Let A B (_fig. 69._,) as before, represent a tube, the bottom of which is equal to a square foot, and let P be a piston in it, resting upon a cubic inch of water spread over the bottom; and let w be an empty vessel, the weight of which exactly counterpoises the piston. By the application of the lamp, the water will be converted into steam of the atmospheric pressure, and the piston will be raised from P to P´, through the height of one foot, the space in the tube beneath it being filled with steam, and the vessel w will have descended through one foot. Let half a ton of water be now poured into the vessel W; its weight will draw the piston P´ upwards, so that the steam below it will expand into a larger space. When the piston P´ was only balanced by the empty vessel W, it was pressed downwards by the whole weight of the atmosphere above, which amounts to about one ton: now, however, half of this pressure is balanced by the half ton of water poured into the vessel W; consequently the effective downward pressure on the piston P´ will be only half a ton, or half its former amount. The piston will therefore rise, until the pressure of the steam below it is diminished to the same extent. By what has been already explained, this will take place when the steam is allowed to expand into double its former bulk; consequently, when the piston has risen to P´´, one foot higher, or two feet from the bottom of the tube, the steam will then exactly balance the downward pressure on the piston, and the latter will remain stationary; the vessel W, with the half ton of water it contains, will have descended one foot lower, or two feet below its first position. Let the steam now be cooled and reconverted into water, and at the same time let another half ton of water be supplied to the vessel W; the pressure below the piston being entirely removed, the atmospheric pressure will act above it with undiminished force; and this force, amounting to one ton, will draw up the vessel W, with its contents. When the piston descends, as it will do, to the bottom of the tube, the ton of water contained in the vessel W will be raised through two perpendicular feet.[49] [Footnote 49: Strictly speaking, the quantity of water supposed in these cases to be placed in the vessel W would just _balance_ the atmospheric pressure. A slight preponderance must therefore be given to the piston, to produce the motion.]