in its different States. 407 



As iron has a strong affinity for oxymnriatic acid, T at- 

 tempted to procure soda by passing steam over a mixture 

 or' iron filings, and muriate of soda intensely heated: and 

 in this way I succeeded in decomposing some or' the salt: 

 hydrogen came over: a little hydrate of soda was formed; 

 and muriate of iron was produced. 



It does not seem improbable, supposing the views that 

 have beeen developed accurate, that, by complex affinities, 

 even potassium and sodium in their metallic form may be 

 procured from their oxymuriatic combinations: for this 

 purpose the oxymuriatic acid should be attracted by one 

 substance, and the alkaline metals by another; and such 

 bodies should be selected for the experiment, as would pro- 

 duce compounds differing considerably in degree of vola- 

 tility. 



I cannot conclude the subject of the application of these 

 doctrines, without asking permission to direct the attention 

 of the Society to some of the theoretical relations of the 

 facts noticed in the preceding pages. 



That a body principally composed of oxymuriatic acid 

 and ammonia, two substances which have been generally 

 conceived incapable of existing together, should be so dif- 

 ficult of decomposition, as to be scarcely affected by any of 

 the agents of chemistry, is a phcenomenon of a perfectly 

 new kind. Three bodies, two of which are permanent 

 gases, and the other of which is considerably volatile, form 

 in this instance a substance neither fusible nor volatile, at a 

 white heat. It could not have been expected that ammonia 

 would remain fixed at such a temperature ; but that it 

 should remain fixed in combination with oxymuriatic acid, 

 would have appeared incredible, according to all the existing 

 analogies of chemistry. The experiments on which these 

 conclusions are founded, are, however, uniform in their re- 

 sults: and it is easy to repeat them. They seem to show, 

 that the common chemical proposition, that complexity of 

 composition is uniformly connected with facility of decom- 

 position, is not well founded. The compound of oxymu- 

 riatic acid, phosphorus, and ammonia, resembles an oxide, 

 such as silex, or that of columbium in its general chemical 

 characters, and is as refractorv when treated by common 

 re-agents ; and except by the effects of combustion, or the 

 agency of fused potash, its nature could not be detected by 

 any of the usual methods of analysis. Is it not likely, rea- 

 soning from these circumstances, that many of the sub- 

 stances, now supposed to be elementary, may be reduced 

 into simpler forms of matter? and that an intense attrac- 



2 C 4 tfodj 



