270 H. M. STANLEY : and complexity, are logical classifications, and should not be confounded or identified with historical classifications. Logical classifications are undoubtedly to be placed under the domain of philosophy, though we sometimes see them connected with systems of logic. Every complete system of philosophy would naturally include a classification of the sciences ; and within the last century especially it has seemed necessary to philosophers to bring their universal principle to bear on a classification of the sciences. Every system of philosophy, as seeking to account for all phenomena by a single unifying principle, must include the sciences as anthropological phenomena. Hegel gives us a complete classification from the point of view of his philosophy, as also does Comte. Mr. Spencer's classification of the sciences does not seem to be in any vital connexion with his system of philosophy. He uses, as the principle of his classifica- tion, increasing concreteness, whereas, the prime principle of his philosophy being evolution, it would be in consonance with his philosophy that he should give this as the principle of classifica- tion, thus attaining the evolutionary or genetic classification of which we have already spoken. The best principle for a logical classification must be the simplest and most comprehensive, and it will be the aim of the rest of this article to set forth what we may term the unitary principle of classification, as one which may have some value. The fact that there are different levels in nature, the higher de- pendent on and made up of the lower, has often been noticed, but it has never been applied to a classification of the sciences. Science has certainly been revealing more and more clearly that nature is a system of interdependent aggregations of units, and the sciences will, we think, be found to have been gradually de- termining themselves as to subject-matter with reference to these units. Mr. Spencer frequently refers to this conception, but never makes use of it as a principle of classification. For in- stance, we take the following extract from First Principles, pp. 383, 384 : " The automatic movements of the viscera, together with the voluntary movements of the limbs and body at large, arise at the expense of certain molecular movements throughout the nervous and muscular tissues, and these originally arose at the expense of certain other molecular movements propagated by the sun to the earth ; so that both the structural and func- tional motions which organic evolution displays are motions of aggregates generated by the arrested motions of units. Even with the aggregate of these aggregates the same rule holds. For among associated men the pro- gress is ever towards a merging of individual actions in the actions of corporate bodies. While, then, during evolution, the escaping motion becomes, by perpetually widening dispersion, more disintegrated, the motion that is for a time retained becomes more integrated. And so, considered dynamically, evolution is a decrease in the relative movements of parts, and an increase in the relative movements of wholes using the words parts and wholes in their most general sense. The advance is from
