in the Magnetic Forces of the Earth. 217 



power. In the second place, the observations show that, in the 

 calmest nights the diminished fall of temperature after midnight 

 is no less certain, than when the equilibrium of the air is disturbed 

 and different masses are brought successively into contact with 

 the surface of the earth. Again, since this law is observed to 

 hold with respect to the temperature of the atmosphere at the 

 earth's surface, the air, when agitated in contact with the earth, 

 must lose from this cause less and less heat as the night progresses. 

 But this is impossible, since the cooling of the earth's surface, 

 from radiation, goes on uniformly, and thus the difference be- 

 tween its temperature and the temperature of the air would in- 

 crease, and therefore the cooling, for the same amount of agitation, 

 should increase. In other words, the cooling of the air, attend- 

 ant upon the same degree and extent of agitation, must be uniform. 

 Unless, therefore, there is generally more wind, or the disturbance 

 of the equilibrium of the air extends to a greater height during 

 the latter than during the fore part of the night, the law above 

 mentioned must have some other cause than that under consid- 

 eration. Now, as a matter of fact, the observations do not dis- 

 close the existence of any general difference, of any amount, be- 

 tween the force of the wind before and after midnight. It ap- 

 pears from the curves showing the variations in the force of the 

 wind at Philadelphia, that there are slight differences, but they 

 sometimes lie in one direction and sometimes in the other. 



In view of all that has now been stated, it may be confidently 

 affirmed, that if the cause of the two anomalous facts connected 

 with the nocturnal loss of temperature be any meteorological 

 phenomenon, it must be the deposition of vapor from the atmos- 

 phere in other forms than that of rain, and chiefly therefore in 

 the form of dew ; and it may be stated farther, that either this 

 must be the actual cause, or it must consist in variations in the 

 amount of heat that is returned towards the surface during the 

 night, by conduction from below, or, in other words, in the laws 

 of the earth's cooling at night, irrespective of all atmospheric 

 influences. 



This result has been reached, it is true, by taking it for granted 

 that, primarily, it is the temperature of the surface of the earth 

 that varies according to the laws which I have stated, and that 

 the temperature of the atmosphere, near the earth's surface, con- 

 forms, of necessity, in its variations, to the same laws, by reason 

 of its relations to the earth's surface. This supposition is in ac- 

 cordance with the notion generally entertained upon this subject, 

 and is not likely to be questioned," but the principles laid down 

 ^ the general discussion preliminary to the present investigation 

 serve conclusively to establish its truth. For it will be seen that 

 the causes which make the nocturnal radiation from the earth's 

 surface into space and the upper regions of the atmosphere uni- 



Seco.nd Series, Vol. VIII, No. 23.— Sept, 1849. 28 



