, which does not appreciably 
shift its position in the presence of alkalies. The shift of approx- 
imately 100 pp, which these authors show on the addition of alka- 
lies, does not take place. What really happens is the formation 
of a new band heading at 405 py. This band increases in per- 
sistence and breadth with the concentration of alkali, while the 
band heading at 300 pp» is correspondingly reduced in persistence. 
(Fig. 1.) When the molecular concentration of the alkali is equal 
to, or greater than, the concentration of the p-nitrophenol, the 
color band has reached approximately its maximum persistence 
and the band of the shorter wave length has entirely disappeared. 
This, in our minds, is capable of only one explanation and that 
is expressed by the chemical equilibrium: 
OH : 
Oo ie) 
MG NZ 
| 
N N 
YN Lidia 
Oh 0 O» OH, 
The band at 300 pp is due to the benzenoid form while the 
color band at 405 pp» is produced by the salt of the quinoid form 
of the compound. 
*Ibid., fig: 6, p. 521. 
