pressure falls.

The idea of using a free piston was proposed as a remedy; it was thought that a piston connected to a crank in the ordinary manner could not move fast enough to utilize the pressure before it was lost. Many inventors proposed to perform work upon a piston free from any direct connection with the crank or shaft of the engine; the explosion after attaining its maximum pressure expends its force in giving velocity to a piston; the velocity so acquired carries it on against atmospheric pressure until the energy is all absorbed, and a vacuum or deficit of pressure exists in the cylinder instead of an excess of pressure. The return stroke is accomplished by the atmospheric pressure, and the work is now done upon the engine shaft on the return only. The method of connecting on the return stroke while leaving the piston free on the out stroke varies, but in many engines the principle was the same.

Barsante and Matteucci, year 1857, British patent No. 1,625, describe the first engine of this kind, but Messrs. Otto and Langen were the first to successfully overcome all difficulties and make a marketable engine of it. Their patent was dated 1866, No. 434. To distinguish it from Otto's later patents, it may be called the rack and clutch engine.

The economy obtained by this engine was a great advance upon the Lenoir. According to a test by Prof. Tresca, at the Paris Exhibition of 1867, the gas consumed was 44 cubic feet per indicated horse power per hour. According to tests I have made myself in Manchester with a two horse power engine, Otto and Langen's free piston engine consumes 40 cubic feet per I.H.P. per hour. This is less than one-half of the gas used by the Hugon engine for one horse power.

The igniting arrangement is a very good modification of Barnett's lighting cock, which I have explained already, but a slide valve is used instead of a cock.

Other engines carried out the same principle in a different manner, including Gilles' engine, but they were not commercially so successful as the Otto and Langen. Mr. F.H. Wenham's engine was of this type, and was working in England, Mr. Wenham informed me, in 1866, his patent being taken out in 1864.

The great objection to this kind of engine is the irregularity and great noise in working; this was so great as to prevent engines from being made larger than three horse power. The engine, however, did good work, and was largely used from 1866 until the end of 1876, when Mr. Otto produced his famous engine, now known as "The Otto Silent Gas Engine." In this engine great economy is attained without the objectionable free piston by a method proposed first by Burnett, 1838, and also by a Frenchman, Millein, in 1861; this method is compression before ignition. Other inventors also described very clearly the advantages to be expected from compression, but none were able to make it commercially successful till Mr. Otto. To him belongs the great credit of inventing a cycle of operations capable of realizing compression in a simple manner.

Starting from the same point as inventors did to produce the free piston engine—namely, that the more quickly the explosive force is utilized, the less will be the loss, and the greater the power produced from a quantity of burning gas—it is evident that if any method can be discovered to increase the pressure upon the piston without increasing the temperature of the flame causing this pressure, then a great gain will result, and the engine will convert more of the heat given to it into work. This is exactly what is done by compression before ignition. Suppose we take a mixture of gas and air of such proportions as to cause when exploded, or rather ignited (because explosion is too strong a term), a pressure of 45 lb. above atmosphere, or 60 lb. per square inch absolute pressure. Then this mixture, if compressed to half volume before igniting and kept at constant temperature, would give, when ignited, a pressure of 120 lb. total, or 105 lb. above atmosphere, and this without any increase of the temperature of the flame.

The effect of compression is to make a small piston do the work of a large one, and convert more heat into work by lessening the loss of heat through the walls of the cylinder. In addition to this advantage, greater expansions are made possible, and therefore greatly increase economy.

The Otto engine must be so familiar in appearance to all of you, that I need hardly trouble you with details of its external appearance. I shall briefly describe its action. Its strong points and its weak points are alike caused by its cycle. One cylinder and piston suffices to carry out its whole action. Its cycle is: First outstroke, gas and air sucked into the cylinder; first instroke, gas and air compressed into space; second outstroke, impulse due to ignition; second instroke, discharge of exhausted gases. When working at full power, it gets one impulse for every two revolutions; this seems to be a retrograde movement, but, notwithstanding, the advantages obtained are very great. The igniting arrangement is in the main similar to that used on the rack and clutch engine. The engine has been exceedingly successful, and is very economical. The Otto compression engine consumes 21 cubic feet of gas per I.H.P. per hour, and runs with great smoothness.

In 1876 I commenced my work upon gas engines, and very soon concluded that the compression system was the true line to proceed upon. It took me two years to produce a workable engine. My efforts have always been directed toward producing an engine giving at least one impulse every revolution and, if possible, to start without hand labor, just as a steam engine does. My first gas engine was running in 1878, and patented and exhibited in 1879. It was first exhibited at the Kilburn Royal Agricultural Society's show.

This engine was self-starting, gave an ignition at every revolution, and ignited without external flame. It consisted of two cylinders, a motor, and a compressing pump, with a small intermediate reservoir. Suitable valves introduced the mixture of gas and air into the pump, and passed it when compressed from the reservoir to the motor cylinder. The igniting arrangement consisted of a platinum cage firmly fixed in a valve port; this cage was heated in the first instance by a flame of gas and air mixed; it became white hot in a few seconds, and then the engine was started by opening a valve.

The platinum was kept hot by the heat derived from the successive ignitions, and, the engine once started, no further external flame was required. I have here one of these platinum cages which has been in use. Finding this method not well suited for small engines, I produced the engine which is at present in the market under my name.

The cycle is different, and is designed for greater simplicity and the avoidance of back ignitions. It also consists of two cylinders, motor cylinder and the displace or charging cylinder. There is no intermediate reservoir. The displace crank leads the motor by a right angle, and takes into it the mixed charge of gas and air, in some cases taking air alone during the latter part of its stroke.

The motor on the outstroke crosses V-shaped parts about from one-sixth to one-seventh from the out end, the displacer charge now passing into the motor cylinder, displacing the exhaust gases by these ports and filling the cylinder and the space at the end of it with the explosive mixture. The introduction of some air in advance of the charge serves the double purpose of cooling down the exhaust gases and preventing direct contact of the inflammable mixture with flame which may linger in the cylinder from the previous stroke. The instroke of the motor compresses the charge into the conical space at the end of the cylinder, and, when fully compressed, ignition is effected by means of the slide I have upon the table.

This system of ignition has been found very reliable, and capable of acting as often as 400 times per minute, which the Otto ignite is quite incapable of doing. By this cycle the