ON DEW, &c. 199 



laid a thermometer on any downy substance in 

 contact with the snow, he would, in all proba- 

 bility, have found a cold indicated by it at least 

 20 greater than that of the air, as marked by a 

 naked instrument, and consequently at least 22 

 greater than the real cold of the surrounding 

 atmosphere. 



3. Mr. Wilson's place of observation was not 

 very favourable to the occurrence of a great 

 cold, from radiation of heat at night, it being 

 near to a large smoky city, in the immediate 

 vicinity also, as appears to me from what he 

 says of it, of one or more considerable buildings, 

 and in a climate abounding in moisture. 



4. None of Mr. Wilson's experiments, in 

 which a very great degree of cold occurred, 

 were made within an hour or two after sunset, 

 during which time, according to my observa- 

 tion, the most considerable differences between 

 the temperatures of the air, and of bodies on 

 the surface of the earth, commonly happen. 



If, then, such experiments should be made 

 in an atmosphere still colder than that, in which 

 Mr. Wilson made his, on a large plain remote 

 from any city, and free from objects of every 

 kind, that are elevated above the ground, and 

 in a country remarkable for the dry ness of its 

 air, all which circumstances may be found in 

 Russia during the winter j a difference of at 



