nation either by the curvature of the banks, or by some physical irregularity of the channel." It was determined by actual experiment that the greatest velocity of current is at the surface and the least at the bottom, and that the increase of velocity "is in the simple ratio of the distance from the bottom." This decides that the mean velocity of a stream is to be found at half its depth. A result perfectly consistent with the previously expressed law that surface velocity is proportionate to depth, it is in fact a corollary, and one that was verified by experiment.
For determining the velocity of a stream, M. Révy employed an instrument called the current-metre. This consisted of a propeller-like wheel attached to a long axle and made to turn with rapidity when immersed in a running stream. The wheel is fastened to a kind of rudder which keeps it always against the stream. The upper end of the axle moves a set of cogs which turn a couple of indices upon a dial-plate. M. Révy is not the inventor of the instrument, but he improved it so as to greatly increase its utility. He first ascertained the correct reading of the metre by moving it through still water, and next made an extensible addition to the axle, so that the instrument could be adjusted to any depth.
An interesting verification of M. Révy's discoveries with regard to the velocity of currents was unexpectedly made. "In two successive measurements of the current of the La Plata, he found that there was a decrease of surface velocity and an accompanying increase of lower velocity. This at first seemed to contradict his previously-obtained results, but on further examination it was discovered that the depth at those points was sixteen inches greater. The water was banked up by its own mass so that the surface fall was less than at the point of previous measurements. "In obedience to known hydraulic law, this decreased surface-inclination was indicated by decreased velocity of surface-current. But the power of the whole moving mass was greater in proportion to its depth. And thus, while to the superficial gaze the velocity was less, the mean velocity was greater; and the river swept with more resistless energy over its bed. This luminous observation is the opening of an entirely new chapter in hydraulic science. It is, in fact, a case of the law of relation of speed to depth; but it is one that could scarcely have been arrived at by theory; although, now that it is experimentally ascertained, its theoretic reason is ascertainable."
These results, of course, require further verification elsewhere, but they cannot fail to be of the highest value both to experimental and practical science, for they will at any rate furnish method, and give new impetus to hydraulic investigation. Should they be verified, and they probably will be, it will follow that wherever water-power is desired, as in mill-races, or abundance of water-supply, as in aqueducts, depth of way should be secured at the expense of breadth.
