effects are strictly proportioned to causes. I am obliged, therefore, to seek, in the image which I call my body, and which follows me everywhere, for changes which shall be the equivalents—but the well-regulated equivalents, now deducible from each other—of the images which succeed one another around my body: the cerebral movements, to which I am led back in this way, again are the duplicates of my perceptions. It is true that these movements are still perceptions, 'possible' perceptions,—so that this second hypothesis is more intelligible than the first; but, on the other hand, it must suppose, in its turn, an inexplicable correspondence between my real perception of things and my possible perception of certain cerebral movements which do not in any way resemble these things. When we look at it closely, we shall see that this is the reef upon which all idealism is wrecked: there is no possible transition from the order which is perceived by our senses to the order which we are to conceive for the sake of our science,—or, if we are dealing more particularly with the Kantian idealism, no possible transition from sense to understanding.—So my only refuge seems to be ordinary dualism. I place matter on this side, mind on that, and I suppose that cerebral movements are the cause or the occasion of my representation of objects. But if they are its cause, if they are enough to produce it, I must fall back, step by step, upon the material-
