MELLONI OX THE NOCTURNAL COOLING OF BODIES. 473 



we read tlie following : — " For dew, it is sufficient to remark, 

 that the temperature of the air being, for example, 15° at a cer- 

 tain period of the night, there Avill be bodies at 14°, others at 13°, 

 and the most radiating will even be at 7° and 6° or 5°, if suitably 

 situated. Then if the air is very moist, that is if the dew-point 

 is near 15°, nearly all the bodies will have dew upon them, the 

 warmest in small quantity, and the colder in greater proportion. 

 If the air is less humid, if, suppose, the dew-point is at 10°, 

 those bodies which are higher than 10^ will remain dry, those 

 at less than 10° will be more or less covered with dew. Lastly, 

 if the air is extremely dry, if the dew-point is below 5°, all the 

 bodies will remain dry, the coldest as well as tlie warmest." 

 After the experiments and considerations preceding, it does not 

 appear to me to be any longer allowable to say that plants and 

 bodies ordinarily moistened by dew are cooled 8° and 10° below 

 the temperature of the air. The observations of Wells, Wilson 

 and Pouillet are accurate, and certain bodies placed near the 

 surface of the earth may indeed descend 8° or 10° below a 

 thermometer situated at a height of 4 or 5 feet ; but we cannot 

 conclude from these observations that the differences obtained 

 indicate the depression of temperature of the radiating body 

 below the medium in which it is plunged; for the nocturnal 

 cold of vegetable leaves, as we have just said, does not exceed 

 the 2° of the Centigrade thermometer, and is therefore four or 

 five times less than the cold admitted in works on physics and 

 meteorology. 



This great reduction in the difference of temperature between 

 the air and the radiating bodies by no means implies that the 

 principle of the condensation of atmospheric vapour in conse- 

 quence of simple nocturnal radiations is erroneous ; and to prove 

 it, the following observation of Saussure is sufficient, which I 

 have several times had occasion to verify in the course of my 

 numerous researches. When the dew begins to appear, the 

 hair-hygrometer introduced into the stratum of air contiguous to 

 the soil marks from 90^ to 98°. Hence the surrounding space 

 is vciy nearly in a state of saturation ; consequently it is not 

 necessary that plants and bodies of all kinds situated near the 

 surface of the earth should attain any very great degree of cold 

 in order to precipitate the aqueous vapour; but the cooling of 

 T" or 2°, acquired by plants under the influence of a clear sky, 

 feeble as it is compared to the H° or 10° hitherto supposed, will 



