822 COSMOS. 



conditions under which, in a globular or spheroi dally flattened 

 stellar cluster, the rotating crowded suns, whose specific 

 density is greater towards the centre, constitute a system of 

 equilibrium; 88 this difficulty increases still more in those 

 circular, well-defined, planetary, nebulous discs which exhibit 

 a perfectly uniform brightness, without any increase of in- 

 tensity towards the centre. Such a condition seems to depend 

 less upon sphericity of form (the state of aggregation of 

 many thousand small stars,) than on the existence of a 

 gaseous photosphere, which is supposed, as in our Sun, to 

 be covered with a thin, untransparent, or very faintly illu- 

 minated stratum of vapour. Does the light in the planetary 

 nebulous disc appear to be thus uniformly diffused, simply in 

 consequence of the great distance, which causes the difference 

 between the centre and the margins to disappear ? 



The fourth and last order of regular nebula? comprises Sir 

 William Herschel's nebulous stars, i. e. true stars surrounded 

 by a milky nebula, which is very probably connected with, 

 and dependent upon, the central star. Very different opinions 

 exist as to whether the nebula, which, according to Lord 

 Rosse and Mr. Stoney, appears to be perfectly annular in 

 some of these groups (Philos. Transact, for 1850, pi. xxxviii. 

 figs. 15 and 16), is self-luminous, forming a photosphere 

 like our Sun, or whether) which, however, is less pro- 



63 On the development of the dynamic relations manifested 

 in the partial attractions in the interior of a globular cluster 

 of stars, which appears in a telescope of weak power, as a 

 round nebula increasing in density towards the centre, see 

 Sir John Herschel, in Outlines if Astronomy, 866 and 

 872; Observations at the Cape, 44, 111 to 113; Philos. 

 Transact, for 1833. p. 501 ; Address of the President in the 

 Report of the Fifteenth Meeting of the British Association, 

 1815, p. xxx vii. 



