MECHANICS. 
if it were at liberty to move, It would be drawn back till 
the flat furface was reftored. Let us fuppofe a fmall flick 
of piece of g'.at's to be loofe within the tube, and to pro¬ 
trude into the vacant fpace beyond the furface of the al¬ 
cohol. The fluid will be attrafted alfo by this glafs, and 
form a concave between its furface and that of the bore 
of the tube ; but the fmall interior piece, being quite at 
liberty to move, will be drawn towards the fpirit fo long 
as the attractive force poflefles any activity 5 that is, fo 
long as any additional fluid hangs round the glafs; or, in 
other words, until the end of the flick of glafs is even 
with the furface. Whence it is feen, that the fmall piece 
of glafs will be refilled in any aftion that may tend to 
protrude it beyond the furface of the fluid ; and, if this 
refinance be greater than the force required to Aide it 
along in the tube (as in faCt it is), the piece mud be Aided 
along as the alcohol contrafts, fo as always to keep the 
piece within the fluid. And this fa ft is accordingly ob¬ 
ferved to take place.” Nich. jour. N° 47. N. S. 
Mr. Leflie, well known for his ingenious “Experimen¬ 
tal Enquiry into the Nature and Propagation of Heat, M 
has invented a differential thermometer for the meafurement 
of minute variations of temperature. It confilts of two 
tubes, each terminating in a fmall bulb of the fame dimen¬ 
sions, joined by the blow-pipe, and bent in the form of a 
U, a fmall portion of dark-coloured liquor having pre- 
vioully been introduced into one of the balls. After 
many trials, the fluid belt adapted to the purpofe was 
found to be a folution of carmine in concentrated ful- 
phuric acid. By managing the included air with the heat 
of the hand, this’red liquor is made to ftand at the required 
point of the oppofite ’tube. This is the zero of a fcale 
fattened to that tube, and divided into equal parts above 
and below that point. The instrument is then fixed on a 
Hand. It is manifeft, that when the liquor is at relt, or 
points at zero, the column is preffed in oppofite direc¬ 
tions by two portions of air equal in elafticity, and con¬ 
taining equal quantities of caloric. Whatever heat, then, 
may be applied to the whole inftrument, provided both 
bulbs receive jt in the fame degree, the liquor mult re¬ 
main at reft. But, if the one ball receives the flighted: 
excefs of temperature, the air which it contains will be 
proportionally expanded, and will pufli the liquid again ft 
the air in the other bulb with a force, varying as the dif¬ 
ference between the temperatures of thole two portions 
of air; thus the equilibrium will be deftroyed, and the 
fluid will rife in the oppofite tube. The degrees .of the 
fcale through which it pnffea will mark the fucceflive 
augmentations in the temperature of the ball which is 
expofed to the greateft heat. So that this inftrument is a 
balance of extreme delicacy for comparing the tempera¬ 
tures of its two fcales. 
The thermometers hitherto defcribed are very limited 
in their extent. They indeed point out to us the lowelt 
degrees.of heat which are commonly obferved even in 
cold climates 5 but they by no means reach to thofe de¬ 
grees of heat which are very familiar to us. The mer¬ 
curial thermometer extends no farther than to 600 of 
Fahrenheit’s fcale, the heat of boiling mercury; but we 
are lure that the heat of folid bodies, when heated to ig- 
- nition, or till they emit light, far exceeds the heat of 
boiling mercury. 
