But it is not only the finenefs of the parts of a
fubftance, and its being a non-conductor, which
render it proper to be employed in the formation
of covering to cenfine Heat ; there is ftill another
property more occult, which feems to have great
influence in rendering fome fubftances better fitted
for this ufe than others ; and this is a certain at
traction which fubfifts between certain bodies and
air. The obftinacy with which air adheres to the
fine fur of beafts and to the feathers of birds, is
well known ; and it may eafily be proved that this
attraction muft aflift very powerfully in preventing
the motion of the air concealed in the interftices of
thofe fubftancesj and confequently in impeding the
paflage of Heat through them.

Perhaps there may be another ftill more hidden
caufe which renders one fubftance better than an-
other-for confining Heat. I have fiiown by a direct
and unexceptionable Experiment, that Heat can
pafs through the Torricellian vacuum,* though
with rather more difficulty than in air (the con
ducting



* See my Experiments on Heat, publifbcd In the Philofophical
Tranfaftions, vol. Ixxvi-



and the Economy of Fuel. 59

duffing power of air being to that of a Torricel
lian vacuum as 1000 to 604, or as ro to 6, very
nearly) but if Heat can pafs where there is no air,
it muft in that cafe pafs by a medium more fubtile
than air ; a medium which moft probably per
vades all folid bodies with the greateft facility, and
which muft certainly pervade either the glafs or the
mercury employed in making a Torricellian va
cuum.

Now, if there exifts a medium more fubtile than
air, by which Heat may be conducted, is it not pof-
fible that there may exift a certain affinity between
that medium and fenfible bodies ? a certain attrac
tion or coheiion by means of which bodies in gen
eral, or fome kinds of bodies in particular, may,
fome how or other, impede this medium in its ope
rations in conducing or tranfportingHeat from one
place to another ? It appeared from the refult of
feveral of my Experiments, of which I have given
an account in detail in my paper before mentioned,
publiflied in the year 1786 in the Lxxvith Vol. of
the Philofophical Tranfaclions, that the conducing
power of a Torricellian vacuum is to that of air as
604 to 1000 t but I found by a fubfequent Ex
periment, (fee my fecond Paper on Heat, publiflied
in the Philofophical Traniactions for the year
1792) -that 55 parts in bulk of air, with i part,
of fine raw filk, formed a covering for confining
Heat, the conducting power of which was to that
of air as 576 to 1284 ; or as 448 to 1000. Now,
from the refult of this laft mentioned Experiment,

it



60 Of the Management of Fire,

it fliould feem that the introduftion into the fpacc
through which the Heat paffed, of fo fmall a quan
tity of raw filk as - part of the volume, or capa
city of that fpace, rendered that fpace (which now
contained 55 parts of air and i part of filk) more
impervious to Heat than even a Torricellian va
cuum. The filk muft therefore not only have
completely deflroyed the conducting power of the
air, but muft alfo at the fame time have very fenfi-
bly impaired that of the etherial fluid which pro
bably occupies the interftices of air, and which
ferves to conduct Heat through a Torricellian va
cuum : For a Torricellian vacuum was a better
conductor of Heat, than this medium, in the pro
portion of 604 to 448. But I forbear to enlarge
upon this fubject, being fenlible of the danger of
reafoning upon the properties of a fluid whofe ex-
iftence even is doubtful ; and feeling that our
knowledge of the nature of Heat, and of the man
ner in which it is communicated from one body to
another, is much too imperfect and obfcure to en
able us to purfue thefe fpeculations with any prof-
peel: of fuccefs or advantage.

Whatever may be the manner in which Heat is
communicated from one body to another, I think it
has been fufficiently proved that it paffes with great
difficulty through confined air ; and the knowledge
of this facl: is very important, as it enables us to
take our meafures with certainty and with facility
for confining Heat, and directing its operations to

ufeful purpofes.

But



and ths Economy of Fuel. 61

But atmofphcric air is not the only non-conductor
of Heat. All kinds of air, artificial as well as na
tural, and in general all elaftic fluids, fteam not ex-
cepted, feem to poffefs this property in as high a de
gree of perfection as atmofpheric air.

