For instance, to take a sensational example, on one occasion in the midst of hilarity while singing, laughing, etc., she suddenly became depressed and burst into tears. What happened was this: It was a sorrowful anniversary, and in the “lower stratum” sad memories had been recurring during the period of hilarity. These memories had come into consciousness early in the morning, but she had resolutely put them out of her mind. They had, however, kept recurring in the lower stratum, and suddenly emerged into the upper stratum of consciousness with the startling effect described. More commonly, however, the emergence of the lower stratum is simply a shifting play of thought. It is interesting to note that censored thoughts and temptations are apt to go into the lower stratum and here with their affects continue at play. _These sometimes reappear as dreams._ Footnote 89: The validity of the evidence of memory as applied to subconscious processes needs to be carefully weighed. It is a question of method, and if the method is fallacious all conclusions fall to the ground. In the sciences of normal psychology and psychiatry and psychopathology, the data given by memory are and necessarily must be relied upon to furnish a knowledge of the content of mental processes and the mental symptoms, and all methods of psychological analysis are based on the data of memory. Without such data there could be no such sciences. As a matter of experience the method is found to be reliable when properly checked by multiple observations. If by special methods of technique mental processes, which do not enter the awareness of the moment, are later brought into consciousness as data of memory, are these data _per contra_ to be rejected as hallucinatory? This is what their rejection would mean. Now, as a fact, there are phenomena, like coconscious personalities, which compel the postulation of coconscious processes. If this is the case, if there are coconscious processes which do not enter awareness, it would be the strangest thing if there were not conditions of the personality in which a memory of these processes could be obtained. This fact would have to be explained. The bringing of coconscious processes into consciousness as data of memory does not seem therefore to be anything _a priori_ improbable and there would seem to be no reason why the memory of them should be more unreliable than that of conscious processes in the forms of attention. Indeed, if the fringe of consciousness be regarded as coconscious, it is an every-day act common to everybody. Such data necessarily should be checked up by multiple observations. Footnote 90: For examples of this kind, see Prince, Experiments to Determine Coconscious Ideation, _Journal of Abnormal Psychology_, April-May, 1908. Footnote 91: The Dissociation, pp. 55, 56. Footnote 92: