404 Dr. Dalton on the Constitution of the Atmosphere, 



Average of thirteen experi- 1 20*92 oxygen 

 ments on town air, gave j per cent. 



On the last-mentioned day I received a bottle of air from 

 the summit of Snowdon through the care and attention of 

 my friend and pupil Mr. John Hall. It was corked and well 

 sealed with wax ; when opened under water a due portion of 

 that fluid entered. 



The average of the first two experiments gave 20*44 oxy- 

 gen per cent. 



The rest of the air after these two experiments was divided 

 into two portions, and entered into two phials for examination. 

 These were analysed a week or two afterwards. 



Average of five experiments with first phial gave 20*25 

 oxygen per cent. 



Average of four experiments, which emptied the first phial, 

 gave 19*98 oxygen per cent. 



Average of seven experiments of second phial gave 20*3 

 oxygen per cent. ; and a considerable portion was left. 



Average of the town air was during these experiments 

 nearly 2 1 oxygen per cent. 



I am not aware of any cause why this air was so much in- 

 ferior in oxygen to that on former occasions. 



1831, July 4. — Helvellyn air brought down from the summit 

 by me; wind S.W., with rain and fog. 



1. July 21. — Mixed two ounce measures of this air with 

 one of hydrogen, so as to make six separate and successive 

 explosions ; the hydrogen had ^^ths of a grain measure per 

 cent, of oxygen, and this is allowed for in the corrected results. 

 These results on the average gave 20*57 oxygen per cent. ; 

 the highest was 20*68, and the lowest was 20*43. 



The residues of the six explosions were collected, and found 

 to have 5 per cent, of hydrogen and 1 in 120 of oxygen. 



2. Mixed equal volumes of this Helvellyn air and the same 

 bottle of hydrogen used above, and fired the mixture in suc- 

 cessive portions. The average of six experiments gave 20*8 

 per cent, of oxygen. No oxygen was found in the residue. 



By comparing the results of 1 and 2, it would seem that 

 more oxygen is reduced from common air by firing equal 

 volumes of common air and hydrogen than by firing one 

 volume of common air with half a volume of hydrogen. 



August 23. — Mixed 100 measures of town air and 120 of new 

 pure hydrogen; this fired gave 21*5 oxygen per cent.; there 

 was no oxygen in the residue. This would seem to point out 

 ^J^th of oxygen in the hydrogen, yet nitrous gas scarcely 

 manifested so much. 