That fteam is not a conductor of Heat, I proved
by the following Experiment : A large globular
bottle being provided, of very thin and very tranf-
parent glafs, with a narrow neck, and its bottom
drawn inward fo as to form a hollow hemifph^re
about 6 inches in diameter ; this bottle, which was
about 8 inches in diameter externally, being filled
with cold water, was placed in a mallow diili, or
rather plate, about 10 inches in diameter, with a
flat bottom formed of very thin meet brafs, and
raifed upon a tripod, and which contained a fmall
quantity (about T a - of an inch in depth) of water ;
a fpirit lamp being then placed under the middle of
this plate, in a very few minutes the water in the
plate began to boil, and the hollow formed by the
bottom of the bottle was filled with clouds of fteam,
which, after circulating in it with furprifmg ra
pidity 4 or 5 minutes, and after forcing out a good
deal of air from under the bottle, began gradually
to clear up. At the end of 8 or i o minutes (when,
as I fuppofed, the air remaining with the fteam in
the hollow cavity formed by the bottom of the
bottle, had acquired nearly the fame temperature
as that of the fteam) thefe clouds totally difap-
peared ; and, though the water continued to boil
with the utmoft violence, the contents of this hol
low



6z Of the Management of Fire,

low cavity became fo perfectly invifible, and fo
little appearance was there of fleam, that, had it
not been for the ftreams of water which were con
tinually running down its fides, I fhould almoft
have been tempted to doubt whether any fleam
was actually generated.

Upon lifting up for an inftant one fide of the
bottle, and letting in a fmaller quantity of cold
air, the clouds inflantly returned, and continued
circulating feveral minutes with great rapidity, and
then gradually difappeared as before. This Expe
riment was repeated feveral times, and always with
the fame refult ; the fleam always becoming vifible
when cold air was mixed with it, and afterwards
recovering its tranfparency when, part of this air
being expelled, that which remained had acquired
the temperature of the fleam.

Finding that cold air introduced under the bottle
caufed the fleam to be partially condenfed, and
clouds to be formed, I was defirous of feeing what
vilible effecls would be produced by introducing a
cold folid body under the bottle. I imagined that
if fteam was a conductor of Heat, fome part of the
Heat in the fleam pafling out of it into the cold
body, clouds would of courfe be formed ; but I
thought if fleam was a ntn-cenduftof of Heat,
that is to fay, if one particle ofjleam could not com
municate any part of its Heat to its neighbouring par-
ticles, in that cafe, as the cold body could only
affecl the particles of fleam aclually in contacl 'with
it, no cloud would appear ; and the refult of the

Experiment






and the Economy of Fuel. 63

Experiment fhowed that fleam is in fact a non-con-
duff or of Heat ; for, notwithftanding the cold body
ufed in this Experiment was very large and very
cold, being a folid lump of ice nearly as large as
an hen's egg, placed in the middle of the hollow
cavity under the bottle, upon a fmall tripod or
fland made of iron wire ; yet as foon as the clouds
which were formed in confequence of the unavoid
able introduction of cold air in lifting up the bottle
to introduce the ice, were diilipated, which foon
happened, the fleam became fo perfectly tranfparent
and invifible, that not the fmalleft appearance of
cloudinefs was to be feen any where, not even about
the ice, which, as it went on to melt, appeared as
clear and as tranfparent as a piece of the fineft rock
cryflal.

This experiment, which I firft made at Florence,
in the month of November, 1793, was repeated
feveral times in the prefence of Lord Palmerflon,
who was then at Florence, and Monf. de Fontana.*

In

* Tke bottle made ufe of in this Experiment, though It appeared
very large externally, contained but a very fmall quantity of water,
owing to its bottom being very much drawn inwards. As the hol
low cavity under the bottom of the bottle (which, as I juft obferved,
was nearly in the form of a hemifphere, and 6 inches in diameter)
ferved as a receiver for confining the fteam, which rofe from the
boiling water in the plate, it may perhaps bs imagined that a common
glafs receiver in the form of a bell, fuch as are ufed in Pneumatical
Experiments, might anfwer as well as this bottle ; I thought fo my-
felf, but upon making the experiment I found my miflake. A
common receiver will anfwer perfectly well for confining the fleam,
but the glafs foon becomes fo hot that the drops of water which are
formed upon its internal furface, in confequence of the condcnfation
of the fteam, inftead of running down the fides of the receiver in clear

tranfparent



64 Of the Management of Fire,

In thefe Experiments the air was not entirely ex
pelled from under the bottle ; on the contrary, a
confiderable quantity of it remained mixed with
the fteam even after the clouds had totally difap-
peared, as I found by a particular Experiment
made with a view to afcertain that fad ; but that
circumftance does not render the rcfult of this Ex
periment lefs curious, on the contrary ; I think it
tends to make it more furpriling. It fhould feem
that neither the mafs of fteam, nor that of air, were
at all cooled by the body of ice which they fur-
rounded, for if the air had been cooled, (in mafs)
it feems highly probable that the clouds would have
returned.

