o8 ASTRONOMY 



ment be some stars on their way to the centre and others on their way from it, and the 

 perspective appearance of these two classes would be precisely the same as that which 

 has given rise to the hypothesis of two star-streams. Parallel lines and lines converging 

 to a point have the same perspective appearance: we can only differentiate them by a 

 knowledge of the distances. Now it is just in respect of distance that our knowledge of 

 the stars is most deficient. Direct measures take us only a little way from our sun, and 

 to probe further into space we must trust hypotheses such as those of Hertzsprung and 

 Plummer above mentioned. Using Plummer's figures, Turner finds indications of an 

 actual acceleration towards a system-centre which he fixes provisionally in direction 

 RA 94 Dec + 12 at a distance of qo light years. Confirmation of these figures is also 

 forthcoming from an entirely independent argument from the distribution of double 

 stars in various directions, put forward by T. Lewis in 1906: and again from the general 

 distribution of all stars. 



The congestion at the centre can be avoided if there is a sufficient departure from 

 accurately radial motion, which would be quite consistent with observed facts. Were 

 the radial motions undiluted, the phenomenon would have earlier attracted general 

 attention, and would not have been at first denied by so acute an observer as L. Boss. 

 (His objection has been since withdrawn.) As it is they are largely diluted by cross 

 motions, so that they only emerge as averages from the discussion of numbers of stars. 

 These cross motions suffice to carry the stars wide enough of the centre to avoid collision 

 and may even suffice in certain cases (for instance in the case of the helium stars) to 

 destroy the "' two-stream " appearance altogether. But it is quite possible that they 

 have developed in the course of ages of interaction between the individual stars, and that 

 if the system were originally formed from a vast nebula condensing under gravitation, 

 the condensation may have been very nearly along radial lines, resulting ultimately in 

 a sudden congestion at the centre which would appear to distant systems as a " new 

 star." Such a rapid congestion would be followed by an explosive expansion, the frag- 

 ments being dispersed to something short of their original distance: gravitational attrac- 

 tion would ultimately draw them together again, and so on. There would be a sequence 

 of explosions of decreasing violence, seeing that on each occasion energy would be dis- 

 sipated. If the fragments became stars, the earliest-formed stars would have the largest 

 velocities, as Campbell has found. The speculation is worthy of note if only for the 

 reason that it shows the possibility of explaining Campbell's remarkable result without 

 having recourse to any hypothesis not already familiar. 



It is mentioned above that our direct knowledge of the distances of the stars is very 

 limited; but progress is being made, and the last year or two have seen some notable 

 additions to the published material, especially in the work of A. R. Hinks and H. N. 

 Russell at Cambridge and Princeton, and of Frank Schlesinger at the Yerkes Observa- 

 tory. An interesting confirmation of Boss's parallax (o".o253) for his Taurus cluster, 

 found by combining spectroscopic velocities with proper motions, was obtained by the 

 direct determination (o".o23) of Kapteyn. 



The great 5-foot reflector of the Mount Wilson Observatory has already been effec- 

 tively employed on the spectra of the stars, chiefly for types A and B, which are of special 

 importance in the study of " star-streaming." In the course of this work more than 50 

 stars with variable velocities have been found; and attention is drawn to 7 stars moving 

 with exceptional speed, including the" runaway star" Groombridge 1830,- for which 

 a velocity of 343 km. per sec. is found. 



General, Among losses by death must be specially mentioned that of Henri Poin- 

 care (July 17, 1912), a mathematician who illuminated a wide range of sciences; and that 

 of Lewis Boss (Oct. 5, 1912), whose patient work has put our knowledge of the posi- 

 tions and motions of the stars on a new basis. 



The scientific papers of Sir W. Herschel (1738-1822) have .been collected and pub- 

 lished in two large quarto volumes by the Royal Society and the Royal Astronomical 

 Society. 



The meeting of the International Union for Co-operation in Solar Research on Mount 



