In 1737, Jonathan Hulls issued a pamphlet proposing to construct a boat to be moved by steam power, for the purpose of towing vessels out of harbors against tide and winds. In his plan the paddle-wheel was used, and was secured to a frame placed far out over the stern of the boat. It was given this position by the inventor because water fowls propelled themselves by pushing their feet behind them. In 1787, Mr. James Rumsey, of Shepherdstown, Virginia, constructed and navigated the first steamboat in actual use. His boat was eighty feet in length, and was propelled by means of a vertical pump in the middle of the vessel, by which the water was drawn in at the bow and expelled at the stern through a horizontal trough in her hull. The engine weighed about one third of a ton, and the boat had a capacity of about three tons burthen. When thus laden, a speed of about four miles an hour could be attained. The boiler held only five gallons of water, and needed but a pint at a time. Rumsey went to England to exhibit his plan on the Thames, and died there in 1793. About the same time the Marquis de Joffrey launched a steamer one hundred feet long on the Loire, at Lyons, using paddles revolving on an endless chain, but only to find his experiment a failure. In December, 1786, John Fitch published the following account of a steamer with which he had made several experiments on the Delaware, at Philadelphia, and which came nearer to success than any thing that had at that time been invented: "The cylinder is to be horizontal, and the steam to work with equal force at each end. The mode by which we obtain what I term a vacuum is, it is believed, entirely new, as is also the method of letting the water into it, and throwing it off against the atmosphere without any friction. It is expected that the cylinder, which is of twelve inches diameter, will move a clear force of eleven or twelve cwt. after the frictions are deducted: this force is to be directed against a wheel of eighteen inches diameter. The piston moves about three feet, and each vibration of it gives the axis about forty revolutions. Each revolution of the axis moves twelve oars or paddles five and a half feet: they work perpendicularly, and are represented by the strokes of a paddle of a canoe. As six of the paddles are raised from the water, six more are entered, and the two sets of paddles make their strokes of about eleven feet in each revolution. The crank of the axis acts upon the paddles about one-third of their length from their lower ends, on which part of the oar the whole force of the axis is applied. The engine is placed in the bottom of the boat, about one-third from the stern, and both the action and reaction turn the wheel the same way." Fitch was unfortunate in his affairs, and became so disheartened that he ceased to attempt to improve his invention, and finally committed suicide by drowning himself in the Alleghany River at Pittsburgh.