The problem was to obtain complete displacement by means of solid water without any admixture of free air, the expansion of which as the plunger was withdrawn would reduce the efficiency of the air-pump. To effect this object the air must be prevented from mingling with the water, and must be delivered into the hot-well first. This was accomplished by two means: First, placing the condenser as well as the hot-well above the air-pump chamber, as already stated, and secondly, inclining the bottom of the condenser, so that the water would pass through the inlet valves at the side farthest from, and the air at the side nearest to, the hot-well. Thus the air remained above the water, and as the latter rose it sent the air before it quite to the delivery valves. Pains were taken to avoid any place where air could be trapped, so it was certain that on every stroke the air would be sent through the delivery valves first, mingled air and water, if there were any, next, and the solid water last, insuring perfect displacement. I have a friend who has often asked me, with a manner showing his conviction that the question could not be answered, “How can you know that anything will work until you have tried it?” In this case I _did_ know that this condenser would work at rapid speed before I tried it. The event proved it, and any engineer could have seen that it must have worked. The only question in my mind was as to the necessity of the springs behind the delivery valves. Experiment was needed to settle that question, which it did in short order. At the speed at which the engine ran, the light springs improved the vacuum a full pound, showing that without them these valves did not close promptly. The following important detail must not be overlooked. The rubber disk valves were backed by cast-iron plates, which effectually preserved them from being cut or even marked by the brass gratings. These plates were made with tubes standing in the middle of them, as shown. These tubes afforded long guides on the stems, and a projection of them on the under side held the valves in place without any wear. They also determined the rise of the valves. The chambers, being long and narrow, accommodated three inlet and three outlet valves. The jet of water struck the opposite wall with sufficient force to fill the chamber with spray. When the plans for this condenser were completed, and the Evan Leigh engine had been vindicated, I felt that the success of the high-speed system was assured, and looked forward to a rapidly growing demand for the engines. We got out an illustrated catalogue of sizes, in which I would have put the condenser, but the firm decided that it would be better to wait for that until it should be on the same footing with the engine, as an accomplished fact.