VAEIABLE STAES ROBINSON 123 



These colors projected on a screen back of the prism constitute a 

 " spectrum " whose general structure depends upon the nature and 

 the source of the incident light. In the case of ordinary starlight, 

 where a cooler atmosphere surrounds the source, very narrow widths 

 of light are apj^arently missing, thus giving rise to a dark-line spec- 

 trum, the relative positions of the lines depending upon the chem- 

 ical elements and conditions of temperature and pressure in the 

 source of light, thereby revealing important data on the stars and 

 their atmospheres. It may also happen that all the spectral lines may 

 be shifted toward the red or toward the violet end of the spectrum. 

 Wave lengths from a given source are all shortened by motion 

 toward the observer for the reason that the observer meets more 

 waves per second than he would w^ere he at rest relative to the source. 

 This shortening of wave lengths indicating motion toward the ob- 

 server accounts for the shift tow^ard the violet end of the spectrum 

 constituting the shorter wave lengths, and conversely a shift toward 

 the red indicates motion away from the observer. Thus the obser- 

 ver is able to distinguish motions of the stars in the line of sight 

 and to measure their velocities. This fact alone is a very important 

 one in the study of variable stars and of the very close double star 

 systems, called spectroscopic binaries. 



From the spectra of novae the astronomer is able to detect the 

 presence of gases rushing out from these stars with velocities as 

 high as 1,200 miles per second. We are led to conclude, therefore, 

 that the great increase in brightness of these stars is due partly to 

 an increase in diameter caused by some kind of explosion. The fre- 

 quency of these explosions suggests to us that possibly similar dis- 

 turbances occur within every star, at least once in its lifetime. If so, 

 what warning are w^e to have of the approach of an explosion of 

 the sun? The sun is only a star among the 30,000 million others 

 which astronomers believe are in our Milky Way system. Why then 

 should it be the exception to the other stars? What if it should 

 explode? We frequently see spots on the sun large enough to hold 

 our little earth; we might therefore be justified in concluding that 

 in a solar conflagration, similar to that wdiich apparently takes place 

 in a nova, the earth would disappear almost instantaneously. 



As a representative of a second type of varial)le there is none bet- 

 ter than Mira, the Wonderful, a star in the constellation of Cetus, 

 which changes continuously in brightness, reaching a maximum and 

 minimum periodically. The variations of Mira were first seen by the 

 Dutch astronomer, Fabricus, in 1596; a star of moderate bright- 

 ness appeared where one hacl not previously been seen. In a few 

 weeks it had faded awa}' and it was therefore thought to be a nova, 

 until the year 1638 when it was seen again by another Dutch astron- 



