He showed that at high speeds waves of two different characters are produced: the one class largest at the bow, which separate from the ship, decreasing in successive undulations without afterward affecting her progress; the other, those in which the wave-crests are at right angles to the ship’s course, and the positions of these crests have a very telling effect upon the resistance. As the ship’s speed is increased the spaces between the crests of these lengthen in unison with the speed, and it has been shown that when the speed is such that a wave-crest would be at the middle point of the after body (or quarter) the wave-making resistance is least, and that it is greatest when the hollow appears at this point. A ship must therefore be of a length that depends largely upon the length of wave which at a high speed she will tend to produce in order that she may be driven at such a speed without an expenditure of power disproportionate to the effect produced. This length, if very high speeds are desired, is best wholly taken up in fining the entrance and run, leaving no parallelism of middle body, and broadening and deepening the ship to keep the necessary displacement. The wave-action at several speeds is well shown in the illustrations, which are from instantaneous photographs, showing the Chilian cruiser Esmeralda at her full speed of 18 knots, when on her trial off Newcastle-upon-Tyne, the Giovanni Bausan, of the Italian navy (almost a sister ship to the Esmeralda), at a moderate speed, and H.M.S. Impérieuse, at about 17-1/4 knots. [See illustration, p. 64.] The following are the principal details of the Esmeralda and Impérieuse: Displacement. Length. Beam. Draught. Horse-power. Esmeralda 3,000 270 42 18.3 6,500 Impérieuse 7,390 315 62 26.0 10,180 The eddy-making resistance is greater or less, of course, as the form is blunted or finer, and there is less resistance with a blunt bow and finely formed after-body than were the two reversed. Our practical towing friends will be glad to know that Mr. Froude substantiates their oft-reiterated assertion that a log tows more easily butt-end foremost. In the Merkara, a merchant ship built by Mr. Denny, of 3,980 tons, 360 feet length, 37.2 feet breadth, and 16.25 feet draught, this resistance is, at all speeds, about eight per cent. of the surface friction, which at the maximum speed of thirteen knots, at which she was intended to be run, still formed nearly eighty per cent. of the whole resistance.