3. It has long been observed that, in the great majority of cases, mental disease begins by affective disturbances and that the intellectual aberrations only make their appearance later. Much more recently, a doctrine has been propounded which tends to seek the primary cause of these affective disturbances in a self-intoxication—_i.e._, in “the disorders produced in the interior of the organism by the excessive formation or the morbid retention of normal poisons; in particular, by those originating in the digestive canal and the urine.” Nutritive troubles through acceleration, retardation, or perversion, are assigned as the most general cause. In support of this, we are referred to the relations between melancholia, hypochondria, a pessimistic disposition, with hyperchlorhydria of the stomach, and the good results of purgative medicines; the numerous mental modifications coinciding with organic chemical modifications—_e.g._, certain attacks of mania in arthritic subjects. “One characteristic of the mental state of diabetic patients is the way in which the fluctuations of the mental state correspond with those of the sugar, and the barometric (if one may so call it) influence of the composition of the urine on the moral disposition.” This liquid, in mania, loses, to a great extent, its toxic character, in consequence of the morbid retention of normal poisons which are no longer eliminated.[84] ----- Footnote 84: Bouchard, _Leçons sur les auto-intoxications_; _Leçons sur les maladies par ralentissement de nutrition_. Régis, _Traité des maladies mentales_, pp. 112, 415, 423, etc. Féré, _Pathologie des émotions_, pp. 264, 495 _et seq._ ----- A long enumeration of facts bearing on this as yet insufficiently studied question would be here out of place. Besides, it could only be of real value if systematic—_i.e._, if under the heading of each emotion were grouped the physiological facts invariably accompanying it, and all the chemical modifications exclusively peculiar to it. We have only included the chemical conditions in our study, in order to penetrate as far as possible into the most general conditions of affective life, and show once more why it betrays the inmost constitution of the individual.