upon the furface of a baiin of water, not only
fwims upon the water without being wet, but it
prevents other bodies from being wet which are
plunged into the water through it ; fo that a piece
of money, or other folid body, may be taken from
the bottom of the bafin by the naked hand, with
out wetting the hand ; which is one of the tricks
commonly mown by the jugglers, in the country :
this meal covers the hand, and defcending along
with it to the bottom of the bafin, defends it from
the water. This fubftance has the appearance of
an exceeding fine, light, and very moveable yellow
powder, and it is very imflammable ; fo much fo,
that being blown out of a quill into the flame of a
candle, it flames like gunpowder, and it is made
ufe of in this manner in our theatres for imitating
lightning.
Conceiving that there muft have been a ftrong
attraction between this fubftance and air, and fuf-
pecling, from fome circumftances attending fome
of the foregoing Experiments, that the w T armth of
a covering depends not merely upon the finenefs of
the
in various Sitbftances. 449
the fubftancc of which the covering is formed, and
the difpolition of its parts, but that it arifes in fome
meafure from a certain attraction between the fub-
fiance and the air which fills its interftices, I thought
that an Experiment wiihfemen lycopodii might pof-
fibly throw fome light upon this matter j and in
this opinion I was not altogether miftaken, as will
appear by the refults of the three following Expe
riments.
The bulb of the thermometer furrounded by
256 grs. of femen lycopodii.
Heat loft.
Cooled.
Cooled.
Heat
acquired.
Heated.
Exp. No. 30.
Exp. No. a8.
Exp. No. 29.
70
..
___
j_
60
1 4 6"
157"
10
230-
50
162
1 60
20
68
40
175
170
30
63
30
20 9
203
40
76
20
284
288
50
121
10
502
513
60
3 l6
70
1585
Total times.
1478
I 49 I
2459
In the laft Experiment (No 30) the refult of
which was fo very extraordinary, the inftrument
was cooled to o in thawing ice, after which it was
plunged fuddenly into boiling water, where it re
mained till the inclofed thermometer had acquired
the Heat of 70, which took up no lefs than 2456
feconds, or above 40 minutes ; and it had remained
in
450 Of the Propagation of Heat
in the boiling water full a minute and an half before
the mercury in the thermometer mowed the leaft
lign of riling. Having at length been put into mo
tion, it rofe very rapidly 40 or 50 degrees, after which
its motion gradually abating became {b flow, that
it took up 1585 feconds, or fomething more than
26 minutes, in riling from 60 to 70, though the
temperature of the medium in which it was placed
during the whole of this time was very nearly 80 j
the mercury in the barometer {landing but little
ihort of 27 Paris inches.
All the different fubftances which I had yet made
ufe of in thefe Experiments for furrounding or
covering the bulb of the thermometer, fluids ex-
cepted 5 had,in a greater, or in alefs degree confined
the Heat, or prevented its palling into or out of the
thermometer fo rapidly as it would have done, had
there been nothing but air in the glafs globe, in the
centre of which the bulb of the thermometer was
fufpended. But the great queftion is, how, or in
what manner, they produced this effed ?
And firil, it was not in confequence of their own
non -con dueling powers, (imply confidered ; for, if
inftead of being only bad conductors of Heat, we
fuppofe them to have been totally impervious to
Heat, their volumes or folid contents were fo ex
ceedingly fmall in proportion to the capacity of the
globe in which they were placed, that, had they
had no effect whatever upon the air filling their in
ter dices, that air would have been fufficient to have
conducted ail the Heat communicated, in lefs time
than was aclually taken up in the Experiment.
The
in various Sub/tances. 45*
The diameter of the globe being 1,6 inches, its
contents amounted to 2,14466 cubic inches j and
the contents of the bulb of the thermometer being
only 0,08711 of a cubic inch, (its diameter being
0^55 of an inch) the fpace between the bulb of the
thermometer and the internal furface of the globe
amounted to 2,14466 0,087 11:122,05755 cu bic
inches ^ the whole of which fpace was occupied
by the fubflances by which the bulb of the ther
mometer was furrounded in the Experiments in
queftion.
