In the tests just described we had to do only with such ideas or concepts as normally were associated with some stereotyped form of expressive movement (see page 106). I now chose a group of ideas which are not normally associated with a particular form of motor expression peculiarly characteristic of them, and sought to establish artificially such a connection with some arbitrary movement, without consciousness of the process on the part of the subject. Thus I asked one subject (Miss St.), who had no intimation of the aim of the tests, to think of the following words in any order she might choose: "Ibis" (ibis), "Irbis" (panther), "Kiebitz" (plover) and "Kuerbis" (pumpkin). I said that I would react to her thoughts by means of arm movements forward and backward to the right and to the left, respectively. 15 out of 20 tests were successful, without the slightest suspicion on the part of the subject (whose whole attention was concentrated on the word-content), that she was giving me the necessary directives in the form of very minute movements of the head and eyes to the right and left, etc. She was greatly astonished that I should be able to guess words so much alike,--(she did not know that the element of likeness was productive of no difficulty). When, during one of the tests, the subject happened to think spontaneously of the movement she was expecting me to make, she became confused, and as a result the number of my sucessful reactions suddenly fell. I never would have discovered the cause, had not the subject enlightened me without my asking.