We now proceed to compare the different modes which have been devised to transmit power from the rider to revolve the wheel; of these there are two principal classes, the simple crank and the lever and clutch. These devices or connecting links relate to the motion of the legs as well as to the power transmitted through them. It is not necessary to treat of the horizontal motion of the limbs, as it is of little consequence provided the rider remains substantially over the work. Power is applied mainly through the vertical resultant, and the consequent fatigue is the effect of the amount of energy given out in a vertical direction. Crank riders acquire a horizontal power, or resultant force, by what we call ankle-motion, which has, to quite an appreciable extent, overcome the most serious inherent defect of the crank device; without this force the dead centre appertaining to the crank, in which the vertical resultant has no power to turn the wheel, would have made it a prey to the champions of other contrivances. The above remarks in regard to horizontal motion and resultant force apply equally well if the rider is not over the work, except in that the phraseology would be different. A man in straightening out his leg can apply power in a certain direction or in a certain line; now, if he is not over the work, this will not be a vertical line; hence the term horizontal motion would have to be called motion at right angles to the line of transmission of power. The importance of the dead centre is too great to be passed over without some further discussion. It would be a source of great satisfaction if a general conclusion could be reached in this crank versus lever and clutch controversy, but aside from the difficulty of drawing our conclusion there is a lack of a specific hypothesis in regard to an important element of the problem,—to wit, that as to the nature of the road and other resistance and consequent speed attainable or usually desirable. There is little doubt but that, so far as present developments show, the crank machine has excelled upon a smooth road and at high speed; yet this very fact leads us to suspect that perhaps for rough roads and at slow speed it might be objectionable, for it is easy to see that all questions of dead centres would eliminate themselves at high speed. Taking a steam-engine, of the crank and pitman type, for example, there is no trouble so long as speed is kept up, but it is well known that a certain velocity must be maintained or the crank will stop at the dead centre, even when provided with a heavy fly-wheel. Now, in a bicycle there is practically no fly-wheel at all, and, to pursue the comparison still further, we know that if the fly-wheel of an engine were removed great trouble would ensue; still it might be possible to keep running if the speed were great enough. It is evident, from common observation, that for intermitting slow and high