off best when as nearly unconscious as the conditions will permit.
This position is further supported by the fact that in no case did the "set" appear until the writing of the list of ten four-letter words had become a series of forty consecutive associated finger movements; that is, until the words, and to a degree the letters, as such had dropped out or, perhaps better, become merged with the movements and the sensations resulting from the movements, so as to be concretely indistinguishable. For example E reported at the close of experimentation that "general bodily sensations and tactual-kinæsthetic sensations, together with visual sensations when the eyes are open, make up the content" of the "set." This and other introspections illustrate that the gradual drift of attention is from the extra-bodily factors, with which attention is concerned at the beginning of the experimentation, toward the intra- bodily sensations as the activity becomes automatic until finally attention to details as such is lost. It is then that the tactual and kinæsthetic elements and the inner speech are observed as the only remaining sensory or ideational elements connected with the activity; and these no longer demand a shift of attention from one to the other but all enter into an organized complex with attention given to it as a whole, and without distinction of inner or outer.
This description holds only for the "practice" writing and not for writing from dictation. At the beginning of experimentation one or two of the subjects thought writing from dictation easier than from the copy, but at the close all agreed that the dictation was much less automatic than the "practice" writing. The reason for this is found in the fact that after the "set" had developed to a certain degree the dictation (a word at a time) periodically interrupted a process which had become continuous and automatic.
The dictation writing at this stage illustrates two of Ach's observations, namely, the presence of inteutionalen Bewegungs-empfindungen and the absence of imagery between the reception of the word and the following movement, (op. cit. p. 146.) In dictation the subject develops, so to speak, a special or temporary "set" for each word by running through the word in inner speech with reference to the letters on the cardboard or the corresponding bulbs or fingers, depending upon the degree of automaticity already attained. In early practice this process was repeated essentially after hearing the word and before beginning to write, but after the appearance of the "set" there was no discoverable imagery intervening between the apperception of the word and the beginning of writing. What is found is simply a psychophysical "set" for the par-
