Solar Liijht on Combustion. 187 



sunshme, and in the other the excess was in favour of the dark^ 

 but the differences are probably withui the hmits ol the expeii- 

 mental inaccuracies incident to such dehcate investigations 



These negative results are the more striking from the tact that 

 if solar hght exercised the decided influence on the process ot 

 combustion which Dr. M'Keever's experiments seem to mdicate, 

 we should expect the effects to be much more marked and con- 

 spicuous when the hght was increased in intensity from e^^A^ to 

 ?mfold by the concentration of a lens. The fact that the rays of 

 the sua traversed the glass lens before they fell on the flame, can 

 scarcely be urged as a possible explanation of the discrepancy ; 

 for Dr. M'Keever obtained analogous results when he employed 

 lanterns. (FzWe experiment No. 5.) 



The obvious variation in the rapidity of combustion on differ- 

 ent days (as exhibited in my experiments) dlustrates m a most 

 striking manner the decided influence exercised on the process 

 by con^paratively slight alterations in the external conditions 

 This fact should inspire us with wholesome caution, and check 

 the spirit of rash generahzation. Throwing out of consideration 

 the possible fluctuations in the rate of burning arising from want 

 of homogeneity m the combustible materials and imperfections 

 in the mechanical arrangements by which they are consumed, 

 there are three external conditions which may be supposed to ex- 

 ercise more or less influence on the rapidity of the process ihese 

 are 1st, barometric pressure : 2nd, the temperature of the air; 

 and 3rd, the amount of aqueous vapour present. 1 propose to 

 consider each of these separately. 



1. Barometric Pressure. 

 From a priori considerations, we should be led to expect that 

 an increase of barometric pressure, through the consequent con- 

 densation of the air, would, cceteris paribus, tend to augment the 

 rapidity of combustion, by furnishing the burning matter with 

 a greater amount of oxygen in a given volume. Unfortunately, 

 direct experiments are wanting to test this in as satisfactory a 

 manner as we should desire. The older experiments m the 

 Boylean vacuum, inasmuch as they relate to the degree of rare- 

 faction at which combustion ceased, do not give us information 

 in re-ard to the rapidity of the process at the various stages ot 

 exhaustion. The same remark applies to the later expenments of 

 Grotthuss, as well as to the admirable -Researches" of Sir Hum- 

 phry DavJ on the "Effects of Rarefaction, by partly removing 

 the Pressure of the Atmosphere, upon Flame and Explosion*. 



* Vide Daw's " Researches on Flame," Phil. Trans for 1817, P- f et 

 ,e,,. AUo' Works of Sir II. Davy,' edited by Dr. John Davy, vol. vi. 

 Y :)\ el seq. London, 1810. 



