Letts & Buake— The Carbonic Anhydride of the Atmosphere. 118 
measure its volume. This principle was originally suggested by 
H. Mangon, in 1875, to Tissandier,' who desired to investigate 
the upper regions of the air during balloon voyages, and was 
employed by the latter for the purpose, necessitating as it did only 
simple operations at the place of observation, and permitting the 
actual determinations to be made later in the laboratory. Mintz 
and Aubin employed as absorbing vessels, glass tubes about 90 cm. 
long and 2 cm. bore, drawn out at each end and filled with pumice 
moistened with strong caustic potash solution. A number of 
these tubes could be prepared on any convenient occasion, and 
when sealed up at their two ends were ready for use at any time. 
When a determination had to be made, the ends of a tube were 
broken off and a given volume of the air aspirated through it at 
a rate of about 3 litres per minute, the total volume being from 
200-300 litres, after which the two ends of the tube were again 
sealed up. 
The apparatus for liberating and measuring the absorbed car- 
bonic anhydride was somewhat complicated, but the essential 
features were as follows:—The ends of the tubes having been 
broken off, one extremity was connected with a mercury pump 
and gas measuring apparatus and the other with a vessel contain- 
ing dilute sulphuric acid, the junction with the latter being closed 
with a pinchcock. A vacuum having been established, acid was 
admitted into the tube (heated by a steam jacket), from which it 
flowed into a boiling flask connected with a Liebig’s condenser, 
and the evolved carbonic anhydride, mixed with a little air, was 
drawn over into the measuring apparatus. Finally the actual 
volume of carbonic anhydride was determined by absorption with 
caustic potash. 
Mintz and Aubin are among the very few observers who have 
tested their method by a blank experiment with a known quantity 
of carbonic anhydride. To do this they aspirated a definite 
volume of air freed from carbonic anhydride through a boiling 
flask containing acid, into which a measured volume of sodium 
carbonate solution of definite strength was gradually added—the 
experiment being so conducted that the resulting mixture of air and 
carbonic anhydride corresponded with ordinary ‘fresh ” air. This 
1 Tissandier, ‘‘ Compt. Rend.,’’ 80 [1875], p. 976. 
