oluntarily for purposes of
experimentation by diagramming before each test the intricate curve of
the varying degrees which the intensity of concentration was to assume.
The types may be described as follows:

I. Here the tension curve rises steadily from beginning to end. This
type preponderates in the case of small numbers. Thus, when I asked the
horse: "How much is 2 plus 4?", the tension increased slowly with every
tap from the moment I began counting, until the final tap was reached,
when it was again relaxed. Externally this relaxation is noticeable as a
slight jerk.

II. In this case the curve does not rise at an equal rate, but rather
more slowly at the beginning and later undergoes a sudden increase, or
the tension increases immediately at the beginning, remains constant for
some time and then ascends to the maximum. This curve is the rule in the
case of large numbers and evidently means economy of physical energy,
for experience soon taught that a steady increase in tension from the
very beginning soon brought it to a level which cannot be long
maintained and usually leads to a premature relaxation. In the case of
very large numbers the alternation of the slight and the sudden increase
may be repeated several times, and at times it may even sink below a
level which has already been attained, thus making a wave-like curve.

III. The third type of curve shows a sudden jump between two units at a
certain point in its course. This may occur in the case of both small
and large numbers but only when the highest or first degree of
concentration is employed (see page 91). Such a jump frequently occurs
in the transition from the tap preceding the last to the last one which
is being eagerly expected. Relaxation--with the upward jerk and raising
of the head--here occurs at the normal time; Hans taps to the end with
his right foot. Oftener still the "jump" described occurs while passing
over to the number just before the last. The goal seems within reach and
the mental tension relaxes, and with it the physical tension,--the head
gives a slight jerk and Hans makes the back-step. Since, however,
another tap is still awaited with some degree of tenseness and, since
complete erection of the head does not follow immediately upon the jerk
of the head, the horse gives another tap with the left foot. Thereupon
occurs the complete relaxation of attention, and the assumption of the
erect posture on the part of the questioner. Tha