_Otto gas engine_ (Fig. 7).—The principle on which this engine is based is known as the Otto cycle, named after Dr. Otto, but first suggested by Beau de Rochas. Since the patents have expired numerous copies and imitations have been brought out, but very few surpass or even equal some of the earlier types. [Illustration: FIG. 7.—Otto Gas Engine.] The explanation of the working of the Otto motor, which we are about to give, will save us from returning to it in the descriptions of analogous types brought out after this famous system. The cylinder is continued in a backward direction so as to form a compression chamber, into which the mixture of gas and air is drawn during the forward stroke of the piston. The mixture is compressed during the return stroke in this chamber, the pressure rising at the end of the stroke to about 3 or 4 atmospheres. At this point in the cycle a flame is brought into contact with the compressed gases and they explode. This explosion raises the temperature of the gases to 1500° C., and drives forward the piston under a pressure of about 150 lbs. to the square inch. During the second return stroke, corresponding to the latter half of the second revolution of the crank, the piston drives out the products of the combustion into the air under a pressure of about an atmosphere. The heating of the cylinder is avoided by keeping water automatically circulating through a jacket surrounding it. This is necessary, because if the cylinder walls became heated, the oil upon them would become decomposed, and lose its lubricating properties. Dr. Otto paid special attention to the efficiency of the engine, and in order to increase it, he diluted the air and gas drawn into the cylinder with a portion of the gases already burnt in the previous stroke. Consequently the explosion at the beginning of the stroke is less violent, and the gases continue burning while the piston moves forward. The cause of this slow combustion has been wrongly attributed to stratification of superimposed layers of gas and air, but it is probably due to the action of the cylinder walls.