Spencer's manner of looking at the social organism may be illustrated in what he says about growth in "social aggregates." When we say that growth is common to social aggregates and organic aggregates, we do not thus entirely exclude community with inorganic aggregates. Some of these, as crystals, grow in a visible manner; and all of them on the hypothesis of evolution, have arisen by integration at some time or other. Nevertheless, compared with things we call inanimate, living bodies and societies so conspicuously exhibit augmentation of mass, that we may fairly regard this as characterizing them both. Many organisms grow throughout their lives; and the rest grow throughout considerable parts of their lives. Social growth usually continues either up to times when the societies divide, or up to times when they are overwhelmed. Here, then, is the first trait by which societies ally themselves with the organic world and substantially distinguish themselves from the inorganic world.[28] In this same way, comparing the characteristic general features of "social" and "living bodies," noting likeness and differences, particularly with reference to complexity of structure, differentiation of function, division of labor, etc., Spencer gives a perfectly naturalistic account of the characteristic identities and differences between societies and animals, between sociological and biological organizations. It is in respect to the division of labor that the analogy between societies and animals goes farthest and is most significant. This division of labour, first dwelt upon by political economists as a social phenomenon, and thereupon recognized by biologists as a phenomenon of living bodies, which they called the "physiological division of labour," is that which in the society, as in the animal, makes it a living whole. Scarcely can I emphasize enough the truth that in respect of this fundamental trait, a social organism and an individual organism are entirely alike.[29] The "social aggregate," although it is "discrete" instead of "concrete"--that is to say, composed of spatially separated units--is nevertheless, because of the mutual dependence of these units upon one another as exhibited in the division of labor, to be regarded as a living whole. It is "a living whole" in much the same way that the plant and animal communities, of which the ecologists are now writing so interestingly, are a living whole; not because of any intrinsic relations between the individuals who compose them, but because each individual member of the community, finds in the community as a whole, a suitable milieu, an environment adapted to his needs and one to which he is able to adapt himself.