822 The American Naturalist. [September, 
with some subjects, the illusion developed after some practice in the ex- 
periments. This illusion sometimes becomes so frequent that no 
thresholds can be determined ; the subject answers, in response to all 
stimuli, whether of one or of two points, “two points.” (3) A long 
series of experiments showed that this illusion is, for the most part, a 
result of suggestion of some kind; a suggestion which the subject gets 
either from the operator, from the nature of the experiments, or by 
auto-suggestion. It was found that the frequency of the illusion, 
and even its occurrence at all, could be influenced to a marked degree 
by suggestion. In some cases the illusion could be prevented by the 
subject’s discovering the suggestive influence and freeing himself from 
it. (4) Subjects were found who, to start with, gave constant thres- 
holds as long as nothing was suggested to them in regard to the object 
and method of the experiments, but by a suggestion from the operator, 
they were led to show a very rapid reduction of the threshold. After- 
ward, by freeing themselves from the influence of the suggestion, they 
returned, in some cases, to the old constant threshold, freeing them- 
selves at the same time from the illusions which had developed as one 
result of the suggestion. — 
All of these facts go to indicate that both the reduction of the 
threshold by practice and the illusion of two points are the results of 
suggestion in some form. In every instance of the perception of space 
relations by touch, there seems to be involved a process of assimilation 
in which a visual or motor image is the assimilating, and the tactual 
sensations the assimilated, elements. In ordinary life, we test contin- 
ually our tactual sensations by visual images, turning the eyes to look 
at the spot touched. In these experiments this was rendered impossi- 
ble by the fact that the subject could not see the spot on which the ex- 
periments were performed, as this was concealed from him by a screen. 
Hence the place of these images is supplied by memory images con- 
nectedjwith the tactual sensations by past experience, i. e., by associa- 
tion. In our experiments the assimilating visual or motor copies of 
past experiences are not called up by the association with the tactual 
sensationsjalone, but are suggested by other factors. In the localiza- 
tion of a single point, the “local sign” involved is not to be conceived 
of as a simple quality of the tactual sensation, but is rather a relation 
of association between the tactual sensation and some visual or motor 
image. 
Another series of experiments was undertaken in the Princeton 
Laboratoryjby Dr. C. W. Hodge and myself, in which the two stimuli 
were successive, instead of simultaneous, as in the above experiments. 