But though thefe fubftances occupied this fpace,-
they were far from filing it ; by much the greater
part of it being filled by the air which occupied the
interftices of the fubftances in queftion. In the
Experiment No. 4, this fpace was occupied by 1 6
grains of raw iilk ; and as the fpecific gravity of raw
filk is to that of water as 1 734 to 1000, the volume
of this filk was equal to the volume of 9,4422
grains of water ; and as i cubic inch of water weighs
253,185 grains, its volume was equal to -^1-1-
=0,037294 of a cubic inch; and, as the fpace
it occupied amounted to 2,05755 cubic inches, it
appears that the filk filled no more than about ~j
part of the fpace in which it was confined, the
reft of that fpace being filled with air.
In the Experiment No. i, when the fpace between
the bulb of the thermometer and the glafs globe,
in the centre of which it was confined, was filled
with nothing but air, the time taken up by the
thermometer in cooling from 70 to 10 was
Voi<. II. M m m 576
452 Of the Propagation of Heat
576 feconds ; but in the Experiment No. 4, when
this fame fpace was filled with 54 parts air, and
i part raw filk, the time of cooling was 1284 fec
onds.
Now, fuppofing that the filk had been totally in
capable of conducting any Heat at all, if we fuppofe,
at the fame time, that it had no power to prevent
the air remaining in the globe from conducting it,
in that cafe its prefence in the globe could only have
prolonged: the time of cooling in proportion to the
quantity of the air it had difplaced to the quantity
remaining, that is to fay, as i is to 54, or a little
more than 10 feconds. But the time of cooling
was actually prolonged 708 feconds (for in the
Experiment No. i, it was 576 feconds, and in the
Experiment No. 4, it was 1284 feconds, as has juft
been obferved ;) and this mows, that the filk not
only did not conduct the Heat itfelf, but that it pre
vented the air by which its interfaces were' filled
from conducting it ; or, at leaft, it greatly weakened
its power of conducting it.
The next queftion which arifes is, how air can
be prevented from conducting Heat ? and this
neceffarily involves another, which is, how does
air conduct Heat ?
If air conducted Heat, as it is probable that the
metals and water, and all other folid bodies and
unelaftic fluids conduct it, that is to fay, if its par
ticles remaining in their places, the Heat paffed
from one particle to another, through the whole
rnafs.
in various Sub/dances. 453
mafs, as there is no reafon to fuppofe that the pro
pagation of Heat is neceflarily in right lines, I can
not conceive how the interpoiition of fo fmall a
quantity of any foiid body as - part of the volume
of the air could have effected fo remarkable a
diminution of the conducting power of the air,
as appeared in the Experiment (No. 4) with raw
filk, above-mentioned.
If air and water conducted Heat in the fame
manner^ it is more than probable that their con
ducting powers might be impaired by the fame
means ; but when I made the Experiment with
water, by filling the glafs globe, in the centre of
which the bulb of the thermometer was fufpended,
with that fluid, and afterwards varied the Experi
ment, by adding 16 grains of raw filk to the
water, I did not find that the conducting power
of the water was fenfibly impaired by the prefence
of the filk.*
But we have jufl feen that the fame filk, mixed
with an equal volume of air, diminimed its con-
dueling power in a very remarkable degree ; con-
fequently, there is great reafon to conclude that
water and air conduct Heat in a different manner.
But the following Experiment, I think, puts
the matter beyond all doubt.
It
* The Experiment here mentioned was made in the year i;8/j
tmt the refult of a more careful invefligation of the fubje& has fmce
Ihown that Heat is not propagated in water in the manner here
fuppofed. (SeeElTay VII.)
454 Of the Propagation of Heat
It is well known, that the power which air
poffeffes of holding water in folution is augmented
by Heat, and diminifhed by cold, and that, if hot
air is faturated with water, and if this air is after
wards cooled, a part of its water is neceflarily de-
pofed.
I took a cylindrical bottle of very clear tranf-
parent glafs, about 8 inches in diameter, and 12
inches high, with a fhort and narrow neck, and
fufpending a fmall piece of linen rag, moderately
wet, in the middle of it, I plunged it into a large
veffel of water, warmed to about 100 of Fahren
heit's thermometer, where I fufiered it to remain
till the contained air was not only warm, but thor
oughly faturated with the moifture which it attract
ed from the linen rag, the mouth of the bottle being
well flopped up during this time with a good cork;
this being done, I removed the cork for a moment,
to take away the linen rag, and flopping up the
bottle again immediately, I took it out of the warm
water, and plunged it into a large cylindrical jar,
about 12 inches in diameter, and 16 inches high,
containing juft fo much ice-cold water, that, when
the bottle was plunged into it, and quite covered
by it, the jar was quite full.
