206 H. SIBGWICK : actual condition exhibits. But whatever success they may have had in tracing back the past states of the physical uni- verse has not really helped them a step towards a philo- sophical solution of this problem : all they have done is to change one particular mode of arbitrariness and irregularity for another no less apparently unaccountable. This negative result, indeed, is not always plain at first sight. E.g., when we first consider the formula in which Mr. Spencer generalises the process through which the physical universe has passed, and contemplates matter " passing from an indefinite, incoherent homogeneity to a definite coherent heterogeneity," it seems at first as if our complex of arbitrary differences would be ultimately simplified away if we could retrace this process far enough back. But reflection shows that the " indefiniteness " which Mr. Spencer attributes to primeval matter is not a condition of matter as we conceive it to have existed, but only relates to its apprehension by our limited intellects if we conceive any particle of matter as existing at all, we of necessity conceive its spatial and dynamic relations as perfectly definite. Similarly, we are forced to conceive every particle of matter as always in a sense coherent that is, connected by dynamic relations with every other particle ; and whatever heterogeneity the whole aggregate now possesses requires us to suppose a cor- responding heterogeneity at every point of the process of com- plex motion through which it has passed in time. The pro- cess which Mr. Spencer describes as a process from the homogeneous to the heterogeneous is a process which may increase the amount of difference between the parts of space compared, in respect of their occupation by matter ; but it is not a process that can originate difference. With whatever confidence we may give the rein to the most audacious of speculative astronomers, and under his guidance sweep back through eras of time to the most diffused of nebulae, we shall yet find in the nebula with which we leave off a complex of apparently arbitrary and irregular differences, needing explanation just as much or just as little as the particu- larities of our actual planet, rolling in the " glearn of a million millions of suns ". 1 1 1 saying this, I do not mean in any way to depreciate the interest and importance of attempts to trace out the past history of the cosmos, by speculative geology and speculative astronomy combined : I merely point out that, whatever degree of success may crown such efforts, there is no pros- pect that they will either tend to solve the philosophical pro- blem prescribed by the actual particularity of the cosmos, or
