We have already noticed that the most which Newcomen could get out of his engine was an up-and-down movement, which was all very well for the purpose for which it was intended, namely, pumping up water, but before it was applicable for propelling a ship the power had to be adapted to give a rotary motion. The accompanying illustration, which is taken from Hulls’ specification for his patent, and reproduced in the booklet mentioned above, will afford some idea of his proposal. In the lower half of the picture the “tow-boat” is seen in imagination hauling an eighteenth century full-rigged ship, a performance which in actual truth she never achieved. There is, in fact, some doubt as to whether Hulls ever did put the idea to a practical test. Admiral Preble, a distinguished American Naval officer, in his “Chronological History of the Origin and Development of Steam Navigation,” published in Philadelphia in 1883, a volume which contains a vast amount of interesting detail up to that date, says that Hulls did not produce a satisfactory experiment. Scott Russell, one of the greatest authorities on such matters in the nineteenth century, affirmed that Hulls did carry out his theory in definite shape, and the recent “Dictionary of National Biography” also states that at any rate he experimented with a vessel on the River Avon in the neighbourhood of Evesham in 1737. One thing is certain, that whatever merits the proposition might have had in certain respects, it was, commercially, a complete failure. On the other hand, in enunciating a method of converting the rectilineal motion of the piston-rod into a rotary movement Hulls undoubtedly showed the direction in which others were to follow. In the upper half of the illustration of Hulls’ drawing, beginning at the bottom right-hand corner, we see the details of his “machine.” _P_ is the pipe which comes from the furnace and brings the steam to _Q_, the cylinder in which the steam was also condensed. (This last remark is important to bear in mind, as we shall see later to what extent this feature was modified.) The point marked _R_ is the valve which enables the steam to be cut off from entering the cylinder whilst that amount of steam which has already been allowed to go in is being condensed. The other small pipe _S_ conveys the cooling water which condenses the steam in the cylinder, and _T_ is the cock which lets in the condensing water after the cylinder is full of steam and the valve is shut. _U_ is the rope which is fixed to the piston that slides up and down the cylinder, and this is the same rope that goes round the wheel _D_ in the machine shown in the larger illustration.