DEVELOPMENTS IN ASTRONOMY PLASKETT. 265 



parent nebula which, while condensing, separates into two masses, 

 and that these masses, as condensation proceeds, and by the influ- 

 ence of tidal action, gradually increase their distance from one 

 another, this being accompanied, of course, by an increase in the 

 period and also, as the results show, by an increase in the eccen- 

 tricity — a greater departure from the circular form — of their orbits. 

 There is no sharp line of distinction between spectroscopic and visual 

 binary orbits except that the latter have much longer periods and 

 generally higher eccentricities. 



The information already obtained, and that which will in the near 

 future be obtained, about these spectroscopic binary systems, has a 

 most important bearing on the problem of the constitution of the 

 sidereal universe, and we must now come to consider recent progress 

 in our knowledge of the extent and form and motion of its parts. 

 This is certainly the most important problem in astronomy, as practi- 

 cally all observing data, whether astrometrical or astrophysic al, 

 whether dealing with the absolute positions, proper motions, and 

 radial velocities of the stars, with their distances, dimensions, and 

 densities, with their evolution and spectral type, or with the investi- 

 gation of variables and binary systems, are all either directly or 

 indirectly obtained with this end in view and all are, undoubtedly, 

 directly of use in its solution. As I said in the early part of the 

 paper, there has been no time when so many different investigations 

 were converging toward this end, and I will try and. give you some 

 details of the principal results. 



One of the most striking of recent advances has been the discovery 

 of star drifts and star streams in the sidereal universe. These have 

 been discovered by statistical methods applied in the discussion of 

 the absolute positions and proper motions of stars and also by the 

 aid of their radial velocities. The one man to whom we owe more 

 than any other the development of this work is Prof. Kapteyn, who 

 is director of what is called the astronomical laboratory of Groningen, 

 where the instruments of research are not telescopes and spectro- 

 scopes but measuring machines and mathematical tables, where no 

 observations are taken but photographs are measured and observa- 

 tions discussed. I will try and give you a general idea of the present 

 state of our knowledge in regard to the motions of the stars. Although 

 we call them the fixed stars the term is a misnomer, for they are all 

 in motion. We can measure this motion in two components. First, 

 the motion at right angles to the line of sight, across the sky, deter- 

 mined from successive observations of the star's position in the sky 

 and measured by the change of position in seconds of arc in a year or 

 a century. The change of position varies between about 9" per 

 year and 0; the average annual proper motion, as it is called, for 

 first magnitude stars being \" and for sixth magnitude about 2 V' 



