[ 71 ] 



air caufed by eledricity, I fhall conclude by flating the fads on 

 which this fiippofition is founded. 



First. It is certain that inflammable air is produced, parti- 

 cularly between the tropics, by many natural operations, fuch 

 as the putrefadiion of animal and vegetable fubftances, volcanoes, 

 &c. and that this air is lighter than any other, and confequently 

 occupies the higheft regions of the atmofphere ; and hence Mr. 

 SaufTure and others have found the air on the higheft mountains 

 .lefs pure than that on the plains, and its eledricity ftronger. 



Secondly. It is allowed by Dodlor Halley and others who 

 have treated of the trade winds, that the higheft air between 

 the tropics is thrown off on both fides towards the poles, and 

 of this I think I have given fufEcient proof; therefore it is 

 inflammable air that is chiefly thrown off towards the poles. 



Thirdly. It is certain that the northern lights are the higheft 

 of all meteors, though they fometimes extend pretty low into 

 the inferior atmofphere ; and Dodor Franklin's conjedure, that 

 they proceed from eledricity, is at prefent generally followed by 

 all meteorologifts. A detail of their reafons I muft omit, as it 

 would occafion too great a digreffion from the prefent fubjed. 



Fourthly. It is certain that, after the appearance of an 

 aurora borealis the barometer commonly falls. This obfervation 

 was firft made by Mr. Madifon in America * ; and I have feen 



* II. Phil. Tranf. Philad. p. 142. 



it 



