speeds an engine, or any other machine, constructed without a fly-wheel must be provided with some means for continuing the power or carrying it over what would otherwise be dead centres. Multiple cylinders and rotary engines are made to serve this purpose. The commonly accepted idea that a cycle for racing purposes upon a smooth road is a certain guide as to the requirements under other conditions is therefore hardly justifiable. For best results the form of mechanism used as the connecting link between the legs of nature and the wheel of mechanics must be determined, or at least be modified, by the conditions under which we intend to work. This problem is not at all confined to the art of cycling, it appears in many departments of mechanics. The same question has been mooted in respect to sewing-machines, and non-dead-centre attachments have been made and used upon them, but naturally the demand was not urgent, as this machine comes within the realm of high-speed devices with fly-wheel and evenly-running resistance. In scroll sawing by foot-power and in portable forges, non-dead-centre clutches are used with great effect. Hence our general mechanical experience makes it safe to say that such modes of continuous application of power have valuable uses applicable to this problem. It is not attempted to set up a definite unequivocal comparison or dictum in this matter as applied to cycles, for it is the desire of the writer and his right to make conclusions comparable only to the proofs recognized in practice, which in this case, in the cycle art, appear to be in favor of the crank machine. However, the writer’s opinion, based upon his theory and individual experience, is that we have more to fear from the weight, complication, and friction of parts in the lever and clutch than from the inherent principle of transmitting power upon which it works, and that some non-dead-centre device will finally prevail in the best all-around road cycles, if it can be relieved of purely mechanical objections which somehow seem to be naturally coupled with it. If the writer’s conclusion in this respect is tenable, the induction would follow that such a system, or connecting link, forms the most economical mode of applying power. The body can stand a steady, even pull upon its energy better than uneven intermitting jerks, and I feel sure new riders who have not acquired the ankle-action on the crank cycle will agree in this. This theory will apply to hill-climbing, in which lever and clutch machines have made so enviable a reputation. The rider has in clutch machines an even, steady resistance during the entire downward thrust, and he does not have to get all his power doubled up into a few inches of motion.