yields, Alhalis, and their Compounds. 97 



— a conibiiiatiou certainly altogether improbable; and any ar- 

 rangement that can be conceived scarcely lessens the difficulty. 

 Mr. Dalton endeavoured to obviate this, by supposing, that in 

 the analysis of Berzelius the hydrogen is under-rated. But the 

 reverse is the case. The solution may now be easily given. In 

 the composition which properly constitutes oxalic acid, the pro- 

 portion of hydrogen is sufficiently large to present no difficulty. 

 And in what was considered as real oxalic acid existing in the dry 

 oxalates, there is no reason to suppose that hydrogen exists. It 

 is also obvious, that the proportion of oxygen and carbon in a 

 dry oxalate is that constituting carbonic acid ; for although in the 

 action of the acid on the base a portion of its oxygen is abstracted 

 with its hvdrogen, a corresponding portion of oxygen is added 

 from the base or metallic oxide, and a ternary compound is esta- 

 blished. 



The proportion of hydrogen indicated in the composition of 

 oxalic acid is not conformable to either of the two proportions of 

 carbon and hyilrogen, which constitute the two compounds at 

 present admitted as constituting the only definite compounds of 

 these elements, the carburetted and supercarburetted hydrogen. 

 It is much less even than that in the latter, which contains the 

 lower proportion. Yet there is every reason to conclude, from 

 the law which has been illustrated in reviewing the composition 

 of sulphuric acid, that it must exist in a definite relation to tke 

 simple radical of the acid, that is to the carbon, conformable to 

 the other relations which subsist between them. It follows, there- 

 fore, either that there is an error of analysis, in consequence of 

 which the proportion of hydrogen is greatly underrated, or that 

 there are other definite proportions in which carbon and hydro- 

 gen combine, than those which are at present admitted. The 

 coincidence in the results of the analysis by Gay-Lussac and by 

 Berzelius, in a great measure precludes the former supposition ; 

 and indeed an error so great would require to be assumed, as 

 cannot be supposed. The other conclusion therefore follows : it 

 is rendered more probable by other considerations, vvhicii give 

 force to the opinion that hydrogen and carbon enter into more 

 numerous proportions than have been assigned : And it is nearly 

 established by tlie results of this case itself. Supercarburetted 

 hydrogen is composed of 100 of cartoon with 17"5 of hydrogen; 

 carburetted liydrogen of 100 with .'J.'). In oxalic aciil, 2()-5 of 

 carbon are combined, according to the analysis of Berzelius, with 

 2\) of hydrogen, which is in the proporiion of 100 to !i'-l. Now 

 this deviates little, and not more than what may fairly lie refer- 

 red to inaccuracy in the estinuition of the proportions in one or 

 other of the compounds, from the fourth of the highest propor- 



Vol. 54. No. 2.j(). Auii, ISUi. G tioii 



