“The combination of the manifold of perception” is repeatedly given as the essential operation of the understanding, by means of its twelve categories. Yet this is never adequately explained, nor is it shown what this manifold of perception is before it is combined by the understanding. But time and space, the latter in all its three dimensions, are _continua_, _i.e._, all their parts are originally not separate but combined. Thus, then, everything that exhibits itself in them (is given) appears originally as a _continuum_, _i.e._, its parts appear already combined and require no adventitious combination of a manifold. If, however, some one should seek to interpret that combining of the manifold of perception by saying that I refer the different sense-impressions of one object to this one only—thus, for example, perceiving a bell, I recognise that what affects my eye as yellow, my hand as smooth and hard, my ear as sounding, is yet only one and the same body,—then I reply that this is rather a consequence of the knowledge _a priori_ of the causal nexus (this actual and only function of the understanding), by virtue of which all those different effects upon my different organs of sense yet lead me only to one common cause of them, the nature of the body standing before me, so that my understanding, in spite of the difference and multiplicity of the effects, still apprehends the unity of the cause as a single object, which just on that account exhibits itself in perception. In the beautiful recapitulation of his doctrine which Kant gives at p. 719-726 or V. 747-754 of the “Critique of Pure Reason,” he explains the categories, perhaps more distinctly than anywhere else, as “the mere rule of the synthesis of that which empirical apprehension has given _a posteriori_.” It seems as if here he had something in his mind, such as that, in the construction of the triangle, the angles give the rule for the composition of the lines; at least by this image one can best explain to oneself what he says of the function of the categories. The preface to the “Metaphysical First Principles of Natural Science” contains a long note which likewise gives an explanation of the categories, and says that they “differ in no respect from the formal acts of the understanding in judging,” except that in the latter subject and predicate can always change places; then the judgment in general is defined in the same passage as “an act through which given ideas first become knowledge of an object.” According to this, the brutes, since they do not judge, must also have no knowledge of objects. In general, according to Kant, there are only conceptions of objects, no perceptions. I, on the contrary, say: Objects exist primarily only for perception, and conceptions are always abstractions from this perception. Therefore abstract thinking must be conducted exactly according to the world present in perception, for it is only their relation to this that gives content to conceptions; and we must assume for the conceptions no other _a priori_ determined form than the faculty of reflection in general, the nature of which is the construction of conceptions, _i.e._, of abstract non-perceptible ideas, which constitutes the sole function of the _reason_, as I have shown in the first book. I therefore require that we should reject eleven of the categories, and only retain that of causality, and yet that we should see clearly that its activity is indeed the condition of empirical perception, which accordingly is not merely sensuous but intellectual, and that the object so perceived, the object of experience, is one with the idea, from which there remains nothing to distinguish except the thing in itself.