The remits of thefe Experiments compared with
thofe formerly alluded to, in which I had endeav
oured to afcertain the mod advantageous forms
for boilers, opened to me an entirely new field for
fpeculation and for improvement in the Manage
ment of Fire. They fhowed me that not only cold
air, but alfo hot air, and hot fteam, and hot mix
tures of air and fteam, are non-conduftors of
Heat; confequently that the hot vapour which

rifes

tranfparent dreams, form blotches and ftreaks, which render the glafs
fo opaque that nothing can be feen diftin&ly through it ; and this of
courfe completely fruftrates the main defign of the Experiment ; but
cold water in the bottle keeping the glafs cool, the sondenfation of
the fteam upon the fides of the hollow cavity formed by the bottom
of the bottle, goes on more regularly, and the dreams of water which
are continually running down the fides of the glafs, uniting together,
form one tranfparent flieet of water, by which means every thing that
goes on under the bottle may be diftin&ly feen.



find the Economy of Fuel. 65

rifes from burning Fuel, and even the fame itfelf,
is a non-conduftor of Heat.

This may be thought a bold afiertion, but a little
calm reflection, and a careful examination of the
phenomena which attend the combuftion of Fuel,
and the communication of Heat by flame, will mow
it to be well founded ; and the advantages which
may be derived from the knowledge of this fad
are of very great importance indeed. But this
fubjecl: deferves to be thoroughly inveftigated.



VOL. II. K



66 Of the Management of Fire ,



CHAP. IV.

Of the MANNER in which HEAT is COMMUNI
CATED by FLAME to other Bodies. Flame acts on
Bodies in the fame Manner as a hot Wind. The
EffecJ of a Blow-pipe in increafmg the Aftivity of
Flame explained^ and illuftrated by Experiments.
A Knowledge of the Manner in which Heat is
communicated by Flame nece'Jfary, in order to deter
mine the moft advantageous Forms for Boilers.
General Principles on which Boilers of all Dimen*
fions ought to be conftrufted.

IF FLAME be merely vapour, or a mixture of
air and fleam heated red-hot, as air and fleam
are both non-conductors of heat, there feems to be
no difficulty in conceiving that Flame may, not-
withflanding its great degree of heat, flill retain the
properties of its component fluids, and remain a
non-conduftor of heat. The non-conducting power
of air does not appear to be at all impaired by be
ing heated to the temperature of boiling water ;
and I fee no reafon why that property in air, or in
any other claftic fluid, mould be impaired by any
augmentation of temperature however great. If
fleam, or if air, at the temperature of 2 1 2 degrees
of Fahrenheit's thermometer^ be a non-conductor

of



and the Economy of Fuel. 67

f heat, why fliould it not remain a non-conductor
at that of 1000 degrees, or when heated red-hot ?
I confefs I do not fee how a body could be deprived
of a property fo effential, without being at the fame
time totally changed ; and I believe nobody will
imagine that either air or fleam undergo any chy-
mical change merely by being heated to the temper
ature of red-hot iron. But without infilling upon
thefe reafonings, however concluiive I may think
them, I mall endeavour to mow, from experiment
and obfervation, in fhort to frove, that Flame is in
fact a non-conductor of heat.

Taking it for granted, what I imagine will not
be denied, that air is a non-conduclor of heat, at
leaft in the fenfe I have ufed that appellation, I mall
endeavour to iliow that Flame acts precifely in the
fame manner as a hot wind would do in commu*
nicating heat, and in no other way ; and if I fuc-
ceed in this, I fancy I may confider the proportion
as fufficiehtly proved.