As the jar was of very fine tranfparent glafs, as
well as the bottle, and as the cold water contained
in the jar was perfectly clear, 1 could fee what
pafled in the bottle mofl diflindly ; and having
taken care to place the jar upon a table near the
window,
in various Subftances. 455
window, in a very favourable light, I fet myfelf to
obferve the appearances which mould take place,
with all that anxious expectation which a conviction
that the refult of the Experiment muft be decifive,
naturally infpired.
I was certain, that the air contained in the bottle
could not part with its Heat, without at the fame
time, that is to fay, at the fonts moment, and in the
fame place, parting with a portion of its water ; if,
therefore, the Heat penetrated the mafs of air from
the centre to the furface, or pa/fed through it from
particle to particle, in the fame manner as it is prob*
able that it paffes through water, and all other
uneiaftic fluids,* by far the greateft part of the air
contained in the bottle would part with its Heat,
when nrt aftually in contaft with the glafs, and a
proportional part of its water being let fall at the
fame time, and in the fame place, would neceffarily
defcend in the form of rain ; and, though this rain
might be too fine to be viiible in its defcent, yet I
was fure I fhould find it at the bottom of the bottle,
if not in vifible drops of water, yet in that kind of
cloudy covering which cold glafs acquires from a
contact with hot fleam or watery vapour.
But if the particles of air, inflead of communi
cating their Heat from one to another, from the
centre to the furface of the bottle, each in its turn,
and for itfelf, came to the furface of the bottle, and
there
* This opinion refpe<5Ung the manner in which Heat is propagated
in water, and other unelaftic fluids, was afterwards found to be erro
neous, as has been lho\vn in the preceding Eff?.y.
456 Of the Propagation of Heat
there depofited its Heat and its water, I concluded
that the cloudinefs occafioned by this depofit of
water would appear all over the bottle, or, at leaft,
not more of it at the bottom than at the iides,
but rather lefs ; and this I found to be the cafe
in fact.
The cloudinefs firft made its appearance upon
the fides of the bottle, near the top of it ; and
from thence it gradually fpread itfelf downwards,
till, growing fainter as it defcended lower, it was
hardly vifible at the diftance of half an inch from
the bottom of the bottle ; and upon the bottom
itfelf, which was nearly flat, there was fcarcely the
fmalleft appearance of cloudinefs.
Thefe appearances, I think, are eafy to be ac
counted for. The air immediately in contact with
the glafs being cooled, and having depciited a part
of its water upon the furface of the glafs, at the
fame time that it communicates to it its Heat,
Hides downwards by the iides of the bottle in con-
fequence of its increafed fpecific gravity, and,
taking its place at the bottom of the bottle, forces
the whole mafs of hot air upwards ; which, in its
turn coming to the iides of the bottle, there de-
pofits its Heat and its water, and afterwards bend
ing its courfe downwards, this circulation is con
tinued till all the air in the bottle has acquired the
exact temperature of the water in the jar.
From hence it is clear why the firft appearance
of condenfed vapour is near the top of the bottle,
as alfo why the greateft collection of vapour is in
that
in various Sub/lances. 457
that part, and that fo very fmall a quantity of it is
found nearer the bottom of the bottle.
This Experiment confirmed me in an opinion
which I had for fome time entertained, that, though
the particles of air individually, or each for itfelf,
are capable of receiving and Iran/porting Heat, yet
air in a quiefcent ftate, or as a fluid wliofe parts are
at reft with refpect to each other, is not capable of
conducting it, or giving it a paffage ; in fhort, that
Heat is incapable of paffing through a mafs of alr^
penetrating from one partkle of it to another, and
that it is to this circumftance that its non-conducl-
ing power is principally owing.
It is alfo to this circumftance, in a great meafure,
that it is owing that its non-conducting power, or
its apparent warmth when employed as a covering
for confining Heat, is fo remarkably increafed upon
its being mixed with a fmall quantity of any very
fine, light, folid fubftance, fuch as the raw filk,
fur, Eider down, &c. in the foregoing Experi
ments : for as I have already obferved, though
thefe fubftances, in the very fmall quantities in
which they were made ufe of, could hardly have
prevented, in any confiderable degree, the air from
conducting, or giving a paffage to the Heat, had it
been capable of paffing through it, yet they might
very much impede it in the operation of tranfport-
ing it.
