OF ARTS AND SCIENCES. 65 



matic trough. In so far as charcoal is concerned, De Saussure * has 

 long ago shown the extreme probability that it is oxidized by the 

 air, even when dry, as it is when wet. Neither do we wish to 

 assert that it is impossible to deprive air of every atom of its car- 

 bonic acid. We insist only upon the facts, that it is a matter of no 

 inconsiderable difficulty to do this, that Karsten's apparatus was en- 

 tirely inadequate, and that nothing in his paper would indicate that he 

 has allowed for this source of error. 



It should be distinctly borne in mind, that in the experiments of 

 Karsten, as well as in our own, the question raised is not at all whether 

 the amount of carbonic acid which escapes absorption can be esti- 

 mated with the balance ; for so long as the experiments are qualitative 

 only, and conclusions are based upon the precipitate which is formed 

 in lime-water, it is clearly necessary to remove every trace of carbonic 

 acid from the air employed, no matter how " imponderable " this trace 

 may be. We do not believe that the carbonic acid which escapes absorp- 

 tion in ordinary experiments can be of sufficient amount to be mentioned 

 as a source of inaccuracy in the determination of the cai'bonic acid 

 of the air, by the method which has been used by so many eminent 

 chemists ; for the extent of the error thus introduced must be far less 

 than that of several others to which the absorption process, as com- 

 monly employed, is exposed, and which have been pointed out by 

 Hlaziwetz,t and in part also by the brothers Rogers. J 



So far as we know, those observers who have previously touched 

 upon this subject have been occupied with quantitative considerations 

 only. They have, therefore, very properly rested content, when by 

 experiment they have satisfied themselves that the last potash-tube of 

 their series no longer increased in weight during the space of time 

 occupied by a single experiment. § It must, however, be evident to 

 any one who will perform the experiment, that the presence of an 

 amount of carbonic acid which could not be detected by any weighing 

 of potash-tubes may readily be made manifest by precipitating it as 

 crystallized carbonate of lime. In this connection it should be men- 



* Gilbert's Ann. der Phys., 1814, XLVII. 119, note, 

 t Wiener Akad. Bericht, 1856, XX. 189. 



X Am. J. Sci., 1848, [2.] V. 115 ; Edin. New Phil. Journ., XLIV. 150. 

 § Compare, for example, Dumas and Stass, Sur la v&itable Poids atomique du 

 Carbone, Ann. Ch. et Phys., [3.] I. 18. 

 VOL. V. 9 



