312 CARNEGIE INSTITUTION OF WASHINGTON. 



Though the stellar galaxy shows signs of a disturbed distribution of 

 stars and star-motions, it presumably conforms in a general way to the 

 state of equihbrium normal to an oblate spheriod of stars of initial 

 random motion, coordinated with the motions implied by its oblate 

 form. Assuming this, there should be (1) a common sphere of control 

 for the whole assemblage, whose form should be an oblate spheroid 

 conformable to the general configuration of the galaxy itself, and 

 (2) special spheres of individual control for each star, nebula, and other 

 body within the galaxy, all of which sweep in individual orbits through 

 the enveloping galactic sphere of control. These spheres of control 

 represent the dynamic concentration of gravitative influences that 

 condition the deployment of nebulse regarded as products of dispersion. 

 It is a corollary of this postulate of the centralized arrangement of grav- 

 itation that the extent of the sphere of control of any star — or other 

 body— will be determined in part by the position of such star in the 

 enveloping sphere of control of the galaxy, and in part by the rela- 

 tions of the spheres of control of neighboring stars to it. It is also 

 a vital part of the corollary that the deployment of a star — or other 

 concentration of matter — into a nebula by dispersion must conform 

 to the forces of generation in so far as the field of deployment ]ies within 

 the sphere of control of the parent body, but that, so soon as the dis- 

 persion passes beyond the border of such parental sphere of control, 

 it is subject to distortion and to possible destruction by the gravita- 

 tive influences of neighboring spheres of control or of the enveloping 

 sphere of control of the galaxy. 



There are intermediate orders of enveloping spheres of control, but 

 they need not be brought in here to complicate the case. It is sufficient 

 here to note that the development of a nebula of definite form is rad- 

 ically dependent on the sphere of control within which it is generated. 

 Beyond such sphere it is likely to be distorted, if not dissipated. 



1 . Now, if our galactic system approaches the consistent organization 

 assigned to it, the spheres of control of its constituent stars will, as a 

 rule, be smallest in the zone of greatest galactic attraction, and hence 

 in that zone systematic nebular deployment from a given star is most 

 restricted. Moreover, the restricted nebula there formed should — 

 other things being equal — pass most rapidly on in its evolution from 

 its initial state toward its later states. This zone of maximum galactic 

 attraction is thus Hkely to take on a relatively barren aspect so far 

 as nebulse are concerned. Great symmetrical nebulas, hke the giant 

 spirals, are essentially incompatible with the conditions of genesis that 

 prevail in this zone. Small spiral nebulse with dimensions confined to 

 the restricted spheres of control of the parent stars should find free con- 

 ditions of deployment in this zone, but they would be under stress to 

 pass promptly from their distinct spiral form into a discoidal form. 



2. In the centroidal region of the galaxy the individual spheres of 

 control of stars grow larger in proportion as the centroid is approached, 



