RELATIONS OF TIME AND SPACE. 141 



experience that within the same portion of duration, as far as can be 

 discerned, there is room for the same or simihir events to occur in 

 the same ^vay. It is the result also of all experience, so far as we 

 can trace the matter, that one portion of duration is like another, so 

 that no portion of it goes on fibster than any other. Now, it is the 

 uniform continuous flow of duration, thus parcelled out, that consti- 

 tutes absolute time. Relative time is a portion of duration whose 

 limits are marked by events. You have something to mark when 

 that duration begins and when it ends. Absolute time will flow on 

 whether you mark it or not. When that portion of time is marked 

 by events, so that you know where certain periods of time begin 

 and end, it becomes relative time. As then we must speak of the 

 limit when a portion begins, not when it ends, it is desirable that we 

 should distinctly indicate what we mean by limit, whether of space 

 or duration. 



Now with regard to the limit of anything that occupies space. The 

 outside of this table before me is no part of the table; it is only pre- 

 cisely where the table ends that outer space begins. The outside is 

 not someivhat in the same sense with the table itself, but onl}^ some- 

 where — where the table ends, and the outer space begins. But this 

 is still sometvJiat in its own sense. It has breadth, it has length and 

 superficial extent, such as may be measured by a square foot. But 

 the outer edge is not somewhat even in the sense that it covers some- 

 thing; it merely penetrates space, instead of dividing it. The very 

 end of the outer edge is not somewhat even in the sense of a line, for 

 although the outer edge was long, penetrating space, the very end of 

 it is only someiohere. Thus it is with the centre of a sphere. It is 

 not somewhat. When we depart from it at all we forsake the central 

 position we occupied. But you may ask is it not somewhere? Oh, 

 yes; precisely at the middle; at the same distance from ihe surface 

 on all sides. So also with the midnight, Avith v/hich Tuesday shall 

 end and Wednesday begin. It is not somewhat in duration. It does 

 not last at all. When midnight arrives, that instant the following day 

 will be here. It is not someivliat in duration, but only someivhere; it 

 is at the precise time when one day ends and another begins that, 

 if I may be permitted to coin a word for the occasion, it is somewhen. 

 It lasts not at all, and therefore is only a point of time, that is an 

 instant. There is a distinction between an instant and a moment, 

 for a moment is a real portion of duration — a small but indefinite por- 

 tion — while an instant is no portion of duration whatever. 



The natural measure of time, Lord Karaes tells us, is the succession 

 of our thoughts. When the varied incidents of a pleasant journey 

 pass before the mind in review, the interval thus spent appears to 

 have been a long one. But if during a tedious journey, or a fit of the 

 toothache, or, to come still nearer home, a dull lecture, we find time 

 to be long, it will certainly not be long in review. So, then, time 

 may appear long or short, according to circumstances, and it is per- 

 fectly evident that this rule of Lord Kames cannot answer as a good 

 measure of time. It has been also supposed that the insects which 



