Scientific Intelligence. 



CIS^TIFIC INTELLIGENCE. 

 L Chemistey and Physics. 



• the fact already remarked by other observers that the 

 lines in the spectrum of the aurora borealis do not coincide with 

 those of any known element, and has endeavored to give a rati 

 explanation oyfthe want of coincidence without assimiino- th. 

 eijce of unknown elements in the earth's atmosphere. If th- 

 in question are really of electrical nature and are product d 1 

 ignition of highly rarefied gases, the ignition must take pln'v 

 so low a temperature that it would be impossible to observe \\x 

 lines in Geissler's tubes. Hence Zollner thinks that the speetnuii 

 of the aurora does not correspond with any known spectrum of tlio 

 atmospheric gases, because it is a spectrum of a ditferent order 

 which cannot at present be produced artificially. If, at a given 

 temperature, Ai and Ei represent respectively the values of the 

 absorptive and emissive powers for the wave length i for the unit 

 of thickness and density, m and (t the thickness and density of the 

 luminous layer of gas, we have, according to Zollner, for the 

 brightness E of the part of the spectrum corresponding to i, tlie 



expression E = [ 1 — ( 1 — A A) ] -— . 



T a given gas and given temperatu: 



alue of the product m a. In a Geissler's tube t 



'. the diameter of tlie narrow part of the 



, and regard tliis as the unit of thickness m, and take as 



unit of density the density of tlie ( 



temperature of 0° and pressure of I'""' of mercury, it now «c 

 ignite the air in the tube by an inductorium, then at a constant 

 temperature the spectrum would remain qualitatively and quanti- 

 tatively unchanged if the diameter of the narrow tube were 

 nicreased from 1""" to 1000"'"\ and if at the same time the pressure 

 on the gas were diminished to i oVir"""- 



Wiilhier found that in the case of nitrogen the spectrum became 

 sufficiently luminous for spectroscopic examination only Avhen the 

 pressure was diminished to 46'""\ while in the case of oxygen the 

 necessary diminution was to 28 to 30"^'". If we assume that ni 

 Wulln 



thickness ana under a pn 

 to observe the atmosphei 

 this the thicknesses of the luminous layers whicli occur m n'^^ 

 case of the aurora, which are of course far greater, and, at a Uv-- 

 tance from the zenith, to be estimated in miles. If we take tne 

 thickness of such a layer as only 1 kilometer, then at the ku 

 temperature as in the last case, the density would have to he mv) 

 the one-millionth of that of the air enclosed in the tube, and tlier 

 fore exert a pressure of only 0-00005">"' at 0° C, in order that tn 