But there is another circumftance which it is
neceffary to take into the account, and that is the
attraction which fubfifts between air and the bodies
above-
45 8 Of tLs Propagation of Heat
above-mentioned, and other likefubftances, confti-
tuting natural and artificial clothing. For, though
the incapacity of air to give a paffage to Heat in
the manner folid bodies permit it to pafs through
them, may enable us to account for its warmth
tinder certain circumftanees, yet the bare admif-
fion of this principle does not feem to be fufficierit
to account for the very extraordinary degrees of
warmth which we find in furs and in feathers, and
in various other kinds of natural and artificial
clothing ; nor even that which we find in fnow 5
for if we fuppofe the particles of air to be at liberty
to carry c^'the Heat wKich thefe bodies are meant
to confine, without any other obftru&ion or hin
drance than that arifing from their w inertia^ or
the force neceffary to put them in motion, it feems
probable that the fuccefiion of frefh particles of cold
air, and the confequent lofs of Heatj would be
much more rapid than we find it to be in fact;
That an attraction, and a very ilrong one, ac
tually fubfifts between the particles of air, and the
fine hair or furs of beafts, the feathers of birds,
wool, &c. appears by the obftinacy with which
thefe fubftances retain the air which adheres to
them, even when immerfed in water, and put
under the receiver of an air-pump ; and that this
attraction is eflential to the warmth of thefe bodies,
I think is very eafy to be demonftrated.
In furs, for inftance, the attraction between the
particles of air, and the fine hairs in which it is
concealed, being greater than the increafed elaf-
tkitV,
in 'various Subftances, 459
ticity, or repulfion of thofe particles with regard
to each other, arifing from the Heat communicated
to them by the animal body, the air in the fur,
though heated, is not eafily difplaced ; and this
coat of confined air is the real barrier which de
fends the animal body from the external cold.
This air cannot carry off the Heat of the animal,
becaufe it is itfelf confined, by its attraction to the
hair or fur ; and it tranfmits it with great difficul
ty, if it tranfmits at all, as has been abundantly
fhewn by the foregoing Experiments.
Hence it appears why thofe furs which are the
fineft, longeft, and thickeft, are likewife the warm-
eft ; and how the furs of the beaver, of the otter,
and of other like quadrupeds which live much in
water, and the feathers of water-fowls, are able to
confine the Heat of thofe animals in winter, not-
withftanding the extreme coldnefs and great con-
dueling power of the water in which they fwim.
The attraction between thefe fubftanc.es, and the
air which occupies their interftices, is fo great, that
this air is not diflodged even by the contact of
water, but remaining in its place, it defends the
body of the animal at the fame time from being
wet, and from being robbed of its Heat by the fur-
rounding cold fluid, and it is pofiible that the
preffure of this fluid upon the covering of air con
fined in the interftices of the fur, or feathers, may
at the fame time increafe its warmth, or non-con
ducting power, in fuch a manner that the animal
may not, in fact, lofe more heat when in water, than
e II, N n n when
46 Of the Propagation of Heat
when in air : for we have feen by the foregoing
Experiments, that, under certain circumftances,
the warmth of a covering is increafed, by bringing
its component parts nearer together, or by in-
creafing its denfity even at the expenfe of its thick-
nefs. But this point will be further invefligated
hereafter.
Bears, wolves, foxes, hares, and other like
quadrupeds, inhabitants of cold countries, which
do not often take the water, have their fur much
thicker upon their backs than upon their bellies.
The heated air occupying the interftices of the
hairs of the animal tending naturally to rife up
wards, in confequence of its increafed elafticity,
would efcape with much greater eafe from the
backs of quadrupeds than from their bellies, had
not Providence wifely guarded againft this evil by
increafing the obftructions in thofe parts, which
entangle it and confine it to the body of the animal.
And this, I think, amounts almoft to a proof of
the principles affumed relative to the manner in
which Heat is carried off by air, and the caufes of
the non-conducting power of air, or its apparent
warmth, when, being combined with other bodies,
it acts as a covering for confining Heat.
The fnows which cover the furface of the earth
in winter, in high latitudes, are doubtlefs defigned
by an all-provident Creator as a garment to defend
it againft the piercing winds from the polar re
gions, which prevail during the cold feafon.
