218 



Wm. A. Norton on the Variations 



form, will also tend to make the radiation of the air, in the same 

 direction, uniform, and on a calm night the radiation from the 

 air to the earth ought gradually to increase, inasmuch as the 

 difference of temperature of the two ought slowly to increase, 

 (radiation alone being considered,) and for small differences of 



temperature 



From 



which it appears that the laws in question are not primarily true 

 of the temperature of the atmosphere. 



It was also taken for granted that there was no material varia- 

 tion in the radiating power of the earth's surface from one season 

 to another, by reason of frost or more or less humidity, or changes 

 in the state of vegetation or any other cause. That, in point ot 

 fact, the diminished loss of temperature at night in the winter 

 cannot be owing to any such cause as this, will be seen on in- 

 specting the following tabular statements. 



Average fall of thermometer from 9 p. m. to 4 a. m. at Philadelphia. 



July, 



August, . . . 

 September, 

 October, . . 

 November, 

 December, . 



1842. 



4°40 



• • 



G -00 

 2 -57 

 2 61 



1843 



o u -35 



4 -04 



1 *5o 



2 '43 



1844. 



5 18 



5 02 



4 -84 



3 



3 -40 



2 -01 



•77 



Average 



4 c -97 



5 

 4 



2 



■02 

 84 

 60 

 50 



2 *35 



Mean Diurnal Variations of Temperature at Halle, Gottingen, 



Padua, and Forth Leith {near Edinburgh.) 



• • • • 



Halle, 

 Gottingen,. fc . . 



Padua, 



Forth Leith, . . 



It will be observed that the nocturnal loss of temperature 

 steadily decreases from August to December, and that there is no 

 cause tending to diminish the radiating power of the earth's sur- 

 face that operates in the same steady manner at all places- during 

 this interval of time. 



I conclude, therefore, that the cause of the nocturnal secondary 

 variations of the horizontal force must either consist in variations 

 in the amount of vapor deposited from the atmosphere, or be in 

 some way connected with the upward flow of heat below the 

 earth's surface. This upward flow of heat is not attended with^| 

 any variations in the total amount of heat lost, and we therefore \ 

 can connect the secondary variations with it, only by introducing 

 some new hypothesis. Besides there exist good reasons for be- 

 lieving that it cannot be the cause of the observed variations in the 

 nocturnal loss of temperature, and therefore cannot be the cause of 

 the secondary variations of the horizontal intensity. In the first 





