the end of the stroke with a certain amount of unexpended momentum stored up in it, if this energy is expended by producing pressure on the crank, then it will not only be a waste of energy but a double waste by retarding the motion of the crank. If, however, this energy can be absorbed by compressing steam which will fill the clearance spaces, it will not only prevent the retarding effect referred to, but the energy in the piston and other parts will be converted into steam pressure, which will be given out in useful work during the next stroke. It would, of course, be impossible to arrest the motion of the piston instantly, and therefore its momentum is gradually absorbed from the time compression begins until it reaches the end of the stroke. As the energy of a moving body is equal to its weight multiplied by the square of its speed, it is obvious that to overcome this a different amount of compression would be required for each speed, and also that it must be adjusted to the weight of the moving parts. Such exact adaptation is not practicable on locomotives, nor does the link motion enable us to alter the amount of compression with so much exactness: but the explanation shows the value of increasing the amount of compression with the speed, which fortunately the peculiarities of the shifting-link motion enable us to do without difficulty. [Illustration: _Fig. 139._] QUESTION 224. _What cause produces the form of diagram represented by Fig. 139?_ _Answer._ It is produced by excessive compression, which causes the pressure in the cylinder to rise above boiler pressure before pre-admission begins. As soon as the port is opened, part of the steam in the cylinder flows back into the steam-chest, and thus the pressure is reduced, as shown by the diagram. QUESTION 225. _How can we determine whether the steam is distributed in the cylinders to the best advantage, and how can we discover the fault, if there is one, in the link motion?_ _Answer._ The indicator will show the action of the steam in the cylinder, and motion-curves drawn with the instrument described in answer to Question 192 will show the exact movement of the valve. By comparing the indicator diagram with the motion-curves, the one will show the defects in the other.[62] [62] See description of Richards’ Improved Steam Engine Indicator, with directions for its use, by Charles T. Porter, London. QUESTION 226. _To what extent can the movement of the valve be modified by alterations in the proportions of the link motion?_