The justification of the title of this chapter must be reached along other lines. The tedious, because entirely new, demonstration of the earlier part of my work has shown that the human memory stands in intimate relation with things hitherto supposed unconnected with it--such things as time, value, genius, immortality. I have attempted to show that memory stands in intimate connection with all these. There must be some strong reason for the complete absence of earlier allusions to this side of the subject. I believe the reason to be no more than the inadequacy and slovenliness which hitherto have spoiled theories of memory. I must here call attention to a theory first propounded by Charles Bonnet in the middle of the eighteenth century and towards the end of the nineteenth century, specially insisted upon by Ewald Hering and E. Mach. This theory regarded the human memory as being only a special case of a property common to all organised matter, the property that makes the path of new stimuli rather easier if these resemble stimuli that have acted at some former time. The theory really makes the human memory an adaptation in the sense of Lamarck, the result on the living organism of repeated stimulation. It is true that there is a point in common between the human memory and the increase of sensitiveness caused by the repeated application of a stimulus; that identical element consists in the permanence of the effect of the first stimulation. There is, however, a fundamental difference between the growth of a muscle that is much used or the adaptation of the eater of arsenic or morphia to increased doses, and the recollection of past experiences by human beings. In the one case the trace of the old is just to be felt in the new stimulation; in the other case, by means of the consciousness, the old situations are actually reproduced with all their individuation. The identification of the two is so superficial that it is a waste of time to dwell longer on it. The doctrine of association as the theory of memory is linked with the foregoing physiological theory as a matter of history, through Hartley, and, as a matter of fact, because the idea of habit is shared by the two. The association theory attributes memory to the mechanical play of the linking of presentations according to four laws. It overlooks the fact that memory (the continuous memory of man) is a function of the will. I can remember a thing if I really _will_. In the case of hypnosis, when the recollection of all that has been forgotten is induced, an outside will replaces the will of the subject. It is will that sets in action the chains of association, and we have to deal here with something deeper than a mechanical principle.