ill

^/
*

C'

BOOK 546.D 17 pt c.
DALTON # NEW SYSTEM OF CHEMICAL
PHILOSOPHY
i

T1S3 0Q],33t,0M

NEW SYSTEM

CHEMICAL PHILOSOPHY,
PART

BY

I.

JOHN DAL TON,

D.C.L., LL.D., F.R.SS., L.

&

E.,

M.R.I.A.

MEMBER OF THE ROYAL ACADEMY OF

SCIENCES, PARIS; PRESIDENT OF THE LITERARY

AND PHILOSOPHICAL SOCIETY, MANCHESTER; MEMBER OF THE ROYAL ACADEMIES
OF SCIENCE OF BERLIN, AND MUNICH, AND OF THE NATURAL HISTORY SOCIETY
OF MOSCOW. HONORARY MEMBER OF THE MEDICAL AND CHIRURGICAL
SOCIETY OF LONDON, OF THE ROYAL MEDICAL SOCIETY EDINBURGH,
THE ROYAL STATISTICAL SOCIETY OF FRANCE; OF THE
PHILOSOPHICAL SOCIETIES OF BRISTOL, CAMBRIDGE,

LEEDS, SHEFFIELD, YORKSHIRE, &C., &C.

SECOND EDITION.

LONDON:
JOHN WEALE, ARCHITECTURAL LIBRARY,
HIGH HOLBORN.
PRINTED BY SIMPSON AND GILLETT, BROWN STREET, MARKET STREET, MANCHESTER.

1842.

I'?

u>

^(7

TO THE

PROFESSORS OF THE UNIVERSITIES

AND

OTHER RESIDENTS
OF

EDINBURGH AND GLASGOW,

WHO GAVE
Ziftix attfntiott

an& ^ncottragem^nt

TO THE

LECTURES ON HEAT AND CHEMICAL ELEMENTS,

DELIVERED IN THOSE CITIES IN 1807;
AMD

TO THE MEMBERS
OF THE

LITERARY AND PHILOSOPHICAL SOCIETY,

OF MANCHESTER,
WHO HAVE

UNIFORMLY PROMOTED

ijts
IS

HIS RESEARCHES,

wmovk

RESrECTFULLY INSCRIBED, BY

THE AUTHOR.

PREFACE TO THE FIRST EDITION.

It was the author*s intention

to publish

publish
to

it

it

in

entire in

two

when

one volume

this

work was put

to press

now induced to
which it may be proper

but he

parts, for reasons

is

announce.
Various essays of his were read before the Literary and

Philosophical Society of Manchester, chiefly on heat and

elastic fluids,

Memoirs,

and were published

in the

5th

The new views which

in 1802.

loped, were considered both curious

Volume of

their

these essays deve-

The

and important.

essays were re-published in several Philosophical Journals,

and soon

French and German, and

after translated into

cir-

culated abroad through the

medium

The author was not remiss

in prosecuting his researches, in

of the foreign Journals.

which he was considerably assisted by the application of

principles derived

from the above essays.

In 1803, he was

gradually led to those primary Laws, which seem to obtain

in

regard to heat, and to chemical combinations, and which

it is

the object of the present

brief outline of

them was

work

first

to exhibit

and

elucidate.

publicly given the ensuing

winter in a course of Lectures on Natural Philosophy, at the

Royal Institution

in

London, and was

the Journals of the Institution

whether that was done.

left for

but he

is

publication in

not informed

The author has ever

since been

occasionally urged by several of his philosophical friends to

lose

no time

in

communicating the

the public, alleging, that

own

results of his enquiries to

the interests of science,

reputation, might suffer

by

and

delay. In the spring of

his

807,

PREFACE.

VI
he was induced

to offer the exposition of the principles herein

contained in a course of Lectures, which were twice read in

On

Edinburgh, and once in Glasgow.

these occasions he

was honoured with the attention of gentlemen, universally

acknowledged
tific

to

attainments

be of the

first

respectability for their scien-

most of whom were pleased to express their

desire to see the publication of the doctrine in the present

Upon

form, as soon as convenient.

the author's return to

Manchester he began to prepare for the press.

experiments required to be repeated

made

to be

manner was

other

Several

new ones were

almost the whole system, both in matter and

to be new,

for the composition

and consequently required more time

and arrangement.

These considerations,

together with the daily avocations of profession, have delayed

the

work nearly a year

and, judging from the past,

require another year before

mean

it

can be completed.

time, as the doctrine of heat,

it

may

In the

and the general principles

of Chemical Synthesis, are in a good degree independent of

the future details, there can no great detriment arise to the
author, or inconvenience to his readers, in submitting what
is

already prepared, to the inspection of the public.

May,

1808.

CONTENTS OF PART FIRST

PAGE

Chap.

I.

Section

On
1.

Heat or Caloric

On

Temperature, and the instruments

for measuring
2.

On Expansion

3.

On

4.

On

5.

On

23

hy heat

the specific heat

the

it

Theory of

47

of bodies
the specific heat

of

^^

elastic fuids

Quantity of heat evolved

the

by

combustion^ ^c
6.

On

the

75

natural Zero of temperature,

82

or absolute privation of heat

7.

On

the

motion and communication of

heat,

arising

from

inequality

of

temperature
.

99

8.

On

the Temperature of the atmosphere 123

9,

On

the

Phenomena of

of water

the

Congelation
133

CONTENTS.

11

PAGE
CifAP.

II.

Section

On
1.

the Constitution

On

of Bodies

constitution

the

of pure

141
elastic

145

fluids
2.

On

constitution

the

of mixed

elastic

150

fluids
3.

On

the constitution

of

liquids,

and

the

mechanical relations betwixt liquids

and
4.

Chap.

III.

On

On

elastic fluids

the constitution

of solids

Cheinical Synthesis

Exjilanation of the Plates

94

208
211

217

PREFACE TO THE SECOND EDmON.

The

first

Edition of this part of the work having been out

of print for some years, the Author has been induced at the
request of several of his friends to publish a second Edition,

without making any material alteration in

The

it.

feeble state of the Author's health, resulting

long and severe

illness

which he experienced about

ago, has greatly interfered with his scientific pursuits

is

now preparing
1.

2.

from a

five years
;

but he

for publication the following Essays, viz.:

On Microcosmic
On the Sulphate

Salt.

of Magnesia, and the Diphosphate of

Soda.
3.
4.

On the Acid," Base and Water, of different Salts.

On a simple and easy method of analysing Sugar.
March,

1842.

NEW SYSTEM
OP

CHEMICAL PHILOSOPHY
CHAPTER

I.

ON HEAT OR CALORIC.
The

most probable opinion concerning the na-

ture of caloric,

is,

that of

its

being an

elastic fluid

of great subtilty, the particles of which repel one

another, but are attracted

When

all

by

all

other bodies.

surrounding bodies are of one tem-

perature, then the heat attached to them

quiescent state
in

any two bodies

ther

we

is

in

the absolute quantities of heat

in this case are not equal,

whe-

take the bodies of equal weights or of

equal bulks.
affinity for

Each kind of matter has its peculiar

heat, by which it requires a certain

portion of the

fluid, in

order to be in equilibrium

ON HEAT OR CALORIC.

Were

with other bodies at a certain temperature.

the whole quantities of heat in bodies of equal

weight or bulk, or even the relative quantities,

accurately ascertained, for any temperature, the

numbers expressing those

stitute a table

quantities

would con-

of specific heats, analogous to a

and would be an im-

table of specific gravities,

Attempts of

portant acquisition to science.

this

kind have been made with very considerable success.

the specific heats, could they be thus

Whether

obtained for one temperature, would express the

relation at every other temperature, w^hilst the

bodies retained their form,

From

moment.

there seems

but
cific

it is

is

an enquiry of some

the experiments hitherto

little

doubt of

its

made

being nearly so

perhaps more correct to deduce the spe-

heat of bodies from equal hulks than from

equal weights.

methods

It is

very certain that the two

will not give precisely the

because the

same

results,

expansions of different bodies

by

equal increments of temperature are not the same.

But before

we should

this subject

first settle

can be

what

by the word temperature.

is

w^ell

considered,

intended to be meant

ON TEMPERATURE.

SECTION

I.

ON TEMPERATURE,
AND THE INSTRUMENTS FOR MEASURING

The

IT.

notion of the specific heat of bodies and

may be

of temperature,

well conceived from a

system of cylindrical vessels of different diameters

connected with each other by pipes

at the bottom,

and a small cylindrical tube attached

tem,

liquid,

to the sys-

capable of holding water or any other

all

and placed perpendicular

(See Plate

Fig. 1.)

1.

The

to the horizon.

cylinders are to re-

present the different specific heats of bodies

and

the small tube, being divided into equal parts,

to represent the

rises

to

and

them

the same level in

thermometer
poured

thermometer or measure of tem-

If water be poured into one vessel

perature.

in,

in the

is

if

all,

and

it

in the

equal portions be successively

there will be equal rises in the vessels

tube

the water

is

to represent heat or caloric.

notion, then,

it is

obviously intended

According

to this

evident that equal increments

of heat in any body correspond to equal incre-

ments of temperature.
This view of the subject necessarily requires,
that

if

two bodies be taken of any one tempera-

ON TEMPERATURE.

4
ture,

and then be raised

to

any other temperature,

the additional quantities of heat received

by each

will

be exactly proportioned to the whole quanti-

ties

of that fluid previously contained in them.

This conclusion, though

it

tent with facts in general,

For, in elastic

true.

may

is

be nearly consis-

certainly not strictly

well known, an

fluids, it is

increase of bulk occasions an increase of specific

heat,

though the weight and temperature continue

the same.

It is

probable then that solids and

liquids too, as they increase in bulk

by

heat, in-

crease in their capacity or capability of receiving

more.

This circumstance, however, might not

affect the conclusion above,

provided

all

bodies

increased in one and the same proportion by heat

but as this

is

not the case, the objection to the

conclusion appears of validity.

Suppose

it

were

allowed that a thermometer ought to indicate the

accession of equal increments of the fluid deno-

minated

caloric, to the

shew the temperature

body of which

it

was

to

suppose too that a mea-

sure of air or elastic fluid was to be the body

query, whether ought the

air to

expand by the temperature, or

to

the same space of one measure

me

be suffered to

be confined to
It

appears to

the most likely in theory to procure a stand-

ard capacity for heat by subjecting a body to heat.

ON TEMPERATURE,
whilst

m =

its

hulk

is

the quantity of heat necessary to raise the

elastic fluid 10 in
711

-\-

= the

when

10,

Let

kept constantly the same.

temperature in

this case

then

quantity necessary to raise the same,

suffered to expand,

d being the

diff'er-

ence of the absolute quantities of heat contained

by the body

in the

two

Now,

cases.

-^

m =

the

quantity of heat necessary to raise the tempera1 in the first case

ture

be the quantity necessary

be

will

but -^
in the

a less quantity in the

(m + d) cannot

second case

it

lower degrees, and

a greater in the higher.

If these principles be

may be

applied to liquids and

admitted, they
solids

a liquid, as water, cannot be raised in

temperature equally by equal increments of heat,

unless

it is

confined within the same space

by an

extraordinary and perhaps incalculable force

we

suffer

it

to take its ordinary course of

sion, then, not equal, but increasing

of heat will raise

suflScient force

its

to

if

expan-

increments

temperature uniformly.

were applied

If

condense a liquid

or solid, there can be no doubt but heat would be

given out, as with elastic

It

fluids.

may perhaps be urged by some that the

ference of heat in condensed and rarefied

by analogy probably
liquids

and

solids, is

in

air,

dif-

and

the supposed cases of

too small to have sensible

ON TEMPERATURE.

influence on the capacities or afiinities of bodies

for heat

that the eff'ects are such, as only to raise

or depress the temperature a few degrees

perhaps the whole mass of heat

air

supposed

and that a
contain 2005 of tem-

to

perature being rarefied

lost 5 of

its

granted

if

become 2000,

till it

may

or

still

be considered as

capacity invariable.

This may be

temperature,

having

when

equivalent to

is

two or three thousand such degrees

volume of

the data are admissable

but the true

changes of temperature consequent to the condensation and rarefaction of air have never been

determined.
Vol.

eff'ects

density of
easily

is

affected as if in

of 50 higher or lower temperature

medium

am

the process of admitting

vacuum, and of liberating condensed

the inclosed thermometer

but the

Mem.

have shewn, (Manchester

5, Pt. 2.) that in

air into a
air,

air,

of instantaneously doubling the

or replenishing a vacuum, cannot

be derived from those or any other

acquainted with

they

may perhaps

facts I

raise the

temperature one hundred degrees or more.

great heat produced in charging an air-gun

The
is

proof of a great change of capacity in the inclosed

air.

Upon

that the

the whole then

it

may be

concluded,

change of bulk in the same body by

change of temperature,

is

productive of consider-

ON TEMPERATURE.
able effect on

we

capacity for heat, but that

its

are not yet in possession of data to determine

effect

on

elastic fluids,

M. De Luc

solids.

and

still

found, that in mixing equal

at the freezing

weights of water

peratures, 32 and

and boiling tem-

212, the mixture

indicated

nearly 119 of Fahrenheit's mercurial

meter

its

on liquids and

less

but the numerical mean

is

thermo-

122

if

he

had mixed equal hulks of water at 32 and 212,

he w^ould have found a mean of 115.

means determined by experiment

ways are probably too high

for,

in

Now

the

both these

water of these

two temperatures being mixed, loses about l-QOth

of

its

bulk

this

condensation of volume (whe-

ther arising from an increased affinity of aggre-

com-

gation, or the effect of external mechanical

pression,

is

all

one) must expel a quantity of

heat, and raise the temperature above the true

mean.

It is

not improbable that the true

temperature between 32 and 212

may be

mean
as

low

as 110 of Fahrenheit.
It

has been generally admitted that

tions of

any

liquid, of equal

if

two por-

weight but of

differ-

ent temperatures, be mixed together, the mixture

must indicate the true mean temperature

that instrument which corresponds with

accurate oneasure of temperature.

it

But

is

if

and
an
the

ON TEMPERATURE.

preceding observations

be

correct,

it

may be

questioned whether any two liquids will agree in

giving the same

mixed

mean temperature upon being

as above.

In the present imperfect mode of estimating

temperature, the equable expansion of mercury
is

adopted as a scale for

measure.

its

not be correct for two reasons

the

mean by

the mixture

1st.

of water of different temperatures

is

This can-

always below

the mercurial thermometer

for in-

stance, water of 32 and 212 being mixed, gives

119 by the thermometer

whereas

it

appears

from the preceding remarks, that the temperature

of such mixture ought to be found above the

122

mean

2nd. mercury appears by the most recent

experiments to expand by the same law as water;

namely, as the square of the temperature from the
point of greatest density.

The apparently equal

expansion of mercury arises from our taking a

small portion of the scale of expansion, and that at

some distance from the freezing point of the liquid.

From what has been remarked it appears that

we have not yet any mode

ascertaining what

is

easily practicable for

the true

mean between any

two temperatures, as those of freezing and boiling water

nor any thermometer which can be

considered as approximating nearly to accuracy.

ON TEMPERATURE.

Heat

a very important agent in nature

is

it

cannot be doubted that so active a principle must

be subject to general laws.

indicate otherwise,
sufficiently

it is

If the

phenomena

because we do not take a

comprehensive view of them.

Philo-

sophers have sought, but in vain, for a body that

should expand uniformly, or in arithmetical pro-

by equal increments of heat liquids

have been tried, and found to expand unequally,
all of them expanding more in the higher temgression,

peratures than in the lower, but no two exactly

Mercury has appeared

alike.

variation, or
sion,

to

have the least

approach nearest to uniform expan-

and on that and other accounts has been ge-

nerally preferred in the construction of thermo-

meters.

