Similarly with conceptual time. Here, as there is only one dimension, the difficulty is less obvious, but it is no less real. In conceptual time there is absolutely no means of distinguishing before from after, past from future. For the past means the direction of our memories, the direction qualified by the feeling of “no longer”; the future is the direction of anticipation and purposive adaptation, the direction of “not yet.” And, apart from the reference given by immediate feeling to the purposive life of an individual subject, these directions cannot be discriminated. In short, conceptual time and space are essentially relative, because they are systems of relations which have no meaning apart from qualitative differences in the terms which they relate; while yet again, for the purpose of the conceptual construction which yields them, the terms have to be taken as having no character but that which they possess in right of the relations.[148] One other feature of the space and time construction is sufficiently important to call for special mention. Space and time are commonly thought of as _unities_ of some kind. All spatial positions, it is usually assumed, fall within one system of space-relations; all dates have their place in one all-inclusive time. This character of unity completes the current conception of the spatial and temporal order. Each of those orders is a unity, including all possible spatial or temporal positions; each is an endless, infinite, continuous series of positions, which all are purely relative. There are other peculiarities, especially of the current concepts of space, with which it is not necessary to deal here, as they are of an accidental kind, not arising out of the essential nature of the process by which the conception is constructed. Thus it is probably a current assumption that the number of dimensions in space is three and no more, and again that the Euclidean postulate about parallels is verified by its constitution. As far as perceptual space is concerned, those assumptions depend, I presume, upon empirical verification; there seems to be no reason why they should be made for the conceptual space-order, since it is quite certain that a coherent science of spatial relations can be constructed without recourse to them.[149] § 5. The question now is, whether the whole of this spatial and temporal construction is more than imperfect, and therefore contradictory, appearance. I will first state in a general form the arguments for regarding it as appearance, and then proceed to reinforce this conclusion by dealing with some special difficulties. Finally, I propose to ask whether we can form some positive conception of the higher order of Reality of which the spatial and temporal series are phenomenal.