THE ECLIPSE OF 1871. 115 



images of the source of light. If the source of light 

 is the sun or any object shining with all the colours, 

 the different images will overlap and he will see simply 

 1 Newton's spectrum,' a rainbow-tinted streak of ex- 

 treme beauty and splendour, but nevertheless what 

 the spectroscopist describes as an ' impure spectrum,' 

 because in it a multitude of overlapping images are 

 present. If, however, the source of light emits rays of 

 certain colours only, then there will be separate images 

 of thee colours, each clearly discernible in all its 

 details. For example, let us suppose that in a little 

 conical flame of great heating power but small lumi- 

 nosity, a chemist places a small quantity of sodium and 

 lithium. Then when he looks at the flame through a 

 spectroscope without using a slit he will see a little 

 conical yellow flame, and close by it a little conical and 

 rather faint orange flame, and further away a little 

 conical red flame ; whereas if he had had a fine slit to 

 his spectroscope he would have seen three fine lines, a 

 yellow one due to the sodium, and two lines, one orange 

 and the other red, due to the lithium. , 



Now if the reader has followed this brief but nece's- 

 sary explanation, he will see that the astronomer 

 possessed the means of at once solving the difficulty of 

 the corona. So long as he used a slit he obtained a 

 bright green line which might not come from the 

 corona, but from the illuminated sky in the same 

 direction ; but if he removed the slit and then saw a 

 green image of the corona, he would no longer be in 

 doubt. For the illumination of the sky could not form 



i 2 



