Chemical Analysis of the Atmosphere. 285 



If the lime is too dry, it absorbs the carbonic acid very imperfectly ; 

 if too moist, it is difficult to make it pass the bend of the tube. By 

 a few trials the right point may be easily obtained. 



2. The sulphuric acid included in the end of the tube, is placed 

 in the same manner as in the determination of moisture. From fifty 

 to sixty drops are always sufficient. The effect of this acid is evi- 

 dent. The lime in passing from the stale of hydrate to that of car- 

 bonate, abandons its water, which would cause a diminution of weight, 

 as well as the portion of water (still greater) carried along by the cur- 

 rent of air passing through the tube. It is in order to retain this wa- 

 ter that the hygrometric portion of the tube becomes necessary. 



3. The air before it enters into the tube containing the lime, ought 

 to pass through a common hygrometric tube, in order to deposit its 

 water, which, without this precaution, would be confounded with the 

 carbonic acid. 



4. All the weighings ought to be done by substitution, that is to say, 

 in taring [en tarant) a known weight, — for example, .2 of a gramme, 

 with the tubes, and on the same plate of the balance, and in deducting 

 from this weight, that required to be substituted for it after the oper- 

 ation, — the difference indicating the increase of weight of the tube 

 produced by the operation. We should never neglect to wipe the 

 tubes immediately before weighing, since the moisture which they 

 attract from the air, may, in a certain time, occasion an error of some 

 milligrammes, as I have ascertained by direct experiment. 



5. The volume of air operated upon ought to be large, in order 

 that the increase of weight of the lime may not be too small. In 

 drawing 12972.5 grammes of water from the vessel A, I have ob- 

 tained, in ordinary circumstances, an increase of seven to nine milli- 

 grammes. It is therefore preferable to operate upon a volume of 

 twice this size, which may be done without difficulty in the course 

 of fifty minutes. 



6. To calculate the proportion of carbonic acid in the air analysed, 

 we set out with the specific gravity of this gas, as determined by Ber- 

 zelius and Dulong, and which gives for 1.97978 grammes of carbo- 

 nic acid its volume equal to one thousand cubic centimetres at 0° 

 and at 0.76 of the barometric column, (=29.52 in.) The volume 

 found by calculation is reduced to the volume which it would have 

 at the temperature of the vessel A and the prevailing height of the 

 barometer, and from the sum representing the volume of analysed 

 air, we obtain, by the rule of proportion, the volume of carbonic acid 

 in a given volume of air, e. g. in ten thousand parts. The reduction 



