246 ON TMB ALKALINE METALLOIDS. 



tionof hidro. Aictallic base, the potassium obtained by tliis procesb has 

 8^^ been observed to have a greater specific gravity, than that 



obtained by the action of galvanism. This has been as- 

 cribed to a slight alloy of iron, whicli has not however been 

 detected. It is not improbable, that it may be owini^ to 

 it? containing less hidrogen, a quaijlity of hidrogen being 

 disengaged in the process by which it is formed, while no 

 fevolution of this principle is observed in the pro 1 action of 

 potassium by galvanism. The whole hidrogen of the wa- 

 ter therefore, which existed in the potash, will not be con- 

 tiiined in the potassium formed in this way; and to this may 

 be owing that slight superiority of alkalitie strength, ob- 

 served in the above experiments, of the potash formed from 

 potassium. 



One or other of these theories — either that proposed by 

 Gay-Lussac and Thenard, or that which I have stated, it 

 appears to me must be adopted; and the latter further ap- 

 pears to me in several respects preferable to the former. 



Arguments in There is no analogy in favour of the assumption, tlmt the 

 favour of the alkaline metals are compounds of the alkalis with hidrogen, 

 thesis, ^ '^^°" ^^^ rather an improbability. There are no grounds whence 

 we can infer, that the common metals are compounds of 

 what we at present regard as their oxides with hidrogen; 

 and unless we make this assutciption* there is an iiiiproba- 

 bility in the hypothesis, that the alkaline metals, which 

 have all the mecailic properties, are compounds of this kind. 

 An analogy can be traced connecting the earths with the 

 common metallic oxides, and a similar analogy connect* 

 the alkalis with the earths. This leads to the conclusion, 

 that all these substances are of similar chemical constitu- 

 tion; that the earths and th« alkalis therefore, like the 

 metallic oxides, are compounds of metallic ' bases with 

 ©xigen. 



The hypothesis I have advanced is in conformity with 

 both the-e analogies; it has therefore all the advantages, 

 which belonged to the theory given by Mr. Davy; and, 

 independent of the above facts, which appear to be subver-- 

 pive of that theory, it has some advantages which Mr. 

 Pavy's cannot claim. In particular the levity of the alka- 

 line 



