NATURE AND DECOMPOSITION OF THE FIXED ALKALIS. 399 



8'3=10'2; and 11-2 — 10-2 = 1; all the estimations being 

 made according to the standard temperature and pressure. 



So that in this experiment on the decomposition of Products only 

 ammonia, the Aveight of the gasses evolved is less by nearly \^'^ therefor* 

 ■jV than that of the ammonia employed; and this loss can ^* oxigen. 

 only be ascribed to the existence of oxigen in the alkali ; 

 part of which probably combined with the platina wires 

 employed for electrization, and part with hidrogen. 



After these ideas the oxigen in ammonia cannot well be Ammonia pro- 

 estimated at less than 7 or 8 parts in the hundred: and i^niVre UiaiiTs 

 possibly exists in a larger proportion, as the gassos evolred 

 may contain more water than the gas decomposed, which 

 of course would increase their volume and their absolute 

 weight*. 



In supposing ammonia a triple compound of nitrogen Supposing it» 

 hidrogen, and oxigen, it is no less easy to give a rational "'^ VTh' 

 account of the phasnomena of its production and decompo- phenomena 

 sition, than in adopting the generally received hypothesis ^^^'"''^*^*^°^^ 

 of its composition. 



Oxigen, hidrogen, and nitrogen are always present in 

 cases in which volatile alkali is formed ; and it usually ap- 

 pears during the decomposition of bodies in which oxigen 

 is loosely attached, as in that of the compounds of oxigen 

 and nitrogen dissolved in water. 



At common temperatures under such favourable circum- Ammonia 

 stances, the three elements may be conceived capable of ^^^ ^^j^^^.-^j^ 

 combining, and of remaining in union: but at the heat of compomid 

 ignition the affinity of hidrogen for oxigen prevails over the 

 complex attraction, Avater is formed, and hidrogen and 

 nitrogen are evolved ; and according to these conclusions, 

 ammonia will boar the same relations to the fixed alkalisj 

 as the vegetable acids with compound bases do to the mineral 

 ones with simple bases. 



100 cubical inches of nitrogen weigh, at the standard temperature 

 and pressure, 29"0 grains, 



* In the present state of our knowledge, perfectly coiToct data Quantity of 

 r ,■ t L 1,1 T • 1 • • < water in gasses 



for proportions cannot probably be gamed ni any expenmeiits on ^^ ^^ knowa 



the decomposition of ammonia^ as it seems impossible to ascertain only by elec- 

 the absolute quantity of water in this gas ; for electrization, ac- trization. 

 cording to Dr. Henry's ingenious researches, olfers the only means 

 Jcpown of ascertaining the quantity of water in gasses. 



5 Oxigen 



