206 
Transactions of the Boyal Society of South Africa. 
p. 1232) by oxidising an ethereal solution of benzidine with lead peroxide. 
The constitution of the hydrated base must be either 
of which the former is the more probable. 
The blue chromate and ferricyanide are very beautiful substances, 
bearing an obvious analogy in lustre and felted crystallisation to the 
similarly-constituted bodies coerulignone and tetraminodiphenoquinone. 
Substances of this class give an intense coloration, even in traces, to 
concentrated sulphuric acid, and the base of "benzidine chromate," 
diphenoquinonediimine, gives an intense crimson shade exactly the same 
as that given by a specimen prepared by Willstatter's method. On the 
other hand, the original blue chromate gives only an orange colour, 
probably owing to the Cr03 interfering with the reaction. 
As an example of the difficulties attached to the investigation, it may 
be mentioned that the blue substance can scarcely be dried without 
undergoing some degree of internal oxidation, with the result that the 
base prepared from it is always contaminated with the lower oxides of 
chromium, as well as an intractable black organic substance which resists 
reduction. The base itself alters on dehydration and is decomposed in 
the warm by the hydrolytic solvents, whilst it is only sparingly soluble in 
the indifferent ones. In addition, the proportion of combined CrOg is 
somewhat variable in different preparations. In fact a considerable 
portion of the work on this substance deals with its alteration-products 
rather than with itself. 
Some preliminary work, done before its behaviour was fully realised, 
may be first mentioned. By dissolving benzidine in excess of very dilute 
acetic acid at 35° and adding chromate in excess (but not so as to render 
alkaline) and leaving overnight, a yield of 154 per cent, of the blue sub- 
stance washed out with water, alcohol, and ether and dried in vacuo was 
obtained. Analysis of this product gave 0 = 49-6 per cent., H = 4-85 per 
cent., Or = 14-7 per cent., N (Kjeldahl) 8*75 and 9*15 per cent. 
Diphenoquinonediimine chromate OiaHiaNgCrO^ would contain 48-0 
per cent, carbon, 4*0 per cent, hydrogen, 9'3 per cent, nitrogen, and 17*3 
per cent, chromium, and the theoretical yield is 163 per cent., so that for 
some obscure reason the product in this case contained only about f of the 
normal proportion of OrOg. On treating the dried specimen with alkali, 
about ^ of the benzidine originally taken was recovered from it. At the 
low temperature of preparation it may therefore possibly have contained 
a small quantity of true benzidine chromate OiaHj^NaCrO^. Under the 
circumstances direct calculation of a formula from the analysis may be 
