VABIABLE STAES ROBINSON 127 



was SO far away that light received from the members was over 

 200,000 years old; or, the cluster is said to be 200,000 light years 

 distant. In the opposite direction, clusters were observed at distances 

 of the order of 100,000 light years. For the first time in the history 

 of astronomy investigators could say definitely that the universe had 

 been sounded, and that its diameter was so large that 300,000 years 

 were required for light to traverse it. Yet only a short time later, 

 Hubble, at Mount Wilson, began to count variable stars, such as 

 Cepheids and novae, by the dozens in the Andromeda nebula, iden- 

 tifying it as a great collection of stars, sometimes calletl an •' Island 

 Universe." Thus he w^as able to measure distances more than three 

 times as large as those previously determined. From this great col- 

 lection of stars, which to the unaided eye appears as a hazy spot in 

 the sky, nearly one million years is required for light to reach us. 

 Some idea of the distance of this far-away cloud of stars can be 

 gained when it is recalled that light from the sun reaches us in 

 only eight minutes (fig. 3). 



The variable stars hitherto discussed are all intrinsically variable 

 and in this way differ from the class of objects represented by Algol 

 and Beta Lyrae. Algol, or Beta Persei, may have been recognized as 

 a variable by the Arabs, since it was called by them al Ghnl^ or the 

 demon, possibly on account of its apparent capricious drops in lumi- 

 nosity. The first historical evidence of its variability, however, 

 seems to rest with Montanari, who noticed its variability in 1669, 

 without attempting any careful study of the star. It was observed 

 again in 1782 by a young Englishman, John Goodricke, a deaf mute, 

 who not only determined its period of variation but also correctly 

 suggested the cause. Beta Lyrae was discovered two years later by 

 Pigot, a friend of Goodricke. Each of the two systems just men- 

 tioned owes its variability to periodic eclipses by the component star. 

 In the case of Algol, one of the components is faint relative to the 

 other, which is called the primary, whereas in the case of Beta 

 Lyrae both components are approximately of the same brightness 

 and relativeh' much closer together than the components of Algol. 

 In fact, for many systems of the Beta Lyrae type, the components, 

 which are generally of the same brightness, are very nearly in actual 

 contact. The range of periods of eclipsing stars, including those of 

 the Algol and Beta Lyrae types, agrees very closely with the range 

 of periods of Cepheids. The amount of change in brightness is also 

 about equal to that of Cepheids. For eclipsing stars, the amount of 

 change depends upon the relative brightness of the two components 

 of the system and upon the orientation of the plane in which they 

 revolve to the line of sight. 



