OUR REVOLVING "ISLAND UNIVERSE" — SKILLING 133 



astonishing distances. But in some respects the best stars of all to 

 show rotation of the galaxy are the Cepheid variables, the stars that 

 gave distances to the spiral nebulae as mentioned above. Joy, at the 

 Mount Wilson Observatory, has studied about 150 of these, dis- 

 tributed mainly along the Milky Way. The results thus obtained 

 are very dependable because of the unusually reliable manner, ex- 

 plained above, in which their distances can be determined. 



Motions of these stars could theoretically be measured in two ways : 

 by getting the angular distance they travel across the sky in a cer- 

 tain length of time — their "cross motion" — or by finding the speed 

 in miles a second by which they come closer to us or move farther 

 away — "radial motion" or "motion in the line of sight." The spectro- 

 scope will measure this line-of-sight speed. The lines of the spectrum 

 of an approaching or receding star are shifted a little toward the blue 

 or the red end of the spectrum as a result of the motion. This is 

 often called the Doppler shift. The greater the speed of the star 

 the greater the shift. 



Cross motion is not useful in studying galactic rotation, as many 

 years must elapse between the taking of two photographs of the 

 stars to show any appreciable change in their positions. But the 

 spectrum of a star can be photographed in a few minutes (or hours, 

 depending on its dimness) and the star's radial velocity is found 

 immediately. 



The general principle of Oort's method may be readily understood 

 by considering the solar system. A planet, such as Venus, when 

 nearest the earth and directly between us and the sun has cross 

 motion because it goes faster than the earth, but it has no radial 

 motion that would be indicated by the spectroscope because it is keep- 

 ing parallel to the earth's orbit and so is not changing its distance 

 from the earth. Likewise Mars when nearest the earth on the opposite 

 side from the sun is not changing its radial distance, although it has 

 cross motion because it moves slower than the earth. 



If we assume another planet a little in advance of the earth, or 

 behind it, in the same orbit, the hypothetical planet would show 

 no radial motion. The spectroscope would show it as a stationary 

 object. But in all other directions except these four — toward the 

 center, away from the center, straight ahead, or straight behind — 

 planets would appear to be coming closer to us or getting farther 

 away from us, and the spectroscope could measure this speed of 

 approach or recession. 



So it is with the stars. Those directly ahead of us or behind us 

 in our revolution about a common galactic center, and those toward 

 the center and away from the center, show no motion in the line 

 of sight to be measured by the spectroscope. (This would be true 



