36 CARNEGIE INSTITUTION 



able, we are unable to suggest a definite program other than the 

 restricted one which follows. 



In the first line it seems desirable that an attempt be made to 

 measure the parallaxes of stars known to be in sensible motion. As 

 the limit of such motion for a comparatively restricted work, one 

 might take o."2 or even o."i. In order to identify the stars having 

 such motions, and to measure the motions themselves, the sources 

 of information are now very scanty. It would be almost a necessity 

 to carry through (2) of the program — the meridian observation of 

 telescopic stars down to the ninth magnitude. If this were not 

 done it would still be necessary to make accurate meridian obser^'a- 

 tions of all stars observed for parallax. The number which would 

 be at present available under this plan and south of — 20° of declina- 

 tion would probably be considerably under 500, though others 

 could doubtless be found without much difficulty. We would 

 recommend the use of a photographic telescope of relatively long 

 focal distance for this purpose. It should be of the highest optical 

 perfection. It is quite possible that a three lens or four lens combi- 

 nation would be advisable, in order to get sharp, round images 

 over an area of at least four square degrees if possible. We suggest 

 a telescope of 18 inches aperture and about 30 feet focal length. 



It would also be highly desirable to measure with the heliometer 

 the parallax of many stars brighter than the sixth magnitude and 

 south of — 20° of declination. A seven inch heliometer similar to 

 those in use at the Cape and at New Haven would be suitable for 

 this purpose. The stars selected should be chiefly those distin- 

 guished for proper motion — several hundred in number. This 

 would be in effect an extension of the parallactic survey which has 

 been carried on so successfully at New Haven within recent years. 



We think that the services of two observers on this work for a 

 period of at least eight years would result in an extremely valuable 

 contribution to the solution of the stellar problem. 



For some years the project of a general Parallax Durchmusterung 

 has been brought to the attention of astronomers. This contem- 

 plates nothing less than the determination of the approximate rela- 

 tive parallax of every star down to the ninth or tenth magnitude. 

 The inquiry would mean something like this : If we take as the 

 unit of distance the average star of the ninth magnitude, what are 

 the relative distances of other classes of stars, and what individual 

 stars are especially near the solar system ? A program for this pur- 

 pose is described in the letter which Professor Kapteyn has addressed 



