462 Clusters and Nebulce. 



From a great many occasional inspections of this superb 

 cluster, H. inclined to attribute the appearance of two sizes of 

 stars to " little groups and knots of the smaller size lying so 

 nearly in the same visual line as to run together by the aber- 

 rations of the eye and telescope : this explanation of an ap- 

 pearance often noticed in the descriptions of such clusters is « 

 corroborated in the present instance by the distribution of 

 these apparently larger stars in rings or mesh-like patterns, 

 chiefly about the centre, where the stars are most crowded." 

 This ingenious supposition does not, however, quite account 

 for the equality in size of the larger stars, which, as necessarily 

 composed of groups varying in number, would, it might be 

 expected, show more variety in brightness also. But how- 

 ever correct it may be in this or other instances, it is evidently 

 not applicable to cases such as we have recently described, 

 M 3, and especially M 5, in which the larger stars occur as 

 frequently among the stragglers, where such coincidences must 

 of necessity be far less common. Computation, or even 

 graphical projection, would show what, on the hypothesis of 

 visual coincidence, ought to be the ratio of increase in such 

 combinations as we approach the centre of the mass : the pre- 

 liminary assumption of symmetry, whether in equidistant 

 arrangement or progressive condensation, would, of course, 

 be seldom fulfilled; but the effect of irregularities would be 

 limited, and might be allowed for within certain bounds of pro- 

 bability. And by working steadily on in this direction we might 

 approximate more nearly to a true idea of the internal structure. 



To those possessed of adequate instruments — to which 

 must be added, a knowledge of the effects of perspective — the 

 investigation of the mode of combination and distribution in 

 these grand and mysterious aggregations may be pointed out as 

 an interesting pursuit. We seem already to be upon the trace 

 of some general laws. H. has taught us to look for larger 

 and ruddier stars in a central position ; the Earl of Rosse has 

 pointed out a tendency to curvature in the outlying branches, 

 and the occasional presence of dark rifts or " lanes ;" and it 

 may fairly be expected that persevering examination, careful 

 drawing, and systematic comparison of the principal clusters, 

 may lead to the detection of other peculiarities, of some sig- 

 nificance it may be, at least, to astronomers as yet unborn. 

 However advanced we may deem ourselves — as we unques- 

 tionably are — in some, and those very important respects, in 

 others we must even now be satisfied with laying foundations. 

 The " adhuc plus ultra est" of Kepler will never be out of 

 date. As it has been given to us to raise a noble superstruc- 

 ture on the labours of earlier workers, so we must be content 

 to do in turn the same preliminary office for future generalizers 



