It will thus be seen that more than half of the heat is communicated to the water in the jacket. Now, this is the opposite of the steam-engine, where the jacket is used to transmit heat _to_ the cylinder, and not _from_ it. This cooling is rendered necessary, because without it the oil would be carbonized, and lubrication of the cylinder rendered impossible. Indeed, a similar difficulty has occurred with all hot-air engines, and is, I think, the reason they have not been more generally adopted. I felt this so strongly that, for some time after the introduction of the gas-engine, I was very cautious in recommending those who consulted me to adopt it. I was afraid that the wear and tear would be excessive. I have, however, for some time past been thoroughly satisfied that this fear was needless; as I am satisfied that a well-made gas-engine is as durable as a steam-engine, and the parts subject to wear can be replaced at moderate cost. We have no boiler, no feed pump, no stuffing-boxes to attend to--no water-gauges, pressure-gauges, safety-valve, or throttle-valve to be looked after; the governor is of a very simple construction; and the slide-valves may be removed and replaced in a few minutes. An occasional cleaning out of the cylinder at considerable intervals is all the supervision that the engine requires. The very large percentage of heat absorbed by the water-jacket should point out to the ingenuity of inventors the first problem to be attacked, viz., how to save this heat without wasting the lubricant or making it inoperative; and in the solution of this problem, I look for the most important improvement to be expected in the engine. The most obvious contrivance would be some sort of intercepting shield, which would save the walls of the cylinder and the rings of the piston from the heat of the ignited gases. I have just learned that something of the kind is under trial. Another solution may possibly be found in the employment of a fluid piston; but here we are placed in a dilemma between the liquids that are decomposed and the metals that are oxidized at high temperatures. Next, the loss by radiation--15 per cent.--seems large; but this is to be attributed to the fact that the inside surface of the cylinder is at each inward stroke exposed to the atmosphere--an influence which contributes to the cooling necessary for lubrication. The remaining 15 per cent., which is carried away by the exhaust, is small compared with the proportion passing away with the exhaust steam of a high-pressure or the water of a condensing engine. As the water in the jacket can be safely raised to 212 deg. Fahr., the whole of the jacket heat can be utilized where hot water is required for other purposes; and this, with the exhaust gases, has been used for drying and heating purposes.