200 PROCEEDINGS OF THE AMERICAN ACADEMY 



pale as to be hardly perceptible except with a considerable thickness 

 of liquid, it was only necessary to transfer the precipitate to a mortar, 

 rub it with water and throw it upon a fresh filter, in order greatly to 

 intensify the yellow color of the filtrate. Believing that the chromic 

 acid which colored the wash-water was derived from the decomposition 

 of the precipitate, we stopped the washing on the eighth day, dried the 

 precipitate at 50° — 55°, and analyzed it by the first method described 

 above. The color of the solution of the substance in dilute nitric acid 

 was red-brown, and of the precipitate produced by ammonia, dirty 

 green, which washing changed to the usual color of chromic oxide. 

 The result of the analysis was, — 



CrjOs = 53.31 per cent. 



CrOa = 18.64 " " 



HO (by difference) = 28.05 « " 



In regard to the ratio of the chromic oxide to the chromic acid, this 

 result is very much like that of the first analysis given above (page 

 193), and leads to the same conclusions, viz, that the original precipi- 

 tate is a compound of chromic oxide and chromic acid, from which 

 prolonged washing abstracts chromic acid to an indefinite extent. We 

 next prepared a new precipitate by passing nitric oxide through a di- 

 lute solution of bichromate of potash, and, without washing it at all, 

 pressed it between folds of filter-paper, dried it by exposure to the air, 

 and analyzed it by Rose's method, above desci'ibed, with the following 

 result : — 



Cr^Oa = 28.55 per cent. 



CrOs = 29.33 « « 



HO (by difference) = 42.12 " " 



The object of this analysis was to prove that, even when the precipi- 

 tate was contaminated with the chromates of the filtrate, the proportion 

 of the chromic acid to the chromic oxide fell far below that which 

 would exist in a compound containing two equivalents of chromic acid 

 to one of chromic oxide, thus making it probable that the true formula 

 of the precipitate is OaOa CrOs- 



To determine the exact proportions of chromic oxide and chromic 

 acid in the original substance evidently requires a different process of 

 analysis, and the third method of analysis described above is applicable 

 to this purpose, with this advantage over the case to which we first 



