From these facts there results a third fact, the consequence of the first two: the consciousness, which directs the limb withdrawn from the personality, appears to have more considerable resources—at least from a memory point of view—than the normal consciousness. If it be correct to speak of apparent disaggregation in that which concerns the conscious normal personality, it seems to me that this expression ceases to represent the facts, as soon as it can be demonstrated, that the consciousness manifested by the automatism is more extensive than the normal consciousness. If we are to attach a precise meaning to language—and Janet’s language is so clear and simple that we may not accuse this elegant and remarkable writer of want of precision—the idea of disaggregation implies the division of the personal consciousness into elementary parts, according to definition, lesser than the whole. This phenomenon is frequently observed, _e.g._ when automatic writing shows itself to be incapable of logical co-ordination, of which I have given examples; sometimes there is no trace of thought, properly so called, _e.g._ when the sensitive confines himself to repeating _sine die_ the same letter, or traces nothing but lines, and strokes, etc. But can we consider the case as one of veritable disaggregation where the hand, withdrawn from normal consciousness, appears to dispose of a greater mass of souvenirs than the normal consciousness does? Janet himself has verified the fact, and gives some examples of it in his work, _Névroses et idées fixes_, vol i. After that, is it not contradictory to say (_Automatisme psychologique_, p. 452): ‘The result of our studies has been to bring back the diverse phenomena of automatism to their essential conditions—most of these phenomena depend upon a state of anæsthesia or abstraction. This state is connected with the narrowing of the field of consciousness, and this narrowing itself is due to the feebleness of synthesis and the disaggregation of the mental compound into diverse groups smaller than they should normally be. These diverse points are easy to verify; the state of abstraction, incoherence, of disaggregation, in a word, of suggestible individuals has often been pointed out.’ How can a group, smaller than the mental compound of which it forms one of the parts, be more considerable than that compound? How can a part be greater than its whole? This is, nevertheless, a fact easily verifiable in the domain of memory and sometimes in that of intelligence. Janet’s theory explains only some of the observable facts; it is only partially true. It suffices to compare the quotation I have just given with what he says in his work, _Névroses et idées fixes_, vol. i. p. 137: ‘The souvenir even in somnambulism only exists if the patient be oblivious to everything and replies automatically to questions, by the mechanical association of ideas without reflection, without the personal perception of what he is doing.