2 24 AIR 
nnd the quickfilver driven by it from the baton up into 
the gauge, till the air remaining in it become of the fame 
d'enlity.as the air without; and, as the air always takes the 
higheft place, the tube being uppermoft, the expanlion will 
be determined by the number of divifions occupied by 
the air at the top. This air-pump is made to a6l alfo as 
a condenfing engine, as fome German machines had done 
before, by the very Ample apparatus of turning a cock. 
By means of this gauge, Mr. Smeatou judged that his 
machine was incomparably better than any former ones, as 
it feemed to rarefy the air in the receiver 1000, or even 
2Coo, times, while the belt of the former conftnuftion on¬ 
ly rarefied about 140 times; and lo the cafe has fince been 
always underflood, an implicit confidence being placed in 
Mr. Smcaton’s accuracy, till the fallacy was accidentally 
detefled in the manner related at large by Mr. Nairne, in 
the Philof. Tranf. for the year 1777. This accurate and 
ingenious artift, wanting to make trial of Mr. Smeaton’s 
pear-gauge, executed an air-pump of his improvedcon- 
•ftrudfion, in the. beft manner poilible; which, in various 
experiments made with it, appeared, by the pear-gauge, 
to rarefy the air to an amazing degree indeed, being at 
times from 4,000 to 10,600, or 50,000, or even 100,000, 
tunes rarefied. But, upon meafuring the fame expanlion 
by the ufual long and fliort tube gauges,, which both ac¬ 
curately agreed together, he found that thefe never (hew¬ 
ed a rarefaction of more than 600 times ; widely different 
from the fame as meafured by the pear or internal gauge, 
by experiments often repeated. 
“ Finding (fays Mr. Nairne) (till this di (agreement be¬ 
tween the pear-gauge and the other gauges, I tried a va¬ 
riety -of experiments ; but none of them appeared to me 
fatisfattory, till one day in April, 1776, fhewing an expe¬ 
riment with one of thefe pumps to the Hon. Henry Ca- 
vendifh, Mr. Smeaton, and leveral other gentlemen of 
the Royal Society, when the two gauges differed fome 
thoufand times from one another, Mr. Cavendifh account¬ 
ed for it in the following manner:—‘ It appeared (hefaid) 
from fome experiments of his father’s, lord Cavendifh, 
that water, w henever the prelfure of the atmofphere on 
it is diminifhed to a certain degree, is immediately turned 
into vapour, and is as immediately turned back again in¬ 
cowater on relioring the preflure. This degree of pref- 
fure is different according to the heat of the water : when 
the heat is 72 0 of Fahrenheit’s fcale, it turns into vapour 
as foon as the preflure is no greater titan that of three 
quarters of an inch of quickfilver, or about one-fortieth 
of the ufual prelfure of the atmofphere; but, when the 
heat is only 41 °, the prelfure mud be reduced to that of a 
quarter of an inch of quickfilver before the water turns 
into vapour. It is true, that water expofed to the open 
air, will evaporate at any heat, and with any prelfure of 
tire atmofphere ; but that evaporation is intirely owing to 
the afition of the air upon it; whereas the evaporation 
here fpoken of, is performed without any alfiftance from 
the air. Hence it follows, that when the receiver is ex- 
Jtaufted to the above-mentioned degree, the moilture ad¬ 
hering to the different parts of the machine will turn in¬ 
to vapour, and fupply the place of the air, which is con¬ 
tinually drawn away by the working of the pump ; fo 
that the fluid in the pear-gauge, as well as that in the re¬ 
ceiver, will confiff in a good meafure of vapour. Now, 
letting the air into the receiver, all the vapour within the 
pear-guage will be reduced to water, and only the real air 
will remain uncondenfed; confequently the pear-gauge 
fhews only how much real air is left in the receiver, and 
riot how much the preffure or fpring of the included 
fluid is diminifhed; whereas t;he common gauges fhew 
how much the prelfure of the included fluid is diininifh- 
ed, and that equally, whether it conlifts of air or of vapour.” 
Mr. Cavendifh having explained fo fatisfaCloril.y the 
caufe of.the difagreement between the two gauges, Mr. 
Nairne considered tltat, if lie were to avoid moifture as 
much as polfible, the two gauges ftiould nearly agree. 
