Ending here this exposition, somewhat too speculative and running into details inappropriate to a work of this kind, it suffices to note the most general facts concerning metabolism. Regarded as a whole it includes, in the first place, those anabolic or building-up processes specially characterizing plants, during which the impacts of ethereal undulations are stored up in compound molecules of unstable kinds; and it includes, in the second place, those katabolic or tumbling-down changes specially characterizing animals, during which this accumulated molecular motion (contained in the food directly or indirectly supplied by plants), is in large measure changed into those molar motions constituting animal activities. There are multitudinous metabolic changes of minor kinds which are ancillary to these--many katabolic changes in plants and many anabolic changes in animals--but these are the essential ones.[13] CHAPTER IV.[14] PROXIMATE CONCEPTION OF LIFE. § 24. To those who accept the general doctrine of Evolution, it need scarcely be pointed out that classifications are subjective conceptions, which have no absolute demarcations in Nature corresponding to them. They are appliances by which we limit and arrange the matters under investigation; and so facilitate our thinking. Consequently, when we attempt to define anything complex, or make a generalization of facts other than the most simple, we can scarcely ever avoid including more than we intended, or leaving out something which should be taken in. Thus it happens that on seeking a definite idea of Life, we have great difficulty in finding one that is neither more nor less than sufficient. Let us look at a few of the most tenable definitions that have been given. While recognizing the respects in which they are defective, we shall see what requirements a more satisfactory one must fulfil. Schelling said that Life is the tendency to individuation. This formula, until studied, conveys little meaning. But we need only consider it as illustrated by the facts of development, or by the contrast between lower and higher forms of life, to recognize its significance; especially in respect of comprehensiveness. As before shown, however (_First Principles_, § 56), it is objectionable; partly on the ground that it refers not so much to the functional changes constituting Life, as to the structural changes of those aggregates of matter which manifest Life; and partly on the ground that it includes under the idea Life, much that we usually exclude from it: for instance--crystallization. The definition of Richerand,--"Life is a collection of phenomena which succeed each other during a limited time in an organized body,"--is liable to the fatal criticism, that it equally applies to the decay which goes on after death. For this, too, is "a collection of phenomena which succeed each other during a limited time in an organized body."