Water has been

rejected, as the most

unequally expanding liquid yet known.

the publication of

my

sion of elastic fluids

Since

experiments on the expan-

by

heat,

and those of

Gay

Lussac, immediately succeeding them, both de-

monstrating the perfect sameness in

all

ently elastic fluids in this respect,

it

permanhas been

imagined by some that gases expand equally; but

this is not corroborated

by experience from other

sources.

Some time ago

it

occurred to

me

as probable,

that water and mercury, notwithstanding their

ON TEMPERATURE.

10

apparent diversity, actually expand by the same

law,

and that the quantity of expansion

is

as the

square of the temperature from their respective

Water very nearly accords with

freezing points.
this

law according to the present scale of temper-

and the

deviation observable

is

exact-

ly of the sort that ought to exist, from the

known

ature,

little

error of the equal division of the mercurial scale.

By

prosecuting this enquiry I found that the mer-

curial

and water

scales divided according to the

principle just mentioned,

as far as they
will

would perfectly accord,

were comparable

probably extend to

all

and that the law

other pure liquids

but not to heterogeneous compounds, as liquid solutions of salts.

If the law of the expansion of liquids be such

as just mentioned,

it is

natural to expect that other

phenomena

of heat will be characteristic of the

same law.

It

may be

Force of Steam (Man.

seen in

Mem.

my

Vol.

5,

Essay on the
Part 2.) that

the elastic force or tension of steam in contact

with water, increases nearly in a geometrical progression to equal increments of temperature, as

measured hy the common inercurial

not a

little

surprising to

me

scale

; it

was

at the time to find

such an approach to a regular progression, and

was then inclined

to think, that the

want of per-

ON TEMPERATURE.
was owing

feet eoineidenee

division of the received

awed by
to

11

to inaeeuraey in the

thermometer

but over-

who seemed

the authority of Crawford,

have proved past doubt that the error of

the thermometer nowhere amounted to

more than

one or two degrees, I durst not venture to throw

out more than a suspicion at the conclusion of the
essay, on the expansion of elastic fluids
that the error
^

had determined

was probably 3 or
;

heat,

De Luc

to admit of an error in the sup-

posed mean, amounting to

rantable.

4, as

by

However

it

12,

seemed unwar-

now appears

that the force

of steam in contact with water, increases accurately in

geometrical progression to equal increments

of temperature, provided those

increments

are

measured by a thermometer of water or mercury,

the scales of which are divided according to the

above-mentioned law.

The

force of

by the above
air to

Steam having been found

law,

do the same

manently

was natural

it
;

elastic fluid)

for, air

to

vary

to expect that of

(meaning any per-

and steam are essentially

the same, diff"ering only in certain modifications.

Accordingly
pands

in

it

was found upon

trial

that air ex-

geometrical progression to equal incre-

ments of temperature, measured

as above.

Steam

ON TEMPERATURE.

12

detached from wacer, by which

it is

rendered

in-

capable of increase or diminution in quantity, was

Gay

found by

Lussac, to have the same quantity

of expansion as the permanently elastic

had formerly conjectured that

air

fluids.

expands as the

cube of the temperature from absolute privation,

as hinted in the essay above-mentioned,

but I

am now obliged to abandon that conjecture.

The union of so many analogies in favour of the
preceding hypothesis of temperature
sufficient to establish

it

is

almost

but one remarkable

trait

of temperature derived from experiments on the

heating and cooling of bodies, which does not

accord with the received scale, and which, nevertheless,

claims special consideration,

body in cooling

is,

that

loses heat in proportion to

its

excess of temperature above that of the cooling

7nediwn

or that the temperature descends in

geometrical progression in equal moments of time.

Thus

if

body were 1000 above the medium

the times in cooling from 1000 to 100, from 100

to 10,

and from 10

to 1,

This, though nearly,

adopt the

common

is

ought

all to

be the same.

not accurately true,

scale, as is well

known

if
;

we
the

times in the lower intervals of temperature are

found longer than

in the

upper

but the

new

scale

ON TEMPERATURE.

13

proposed, by shortening the lower degrees, and

lengthening the higher,

is

found perfectly accord-

ing to this remarkable law of heat.

Temperature then will be found to have four

most remarkable analogies to support it.
All pure homogenous liquids, as water

1st.

and mercury, expand from the point of their congelation, or greatest density, a quantity always
as the square of the temperature

2nd.

The

as water,

ether, &c., constitutes

progression

from that point.

force of steam from pure liquids,

to

increments

of

geometrical

temperature in

arithmetical progression.
3rd.
is

The expansion

of permanent elastic fluids

in geometrical progression to equal increments

of temperature.
4th.
trical

The

refrigeration of bodies

is

in

geome-

progression in equal increments of time.

A mercurial thermometer graduated according

to this principle will differ

from the ordinary one

with equidiflerential scale, by having

its

degrees smaller and the upper ones larger

lower
;

the

mean between freezing and boiling water, or 122

on the new scale, will be found about 110 on
the old one.
The following Table exhibits the

numerical calculations illustrative of the principles

inculcated above.

14

ON TEMPERATURE.

NEW TABLE OF TEMPERATURE.

ON TEMPERATURE.

15

EXPLANATION OF THE TABLE.

The

first

column contains the degrees of tem-

perature, of which there are supposed to be 180

between freezing and boihng water, according

The concurrence

Fahrenheit.

of so

many

to

analo-

gies as have been mentioned, as well as experi-

ence, indicate that those degrees are produced

by equal increments of the matter of

caloric

but then

it

heat, or

should be understood they

are to be applied to a

body of uniform bulk and

capacity, such as air confined within a given space.

If water, for instance, in

its

ordinary

state, is to

be raised successively through equal intervals of

temperature, as measured by this scale, then unequal increments of heat will

reason of

its

be requisite, by

increased capacity.

ber in the column,

The

175, denotes

first

the point at

which mercury freezes, hitherto marked

The

calculations

68
100.

to 212

By

are

made

above the

comparing

for

last

this

are perceived

ence between 32 and 212

of the

new

scale,

40.

every 10 from

number,

for

every

column with the 5thj

new

the correspondences of the

common one

num-

scale

and the

the greatest differis

observable at 122'

which agrees with 110 of the

ON TEMPERATURE.

16

being 12

old, the difference

above 212, the differences

but below 32 and

become more remark-

able.

The 2nd and 3rd columns

are two series, the

one of roots, and the other of their squares.

are

obtained thus

column,

is

opposite

32,

in

They

the

first

placed in the 3rd, 72, being the num-

ber of degrees or equal parts in Fahrenheit's

scale

from freezing mercury to freezing water

and opposite 212

3rd, being

in the first is placed 252 in the

212 + 40, the number of degrees

(or

rather equal parts) between freezing mercury and

The

boiling water.

square roots of these two

numbers, 72 and 252, are found and placed

opposite to them in the second column.
The
number 8,4853 represents the relative quantity
of real temperature between freezing mercury

and freezing water

and the number 15.8743

represents the like between freezing mercury and

boiling water

consequently the difference 7.3890

represents the relative quantity between freezing

water and boiling water, and 7.3890

~ 18 = .4105

represents the quantity corresponding to

interval of 10.

By

each

adding .4105 successively

to 8.4853, or subtracting

it

from

it,

the rest of

the numbers in the column are obtained, which

are of course in arithmetical progression.

The

ON TEMPERATURE.
numbers

3rd column are

in the

1?
obtained by

all

The

squaring those of the 2nd opposite to them.

unequal

3rd column mark the

differences in the

expansions of mercury due to equal increments

of temperature,

down by

inconve-

being carried

its

intervals of 10 to the point of freezing

mercury, which however

The

The

by the theory.

nient length of the table prevents

4th column

the difference of

common method

is

40%

is

found to be

at

175.

the same as the 3rd, with

to

make

it

conform to the

of numbering on Fahrenheit's

scale.

The

5th column

is

the 4th corrected, on account

of the unequal expansion of Glass

rent expansion of mercury in glass

real,

by the expansion of the

The appa-

is

less than the

glass itself; this,

however, would not disturb the law of expansion

of the liquid, both apparent and real being subject
to the same,
this will

shewn by

provided the glass expands equally;

be shewn hereafter.

De

But

has been

it

Luc, that glass expands

lower half of the scale than the higher

less in the
;

this

must

occasion the mercury apparently to expand more

in the lower half than

of expansion.

By

data, I find, that the

scale, or 122,

what

is

dictated

calculating from

mercury

in the

by the law

De

Luc's

middle of the

ought to be found nearly

3"^

higher

ON TEMPERATURE.

18

than would be, were

however

it

not for this increase.

Not

the effect, I have taken

to over-rate

it

only at 1.7, making the number 108. 3 in the

4th column, 110^ in the 5th, and the rest of the

column
in

is

corrected accordingly.

The numbers

column cannot well be extended much

this

beyond the

interval from freezing to boiling water,

want of experiments on the expansion of glass.

for

By

viewing

this

column along with the

quantity of the supposed error in the

1st,

the

common

may be perceived and any observations on

may be reduced to the new.
The 6th column contains the squares of the

scale

the old thermometer

natural series,

1, 2, 3, &c.,

representing the ex-

pansion of water by equal intervals of tempera-

Thus,

ture.

if

a portion of water at 42 expands

a quantity represented

by 289,

temperature, then at 52

it

expanded

1,

at

will

at

the boiling

be found to have

62, 4 parts, &c., &c.

expands by cold or the abstraction of heat

Water
in the

same way below the point of greatest density,

will

as

be illustrated when we come to consider the

absolute

expansion

of bodies.

The apparent

greatest density too does not happen at 39. 3 old

scale,

but about 42

and the greatest real den-

sity is at or near 36^ of the same.

The

7th column contains a series of numbers

ON TEMPERATURE.

19

Geometrical Progression, denoting the expan-

in

sion of

air,

taken

is

Gay

The volume

or elastic fluids.

at 32

1000, and at 212, 1376 according to

Lussac's and

my own

experiments.

As for

the expansion at intermediate degrees. General

mid-way of

Roi makes the temperature

at

expansion, 11 6 J old scale

from the results of

my

Mem. Vol. 5,
temperature may be esti-

former experiments, (Manch.

Part

page 599) the

2,

mated

at

119i
nity of having
experiments I
will

total

but
air

at

had not then an opportu-

By more

32.

am convinced

that dry air of 32

expand the same quantity from

or 118 of
to 212.

Table

it

common

According

scale, as

if

that to 117^

from the

term

last

to the theory in the

above

appears, that air of 117^" will be 1188,

or have acquired one half

Now

recent

its

expansion.

total

the theory accord so well with experiment

in the middle of the interval, we cannot expect

to

do otherwise

The

it

in the intermediate points.

8th column contains the force of aqueous

vapours in contact with water expressed in inches

of mercury, at the respective temperatures.
constitutes a geometrical progression

the

It

num-

bers opposite 32 and 212, namely, .200 and 30.0

are derived from experiments, (ibid, page 559)

and the rest are determined from theory.

It is

ON TEMPERATURE.

20

remarkable that those numbers do not

the table just referred

that might

from

which was the result of

much

actual experience, so

difference

to,

differ

as 2 in

any part

even exist between two

thermometers of the same kind.

The

9th column exhibits the force of the vapour

of sulphuric ether in contact with liquid ether

which

is

a geometrical progression, having a less

mer Essay on
to correct

the Force of Steam, I

for-

am enabled

one of the conclusions therein contain-

by trusting

the error was committed

my

Since writing

ratio than that of water.

ed

to the

common mercurial thermometer.

Experience confirmed me that the force of vapour

accuracy of the

from water of nearly 212% varied from a change

much as vapour from ether of

Hence I deduced this general law,

of temperature as

nearly 100.

namely,

from

all

*'

that the variation of the force of vapour

liquids

is

the same for the same variation

oftemperature, reckoning from vapour of any given

force."
in

But

now

find that 30 of

the lower part of the

more than 30

common

in the higher

temperature

scale

is

much

and therefore the

vapours of ether and water are not subject to the

same change of force by equal increments of temperature.

and other

The

truth

is,

vapour from water, ether

liquids, increases in force in

geometrical

ON TEMPERATURE.
progression to the temperature

21
but the ratio

is

Ether as manufac-

different in different fluids.

tured in the large way, appears to be a very-

homogeneous

liquid.

have purchased

it

in

London, Edinburgh, Glasgow and Manchester,

at very different times, of precisely the same
quality in respect to

when thrown up
the mercury

its

into a

vapour

namely, such as

barometer would depress

5 inches at the temperature of

68"*.

Nor does it lose any of its effects by time I have

now a barometer with a few drops of ether on the
;

mercury, that has continued with invaried efficacy

The numbers in the

eight or nine years.
column between the temperatures of 20 and 80,

for

are the results of repeated observations on the

above ether barometer for many years

those

above and below are obtained from direct experi-

ment

as far as

from

to 212

the low ones were

found by subjecting the vacuum of the barometer

to

an

artificial

were found
Essays

cold mixture

in

the

manner

and the higher ones

related in

my

former

only the highest force has been consi-

derably increased from what I formerly had

it,

in

consequence of supplying the manometer with

more ether

having been found to leave little

or no liquid when at the temperature of 212";

and

in

it

order to obtain the

maximum

effect

it

is

ON EXPANSION.

22

indispensible to have a portion of liquid remain-

ing in contact with the vapour.

The

10th column shews the force of vapour

from alcohol, or rather common

termined by experiment

This

vapour of water.

spirit of

in the
is

this case is a

as the

not a geometrical pro-

gression, probably because the liquid

and homogeneous.

wine, de-

same way

is

not pure

suspect the elastic fluid in

mixture of aqueous and alcoholic

vapour.

SECTION

2.

EXPANSION BY HEAT.
One

important effect of heat

of bodies of every kind.

is

Solids are least ex-

panded, liquids more, and elastic

all.

The

many

the expansion

fluids

quantities of increase in bulk

instances

been determined

most of

have

in

but partly

through the want of a proper thermometer,

little

general information has been derived from particular

The

experiments.

force

necessary to

counteract the expansion has not been ascertained, except in the case of elastic fluids
is

no doubt

it is

all

but there

The

quantity and

permanent

elastic fluids

very great.

law of expansion of

ON EXPANSION.
have already been given

it

23

remains then to ad-

vert to liquid and solid bodies.

In order to understand the expansion of liquids,

expedient to premise certain propositions

it is

Suppose a thermometrical vessel of glass,

1st.

with any liquid up to a

metal, &c., were filled

certain

mark

in the

stem

and that

it

was known

the vessel and the liquid had precisely the same

expansion, bulk for bulk, with the same change

of temperature
little

then

it

must be evident upon a

consideration, that whatever change of tem-

perature took place, the liquid must remain at the

same mark.
2nd. Suppose as before, except that both bodies

expand uniformly with the temperature, but the

liquid at a greater rate than the vessel
is

then

it

evident by an increase of temperature, the

liquid
tity

would appear

to ascend uniformly a quan-

equal to the difference of the absolute expan-

sion of the

two bodies.

3rd. Suppose as in the last case, but that the

liquid

expands

liquid

would then descend, and that uniformly by

at a less rate than the vessel

the

an increase of temperature, a quantity equal to the

difference of the absolute expansions.
4th.

Suppose as before, only the vessel now

expands uniformly, and the liquid with a velocity

ON EXPANSION.

