WATER. 



barrel, fo that when it is let down it will occupy the whole 

 interior fpace of the barrel, except that fmall fpace which 

 is left between the infide of the barrel and the pitton, to 

 avoid aftual contaft. The bottom of the barrel has a valve 

 in it which opens upwards, and a pipe proceeds from the 

 lower part to convey away the water to the refervoir into 

 which it is to be raifed by tlie pump. This pipe is pro- 

 vided with a valve, to prevent the return of any water 

 which has pafled through it, but the greateft elevation of 

 the water in the refervoir muft not be quite fo great as the 

 top of the barrel. When this pump is fixed for work, the 

 lower end of the barrel muft: be immerfed in the water of 

 the well at lead as much as the whole length of the ftroke, 

 fo that the lower end of the pifton will never rife above the 

 furface of the water in the well, and upon this circumftance 

 the aftion of the pump depends ; for when the pillon is 

 drawn up, the water flows through the valve in the bottom 

 by its gravity, and fills the fpace which is left by the draw- 

 ing up of the pifton ; when the pifton defcends, it difplaces 

 from the barrel all this water, and forces it up the fide-pipe 

 into the refervoir. It is true that a fmall portion of water 

 rifes in the fpace between the barrel and the pifton, but 

 this fmall quantity cannot efcape, becaufe the top of the 

 barrel rifes higher than the furface of the water in the re- 

 fervoir. 



Dr. Robinfon, who we believe firft defcribed this pump, 

 obferves that it is free from all the difficulties which are 

 experienced in common pumps, from want of being air- 

 tight. Another is, that the quantity of water raifed is 

 very nearly equal to the power expended ; for if there is 

 any want of accuracy in the work, which occafions a dimi- , 

 nution of the quantity of water difcharged, it alfo makes an 

 equal diminution in the force which is neceffary for pufliing 

 down the plunger. The dodlor mentions a machine, con- 

 fift;ing of two fuch pumps, the piftons of which were fuf- 

 pended from the arms of a long beam or lever, the upper 

 fide of which was formed into a walk, with a rail on each 

 fide. A man ftood on one fide of the centre of the lever, 

 until the pifton of the pump at that end funk to the bottom 

 of its barrel, and of courfe the pilton of the pump on the 

 oppofite fide of the centre was drawn up ; he then walked 

 flowly up to the other end of the walk upon the beam or 

 lever, the inclination being about twenty-five degrees at 

 firft, but gradually diminiftied as he went along, and pafled 

 on the oppofite fide of the centre of motion, fo as to change 

 the load of the beam. By this means he made the pifton at 

 the other end go to the bottom of its barrel, and fo on al- 

 ternately, with the eafieft of all exertions, and what a man 

 is moft fitted for by his ftrufture. With this machine a 

 feeble old man, weighing no pounds, raifed 7 cubic feet 

 of water 11-5 feet high every minute, and continued work- 

 ing eight or ten hours every day. A ftout young man, 

 weighing nearly 135 pounds, raifed 83- cubic feet to the 

 fame height ; and when he carried 30 pounds conveniently 

 flung about him, he raifed 9^ feet to this height, working 

 ten hours a day, without greatly fatiguing himfelf. This 

 exceeds Defagulitrs' maximum of a hogftiead of water ten 

 feet high in a minute, in the proportion of 9 to 7 nearly. 

 This pump is limited to very moderate heights, and in fuch 

 fituations it is very effeftual. 



The mercurial pump is a fpecics of lift -pump, in which 

 mercury is employed to make a clofe fitting between the 

 pifton and the barrel, and thus avoid the friftion of leathers, 

 and prevent lofs of water. 



This pump was originally invented by Mr. Jofliua Haf- 

 kins, and was improved by Defaguliers, who defcribed it in 



the Philofophical Tranfaftiont for 1722, N* 370. p. 5 ; 

 and he has alfo given every detail of the conftruftion in his 

 Experimental Philofophy, vol. ii. p. 491. 



In this pump the barrel is inverted, that is, it is open at 

 the bottom, like the firft lift-pump which we have men- 

 tioned ; and it has alfo two pipes communicating with the 

 upper end of the barrel, one to bring up the water from 

 the well, and the other to carry it up to the refervoir : each 

 pipe is provided with its valve, to prevent the return of the 

 water. The barrel muft be made of iron, and as thin as is 

 confiftent with the ftrength of the metal. The pifton is a 

 cylindrical plug of wood, fitted to the barrel fo as to fill it, 

 but not to touch the fides. This pifton is fixed perpendi- 

 cularly in the centre of a hollow cylinder of iron, vihich is 

 rather larger within than the outfide of the pump-barrel, fo 

 that an annular fpace is left all round between the fodid 

 pifton or plug and the infide of the cyhnder, into which fpace 

 the pump-barrel can enter, and will fill it very nearly. The 

 annular fpace is then filled with mercury. This compound 

 piece, confifting of the hollow cylinder, with the fmaller 

 fohd cylinder within it, forms the pifton ; and to this the 

 power which is to work the pump is applied by means of 

 chains, which fufpend it from the fhort arms, fo that if the 

 lever is moved, the pifton will rife up and down. When 

 the pifton is applied in its place, and the inverted pump, 

 barrel is received into the annular fpace between the folid 

 and hollow cylinders, the mercury therein will make a clofe 

 fitting between the folid pifton and the infide of the barrel, 

 fo as to prevent any water paffing between them ; and the 

 afcent and defcent of the pifton will produce an alternate 

 contraftion and dilatation of the internal capacity of the 

 working barrel, in the fame manner as a folid pifton would 

 do, if it was clofely fitted to the infide of the barrel with 

 leather all round. 



As the water exerts a preffure on the mercury, to force 

 it out of the annular fpace in which it is lodged, the depth 

 of the annular fpace and length of the barrel which defcends 

 into it muft be adapted to the height to which the water is 

 intended to be elevated ; fo that the column of mercury 

 which it will contain, without raifing the mercury fo high ai 

 to run over the edge of the external cyhnder, (hall always 

 exceed one-thirteenth part of the height to which the water 

 is to be elevated ; the weight of mercury being more than 

 thirteen times the weight of an equal quantity of water. 



That there may be lefs mercury ufed, the pump-barrel 

 (hould be made of plate-iron, turned on the outfide, and 

 bored within ; the outer cylinder of the pifton ftiould be 

 bored, and the inner one turned ; and if the work be well 

 performed, eight or ten pounds of mercury will be fuffi- 

 cient, though the bore of the barrel, or diameter of the 

 column of water which is raifed, is fix inches. Lefs than 

 fix pounds of mercury would fuffice, if there were two bar- 

 rels, in order to keep a conftant ftream. This will very 

 much leflen the expence of mercury, which would otherwife 

 be an objedlion againft this pump ; and by making the inner 

 and outer cylinder of hard wood, as box, or lignum vita, the 

 expence may ftill be reduced. But if the engine be very 

 large, caft-iron bored will be proper for the outer cyhnder, 

 and caft-iron turned on the outfide for the inner cyhnder or 

 plug, and hammered iron bored and turned for the middle 

 cylinder. 



There is an objeftion, which feems at firft to take off the 

 intended advantage of this engine, viz. that inftead of the 

 friAion of the leather of a pifton, when wc hft up the pifton 

 to make a ftroke, the rcfiftancc neceffary to make the mer- 

 cury to rife on the outfide of the barrel in the outer cylinder 



of 