The effect of a blaft of cold air in cooling any
hot body expofed to it is well known, and the
caufes of this effect may eafily be traced to that
property of air which renders it a non-conduclor
of heat ; for if the particles of cold air in contact
with a hot body, could, with perfect facility, give
the heat they acquire from the hot body to other
particles of air by which they are immediately
furrounded, and thefe again to others, and fo on,
the heat would be carried off as f aft as the hot body
could part with //, and any motion of the particles

of



68 Of the Management of Fire,

of the air, any wind, or blaft, would not fenfibly
facilitate or haften the cooling of the body ; and
by a parity of reafoning it may be fhown, that if
Flame were in fact a perfect conductor of heat, any
cold body plunged into it would always be heated
as fqft as that body could receive heat ; and neither
any motion of the internal parts of the Flame, nor
the velocity with which it impinged againft the cold
body, could have any fenfible effect either to facil
itate or accelerate the heating of the body. But if
Flame be a non-conductor of heat, its action will be
exactly fimilar to that of a hot wind, and confe-
quently much will depend upon the manner in
which it is applied to any body intended to be
heated by it. Thofe* particles of it only which are
in actual contact with the body will communicate
heat to it ; and the greater the number of different
particles of the Flame which are brought into con.
tact with it, the greater will be the quantity of heat
communicated : Hence the importance of cauling
the Flame to impinge with force againft the body to
be heated, and to ftrike it in fuch a manner that its
current may be broken, and that whirlpools may be
formed in it ; for the rapid motion of the Flame
caufes a quick fucceffion of hot particles ; and admit
ting our affumed principles to be true, it is quite
evident that every kind of internal motion among
the particles of the Flame by which it can be agitat
ed, mult tend very powerfully to accelerate the
communication of the heat.

The



and the Economy of Fuel. 69

The effect of a blow-pipe is well known, but I
do not think that the manner in which it increafes
the aftion of Flame ^ has ever been fatisfactoriiy ex
plained. It has generally been imagined, I be
lieve, that the current of frefh air which is forced
through the Flame by a blow-pipe actually increafes
the quantity of heat ; I rather fuppofe it does little
more than direct the heat aftually exifting in the
Flame to a given point. A current of air cannot
generate heat, without at the fame time being de*
compofed ; and in order to its being decompofed
in a fire, it muft be brought into actual contact
with the burning Fuel, or at leaft with the unin-
flamed inflammable vapour which rifes from it :
But can it be fuppofed that there can be any thing
inflammable, and not actually inflamed, in the
clear, bright, and perfectly tranfparent Flame of a
wax candle ? A blow-pipe has however as fenfible
an effect, when directed againft the clear Flame of a
wax candle, as when it is employed to increafe the
action of a common glafs-worker's lamp.

Conceiving that the difcovery of the manner in
which the current of air from a blow-pipe ferves to
increafe the intensity of the action of the Flame,
could not fail to throw much light upon the fub-
ject under confideration, namely, the inveftiga-
tion of the manner in which heat is communicated
to bodies by Flame, I made the following Experi
ments, the refults of which I conceive to be de-
cifive.

Concluding



70 Of the Management of Fire,

Concluding that the current of air from a blow-
pipe, directed againft the Flame of any burning
body, could tend to increafe the intensity of the
action of the Flame only in one or both of thefe two
ways, namely, by increafing its a&ion upon the
body againft which it is directed ; or by actually
increafing the quantity of heat generated in the
combuftion of the Fuel ; a method occurred to me
by which I thought it poffible to determine, by ac
tual experiment, to which of thefe caufes the effect
in queftion is owing, or how much each of them
might contribute to it. To do this, I filled a large
bladder, containing above a gallon, \viihjixed air,
which, as is well known, is totally unfit for fup-
porting the combuftion of inflammable bodies, and
which, of courfe, could not be fufpected of adding
any heat to a Flame againft which a current of it
fhould be directed ; I imagined therefore that if a
blow-pipe fupplied with this air, on being directed
againft the Flame of a candle, mould be found to
produce nearly the fame effect as when common
air is ufed for the fame purpofe, it would prove to
a demonftration that the augmentation of the in-
tenfity of the action, or activity of the Flame which
arifes from the ufe of a blow-pipe, is owing to the
agitation of the Flame, to its being directed to a
point, to the impetuofity with which it is made
to {hike againft the body which is heated by it,
and to the rapid fucceffion of frefti particles of this
hot vapour, and not to ifty po/itive increafe of heat.

A blow-



and the Economy of Fuel. 71

A blow-pipe being attached to the bladder con
taining fixed air, the end of this pipe was directed
to the clear brilliant flame of a wax candle, which
had juft been muffed ; and, by comprefling the
bladder, the Flame was projected againft a fmall
tube of glafs, which was very foon made red-hot,
and even melted.