24

uniformly accelerated, commencing from rest

in

temperature be added uniformly, the

this case if

liquid will appear to

descend with a velocity uni-

formly retarded to a certain point, there to be

and afterwards

stationary,

to ascend with an uni-

formly accelerated velocity, of the same sort as

the former.
liquid

For,

expands

is

as the velocity with

which the

uniformly accelerative,

successively pass through

all

in

to

some

the same as that of the vessel, and

therefore, for that

stationary

must

degrees, from

any assigned quantity, and must therefore

moment be

it

moment, the

previously to

liquid

that

must have descended by the

must appear

time

the liquid

third proposition,

and must afterwards ascend, by the 2nd, but not

uniformly.

Let the absolute space expanded by

moment

the liquid at the

denoted by

1,

of equal velocities be

then that of the vessel in the same

time must be 2

because the velocity acquired

by an uniformly accelerating

move a body through

It follows

time.

sunk

1,

force, is such as to

twice the space in the same

then that the liquid must have

being the excess of the expansion of the

vessel above that of the liquid.

Again,

let an-

other portion of temperature equal to the former

be added, then the absolute expansion of the

quid will be

4,

li-

reckoned from the commencement;

ON EXPANSION.

25

and the expansion of the vessel

also

place of the liquid will be the same as at

and therefore

2nd
to

it

must apparently ascend

the

first,

by the

Let a third portion of heat equal

portion.

one of the former be added, and

it

will

make

the total expansion of the liquid 9, or give 5 additional expansion,

from which deducting

2, that

of the vessel, there remains 3 for the apparent

expansion by the 3rd portion

will

in like

manner 5

be due for the 4th, and 7 for the 5th, &c.,

But the

being the series of odd numbers.

ag-

gregate of these forms a series of squares, as

Hence

well known.

is

the apparent expansion will

proceed by the same law as the

real,

only starting

If the law of expan-

from a higher temperature.

sion of the liquid be such that either the addition

or abstraction of temperature, that

is,

either heat

or cold produces expansion alike, reckoned from

the point of greatest density

expansion will
the real.

then the apparent

when the liquid is at the lowest

scale, we withdraw a portion of heat,

For,

point of the
it

still

be guided by the same law as

ascends to

if

or

is

in

the circumstance of

greatest density, and no expansion as at the com-

mencement if then we withdraw another portion,

it will expand 1 by hypothesis, but the vessel will
contract 2, which must make the apparent expan;

ON EXPANSION.

26

sion of the liquid 3

be

5,

by another

The

by another portion

will

it

7? &c., as before.

truth of the above proposition

may be

otherwise shewn thus

Let

1,4, 9? 16, 25, &c., represent the absolute

expansions of the liquid, and p, 2 p, S p, 4p, 5p,

by equal increments of

&c., those of the vessel

temperature, then

25

p,

p, 4

mon

p, 5

by

p, 7

p,

series of square

But

is 2.

it

is

com-

demon-

numbers, whose roots are

progression,

progression, and that the

is

metical series

3p,

1,

difference of

equal to twice the

square of the difference of the roots.

the square of

in

form an arithmetical

common

the terms of this progression

we have

Hence,

as

the above arith-

p, &c., equal to the dif-

ferences of a series of squares, the

ence of the roots of which

Now

p, &c.,

algebraists, that the differences of a

arithmetical

2= twice

in arithmetical progression, the

difference of which

strated

4 p,

the differences of these last

namely 3

form a series

^ p, 16

represent the apparent ex-

&c., will

pansion of the liquid

quantities,

2p, 9

common

differ-

is 1

to apply these principles

solid bodies are

generally allowed to expand uniformly within the

common range
quantity

is

of temperature

so small

at all events the

compared with the expansion

ON EXPANSION.

27

of liquids, such as water, that the deviation from

many

cases.

Water being supposed to expand according

to the

uniformity cannot require notice in

square of the temperature from that of greatest

density,

Cor.

we may derive the following conclusions.

1. The laws of uniformly accelerated mo-

tion, are the

same as those of the expansion of

water, whether absolute or apparent, the time in

one denoting the temperature

in the other,

the space denoting the expansion

time or temperature,

expansion

?;

= velocity,

that

and

.v

and

is, if ^

= space or

then,

f^ or ti\ or if are as s.

Cor.

is

as

s IS

SiS

t f

s is as

t,

2.

maximum

The

being supposed constant, &c.

real

density for any

temperature,

is

of water

expansion

number

of degrees of

the same as the apparent expan-

sion from apparent greatest density in

for the
if

from

same number of degrees.

any vessel

For instance,

water in a glass vessel appears to be of great-

est density, or descends lowest at 42 of

scale,

and appears

from thence

to

to

212

expand

-aV

then

it

of

common

its first

may be

volume

inferred

ON EXPANSION.

28

expansion of water from greatest

that the real

density by 170

-^ of

is

its

volume

absolute expansion of water

is

so that the

determinable this

way, without knowing either at what temperature

density

its

greatest, or the expansion of the

is

vessel containing

it.

Cor. 3. If the expansion of any vessel can be obtained

is

then

may

the temperature at which water

of greatest density be obtained and vice versa.

This furnishes us with an excellent method of

as-

certaining both the relative and absolute expansion

of

all solid

bodies that can be formed into

vessels capable of holding water.

Cor.

4. If

maximum
a

the apparent expansion of water from

density for 180 were to be equalled by

body expanding uniformly,

its

velocity

equal to that of water at 90 or mid- way.

any

solid

body be found

to

sion as water at 10 from

must be

And

if

have the same expan-

max. density

then

its

expansion for 180 must be \ of that water, &c.

Because

By

in

water v

is

as

t,

he.

graduating several glass thermometer ves-

sels, filling

them with water, exposing them

to dif-

ferent temperatures, and comparing results, I have

found the apparent expansion of water

for

every 10 of the

henceforward

common

call it)

in

glass

or old scale (as I shall

and the new one, as follows

ON EXPANSION.

EXPANSION OF WATER.
OLD SCALE.

29

ON EXPANSION.

30
for 1804,

Dr. Hope has given a paper on the con-

traction of water

by heat

in

low temperatures.

(See also Nicholson's Journal, Vol. 12.)

paper we find an excellent history of

In this

facts

and

opinions relative to this remarkable question in

There appear

physics, with original experiments.

to

have been two opinions respecting the tempera-

ture at which water obtains

the one stating

32

be

to

it

the other at 40.

cation of the above

its

maximum

density;

at the freezing point, or

Previously to the publi-

essay, I

had embraced the

opinion that the point was 32, chiefly from some

Dr, Hope

experiments about to be related.

argued from

own experiments

his

and upon re-examination of

to the subject,
I

found them

favour of

in

My attention was again turned

the other opinion

all

to

facts,

concur in giving the point of

greatest density at the temperature 36, or mid-

way between
two

the points formerly supposed.

In

letters inserted in Nicholson's Journal, Vol.

13 and

14, I

endeavoured

to

Hope's experiments supported

and no

other.

I shall

shew
this

now shew

that

Dr.

conclusion

that

my own

experiments on the apparent expansion of water

in different vessels, coincide

with them in esta-

blishing the same conclusion.

The

results of

my

experiments, without those

ON EXPANSION.

31

deductions, were published in Nicholson's Journal,

Vol. 10.

Since then some small additions

and corrections have been made.

It

may be

observed that small vessels, capable of holding

one or two ounces of water, were made of the

different materials,

and such as that glass tubes

could be cemented into them

and act

so as to resemble

The

meter.

as

when full of water,

a common thermo-

observations follow:
Water

stationary.

Corresponding points
of expansion.

Brown

Common white ware, and stone

I

ware,

&

44*

32

&

48

&
&
&
&
&
&
&

52i

at 32

at 38

earthen ware

40

3 Flint glass

42

32

4 Iron

32

5 Copper

42+
45+

6 Brass

45i

32

7 Pewter

46

32

8 Zinc,

48

32

9 Lead

49

32

As

32

53
59

60
60i
64 +
67

the expansion of earthen ware by heat has

never before been ascertained, we cannot make

use of the

first

and second experiments

the temperature of greatest density

can learn from them

is,

that the

all

to find

that

we

point must be

below 38.

According

to

Smeaton, glass expands yVo^ in

consequently it
temperature

length for 180 of

32

ON EXPANSION.

expands

-^

But water expands ^i^ or

in bulk.

more than 18 times as much therefore

the mean velocity of the expansion of water (which
rather

is

that at 90, or half

that of glass, which

water at 42
the former

through

this

equal to the expansion of

is

must therefore be yV of

last

of the temperature to the mean, or

of

5,

new

above the temperature

greatest density.

not be accurate

scale
at

The

38.

which

4 of old scale,
it is

absolutely of

This conclusion however canfor, it

appears from the preced-

ing paragraph that the

below

18 times more than

is

consequently water of 42 has passed

-Ys

^ of 90

way)

temperature must be

inaccuracy arises, I

have no

doubt, from the expansion of glass having been

under-rated by Smeaton

not from any mistake

of his, but from the peculiar nature of glass.

Rods and tubs of glass

annealed

are seldom

hence they are

if

ever properly

in a state of violent

energy, and often break spontaneously or with a

slight scratch of a file

expand more than

rods,

that thin bulbs should

tubes have been found to

and

it

might be expected

expand more

they do not require annealing

great strength of thin glass,

its

still,

because

hence too the

being less

brittle

and more susceptible of sudden transitions of

temperature.

From

the

above

experiments

it

ON EXPANSION.

33

seems that the expansion due

to glass, such as

the bulbs of ordinary thermometers,

very

is

little

less than that of iron.

Iron expands nearly

heat, or aiyin bulk
;

temperature

this is

hence 90

sion of water

from 42 +

6 of

leaves

-r-

12

7^ of true mean

common scale this taken

of common scale for the
;

36''

to iron as

is

of

nearly -^ of the expan-

temperature at which water

Copper

180''

by

in length

is

of greatest density.

2 in expansion

there-

fore if 6 be the allowance for iron, that for copper

must be

hence 45

""

36, for

the

temperature as before.
Brass expands about sV more than copper ;
shall have 45l
9i = 36, for the
temperature as above.

hence we

Fine pewter

is

to iron as 11

according to Smeaton

6 in expansion,

pewter

but this being a mixed metal,

much

as

Smeaton

much

scale

= 35%

it

is

as

if

we may

hence water expands 5^ times

zinc
and 90 -h 5| = 17 of new
:

13l of old scale

whence 48

13| =

34^ or the temperature derived from zinc.

seems highly probable that

E

not

to be relied upon.

Zinc expands ttt in bulk for 180,

credit

11

temperature as derived from the vessel of

for the

so

hence 46

in this case the

It

expan-

34

ON EXPANSION.

sion of the vessel

is

over-rated

was found to

it

be less than that of lead, whereas Smeaton makes

it

The

more.

it

made

vessel was

Hodson and

malleable zinc of

contains a portion of

tin,

of the patent

Perhaps

Sylvester.

which

will

account for

the deviation.

Lead expands

tt^ of

its

bulk for 180

^=

gives 90 -h
scale

161 of

whence 49

new

13 =

scale

water

much

therefore expands about 5^ times as

this

13 of old

36, as before.

From these experiments it seems demonstrated,

that the greatest density of water

is

near

at or

the 36 of the old scale, and 37 or 38 of the

scale:
is

new

and further, that the expansion of thin glass

nearly the same as that of iron, whilst that of

stone ware

is |,

and brown earthen ware J of the

same.

The

apparent expansion of mercury

in a ther-

mometrical glass for 180 I find to be .0163 from

1.

That of thin glass may be stated

2To>

which

is

at

rather less than iron, y^y.

.0037

Con-

sequently the real expansion of mercury from

32 to 212
or

sV'

authors

is

equal to the

De Luc

makes

make

less

it

it

sum

of these

.02

.01836, and most other

because they have

der-rated the expansion of glass.

all

un-

Hence we

derive this proportion, ,0163: 180:: .0037: 41

35

ON EXPANSION.

nearly, which expresses the effect of the expansion of glass on the mercurial
is,

thermometer

that

the mercury would rise 41 higher on the scale

at the

temperature of boiling water,

had no expansion.

De

if

the glass

Luc makes the expansion

of a glass tube from 32 to 212

.00083 in

length, and from 32' to 122^ only .00035.

This

inequality arises in part at least, I apprehend, from

the want of equilibrium in the original fixation of

glass tubes, the outside being hard

when the inside

is soft.

Liquids

may be denominated pure when they

are not decomposed by heat and cold.

deemed such because

affected by temperature.

of salts in water cannot be

their

constitution

Thus,

if

is

Solutions
;

a solution of sulphate of soda in water

be cooled, a portion of the

salt crystallizes,

and

leaves the remaining liquid less saline than before

whereas water and mercury, when partially congealed, leave the remaining liquid of the same
quality as before.

Most

acid liquids are similar

Alcohol as we

to saline solutions in this respect.

commonly have

it, is

a solution of pure alcohol in

a greater or less portion of water

would be affected by congelation

tions.

Ether

is

like other solu-

one of the purest

water and mercury.

and probably

liquids,

Oils, both fixed

and

except

volatile,

ON EXPANSION.

36

are probably for the most part impure, in the sense

we

use

it is

Notwithstanding these observations,

it.

remarkable how nearly those liquids approxi-

mate to the law of expansion observed

Few

and mercury.

authors have

ments on these subjects

experi-

My

own

in

inves-

have been chiefly directed to water and

tigations

mercury

my

made

and their results

several instances are incorrect.

of

water

in

but

it

may be proper

to give the results

enquiries on the other liquids as far as they

have been prosecuted.

Alcohol expands about ^ of

from

this liquid are

given by

but the results of

accord with

35 parts

The

to 172.

my

as to fire

The

to

I find

an indefinite

this is

judge

it

must have been

temperature 50 became 1079

scale: at 110 the alcohol

is

common

.86, I find

1072 at 170'

mercurial

at 1039, or half a

below the true mean.

is

test.

1000 parts of alcohol of .817

temperature 170 of the

gravity

for the se-

strength of his alcohol was such

sp. gravity at the

division

to 212;

him alcohol expands

and 45 parts

gunpowder but

very weak.

from 32

experiments do not seem to

From my experiments

at the

De Luc

for the first 90,

cond 90.

bulk for 180,

relative expansions of

According

his.

its

1000 parts

at 110' the bulk

is

When

the sp.

at 50

become

1035 + whence
,

ON EXPANSION.

3?

the disproportion of the two parts of the scale

much

not so

in this case as

sp. gravity is .937, 1 find

and 1029^

170,

at

at

gravity

is

to 37.

29^

hence the
to 32^.

45

and 1017^
which

is

at 110,

the same as

It is true

hol.

ratio of

When

the

become 1040

at

giving a ratio of 35 to

De Luc

gives for alco-

he takes an interval of tempera-

ture == 180, and I take one for 120 only

still it is

is

the

.967, answering to 75 per cent,

water, I find 1000 parts at 50

170,

When

1000 parts become 1062

110

the expansion becomes

sp.

35

impossible to reconcile our results.

but

As

the expansion of alcohol from 172 to 212 must

have been conjectural, perhaps he has over-rated

In reporting these results I have not taken

it.

into account the expansion of the glass vessel, a

large thermometrical bulb, containing about 750

grains of water, and having a tube proportionally

wide

consequently the real expansions must be

considered as more rather than less than above

stated.

The

graduation of the vessel having been

repeatedly examined, and being the same that

was used
I

in

determining the expansion of water,

can place confidence in the results.

Particular

care was taken in these experiments to have the

bulb and stem both immersed in water of the pro-

posed temperature.

ON EXPANSION.

