218 



CARNEGIE INSTITUTION OF WASHINGTON. 



(2) The directions of the axes of least motion for the various groups 

 cluster widely about the poles of the galaxy, the remoter ones being a 

 little over 20° distant from it. 



(3) The directions of greatest motion (vertices) show the same 

 systematic arrangement found by Mr. Raymond by Schwarzschild's 

 method (A. J. 676). The vertices of types A, F, and G lie well to the 

 south of those of types K, M, and "X." ('*X" did not appear in the 

 fonner discussion, and G was anomalous. It here falls into line.) 

 There seems to be no longer reason to doubt the reahty of this phe- 

 nomenon. 



Table 1. 



Pole of galaxy 190?33,+27?35. ^X =(N, O. and unknown). 



(4) As a by-product of the investigation we have a detennination of 

 solar motion by the method of Bravais. The solar apices for A and F 

 types are in lower declinations and smaller right ascensions (smaller 

 galacitic longitudes) than those of types K to X. This has been 

 shown before by several investigators using different methods. The 

 several values of M differ considerably; but this seems to be little more 

 than the effect of different average parallaxes of the groups, except in 

 the case of type G, where M seems to be actually small, perhaps 

 because of the presence of a large proportion of stars approximating 

 to the sun's motion (solar group). 



(5) The velocity-figure for type A is very elongate and much 

 compressed toward the galactic plane. In "later" types the dimen- 

 sion in the preferential direction increases comparatively little. 

 The other two dimensions increase much more; the intermediate axis 

 increases steadily throughout the chain of types, or nearly so. The 

 least axis, the one perpendicular to the galaxy, increases very abruptly 

 between A and F, but does not continue to increase. Type M shows a 

 reversion to the shape of velocity-figure (but not to the size or orienta- 

 tion) shown by type A; indeed, there is a hint that the least axis passes 

 through a maximum somewhere in the region F to K. 