Having repeated this experiment feveral times,
and having found how long it required to melt the
tube when the Flame of the candle was forced
againft it by a blaft of jixed air, I now varied the
experiment, by making ufe of common atmofpheric
air, inftead of fixed air j taking care to employ the
fame candle and the fame blow-pipe ufed in the for
mer experiments, and even making ufe of the
bladder, in order that the experiments being ex
actly fimilar, and differing only in the kinds of air
made ufe of, the effect of that difference might be
difcovered and eftimated.

The refults of thefe experiments were moil per-
fectly conclufive ; and proved in a decilive man
ner, that the effect of a blow-pipe, when applied to
clear Flame, arifes not from any real augmentation
of heat, but merely from the increafed activity of
the Flame, in confequence of its being impelled
with force, and broken in eddies on the furface of
the body againft which it is made to act ; the effect
of the blow-pipe in thefe experiments being to all
appearance quite as great when fixed air was made
ufe of, (which could not increafe the quantity of
heat,) as when atmofpheric air was ufed.

But:



72 Of the Management of Fire,

But conceiving the determination of this queftion
relative to the manner in which Flame communicates
heat, to be a matter of much importance, I did
not reft my inquiries here : I repeated the exper
iments very often, and varied them in a great
number of different ways ; fometimes making ufe
of fixed air ; fometimes of atmofpheric air ; and
at other times ufing dephlogifticated air ; and
common air rendered unfit for the fupport of ani
mal life and of combuftion, by burning a candle in
it till the candle went out.

It would take up too much time to give an ac
count in detail of all thefe experiments ; I mail
therefore content myfelf with merely obferving,
that they all tended to mow that the effecl: of a
blow-pipe ufed in the manner here defcribed, is owing
to the direction and velocity it gives to the Flame
againft which it is employed, and not to any real
increafe of heat.

It muft be remembered that the principal object
I had in view in thefe experiments was to difcover
the manner in which Flame communicates heat to
other bodies, and by what means that communi
cation may be facilitated. Were it required to in
creafe the intenfity of the heat by blowing the fire ^ the
current of air muft be applied in fuch a manner as
to expedite the combuftion ; it muft be directed to
the inflamed furface of the burning Fuel, and not
to the red-hot vapour or flame which rifes from it,
and in which the combuftion is moft probably
already quite complete ; and in this cafe there is

no



and tie Economy of Fuel. 73

no doubt but the effeft produced by blowing would
depend much upon the quality of the air made

ufe of.

The refults of the forgoing experiments with
the blow-pipe will, I am confident, be thought
quite conclufive by thofe who will take the trouble
to confider them attentively, and the advantages
that may be derived from the knowledge of the
fad eftablifhed by them are very obvious. If Flame *
or the hot vapour which rifes from burning bodies,
be a non-conductor of heat ; and if, in order to
communicate its heat to any other body, it be ne-
ceffary that its particles individually be brought into
aftual contact with that body ; it is evident that
the form of a boiler, and of its fire-place, muft be
matters of much importance; and that that form
muft be moft advantageous, which is beft calculated
to produce an internal motion in the Flame, and to
bring alternately as many of its particles as poflible
into contact with the body which is to be heated by-
it. The boiler muft not only have as large a fur-
face as poilible, but it muft be of fuch a form as
to caufe the Flame which embraces it to impinge
againft it with force to break againft it and to
play over its furface in eddies and whirlpools.

It is therefore againft the bottom of a boiler, and
not againft its fides, that the principal efforts of the
Flame muft be directed ; for when the flame, or hot
vapour, is permitted to rife freely by the vertical
fides of a boiler, it flides over its furface very ra
pidly, and there being no obftacle in the way to
L break



74 Qf the Management of Fire,

break the Flame into eddies and whirlpools, it glides
quietly on like a ftream of water in a fmooth canal;
and the fame hot particles of this vapour which
happen to be in immediate contact with the fides of
the boiler at its bottom or lower extremity, being
continually prefled againft the furface of the boiler
as they are forced upwards by the rifing current,
prevent other hot particles from approaching the
boiler ; fo that by far the greateft part of the heat
in the Flame and hot vapour which rife from the
Fire, inftead of entering the boiler, goes off into
the atmofphere by the chimney, and is totally loft.
The amount of this lofs of heat, ariiing from
the faulty conftruction of boilers and their fire
places, may be eftimated from the refults of the
Experiments recorded -in the following Chapter.



:



and tie Economy of Fuel. 75



CHAP. V.

An Account of Experiments made with Boilers and
Fire-places of various Forms and Dimenfions ; to