As regards those destructive tendencies which are to anger what phobias are to fear, a problem presents itself, the only psychological problem: that of their origin or cause. This question I divide into two: How do they arise? How do they take a determinate direction? I. To explain the origin and appearance of irresistible impulses, most writers have recourse to the hypothesis of degeneration. As it is also called in to explain the converse phenomenon of phobias, it becomes necessary to be a little more precise. Without entering for the moment on the discussion of the different interpretations of this vague word, degeneration, let us take it as synonymous with dissolution or regression. The ideal of heredity, as a conservative principle, is to transmit under a healthy form a healthy organisation, _i.e._ (so far as our subject is concerned), one with harmonious and convergent tendencies. If dissolution is total, we have the idiot, or the dementia patient. If it is partial, we have a breach of equilibrium in favour of one or more tendencies. This disaggregation is not fortuitous; it has a retrogressive character, it is a return to the reflex movements. It approaches the character of the animal, the idiot, or the imbecile; it goes back to that stage of psychic life when the will under its higher form, the arrestive power, was not yet constituted. II. In any case, there remains the principal question: Why was such a tendency predominant? What causes determined the particular direction taken by retrogression—homicide in one case, suicide or erotomania in another? Attempts have been made to explain this by alleging that every irresistible impulse results from the excessive irritation of an _isolated_ group of brain-cells. Besides being purely hypothetical, this explanation is, in spite of its apparent precision, extremely vague. Is there an isolated group of homicidal, or one of kleptomaniac cells? This explanation is really too simple. As far as we can penetrate the very obscure psychological genesis of the destructive impulses (and this may be held to apply to the whole group of irresistible tendencies), we find two sorts of causes at work, the essential and the accidental.