ON DEW, &c. 255 



than that of the atmosphere, provided the 

 oblique aspect of the sky was equally excluded 

 from the spots on which my thermometers were 

 laid. 



On the other hand ; a difference in tem- 

 perature, of some magnitude, was always ob- 

 served on still and serene nights, between bodies 

 sheltered from the sky by substances touching 

 them, and similar bodies, which were sheltered 

 by a substance a little above them. I found, 

 for example, upon one night, that the warmth 

 of grass, sheltered by a cambric handkerchief 

 raised a few inches in the air, was 5 greater, 

 than that of a neighbouring piece of grass, 

 which was sheltered by a similar handkerchief 

 actually in contact with it. On another night, 

 the difference between the temperatures of two 

 portions of grass, shielded in the same manner, 

 as the two above-mentioned, from the influence 

 of the sky, was 4. Possibly, experience has 

 long ago taught gardeners the superior advan- 

 tage of defending tender vegetables, from the 

 cold of clear and calm nights, by means of sub- 

 stances not directly touching them ; though I 

 do not recollect ever having seen any con- 

 trivance for keeping mats, or such like bodies, 

 at a distance from the plants, which they were 

 meant to protect. 



