After sensations and images, we have to name among the phenomena of consciousness, the whole series of affective states--our pleasures and our pains, our joys and our griefs, our sentiments, our emotions, and our passions. It is universally admitted that these states are of a mental nature, for several reasons. (1) We never objectivate them as we do our sensations, but we constantly consider them as indwelling or subjective states. This rule, however, allows an exception for the pleasure and the pain termed physical, which are often localised in particular parts of our bodies, although the position attributed to them is less precise than with indifferent sensations. (2) We do not alienate them as we do our indifferent sensations. The sensations of weight, of colour, and of form serve us for the construction of bodies which appear to us as perceived by us, but as being other than ourselves. On the contrary, we constantly and without hesitation refer our emotional states to our _Ego_. It is I who suffer, we say, I who complain, I who hope. It is true that this attribution is not absolutely characteristic of mental phenomena, for it happens that we put a part of our Ego into material objects, such as our bodies, and even into objects separate from our bodies, and whose sole relation to us is that of a legal proprietorship. We must guard against the somewhat frequent error of identifying the Ego with the psychical. These two reasons sufficiently explain the tendency to see only psychological states in the emotional ones; and, in fact, those authors who have sought to oppose mind to matter have not failed to introduce emotion into their parallel as representing the essence of mind. On this point I will recall the fine ironical image used by Tyndall, the illustrious English physicist, to show the abyss which separates thought from the molecular states of the brain. "Let us suppose," he says, "that the sentiment love, for example, corresponds to a right-hand spiral movement of the molecules of the brain and the sentiment hatred to a left-hand spiral movement. We should then know that when we love, a movement is produced in one direction, and when we hate, in another. But the Why would remain without an answer." The question of knowing what place in our metaphysical theory we ought to secure for emotion seems difficult to resolve, and we even find some pleasure in leaving it in suspense, in order that it may be understood that a metaphysician is not compelled to explain everything. Besides, the difficulties which atop us here are peculiarly of a psychological order. They proceed from the fact that studies on the nature of the emotions are still very little advanced. The physical conditions of these states are pretty well known, and their psychical and social effects have been abundantly described; but very little is known as to what distinguishes an emotion from a thought.