THE ASTRONOMY OF THE INVISIBLE 379 



the other there is a chance formation of a double system, whose 

 existence may be relatively that of a moment, or of an aeon. 

 The conception of the helter-skelter flight of the stars, com- 

 bined with the extension of the law of attraction, long ago 

 suggested that the number of such binary and multiple systems 

 must be very great. This involved a corresponding, though 

 lesser, number of actual collisions. In a very striking way 

 has the spectroscope confirmed these conclusions. 



Long ago the sudden appearance of a very bright star in 

 the skies obtained its due meed of wonder and surmise. The 

 Star of Bethlehem may have been more or less of a myth. One 

 of the best known was the famous star of Tycho. It was first 

 seen on jth November 1572. Four days later it had grown 

 to the first magnitude. It continued to increase in brilliancy, 

 finally rivalling Venus at its brightest, and becoming visible 

 in full daylight. A month later it had begun gradually to fade, 

 and in the following May it had disappeared from view. The 

 telescope was then, of course, unknown. 



This is the type of these apparitions. Formerly they were 

 supposed to be very rare. Miss Clerke, in her System of the 

 Stars, was able to list only ten which were known up to 

 the end of the fifteenth century that is, until the coming of 

 the telescope. Not many more have been found since ; but 

 it is notable that ten or twelve of these have come within 

 the last century. One observer, Mrs. Fleming, has made the 

 subject her speciality, and she already has eight to her credit. 



There is, of course, no reason to suppose that they have 

 been any more frequent within the last twelve or fifteen years 

 since Mrs. Fleming began, than in the twelve or fifteen years 

 preceding, or than in thousands or millions of years. The 

 difference is simply due to perfected methods of observation. 

 Doubtless if these methods could be still further refined, these 

 conflagrations would be found to be daily and conceivably 

 hourly happenings. We do not observe them simply because they 

 are for the most part so remote that they do not strike the 

 attention even of observers with telescopes of enormous power. 

 For the most part actual visual observation is quite unable 

 to cope with the problem, and resort is had, therefore, to star 

 photographs and spectrographs. 



These new stars or nova, as they have come to be called, 

 seem to have a rather characteristic spectrum, so that when an 