And in fact they were found to do fo, eacli (hewing, a ra- 
* A I R 
refaction of about 600, when all moifture was perfectly 
cleared away from the pump, and the plate and the edges 
of the receiver w'ere fecured by a cement inftead of fet- 
ting it upon a foaked leather, as in the ufual way. But, by 
future experiments, Mr. Nairne found that the fame ex¬ 
cellent machine would not exhauft more than fifty or lixty 
times, when the receiver was fet upon leather foaked in 
water, the heat of the room being about 57 0 . And from 
the whole, Mr. Nairne concludes that the air-pump of 
Otto Guericke, and thole contrived by Mr. Gratorix and 
Dr. Hook, and the improved one by M. Papin, both ufed 
by Mr. Boyle; as alfo Haukfbee’s, s’Gravefande’s, Mu- 
chenbroeck’s, and of all thofe who-have ufed water in 
the barrels of their pumps, could never have exhaufted 
to more than-between forty and fifty, if the heat of the 
place was about 57 0 ; and although Mr. Smeaton, with 
his pump, where no water was in the barrel, but where 
leather (caked in a mixture of water and fpirit of wine 
was ufed on the pump-plate, to fet the receiver upon, mav 
have exhaufted all but a thoufandth, or even a ten-thoul 
fandth, part of the common air, according to the teftimo- 
ny of his pear-gauge; yet fo much vapour nutft have ari- 
fen from the wet leather, that the contents of the receiver 
could never be lefs than a feventierh or eightieth part of 
the denlity of the atmofphere. But, when nothing of 
moifture is ufed about this machine, it will, when in its 
great eft perfection, rarefy its contents of air about 6oo 
times. 
It is evident that by means of thefe two gauges we can 
afeertain the feveral quantities of vapour and permanent 
air which make up the contents of the receiver, after the 
exhauftion is made as perfect as can be ; for the ufual ex¬ 
ternal gauge determines the whole contents, made up of 
the vapour and air, whilft the .pear-gauge (hews the quan¬ 
tity of real permanent air ; confequently the difference is 
the quantity of vapour. 
The principal caufe which prevents this pump from 
exhaufting beyond the limit above-mentioned, is the weak¬ 
ened elaflicity of the air within the receiver, which, de- 
crealing in proportion as the quantity of the air within is 
diminifhed, becomes at laft incapable of lifting up the 
valve of communication between the receiver and the bar¬ 
rel ; and confequently no more air can then pafs from the 
former to the latter. 
Several ingenious perfons have ufed their endeavours to 
remove this imperfection in the belt air-pumps. Amongft 
thefe, one Mr. Haas has fucceeded tolerably well; having, 
by means of a contrivance to open the communication 
valve in the bottom of the barrel, made his machine fo 
perfeCt, that, when every thing is in the greatelt perfec¬ 
tion, it rarefies the contents of the receiver as far as 1000 
times, even when meafured by the exterior gauge. 
But the imperfections of the air-pump have more re¬ 
cently been removed by an ingenious contrivance of Mr. 
Cuthbertfon, a mathematical-inftrument maker at Amfter- 
dam, now of London, whofe air-pump has neither cock$ 
nor-valves, and is fo conftruCted, that what fupplies their 
place has the advantages of both, without the inconveni¬ 
ences of either. He has alfo made improvements in the 
gauges, by means of which he determines the height of 
the mercury in the tube, by which the degree of exhauf¬ 
tion is indicated, to the hundredth part of an inch. And 
to obviate the inconvenience of the elaftic vapour ariling 
from the wet leather, upon which the receiver is placed, 
for common experiments, lie recommends the ufe of lea¬ 
ther drelfed with allum, and foaked in hog’s lard, which 
lie found to yield very little of this vapour; but, when the 
utmoff degree of exhauftion is required, his advice is, to 
dry the receiver well, and fet it upon the plate without 
any leather, only fmearing its outer edges with hog’s lard, 
or with a mixture of three parts of hog’s lard and one of 
oil. But the ufe of the leather has long been laid alide 
by our Englifh inftrument-makers, a circtmiftance which 
probably had not come to Mr. Cuthbertfon’s knowledge. 
An account of this inftrument, and of lbme experiments 
. performed 
