In some cases, after the agent had drawn an animal, a bird, or some
other creature possessing animal life, the percipient’s drawing was
successful, partly successful or at least suggestive in shape; in many
instances it was a flat failure. But as examination proceeded it began
to appear that a number of the failures represented some other form of
the animal kingdom, however diverse. A careful canvass was made,
including the material in hand produced subsequent to that in the
Sinclair book, embracing in all 388 experiments; drawings of human
beings, animals, birds, fishes, insects, and parts of bodies, as a hand
or a leg, were included.
The Agent drew 103 such out of 388.
The Percipient drew 98 such out of 388.
There were found to be 39 correspondences;[22] that is, in 39 cases,
where the agent drew some animal form or part thereof, the percipient
also drew some animal form or part thereof. If out of a total of 388,
the agent makes 103 drawings of this character, chance would give about
26 correspondences, so defined, among the 98 reproductions. In fact,
there are 39, another proof, by a peculiar test, that something more
than chance is in operation.
Now let us make another test, this time including the material only up
to the close of the period covered by the book, and not insisting, as we
have done above, on strict recognition of reproductions, but stating
precisely how they compare with the originals in form.