8 Introductory Chapter [CH. i 



axes through the centre, these axes .being in general perpendicular to the 

 longest dimensions of the nebulae. In some cases the motion is more com- 

 plicated than a pure rotation ; thus in N.G.C. 6543 the outer portions seem 

 to have velocities much smaller than those of the central parts. 



The parallax of the planetary nebula N.G.C. 7662 (Plate I) has been found 

 at Mount Wilson to be 0"'023, from which its diameter may be calculated to 

 be 19 times that of the orbit of Neptune. That of the ring nebula in Lyra, 

 N.G.C. 6720 (Plate I) has been found to be 0"'004, but with a probable error 

 comparable to the whole; the corresponding greatest and least diameters 

 are 330 and 250 times those of the orbit of Neptune*. 



Star-clusters 



Yl5. "Further uniformities of formation are to be found in star-clusters. 

 The uniformities are not so definite as those we have just been considering, 

 but are quite definite enough to suggest common origins. There are a great 

 variety of star-clusters, shading often imperceptibly into one another, but they 

 may be classified into three broad types : globular clusters, open clusters 

 and moving clusters f. 



The globular clusters are dense aggregates of stars shewing very great 

 condensation towards the centre. They are approximately globular in form, 

 although Pease and ShapleyJ: have recently found that out of six supposed 

 globular clusters which it was possible to study in detail, five shewed a 

 pronounced departure from the spherical form, being apparently of a flattened 

 or spheroidal form. A similar absence of complete symmetry in some clusters 

 had been previously noticed by Bailey. Bailey has also made counts of the 

 stars in some of these globular star-clusters, and it has been shewn by 

 Plummerll and von Zeipel'F that the law of distribution is approximately 

 uniform. The procedure has been criticised by Shapley** on the grounds 

 that only a few of the brightest stars are included in such counts, but however 

 this may be, there is no question that there is a uniformity of some kind. 

 The number of known globular clusters is at most about 100: Bailey t| gives 

 the number of. " definitely globular" clusters as 76, while Melotte estimates 

 the number as 82. Practically all of these had been discovered by the time 

 of the Herschels. 



* Van Maanen, Ast. Soc. Pac. 171 (Oct. 1917). 



t Shapley, Contributions from the Mount Wilson Solar Observatory, No. 115 (1916), where an 

 excellent summary is given ; also P. J. Melotte, "A Catalogue of Star-clusters," Mem. It. A. 8 70 

 (1915), p. 175. 



* Xat. Acad. of Sciences, 3 (1917), p. 96, and Astrophys. Journ. 45 (1917), p. 225. 



Harvard Coll. Observatory Annals, 76, No. 4. \\ Monthly Notices R.A.8. 76, p. 107. 



U K. Svenka. Vetensk. Acad. Handl. Bd. 51, No. 5. 

 ** Observatory, 39 (1916), p. 452. 

 -ft Harvard Coll. Observatory Annals, 76, p. 43. 