38

As

alcohol of .817 sp. gravity contains at least

8 per cent, water,

that a

it is

fair to infer

from the above

thermometer of pure alcohol would

apparent degree

differ

in

interval of temperature from 50 to 170.

when we consider

no

from one of mercury in the

But

that the relative expansions of

glass,

mercury and alcohol

as 1,

5^ and 22 respectively,

for this interval, are

it

must be obvious

that the inequality of the expansion of glass in the

higher and lower parts of the scale, which tends

to equalize the apparent expansion of

has

little

influence on alcohol,

presumed

that a spirit

equable in

its

by reason of

Hence

comparative insignificance.

mercury,

it

its

may be

thermometer would be more

divisions than a mercurial one, in

This

a vessel of uniform expansion.

it

ought to

be by theory, because the point of greatest density,

or congelation of alcohol,

is

below that of

mercury.

Water being densest

at 36,

very remote temperature below,

and alcohol
it

was

to

at a

be ex-

pected that mixtures of these would be densest at

intermediate temperatures, and those higher as
the water prevailed
tion, so

thus

we

find the dispropor-

observable in the expansion of water,

growing greater and greater

they approach to pure water.

in the

mixtures as

ON EXPANSION.

Water saturated with common

follows

1000 parts at 32

122

at

39
expands

salt

become 1050

at

as

212

nearly 1023, which gives the ratio

it is

of 23 to 27 for the corresponding equal intervals

This

of mercury.

nearly the same as

is

ratio of 36.3 to 43.7.

congeal

at

7,

This solution

De Luc's
is

said to

and probably expands nearly

as

the square of the temperature from that point. It

from most other saline solutions

differs

Olive and linseed

oils

by 180 of temperature
pansion of olive
with

in

regard

expansion by temperature.

to its

me

it is

oil

more

expand about 8 per

;

De Luc

cent,

finds the ex-

nearly correspond to mercury

disproportionate, nearly agree-

ing with water saturated with

salt.

Oil of turpentine expands about 7 per cent, for

180

expands much more

it

in the

in the lower part of the scale, as

it

higher than

ought to do,

the freezing point being stated at 14 or 16.

ratio is

somewhere about 3

have

that oil of turpentine boils at 560

not

it

know how

below 212,

to 5.

The

Several authors

the mistake originated

but

it

do

boils

like the rest of the essential oils.

]

.85 expands about

6 per cent, from 32 to 212.

It accords with

Sulphuric acid,

mercury

as

the scale.

sp.

gravity

nearly as possible in every part of

Dr. Thomson says the freezing point

ON EXPANSION.

40

of acid of this strength

whence

it

mercury.

is

at

36

accords with the same

I find that

at 45,

and

even the glacial sulphuric

acid, or that of 1.78 sp. gravity,

congealed

or below

law as water

which remains

expands uniformly, or nearly

like the other, whilst

it

continues liquid.

Nitric acid, sp. gravity 1.40, expands about 11

per cent, from 32 to 212

ly of the

same rate

the expansion

as that of

is

near-

mercury, the

dis-

proportion not being more than 2? to 28 or thereabouts.

is

The

freezing point of acid of this strength

near the freezing point of mercury.

Muriatic acid,

sp.

gravity 1.137, expands about

6 per cent, from 32 to 212

it is

more

dispro-

portionate than nitric acid, as might be expected,

The

being so largely diluted wath water.

is

ratio

nearly 6 to 7.

Sulphuric ether expands after the rate of 7 per

cent, for 180 of temperature.

pared the expansion of

have only com-

this liquid

mercury from 60 to 90.

with that of

In this interval

it

accords so nearly with mercury that I could perceive no sensible difference in their rates.
said to freeze at

From what

It is

46.

has been observed

it

may be

seen

that water expands less than most other liquids

yet

it

ought to be considered as having

in reality

ON EXPANSION.

41
Alcohol and

the greatest rate of expansion.

nitric acid,

expand

which appear to

not excel, or even equal water,

their expansion

density, and

much, do

we

estimate

from the temperature of greatest

compare them with water

cumstances.
at

if

so

It is

in like cir-

because we begin with them

100 or 200 above the point of greatest density,

and observe their expansion

for 180 further, that

they appear to expand so largely.

Water,

if it

continued liquid, would expand three times as

much
the

in the

first,

second interval of 180 as

reckoning from

does in

it

36.

EXPANSION OF SOLIDS.
No

general law has hitherto been discovered

respecting the expansion of solid bodies

elastic fluids

and liquids appear

to

but as

be subject to

their respective laws in this particular,

we may

confidently expect that solids will be found so too.

As

it

may be presumed

that solids

undergo no

change of form, by the abstraction of heat,

may

probable that whatever the law

be,

it

it

is

will

respect the point atwhich temperature commences,

or what

may be

called, absolute cold.

our present business to enquire

point

is

but

it

may be observed

It is

how low

not
this

that every phe-

ON EXPANSION.

42

nomenon

indicates

lower than
it

may

it

to

be very low, or much

commonly apprehended.

is

Perhaps

hereafter be demonstrated that the inter-

val of temperature from 32 to 212 of Fahrenheit,

constitutes the 10th, 15th, or 20th interval from

absolute cold.

Judging from analogy, we may

conjecture that the expansion of solids

sively increasing with the temperature

ther

it

fluids,

is

is

progres-

but whe-

a geometrical progression as elastic

or one increasing as the square of the tem-

perature, like liquids, or as the 3rd or any

of the temperature,
absolute cold,

it

if it

still

must appear

power

be estimated from

to

be nearly uniform,

or in arithmetical progression to the temperature,

for so small

as that

and remote an interval of temperature

between freezing and boiling water. The

truth of this observation will appear from the fol-

lowing calculation

let us

question to be the 15th

suppose the interval in

then the real tempera-

ture of freezing water will be 2520, the mid-way

to boiling 2610,

and boilingwater 2700, reckoned

from absolute cold.

14)^

43

ON EXPANSION.

Now

the differences above represent the ratios

of expansion for 90 of temperature

they are in

the former case as 57 to 59, and in the latter as

14 to 15 nearly.

supposed

mean

is

But the temperature being

be measured by the new

to

about 110 of the old scale

scale, the

therefore

the expansion of solids should be as 57 or 14 from

32 to 110, and as 59 or 15 from 110 to 212 of

If these conjectures

the old scale.

be

right, the

expansion of solids ought to be something greater

in the lower part of the old scale,
less in the

and something

Experience

higher part.

present

at

For

does not enable us to decide the question.

all practical

purposes we

may adopt

the equable expansion of solids.

the notion of

Only

glass has

been found to expand increasingly with the temperature, and this arises probably from
constitution, as has

its

peculiar

been already observed.

Various pyrometers, or instruments for measuring the expansion of solids, have been invented, of

which accounts maybe seen

Their object

losophy.

is

in books of natural phi-

to ascertain the

sion in length of any proposed subject.

expan-

The

longi-

tudinal expansion being found, that of the bulk

may be

derived from

much. Thus,

if

it,

and

will

be three times as

a bar of 1000 expand to 1001

by a

certain temperature; then 1000 cubic inches of the

same

will

become 1003 by the same temperature.

ON EXPANSION.

44

The

following Table exhibits the expansion of

the principal subjects hitherto determined, for

180^ of temperature

that

The bulk and

Fahrenheit.

at 32 are denoted by

from 32

is,

to

212 of

length of the articles

1.

EXPANSION.
In bulk.

SOLIDS.

Brown earthen

ware....

Stoneware
Glass

rods and tubes,

bulbs (thin)

In length.

0012=^
.0025=^
.0025=^
.0037=^

Platinum

.0034=^
.0038=^
.0042=^
.0042=^

Steel

Iron

Gold
Bismuth
Copper

240
12
8

10

115
8 8 2
1

T9~0
1

TTT
TTT
1
1

.0051

100

5 8 8
1

33

Silver

.0056=,,,
.0060=1^0

Fine Pewter.

.0068=-ri^

440

Brass

'0074=^
.0086=^
0093=^

Tin

Lead
Zinc
LIQUIDS.

Mercury
Water
Water sat. with

.0200=3^
.0466= 2iy
salt.

Sulphuric acid
Muriatic acid
Oil of turpentine
Ether

Fixed oils
Alcohol

.0500=^
.0600=^
.0600=:iV
.0700=-iV

.0700=^
.0800=yi^

Nitric acid

.1100= i
.1100= ^

ELASTIC FLUIDS.
Gases of all kinds

.376

Smealon,

Elhcott.

4
t

Boida.

1
1

Too"

3 2 2

on expansion.

45

Wedgwood's thermometer.

The

spirit

thermometer serves

greatest degrees of cold

we

to

measure the

are acquainted with,

and the mercurial thermometer measures 400

above boiling water, by the old

new

250 by the

mercury
very

An
is

the best

clay,

and

This

is

short of red heat, and

tempera-

instrument to measure high tempera-

very desirable

we have yet

improvement.

for

or about

one, at which temperature the

far short of the highest attainable

ture.

tures

boils.

scale,

composed

and Mr. Wedgwood's

but there

is still

is

great room

Small cylindrical pieces of

in the

manner of earthen ware,

slightly baked, are the thermometrical pieces.

When

used, one of

them

is

exposed

in a crucible

to the heat

proposed to be measured, and

cooling,

found to be contracted, in proportion

it is

to the heat previously sustained

after

the quantity of

contraction being measured, indicates the temperature.

The whole range

of this thermometer

is

divided into 240 equal degrees, each of which

is

calculated to be equal to 130 of Fahrenheit.

The

lowest, or 0,

is

found about 1077 of Fah-

renheit (supposing the

common

scale continued

above boiling mercury,) and the highest 32277.

According

to the

new views

of temperature in the

ON EXPANSION.

46

preceding pages, there

numbers are much too

The

is

reason to think these

large.

following Table exhibits some of the

more

remarkable temperatures in the whole range, according to the present state of our knowledge.
Wedg.

Extremity of Wedgwood's thermometer

240

Pig

150

and

iron, cobalt

nickel, melt

from 130 to

Greatest heat of a Smith's forge

125

Furnaces for glass and earthen ware, from 40 to

124

Gold melts

32

Settling heat of flint glass

29

Silver melts

28

Copper melts

27

Brass melts

21

Diamond burns
Red heat visible

14
in day-light
Fahrenheit,
old scale.

Hydrogen and charcoal burn 800

to

1000

Antimony melts

809

Zinc

700

Lead

612

Mercury boils

600

Linseed

600

oil boils

Sulphuric acid boils

590

Bismuth

476

Tin

442

Sulphur burns slowly

303

Nitric acid boils

240

Water and

212

essential oils boil...

ON SPECIFIC HEAT.

4?
Fahrenheit.

Bismuth 5

parts, tin 3

and lead

2,

210

melt

Alcohol boils

174

Bees wax melts

142

Ether boils

98

Blood heat 96 to

98

Summer

heat in this climate 75 to

Sulphuric acid

(^1.78)

80

when congealed, begins

to melt..

45

Mixture of ice and water

32

Milk freezes

30

Vinegar freezes

28

Strong wines freeze about

20

Snow

3 parts, salt 2

Cold observed on the snow

at

10

Kendal, 1791

Ditto at Glasgow, 1780

Mercury

freezes

Greatest

artificial

23
39

90

cold observed

SECTION

3.

ON THE

SPECIFIC

HEAT OF

BODIES.

If the whole quantity of heat in a

measure of

water of a certain temperature be denoted by

that in the

noted by

same measure of mercury

.5

nearly

hence the

by

and

be de-

specific heats of

water and mercury, of equal hulks,

fied

will

1,

may be

signi-

.5 respectively.

If the specific heats

be taken from equal weights

ON SPECIFIC HEAT.

48

of the two liquids

and .04 nearly

then they will be denoted by

because we have to divide

.5

by

13.6, the specific gravity of mercury.

That bodies
is

differ

much

in their specific heats,

manifest from the following facts.

If a

1.

measure of mercury of 212 be mixed

with a measure of water of 32, the mixture will

be

far

below the

mean temperature.

If a measure of

2.

mercury of 32 be mixed

with a measure of water of 212, the mixture will

be

far
3.

above the mean.

If two equal

and

like vessels

be

the

filled,

one with hot water, the other with hot mercury

the latter will cool in about half the time of the

former.
If a

4.

measure of sulphuric acid be mixed with

a measure of water of the same temperature, the

mixture

will

assume a temperature about 240

higher.

These
ous

facts clearly

affinities for heat,

shew that bodies have

and that those bodies which

have the strongest attraction or

possess the most of

vari-

it

affinity for heat,

in like circumstances

in

other words, they are said to have the greatest

capacity for heat, or the greatest specific heat.

It is

found too that the same body changes

capacity for heat, or apparently assumes a

its

new

ON SPECIFIC HEAT.
affinity,

with a change of form.

This no doubt

new arrangement

or disposition of

arises from a
its

by which

ultimate particles,

of heat are influenced

on becoming
heat,

49

Thus

a solid body, as ice,

liquid, acquires a larger capacity for

even though

liquid, as

their atmospheres

its

bulk

is

diminished

water, acquires a larger

and a

capacity for

heat on being converted into an elastic fluid

last increase is occasioned,

by

ly

we may

this

conceive, sole-

being increased in bulk, in consequence

its

of which every atom of liquid possesses a larger

sphere than before.

very important enquiry

body

same

in the

capacity

state

is,

w^hether the same

undergoes any change of

by change of temperature.

Does

water,

for instance, at 32 possess the same capacity for

heat, as at 212,

degrees

and through

all

the intermediate

Dr. Crawford, and most writers

after

him, contend, that the capacities of bodies in such

As

circumstances are nearly permanent.

line of doctrine this

may be

admitted

requisite, if possible, to ascertain,

small change of capacity induced

is

an out-

but

it

is

whether the

by temperature,

such as to increase the capacity, or to diminish

and

also,

whether the increase or diminution

it

is

uniform or otherwise.

it is

of

little

Till this point

is

settled,

use to mix water of 32 and 212%

ON SPECIFIC HEAT.

50

mean tempera-

with a view to obtain the true

ture.

That water increases

in its capacity for heat

with the increase of temperature, I consider de-

monstrable from the following arguments

ist.

measure of water of any one temperature being

mixed with a measure

the mixture

is

condensation

any other temperature,

at

two measures.

less than

of volume

is

certain

Now

mark of

diminution of capacity and increase of temperature,

whether the condensation be the

effect of

chemical agency, as in the mixture of sulphuric

acid and water, or the effect of mechanical pres-

When

sure, as with elastic fluids.

2nd.

body suddenly changes

capacity

of form,

it

always from a

is

the temperature ascends

and vapour.

its

3rd.

the same

by a change

less to a greater, as

for instance, ice, water,

Dr. Crawford acknowledges

from his own experience, that dilute sulphuric

acid,

and most other

liquids

he

tried,

were found

to increase in their capacity for heat with the in-

crease of temperature.

Admitting the force of these arguments,

lows that when

it fol-

water of 32 and 212 are mixed,

and give a temperature denoted by 119 of the

common thermometer, we must conclude that the

true mean temperature is somewhere below that

ON SPECIFIC HEAT.
degree.

have already assigned the reasons why

mean

I place the

With respect

at 110.

the question whether water

to

uniformly or otherwise in

varies

am

51

its

capacity, I

inclined to think the increase, in this respect,

will

be found nearly proportional

to the increase

in bulk,

and consequently

will

be four times as

much

212 as at the mean.

Perhaps the ex-

at

pressions for the bulk

if so,

may

serve for the capacity;

the ratios of the capacities at 32, 122 and

212 of the

and 23.

new

scale,

may be denoted by

22,

22|

should rather expect, however, that

the ratios are

much nearer

and that 200,

equality,

201 and 204, would be nearer the truth.*

* In the Lectures

delivered in Edinburgh and Glasgow

in the spring of 1807, I

city of water at 32

The

gave

it

as

was to that

my

opinion that the capa-

at 212, as 5 to 6, nearly.

opinion was founded on the fact I had just before ob-

served, that a small mercurial thermometer at the temperature 32 being plunged into boiling water, rose to 202 in

15"

but the same at 212 being plunged into ice-cold water,

was 18"

in descending to 42

estimating the capacities to

be reciprocally as the times of cooling,

to 6.

