84 Dr. Kane on the Compounds of Ammonia. 



having the four equivalents of oxide all alike, yet I cannot consider such an 

 arrangement as being excluded. 



Indeed an idea which was suggested to me by the mercurial nitrates is, that 

 the constitution of the nitrates may be better shovra by vfriting the formula of 

 their class as follows : 



RO . NO5 . KO "1- 2 RO. 



R being either water or metallic oxide. There is then 



and the red basic nitrate, 

 and still further, 



H^O . NO5 . H^O -j- 2 HO, 

 H^O . NO5 . HO -j- 2 H^O, 



Hg-O. NO5 . H^O + 2vigO -\- 2 H^O. 



H^O . NOj . HAC? -|" 2Hg'0, 

 HOO . NO5 . Hg'AC? + 2 UgO. 



As in the copper and bismuth nitrates, no water whatsoever can be separated 

 without a corresponding quantity of acid being set free, it is difficult to ascertain 

 whether one atom of the water is more firmly attached to the acid than the other 

 two ; but in the case of nitrate of magnesia, Graham has found that two of the 

 equivalents of water may be separated much more easily than the third, and 

 hence its formula should be, as in the quicksilver series, 



M^O . NO5 . HO 4" 2 HO. 

 Mg-O.NOj H0-f-2M^0. 



Thus connecting still further mercury with the metals of the magnesian class. 

 This form of expression for the nitrates connects them much more closely with 

 the sulphates than the older view, and the equivalent to the right of the acid 

 evidently replaces the saline water of the magnesian sulphates. Thus a sulphate 

 of that group is generally, though not always, 



RO . SO3 . no -j- 2?ZH0, 

 n being a whole number. 



In the sulphates the most common form of basic constitution approaches still 

 more closely to the type of basic nitrates than in the neutral state ; thus. 



