424 ASTROMETRY 



certainly only be on the condition that we seek to obtain relative 

 parallaxes. For who will dare to prophesy that we shall be able to 

 measure absolute parallaxes of such an amount in many centuries 

 to come? 



Still, it is absolute parallax we want. Therefore the difficulty will 

 remain: how are we to get at the parallax of the stars of comparison? 



In my conviction the part to be acted by direct determination of 

 parallax in our investigations about the general structure of the 

 system will be : to furnish the most powerful and most reliable check 

 on the results of the statistical methods. 



These methods furnish absolute parallaxes. 



For the bulk of the stars the task of the direct determination of 

 parallax will be to decide whether or no they lead to the true differ- 

 ences of parallax. 



For them it cannot do more, and the task, as it is, is already by no 

 means a light or an unworthy one. 



It becomes enormously more difficult if we take into consideration 

 that very small but interesting part of the stars for which it can do 

 more. 



We shall be able to measure directly absolute parallaxes, practically 

 independent of any theory, for the stars nearest to the solar system. 

 Where the parallax exceeds, say, 0"05, the uncertainty of the dis- 

 tance of the stars df comparison will be practically insignificant, if 

 only due care is taken as to the number and choice of these stars. 



The number of stars brighter than the tenth magnitude within 

 the distance corresponding with this parallax may be evaluated at 

 some two thousand. 



It will be a noble and still not over-heavy task to determine 

 directly the parallax of these stars. This determination would furnish 

 a foundation, independent of any hypothesis, for the astronomy of the 

 regions of the stellar world nearest our terrestrial abode. 



The main difficulty, however, lies not in this determination. More 

 difficult it will be to find out which, among the million stars brighter 

 than the tenth magnitude, are the two thousand stars to be meas- 

 ured. 



I have tried to show elsewhere that even this difficulty, great as it 

 undoubtedly is, may be overcome without overtasking the practical 

 astronomers of the present day. Since that time the kindness of 

 several scientific men has enabled me to ascertain that by using 

 telescopes of longer focus the work may be diminished to a third or 

 less of what it would demand with carte du del instruments. So in 

 this direction also the future looks hopeful. 



At the end of this too long lecture I hope you will agree with the 

 conclusion: 



Time has come to undertake a general attack on the mysterious 



