comparison of the volume of the air with the height of the mercurial column demonstrated the diminution of the volume of the air to be precisely in the ratio of the pressure, so that the law of Marriotte is rigidly correct, even when extended to the extreme case where the air is reduced to less than part of its usual volume.

This preliminary process having been successfully terminated, the enormous column of glass was now laid aside, and the manometer, with its reservoir of mercury, transported to the court of the Observatory, for the purpose of being attached to the experimental boiler. Figure 18 shows the manometer in situ. An iron tube d d', g', composed of gun barrels welded together, connects the cover of the boiler a, with the reservoir of the manometer f, so as to conduct the pressure of the steam to the surface, which formerly had sustained the mercurial column. The vacant space above the mercury was filled with water, which, by condensation from a stream of water on the outside, was kept full to the constant height v. A column of water contained in the glass tube ≈ ≈, and constantly replenished, preserved the column of air, and other parts of the apparatus, at a constant temperature, indicated by a thermometer. A tube o p, of glass, communicating with the reservoir of mercury above and below, indicates, on the scale l, m, the variation of level arising from the recession of the mercury into the manometer tube.

To ascertain the temperature of the water and steam of the boiler, it had been considered sufficient in the ruder experiments of earlier observers to insert thermometers directly into the boiler itself. Every one who has an acquaintance with these instruments knows, that any difference of pressure on the glass produces a false indication of the instru

ments, so that even the few inches of mercury in the instrument itself, when inverted, alter its indications, and a

The following Table contains the results of Thirty of the most unexceptionable Experiments :

slight pressure of the finger would raise it a degree; the inaccuracy of the old method, when used under a pressure of 70 or 80 feet of mercury, or 450 pounds on every inch of the immersed surface of the instrument would have been great. The French academicians avoided this error, by immersing strong iron tubes t t, (figs. 18 and 19,) in the water and steam, in which the thermometers, surrounded by liquid metal, were kept in close communication with the heat of the fluids, without exposure to their force. By adopting only very slow variations of temperature, the error arising from the motion of heat was rendered insensible.

A table of temperatures, from 1 to 50 atmospheres, calculated in coincidence with the experiments of the French academicians, and adapted to English measures, is given by us in Article 57, for the purpose of convenient practical reference.

33. The latest series of experiments on the elastic force of high-pressure steam, we owe to America. At the request of the Hon. S. D. Ingham, Secretary of the Treasury of the United States, a committee of the Franklin Institute, of the State of Pennsylvania, was appointed "to examine into the causes of the explosions of the boilers used on board of steam-boats, and to devise the most ef fectual means of preventing the accidents, or of diminishing the extent of their injurious effects." Among other subjects, such as the strength of boilers, the construction of safety-valves, to which we shall refer in another place, this committee took into consideration the elastic force of high-pressure steam at different temperatures. Funds were placed at their disposal by the House of Representatives, and the committee consisted of such a combination

of scientific and practical men, as to give high authority to their results. On the 1st day of November 1830, the subject was placed in the hands of the following gentlemen - Professor Alex. Dallas Bache, Mr Benjamin Reeves, Mr W. H. Keating, Mr M. W. Balwin, Mr S. V. Berrick, and Isaiah Sukens.

We shall enter more fully on the description of their apparatus of experiment than we should otherwise have done, because we shall have frequent reference to make to the whole of their experiments, not only in this article,

Fig. 20.

N

but in our article on the Steam-Engine, where we treat of explosions of boilers and their causes.

The boiler used by the committee is represented in figs. 20, 21, 22. It is a cylinder, twelve inches in

internal diameter, two feet ten inches and a quarter in length within, and a quarter of an inch thick, of rolled iron, with the ends riveted in the usual manner. Fig. 21 is a side view. Figs. 20 and 22 are end views of the boiler, and of the apparatus connected with it. The boiler was placed horizontally in a furnace, the fire surface extend