In 1787, Mr. Patrick Miller, of Dalwinston, Scotland, designed a double vessel, propelled by a wheel placed in the stern between the two keels. This boat is said to have been very successful, but it was very small, the cylinder being only four inches in diameter. In 1789, Mr. Miller produced a larger vessel on the same plan, which made seven miles per hour in the still water of the Forth and Clyde Canal, but it proved too weak for its machinery, which had to be taken out. It was in the face of these failures that Fulton applied himself to the task of designing a successful steamboat. During his residence in Paris he had made the acquaintance of Mr. Robert R. Livingston, then the American minister in France, who had previously been connected with some unsuccessful steamboat experiments at home. Mr. Livingston was delighted to find a man of Fulton's mechanical genius so well satisfied of the practicability of steam navigation, and joined heartily with him in his efforts to prove his theories by experiments. Several small working models made by Fulton convinced Mr. Livingston that the former had discovered and had overcome the cause of the failure of the experiments of other inventors, and it was finally agreed between them to build a large boat for trial on the Seine. This experimental steamer was furnished with paddle wheels, and was completed and launched early in the spring of 1803. On the very morning appointed for the trial, Fulton was aroused from his sleep by a messenger from the boat, who rushed into his chamber, pale and breathless, exclaiming, "Oh, sir, the boat has broken in pieces and gone to the bottom!" Hastily dressing and hurrying to the spot, he found that the weight of the machinery had broken the boat in half and carried the whole structure to the bottom of the river. He at once set to work to raise the machinery, devoting twenty-four hours, without resting or eating, to the undertaking, and succeeded in doing so, but inflicted upon his constitution a strain from which he never entirely recovered. The machinery was very slightly damaged, but it was necessary to rebuild the boat entirely. This was accomplished by July of the same year, and the boat was tried in August with triumphant success, in the presence of the French National Institute and a vast crowd of the citizens of Paris.