WATER. 



, It (hould follow, from experiments on the nature of air, 

 that the column F D is counterpoifed by the comprefled air 

 in the inverfe ratio of 104 to 32 : hence the volume of air 

 contained in the vefTel A and the pipe L M, equal to 58^ 

 cubic feet, muft be reduced to 18 cubic feet, before the 

 elafticity will be equal to the preffure of the column C F 

 104 feet; but by increafmg the compreflion a little more, 

 the water in B will be made to flow out through F. 



If, at the moment the vefTel A is full of water, the 

 cock H be Hiut, the water will continue to flow through F, 

 until the air occupies a fpace of 1 8 cubic feet in the veffel B, 

 and in the pipe L M ; the elafticity of the air will then be 

 in equilibrio with the column FD, and the efflux of the 

 water through F will ceafe. In this manner, not above 

 17 cubic feet of water are evacuated at each draught, and 

 io| cubic feet are conftantly left in the veflel B. 



But if the cock H is not fhut the very moment that the 

 veflel A is full, the water in A will follow the air through 

 L M, and, before it gets to the vefTel B, will raife one 

 cubic foot more out of that vefTel. After the water from 

 A enters into the vefTel B, the difcharge at F will not be 

 the water of B, but the water of A defcending and afcend- 

 ing again by a ufelefs circuit, until H be (hut ; which being 

 done, the water will continue to flow at F, until the re- 

 mainder of 1O5 cubic feet is expelled from B by the air 

 contained in it. The moment when the water from A has 

 defcended into the lower vefTel B may eafily be known, by 

 the velocity of the efflux at F becoming fuddenly three times 

 greater. 



That this is aftually the cafe is proved, becaufe fometimes 

 315 cubic feet are difcharged ; which quantity exceeds the 

 capacity of the vefTel B by more than 4 cubic feet. 



This inconvenience might eafily have been prevented, by 

 giving to the pipe S a diameter of i8 inches ; for then there 

 would have remained only the juft fpace of i8 cubic feet for 

 the comprefTed air. 



The height of the column T to the loweft of the two 

 refervoirs at O is 143 feet, which, taken upon the diameter 

 of the veflel A as a bafe, is equal to the weight of 822^ 

 cubic feet, and would comprefs the air into a fourth ; or, 

 when the water is defcending into the lower vefTel B, into a 

 feventh part of its natural fpace, provided it were equally 

 refifted at F. The vefTel A becomes filled at a mean in 

 8 Teconds ; and in twice that fpace of time, 1 7 cubic feet 

 are evacuated through F. 



The power of the column of 260 feet from the moft ele- 

 vated refervoir, afting within the vefTel A, is equivalent to 

 the weight of 1495 cubic feet of water. It can raife a 

 greater quantity, if the vefTel B be fo conftrufted as to 

 allow no more than a juft fpace to the comprefTed air. If 

 the vefTel A were filled in 4 feconds, then 17 cubic feet of 

 water would be difcharged through F in twice that time, 

 and the air would be reduced into an eighth, and, during 

 the defcent of the water of the vefTel A into the lower vefTel 

 B, into an eleventh part of its bulk. But this makes no 

 alteration as to the quantity of the effeft ; and when water 

 ceafes to flow out at F, there will always remain 105 cubic 

 feet of water in the vefTel B. 



Two men are required to attend it, but it would be very 

 eafy to conneft the levers of the cocks above and below, fo 

 as to require only one man to work the whole fet ; and in- 

 deed there would be little difficulty in making the machine 

 work itfelf fafely, without any attendant, except to fet it 

 off at firft, or flop it when requifite. The machinery for 

 this purpofe has been propofed by Mr. Bofwell. See Nichol- 

 fon's Journal, 4to. iv. 117. 



From what has been faid, it is evident that this machine, 

 Vol. XXXVIII. 



though i± anfwers the author's intention, is fo deficient ar, 

 to the effeft the fame fall of water might produce, as to bear 

 fcarce any proportion ; and there is a defeft in the principle 

 of the machine, -viz. that the air will require a confiderable 

 fliare of the power to comprefs it, and this air muft be fuf- 

 fered to efcape, before the vefTels can be refilled to repeat 

 the aftion ; in confequence, all the power taken to comprefs 

 the air is loft, and expands itfelf in forcing out a ftrong 

 blaft of air at the difcharging cock, without producing, any 

 ufeful effeft. Notwithftanding this defeft, the cheapnefa 

 and eafe of conftrufrtion, and the little wear and tear, to- 

 gether with the facility with which it may be made to work 

 and ftop for very fliort periods of time, are powerful recom- 

 mendations of this machine, in fuch places as afford the re- 

 quifite fall of fuperior water, and do not require a higher 

 fingle lift than 15 or 20 fathoms. 



A curious phenomenon has been obferved in this machine, 

 when it is near the end of its operation, that is, when nearly 

 the whole of the water has been raifed out of the lower 

 vefTel B, and the cock E be opened to give vent to the 

 comprefTed air, and beforethe cock L is fhut, fo that the 

 air is followed up by the water, then if a hat or miner's 

 bonnet be prefented to the aperture E, the aqueous vapours 

 difperfed through the comprefTed air, and perhaps alfo, 

 fays M. Jars, part of thofe of the external air are con- 

 denfed in the bonnet in the form of very white and compaA 

 ice, very much refembling hail, and not eafily feparated 

 from the bonnet. It foon melts, which is not to be won- 

 dered at, as the temperature of the place itfelf is not cold. 

 MefTrs. Du Hamel and Jars remained in Hungary from 

 January to July 1758, and obferved the fame phenomenon 

 at all feafons ; but as they had no thermometer, they could 

 not make a number of experiments, which might have been 

 of value in the inveftigation of the fubjeft. 



It is obferved that the air iffues out with fuch impetuofity, 

 that the workman could not hold the bonnet at the diftance 

 of a few inches from the aperture, as he does in this experi- 

 ment, if he were not fupported behind. The ice is much 

 more compaft, if the cock be only in part opened. 



When the cock at which the air is difcharged is opened, 

 it rufhes out with prodigious violence, and the drops of 

 water are changed into hail or lumps of ice. It is a light 

 ufually fhewn to ftrangers, who are defired to hold their 

 hats, to receive the blafts of air : the ice comes out with 

 fuch violence as frequently to pierce the hat like a piftol 

 bullet. This rapid congelation is a remarkable inftance of 

 the general faft, that air, by fuddenly expanding, generates 

 cold ; its capacity for heat being increafed. 



The formation of the ice and fnow, when the condenfed 

 air rufhes out of this machine, has been explained in a dif- 

 ferent way ill almoft every fyftem of philofophy. It ap- 

 pears to us to be a neceffary confequence of the condenfed 

 air, on rufliing out into the open air. 



The air of the atmofphere, and the water when taken into 

 the macfiine, are nearly of the fame temperature ; and it 

 may be confidered that each cubic foot of water and of air 

 contains fome certain quantity of heat or caloric ; but they 

 will readily impart a portion of this heat to any body contain- 

 ing a lefs degree than themfelves, or they will abforb or take 

 up heat from any body containing a greater proportion of 

 heat than themfelves, in confequence of that property of heat, 

 by which it will diftribute itfelf equally among all bodies 

 which are in contaft with each other. By the aftion of the 

 machine, the air is comprefTed into one-third of the fpace it 

 before occupied, and the fhare of heat contained in that air 

 is hkewife concentrated or thrown into a third of the fpace, 

 and ill confequence becomes more intcnfe. Some part of 

 I the 



