WATER. 



of the pifton is at leaft as great as the friftion we avoid. 

 Defaguliers fays, that refiftance is never greater than the 

 weight of a concave cylinder of mercury, whofe height is 

 the greateft to which the mercury rifes in the faid outer 

 cylinder, and the bafe is the area of the barrel itfelf. This 

 weight in a pump of 6 inches bore is equal to 57^ pounds, 

 and, therefore, it would appear to be greater than the re- 

 fiftance arifing from the friftion of a pifton. But if it be 

 confidered, that in the defcent of the pifton for fucking, the 

 mercury fhifts immediately into the infide of the barrel, 

 rifing to the fame height therein, and ftill keeping the fame 

 bafe, the weight of 57-^ pounds helps to prefs down the 

 pifton, and faciUtates the overcoming of the force of the 

 atmofphere, or fuftion of the pump ; confequently, the 

 weight of the mercury being balanced is no hindrance, 

 whether the pump works with a double or with a fingle 

 barrel. 



There remains only then the hindrance by lofs of time, 

 whilft the mercury changes from the outfide to the infide of 

 the barrel, at the beginning of any ftroke. Defaguliers 

 ftates this to be one-fifty-fecond part of the ftroke, and that 

 he found tlie beil pumps then in ufe generally loft near 

 one-fifth of the water that they ought to have given, ac- 

 cording to their number of ftrokes. 



Notwithftanding the high terms in which this author and 

 others have fpoken of the mercurial pump, it can only be 

 confidered as an ingenious fuggeftion, for the expence of 

 mercury would be too great for the aftual application of 

 any fuch machine in praftice ; and in refpeft to friftion, it 

 would have a confiderable (hare of refiftance in plunging 

 the pifton into the mercury, although there would be no 

 aftual rubbing of hard fubftaiices together. This refiftance 

 would arife in the rapid running of the mercury from the 

 infide of the barrel to the outfide, and back again, at the 

 beginning of each ftroke. 



The machine is exceedingly ingenious and refined, and 

 there is no doubt but that its performance will exceed that 

 of any other pump which raifes the water to the fame 

 height, becaufe there can be no want of tightnefs in the 

 pifton, and friftion is in a great meafure avoided. But 

 thefe advantages are but trifling. The expence would be 

 enormqus ; for with wliatever care the cylinders are made, 

 the interval between the inner and outer cylinders muft 

 contain a very great quantity of mercury. The middle cy- 

 Lnder muft be made of iron-plate, and without any feara, 

 for mercury diflblves every kind of folder. For fuch rea- 

 fons, it has never come into ufe. But although we have 

 profeffed to defcribe only the machines in aftual ufe, it 

 would have been unpardonable to have omitted the defcrip- 

 tion of an invention, which is fo original and ingenious ; 

 and there are fome occafions where it may be of ufe, fuch 

 at nice experiments for illuftrating the theory of hydrauhcs : 

 it would be the beft pifton for meafuring the preftiires of 

 water in pipes, being in faft the fame principle as the baro- 

 meter. 



Seflor pumps are thofe in which the pifton is made to 

 move upon a centre, like a door upon its hinges. The 

 pifton is inclofed within a veflel ftiaped like the feftor of a 

 circle, which forms the body of the pump, and which is 

 divided by the pifton into two compartments. The pifton 

 is fitted, fo that it can move backwards and forwards on its 

 centre of motion, without fuff'ering any water to pafs by it ; 

 and by this motion it will alternately enlarge or contraft the 

 capacities of the two compartments, fo as to draw in water 

 through pipes and valves properly fituated, and force it out 

 again at other pipes. Thefe kinds of pumps are difficult to 



conftruft, and have no advantages over the pumps with 

 ftraight barrels, except for the engines for extinguifhing 

 fire. See that article for a defcription and figure of Mr. 

 RovsTitree's, which is one of the beft of this kind. 



Rotatl've Pumps — As moft of the firft movers for hy- 

 draulic machinery aft with a rotative motion, it would be 

 very defirable to have a pump which would at once employ 

 the rotative force to the purpofe of raifing water. Many 

 fchemes have been propofed, and much ingenuity difplayed 

 in thefe inventions ; but hitherto no one has been brought 

 to fuch perfeftion as to be equal to the pumps with ftraight 

 barrels. In Ramelli's work, pubhftied in 1588, feveral 

 rotative pumps are defcribed ; and Leopold has made a 

 coUeftion of them in his " Theatrum Machinarum Hydrau- 

 hcarum," vol. i. They are all upon one common prin- 

 ciple, viz. a hollow cylinder or drum clofed on all fides ; 

 within this another fmaller cylinder is inclofed, and the in- 

 terior cylinder is placed out of the centre of the hollow cy- 

 linder, fo that the interior cylinder touches the hollow one 

 at one point of the circumference ; but at all other points 

 there is a confiderable fpace between the two. The interior 

 cylinder is provided with four or fix valves or leaves, which 

 are united to it by hinges, and, when folded clofe up to the 

 cylinder, will form a fmooth and circular circumference ; 

 but if the leaves are opened out, they will reach to the in- 

 terior firface of the hollow cylinder. When the interior 

 cylinder is turned round by a handle applied to the axis, the 

 valves fweep round within the hollow cylinder, and in this 

 motion perform the office of piftons, becaufe tliey clofe up 

 to the internal cyhnder, in proportion as they approach to- 

 wards the point where the internal cylinder touches the 

 hollow cylinder ; and the fame vanes open out again, after 

 they have pafTed that point. In this way the fpaces between 

 the valves form a number of cavities, which alternately ex- 

 pand and contraft in their capacity, and in confequence 

 they will draw up water through a pipe which is inferted 

 into the hollow cylinder, and force it out at another pipe, 

 fo as to raife up a continual ftream. 



The machine is fometimes varied, by making the hollow 

 cylinder of an elliptical form : in other cafes, the valves, 

 inftead of moving upon hinges, are made to Hide in ftraight 

 lines from the centre of the revolving cylinder ; but in either 

 cafe, the aftion is the fame. The common defeft of all 

 thefe rotatory pumps is, that it is very difficult to pack them 

 fo as to be tight, and they have more friftion than any other 

 kind of pump. 



The centrifugal pump, invented by Mr. Erfliine, may be 

 called a rotative pump, but it is on a different principle 

 from all other pumps. A perpendicular pipe has another 

 joined to it, in form of the letter T : the lower end of this 

 pipe being immerfed in water, and the whole filled with 

 water, it is turned round on the perpendicular item as an 

 axis ; the water contained in the horizontal arms will, by its 

 centrifugal force, fly out, and draw a conftant ftream of 

 water up through the perpendicular pipe. See Centrifugal 

 Pump. 



Spiral pump, or Zurich machine, is a hollow drum or cy- 

 linder turning on a horizontal axis, and partly plunged in a 

 ciftern of water, like a very large grind-ftone. The interior 

 fpace of this cylinder or drum is formed into a fpiral canal, 

 by a plate coiled up within it, like the main-fpring of a 

 watch in its box, only that the fpirts are fituated at a given 

 diftance from each other, fo as to form a fpiral paffage of 

 uniform width. [See Jig. 11. Plate Water-Works.) This 

 fpiral partition is well joined to the two circular ends of the 

 cylinder, and no water can efcape between them. The 



inner 



