122 PROCEEDINGS OF SECTION B. 
More Exact ANALYSES. 
Whilst these developments were being pursued with 
scientific zeal, another band of workers were determining 
afresh the original data, and passing them through the 
crucible of more and more refined experiment. This de 
partment alone represents a hundred years of laborious work. 
Gay Lussac and Humboldt (by explosion with hydrogen), 
and a number of other chemists in Britain and on the Con- 
tinent, repeated the analyses, and found the composition of 
the air to scarcely vary from 21 per cent. of oxygen and 79 
per cent. of mtrogen by volume. Later, more accurate 
endiometric methods were devised, especially by Bunsen, 
and extensively employed by himself, Lewy, Angus Smith, 
and many others, in the course of extended researches; 
20-97 to 20-84 per cent. of oxygen, with probable error of 
0-03, may be taken as Bunsen’s numbers, varying very little 
from Cavendish’s results. 
The weight determinations by Dumas and Boussingault 
gave for normal air freed from moisture and carbon dioxide 
23 of oxygen to 77 of nitrogen, and there it remained. 
But, as we have seen, the normal atmosphere was known 
to contain aqueous vapour, carbon dioxide, and ammonia. 
These constituents were similarly determined, with the 
greatest exactness, by volume and weight methods. The 
aqueous vapour was naturally found to vary very consider- 
ably, the amount for saturation varying with the temperar 
ture ; the carbon dioxide, for so long regarded as constituting 
four parts in 10,000 of air, by volume, has been, by more 
rigid methods of analysis, reduced to three parts, or even 
less, per 10,000 of normal air. The ammonia, in combina- 
tion, has yielded, say, five parts per ten million of air, by 
volume; nitrous and nitric acids, probably as ammonia 
salts, occur in only minute traces. All these experimental 
results may be regarded as a wider confirmation, by more 
exact experimental determinations, of the results obtained 
towards the end of the 18th century. 
In 1840 Schonbem of Bale discovered ozone, an oxidising 
agent of great energy, and which proved later to be an allo 
tropic modification of oxygen. This interesting substance, 
with its phantom brother antozone, has been the subject of 
much controversy. Antozone had to retire in favour of 
another equally interesting substance, peroxide of hydrogen. 
Ozone, however, retained its popularity from the moment 
of its discovery, and has been generally acknowledged to 
exist in atmospheric air, playing the réle of purifier. 
