362 ATMOSPHERIC ELECTRICITY. 



the density of the media traversed by the spark of our machines singu- 

 larly influences the brightness and the color of the light produced ; 

 that when this density is very small the light becomes diffuse and 

 reddish ; that in the case where it is considerable tlie light is con- 

 tracted and brilliant. The electrical sparks of our machines have 

 also different appearances, and especially vary their color according 

 to the hygrometric state of the air. They are also generally the 

 more brilliant the better the substances between which they take 

 place conduct electricity. Finally, in the electric explosion occurring 

 between the strata of a cloud lying above another^ it may happen, as M. 

 Kaemtz* observes, that on account of the thickness of the lower cloud 

 the light of the flashes, however intense it may be supposed, produces 

 on the eye of the observer only a feeble and diffused light. 



The experiments of Mr. Wheatstone show that the duration of the 

 linear streaks of lightning and of dift'used flashes does not exceed the 

 thousandth part of a second. It is not the same with the globular dis- 

 charges, which are sometimes visible for several seconds, and pass from 

 the clouds to the earth with sufficient slowness for the eye plainly to 

 follow their course and ascertain their velocity. Their existence 

 appears to be now well established. Schiiblerf and M. Kaemtz| men- 

 tion flashes of lightning which had the appearance of a current of fire as 

 large as the arm, terminated by a much larger and more brilliant ball ; 

 and Muncke§ saw a discharge which descended vertically, and which 

 must have been about 200 feet long, transformed, under his eyes, into 

 a great number of small balls. These globular discharges, which some- 

 times leave along their passage inflamed portions, appear frequently 

 in the midst of volcanic storms ;|| the same is the case with linear 

 streaks of lightning ; they often become subdivided, and after their 

 explosion the}'' commonly leave a strongly sulphurous smoke. We 

 are as yet ignorant of the cause of the formation of these meteors ; they 

 seem to be the agglomerations of ponderable substances strongly 

 charged with electricity. M. Arago^ thought that it might not be 

 impossible that the lightning in passing through the air sometimes 

 produces instantaneously a sort of half union of substances which may 

 exist in this medium, and thus produce these globular discharges. 

 There are many facts which favors this opinion. First, it is established 

 that lightning, when on its passage through the air, produces chemical 

 combinations, as is proved by the presence of nitric acid in the rain 

 ■water gathered during storms.** Then the observations of M. Fusi- 

 Dieri,tt confirmed by those of M. Boussingault, on the presence of 

 metallic iron, of iron in different degrees of oxydation, and of sulphur 

 in the powdery deposits which surround the orifices in the bodies 

 through which lightning has forced a passage ; the fact that hail- 

 stones have for a nucleus small fragments of sulphuret of iron,|J the 



"" Lehrbuch der Meteorologie, tome IT, p. 42D. 



f Journal de Schweigger, tome XLI, p. 36; ]82i. 



J Lehrbuch dcr Meteorologie, tome II, p. 427. 



§ Dictionnaire de Gchler, tome I, p. 1000. 



II Philosophical Transactions for 1795, part I, p. 82. 



ijAnnuaire pour IS.TS, p. 426. 



«~ Annales de Chimie et de Phj-sique, tome XXXV, p. 329. 1827. 



If Bibliotheque Universelle, tome XLVIII, p. 371, and tome XLIX, p. 1. 1831 and 1332. 



jj Bibliotheque Universelle, Sciences et Art-s, tome XVIII, p. 78. 1821. 



