necessarily forced to admit at once that space must be infinite too; for had space any boundary, then, since the molecules do not admit of being crowded together beyond a certain extent, it would be impossible that they could exist in infinite abundance. Adopting the sound principle that we need not assume more than is necessary to explain the phenomena actually presented by our consciousness, it seems to me to be clear that the number of molecules of matter in the universe must be finite. The row of figures which would express that number, whatever it may be, is the most remarkable descriptive constant which the universe possesses. It matters not for our present argument what may be the range of figures by which this number can be expressed. It may not be too large to be written even on the thumb-nail by the compendious method of notation now in general use.
Let us next see whether we can learn anything as to the extent of space itself. It is apparent that we seem to be in the presence of about equal difficulties whether we attempt to think of space as finite or as infinite; for, imagine that you go up in a straight line into the sky, and on, and on, and on, in thought for millions of miles, it would seem that the journey ought to be endless; for, supposing that you try to conceive a boundary, the imagination will equally suggest that there is something on the other side of that boundary from which you can commence again. It appears almost equally impossible to suppose that the journey could be carried on for ever as to suppose that it could ever be brought to an end. It was, however, long ago shown by Kant that space was rather to be regarded as a form in which the human mind was compelled to regard objects than as a self-existing fact
