Mrs. Holland’s script records, as investigation seemed to show, the exact words expressing a personal sentiment contained in a letter written to her _twenty years_ before and long forgotten. The letter proving this was accidentally discovered.[10] The following instance of a forgotten experience is, in itself, common enough with everybody, but its recovery by automatic writing illustrates how conservation of the thousand and one simply _forgotten acts of everyday life_ may still persist. It forces, too, a realization of the reason why it is possible that though an act may be forgotten at any given moment it may later at any time flash into the mind. It is still conserved. B. C. A. had been vainly hunting for a bunch of keys which she had not seen or thought of for four months, having been in Europe. One day, soon after her return, while writing a letter to her son she was interrupted by her hand automatically and spontaneously writing the desired information. The letter to her son began as follows: “October 30, 19—. Dear Boy: I cannot find those keys—have hunted everywhere”.... [Here the hand began to write the following, automatically.] “_O, I know—take a pencil_” [Here she did as she was bidden] “_you put those keys in the little box where X’s watch is._” In explanation B. C. A. sent me the following letter: “The keys were found in the box mentioned. I had hunted for them ever since coming home, October 4th. One key belonged to my box in the safety deposit vault and I had felt very troubled and anxious at not being able to find them. I have no recollection now of putting them where I found them.” [Nor was recollection subsequently recovered.] I could give from my own observation if it were necessary as many instances as could be desired of “automatic” reproductions of forgotten experiences of one kind or another the truth of which could be verified by notebook records or other evidence. By a forgotten experience of course is meant something more than what cannot for the moment be voluntarily recalled. I mean something that cannot be remembered at any moment nor under any conditions, _even after the memory has been prodded by the reproduction in the script_—something that is apparently absolutely forgotten. The experience may not only be of a trivial nature but something that happened long in the past and of the kind that is ordinarily absolutely forgotten. I have often invoked the automatic writing (memories) of the subject to recover data elicited in the past in psychological examinations but which both I and the subject had forgotten. Reference to notes always verified the automatic memories. The records of automatic writing to be found in the literature are rich in reproductions showing conservation of forgotten experiences. In fact, given a good subject who can write automatically it is easy to obtain experimentally evidence of this kind at will.