At the end of last century Lavoisier and Laplace, and after them, down to our own time, Dulong, Desprez, Favre and Silbermann, Andrews, Berthelot, Thomson, and others, devoted much time and labor to the experimental determination of the heat of combustion and the laws which governed its development.  Messrs. Favre and Silbermann, in particular, between the years 1845 and 1852, carried out a splendid series of experiments by means of the apparatus partly represented in Fig. 1 (opposite), which is a drawing one-third the natural size of the calorimeter employed.  It consisted essentially of a combustion chamber formed of thin copper, gilt internally.  The upper part of the chamber was fitted with a cover through which the combustible could be introduced, with a pipe for a gas jet, with a peep hole closed by adiathermanous but transparent substances, alum and glass, and with a branch leading to a thin copper coil surrounding the lower part of the chamber and descending below it.  The whole of this portion of the apparatus was plunged into a thin copper vessel, silvered internally and filled with water, which was kept thoroughly mixed by means of agitators.  This second vessel stood inside a third one, the sides and bottom of which were covered with the skins of swans with the down on, and the whole was immersed in a fourth vessel tilled with water, kept at the average temperature of the laboratory.  Suitable thermometers of great delicacy were provided, and all manner of precautions were taken to prevent loss of heat.

[Illustration:  THE GENERATION OF STEAM.  Fig 1.]

It is impossible not to admire the ingenuity and skill exhibited in the details of the apparatus, in the various accessories for generating and storing the gases used, and for absorbing and weighing the products of combustion; but it is a matter of regret that the experiments should have been carried out on so small a scale.  For example, the little cage which held the solid fuel tested was only 5/8 inch diameter by barely 2 inches high, and held only 38 grains of charcoal, the combustion occupying about sixteen minutes.  Favre and Silbermann adopted the plan of ascertaining the weight of the substances consumed by calculation from the weight of the products of combustion.  Carbonic acid was absorbed by caustic potash, as also was carbonic oxide, after having been oxidized to carbonic acid by heated oxide of copper, and the vapor of water was absorbed by concentrated sulphuric acid.  The adoption of this system showed that it was in any case necessary to analyze the products of combustion in order to detect imperfect action.  Thus, in the case of substances containing carbon, carbonic oxide was always present to a variable extent with the carbonic acid, and corrections were necessary in order to determine the total heat due to the complete combination of the substance with oxygen.  Another advantage gained was that the absorption of the products of combustion prevents any sensible alteration in the volumes during the process, so that corrections for the heat absorbed in the work of displacing the atmosphere were not required.  The experiments on various substances were repeated many times.  The mean results for those in which we are immediately interested are given in Table I., next column.