56 POPULAE ERRORS. 



apparent effect of the nioon in dispersing the clouds 

 can be easily observed on any moonlit night when 

 the clouds are thin. At such a time the moon 

 always appears to be shining through an open space 

 in the clouds, though the latter may be moving rap- 

 idly. A little attention will convince any one that 

 in such cases the light of the moon has merely 

 illuminated the clouds and rendered them less visi- 

 ble, but no less real. The opposite belief regarding 

 the effect of the moon on temperature is that moon- 

 beams are cold. One proof of this is supposed to 

 exist in the fact that on a clear night, when the 

 moon is shining, plants are more liable to be 

 injured by frost than on cloudy nights, when its 

 "cold rays " are prevented from reaching the earth. 

 That plants are more liable to be killed by frost on 

 clear nights than on cloudy nights is a fact too well 

 known to be disputed, but that the presence of the 

 moon on clear nights has any material effect on the 

 temperature is not established. The misapprehen- 

 sion in this case will be removed by considering 

 that the effect of the radiation of heat from the 

 earth or any other body is to lower its temperature, 

 and that a covering of any kind, as of clouds far 

 above the earth, or of any artificial object nearer 

 to its suf ace, prevents to some extent the diminution 

 of temperature, by retarding radiation. 



The presence or absence of dew likewise'depends 



