Dr. Kane on the Compounds of Ammonia. 11 



by heat red fumes, and also liquid nitric acid, but in constantly decreasing pro- 

 portion. I consequently considered it unnecessary to press the series of analyses 

 further. 



The analyses A and B give the result NO5. Ho -|-3h^o pretty closely, the 

 theoretical numbers being 



63.14 16.13 



} 

 } 



328.20 83.87 



391.34 100.00 100.00 



and I am disposed to consider such as being the real composition of the yellow 

 sub-pernitrate prepared by water not boiling. It will be at once seen that this 

 formula assimilates completely the sub-nitrate of mercury with those of copper 

 and of bismuth, the nature of which has been lately elucidated by the experi- 

 ments of Graham. 



With regard to the red sub-nitrate prepared by boiling water, I am inclined 

 to look upon it, in like manner, as having a definite composition, because, whilst 

 the specimen used in analysis C had been boiled but for a few minutes, and that 

 used in analysis D for some hours, their composition appeared to be quite the 

 same. When heated, this red subsalt certainly yields a trace of water, besides 

 nitrous acid fumes ; but this water is in such small quantity that it might be 

 considered as hygrometric. The quantity of mercury obtained, may serve equally 

 well for one or other of two formulee, thus : 



NOj -f- 6 H^O NO5. HO 4- 7 H^O 



Nitric acid = 7.62 6.52 ] 



Oxide of mercury z= 92.38 92.39 • = 100.0 



Water = 1.09 . 



Although I have always found this red powder to give a trace of water, yet I 

 incline strongly to the first of the above formulae, to which I shall refer when 

 treating of some analogous ammonia compounds. 



As the composition assigned by Grouvelle to the sub-pernitrate falls within 

 the limits of the two bodies which have been just described, it may be supposed 

 that he had examined a mixture of them, and not a pure substance ; this idea I 

 consider probably to be true. 



c2 



