SIPHON. 



K L, fuftains the column of water, L D, in the (hort leg 

 of the fiphon, prefling againlt that air with its perpendicular 

 height D F ; whiHt the column of air M S, prefling up- 

 wards againit the whole of the long or iduing leg D S, 

 which afts according to the height D C, muft yield and 

 fufFer the water to run out, as long as the leg D S is 

 longer, or rather perpendicularly higher, than D G. For 

 fince K and M are fuppofed to be at the top of the atmo- 

 fphere, the columns K L and M S are equal in height and 

 preflure, (for the height of L above S is of no account with 

 refpeft to the whole height of the atmofphere,) as long as 

 M S is afted upon by the defcending water D S, whofe 

 height is from D to S, {e.g. fifteen inches,) an height fu- 

 perior to that of the column D G, {e.g. feven inches,) 

 Supported by the column of air K L, the column M S mult 

 yield to the water iffuing out at S ; and iiowever the furface 

 of the water, E F, defcends, the column K L, by its pref- 

 fure, will always overcome the refinance of the column 

 M S, beca\ife it has a lefs height of water to fullain than 

 M S has. If the mouth of the illuing leg had been at T, 

 the water would hang in equU'ibrio, filhng both legs of the 

 fiphon, when the water is come down in the upper veflel to 

 I H T ; becaufe then the two columns of air, K L and M S, 

 will be afted againft by an equal heie;ht of water in the legs 

 of the fiphon ; but if you raife the ifiuing end of the fiphon 

 (now fuppofed at T) up to the level of u V above I H, 

 the water will run back, up from V to D, and fo out at H 

 in the upper veflel : becaufe then the column M S, having 

 only the height V D to fuftain, will be afted againft with 

 lefs force than the column K L, which is prefled againit by 

 the whole height, D H, fuperior to VD. Since, there- 

 fore, the prefl^ure of the air is the caufe of the water being 

 puftied up into the fiphon, and the difference of its preflure 

 (as one column is afted againft by the water in the fliort leg 

 more weakly than another column of air is afted upon by 

 the water in the long leg) is the caufe of its running con- 

 tinually from one veflel into another, when once fet a-going ; 

 it follows, that the bend, D, of the upper part of the 

 fiphon muft not be above thirty-two feet higher than the 

 water in the upper veflel, becaufe the air cannot fultain a 

 column of water, whofe height exceeds thirty-two feet. 



Mercury will run in a fiphon in the fame manner as water ; 

 but in this cafe, the bend of the fiphon muft not be more 

 than 30,^1^ inches above the ftagnant mercury in the upper 

 veflel ; becaufe, as it is about fourteen times fpecifically 

 heavier than water, it will be lifted up by the preflure of air 

 but the fourteenth part of the height to which the water is 

 lifted. Defag. Exp. Phil. vol. ii. fed. 8. 



But fome authors aflert, that the water continues to flow 

 through the fiphon, even when placed under a receiver, 

 and the air exhaufted from it. 



Some will have it, that there is ftill air enough left in the 

 evacuated receiver, to raife the water to an inch or two. 

 But as both mercury and water are found to fall entirely out 

 of the Toricellian tube, in "vacuo ; the prefigure of the thin 

 remaining air there can never be the caufe of the afceut, 

 both of mercury and water, in the fliorter leg of the fiphon. 

 Hence, as the height of the fiphon is limited to thirty-two 

 feet ; for this only reafon, that air cannot raife water 

 higher ; fome have queftioned whether or not we are in the 

 right in rejefting Hero's method of carrying water, by 

 means of a fiphon, over the tops of mountain?, into an op- 

 pofite valley. For Hero only orders the apertures of the 

 fiphon to be flopped, and water to be poured through a 

 funnel into the angle or meeting of the legs, till the fiphon 

 be full ; when, fliutting the aperture in the angle, and open- 

 ing the other two, the water, he fays, will continue to 



flow. Now, if there only need air for the firll rife of the 

 water into the lefs leg, not for the continuation of the mo- 

 tion, it were pofiible to raife the water much higher than the 

 height of the atmofphere would carry it. 



