We will content ourselves with viewing the individual conceptions of memory and recollection as occurring in particular cases and with considering them, now one, now the other, according to the requirements of the case. We shall consider the general relation of “reproduction” to memory. “Reproduction” we shall consider in a general sense and shall subsume under it also the so-called involuntary reproductions which rise in the forms and qualities of past events without being evoked, i.e., which rise with the help of unconscious activity through the more or less independent association of ideas. Exactly this unconscious reproduction, this apparently involuntary activity, is perhaps the most fruitful, and we therefore unjustly meet with unexceptionable distrust the later sudden “occurrence,” especially when these occurrences happen to defendant and his witnesses. It is true that they frequently deceive us because behind the sudden occurrence there often may be nothing more than a better training and instruction from experienced cell-mates; though very often the circumstances are such that the suspect has succeeded through some released prisoner, or by a blackened letter, in sending a message from his prison, by means of which false witnesses of alibi, etc., are provided. Distrust is in any event justified, when his most important witnesses suddenly “occur” to the accused. But this does not always happen, and we find in our own experience evidence of the fact that memory and the capacity to recall something often depend upon health, feeling, location, and chance associations which can not be commanded, and happen as accidentally as anything in life can. That we should remember anything at all depends upon the point of time. Everybody knows how important twilight may be for memory. Indeed, twilight has been called the visiting-hour of recollection, and it is always worth while to observe the situation when anybody asserts that some matter of importance occurred to him in the twilight. Such an assertion merits, at least, further examination. Now, if we only know how these occurrences constitute themselves, it would not be difficult to study them out and to estimate their probability. But we do not know, and we have to depend, primarily, on observation and test. Not one of the theories applied is supported by experience altogether. They may be divided into three essential groups. 1. What is received, fades away, becomes a “trace,” and is more or less overlaid by new perceptions. When these latter are ever set aside, the old trace comes into the foreground. 2. The ideas sink, darken, and disintegrate. If they receive support and intensification they regain complete clearness. 3. The ideas crumble up, lose their parts. When anything occurs that reunites them and restores what is lost, they become whole again.