194 



Prof. J. Le Conte on the Influence of 



in the rate of combustion do not exceed the probable limits of 

 experimental error. In this case, therefore, the question to be 

 determined is, whether the differences in the rapidity of burning 

 observed on different days can be explained by the variations of 

 the barometer and thermometer ? For this purpose I shall take 

 the average of each pair of experiments, as a nearer approxima- 

 tion to the correct rate of burning on each of the three days. 

 The following Table, in which the relative densities of the air 

 have been calculated by combining the effects of barometric and 

 thermometric oscillations, will serve to illustrate this point* : — 



It will be seen that the rate of combustion increases in a de- 

 cidedly higher ratio than the density of the air. If therefore we 

 assume that the rapidity of burning is, cceteris paribus, in the 

 direct ratio of the density of the air, it follows that some other 

 agency must have cooperated in these eases. 



3. The results of Mr. MitchelFs experiments at different alti- 

 tudes may in like manner be subjected to a similar numerical 

 test. The subjoined Table is constructed from the data contained 

 in that which is given on a previous page : — 



This comparison places in a still stronger light the fact, that 

 the augmentation in the rate of burning increases in a somewhat 

 higher ratio than the density of the air, while at the same time it 

 clearly demonstrates the controlling influence of atmospheric den- 



* In making these calculations I used the following formula, based upon 

 the two weU-kiiown physical laws, that the density of any permanent gas 

 varies directly as the compressing force, and inversely as the volume : — 

 , ,, b . b' . 



l+0-002036(<-32) ' 1 +D-002036(<'— 32) ' 



in which d and rf' represent the densities; b and b' the barometric heights 

 reduced to the freezing-point ; and t and t' the temperatures on Fahren- 

 heit's scale. 



i 



