painful sensations felt by him or threatening him. Against certain painful excitations from within the ego has only the same means of defence as that employed against pain coming from without, and this is the starting-point of important morbid disturbances. In this way the ego detaches itself from the external world. It is more correct to say: Originally the ego includes everything, later it detaches from itself the external world. The ego-feeling we are aware of now is thus only a shrunken vestige of a far more extensive feeling—a feeling which embraced the universe and expressed an inseparable connection of the ego with the external world. If we may suppose that this primary ego-feeling has been preserved in the minds of many people—to a greater or lesser extent—it would co-exist like a sort of counterpart with the narrower and more sharply outlined ego-feeling of maturity, and the ideational content belonging to it would be precisely the notion of limitless extension and oneness with the universe—the same feeling as that described by my friend as ‘oceanic’. But have we any right to assume that the original type of feeling survives alongside the later one which has developed from it? Undoubtedly we have: there is nothing unusual in such a phenomenon, whether in the psychological or in other spheres. Where animals are concerned we hold the view that the most highly developed have arisen from the lowest. Yet we still find all the simple forms alive to-day. The great saurians are extinct and have made way for the mammals, but a typical representative of them, the crocodile, is still living among us. The analogy may be too remote, and it is also weakened by the fact that the surviving lower species are not as a rule the true ancestors of the present-day more highly developed types. The intermediate members have mostly died out and are known to us only through reconstruction. In the realm of mind, on the other hand, the primitive type is so commonly preserved alongside the transformations which have developed out of it that it is superfluous to give instances in proof of it. When this happens, it is usually the result of a bifurcation in development. One quantitative part of an attitude or an impulse has survived unchanged while another has undergone further development.