lake. Besides, the two points being possibly distant from one another, the effect, or the eddy, appears before the cause, the obstacle (Fig. 1, ii). It is in this way, by the appearance of the surface, that the raftsman going down a river can judge, from the variations of speed, of the depth of the stream, and the size and position of reefs hidden under the water, frequently a considerable distance below the point where he is. Is not this the supreme end of art—to cause to be foretold by outer forms what is going on in the domain of invisible things, and to divine the reality without laying it bare? Water lends itself eminently to this end. It obeys mechanical laws, not as a machine which exposes them bluntly and fatally, but with a variety of suggestions and a lightness that leave the field clear for the imagination.
Let us, in our experimental canal, reduce the dam to a single obstacle in the middle of the stream. The eddy-wave, instead of being straight, bends around on either side, and takes the form of a parabola with a Λ more or less open according to the velocity of the stream. Moreover, the branches of the parabola, turned back by the side-walls of the canal, if it is not too broad, take a figure below the obstacle in which the first traces of a lozenge appear (Fig. 2, iii). Lastly, let us place the obstacles on the sides
Fig. 2.—Forms of Eddy-waves.
of the canal. We find that the waves are turned toward the middle of the stream, and that by intercrossing with those coming from the opposite bank, and by their own return, a system of lozenges is produced on the surface of the water (Fig. 2, iv, and Figs. 3 and 4).
In this we see the pre-eminently typical phenomenon of running water. Every inequality of the shore, whether promontory or bay, plant or pebble, stake or bridge pier, is the starting-point for centripetal lines which go on in graceful undulations to lose themselves in the stream, or, meeting with others, to form qua-
