Mills occurred here and there. Some of these had the waterwheel simply built on the river; others had it so arranged as to allow the shaft to be raised or lowered to suit the varying height of water in floods and droughts. Others had it floating on barges. Others, again, had a half weir built diagonally across part of the river; and it was important to look carefully at this wall so as to see on which side it ought to be kept in selecting the best course. In a few cases there was another construction; two half weirs, converged gradually towards the middle of the river, forming a letter V, with its sharp end turned _up_ the stream, and leaving a narrow opening there, through which a torrent flowed, with rough waves dancing merrily in the pool below. I had to "shoot" several of these, and at other times to get out and lower the boat down them, in the manner explained before. On one occasion I was in an unaccountably careless fit, and instead of first examining the depth of the water on the edge of the little fall, I resolved to go straight at it and take my chance. It must be stated that while a depth of three inches is enough for the canoe to float in when all its length is in the water, the same depth will by no means suffice at the upper edge of a fall. For when the boat arrives there the fore part, say six or seven feet of it, projects for a time over the fall and out of the water, and is merely in the air, without support, so that the centre of the keel will sink at least six or seven inches; and if there be not more water than this the keel catches the crest of the weir, and the boat will then stop, and perhaps swing round, after which it must fall over sideways, unless considerable dexterity is used in the management. Although a case of this sort had occurred to me before, I got again into the same predicament, which was made far more puzzling as the fore end of the boat went under a rock at the bottom of the fall, and thus the canoe hung upon the edge, and would go neither one way nor another.[XXIV.] It would also have been very difficult to get out of the boat in this position; for to jump feet foremost would have broken the boat--to plunge in head first might have broken my head on the rocks below. [XXIV.] This adventure was the result of temporary carelessness, while that at the rapids was the result of impatience, for the passage of these latter could probably have been effected without encountering the central wave had an hour or two been spent in examining the place. Let not any tourist, then, be deterred from a paddle on the Reuss, which is a perfectly suitable river, with no unavoidable dangers. [Illustration: "Fixed on the fall."] The canoe was much wrenched in my struggles, which ended, however, by man and boat tumbling down sideways, and, marvellous to say, quite safely to the bottom.