dhound or of the man hurrying to a train or seeking in the crowd
for a friend, there we have to suppose that a center, once aroused to
activity and prevented from complete discharge, remains active by
virtue of energy dammed up in itself. There is pretty good
physiological evidence that this sort of thing is a fundamental fact;
for there are certain rhythmical reflexes, like scratching or
stepping, that, when started going by a momentary sensory stimulus,
keep it up for a time after the stimulus has ceased. There seems to be
no doubt that a nerve center, once aroused, may stay aroused for a
time.

The "dammed-up energy" here is not to be confused with the "stored
energy" spoken of under the head of reactions. We said, in that
connection, that a stimulus released energy stored in the organism.
That, however, was _potential_ energy, dormant within the organism
till aroused; but what we have here in mind is active or _kinetic_
energy. Stored energy is like that of coal in the bin; dammed-up
energy is like that of steam in the boiler.

Dammed-up energy in the nerve centers accounts for the persistence of
a tendency to reaction after the stimulus has ceased. It accounts for
the "delayed reaction" and similar cases. But how shall we account for
preparatory reactions? We have a nerve center in an active state,
tending to discharge into a certain lower motor center, but unable to
do so because a peripheral stimulus is necessary, in addition, in
order to arouse this lower center. Then we find the higher center
discharging into _other_ lower centers, and so giving rise to
preparatory reactions. More precisely, what we find is that the higher
center facilitates the response {84} of certain lower centers to their
proper peripheral stimuli, while inhibiting the response of other
lower centers to their appropriate stimuli. This is the same sort of
thing that we observe in all control exerted by a higher center over a
lower. It means that the higher center, besides its main line of
connection with the lower center that will give the end-reaction, has
minor lines of connection with certain other lower centers; some of
these centers it facilitates and others it inhibits. These connections
between the main and the subordinate centers may have been established
by inborn nature, or by previous training, as will be explained in
later chapters.

The action of the main center on the subordinate centers concerned in
executing preparatory r