72 LIGHT SCIENCE FOR LEISURE HOURS. 



fainter part of the corona, the two together forming a 

 perpetual solar aurora; and in this way we should 

 begin to see the means of explaining the remarkable 

 and undoubted fact that the displays of our terrestrial 

 auroras are associated in a most intimate manner with 

 the condition of the solar surface. For we should be 

 led to regard the recurrence of our auroras as a mani- 

 festation of the same sort of solar action which is more 

 constantly at work amidst the materials constituting 

 the corona and the zodiacal light. 



This view leaves unexplained the bright lines of 

 the coronal spectrum. But as we have every reason 

 for regarding the auroral light as an electrical phe- 

 nomenon, and the bright lines in the auroral spectrum 

 as therefore not due to the presence of vast quantities 

 of glowing vapour, we may extend the same interpreta- 

 tion to the coronal spectrum. In laboratory experi- 

 ments, when the electric spark passes between two iron 

 points, its spectrum shows the lines belonging to 

 vaporized iron, and yet the quantity of iron vaporized 

 by the spark is almost infinitesimally minute. And 

 similarly, if we regard the corona as an electrical phe- 

 nomenon, we get over the difficulty which opposes itself 

 to Professor Harkness' theory, that a large proportion 

 of the corona consists of the luminous vapour of iron. 



The general result would seem confirmatory of these 

 views, according to which the real origin of the coronal 

 light is to be sought in the millions of meteor-systems 

 which undoubtedly circle round the sun, many of them 

 passing (when in perihelion) very close to his globe. 



