rcise of function, or the survival after any
particular kind of activity of a disposition to act again in that
particular way. The revival of a mental impression in the weaker form of
an image is thus, on its physical side, due in part to this remaining
functional disposition in the central nervous tracts concerned. And so,
while on the psychical or subjective side we are unable to find anything
permanent in memory, on the physical or objective side we do find such a
permanent substratum.

With respect to the special conditions of mnemonic revival at any time,
physiology is less explicit. In a general way, it informs us that such a
reinstatement of the past is determined by the existence of certain
connections between the nervous structures concerned in the reviving and
revived mental elements. Thus, it is said that when the sound of a name
calls up in the mind a visual image of a person seen some time since, it
is because connections have been formed between particular regions and
modes of activity of the auditory and the visual centres. And it is
supposed that the existence of such connections is somehow due to the
fact that the two regions acted simultaneously in the first instance,
when the sight of the person was accompanied by the hearing of his name.
In other words, the centres, as a whole, will tend to act at any future
moment in the same complex way in which they have acted in past moments.

All this is valuable hypothesis so far as it goes, though it plainly
leaves much unaccounted for. As to why this reinstatement of a total
cerebral pulsation in consequence of the re-excitation of a portion of
the same should be accompanied by the specific mode of consciousness
which we call recollection of something past, it is perhaps unreasonable
to ask of physiology any sort of explanation.[114]

Thus far as to the general or essential characteristics of memory on its
mental and its bodily side. But what we commonly mean by memory is, on
its psychical side at least, much more than this. We do not say that we
properly recollect a thing unless we are able to refer it to some more
or less clearly defined region of the past, and to localize it in the
succession of experiences making up our mental image of the past. In
other words, though we may speak of an imperfect kind of recollection
where this definite reference is wanting, we mean by a perfect form of
memory something which includes this reference.

Without entering