Solar Light on Combustion. 195 



sity on the phrenomenon of combustion. The extreme rates of 

 burning are as the numbers 100 to 127, while the corresponding 

 densities of the air are as 100 to 121 nearly; in the other cases 

 the approximation to identity in the ratios is still closer. Would 

 the variations in the hygrometric state of the atmosphere, which 

 we have left out of consideration, explain this discrepancy ? In 

 the absence of the experiments necessary for testing this question, 

 it would be premature to hazard any conjecture. I may remark, 

 however, that in the case of Mr. Mitchell's experiments, the cor- 

 rection for the effects of aqueous vapour would probably in one 

 point of view operate in the wrong direction, and thus tend to 

 increase the discrepancy in the ratios. For as the temperature 

 was decidedly higher at the lower stations, it is more than pro- 

 bable that the tension of vapour was greater there than at the 

 upper ones, and consequently that its influence in retarding 

 combustion should be relatively greater at the points nearer the 

 sea-level. This of course would tend to equalize the rates of 

 burning at lower and higher altitudes, when no correction is 

 made for this cause. On the contrary, it is obvious that the 

 influence of vapour having a given tension in altering the relative 

 amount of air in a given volume, must be greater when the baro- 

 meter is low. From this cause the aqueous vapour at the upper 

 stations might have had a greater effect in retarding combus- 

 tion, and thus tended to exaggerate the difference in the rates of 

 burning. 



The comparatively large rate of consumption indicated by my 

 first experiment of the 9th of May (being more than 9 per cent, 

 above the other) was most pi-obably attributable to a combina- 

 tion of causes. All the three external conditions concurred in 

 accelerating the process. The barometer was high, the tempera- 

 ture low, and the atmosphere excessively dry. The last-mentioned 

 condition was accidentally forced upon my attention from the 

 fact, that on that day I failed in an experiment for determining 

 the dew-point by means of Daniell's hygrometer*. 



* Collaterally related to this subject are the eiFects of condensed and 

 rarefied air and of temperature on the process of respiration and the elimi- 

 nation of carbonic acid in men and other warm-blooded animals. M. Le- 

 gallois found that, when warm-blooded animals breathed air under pressure 

 reduced to 11 '811 inches, the amount of oxygen consumed was diminished 

 {Ann. de Chim. et de Phys. vol. iv. p. 113. 181/). M. Theodore Junod's 

 experiments show that condensed air produced dee]) inspirations and an 

 agreeable glow throughout the system, while rarefied air had an opposite 

 effect. {Archives Generates de M^decine, ser. 2. vol. ix. p. 15/. Paris, 18.35. 

 Also Magendie's Report on the same Memoir, Comj'tes Rendus, vol. i. p. (JO. 

 Paris, 1835.) The observations of M. Triger, already referred to, indicate 

 analogous ett'ects on those who laboured in the condensed air. They could 

 do double work without fatigue, and even old asthmatics seemed to re- 

 cover their vigour. {Comptes Rendus, vol. xiii. p. 884 e/ se^. Paris, 1841.) 



02 



