Lerts & Bhake— The Carbonic Anhydride of the Atmosphere. 221 
The vessels before use were thoroughly washed—not with acid, for 
after cleansing a flask with acid, and subsequently using it for the 
absorption of carbonic anhydride, even though the acid had been 
completely washed out, the results were found to be higher than if the 
flask had been merely washed out with water. The flasks were, how- 
ever, very thoroughly washed for the first experiments, and in all 
future experiments they were only rinsed out, and the slight skin of 
barium carbonate with which they were coated allowed to remain. 
As soon as the distilled water had drained from the flasks, they were 
dried by blowing warm air through them by the foot bellows. 
To fill them with the air sample to be examined, the flasks were 
allowed to remain where the air was collected for some time, to come 
to the temperature of their surroundings, and a quantity of air was 
then pumped through them sufficient to displace the original volume 
five times, as Pettenkofer recommends. 
Both flasks were filled as quickly as possible, each experiment being 
made in duplicate. 
The absorbing solution was then added by means of a pipette, the 
flasks securely closed with india-rubber corks, and absorption allowed 
to proceed for two hours. In all the operations connected with the 
absorption and subsequent titration, I have followed Pettenkofer’s 
description of the process as closely as possible. 
After absorption, the solution was rapidly transferred to a very 
narrow beaker, the pipette rinsed out with a small portion of the 
solution, and then an aliquot part—30 cubic centimetres of the 
45 cubic centimetres originally added—titrated with the standard 
oxalic acid, and the amount of carbonic anhydride deduced. 
For the standard oxalic acid solution, the pure acid of commerce 
was taken, and recrystallized quite a number of times after the manner 
indicated by Fresenius. 
About 20 per cent. was first crystallized from the solution. This 
contained the impurities, which in the first case were extremely minute. 
The remainder of the solution was then carefully evaporated and 
allowed to crystallize. 
The resulting crystals were again taken, and they underwent similar 
re-crystallization, until acid was obtained, which, on ignition, just left 
the merest speck of impurity. Its strength was determined by standard 
permanganate solution, which had been factorized by pure ferrous 
ammonium sulphate. From this pure acid a solution was prepared, 
one cubic centimetre of which was equal to one milligram of carbonic 
anhydride. 
R2 
