I 375 1 



tefides the heats below appeared to have been influenced by re- 

 flexion, and the heat above by contad with the wall. Thus on 

 the i6th of June, 1784, the heat below being 68°, the heat above 

 was no more than 60*; fo great a diminution at fo fmall a height 

 could not arife from the natural progreflion of heat, but muft 

 have arifen from the range of a N E. wind which prevailed on part 

 of that day from which the garden was more (heltered. 



Section II. 



Of the Temperature of the Winter Months, 



By winter months I underftand thofe of November, December, 

 January and February. The temperature of thefe months has pre- 

 fented fome extraordinary phasnomena hitherto deemed inexplica- 

 ble, or at leafl: not fully accounted for, though they appear to me 

 conneded with and dependant on a fad fuggefted by that eminent 

 philofopher and mathematician Dr. Halley, fo long ago as the 

 latter end of the feventeenth century. 



The firft phenomenon is, that during the winter months, the 

 temperature of the higher ftrata of the atmofphere is often warmer 

 than that of the lower. 



3 A 2 Thus 



