190 



Dr. W. Marcet. On the 



subsidence of the precipitate, with my instrument the analysis can be 

 made at once, and in a short time. It might be objected that Petten- 

 kofer's method conld be carried out as far as the combination of the 

 carbonic acid with the barium, the fluid being preserved for analysis 

 in well-stoppered bottles. A number of determinations in succession 

 might thus be obtained partially made, the final analysis being put off 

 to a future period. I had to adopt this very plan myself in my 

 experiments on the influence of altitude on the chemical changes of 

 respiration, but found it inconvenient in many ways and sometimes 

 objectionable. I am now alluding to air containing over 1 per cent, of 

 carbonic acid ; so far my experience does not extend to atmospheric 

 air. I hope soon, however, to be able to determine carbonic acid in 

 the atmosphere by the same means, as I have had a much larger in- 

 strument made for that purpose, each bell-jar holding 3 cubic feet of 

 air. 



In order to test the accuracy of my instrument, it was necessary 

 to compare analyses of the same air by this means and by another 

 reliable method, and for that purpose I selected that of Pettenkof er 

 which was carried out with one of the cylinders I have formerly 

 described in the "Journal of the Chemical Society," 1880. 



The following are eighteen of the comparative analyses made by 

 the two methods : — 



Carbonic Acid in 100 parts of Air analysed. 



« 



First Analysis. 



1st — Volumetric method 4*694 



2nd „ „ 4-633 



Mean 4*663 



By Pettenkofer's method 4*642 



Difference 0*021=0*45 per cent. 



Second Analysis. 



Volumetric method . . „ 4'560 



By Pettenkofer's method 4*580 



Difference 0*020=0*4 per cent. 



Third Analysis. 



Volumetric method 4*495 



By Pettenkofer's method 4*435 



Difference 



1*060=1*3 per cent. 



