Intelligence and Miscellaneous Articles. 501 



of air never exceeded io^th, and most frequently was not more J^Ii^n 



loo.ooo''"' i'ti J'vj 



The various samples of air which I analysed were collected in 

 bulb-tubes with the two ends drawn out and open ; the capacity of 

 these tubes was about 100 cub. centims. The air was collected in 

 the following manner: — One of the extremities of the tube was 

 connected, by means of caoutchouc tubing, with a little pair of 

 bellows, which was moved sufficiently long to be certain that the 

 whole of the air contained in the tube was replaced by the atmo- 

 spheric air of the locality, taking the necessary precautions to avoid 

 any mixture of the air of respiration ; as soon as the air was col- 

 lected, the tubes were sealed b(^fore the lamp. 



1 have divided the analyses into three series : — 1st, analysis of the 

 atmospheric air of France ; 2nd, analysis of the atmospheric air of 

 the Atlantic Ocean and Carribean Sea; 3rd, analysis of the air of 

 New Granada. 



The comparison of the results obtained in this investigation with 

 previous ones shows that the constitution of the atmosphere is nearly 

 the same in the New and the Old World. Taking the mean of the 

 analyses, executed in eleven different localities of New Granada, we 

 find that 10,000 vols, of normal atmospheric air contain 4-'008 carbonic 

 acid, 2101*425 oxygen, and 7894<*557 nitrogen, which are nearly 

 the same proportions as those which have been found for normal 

 atmospheric air in various parts of Europe. However, on examining 

 carefully all the experiments hitherto made on the constitution of 

 the atmosphere, it is readily seen that the composition of the air is 

 not absolutely constant. Perceptible differences exist, which vary 

 with the meteorological conditions ; thus after a long rain the car- 

 bonic acid and oxygen are always in smaller proportion than after 

 a long drought; however, these differences are only appreciable 

 when the analysis has been carried out with very great accuracy. 



In the New World, where the seasons are more defined than in 

 Europe, these variations are more easily detected. During the fine 

 season the normal air always contains a little more oxygen and a 

 little more carbonic acid than in the season of the rains. Thus, 

 taking the mean of a large number of analyses, I found that 10,000 

 vols, of normal atmospheric air of Bogota contain — 



With a With a ,\ 



cloudy sky, and bright sky, and 



nO)j 

 during the rains. during^the fine season. '^^'"^ 



Carbonic acid 3*822 4*573 



Oxygen 2099*542 2102-195 



Nitrogen 7896*636 7893*232 ^ ; 



The difference which exists between the atmospheric air of the 

 two seasons is therefore on an average 0*751 for the carbonic acid 

 and 2*653 for the oxygen in 10,000 vols, of air. 



Taking the maximum and minimum of the results obtained in the 

 analyses performed during the two seasons, we have — 