Thefe
in various Sub/lances. 461
Thefe winds, notwhhftanding the vaft tracls of
continent over which they blow, retain their iharp-
nefs as long as the ground they pafs over is covered
with fnow ; and it is not till meeting with the
ocean they acquire, from a contacl with its waters,
the Heat which the fnows prevent their acquiring
from the earth, that the edge of their coldnefs is
taken off, and they gradually die away and are
loft.
The winds are always found to be much colder
when the ground is covered with mow than when
it is bare, and this extraordinary coldnefs is vul
garly fuppofed to be communicated to the air by
the fnow ; but this is an erroneous opinion ; for
thefe winds are in general much colder than the
fnow itfelf.
They retain their coldnefs, becaufe the fnow pre
vents them from being warmed at the expenfe of
the earth ; and this is a ftriking proof of the ufe
of the fnows in preferving the Heat of the earth
during the winter in cold latitudes.
It is remarkable that thefe winds feldom blow
from the poles directly towards the equator, but
from the land towards the fea. Upon the eaftern
coaft of North America the cold winds come from
the north-weft ; but upon the weftern coaft of
Europe they blow from the north-eaft.
That they mould blow towards thofe parts where
they can moft eafily acquire the Heat they are in
fearch of, is not extraordinary ; and that they
ihould gradually ceafe and die away, upon being
warmed
462 Of the Propagation of Heat
warmed by a contact with the waters of the ocean,
is likewife agreeable to the nature and caufes of
their motion ; and if I might be allowed a conjec
ture refpecling the principal ufe of the feas, or the
reafon why the proportion of water upon the fur-
face of our globe is fo great, compared to that of
the land, it is to maintain a more equal tempera
ture in the different climates, by heating or cool
ing the winds which at certain periods blow from
the great continents.
That cold winds actually grow much milder
upon pafling over the fea, and that hot winds are
refrefhed by a contact with its waters, is very cer
tain ; and it is equally certain that the winds from
the ocean are, in all climates, much more tempe
rate than thofe which blow from the land.
In the iflands of Great Britain and Ireland, there
is not the leaft doubt but the great mildnefs of the
climate is entirely owing to their feparation from
the neighbouring continent by fo large a tract of
fea ; and in all fimilar fituations, in every part of
the globe, fimilar caufes are found to produce
fimilar effects.
The cold north-weft winds, which prevail upon
the coaft of North- America during the winter, fel-
dom extend above 100 leagues from the more, and
they are always found to be lefs violent, and lefs
piercing, as they are further from the land.
Thefe periodical winds from the continents of
Europe and North America, prevail moft towards
the end of the month of February, and in the month
of
in various Subftances. 463
of March ; and I conceive that they contribute
very effentially towards bringing on an early fpring,
and a fruitful fummer, particularly when they are
very violent in the month of March, and if at that
time the ground is well covered with fnow. The
whole atmofphere of the polar regions being, as it
were, tranfported into the ocean by thefe winds, is
there warmed and faturated with water : and, a
great accumulation of air upon the fea being the
neceflfary confequence of the long continuance of
thefe cold winds from the fhore, upon their ceafing
the warm breezes from the fea neceffarily com
mence, and, fpreading themfelves upon the land
far and wide, ailift the returning fun in difmantling
the earth of the remains of her winter garment,
and in bringing forward into life all the manifold
beauties of the new born year.
This warmed air which comes in from the fea,
having acquired its Heat from a contact with the
ocean, is, of courfe, faturated with water ; and
hence the warm mowers of April and May, fo
neceffary to a fruitful feafon.
The ocean may be coniidered as the great refer-
voir and equalizer of Heat ; and its benign influ
ences in preferving a proper temperature in the
atmofphere operate in all feafons and in all cli
mates.
The parching winds from the land under the
torrid zone are cooled by a contact with its waters,
and, in return, the breezes from the fea, which at
certain hours of the day come in to the mores in
almoft all hot countries, bring with them refrefh-
46 4 Of the Propagation of Heat
merit, and, as it were, new life and vigor both to
the animal and vegetable creation, fainting and
melting under the exceffive Heats of a burning
fun. What a vafl tract of country, now the moft
fertile upon the face of the globe, would be abfo-
lutely barren and uninhabitable on account of the
excefnve Heat, were it not for thefe refrefhing
fea-breezes ! And is it not more than probable,
that the extremes of heat and of cold in the dif
ferent feafons in the temperate and frigid zones
would be quite intolerable, were it not for the in
fluence of the ocean in preferving an equability of
temperature ?
And to thefe purpofes the ocean is wonderfully