OBSERVATIONS IN FOUR BALLOON ASCENTS. 
;j39 
Dr. Miller’s Analysis of Air collected in the Ascents. 
“King’s College, London, 5 May, 1853. 
“ My DEAR Sir, — The following particulars of my examinations of some of the 
specimens of air collected by Mr. Welsh in the course of the balloon ascents made 
under the superintendence of the Kew Committee of the British Association, may 
not be unacceptable to the Fellows of the Royal Society as supplementary to a part 
of Mr. Welsh’s report and observations. 
“The samples of air collected upon the 26th of August appear to have been taken 
in the most unexceptionable manner, and it was upon these only that my experiments 
were made. The recipients for the air were wide glass tubes, about 5 cubic inches 
in capacity, to each of which a portion of barometric tubing, 3 or 4 inches in 
length, was attached, as a neck that might receive a cap and stopcock, and which 
would admit also of being hermetically sealed afterwards by the blowpipe. Two of 
these tubes v/ere furnished with excellent stopcocks, and were found able to support 
without leakage for twenty-four hours the exhaustion obtained by an air-pump, the 
gauge of which indicated a pressure of 0’5 inch. 
“ Having been thus tested they were exhausted to this extent immediately before 
the ascent took place, and were filled with the specimens to be examined by simply 
opening and then closing the stopcock, the altitude being determined by an observa- 
tion of the barometer at the moment. In the third tube, a Torricellian vacuum was 
obtained, the tube being then sealed and drawn off, so as to admit of being broken 
at a filemark when the air was to be collected ; after the specimen had been thus 
obtained, the aperture was closed by thrusting the neck of the tube into a cap filled 
with softened wax, 
“The tubes were within twenty hours after tlie air had been collected hermetically 
sealed by myself, and the proportions of oxygen and nitrogen determined with great 
care by detonation with hydrogen in ‘ Regnault’s Eudiometer.’ 
“The volumes of oxygen found in the air collected at different altitudes are given 
in the following table : — 
Altitude. Volume of oxygen. 
Air collected at King’s College 20-920 
Tube 2 13,460 feet . . . 20-888 
Tube 3 18,000 feet . . . 20-/47 
Tube (G 1), Torricellian vacuum . . . 18,630 feet . . . 20-888 
“ From these observations it would appear that the composition of the atmosphere, 
as regards the proportion of oxygen and nitrogen, scarcely varies more as we ascend 
through the first half of that atmosphere (for at an altitude of about 3^ miles one-half 
of the atmosphere lies beneath us), than it is found to vary at different spots upon 
the surface: that there is, in fact (as Gay-Lussac had long since announced as the 
result of his experiments, made at a time when the methods of gaseous analysis were 
