He recognises the street as one down which he had walked on the previous day. The child he acknowledges as his brother's child, whom he had seen on the previous evening when visiting his brother. The motor accident reminds him of an accident that had actually occurred a few days before, but of which he had only read an account in a newspaper. Popular opinion is known to be satisfied with this kind of explanation. People say: "Oh, that is why I dreamt such and such a thing!" Obviously this explanation is absolutely unsatisfactory from a scientific standpoint. The dreamer walked down many streets on the previous day; why was this particular one selected? He had read about several accidents; why did he select just this one? The mere disclosure of an antecedent is by no means sufficient; for a plausible determination of the dream presentation can only be obtained from the competition of various determinants. The collection of additional material proceeds, according to the principle of recollection that has been called the _Association Method_. The result, as will easily be understood, is the admission of a mass of multifarious and quite heterogeneous material, having apparently nothing in common but the fact of its evident associative connection with the dream contents, since it has been reproduced by means of this content. How far the collection of such material should go, is an important question from the technical point of view. Since the entire psychic content of a life may be ultimately disclosed from any single starting point, theoretically the whole previous life-experience might be found in every dream. But we only need to assemble just so much material as is absolutely necessary in order to comprehend the dream's meaning. The limitation of the material is obviously an arbitrary proceeding, according to that principle of Kant's whereby to _comprehend_ is "_to perceive to the extent necessary for our purpose_." For instance, when undertaking a survey of the causes of the French Revolution, we could, in amassing our material, include not only the history of medieval France but also that of Rome and Greece, which certainly would not be "necessary for our purpose," since we can comprehend the historical genesis of the Revolution from much more limited material. Except for the aforesaid arbitrary limitation, the collecting of material lies outside the investigator's discretion. The material gathered must now be sifted and examined, according to principles which are always applied to the examination of historical or any empirical scientific material. The method is an essentially comparative one, that obviously cannot be applied automatically, but is largely dependent upon the skill and aim of the investigator.