of his senses.

The late Sir William Siemens worked for many years on combustion engines,
some of his patents on this subject dating back to 1860. In the course of
a conversation I had with him on the subject of his earlier patents, I
asked him why he had entitled one of those patents "steam engine
improvements" when it was wholly concerned with a gas engine using
hydrogen and air in the motive cylinder, the combustion of the hydrogen
taking place in the motive cylinder. He answered me that in 1860 he did
not care to entitle his patent gas or combustion engine simply because
engineers at that time would have thought him mad.

Notwithstanding this widespread incredulity among engineers, and the
apparent novelty of the gas engine idea, fire or combustion engines have
been proposed long, long ago. The first Newcomen steam engine ever set to
work was used by a Mr. Back, of Wolverhampton, in the year 1711.
Thirty-one years before this time, in Paris--year 1680--Huyghens presented
a memoir to the Academy of Sciences describing a method of utilizing the
expansive force of gunpowder. This engineer is notable as being the very
first to propose the use of a cylinder and piston, as well as the first
combustion engine of a practical kind.

The engine consists of a vertical open topped cylinder, in which works a
piston; the piston is connected by a chain passing over a pulley above it
to a heavy weight; the upstroke is accomplished by the descent of the
weight, which pulls the piston to the top of the cylinder; gunpowder
placed in a tray at the bottom of the cylinder is now ignited, and expels
the air with which the cylinder is filled through a shifting valve, and,
after the products of combustion have cooled, a partial vacuum takes place
and the atmospheric pressure forces down the piston to the bottom of its
stroke, during which work may be obtained.

On the board I have made a sketch of this engine. Some years previous to
Huyghens' proposal, the Abbe Hautefeuille (1678) proposed a gunpowder
engine without piston for pumping water. It is similar to Savery's steam
engine, but using the pressure of the explosion instead of the pressure of
steam. This engine, however, had no piston, and was only applicable as a
pump. The Savery principle still survives in the action of the well-known
pulsometer steam pump.

Denys Papin, the pupil and assistant of Huyghens, continued experimenting
upon the production of motive power, and in 1690 p