H E 
■contiguous fhell of air, feel at once the pulfatory influ¬ 
ence i the other particles, which conftitute the general 
jnafs, probably imbibe their fhare of heaty and palfively 
obey the impreflion of their augmented elaliicity.” 
To determine the law of the rate of cooling with regard 
to the temperature, recourfe was again had to experi¬ 
ment and Mr. Leflie, by repeated trials, found that 
at the temperatures io°, 40°, 70 0 , the rates of cooling 
for a metallic lurface- were 2, 3, 4 ; and for a painted 
fnrface, 4, 5, 6 . Consequently, if we increafe the tem¬ 
perature by 30 0 , we increafe the rate of cooling by 1, 
the pulfatory energy remaining the fame. Hence, de¬ 
noting_by P the pulfatory energy, and by an unknown 
quantity .v fome unknown refrigerating caufe, inde¬ 
pendent of the nature of the heated furface,. we have 
temperature being io°. Confequently, 
P—i, and * = 2 — §■= if-, which two numbers f, if-, 
therefore folely denote two effedts, or Ioffes of neat ; 
one due to puliation, a known caufe, the other to caufes 
not yet afcertained. Suppofe, now, the effedt, if to be 
due to two caufes, y, z, one of which. (_y ) varies as the 
excefs of temperature; then, finceat 30 the value of_y— 1, 
at 10 it —f. Confequently, i§=z±-\-z, or z= T % T , the 
meafure ot a third unknown refrigerating caufe. This 
third caufe the author thus afligns : “ The portion of 
heat thus confumed is moll certainly not annihilated : 
neither is it franfported to a diftance, by any fpecies of 
elaftic motion e,xcfied in the encircling fluid. It is, 
therefore, abforbed by the contiguous Ihell of matter, 
and afterwards flowly ditTufed through the extended 
mafs. Air is Hill the foie medium by which heat en¬ 
deavours to maintain the balance among remote or “de¬ 
tached bodies; but here its operation is of a paflive na¬ 
ture, and it receives and conveys the calorific impref- 
fions through its fubftance in the fame manner as a bar 
of iron or any folid’ material. 
“ This completes the analyfis of the refrigerating ac¬ 
tion of air. There are four diHindi inodes in which it 
produces the effedt: three of thefe are al ways conjoined, 
and the fourth only throws in its occafional influence. 
They all confpire to the fame end, but their relative 
fiiares of operation are various and mutable. One fource 
of communication depends on the quality of the heated 
lurface, another on its elevation of temperature, a third 
on the permanent condudting difpofition of the air, and 
the lad arifes from the celerity of impulfe by which that 
adtive fluid may chance to De affedted. The continual 
afcent of the hot, and confequently rarefied, a'.r muff con¬ 
tribute in fome degree, though indiredtly, to accelerate 
the effedt; for it is evident, that the flagnation of a 
warm encircling atmofphere would debilit ice the opeia- 
tion of the combined refrigerating caufes.” Thefe re - 
fults were drawn from experiments made with a hollow 
fphere fix inches in diameter, filled with boiling, water ; 
from whence the portions of heat difcharged every mi¬ 
nute are thus reprefented : 
By abdudtion, the 524*1485111, 
By recefiion, the hy 217.14*725111, and 
By puliation, the 2533*385111 for metallic, 
and the 316'673d for the lurface o'f paper. 
Such is Mr. Leflie’s beautiful analyfis of the procefs 
of the cooling of heated veffels in elaftic fluids, or at- 
mofpheric air :—the real caufe of the difference in the 
fates of cooling is pulfation. This theory the author 
applies to explain an anomaly that fliowed itfelf in one 
of his experiments above recited, viz. when the painted 
fnrface of the canifter was turned towards the refledtor, 
and the bulb of the thermometer was covered with tin¬ 
foil, "the effedt indicated was 22 0 : when the metal fur- 
face was ufed, and the bulb was covered, the effedt in¬ 
dicated was only 12 0 : the procedure then being invert¬ 
ed, the fame eff^dts did not take place. The caufe is 
thus afligned : the metallic bulb cools flower than the 
Vol. IX. No. 5S4. 