In order to remedy this deleft, fir Ifaac Newton, whofe 
c'enius overcame thofe obftacles which ordinary minds 
could not approach, attempted by an ingenious experi¬ 
ment to extend the fcale to any degree required. Hav¬ 
ing heated a mafs of iron red-hot, and expofed it to the 
cold air, he obferved the time wiiich elapfed till it became 
cold, or of the fame temperature with the air ; and, when 
the heat fo far decreafed that he could apply feme known 
meafure (as a thermometer) to it, he obferved the degrees 
of heat loft in given times ; and thence drew the general 
conclulion, that the quantities of heat loft in given fmall 
fuaces are always proportional to the heat remaining in 
the body, reckoning the heat to be the excels by which 
Vol. XIV. No. 1003* 
73S 
it is warmer than the ambient air. So that, taking the 
number of minutes which it took to cool after it came to 
a determined point in an arithmetical progreflion, the de¬ 
crements of the heat of the iron would be continually pro¬ 
portional. Having by this proportion found out the de¬ 
crements of heat in a given time after it came to a known ' 
point, it was eafy, by carrying upwards the fame propor¬ 
tion tothebeginning of its cooling, to determine thegreatelt 
heat which the body had acquired. This proportion of 
fir Ifaac’s was found by Dr. Martineto be fomewhat inac¬ 
curate. The heat of a cooling body does not decreafe ex- 
aftly in proportion to that which the body retains. As 
the refult of many obfervations, he found that two kinds 
of proportion took, place, an arithmetical as well as the geo¬ 
metrical proportion which fir Ifaac Newton had adopted ; 
namely, that the decrements of heat were partly propor¬ 
tional to the times (that is, that quantities of heat are loft 
in equal times), as well as partly in proportion to the re¬ 
maining heat; and that if thefe two are added together the 
rule will be lufficientiy accurate. By the geometrical 
proportion which fir Ifaac Newton adopted he difeovered 
tlie heat of metals red-hot or in fufion. 
The method, purfued by fir Ifaac Newton, was fufiicient 
to form a (bale of high degrees of heat, but was not con¬ 
venient for practical purpofes. Upon thefe principles, 
therefore, M. Mufchenbroek, the celebrated Dutch phi- 
lofopher, and profeffor of natural philofophy at Leyden, 
applied himfelf to form a thermometer which might be ea- 
fily managed. In this he fucceeded, and produced the 
inftrument which is now called a pyrometer, from the 
Greek wvp, fire. The nature and conftruftion of this 
machine may be underftood from the following account. 
If we fuppofe a fmall bar of metal, twelve or fifteen inches 
in length, made faft at one of its extremities, it is obvious 
that if it be dilated by heat it will become lengthened, and 
its other extremity will be pufhed forwards. If this ex¬ 
tremity then be fixed to the end of a lever, the other end 
of which is furnilhed with a pinion adapted to a wheel, 
and if this wheel move a fecond pinion, the latter a third, 
and fo on, it will be evident that, by multiplying wheels 
and pinions in this manner, the la(t will have a veryfenfi- 
ble motion ; fo that the moveable extremity of the fmall 
bar cannot pafs over the hundredth or thoufandth part of 
a line, without a point of the circumference of the laft 
wheel pa fling over feveral inches. If this circumference 
then have teeth fitted into a pinion, to which an index 
is attached, this index will make feveral revolutions, when 
the dilatation of the bar amounts only to a quantity alto¬ 
gether infenfible. The portions of this revolution may be 
meafured on a dial-plate, divided into equal parts; and, by 
means of the ratio which the wheels bear to the pinions, 
the abfolute quantity which a certain degree of heat may 
have expanded the fmall bar can be afeertained ; or, con- 
verfely, by the dilatation of the fmall bar the degree of 
heat which has been applied to it may be determined. 
Such is the conftruftion of Mufchenbroek’s pyrometer. 
It is necelfary to obferve, that a fmall cup is adapted to the 
machine, in order to receive the liquid or fuled matters 
fubjefted to experiment, and in which the bar to be tried 
is immerfed. When it is required to meafure, by this in- 
ftruinent, a confiderable degree of heat, fuch as that of 
boiling oil or fufed metal, fill the cup with the matter to 
be tried, and immerfe the bar of iron into it. The dilata¬ 
tion of the bar, indicated by the index, will point out the 
degree of heat it has a (fumed, and which mult neceflarily 
be equal to that of the matter into which it is immerfed. 
This machine evidently ferves to determine the ratio of 
the dilatation of metals, &c. for, by fubftituting in the 
room of the pyrometric bar other metallic bars of the lame 
length, and then expofing them to an equal degree of heat, 
the ratios of their dilatation will be ftiown by the motion 
of the index, Mufchenbroek has given a table of the ex- 
panfion of the different metals in the fame degree of heat. 
Having prepared cylindrical rods of iron, iteel, copper, 
brafs, tin, and lead, he expofed them firft to a pyrometer 
9 A with 
