(58.) Whether the simple crank or the sun and planet wheel be used, there still remains a difficulty of a peculiar nature attending the continuance of the rotatory motion. There are two positions in which the engine can give no motion whatever to the crank. These are when the end of the beam, the axle of the crank, and the pivot which joins the connector with the crank, are in the same straight line. This will be easily understood. Suppose the beam, connector, and crank to assume the position represented in fig. 15. If steam urge the piston downwards, the point H and the connector H I will be drawn directly upwards. But it must be very evident that in the present situation of the connector H I, and the lever I K, the force which draws the point I in the direction I K can have no effect whatever in turning I K round the centre K, but will merely exert a pressure on the axle or pivots of the wheel. Again, suppose the crank and connector to be in the position H I K (fig. 16.), the piston being consequently at the bottom of the cylinder. If steam now press the piston _upwards_, the pivot H and the connector H I will be pressed _downwards_, and this pressure will urge the crank I K in the direction I K. It is evident that such a force cannot turn the crank round the centre K, and can be attended with no other effect than a pressure on the axle or pivots of the wheel. Hence in these two positions, the engine can have no effect whatever in turning the crank. What, then, it may be asked, extricates the machine from this mechanical dilemma in which it is placed twice in every revolution, on arriving at those positions in which the crank escapes the influence of the power? There is a tendency in bodies, when once put in motion, to continue that motion until stopped by some opposing force, and this tendency carries the crank out of those two critical situations. The velocity which is given to it, while it is under the influence of the impelling force of the beam, is retained in a sufficient degree to carry it through that situation in which it is deserted by this impelling force. Although the rotatory motion intended to be produced by the crank is, therefore, not absolutely destroyed by this circumstance, yet it is rendered extremely irregular, since, in passing through the two positions already described, where the machine loses its power over the crank, the motion will be very slow, and, in the positions of the crank most remote from these, where the power of the beam upon it is greatest, the motion will be very quick. As the crank revolves from each of those positions where the power of the machine over it is greatest, to where that power is altogether lost, it is continually diminished, so that, in fact, the crank is driven by a varying power, and therefore produces a varying motion. This will be easily understood by considering the successive positions of the crank and connector represented in fig. 16.