172 LIGHT SCIENCE FOR LEISURE HOURS. 



may exist in the direction of its length. But if the 

 object is brighter in some parts of its breadth than in 

 others, the spectrum will show corresponding variations 

 of brilliancy across its breadth. Hitherto we have been 

 assuming that all the light from the object is of the 

 same kind, however it may vary in brilliancy. Suppose, 

 however, that the light from the middle of the object 

 gives one kind of spectrum, the light from the outer 

 parts another ; then the spectrum will vary in character 

 as well as in brilliancy across its breadth. Suppose for 

 example, that the middle of the object is gaseous while 

 the outer parts are solid or liquid, then the appearance 

 presented would be two thin streaks of rainbow-tinted 

 light, separated by a dark space l across which would 

 be seen the bright lines belonging to the gaseous central 

 part of the luminous object. 



Now the breadth of the spectrum seen by Dr. Huggins, 

 corresponded with the breadth of the coma so far as 

 the widest parts of the tongue-shaped bands were con- 

 cerned. But the narrower parts were about the width 

 of the nucleus. Therefore the first question to be de- 

 cided was 'this, Is the narrowing of these bands of 

 light towards one extremity, and of the other towards 

 both extremities, to be considered as indicative of any 



1 Our readers will, of course, understand that a slice only of the object 

 is brought under spectroscopic analysis at once. If the whole of a cir- 

 cular object, whose centre was gaseous, were examined at once, the 

 middle streak of the spectrum would exhibit the compound spectrum of 

 the edge and centre of the object. Such an arrangement would clearly 

 be unfavourable to the formation of clear views respecting the character 

 of the object's light. 