Wolfius (Elem. Mathef. tom. ii. p. 348. ed. 1733.) ex- 

 prefsly fays, that this propofed artifice of Hero is very 

 juftly rejefted ; becaufe air is necefl'ary not only for the 

 elevation of the water in the fhorter leg, but for the con- 

 tinuance of its motion ; and he adds, that this method was 

 unfuccefsfuUy tried, where the height was greater than 

 thirty-two feet. 



Defaguliers alfo obferves, that in a crane or fiphon, fuch 

 as that propofed by Hero, of about forty feet high, the 

 water, inftead of running from the upper veflel into the 

 lower, as it would do if the height of the fliort leg were 

 much under thirty-two feet, will in the two legs fall back 

 to the height of thirty-two feet, above the refpeiflive veflels, 

 where it will hang ; the air not being able to fuftain the 

 water above thefe heights, and confequently drive it up over 

 the bend. 



In an experiment of this kind, the water in the legs of 

 the fiphon, unlefs it be purged of its air, will not reft at a 

 height of quite thirty feet above the water in the veflels, 

 becaufe air will extricate itfelf out of the water, and getting 

 above the water in the legs, prefs it downwards, fo that its 

 height will be lefs to balance the prefl"ure of the atmofphere. 

 However, this is certain, that a fiphon of a particular kind, 

 once fet a running, will perfift in its motion, though re- 

 moved into the moft perfeft vacuum our air-pumps will 

 make : or, if the lower orifice of a full fiphon be ftiut, and 

 the whole be thus placed in a receiver, with a contrivance 

 for opening the orifice when the air is exhaufted ; the water 

 will be all emptied out of the veflel, as if it had been in 

 the open air. 



This faft has been fufBciently afcertained by many ap- 

 proved hydroftatical writers. 



Defaguliers informs us, that he made the experiment, 

 both with water and mercury ; for having filled a fiphon, 

 recurved at the extremities of its legs, fucceflively with 

 thofe liquors, and fufpended it by a flip-wire in the receiver 

 of an air-pump, over two fmall jars containing mercury to 

 unequal heights, (and water, when water was ufed in the 

 fiphon,) he exhaufted the air out of the receiver, and then let- 

 ting down the fiphon, fo that its two ends went into the hquor 

 in the jars, the liquor ran from the higher into the lower veflel. 

 He alfo made an experiment in the open air, where the mer- 

 cury ran through a fiphon, whofe bend was more than thirty- 

 one inches above the lower orifice of the fliort leg of the 

 fiphon. But neither of thefe experiments afl^ords a jull ob- 

 jeftion againft the doftrine advanced in the preceding part 

 of this article, to's. that the air is the caufe of the difcharge 

 of hquors from one veflel into another by means of fiphons ; 

 for its running in vacuo was only owing to the attraftion of 

 cohefion, which afts for a fmall height ; becaufe the experi- 

 ment will not fucceed in vacuo, if the fiphon ufed for mer- 

 cury has it« bend fix inches higher than the orifice of the 

 fliort leg, and if the bend of the fiphon for water be two or 

 three feet high ; neither will the laft-mentioned experiment 

 with mercury in the open air anfwer, if the bend of the 

 fiphon be forty inches high : and in all the experiments, the 

 bores of the fiphons muft be very fir.all. Exp. Phil, ubi 

 fupra, p. 168. 



The figure of the fiphon may be varied at pleafure (fee 

 Jigs. 3, 4, 5.), provided only the orifice C be below the 

 level of the (urface of the water to be drawn up ; but ftill 

 the farther it is diftant from it, the fafti-r will the fluid be 

 carriedoff^. And if, in the courfe of the flux, the orifice A 



be 



