CELESTIAL MEASURINGS AND WEIGHINGS. 2I7 



any bounds to the further extension of that principle of 

 systematic subordination which we have ah-eady traced 

 to a certain extent ; and which combines in its fullest 

 conception a unity of plan and singleness of result with 

 an unlimited multiplicity of subordinated individuals, 

 groups, systems, and families of systems. Thus it by no 

 means follows that all those objects which stand classed 

 under the general designation of " nebulae " or " clusters 

 of stars," and of which the number already known 

 amounts to upwards of five thousand, are objects (looked 

 upon from this point of view) of the same order. Among 

 those dim and mysterious existences, which only a prac- 

 tised eye, aided by a powerful telescope, can pronounce 

 to be something diffei'eiit from minute stars, may, for any- 

 thing we can prove to the contrary, be included systems 

 of a higher order than that which comprehends all our 

 nebulae (properly such) reduced by immensity of distance 

 to the very last limit of visibility. And this conception, 

 we may remark, affords something like a reasonable 

 answer to those who have assumed an imperfect ti-ajis- 

 parcncy of the celestial spaces, on the ground that, but 

 for some such cause, the whole celestial vault ought to 

 blaze with solar splendour, seeing that in no direction of 

 the visual ray, if continued far enough, would it fail to 

 meet with a star. Such would no doubt be the case 

 were all space occupied by stars disseminated tln-ough it 

 uniformly, i.e.^ so that the same number of stars should 

 in every region be comprised in the same space. But no 

 such consequence would follow were the law of sidereal 

 distribution such as we have been here descri])ing : a 



