It has been seen that gunpowder was the explosive used to produce a vacuum in Huyghens' engine, and that it was abandoned in favor of gas by Buren in 1823. The reason of departure is very obvious: a gunpowder explosion and a gaseous explosion differ in very important practical points. Gunpowder being a solid substance is capable of being packed into a very small space; the gas evolved by its decomposition is so great in volume that, even in the absence of any evolution of heat, a very high pressure would result. One cubic inch of gunpowder confined in a space of one cubic inch would cause a pressure by the gas it contains alone of 15,000 lb. per square inch; if the heating effect be allowed for, pressures of four times that amount, or 60,000 lb. per square inch, are easily accounted for. These pressures are far too high for use in any engine, and the bare possibility of getting such pressure by accident put gunpowder quite outside the purpose of the engineer, quite apart from any question of comparative cost. In a proper mixture of inflammable gas and air is found an exceedingly safe explosive, perfectly manageable and quite incapable of producing pressures in any sense dangerous to a properly constructed engine. The pressure produced by the explosion of any mixture of gas and air is strictly determined and limited, whereas the pressure produced by the explosion of gunpowder depends greatly upon the relation between the volume of the gunpowder and the space in which it is confined. Engines of the "Lenoir" type are the simplest in idea and construction; in them a mixture of gas and air is made in the cylinder during the first half of the piston stroke, air being taken from the atmosphere and drawn into the cylinder by the forward movement of the piston. At the same time gas entering by a number of holes, and streaming into the air to form an explosive mixture, the movement of a valve cuts off the supply, and brings the igniting arrangement into action. The pressure produced by the explosion acting upon the piston makes it complete its stroke, when the exhaust valve opens exactly as in the steam engine. The Lenoir and Hugon engines, the earlier forms of this type, were double acting, receiving two impulses for every revolution of the crank, the impulse differing from that in a high pressure steam engine in commencing at half stroke. The Lenoir igniting arrangement was complicated and troublesome. I have it upon the table; the mixture was ignited at the proper time by the electric spark produced from a primary battery and Ruhmkorff coil. The Hugon engine was an advance in this respect, using a flame ignited, and securing greater certainty of action in a comparatively simple manner. It is really a modification of Barnett's lighting cock described in his patent of 1838.