There is in all descriptions of pumps a certaiu amount of power required to be exerted, in addition to that which is necessary to raise the quantity of water to be delivered. In an ordinary suction pump this additional power is, on the, ascending stroke, composed of the weight of the water above the of the weight of the piston itself, and of the friction of the various parts. D'Aubuisson calculates that, in well-made pumps, the sum of these elements is equal to about h more than the real statical weight resting on the piston. Ou the descending stroke the piston meets with a resistance arisiug from the contraction of the fluid vein, when the water passes through the clacks, and from the friction of the packing against the sides of the cylinders ; as this resistance is, however, considerably less than that which is overcome on the ascending stroke, little attention is paid to it. In cases where large quantities of water have to be raised by pumps of this description, it is customary to couple them, in order to equalise the power exercised during the upward and downward strokes, by making the motions of the respective pistons reciprocally alternating. The two pumps pour their water into the same head, which thus gives nearly a constant discharge.
The resistances involving loss of power are much more numerous and more complicated in the case of force pumps, for in them not only is the whole column of water to be set in motion, but there is a loss of useful effect : 1st, in consequence of the contraction of the fluid vein at the foot valves ; 2nd, in consequence of the bends, or changes of direction in the line of flow ; 3rd, in the weight and friction of the valves; 4th, in the friction of the pumping machinery ; and Lth, in the friction of the ascending column of water against the sides of the pipes, and in the inertia of the water itself. D'Autetiason gives the formula usually adopted to calculate the value of these resistances (pp. 519 to 525); but he prefaces them by saying. that it 6 impossible to determine them rigorously. It may then suffice to say, that in some miningspumps, where the lift is very great, the sum of these resistances is such as to reduce the useful effect of the pumping machiuery to 'about one-half of the power exercised. In largo force pumps the pistons are usually made of the description known techni cally by the name of plungers; that is to say, they are formed of hollow metal cylinders working in a stuffing-box, but with a small annular space between their sides, and those of the cylinders. Theo retically, the quantity of water raised by each stroke of the plunger would be equal to its sectional area multiplied by the length of the stroke, but in practice there is found to be a considerable loss of effect ; and in cases where plunger pumps are used iu conjunction with lifting pumps (as in deep mines), it is customary to make the former of larger diameter than the latter. In the force-pumps of very deep mines it is necessary to counterbalance the weight of the piston rods, and to take great precautions to prevent their becoming bent under the jars they are exposed to by their own vibration.
The action even of connected pumps being of necessity of an inter mittent character. there is introduced near the point of delivery a vessel, called the air rend, into which the water is forced in quantities than can be discharged by the outlet valve, during the period of inlet. The water then accumulates in the air vessel, and compresses the air contained in it, sometimes even to the extent of producing a pressure equal to three atmospheres. Of course when the water ceases to enter through the valve couneeted with the piston, the reaction of the compressed air maintains the discharge through the foot valve, and thus the discharge is rendered almost continuous. The air vessels also diminish in a potable manner the hydraulic jar on the rising main, produced by the intermittent action of the pumping machinery.
In addition to the suction and forcing pumps, there are several descriptions of machines used for raising water, such as the chain pumps, norias, Archimedian screws, scoops, bucket wheels, Persian wheels, flash wheels, rotary pumps, &c.; and perhaps it would be preferable at once to notice them, oven though in some cases the machinery can hardly be considered to come under the strict definition of the word " pump." Thus, the chain pump consists of a series of flat boards which are attached to an endless chain. and the latter works upon fixed wheels at the top and bottom of the course; tho flat boards in ascending pans through a cylinder, and lift in it a column of water. The Noria is a machine much used in the East for irrigation purposes, and consists of a series of earthen pots ranged upon a rope, working also on fixed wheels; so that the pots dip into, and raise the water from the bottom of the well, to discharge it whan they pass over the top wheel. The A rehhnedian screws, as their name implies, consist of a series of threads, or worms, fixed upon an axis of rotation bearing on a fixed pivot at the lower end, and having a winch handle at the top. There are usually three threads working in a species of close fitting race, or channel, open at the top; the diameter of the thread is usually made from 13 to 26 inches, but occasionally it has been made as much as 6 feet ; the angle made by the thread with the axis of the central shaft (which, it may be added, is usually equal in diameter to ird of that of the thread) varies between 45° and 778°; but in practice an angle of CO' is found to be the most advantageous. The axis of the central shaft is usually placed so as to form an angle of from 30° to 45° with the horizontal lino. These Archiinedian screws are much used in Rolland fur drainage well:a; they are driven by wind-111116, and raise the water very economically, provided the lift be not more than 15 [mt. On the Sambre and Meuse canal a screw of 5 feet 4 indica diameter, and 21 feet 6 inches lift, worked by a steam engine, has given very satisfactory ressults.