A T. 281 
glafs, and confequently is proportionally more affedted by 
the fame impreflion of heat. 
This ingenious philofopher next proceeds to invefti- 
gate the phenomena of refrigeration in non-elaftic fluids; 
and the fluid which he choofes for the fubjedt of his ex¬ 
periments, is water. A hot body placed in water is 
cooled chiefly by two only of the refrigerating caufes 
which act in an elaftic medium ; a portion of the heat is 
uniformly abforbed by the furrounding water, and con- 
dudted through the internal mafs, in the fame manner 
as if it were congealed into folid ice ; and the remain¬ 
ing portion is difcharged by the flow recefiion of the 
heated particles. In water, however, the latv of this 
fecond refrigerating caufe will not be the fame as in air, 
in which it varies limply as the excefs of temperature ; 
and the proof that the law is not the' fame is derived 
from experiment. Thus, fiippofe a ball containing hot 
water, and furnifhed with a thermometer, to be im- 
merfed in a water-bath, and the temperature' of the 
bath to be in three experiments o, 30°, 6o°, and the 
correfponding temperatures of the immerfed ball, 20°, 
50 0 , 8o°; then in the fiVft;experiment the thermometer 
links 10 degrees in fix minutes ; in the fecond 10 de¬ 
grees in 3^ minutes; and in the third, id degrees in two 
minutes; but in air the expreffion for the time b'e- 
ing t — b. | log. f_zi — log* | muft evidently be 
the fame, fince h and H are the fame whether they 
= 10 — o and 20 — o, 012=40 — 30 and 50 — 30, or 2= 
70 — 60 and 80—60. The author attributes this increafe 
in tlie rate of cooling to a property inherent in water* 
namely that of expanding with a rapid acceleration on 
Receiving equal acceflions of heat. If e be the dilatation, 
and h the temperature, then de CC h. d h, confequently 
e CC h 2 , reckoning the commencement of expanfion from 
the point of congelation ; and here Mr. Lellie takes oc- 
eafion to controvert an- idea which has recently been an¬ 
nounced as a curious and remarkable .fait, that the mini- 
num of the expanfion of water takes place, not at tne 
point of congelation-, but a few degrees above that point; 
On a fubjedt of fitch intereft, we lhall give the author’s 
own words : “ I have referred the commencement of 
expanfion in water to the point of congelation. But.it 
now feems generally fuppofed that water is contradted 
into the fmalleft volume about five or fix degrees above 
zero, and, in its defeent beyond this flat ionary limit, 
again undergoes a flight dilatation. I am dilpofed, 
however, to queftion the. accuracy of a principle fo dif- 
cordant and anomalous. I11 fadt the experiments on 
which it is grounded, though fomewhat varied in their 
plan, never give thq. true expanfions of water, but only 
the differences between thole expanfions and the con e- 
fponding<expanfions of glafs. Having filled a thin glafs 
ball, terminating in a fine tubular ftem, with diftilled 
water, and cooled the whole down to the point of con¬ 
gelation, I plunged it into a large bath, wliofe tempe¬ 
rature was four or five degrees above zero. The water 
in the ftem funk at firft confiderably, owing evidently 
to the dilatation of the glafs, and, by confequence, the 
enlarged capacity of the ball; blit it then role a feniible 
fpace, which muft be aferibed to the expanfion of the 
water itfelf. In like manner, when the procedure is re- 
verfed, and the ball, heated up a few degrees, is plunged 
into a bath at the point of congelation, the water riles 
in the ftem as the ball contradts, and then, by its own 
contradtion, partially fubfides. The dilatation of glafs 
by heat is indeed fo very frnall, that in'mod cafes it may 
be fafely difregarded. But the rate with which water 
contradls is perpetually diminilhingas the heat declines ; 
and therefore, at fome particular poirtt, this effedt is 
exadtly counteracted by the oppofite contradtion of the 
glafs, and beyond it the latter muft predominate. Nor 
is it difficult to determine, at lead theoretically, the po- 
fition of that minimum , or limit of apparent condenfation. 
4 C Water 