On more

difference

capacities, as

212

is

mature consideration

occasioned, not so

is

more

by the
fluid

too,

find, that

gave the

am

much by

ratio of 5

persuaded

this

the difference of

different degrees of fluidity.

Water of

than water of 32, and distributes the

temperature with gi-eater

ment

it

facility.

mercury cools

By

a subsequent experi-

thermometer twice as

52

ON SPECIFIC HEAT.
Dr. Crawford, when investigating the accuracy

of the

common thermometer, was

aware, that

if

equal portions of water of different temperatures

were mixed together, and the thermometer

ways indicated the mean,

proof of

its

this

He

accuracy.

was not an

allows that

al-

infallible
if

water

have an increasing capacity, and the mercury ex-

pand increasingly with the temperature, an equation

may be formed

in fact the case in

some degree

have been deceived by

to

capacity of water,

This

so as to deceive us
;

and he appears

Yet the increased

it.

by no means

is

is

sufficient

to

balance the increased expansion of the mercury,

as appears

from the following experiments.

I took a vessel of tinned iron, the capacity of

which was found to be equal to 2

into this

the

sum

oz. of

water

were put 58 ounces of water, making

60 ounces

The whole was

of water.

raised to any proposed temperature, and then two

ounces of ice were put

in

and melted

perature was then observed, as follows

the tem-

+ 2 oz. ice of 32, gave 200^

130 + 2 oz. ice of 32, gave 122
50 + 2 oz. ice of 32, gave 45.3

60

oz.

water of 212

60

oz.

water of

60

oz.

water of

fast

as

water, though

it

has but half

the times in which a thermometer

is

its

capacity for heat

in cooling in fluids, are

not, therefore tests of their specific heats.

ON SPECIFIC HEAT.

From

the

first

53

of these, 30 parts of water lost

11^ each, or 345, and 1 part water of 32 gained

1681

the

difference

345

1681

176|,

expresses the number of degrees of temperature

(such as are found between 200 and 212 of the

old scale) entering into ice of 32 to convert
into

made
150,

water of 32.

Similar

calculations being

we

find in the second,

for the other two,

and

These three result-

in the third, 128.

ing numbers are nearly as

it

follows that as

it

much heat

5, 6,
is

and

7.

Hence

necessary to raise

water 5 in the lower part of the old scale, as is

required to raise

it

7 in the higher,

and

6 in the

middle.*

METHODS OF FINDING THE SPECIFIC HEATS OF

BODIES.

The most
specific

obvious method of ascertaining the

heats of bodies that have no chemical

afiinity for water, is to

mix equal weights of water,

and any proposed body of two known tempera* Perhaps the above results

may account

for the diversity

in authors respecting the quantity of latent heat (improperly

so called) in water.

Respecting the doctrine of Black on

Latent Heat, see an excellent note of Leslie.

529.)

(Inquiry, page

ON SPECIFIC HEAT.

54
tures,

and

mark the temperature of the mix-

to

Thus,

ture.

if

pound of water of

32,

and a

pound of mercury of 212 be mixed, and brought

to a

common

temperature, the water will be raised

m degrees, and the mercury depressed

and

7i

degrees;

be

their capacities or specific heats will

versely as those numbers

heat of water

way

Black,

or,

specific heat of

Irvine,

/i

in-

specific

mercury.

In this

Crawford, and Wilcke, ap-

proximated to the capacities of various bodies.

Such bodies

as

have an

aflSnity for water,

may be

confined in a vessel of

known

ed into water so as

be heated or cooled, as

to

capacity,

and plungin

the former case.

The

results

already obtained by this method

are liable to two objections

sume the
their

1st.

the authors pre-

capacities of bodies while they retain

form are permanent

that

is,

the specific

heat increases exactly in proportion to the temperature

and 2nd, that the common mercurial

thermometer

is

a true test of temperature.

But

it

has been shewn that neither of these positions

is

warrantable.

The

calorimeter of Lavoisier and Laplace was

an ingenious contrivance for the purpose of

vestigating specific heat;

it

was calculated

the quantity of ice which any

in-

to

shew

body heated

to a

ON SPECIFIC HEAT.
given temperature could melt.

It

55

was therefore

not liable to the 2nd objection above.

Unfor-

tunately this instrument does not seem to have

answered well

in practice.

Meyer attempted to

find the capacities of dried

woods, by observing the times in which given

equal volumes of them were in cooling.

These

times he considered as proportionate to the capa-

bulk for bulk

cities

and when the times were

divided by the specific gravities, the quotient re-

(An-

presented the capacities of equal weights.

de Chemie Tom.

nal.

recommended

a similar

mode

given us the results of his

From my own
this

method

Leslie

30).

for

trials

experience I

am

on

has

since

and

liquids,
five of

them.

inclined to adopt

as susceptible of great precision.

The

times in which bodies cool in like circumstances

appear to be ascertainable

mon

this

way with uncom-

exactness, and as they are mostly very

ferent, a

The

very small error

results

mixture

is

of

little

dif-

consequence.

too I find to agree with those

by

and they have the advantage of not be-

ing affected by any error in the

thermometric

scale.

The

formulse for exhibiting the

phenomena of

the specific heats of bodies are best

conceived

from the contemplation of cylindrical vessels of

ON SPECIFIC HEAT.

'^6

unequal bases.

(See plate

Fig. 1).

1.

Suppos-

ing heat to be represented by a quantity of liquid

in

each vessel, and temperature by the height of

the liquid in the vessel, the base denoting the

zero or total privation of heat

then the specific

heats of bodies at any given temperature,

a:,

will

be denoted by multiplying the area of the several

bases by the height or temperature,

.r.

Those

be directly as the bases, or

specific heats too will

as the increments of heat necessary to produce

equal changes of temperature.

Let

and JV

hot bodies

and

the weights of two cold and

their capacities for heat at

the same temperature (or the bases of the cylinders)

the difference of the temperature of

the two bodies before mixture, reckoned in de-

grees

and n

m =

the elevation of the colder body,

the depression of the

ture, (supposing

then

we

warmer

after mix-

them to have no chemical

obtain the following equations.

1.

2.

m =-w c^W C

-\-

n=

d,

WCd

Wn

action);

ON SPECIFIC HEAT.

57

W Cn
4,

If

(7=

c,

If

J^=

w;,

wm

Wd

m=

then, 5.

j^^w

'

m
c

To

then, 6.

C=

find the zero, or point of absolute privation

of temperature, from observations on the change

of capacity in the same body.

and

C=

the greater capacity,

Let

the less,

the

number

m=

of degrees of the less capacity requisite to pro-

duce the change

in equal weights,

n=

ber of degrees of the greater capacity,

the num-

s =

the

whole number of degrees of temperature down to

zero

To

then,

7.

CX

8.

find

a:

= On =

Cn

cm

C-c

~ C-c

771.

the zero from mixing two bodies of

the same temperature which act chemically, and

produce a change of temperature.

Cy

C^

x,

be as before

mixture, and

let

Let

iv,

Wy

M = capacity of the

the degrees of heat or cold

58

ON SPECIFIC HEAT.

produced

then

=
(w+ W)

bodies will be

a:

quantity of heat in both

the

w+C W)

(c

(w-\-

W)

n.

Mn
(cw+CW)^
(w+ W)

9.

and x

It is to

has been

be regretted that so

made

W)

(w +

little

improvement

for the last fifteen years in this

Some

department of science.

most incorrect results are

notice of students

still

of the earliest and

obtruded upon the

though with the least

reflec-

have made great

tion their errors are obvious.

number of experiments with

a view to enlarge,

but more especially, to correct the Tables of

Specific Heat.

It

of the particulars.

may be proper
For

liquids I

some

to relate

used an egg-

shaped thin glass vessel, capable of holding eight

ounces of water

to this

was adapted a cork, with

a small circular hole, sufficient to admit the stem

of a delicate thermometer tube, which had two
small marks with a

file,

the one at 92, and the

other at 82, both being above the cork

when

the cork was in the neck of the bottle, the bulb

of the thermometer was in the centre of the internal capacity.

When

an experiment was made

the bottle was filled with the proposed liquid,

and heated a

little

above 92

it

was then

sus-

ON SPECIFIC HEAT.

pended

the middle of a room, and the time

in

when the thermometer was

accurately noted
92,

and again when

meter

at the

it

was

82, another

at

thermo-

same time indicating the temperature

The

of the air in the room.

vessel was found

The mean
as follow

59

capacity of the glass

oz. of water.

results of several experiments

were

AIR IN THE

KOOM

52.
Minutes.

Water cooled from 92

to 82, in

29

Milk (1.026)

29

Solution of carbonate of potash (1.30)

Solution of carbonate of

Ammoniacal

Common

ammonia

28i
28i

(1.035)

solution (.948)

28i

vinegar (1.02)

Solution of

common

salt,

Solution of soft sugar, 6

27i
88

W. +32

W+4

S. (1.197)

27

26^

S. (1.17)

Nitric acid (1.20)

26^

Nitric acid (1.30)

25i

Nitric acid (1.36)

25

Sulphuric acid (1.844) and water, equal bulks (1.535)... 23i

Muriatic acid (1.153)

21

Acetic acid (1.056) from Acet. Cop

21

Sulphuric acid (1.844)

19i

Alcohol (.85)

19i

Ditto

Hi

(.817)

Ether sulphuric (.76)

15^

Spermaceti

14

oil (.87)

These times would express

accurately

the

ON SPECIFIC HEAT.

60

specific heats of the several bodies,

bulk for bulk,

provided the heat of the glass vessel did not enter

But

into consideration.

proved

to

to f of an

be equal

to f of

as the heat of that

was

an ounce of water, or

ounce measure of

oil, it

evident

is

must consider the heat disengaged

we

in the 1st ex-

periment, as from 8 f ounces of water, and in the

last as

from 9 i ounce measures of

On

oil.

this

account the numbers below 29 will require a small

reduction, before they can be allowed to represent

the times of cooling of equal hulks of the different

liquids

in the last

experiment the reduction

be one minute, and

It

may be proper

less in all the

articles

preceding ones.

to observe, that the

do not depend upon one

sults

will

trial

above re-

of the several

most of the experiments were repeated

several times, and the times of cooling were found

not to differ more than half a minute

indeed, in

general, there was no sensible differences.

air in the

below
I

room was,

52, the

in

any

case, a little

If the

above or

due allowance was made.

found the specific heat of mercury, by mix-

ture with water, and

by the time of

its

cooling in

a smaller vessel than the above, to be to that of

water of equal bulk,

I

as, Sj5 to 1 nearly.

found the specific heats of the metals and

other

solids

after

the

manner of Wilcke and

ON SPECIFIC HEAT.
Crawford
glass

61

having procured a goblet, of very thin

and small stem,

then put v^ater into

found

its

capacity for heat

such that the

it,

vv^ater, to-

gether with the value of the glass in water, might

be equal to the weight of the

was raised

to 212,

The

solid.

solid

and suddenly plunged into the

water, and the specific heats of equal weights of

the solid and the water, were inferred to be in-

versely as the changes of temperature which they

experienced, according to the 6th formula.

Some

regard was paid to the correction, on account of

common thermometer, which was

occasion.
The solids I tried were

the error of the

used on the

copper, lead,

iron,

glass, pitcoal, &c.

zinc,

tin,

The results

antimony, nickel,
differed little

those of Wilcke and Crawford

from

numbers

their

may, therefore, be adopted without any material

error,

till

greater precision can be attained.

In

the following Table 1 have not carried the deci-

mals beyond two places

ence

will not

because present experi-

warrant further extension

the

first

place of decimals may, 1 believe, be relied upon

as accurate,

and the second generally

a few instances

except from

it

may, perhaps, be

this observation, the

of the gases by Crawford, on which

remark.

so,

but in

or 2 wrong;

specific heats
I shall

further

ON SPECIFIC HEAT.

62

TABLE OF SPECIFIC HEATS.

GASES.

21.40*

Hydrogen

Oxygen

Common

equal
eq.
weights, biks.

air...

SOLIDS.

002 Ice

4.73*

006

1.79*

002

Dried woods,
and
other vegetable
substances, from
.45 to

Quicklime
Carbonic acid.

vapour....

LIQUIDS.

Water
Arterial blood

Milk (1.026)
Carbonat. of ammon. (1.035)
Carbonat. of potash (1.30).
Solut. of ammonia (.948)....
Common vinegar (1.02)
Venous blood
Solut. of common salt (1.197)
Solut. of sugar (1.17)
Nitric acid (1.20)
Nitric acid (1.30)
Nitric acid (1.36)
Nitrate of lime (1.40)
Sulph. acid and water, equal b
Muriatic acid (1.153)
Acetic acid (1.056)
Sulphuric acid (1.844)

Alcohol(.85)
Ditto
(.817)
Sulphuric ether (.76)
Spermaceti oil (.87)

Mercury

002 Pit-coal (1.27)

Charcoal
001 Chalk
Hydrat. lime
1.55* 001 Flint glass (2.87)
Muriate of soda
Sulphur
.79*

Azotic.

Aqueous

1.05*

Iron
1.00 1.00 Brass
1.03*

Copper

l.OO Nickel
.98
.98 Zinc
.95
.98 Silver
.75
.98 Tin
1.03
.94 Antimony
.92
.89*
Gold
.78
.77
.76
.68
.63
.62
.52
.60
.66
.35
.76
.70
.66
.52
.04

.93 Lead
.90 Bismuth
.96
.88 Oxides of the metals
.85 surpass the
metals
.87 themselves,
accord.80 ing to Crawford.

JO

.70
.65
.65
.57
.50
.45
.55

eq.

63

ON SPECIFIC HEAT.

REMARKS ON THE TABLE.

The

articles

marked * are from Crawford.

Notwithstanding the ingenuity and address

dis-

played in his experiments on the capacities of the

elastic fluids, there is

reason to believe his results

are not very near approximations to the truth

we can never expect accuracy when

upon the observation of

depends

it

or 2 tenths of a degree

of temperature after a tedious and complicated

Great merit

process.

for the attempt.

The

is

undoubtedly due to him

difference

between

arterial

and venous blood, on which he has founded the

beautiful system of animal heat,

is

remarkable,

and deserves further enquiry.

From

the observed capacities of water, solution

of ammonia, and the combustibles, into

hydrogen
gravity,

which

enters, together with its small specific

we cannot doubt but

that this element

possesses a very superior specific heat.

Oxygen,

and azote likewise, undoubtedly stand high,

as

water and ammonia indicate but the compound

of these two elements denominated nitric acid,
;

being so low, compared with the same joined to

hydrogen, or water and ammonia, we must conclude that the superiority of the two last articles

due to the hydrogen they contain.

The elements, charcoal and sulphur, are remark-

is

chiefly

ON SPECIFIC HEAT.

64

ably low, and carry their character along with

them

into

compounds,

Water appears
for heat of

as

oil,

sulphuric acid, &c.

to possess the greatest capacity

any pure liquid yet known, whether

be compared with equal bulks or weights

it

may be

it

indeed

doubted, whether any solid or liquid

whatever contains more heat than an equal bulk

The

of water of the same temperature.

great

capacity of water arises from the strong affinity,

which both

its

elements, hydrogen, and oxygen,

Hence

have for heat.

that solutions of salts

it is

water, contain generally less heat in a given

in

volume than pure water

volume of water*

increase the

for, salts

as well as the density,

and hav-

ing mostly a small capacity for heat, they enlarge

the volume of the water more than proportional
to the heat they contribute.

