194 REPORTS ON INVESTIGATIONS AND PROJECTS. 



tion the stars A (Aq) of Harvard Annals, vol. 50, vv^ere to a large extent 

 subdivided, one part being found to approach more nearly the helium stars, 

 the other part deviating somewhat more toward the later types. The num- 

 ber of B stars thus gained for the second stream has more than doubled the 

 total number. Still, it will be necessary to strengthen considerably the data 

 for radial velocity of these stars before even an approximate determination 

 of the elements of the stream will be possible. It is planned to obtain such 

 data during the coming year, not only for the helium stars, but also for the 

 classes Ao and Aj. 



(3) The extent and the elements of the Scorpius-Centaurus and Perseus 

 groups, which must be considered as subgroups of Stream I, were carefully 

 investigated. Much attention was paid to the question whether the rela- 

 tively small difference in the stream-motion of these two groups could 

 possibly be due to systematic errors in Boss's Catalogue and in the deter- 

 mination of the radial velocities. The conclusion was reached that such an 

 explanation was untenable and that, consequently, differences in the stream- 

 motion in different parts of the sky must be admitted. With the exception 

 of a relatively small group of stars in Vela, which may form a more inti- 

 mately connected subgroup, nearly all the stars between galactic latitude 

 -J- 30° and — 30° and right-ascensions 5^^ 20"^ and 16^ 30™ appear to belong to 

 one of the two groups mentioned. 



(4) In each of these groups the parallelism of the motion is such that 

 parallaxes can be and were derived for those objects which have a sensible 

 proper-motion ( >o".oi7). Near the vertices the accuracy of these paral- 

 laxes is of course rather small, but for the great majority of the stars they 

 must be fairly reliable. For these stars we thus know both the position in 

 space and the amount and direction of the linear motion. We can also 

 derive the frequency-curve of their absolute luminosities. Such a deter- 

 mination is now being made. 



(5) A beginning has been made in the study of the question as to what 

 distance from our system the star-streams extend. 



(6) It has already been noticed by several astronomers that there is a 

 certain clustering of the helium stars in different parts of the sky. The 

 same thing is also brought out by the study of the motions of these stars 

 and to a certain extent of the A stars. It is the clustering in such groups 

 as the Pleiades, the restricted Perseus group, the larger Perseus group, the 

 Scorpius-Centaurus group, and the Vela group (all probably partial groups of 

 Stream II). These local groups make the investigation of the stream- 

 motion considerably more difficult. It becomes necessary to investigate 

 them, at least to such an extent that they may be approximately separated 

 from each other and from the rest. Such a study led to the discovery 

 already mentioned, that a number of stars in Vela probably form a local 

 group; and further to the belief that the Ursa Major group, already ex- 

 tended by Hertzsprung to widely separated regions of the sky, may even 



