156 LIGHT SCIENCE FOR LEISURE HOURS. 



' It appears in the telescope as a nearly round nebu- 

 losity, in which the light increases rapidly towards the 

 centre, where, on some occasions, I detected, I believe, 

 a small stellar nucleus. Generally, this minute nucleus 

 was not to be distinguished in the bright central part 

 of the comet. The spectrum consists for the most 

 part of three bright bands. The length of the bands 

 in the instrument shows that they are not due alone 

 to the stellar nucleus, but are produced by the light 

 of the brighter portions of the coma. I took some 

 pains to learn the precise character of these luminous 

 bands. When the slit was wide they resembled the 

 expanded lines seen in some gases. As the slit was 

 made narrow the two fainter bands appeared to fade 

 out without becoming more denned. I was unable to 

 resolve them into lines. The middle band, which is so 

 much brighter than the others that it may be con- 

 sidered to represent probably three-fourths, or nearly 

 so,* of the whole of the light which we receive from the 

 comet, appears to possess similar characters. In this 

 band, however, I detected occasionally two bright lines 

 which appear to be shorter than the band, and may be 

 due to the nucleus itself. .... Besides these bright 

 bands there was a very faint continuous spectrum.' 



Interpreting these observations according to the 

 principles which have been already stated, we deduce 

 the following interesting results. 



The nucleus of Brorsen's comet consists of luminous 

 gas. The coma is also gaseous in the neighbourhood 

 of the nucleus, but its outer portions are of a different 