Pure ammonia seems

heat,

to possess a

contains only about 10 per cent.

exhibited pure in a liquid form,

exceed water

the characters of

woods,

all fall

it

If

it

could be

would probably

in this particular.

The compounds

high specific

judging from the aqueous solution, which

of hydrogen and carbon, under

oil,

ether and alcohol, and the

below the two

From numerous experiments

last

mentioned

have lately made,

it

the
ap-

pears that the solutions of anhydrous salts do not increase

the volume of water.

J. D.,

1841.

ON SPECIFIC HEAT,

65

reason seems to be, because charcoal

ment of a low

The
in

acids form an interesting class of bodies

only one

who

be ,66

some other of

his results I find

found to have, whence

cent,

and should

as the

compound

63 per

have nearly as much heat

in

it

it

should seem that the

acid loses the principal part of

This

bining with water.

in muriatic acid,
its

the

remarkable that the water

It is

in acid of this strength is

water, and

is

specific heat of the acid 1.3 to

this with

rather too low.

is

Lavoisier

nearly correct in regard to nitric

is

he finds the

an ele-

specific heat.

regard to their specific heats.

acid

is

is still

its

heat on com-

more observable

which contains 80 per cent, of

specific heat is only .60

whence

not only the heat of the acid gas, but part of that
in the water

is

expelled on the union

this

ac-

counts for the great heat produced by the union

of this acid gas with water.

The

specific heat of sulphuric acid has

well approximatedby several.

make
.43,

it

.34, Lavoisier

.33

been

Gadolin and Leslie

Crawford

finds

it

but he must probably have had a diluted acid.

Common

vinegar, being water with 4 or 5 per

cent, of acid, does not

in its specific heat
at .10

it

diff*er

materially from water

has been stated at .39 and

but such results do not require animad-

ON SPECIFIC HEAT.

66

version.

The

acid I used contained

acetic

per cent, pure acid

Quicklime

Crawford
it

much

be .22

with

oil.

plunged

fixed at .28

(that

1 part, or

by Gadolin

it

experiments ; but I since find

The

have underrated

subject will

much

when

in water,

Hydrat of lime

parts and water

'

heat.

I think they

heat than carbonate of lime,

vessel and

determined by Lavoisier and

find quicklime to impart as

to

is

com-

this acid, therefore, in

bining with water, expels

33

or
is,

or

more

inclosed in a

when mixed
quicklime 3

dry slaked lime)

was .25 by
I

my

is

first

<

have underratedit.

be adverted to in a future section.

SECTION

4.

SPECIFIC HEAT OF
ELASTIC FLUIDS.

THEORY OF THE

Since the preceding section was printed

have spent some time

results already obtained cannot

it is difficult

ments
It is

in considering the consti-

tution of elastic fluids with regard to heat.

yet

off*,

to conceive

The

be relied upon

and execute experi-

less exceptionable than those of

Crawford.

extremely important, however, to obtain the

THEORY OF SPECIFIC HEAT.

exact specific heat of elastic

phenomena

67

because the

fluids,

of combustion and of heat in general,

and consequently a great part of chemical agency,

are intimately connected therewith.

In speaking of the uncertainty of Crawford's

on the specific heat of

results

must not be understood that

ly implicated.

The

all of

reiterated

elastic fluids,

it

them are equalexperiments on

by the combustion of hydrogen, in which it was found that 11 measures of

mixed gases, when fired by electricity heated
the heat given out

20.5 measures of water

2. 4

(page 263) at a

medium, were susceptible of very considerable

accuracy, and are therefore entitled

The comparative

to

heat of atmospheric

credit.
air

and

water, which rested on the observance of nearly

^ of a degree of temperature, is probably not very

from the truth but the very small diflerences

far

in

heats communicated

the

by equal bulks of

oxygen, hydrogen, carbonic acid, azotic gas and

common

air,

together with the great importance

of those differences in the calculation, render the

results
if w^e

very uncertain.

He

suppose the heats imparted by equal bulks

of these gases to be equal,

doctrine.

The

tenor of

estimate the heat of

equal

justly observes, that

weights

of

it

it

will

not affect his

necessarily led

him

to

oxygen high, compared with

acid and aqueous

carbonic

;;

THEORY OF SPECIFIC HEAT.

68

vapour, and of azotic gas or phlogisticated

as

it

was then

called,

under the idea of

its

being

an opposite to oxygen or dephlogisticated

Indeed

air,

air.

his deductions respecting azotic gas, are

not consistent with his experiments

no use of experiments 12 and

13,

for

he makes

which are the

only direct ones for the purpose, but he infers

the heat of azotic

ofas

from the observed

ence between oxygen and

result gives

it

less than half

differ-

common air. The

that of common air

whereas from the 13th experiment, scarcely any

sensible difference was perceived between them.

He

has in

all

probability

much underrated

but his errors in this respect, whatever they

it

may

be, do not affect his system.

When we

consider that

all

elastic fluids are

equally expanded by temperature, and that liquids

and
ral

solids are not so,

it

should seem that a gene-

law for the affection of elastic

fluids for heat,

ought to be more easily deducible and more

simple than one for liquids or solids.

There are

three suppositions in regard to elastic fluids which

merit discussion.
1.

the

Equal weights of

elastic fluids

may have

same quantity of heat under like circumand pressure.

stances of temperature

The

truth of this supposition

several facts

is

disproved by

oxygen and hydrogen upon

their

THEORY OF SPECIFIC HEAT.

union give out

much

69

though they form

heat,

steam, an elastic fluid of the same weight as the

elements composing

under similar circumstances.

unite
acid

Nitrous gas and oxygen

it.

is

stance of low specific heat, with

heat

is

with

it

oxygen

much

given out, which must be principally de-

rived from
contain

Carbonic

formed by the union of charcoal, a sub-

the

oxygen

heat,

little

then the charcoal

if

and the oxygen combining

be reduced, the carbonic acid must be

inferior in heat to

far

an equal weight of oxygenous

gas.

Equal

2.

hulks of elastic fluids

may have

the

same pressure and

saine quantity of heat with the

temperature.

This appears much more plausible

tion

of volume

hydrogen

is

when

a mixture of

converted into steam,

sioned by a proportionate
absolute heat

the same

the diminu-

oxygen and

may be

diminution

may be

occa-

of

the

said of a mix-

The minute
observed by Crawford, may have been

ture of nitrous gas and oxygen.

differences

inaccuracies occasioned

experiments.

But

which render
bable,

if

by the complexity of

this supposition

they

his

there are other considerations

extremely impro-

do not altogether disprove

Carbonic acid contains

its

own bulk

of

it.

oxygen

THEORY OF SPECIFIC HEAT.

70

formation must therefore

the heat given out at

its

be exactly equal

the whole heat previously-

to

contained in the charcoal

on

this supposition

but

by the combustion of one pound of charseems, at least, equal to the heat by the com-

the heat
coal

bustion of a quantity of hydrogen

produce one pound of water, and

or

to,

sufficient to

this last is equal

more than the heat retained by the water,

because steam

is

nearly twice the density of the

mixture from which

elastic

it

produced

is

it

should therefore follow, that charcoal should be

found of the same specific heat as water, whereas

it is

only about ^ of

it.

Were

this

supposition

true, the specific heats of elastic fluids of equal

weights would be inversely as their specific graIf that of steam or aqueous vapour

vities.

represented by

1,

8.4, azote

and carbonic acid

supposition

The

3.

.72,
is

oxygen would be

.64,
.46.

were

hydrogen

But

the

untenable.

quantity of heat belonging to the

ulti-

mate particles of all elastic fluids^ must he the

same under the same pressure and temperature.
It is

evident the

number

of ultimate particles

or molecules in a given weight or volume of one

gas

is

not the same as in another

for, if

equal

measures of azotic and oxygenous gases were

mixed, and could be instantly united chemically.

THEORY OF SPECIFIC HEAT.

71

they would form nearly two measures of nitrous

gas, having the

measures

same weight

as the

two original

but the number of ultimate particles

could at most be one half of that before the union.

No

two

the same

elastic

fluids,

probably, therefore, have

number of particles,

either in the

volume or the same weight.

given volume of any elastic

same

Suppose, then, a

fluid to

be constituted

of particles, each surrounded with an atmosphere

of heat repelling each other through the

medium

of those atmospheres, and in a state of equilibrium

under the pressure of a constant

force,

such as

the earth's atmosphere, also at the temperature

of the surrounding bodies

suppose further, that

by some sudden change each molecule of

endued with a stronger affinity for heat

air
;

was

query

the change that would take place in consequence

of this last supposition

can be given, as
particles

will

it

The

only answer that

appears to me,

is

this.

The

condense their respective atmos-

pheres of heat, by which their mutual repulsion

will

be diminished, and the external pressure

will

therefore effect a proportionate condensation in

the volume of air

neither an increase nor dimi-

nution in the quantity of heat around each molecule, or

around the whole,

the truth of the supposition, or as

called, proposition,

is

Hence
may now be

will take place.

it

demonstrated.

THEORY OF SPECIFIC HEAT.

72
Corel.

The

1.

specific heats of equal

weights

of any two elastic fluids, are inversely as the

of their atoms or molecules.

weights
o
2. The specific heats of equal hulks of elastic
are directly as their specific gravities, and

fluids,

inversely as the weights of their atoms.

3.

Those

have their atoms

elastic fluids that

the most condensed, have the strongest attraction

for heat

the greater attraction

is

spent in accu-

mulating more heat in a given space or volume,

but does not increase the quantity around any
single atom.
4.

When

affinity to

heat

is

two

elastic

form one

elastic atom,

When

disengaged.

thirds of their heat

when

union become

general,

heat retained as

One

three unite, then two

particles

n
will

specific attraction of
eff'ect

in

by chemical

the heat given out

m
it

And

disengaged, &c.

is

elastic
;

one half of their

is

to the

to n.

is

objection to this proposition

per to obviate

same

atoms unite by chemical

be

said,

it

may be pro-

an increase in the

each atom must produce the

on the system as an increase of exter-

nal pressure.

Now

this last is

known

to express

or give out a quantity of the absolute heat ; therefore the former

sion

must do the same.

must be admitted

and

it

This conclu-

tends to establish

the truth of the preceding proposition.

The heat

THEORY OF SPECIFIC HEAT.

<4

expressed by doubling the density of any

fluid

amounts

my

about 50, according to

to

mer experiments

this heat is not so

hundredth part of the whole,

elastic

much

as

the specific heat

interstitial

it

seems

one

be shewn

as will

and therefore does not materially

hereafter,

for-

aifect

be merely the

to

heat amongst the small globular mole-

and scarcely can be said

cules of

air,

to them,

because

it is

to

belong

equally found in a

vacuum

or space devoid of air, as

is

proved by the increase

of temperature upon admitting air into a vacuum.

Before we can apply

this doctrine to find the

specific heat of elastic fluids,

we must

first

ascer-

tain the relative weights of their ultimate particles.

Assuming

at present

after, that if the

be

1,

that of

what

will

be proved here-

weight of an atom of hydrogen

oxygen

will

be

7,

azote

5,

nitrous

gas 12, nitrous oxide 17? carbonic acid 19, am-

moniacal gas
gas

6,

carburetted hydrogen

6, nitric acid 19,

7, olefiant

carbonic oxide 12, sulphu-

retted hydrogen 16, muriatic acid 22, aqueous

vapour
16;

we

8,

ethereal vapour 11, and alcoholic vapour

shall

have the

specific heats of the several

elastic fluids as in the following table.

to

compare them with that of water, we

In order
shall fur-

ther assume the specific heat of water to that of

steam as 6

to 7, or as 1 to

L166.

THEORY OF SPECIFIC HEAT.

74

TABLE OF THE SPECIFIC HEATS OFELASTIC FLUIDS.

Hydrogen

9-333

Olefiant gas

Azote

1.866

Nitric acid

491

Oxygen

1.333

Carbonic oxide

.777

1.759

Sulph. hydrogen

583

Atmos.

air

1.555

424

Nitrous gas

777

Muriatic acid

Nitrous oxide

549

Aqueous vapour

Carbonic acid

491

Ether, vapour

848

1.555

Alcohol, vapour

586

1.333

Water

Ammon.

gas

Carb. hydrogen

Let US now see how

cord with experience.

common

heat of

Crawford found
excels
is

air
it

1.000

far these results will acIt is

remarkable that the

comes out nearly the same

by experiment

also,

the rest as he determined

all

much lower and

1.166

but oxygen

The

azote higher.

as

hydrogen
principles

of Crawford's doctrine of animal heat and combustion, however, are not at

all

affected with the

Besides the reason already assigned for

change.

thinking that azote has been rated too low,

we

see from the Table, page 62, that ammonia, a

compound

of hydrogen and azote, has a higher

specific heat than water, a similar

compound

of

hydrogen and oxygen.

Upon
fact

the whole, there

in regard to the

whether

elastic

is

not any established

specific

or liquid, that

heats of bodies,
is

repugnant to

ON HEAT BY COMBUSTION.
the above table as far as I

know

75

and

it

is

to

be

hoped, that some principle analogous to the one

here adopted,

may soon be extended

to solid

and

liquid bodies in general.

SECTION

5.

ON THE

QUANTITY OF HEAT EVOLVED

BY COMBUSTION.
When

certain bodies

oxygen, the process

and

is

is

unite

chemically with

denominated combustion,

generally accompanied with the evolution

of heat, in consequence of the diminished capacities

of the products.

voisier

and Laplace

The

fine

attempt of La-

to find the quantities of heat

disengaged during different species of combustion,

has not been followed up with the attention
deserves.

Perhaps

this

it

may have been owing to

the supposed necessity of using the calorimeter

of the above philosophers, and to a notion that

results

Much

are not

its

always to be depended upon.

important information may, however, be

obtained on this subject by the use of a very simple apparatus, as will appear from what follows

ON HEAT BY COMBUSTION.

76
I

when

took a bladder, the bulk of which,

tended with
water

air,

was

this

was equal
filled

30,000 grains of

to

with any combustible gas,

and a pipe and stop-cock adapted

to

it

a tinned

capable of containing 30,000 grains

vessel,

water was provided, and

ing found, so

ex-

much water

of

capacity for heat be-

its

w^as

put into

as to

it

make

the vessel and water together, equal to 30,000

grains of water.

The

gas was lighted, and the

point of the small flame was applied to the concavity of the bottom of the tinned vessel,

whole of the gas was consumed

till

the

the increase of

the temperature of the w^ater was then carefully

noted

whence the

effect of the

combustion of a

given volume of gas, of the

common

and temperature,

temperature of an

in raising the

pressure

equal volume of water, was ascertained, except a

very small
this

loss

of heat

method must be

by

radiation, &c.,

liable to,

which

and which probably

does not exceed | or y^th of the whole.

The mean

results of several trials of the differ-

ent gases are stated below

when

the experiments

are performed with due care, there

any sensible differences

same species of

gas.

is

scarcely

in the results with the

The

point of the flame

should just touch the bottom of the vessel.

ON HEAT BY COMBUSTION.
Hydrogen, combustion of

it

77

an equal volume of

raises

4.5

water

Coal gas, or carburetted hydrogen

10

Olefiant gas

14.

Carbonic oxide

4.5

Oil, alcohol,

&c.,

and the

Oil, spermaceti,

and ether, were burned

effect

observed as under

combustion of 10

grs. raised

in a lamp,

30,000 grs.
5.

water
Oil of turpentine

(much smoke unburnt)

Alcohol (.817)

2.9

Ether, sulphuric

3.1

Tallow and wax

5.2

Phosphor.

lOgrs. heated 30,000grs. water

Charcoal

Sulphur

Campli or

3 .5

Caoutchouc

2.1

The

were placed upon a con-

five last articles

venient stand, and burned under the vessel of

water

except charcoal, a piece of which was

and the combustion was

ignited, then weighed,

maintained by a gentle blast from a blow-pipe,

directing the heat as

bottom of the vessel

much

as possible

again weighed, and the loss ascertained

sult

never amounted

generally approached

upon the

after the operation

to 2
it

for ten

nearly.

it

was

the re-

grains, but

ON HEAT BY COMBUSTION.

