The neuroses were the first subject of analysis and for a long time they were the only one. No analyst could doubt that medical practice was wrong in separating those disorders from the psychoses and in attaching them to the organic nervous diseases. The theory of the neuroses belongs to psychiatry and is indispensable as an introduction to it. It would seem, however, that the analytical study of the psychoses is impracticable owing to its lack of therapeutic results. Mental patients are, as a rule, without the capacity for forming a positive transference, so that the principal instrument of analytic technique is inapplicable to them. There are, nevertheless, a number of methods of approach to be found. Transference is often not so completely absent but that it can be used to a certain extent; and analysis has achieved undoubted successes with cyclical depressions, light paranoic modifications and partial schizophrenias. It has at least been a benefit to science that in many cases the diagnosis can oscillate for quite a long time between assuming the presence of a psychoneurosis or of a dementia præcox; for therapeutic attempts initiated in such cases have resulted in valuable discoveries before they have had to be broken off. But the chief consideration in this connection is that so many things that, in the neuroses, have to be laboriously fetched up from the depths, are found in the psychoses upon the surface, visible to every eye. So that the best subjects for the demonstration of many of the assertions of analysis are provided by the psychiatric clinic. It was thus bound to happen before long that analysis would find its way to the objects of psychiatric observation. At a very early date (1896) I was able to establish, in a case of paranoid dementia, the presence of the same ætiological factors and the same emotional complexes as in the neuroses. Jung explained some most puzzling stereotypes in dements by bringing them into relation with the patients’ life histories; Bleuler demonstrated the existence in various psychoses of mechanisms like those which analysis had discovered in neurotics. Since then analysts have never relaxed their efforts to come to an understanding of the psychoses. Especially since it has been possible to work with the concept of narcissism, they have managed, now in this place and now in that, to get a glimpse beyond the wall. Most of all, no doubt, was achieved by Abraham in his elucidation of melancholia. It is true that in this sphere all our knowledge is not yet converted into therapeutic power; but the mere theoretical gain is not to be despised, and we may be content to wait for its practical application. In the long run even the psychiatrists have been unable to resist the convincing force of their own clinical material. At the present time German psychiatry is undergoing a kind of “peaceful penetration” by analytic views. While they continually declare that they will never be psychoanalysts, that