LIFE AND MECHANISM. 43 order to palliate the disease, but with the hope that a relative cure may follow. The digitalis causes the heart to contract powerfully, so as to overcome the obstruction to the circu- lation and thus keep the patient in health. If this were all, its action would be merely palliative. But while it is being given the heart enlarges, so that after a time it becomes sufficiently powerful to carry on the circulation without the artificial stimulus of the digitalis. Now in this case it is only from an abstract point of view that we can regard the organism as passively undergoing cure through the influence of the digitalis ; although death might speedily have resulted had this drug not been given. The role of the digitalis in the process of cure is as much passive as active. In this process the organism actively utilises the temporary effect of the digitalis on the heart, just as much as it passively submits to the action of the drug. It is thus proper to speak of the physiological action of a drug when its administration is regarded from the abstract point of view of physiology. But it is more correct to speak of its therapeutical fu.iiction than of its therapeutical action, since medicine deals with what is concrete. As another among countless instances of the application of the same principle, may be taken the operation of sewing to- gether the ends of a nerve that has been accidentally divided. This is done in order that the two ends of the nerve may be kept in apposition with certainty. After the lapse of a sufficient time the function of the nerve is entirely restored. When one considers what is implied in this restoration of function, the process seems scarcely less wonderful than the reproduction of limbs in the lower animals. Let us suppose that there is a bundle consisting of several bandied insulated telegraph-wires, each wire being exactly like its neighbours, but coming from a different place : and that this bundle is cut in two. It will be most difficult to find the correspond- ing ends ; and one can imagine the difficulty and confusion thus arising. But the case of the division of a nerve is in every way analogous to that of the division of the bundle of telegraph-wires. What happens usually, if not always, is that the nerve-fibrils of the peripheral portion of the nerve disintegrate, and are replaced after a time by prolongations of the fibrils of the central end, which grow down the connective tissue of the nerve from the point of section. One must suppose, either that these prolongations find their way, each to exactly the same spot as that to which the fibril of which it is a prolongation originally went ; or, what seems more likely, that the exact destination of the new fibrils is more or less indifferent, but that the central nerve-