78

In order to exhibit the comparative effects more

it may be proper to reduce the articles
common weight, and to place along with
them the quantity of oxygen known to combine
with them.
The quantity of heat given out may

clearly,

to a

by the number of pounds of

well be expressed
ice

which

it

would melt, taking

it

granted that

for

the quantity necessary to melt ice,

is

equal to

that which would raise water 150^ of the

scale.

The

results

maybe

new

seen in the following

table.

lb.

hydrogen takes
^"^
'"''

carbur. hydrogen

71bs.
^

oxygen, prod. 81bs. water, melts 3201bs.

_
_

5 & acid 85
_ 4.5
88
1.58
acid
25
4.5 w. and
104
_ _
60
_ _
58
4
62
\v.

car.

carb.

car. ac.

3.8

w. &

2.5 phos. acid

60

carb. acid

40

sulph. acid
car. ac.

20

70

ice,

ON HEAT BY COMBUSTION.
tus.

By

reducing Crawford's results to a com-

parative scale with Lavoisier's,

appear as follows

lb.

79

Hydrogen by combustion melts

Phoopborus

they will both

ON HEAT BY COMBUSTION.

80

experiments on the heat produced by the respiration of animals, support this supposition.

Wax

and Oil.

Crawford's

results are a little

lower than mine, which they ought not to be, and

are doubtless below the truth.
tainly cannot

Lavoisier's cer-

This great philoso-

be supported.

pher was well aware of the uncertainty of his

results,

He

and expresses himself accordingly.

seems not

to

have had an adequate idea of the

heat of hydrogen gas, which contributes so

to the quantity

given out by

compares, and expects

its

much

combustion

he

to find an equation, be-

tween the heat given out by burning wax,

&c.,

and the heat given out by the combustion of equal

weights of hydrogen and charcoal in their separate
state

but this cannot be expected, as both hy-

drogen and charcoal

must contain

of combination

when

separate, agree-

less heat than

ably to the general

on

in a state

combination.

law of the evolution of heat

In

fact,

both

Crawford and

Lavoisier have been, in some degree, led away

by

the notion, that oxygenous gas w^as the sole or

principal source of the light

by combustion.

This

is

the

and heat produced

more remarkable of

the former, after he had proved that hydrogenous

gas, one of the

most frequent and abundant com-

bustibles, possessed nearly five times as

much

heat

ON HEAT BY COMBUSTION.
as the

same weight of oxygenous gas.

81

Azote an-

other combustible, possesses as high and probably

higher specific heat than

tallow, alcohol, &c.,

oxygen.

would be

far

Oil,

wax,

from being low

the table of specific heat, provided a table

in

were formed comprehending bodies of every

class.

Charcoal and sulphur rank but low in the table.

Upon

guage of Crawford,
contain

which
the

we cannot adopt

the whole then,

little

is

air,

*'

the lan-

that inflammable

absolute heat," and

produced by combustion

bodies

that the heat

**

derived from

is

and not from the inflammable body."

This language may be nearly right as applied to

the ordinary combustion of charcoal and pitcoal

but cannot be so when

applied

universally to

combustible bodies.

After these remarks

to

it

is

almost unnecessary

add that the heat, and probably the

light also,

evolved by combustion, must be conceived to be

derived both from the oxygen and the combustible

body

know

and that each contributes,

aught we

to the contrary, in proportion to its specific

heat before the combustion.

tion

for

may be made upon

similar observa-

the heat produced

by the

union of sulphur with the metals, and every other

chemical union in which heat

Before we

conclude

this

is

evolved.

section

it

may be

ON THE ZERO OF TEMPERATURE.

82

proper to add, for the sake of those who are more

immediately interested in the economy of

of charcoal, and perhaps also of pitcoal,

cient (if there

were no

fuel,

by the combustion of

that the heat given out

loss) to raise

lib.

is suffi-

45 or 50lbs.

of water from the freezing to the boiling tem-

perature

or

it is

sufficient to

of water into steam.

coal be used, there

heat

lost,

more than

If
is

convert 7 or 8lbs.
this

weight of

a proportionate quantity of

which ought,

if

possible, to

SECTION

be avoided.

6.

ON THE

NATURAL ZERO OF TEMPERATURE;

OR ABSOLUTE PRIVATION OF HEAT.
If

we suppose

body

at the ordinary

tempera-

ture to contain a given quantity of heat, like as a

vessel contains a given quantity of water,

plain that

by

portions, the

the

fluid.

it

is

abstracting successively small equal

body would

It is

finally

be exhausted of

an object of primary importance

in the doctrine of heat to

determine,

how many

degrees of the ordinary scale of temperature a

body must be depressed before it would lose all

its heat, or become absolutely cold.
We have no

ON THE ZERO OF TEMPERATURE.

means of

effecting this

by

83

direct experiment

but

we can acquire data for a calculus, from which the

zero may be approximated with considerable accuracy.

The

data requisite for the calculus are the

exact specific heats of the several bodies operated

upon, and the quantity of heat evolved, or ab-

sorbed by bodies, in cases of their chemical combinations or otherwise.

These data are not

to

be acquired without great care and circumspection;

and hence thegreat diversity of the

obtained in this

difficult investigation.

to some, the zero

the

is

it is

According

estimated to be 900 below

common temperature

others,

results hitherto

according to

whilst,

nearly 8000 below the same. These

are the extremes, but various determinations of

an intermediate nature are

The most
and water

to

be found.

simple case in theory

is

that of ice

supposing the capacities of these two

bodies to be as 9 to 10, at the temperature of 32,

it is

known

that ice of 32 requires as

as would raise water 150, to convert

of 32, or to melt

it.

it

much

heat

into water

Consequently, according to

the 8th formula, page 57, water of 32 must contain ten times as

the

zero

much

heat, or 1500.

must be placed

at

temperature of freezing water.

That

is,

1500 below the

Unfortunately^

ON THE ZERO OF TEMPERATURE.

84

however, the capacity of

mined with
its

ice has not

sufficient accuracy, partly

been deterbecause of

being a solid of a bad conducting power, but

principally because the degrees of the

common

thermometer below freezing, are very erroneous

from the equal division of the

scale.

Besides the one already mentioned, the prin-

have been used

cipal subjects that

gation are,

water

1st,

mixtures of sulphuric acid and

2nd, mixtures of lime and water

mixture or combination of

and

in this investi-

4th, combustion of

charcoal.

Upon

nitric

3rd,

acid and lime

hydrogen, phosphorus and

these

will

it

be necessary to

enlarge.

MIXTURE OF SULPHURIC ACID AND W^ATER.

According

to the

experiments of Lavoisier and

Laplace on the calorimeter, a mixture of sulphuric

acid and water in the proportion of 4 to 3

weight,

determines

the

zero

freezing water, reckoning

at

by

7292 below

by Fahrenheit.

But

a mixture of 4 acid with 5 water, determines the

same

at 2630.

Gadolin made several experiments on mixtures

of sulphuric acid and w^ater, the results of which
are as accurate as can be expected in a
of the kind.

He

first

essay

has not determined the zero

ON THE ZERO OF TEMPERATURE.

85

from his experiments, but taking

it for granted to
be 1400 below the freezing point on the sup-

and water are

position that the capacities of ice

as 9 to 10, he has enquired

how far his experiments

corroborate the same, by comparing the capacities

of the mixtures

by experiment with those

lated from the previous assumption.

are thus curtailed in their utility

given us data

calcu-

His results
but as he has

sufficient to calculate the zero

each experiment,

it

will

from

be proper to see how^

far

they accord with Lavoisier's, or those of others.

Taking the
finds,

by

water at

specific heat of

1,

Gadolin

direct experiment, the specific heat of

concentrated

sulphuric

acid

Craw^ford on heat, page 465)

to
;

be

.339

(See

he then mixes the

acid and water in various proportions, observes

the increase of temperature, and then finds the

capacities

of the mixtures.

data to find the zero

Whence we have

by formula

9?

page 58.

In

giving his numbers, I have changed his scale, the

centigrade, to Fahrenheit's.
Lcid.

ON THE ZERO OF TEMPERATURE.

86

The mean

of these

is

2300, which

is

far be-

yond what Gadolin supposes to be the zero, as

deduced from the relative capacities of ice and
water, and to which he seeks to accommodate
these experiments.

As

the heat evolved upon the mixture of sul-

phuric acid and water

all

three

articles

is

so considerable,

and as

are liquids, and consequently

admit of having their capacities ascertained with

greater precision, I have long been occasionally

pursuing the investigation of the zero from experiments

on

these

liquids.

The

strongest

sulphuric acid of 1.855, I find has the specific

heat .33, and

Acid.

;;

ON THE ZERO OF TEMPERATURE.

87

give the zero less remote than otherwise

appears to be the case both with Gadolin and

I

this

me

have not yet been able to discover the cause of

it

perhaps the capacity of such mixture increases

with the temperature more than in the other cases.

LIME AND WATER.

Quicklime, that

is,

lime recently burned, has a

strong affinity for water

falls,

or

becomes

slaked,

minated hydrat of lime.

to

quicklime than

verize

it,

is

when mixed

proportion an intense heat

is

produced

in

due

the lime

and then may be denoIf

no more water

sufficient to slake

it,

is

put

or pul-

three parts of lime, by weight, form

four parts of hydrat, a perfectly dry powder, from

which the water cannot be expelled under a red

heat.

If

more water

is

added, the mixture forms

mortar, a pasty compound, from which the excess

of water

may be

expelled by a boiling heat, and

the hydrat remains a dry powder.

When hydrat

of lime and water are mixed, no heat

is

evolved

hence the two form a mere mixture, and not a

The heat then which is
chemical compound.
evolved in slaking lime, arises frt)m the chemical
union of three parts of lime and one of water, or

from the formation of the hydrat, and any excess

ON THE ZERO OF TEMPERATURE.

88

of water diminishes the sensible heat produced.

Before any use can be made of these facts for

determining the zero,
determine

the

For

lime.

it

this

becomes necessary

to

of dry hydrat

of

heat

specific

purpose a given weight of lime

the
be slaked with an excess of water
till
the
heat
expelled
by
then
be
must
excess

is

to

hydrat

is

of this powder
or any other

perature, and

By

ingly.

A given weight

i heavier than the lime.

may then be mixed

weight of water

its specific

with the same,

of another tem-

heat determined accord-

a variety of experiments

way, and with sundry variations,

made

in this

I find the specific

heat of hydrat of lime about .40, and not .25 as

in the table,

nearly

.30.

Lime

page 62.

Crawford

itself I

undervalues

mixing cold lime with hot alcohol

find to

be

lime,

by

the lime does

not produce a sufficient effect on the alcohol,

because
I

lime.

it

contains water, which acts upon the

have no doubt a different

would have been found,

poured on hot lime.

if

The

heat evolved in the

formation of hydrat of lime

follows

If 1 oz. of lime

specific heat

cold alcohol had been

may be found

be put into 4

as

oz. of water,

the temperature of the mixture will be raised

100

in this case

1^

oz.

the heat evolved raises

it

hydrat

is

formed, and

together with 3| oz.

ON THE ZERO OF TEMPERATURE.

water 100
heat that

but 3 water contains 7 times the

89

J hydrat of lime does

therefore the

sufficient to raise 8

times the

hydrat 100% or once the hydrat 800.

Whence

heat given out

is

by mixing 3 parts of lime and 1

sufficient to raise the new compound

the heat evolved

of water,
800.

we

is

Applying then the theorem

obtain the zero

in

page 58,

4260 below the common

temperature.

NITRIC ACID AND LIME.

According

to the

experiments of Lavoisier and

Laplace, the specific heat of nitric acid, sp. gr.

1.3, is .661,

and that of lime .217, and a compound

of 9j parts of said acid, and one of lime, is .619But supposing there was no change of capacity

upon combination,

compound should only

this

have the capacity .618; whereas,

in

fact,

the

mixture produces an increase of temperature of

about 180, and therefore ought to be found with
a diminished capacity, or one below .618.
this fact to

inexplicable

be established,

it

Were

would exhibit an

phenomenon, unless

to those

who

adopt the notion o^ free caloric and combined

caloric

existing in the

more properly, of

same body, or

caloric

combined so

to

speak

as to retain

ON THE ZERO OF TEMPERATURE.

90

characteristic properties,

all its

and

com-

caloric

One

bined so as to lose the whole of them.

error in this statement has already been pointed

regard to the capacity of lime.

out, in

adopt the specific heat of lime to be

.30,

we

If

and ap-

we shall find it
common temperature,

ply the theorem for the zero,

to

be 15770 below the

as

deduced from the above data so corrected.

I took a specimen of nitric acid of the specific

gravity 1.2, and found, by repeated

cific

heat to be .76 by weight.

trials, its

spe-

Into 4600 grains

of this acid of 35 temperature, in a thin flask,

657 grains of lime were gradually dropped, and

in one or two
the mixture moderately agitated
;

minutes after 3-4ths of the lime was in and

solved, the

the mixture was beginning to boil

ed to cool

20,

added, and

it

when

it

was

suffer-

the rest of the lime was

again rose to the boiling point

about 15 grains of insoluble residuum were

These were taken

by 15 grains of
and

left

be

out,

and

left.

their place supplied

fresh lime, which

were dissolved,

a clear liquid nearly saturated, of 1.334

sp. gravity.

to

dis-

thermometer rose nearly to 212, and

.69.

The specific heat of this was found

The increase of temperature being

called 200", and the specific heat of lime being

.30,

we

find the zero to be

11000 below the

ON THE ZERO OF TEMPERATURE.

The

temperature.

freezing

91

was

experiment

varied by taking acids of different strengths, and

various proportions of lime, but the results

still

gave the zero more remote than either of the

Perhaps the reason may be
previous methods.
that lime

is still

underrated.

COMBUSTION OF HYDROGEN.
Lavoisier finds the combustion of lib. of hy-

drogen to melt 295lbs. of

my

ice.

The

results of

experience give 320lbs, and Crawford's 480.

Till this fact can

we may
truth.

be more accurately ascertained,

take 400lbs.

as

approximating

Or, which amounts to the same

combustion of

lib.

to.

the

thing, the

of hydrogen takes 71bs. of

oxygen, and gives out heat which would raise

8lbs.

of water 7500.

By

adopting Crawford's

capacities of hydrogen and oxygen, and applying

the theorem, page 58,

from the

common

we

find the

temperature.

But

zero 1290
if

we adopt

the preceding theory of the specific heat of elastic

fluids,

page 72, we
the formation of steam, one

and apply the 4th

must conclude that

in

half of the whole heat of both

out
will

corol.

its

elements

is

given

the conversion of 8lbs. of steam into water,

give out heat sufficient to melt 561bs. of ice

ON THE ZERO OF TEMPERATURE.

92

therefore, one half of the whole heat in lib. of

hydrogen, and 71bs. of oxygen together, or which

is

the same thing, the whole heat in

drogen, or 71bs. of oxygen separately,

344lbs.

of ice

now

if

hy-

lib. of
will

melt

from 688 we take 400,

there remain 288 for the

lbs.

of

ice,

which the

heat in 8lbs. of water, at the ordinary temperature, is sufficient to melt, or the heat in lib.

capable of melting 361bs. of ice

will

is

hence the zero

be 5400 below freezing water.

COMBUSTION OF PHOSPHORUS.

