nothing of the events of the previous day which stimulated the dream; and that the associations of the person analysed are the very source from which we obtain our knowledge of what we call the mental situation. Further, it is especially remarkable, particularly with reference to certain considerations upon which we shall touch later, that the most strenuous opposition has manifested itself again here, over this question of the existence of a symbolic relation between the dream and the unconscious. Even persons of judgement and standing, who in other respects have gone a long way with psycho-analysis, have renounced their adherence at this point. This behaviour is the more remarkable when we remember two things: first, that symbolism is not peculiar to dreams, nor exclusively characteristic of them; and, in the second place, that the use of symbolism in dreams was not one of the discoveries of psycho-analysis, although this science has certainly not been wanting in surprising discoveries. If we must ascribe priority in this field to anyone in modern times, the discoverer must be recognized in the philosopher K. A. Scherner (1861); psycho-analysis has confirmed his discovery, although modifying it in certain important respects. Now you will wish to hear something about the nature of dream-symbolism and will want some examples. I will gladly tell you what I know, but I confess that our knowledge is less full than we could wish. The symbolic relation is essentially that of a comparison, but not any kind of comparison. We must suspect that this comparison is subject to particular conditions, although we cannot say what these conditions are. Not everything with which an object or an occurrence can be compared appears in dreams as symbolic of it, and, on the other hand, dreams do not employ symbolism for anything and everything, but only for particular elements of latent dream-thoughts; there are thus limitations in both directions. We must admit also that we cannot at present assign quite definite limits to our conception of a symbol; for it tends to merge into substitution, representation, etc., and even approaches closely to allusion. In one set of symbols the underlying comparison may be easily apparent, but there are others in which we have to look about for the common factor, the _tertium comparationis_ contained in the supposed comparison. Further reflection may then reveal it to us, or on the other hand it may remain definitely hidden from us. Again, if the symbol is really a comparison, it is remarkable that this comparison is not exposed by the process of free association, and also that the dreamer knows nothing about it, but makes use of it unawares; nay, more, that he is actually unwilling to recognize it when it is brought to his notice. So you see that the symbolic relation is a comparison of a quite peculiar kind, the nature of which is as yet not fully clear to us.