To explain the first conception of this memorable invention; let a tube or pipe, S (_fig._ 19.), be imagined to proceed from the bottom of the cylinder A B to a vessel, C, having a stop-cock, D, by which the communication between the cylinder and the vessel C may be opened or closed at pleasure. If we suppose the piston P at the top of the cylinder, and the space below it filled with steam, the cylinder and steam being at the usual temperature, while the vessel C is a vacuum, and maintained at a low temperature. Then, on opening the cock D, the steam will rush from the cylinder A B through the tube S, and, passing into the cold vessel C, will be condensed by contact with its cold sides. This process of condensation will be rendered instantaneous if a jet of cold water is allowed to play in the vessel C. When the steam thus rushing into C, has been destroyed, and the space in the cylinder A B becomes a vacuum, then the pressure of the atmosphere being unobstructed, the piston will descend with the force due to the excess of the pressure of the atmosphere above the friction. When it has descended, suppose the stop-cock D closed, and steam admitted from [Pg122] the boiler through a proper cock or valve below the piston, the cylinder and piston being still at the same temperature as before. The steam on entering the cylinder, not being exposed to contact with any surface below its own temperature, will not be condensed, and therefore will immediately cause the piston to rise, and the piston will have attained the top of the cylinder when as much steam shall have been supplied by the boiler as will fill the cylinder. When this has taken place, suppose the communication with the boiler cut off, and the cock D once more opened: the steam will again rush through the pipe S into the vessel C, where encountering the cold surface and the jet of cold water, it will be condensed, and the vacuum, as before, will be produced in the cylinder A B; that cylinder still maintaining its temperature, the piston will again descend, and so the process may be continued. (69.) Having carried the invention to this point, Watt saw that the vessel C would gradually become heated by the steam which would be continually condensed in it. To prevent this, as well as to supply a constant jet of cold water, he proposed to keep the vessel C submerged in a cistern of cold water, from which a pipe should conduct a jet to play within the vessel, so as to condense the steam as it would pass from the cylinder.