every one who has travelled in Western China ought to know, the li is not bound by the limitations of the ordinary standard of linear measurement, but is affected to no small extent by the nature of the ground which it purports to measure. Thus, though it may be 3000 li from Lao-wa-t'an to Sui Fu, it does not in the least follow that it is 3000 li from Sui Fu to Lao-wa-t'an, the explanation of this apparent mathematical contradiction being provided by the fact that it is up-hill one way and down-hill the other. Time as well as distance has a direct bearing on the nature of the li, and since it takes longer to travel up-hill from A to B than it does to travel down-hill from B to A, there must obviously be a greater number of li between A and B than there are between B and A. Quod erat demonstrandum.
On New Year's day 1907 I left the temple at Lao-wa-t'an, which had seemed to be a degree less dirty than the inn, and which I had consequently occupied for the night, and crossing the suspension bridge climbed to the summit of a mountain spur over which the track passes, in order to avoid the extra
