158 ESSAY 



times, on its appearance betwixt daybreak and 

 sunrise, found the difference between thermo- 

 meters on grass and in the air, which had been 

 considerable during the night, to diminish 

 greatly. I never, indeed, observed it to vanish, 

 but this I used to impute to the air being not 

 very much obscured. I have now, however, 

 reason to doubt the justness of this conclusion; 

 for on the evening of the 1st of January in the 

 present year, 1814, I found, during a dense fog, 

 while the weather was very calm, a thermo- 

 meter lying on grass, thickly covered with 

 hoarfrost, 9 lower than another suspended in 

 the air, 4 feet above the former. On the fol- 

 lowing evening, when the air was equally calm, 

 but the fog sufficiently attenuated to allow me 

 to see that the sky was almost entirely covered 

 with clouds, the difference between two ther- 

 mometers, similarly placed with the former, 

 was only 1. On comparing the observations 

 of these two evenings, I conclude, that on the 

 first few or no clouds existed above the fog, 

 and consequently that fog, if there be no clouds 

 above it, may, in a very calm air, admit of the 

 appearance of a considerable degree of cold, at 

 night, upon the surface of the earth, in addition 

 to that of the atmosphere. Mr. Six, indeed, 

 says, while speaking of the cold connected with 

 dew, in his paper in the Philosophical Transac- 



