But when steam navigation had become an accomplished fact and steam-propelled vessels were able to undertake ocean voyages, there was the less excuse for the absolute rejection of steam-driven ships for war purposes. The objections raised, and they were certainly serious, were that the paddle-boxes were large and increased the size of the target at which the enemy could aim, that one or other of the paddle-wheels could easily be disabled by a hostile shot, in which case the steamer would be almost helpless, and would be entirely so if both wheels were disabled; that the engines and boilers, being partly above the water-line, were peculiarly vulnerable, and that the only vessels of a size capable of being propelled by paddle engines were too small to be of much fighting value. Long before steam engines were adopted in any form in the navy, numerous experiments had been made in the mechanical propulsion of warships, and some extraordinary methods were brought forward. The remarkable feature of nearly all these experiments is that they would propel a vessel; but the inventors could not be taught, except by the bitter lesson of experimental failure, that an appliance which might attain a certain measure of success with a small boat or a model might be incapable of developing sufficient power to propel a larger vessel. Several inventors, both in this country and America, tried what they could do with oscillating paddles at the stern. Manual power was tried on the frigate _Doncaster_ at Gibraltar, in 1802, to drive a “perpetual sculling machine,” invented by a man named Shorter, and a speed of something under two miles an hour in still water was obtained. Certain of the early experiments in America had an important influence upon the development of the steam warship, though the proof of that influence did not become evident for many years. Colonel John Stevens, who had a small, screw-propelled steamboat on the Hudson, in 1804, and a twin-screw steamer in the following year, designed, in 1813, an iron-clad ship which fully embodied the _Monitor_ type, and was the first ironclad ever worked out for construction. It was Stevens who sent the first steamer on a sea voyage; his vessel, the _Phœnix_, being shut out of the waters of New York by the monopoly which Fulton and his friends had secured, went round to the Delaware by sea in June, 1809, experiencing a gale on the way, which compelled her to seek shelter in Barnegat Inlet. Of Fulton’s early experiments and failures in the matter of submarines and torpedoes it is unnecessary to speak. In 1814, some years after his return to the United States from Europe, where he had been impartially offering his services to the French and then the British, he submitted to the American coast and harbour defence committee plans for a steam warship which was to carry a large number of guns.