Stevens also invented what he called a "rotary engine" which was really an engine constructed on the same principle as the modern turbine engine. It was a small affair which he placed in a skiff, and used for turning the screw-propeller of a boat which was able to travel at a rate of three or four miles an hour on the North River, during the fall of 1802. But Stevens found so much difficulty in packing the blades of this engine without causing too much friction that he finally abandoned it for the more common type of reciprocating engine. But if this little steamboat had its defects, it nevertheless contained the germs of two great features of steam navigation--the screw propeller and the turbine engine, the advantage of the first of which was not recognized for nearly half a century, and the other not until almost a full century later. In 1804 Stevens produced another propeller steamboat, this one using the ordinary type of reciprocating engine, and being notable for having twin screws of a pattern practically identical with the screws now in use. This boat was able to steam at a rate of four miles an hour on many occasions, and at times almost double this rate, according to some observers. The engines of this boat are still in existence, and on several occasions since 1804 have been placed in hulls corresponding as nearly as possible to the original, and have demonstrated that they could force the boat through the water at six or eight miles an hour. These engines in a modern hull were exhibited at the Columbian Exposition at Chicago, in 1893. They supply irrefutable evidence that the practical steamboat had been invented at least three years before Fulton's historic voyage in 1807. Yet no one questions that it was Fulton's, not Stevens', invention that inaugurated steam navigation. Just why this was so is a little difficult to comprehend at this time, unless it was that Stevens' boat was such a small affair that it did not attract the attention it deserved, as did Fulton's larger boat. And yet we should not be guided too much by retrospective judgment. The significant fact remains that Stevens himself did not have entire confidence in his boat, or in the principle of his screw propeller, as is shown by the fact that three years later, while Fulton was building the _Clermont_, Stevens was also constructing a steamboat, not along the lines of his previous inventions, but as a paddle-wheel boat. This leaves little room for doubt that Stevens had not full confidence in the propeller; and when an inventor himself mistrusts his own device, there is little likelihood that anyone else will supply the necessary confidence. This may account for the fact that Stevens found difficulty in securing financial backing for his enterprise; and when such backing was found it was for the construction of the paddle-wheel boat, which was finished a few months after Fulton's boat had solved the problem of steam navigation. FULTON AND THE CLERMONT