§ 15. There is another species of re-distribution among the component matters of organisms, which is not immediately effected by the affinities of the matters concerned, but is mediately effected by other affinities; and there is reason to think that the re-distribution thus caused is important in amount, if not indeed the most important. In ordinary cases of chemical action, the two or more substances concerned themselves undergo changes of molecular arrangement; and the changes are confined to the substances themselves. But there are other cases in which the chemical action going on does not end with the substances at first concerned, but sets up chemical actions, or changes of molecular arrangement, among surrounding substances that would else have remained quiescent. And there are yet further cases in which mere contact with a substance that is itself quiescent, will cause other substances to undergo rapid metamorphoses. In what we call fermentation, the first species of this communicated chemical action is exemplified. One part of yeast, while itself undergoing molecular change, will convert 100 parts of sugar into alcohol and carbonic acid; and during its own decomposition, one part of diastase "is able to effect the transformation of more than 1000 times its weight of starch into sugar." As illustrations of the second species, may be mentioned those changes which are suddenly produced in many colloids by minute portions of various substances added to them--substances that are not undergoing manifest transformations, and suffer no appreciable effects from the contact. The nature of the first of these two kinds of communicated molecular change, which here chiefly concerns us, may be rudely represented by certain visible changes communicated from mass to mass, when a series of masses has been arranged in a special way. The simplest example is that furnished by the child's play of setting bricks on end in a row, in such positions that when the first is overthrown it overthrows the second, the second the third, the third the fourth, and so on to the end of the row. Here we have a number of units severally placed in unstable equilibrium, and in such relative positions that each, while falling into a state of stable equilibrium, gives an impulse to the next sufficient to make the next, also, fall from unstable to stable equilibrium. Now since, among mingled compound molecules, no one can undergo change in the arrangement of its parts without a molecular motion that must cause some disturbance all round; and since an adjacent molecule disturbed by this communicated motion, may have the arrangement of its constituent atoms altered, if it is not a stable arrangement; and since we know, both that the molecules which are changed by this so-called catalysis _are_ unstable, and that the molecules resulting from their changes are _more_ stable; it seems probable that the transformation is really analogous, in principle, to the familiar one named.