on tJie Atomic Weight of Carbon. 213 



order conjointly to submit the atomic weight of carbon, as at 

 present received, to a severe and accurate scrutiny. 



It is known that two of the most distinguished natural phi- 

 losophers, Biot and Arago, have, by means of the direct 

 weighing of carbonic acid, fixed upon the number 1*519 for 

 the specific gravity of that gas. Their experiments were re- 

 peated by Dulong and Berzelius, with whom, as regards skill 

 and talent, conscientiousness and accuracy, no others can be 

 compared. The two last observers found the number 1*524 

 for the specific gravity of carbonic acid ; that obtained by 

 De Saussure is 1*5269. 



The atomic weight of carbon, as calculated from the first 

 of these, is 75*530, and from the other, 76*437. There is no 

 known gas more easily obtained in a pure state, or which can 

 more easily be distinguished from a foreign body, than car- 

 bonic acid. Any admixture of atmospheric air, or of other 

 gases, can only lower its specific gravity. 



Experiments have lately been conducted by Rudberg on 

 the dilatation of gases under the influence of heat*, from which 

 he calculates that the coefficient of dilatation is somewhat 

 less than was previously supposed ; should these experiments 

 be correct, the proof of which still remains to be made known, 

 they do not influence the specific gravities of two gases as 

 determined at the same temperature, even supposing the re- 

 duction to the normal temperature be made according to the 

 coefficient of dilatation as hitherto received ; if weighed at 

 unequal temperatures, however, a difference is observed. 



In the experiments of Dulong and Berzelius, atmospheric 

 air was weighed at 20^° C, and carbonic acid at 18° C. ; con- 

 sequently the reduction of the gas to 0° C, according to the 

 former coefficient of dilatation, gives the weight of air some- 

 what too high, and since this, in an equal volume, represents 

 the diviwry the specific gravity of carbonic acid is estimated 

 rather too low ; in all cases, however, the differences fall 

 within the limits of the errors of observation. 



When we remember that the determinations of the specific 

 gravities of these gases were conducted with the same balloon, 

 the same scales and weights, and at temperatures varying 

 very little from each other, we ought not to call in question 

 their correctness without the strongest and most convincing 

 reasons. 



During the last twelve years, a great number of weighings 

 of the vapours of volatile bodies very rich in carbon, have 

 been undertaken in reference to this point by Gay-Lussac, 

 * See Taylor's Scientific Memoirs, vol. ii. pp. 507, 514, 543. 



