One further preliminary is requisite. We must for a moment return to those physico-chemical data set down in the first chapter of this work (§§ 7–8). We there saw that the complex and large-atomed colloids which mainly compose living organic matter, have extremely little molecular mobility; and, consequently, extremely little power of diffusing themselves. Whereas we saw not only that those absorbed matters, gaseous and liquid, which further the decomposition of living organic matter, have very high diffusibilities, but also that the products of the decomposition are much more diffusible than the components of living organic matter. And we saw that, as a consequence of this, the tissues give ready entrance to the substances which decompose them, and ready exit to the substances into which they are decomposed. Hence it follows that, under its initial form, uncomplicated by nervous and other agencies, the escape of effete matters from the organism, is a physical action parallel to that which goes on among mixed colloids and crystalloids that are dead or even inorganic. Excretion is a specialized form of this spontaneous action; and we have to inquire how the specialization arises. Two causes conspire to establish it. The first is that these products of decomposition are diffusible in widely different degrees. While the carbonic acid and water permeate the tissues with ease in all directions, and escape more or less from the exposed surfaces, urea, and other waste substances incapable of being vaporized, cannot escape thus readily. The second is that the different parts of the body, being subject to different physical conditions, are from the outset sure severally to favour the exit of these various products of decomposition in various degrees. How these causes must have co-operated in localizing the excretions, we shall see on remembering how they now co-operate in localizing the separation of morbid materials. The characteristic substances of gout and rheumatism have their habitual places of deposit. Tuberculous matter, though it may be present in various organs, gravitates towards some much more than towards others. Certain products of disease are habitually got rid of by the skin, instead of collecting internally. Mostly, these have special parts of the skin which they affect rather than the rest; and there are those which, by breaking out symmetrically on the two sides of the body, show how definitely the places of their excretion are determined by certain favouring conditions, which corresponding parts may be presumed to furnish in equal degrees. Further, it is to be observed of these morbid substances circulating in the blood, that having once commenced segregating at particular places, they tend to continue segregating at those places. Assuming, then, as we may fairly do, that this localization of excretion, which we see continually commencing afresh with morbid matters, has always gone on with the matters produced by