A word designates rather a current of thought than one or several halts on a logical path. Here again a dynamic continuity exists previous to the parcelling out of the acceptations. What, then, should be the attitude of the mind? A supple moving attitude more attentive to the curve of change than to the possible halting-points along the road. But this is not the case at all; the effort would be too great, and what happens, on the contrary, is this. For the spectrum a chromatic scale of uniform tints is very quickly substituted. This is in itself an undesirable simplification, for it is impossible to reconstitute the infinity of real shades by combinations of fundamental colours each representing the homogeneous shore, which each region of the spectrum finally becomes. However cleverly we proportion these averages, we get, at most, some vulgar counterfeit: orange, for example, is not a mixture of yellow and red, although this mixture may recall to those who have known it elsewhere the simple and original sensation of orange. Again, a second simplification, still more undesirable, succeeds the first. There are no longer any colours at all; black lines serve as guide-marks. We are therefore with pure concepts decidedly in full symbolism. And it is with symbols that we shall henceforward be trying to reconstruct reality. I need not go back to the general characteristics or the inconveniences of this method. Concepts resemble photographic views; concrete thickness escapes them. However exact, varied, or numerous we suppose them, they can certainly recall their object, but not reveal it to any one who had not had any direct intuition of it. Nothing is easier than to trace the plan of a body in four dimensions; all the same, this drawing does not admit "visualisation in space" as is the case with ordinary bodies, for want of a previous intuition which it would awaken: thus it is with concepts in relation to reality. Like photographs and like plans, they are extracted from reality, but we are not able to say that they were contained in it; and many of them besides are not so much as extracts; they are simple systematised notes, in fact, notes made upon notes. In other terms, concepts do not represent pieces, parts, or elements of reality. Literally they are nothing but simple symbolic notations. To wish to make integral factors of them would be as strange an illusion as that of seeing in the co-ordinates of a geometric point the constitutive essence of that point. We do not make things with symbols, any more than we should reconstruct a picture with the qualifications which classify it. Whence, then, comes the natural inclination of thought towards the concept? From the fact that thought delights in artifices which facilitate analysis and language.