PKESIDENTIAL ADDRESS SECTION A. 35 



star photographs taken about a quarter of a century ago with 

 repeat photographs taken recently has made the detection of star 

 movements across the line of sight much more certain and less 

 laborious, and in the hands of Innes and Wood at the Union Obser- 

 vatory, Johannesburg, is yielding results of great value. Star 

 movements in the radial lines from the observer to the star are 

 in the peculiar domain of the spectroscopist. A star moving in a 

 radial line away from us is lengthening its waves of light, 

 and a star moving towards us is shortening them, just as the 

 whistle of an express train is noticed to be higher in pitch when 

 it is approaching us and lower when it is receding from us, so is 

 the pitch or wave-length of light altered by motion. 



An observer at a rock lighthouse will see the sea-waves passing 

 him at a certain rate; a steamer going against the same incoming 

 waves will meet more (or shorter) waves, and another steamer 

 going with the waves will be overtaken by fewer (or longer) waves 

 in a given time, than the stationary observer in the lighthouse. 



This alteration of wave-length of light from its normal, due 

 to motion either of the observer or the source, has the effect of 

 shifting a line from its normal place in the spectrum, to the violet 

 if the motion is one of approach and to the red if it is one of 

 recession. The shift is microscopic, but accurate measurements 

 easily show the oscillation of a star line to either side of its normal 

 place, due to the motion of the earth in its orbit, and this shift 

 has been used to determine the speed of the earth's motion and 

 therefore the size of the earth's orbit and the distance of the sun. 

 In the same way the motion of the sun in space manifests itself 

 by the stars in one-half of the sky appearing, in the mean, to be 

 approaching and those in the other half to be receding. The 

 direction in which our solar system is travelling is now known 

 with a fair degree of accuracy. Our observations of star move- 

 ments in these radial lines must be corrected for the motions of 

 the earth and sun before we can deal with the real movements of 

 the stars themselves. These spectroscopically determined radial 

 movements are expressed in kilometres per second, but the trans- 

 verse movements (technically called "proper-motions") are 

 expressed in seconds of arc per year or per century. The "proper 

 motion" expressed in angular measure may be fast or slow, and 

 we have no means of knowing which until the distance of the star 

 concerned is known. When the distance is known, then we can 

 convert seconds of arc into kilometres per second, and combining 

 the radial and transverse motions expressed in the same units we 

 can derive the real motions of the stars. 



It is evident, therefore, that before we obtain a knowledge 

 of the real motions of the stars we must know their distances, and 

 when we know their distances we must, in order to compare their 

 real magnitudes, bring them — in our calculations — to one common 

 distance in order to compare them one with the other. 



Star Magnitudes. 

 We shall see later that a knowledge of star distances leads to 

 information about the absolute magnitudes of stars, and then 



