Figs. 5, 6, and 7 show why, in those days of rope cables, a ship was always breaking the cable where it bent at right angles just outside the hawse-hole. All the strain was on the outer strands of the rope at a bc, Fig. 7, and as they broke the others followed one by one. His remedy for this was to have a large wheel or pulley in the hawse-hoi

Figs. 8 and 9 show how a vessel with a leak at first fills very rapidly, so that the crew, finding they cannot gain on the water with the pumps, take to their boats. But if they would remain they would find after a while that the quantity entering would be less as the surfaces without and within became more nearly equal, and that the pumps would, now be able to prevent it from rising higher. The water would also begin to reach light wooden work, empty chests, and water-casks, which would give buoyancy, and thus the ship could be kept afloat longer than the crew at first expected. In this connection he calls attention to the Chinese method of watertight compartments which Mr. Le Roy had already adopted in his boat on the Seine.

Fig. 12 is intended to show the loss of power in a paddle-wheel because the stroke from A to B is downward and from D to X upward, and the only effective stroke is from В to D. A better method of propulsion, he thinks, is by pumping water out through the stern, as shown in Figs. 13 and 14.

Figs. 15, 16, 17, 18, 19, 20, 21, and 22 illustrate methods of making floating sea anchors by which to lay a vessel to in a gale. Fig. 24 shows how a heavy boat may be drawn ashore by bending the rope from