M. Fraunhofer on the Laws ()f Light. 251 



bill both appear evidently to have been very much affected by 

 the heat of the sun. 



The experiments at and immediately after sunset prove 

 the rapid ascent of the heated air from below that takes place, 

 and the descent of cold air from above that produces the sud- 

 den chill we feel at that time. I have frequently observed, 

 that, on insulated heights, the thermometer appeared to rise a 

 little about the time of sunset, and even when that was not 

 the case, that it did not fall so soon nor so fast as on the plain 

 below. So that it often happens where the difference of 

 height between the two stations is not very great, that the an- 

 just after sunset is warmer above than it is below. I have ob-^ 

 served this in hard frost, and in very warm weather. 



The foregoing very meagre experiments are all that I have 

 yet been able to make with my balloon, as I. have in vain at- 

 tempted to discover some means of protecting the thermome- 

 ter from the influence of the sun. But I shall be very happy, 

 if you will come here in the summer, to pursue them further 

 with you. 



You will observe, that, from a quarter after five o'clock, till 

 three quarters after seven, the temperature of the air, at 1340 

 feet above the earth, was only lowered three-fourths of a de- 

 gree in those two hours and a half. Whilst, in the same 

 time, the thermometer below fell six and a fourth degrees. 

 The difference between the height of the hill and of the ground 

 before the house may in part explain the smallness of the 

 change of temperature in that time, but this would apply 

 equally to both thermometers. 



MiNTo, 5th April 1827. • 



Art. IX. — A short account of' the results of recent Experiments 

 upon the Laws of Light and its Theory. By M. Le Che- 

 valier JFraunhofer, Member of the Royal Bavarian 

 Academy of Sciences at Munich. (Concluded from Vol. 

 vii. p. 113.) 



It would be difficult for the most ingenious natural philoso- 

 pher to derive immediately from the results of these experi- 

 ments a lax0 of the phenomena. 



