Solar Light on Combustion, 



193 



two effects to a quantitative estimation, by using Mariotte's law 

 and Regnault^s coefficient of expansion for air. 



1. Presuming that each set of experiments made by Dr. 

 M'Keever, alternately in the dark and in the sunshine, were 

 performed on the same day, and therefore under identical baro- 

 metric conditions, we may form some estimate of the adequacy 

 of temperature to account for the difference in the rate of burn- 

 ing observed by him. The subjoined Table, which I have con- 

 structed from the data previously given, will place this in a clear 

 light :- 



The remarkable discrepancies indicated by the numbers in the 

 first column of ratios afford a striking illustration of the exist- 

 ence of some disturbing cause, tending to vitiate the accuracy 

 of these experiments. But a glance at the numbers contained 

 in the two columns of ratios is sufficient to show that tempera- 

 ture alone is entirely inadequate to account for the diminished 

 rate of combustion in the sunshine. A remarkable difference is 

 observed in the rate of consumption in expei'iments 1 and 2. 

 No. 1 was made with a " green wax taper," and No. 2 with a 

 " taper;" but, as from the context, the second experiment appears 

 to be a repetition of the first, the presumption is, that the same 

 kind of taper was used in both cases. The rate of burning in 

 experiments Nos. 3 and 4 was determined by the time required 

 to consume a given length ; and as one of them was made with a 

 mould-candle and the other with a taper, no comparison can be 

 extended to them, so far as the rates of consumption in these 

 two cases are concerned. The irregularities exhibited in these 

 results probably arose from the agitations of the atmosphere, 

 which were incident to the method of exposing the burning 

 body to the sunshine in the open air. As the excess of the con- 

 sumption in the dark varied from 5 to 11 per cent., whereas the 

 excess in the density of the air was only 2'3 per cent., it is evi- 

 dent that some other cause than temperature must be evoked to 

 explain the difference. 



2. In my experiments, the conditions were such as to elimi- 

 nate the effects of temperature on the results obtained in the dark 

 and in the sunshine on any given day; and it has been shown 

 that for each pair of experiments thus conducted, the variations 



Phil. Mag. S. 4. Vol. 16. No. 106. Sept. 1858. 



