Suppose, for instance, that we have moved the driving wheel forward until the crank stands upright at a right angle to the bore of the cylinder, the resistance to motion of the piston and crosshead has caused the crank pin to bed against the half-brass nearest to the cylinder, all the play or lost motion is then between the other half-brass and the crank pin. When, however, the engine is at work and the piston is driving the crank pin, instead of being driven by it, the lost motion will exist between the crank pin and the half-brass nearest to the cylinder, and the contact will exist between the crank pin and the other brass. The difference in the position of the piston, caused by this lost motion, may be ascertained by moving the piston back and forth until the crank pin contacts with first one and then the other half-brass. It is sometimes attempted to remedy the defect due to this lost motion by moving the crank pin past the dead center and then moving it back to the dead center, so that while on that center the play or lost motion in the connecting rod is taken up. This is all very well so far as the connecting rod and piston is concerned, and will cause them both to stand on their respective dead centers with the lost motion taken up; but, in moving the wheel back to the dead center, we have given full liberty to all the lost motion in the various parts of the valve motion or gear, as already explained, in reference to moving the eccentric upon the shaft. As there are so many more parts in the valve gear, in which lost motion may occur, it is manifestly preferable to take up that play by moving the driving wheel in a continuous direction, rather than to move the latter back to accommodate any play there may be in the connecting rod.