55 
on the Nature of certain Bodies. 
will be found, that by the decomposition of 11 cubical inches 
of ammonia equal to 2.05 grains, there is generated 3.6 cu- 
bical inches of nitrogene equal to 1.06 grains, and 9.9 cubical 
inches of hydrogene, which added to that disengaged in the 
first operation equal to about 6.1 cubical inches, are together 
equal to .382 grains ; and the oxygene added to 3.5 grains of 
potassium would be .6 grains, and the whole amount is 2.04 
grains; and 2.05 — 2.04 = .01. But the same quantity of 
ammonia, decomposed by electricity, would have given 5.5 
cubical inches of nitrogene equal to 1 .6 grains, and only 14 
cubical inches of hydrogene* equal to .33, and allowing the 
separation of oxygene in this process in water, it cannot be 
estimated at more than .11 or .12. 
So that if the analysis of ammonia by electricity at all ap- 
proaches towards accuracy; in the process just described, there 
is a considerable loss of nitrogene, and a production of oxy- 
gene and inflammable gas. 
And in the action of water upon the residuum, in the expe- 
riment page 52, there is an apparent generation of nitro- 
gene. 
How can these extraordinary results be explained ? 
The decomposition and composition of nitrogene seem 
proved, allowing the correctness of the data ; and one of its 
elements appears to be oxygene ; but what is its other ele- 
mentary matter ? 
Is the gas that appears to possess the properties of hydro- 
gene, a new species of inflammable aeriform substance? 
Or has nitrogene a metallic basis which alloys with the iron 
or platina ? 
* See Phil. Trans. 1808, p. 40. 
