,
by giving us the consciousness of change, make us dimly aware of the
numerical richness of our experiences. But, however this be, there is no
doubt that, in glancing back on such a succession of exciting
transitions of mental condition, time appears to expand enormously, just
as it does in looking back on our dream-experience, or that rapid series
of intensified feelings which, according to De Quincey and others, is
produced by certain narcotics.

The reason of this is plain. Such a type of successive experience offers
to the retrospective imagination a large number of distinguishable
points, and since this mode of estimating time depends, as we have seen,
on the extent of the process of filling in, time will necessarily appear
long in this case. On the other hand, when we have been engaged in very
ordinary pursuits, in which few deeply interesting or exciting events
have impressed themselves on memory, our retrospective picture will
necessarily be very much of a blank, and consequently the duration of
the period will seem to be short.

I observed that this retrospective appreciation of time depended on the
degree of connection between the successive experiences. This condition
is very much the same as the other just given, namely, the degree of
uniformity of the experiences, since the more closely the successive
stages of the experience are connected--as when, for example, we are
going through our daily routine of work--the more quiet and unexciting
will be the transition from each stage to its succeeding one. And on the
other hand, all novelty of impression and exciting transition of
experience clearly involves a want of connection. Wundt thinks the
retrospective estimate of a connected series of experiences, such as
those of our daily round of occupations, is defective just because the
effort of attention, which precedes even an imaginative reproduction of
an impression, so quickly accommodates itself in this case to each of
the successive steps, whereas, when the experiences to be recalled are
disconnected, the effort requires more time. In this way, the estimate
of a past duration would be coloured by the sense of time accompanying
the reproductive process itself. This may very likely be the case, yet I
should be disposed to attach most importance to the number of
distinguishable items of experience recalled.

Our representation of the position of a given event in the past is, as I
have tried to show, determin