By describing a circle to represent the movement of the crank, and at a distance from it drawing to scale an arc of a circle to represent the movement of the slay, we can prove the foregoing remarks. In Fig. 24 the circle referred to is shown, and also the line A E representing the stroke of the slay. By observing the position of the ends of the connecting rod E^1, when in contact with the circle at the back centre, and also when the slay is at front centre, we find that it has travelled over from E^1 to A^1, which is more than half of the circle. Then assuming that the crank-shaft runs at one even speed, we would say that the slay takes longer time going forward than coming back. Again, by measuring the movement of the slay from A to B at the front, and from D to E at the back, these parts are found of equal length, but by extending from these points our crank-arm of equal length in every case, we find that to move the slay from B to A and back occupies a space on the circle from B to B^1, while to move the slay from D to E and back occupies a space from D^1 to D^1 obviously greater, from which fact we assume that the slay occupies a longer time at the back than at the front, an arrangement purposely contrived so as to get theoretically a longer dwell. The difference between the arcs D^1 D^1 and B B^1 is approximately 15°. Calculating at 180 picks per minute, we get the difference between the time of dwell at front and back to a seventy-secondth part of a second--to a practical mind not a very great consideration. [Illustration: FIG. 24.] The shuttle race is made of hard wood laid on the beam, and in addition to the depression at the centre is also rather wider at the middle of the race--at that point bulging out slightly against the reed, which is kept in contact with it by the slay cap at the top and the reed case at the bottom. Care must be taken to keep the ends of the reed from projecting in front of backboard, or the shuttle will be thrown out. LOOSE AND FAST REEDS. To obtain a firm blow at the cloth the reed is made to strike it at right angles, for if the angle were more obtuse or acute a loss of force would be entailed by a consequent downward or upward stroke imparted to the cloth, the line of action not being in the same plane as the line of reaction. With regard to the firmness of the beating up, we have to consider the merits of fast and loose reed looms.