Lerts & Bhake—The Carbonic Anhydride of the Atmosphere. 189 
Any diminution of atmospheric carbonic anhydride by the 
action of vegetation should correspond with an increase of atmos- 
pheric oxygen, of precisely the same amount, and it is therefore of 
interest to enquire into the variations which have been observed in 
the quantity of atmospheric oxygen. ‘That variations do occur, 
has been recognized by all who have undertaken analyses of air; 
but, as in the case of atmospheric carbonic anhydride, the older 
observers found a larger variation than actually occurs. Some of 
their more important results were as follows! :— 
Percentage of Oxygen. 
Variation. 
Maximum. | Minimum. 
Gay Lussac and Humboldt, . : : 3 21°20 20-90 0°30 
De Saussure, F ; : : ‘ ; MoT 20:98 0°17 
Regnault, . : ‘ : ‘ : : 21-00 20-90 0-10 
Bunsen, . 4 : : : F : 21:18 20°80 0°38 
Coming to more recent work on the subject we have— 
Percentage of Oxygen. 
Date. Observer. Seren ae al MAT at Olt. 
Maximum. | Minimum. 
1875-76 Jolly, 2 5 6 ; 5 20:960 20-470 0°490 
1877 29 6 0 é ‘ : 21-010 20°530 0°480 
1883-84 Kreusler, . 9 3 : 5 20°991 20-867 0°124 
55 33 (deducting exceptional 20°940 20°880 0-060 
amounts), 
1885 Hempel, 6 6 6  § oi] COO. |) Bos 0-094 
In 1886, the excellent idea occurred to Hempel of inviting the 
cooperation of several observers in different parallels of latitude, to 
collect samples of air at the same time (making due allowance for 
differences in longitude), for the purpose of oxygen determinations. 
1 The figures are taken, partiy from Roscoe and Schorlemmer’s ‘‘ Treatise on 
Chemistry,” and partly from Angus Smith’s ‘‘ Air and Rain.”’ 
P2 