One pound

of phosphorus requires Ijlb. of

oxygen, and melts 661bs. of

heat of phosphorus

is

one may suppose

it

From

whole heat

to

in

have as much heat as

is

nearly half as

the last article

each

melt 50lbs. of ice

specific

not known; but from analogy

wax, tallow, &c., which

water.

The

ice.

lb.

of

as

seems, that the

it

oxygen

is

sufficient to

whence the whole heat

articles,

previous to combustion,

melt 75

+ 18 =

93lbs. of ice.

oil,

much

is

both

in

sufficient to

From which

de-

ducting 66, there remains 27 for the pounds of

ice,

which the heat

ought to melt.

in 2.5lbs.

of phosphoric acid

This would give the

specific heat

of that acid .30, a supposition not at

all

impro-

ON THE ZERO OF TEMPERATURE.

The

bable.

93

combustion of phos-

result of the

phorus seems then to corroborate that from hydrogen.

COMBUSTION OF CHARCOAL.
Crawford's data are, specific heat of charcoal

oxygen 4.749, carbonic acid 1.0454, and the

.26,

heat given out by burning lib. of charcoal = 69lbs.

ice

10350\

doubt, that

oxygen

It

to convert

the zero

by

now

beyond

established

of charcoal requires 2.61bs. of

lib.

these data,

is

it

4400.

From
we deduce

into carbonic acid.

the theorem, page 58,

But Crawford himself has not

noticed this deduction.

If

we adopt

the theory

of specific heat, and the table founded on

it,

com-

bined with the supposition of the zero being 6000

below the common temperature, (see page 74),

we

shall

have from the general formula,

this equa-

tion,

(1+2.6) X .491 X h
i

X .26

2.6 X 1.333

3.6

x .461

6000^

where h represents the degrees of temperature

which the combustion of
raise the

From

lib. of charcoal

would

product, or 3.61bs. of carbonic acid.

this,

is

found

6650'.

But

this

heat

ON THE ZERO OF TEMPERATURE.

94
would

raise

3265.

or

it

of water

3.61bs.

Or it would raise

would melt 78lbs. of

6650 x .491
11750

lib. of water,

Lavoisier finds

ice.

it

69.

So that the supposed distance of the zero

is

not

the effect

and Crawford

961bs.,

finds

discountenanced by the combustion of charcoal,

as far as the theory

is

COMBUSTION OF

We

concerned.

OIL,

WAX, AND TALLOW.

do not know the exact constitution of these

compounds, nor the quantity of oxygen which

they require

but from the experiments of La-

voisier, as well as

am

from some attempts of

my

own,

inclined to think, that they are formed of

about 5 parts of charcoal and

of hydrogen

by

weight, and that 6 parts require 21 of oxygen for

their combustion, forming

acid and 8 of water.

zero

is

ice,

By

is

sufficient to

applying Cor.

shall find the heat in

of

be supposed that the

1, at

oxygenous gas

in carbonic acid, 22.3lbs.

oil,

melt 461bs.

then the heat in steam will be sufficient to

melt 53lbs.

and

it

6900 below freezing water, or that the

heat in water of 32,

of

Let

19 parts of carbonic

The

page 72, we

heat in lib.

&c., equal to half that of water

which being added

60.5lbs.,

23lbs.,

to 211.7, the heat in 3.5lbs. of

ON THE ZERO OF TEMPERATURE.

oxygen, gives 234.7lbs. of
all

oxygen

but the products

the heat in lib. of

1.31b. of water,

and

of ice

and 3.5 of

oil

of combustion are

3.2lbs. of carbonic acid, to-

gether containing as
131.2lbs.

would be

ice, whicli

melted by

95

much

this

heat as would melt

being subtracted from

234.7, leaves 103.5 for the ice to be melted

by

the heat evolved during the combustion of lib. of

wax, or tallow, which agrees with the experi-

oil,

ment.

The

conclusion then supports the sup-

position, that the zero

is

6900 below freezing

water.

COMBUSTION OF ETHER, &C.

I

have pretty accurately ascertained the pro-

ducts of the combustion of lib. of ether to be

1.75 water, and 2.25 carbonic acid, derived from

its

union with 3lbs. of oxygen.

By

instituting a

calculation similar to the above, but on the sup-

position of the zero being 6000 below freezing

water, I find the heat given out on the combustion of ether,

ought to be

671bs. of ice

observed to be 62, and the difference

be attributed to the

loss unavoidable in

it

may

was
well

my method

of observation.
I

might here enquire into the results of the

combustion of the other articles mentioned

table,

page 78, as

far as

in

the

they affect the present

ON THE ZERO OF TEMPERATURE.

96
question

but I consider those above noticed as

From

the most to be depended upon.

of olefiant gas

body

in the

we may

gaseous

state,

more heat than when

and

does not give out much

in a liquid state

olefiant gas certainly

their constitution

the result

learn, that a combustible

do not

differ

for, oil

much

in

one would therefore have ex-

pected the same weight of olefiant gas to have

yielded more heat than

oil,

because of the heat

required to maintain the elastic state

seem

but

it

should

that the heat requisite to convert a liquid to

an elastic

fluid,

but a small portion of the

is

whole, a conclusion evidently countenanced by

the experiments and observations contained in
the preceding pages.
It

of

may be proper now

my

into

to

draw up the results

experience, reported in the present section,

one point of view.

Zero below
32

Fahrenheit.

From

a mixture of 5.77 sulphuric acid

water

6400

4150

1.6

6000

lime

4260

7 nitric acid

From

and

the combustion of hydrogen

lime

11000

5400

phosphorus

5400

charcoal

6000

oil,

wax, and tallow

ether

6900
6000

ON THE ZERO OF TEMPERATURE.

97

We

are au-

The mean

of

thorised then,
to

all

till

these

6150.

is

something more decisive appear,

consider the natural zero of temperature as

being about 6000 below the temperature of

according to

freezing water,

The

Fahrenheit's scale.

results are not greater than

to inaccuracies,

lieve

it

wiir

the

divisions

what may be ascribed

except the 2nd and 5th.

be impossible

to each other, unless

of

diiferences of the above

I be-

to reconcile these

two

upon the supposition of

it is

a change of capacity with change of temperature

in

one or both of the mixtures.

This deserves

farther enquiry.

HEAT PRODUCED BY PERCUSSION AND FRICTION.

The

heat produced by the percussion and

tion of solid bodies, arises

fric-

from one and the same

cause, namely, from a condensation

of volume,

and consequent diminution of capacity of the excited

body

exactly in the same

condensation of

known

air

fact, that iron

manner

produces heat.

It is a well

and other metals, by being

hammered, become hot and condensed

at the

same time

as the

and

if

has not been observed

in

volume

a diminution of capacity
it

is

because

it is

small,

and has not been investigated with

sufficient ac-

That a change of capacity

actually takes

curacy.

ON THE ZERO OF TEMPERATURE.

98

when it is considered,
once hammered in this way,

place cannot be doubted,

that a piece of iron
is

unfitted for a repetition of the

been heated

in

fire

efi*ect, till it

and cooled

has

gradually.

Count Rumford has furnished us with some important facts on the production of heat by friction.

He

found that in boring a cannon for 30 minutes,

the temperature was raised 70

and that

it

suf-

fered a loss of 837 grains by the dust and scales

torn

On

off",

which amounted to g^part of the cylinder.

the supposition that all the heat

was given

out hy these scales, he finds they must have lost

66360 of temperature

But

ed.

this is manifestly

the subject

at the

an incorrect view of

how should hammering

body red hot without any

loss of scales ?

The

fact

less

condensed by the violence used

is,

same time

the heat excited does not arise from

the scales merely, else

make

when

specific heat not sensibly diminish-

he found their

the whole mass of metal

is

more or

in boring,

temperature of 70 or 100

is

too

small to produce a sensible diminution in

its

ca-

and a

rise of

pacity for heat.

Does Count Rumford suppose,

that if in this case the quantity of metal operated

upon had been

lib.

and the dust produced the

same as above, that the whole quantity of heat

evolved would have been the same

ON THE MOTION OF HEAT.

The phenomena

99

of heat produced by friction

and percussion, however,

sufficiently

shew

that

the zero of temperature cannot be placed at so

small a distance

common

as

1000 or 1500 below the

temperature, as has been determined by

some philosophers.

SECTION

7.

ON THE

MOTION AND COMMUNICATION OF

HEAT,
ARISING FROM INEQUALITY OF TEMPERATURE.

As from
bodies

is

various sources the temperature of

liable to

perpetual fluctuation,

it

becomes

of importance to determine the nature of the

motion of heat in the same body, and

in its

from one body to another, arising from

passage

its

inces-

sant tendency to an equilibrium.

solid bar

posed to the
the

air,

being heated

air,

the heat

at

is

one end, and ex-

partly dissipated in

and partly conducted along the bar, ex-

hibiting a gradation of temperature from the hot

to the cold end.

varies

This power of conducting heat

greatly, according

to

the nature

of the

ON THE MOTION OF HEAT.

100
subject:

in

which are good conductors of

electricity, are like-

wise good conductors of heat

When

and those bodies

general, metals,

a fluid

is

heated at

and vice versa.

surface, the heat

its

gradually and slowly descends in the same manner

as along a solid

and

fluids

seem

to

have a

diff'er-

ence in their conducting power analogous to that

But when the heat

of solids.

bottom of a vessel, containing a

very diflerent
in

is

fluid,

form an ascending current and

the contiguous particles, but

temperature

perature in the mass

and

still

is

in their ascent to

retaining a supe-

observed at the surface,

is first

till

ebullition in liquids, at

temperature

specific gravity,

so that the increase of tem-

constantly greatest there

ment of

is

fluid,

rise to the surface,

communicating a portion of heat

is

the case

the heated particles of the

consequence of their diminished

riority of

applied to the

uniform.

the

commence-

which period the

The conducting power

of fluids then arises from two distinct sources; the

one

is

the

same

as in solids, namely, a gradual

progress of the heat from particle to particle,

exclusive of any motion of the particles themselves

the other arises from the internal motion

of the particles of the

fluid,

by which the extremes

of hot and cold are perpetually brought into contact,

and the heat

is

thus

diffused

with great

ON THE MOTION OF HEAT.

The

celerity.

effectual
led,

latter source

much more

so

is

lOl

than the former, that some have been

though without

sufficient reason, to

doubt the

existence of the former, or that fluids do convey

heat in the same manner as solids.

Nothing appears, then, but that the communication of heat from particle to particle,
in the

same way

dity of

diffusion in fluids,

its

is

an hydrostatical law. But there

by which heat
and through

is

and by

fluids,

another method

By

which demands our

we receive the heat

when in a room, we re-

this

this,

ceive the heat of an ordinary

the radiation of heat
is

the rapi-

propagated through a vacuum,

elastic

particular notice.

of the sun

be ascribed to

to
is

performed

is

in fluids as in solids

It is called

fire.

and the heat, so propelled,

called radiant heat.

Till lately

light

we have been used

to consider the

and heat of the sun as the same thing.

But

Dr. Herschel has shewn, that there are rays of

heat proceeding from the sun, which are separable

by a prism from the rays of

light

ject to reflection, like light

but in a less degree, which

separability from light.

heat

is

not known, but

it

the same as that of light,

is

and

they are subto

refraction,

the cause of their

The velocity of radiant

may be presumed to be
till

something appears

ON THE MOTION OF HEAT.

102

to the contrary.

An

ordinary

fire,

red hot char-

coal, or

indeed any heated body, radiates heat,

which

capable of being reflected to a focus, like

the

is

and heat of the sun

light

seem

to

be not of

sufficient

but

energy

it

should

to penetrate

other transparent bodies so as to be

glass, or

refracted to an eflScient focus.

Several

new and important

facts relative to the

been ascertained by

radiation of heat, have lately

on Heat."

Enquiry
Having invented an ingenious and

delicate air

thermometer, well adapted for the

Professor Leslie, and published in his

*'

purpose, he was enabled to mark the effects of

radiation in a great variety of cases
stances, with

been done.

and circum-

more precision then had previously

Some

of the principal facts respecting

the radiation of heat, which have either been dis-

covered or confirmed by him,

it

will

be proper to

mention.
1.

If a given vessel be filled with hot water,

the quantity of heat which radiates from

chiefly

it,

depends

upon the nature of the exterior surface of

Thus,

the vessel.

if

a canister of tinned iron be

the vessel, then a certain quantity of heat radiates

from

it

if

the said vessel be covered with black

paint, paper,

times as

glass,

much heat

&c.,

it

will

then radiate 8

in like circumstances.

ON THE MOTION OF HEAT.

with

thermometer be covered

If the bulb of the

2.

103

the impression of the radiant heat

tinfoil,

is

only |th of that upon the glass surface.

3.

metallic mirror reflects 10 times as

heat from an ordinary

fire,

body, as a similar glass mirror does.

found to

reflect the heat

from

its

This

Here then

is

last is

anterior surface,

and not from the quicksilvered one, which

most essential

much

or from any heated

in reflecting solar light

is

the

and heat.

a striking difference between solar

and culinary heat.

From

these facts

other bodies which

it

appears, that metals and

are eminently disposed to

reflect radiant heat, are not disposed to absorb

in

any remarkable degree

paper, glass, &c., are disposed to absorb

consequently to radiate

it

it

whereas, black paint,

it,

and

again in proper circum-

stances.

Screens of glass, paper,

4.

tinfoil, &c.,

being

placed between the radiating body and the reflector,

were proved

completely

to

intercept the radiant heat

but being heated themselves by the

direct radiant heat, in time the

aff*ected

by

their radiation.

thermometer was

The

heat radiating

from hot water, does not then seem capable of

being transmitted through glass, like the solar
heat.

ON THE MOTION OF HEAT.

104
5.

Radiant heat suffers no sensible loss

passage through the

air

in its

a greater or less radiant

body produces the same

provided

effect,

it

sub-

tends the same angle at the reflector, agreeing

with light in this respect.
6.

The

intensity of reflected heat diminishes

inversely as the distance

the same at
differs

all

distances

whereas, in

from that of light;

light,

it is

the focus of heat, too,

it is

nearer the reflector;

the heating effect diminishes rapidly in going out-

wards, but slowly in going inwards towards the

reflector.

This

seems

intimate the want of

to

perfect elasticity in radiant heat.

7.

hollow globe of

meter, being

filled

four inches in dia-

tin,

with hot water, cooled from

35 to 25 centigrade in 156 minutes

the same

painted with lamp-black, cooled from 35 to 25

in

81 minutes.
8.

air,

When

The

a heated

air of the

body

is

room was

15.

whirled through the

the additional cooling effect

is

directly pro-

air the rate of cooling of a

hollow glass

portional to the velocity.

1).

globe

In

filled

with hot water, and that of the same

globe covered with

temperatures.

The

tinfoil, is

low temperatures, and less

the present ease,

not constant at

disproportion
in

Mr. Leslie

is

high.

all

greater in

Thus,

in

finds the variable

ON THE MOTION OF HEAT.

be as 105 + A for glass, and as 50 + A for

ratio to
tin,

105

where h represents the elevation of temperaAccording

ture in degrees.

to this the rate of

cooling of a vitreous and a metallic surface

nearly the same at very high temperatures

is

nearly as 105 to 50,

No

when h

is

very

is

but

little.

differences are observed in their rates of cool-

ing in water.

and

10. After a long

intricate,

but ingenious

Mr. Leslie finds the cooling power

upon a hollow sphere, six inches in

investigation,

of the air

diameter, and filled with boiling water, to be as

follows

namely,

in

each minute of time the

fluid

loses the following fractional parts of its excess

of temperature, by the three distinct sources of

refrigeration in the air undermentioned

By

abduction, that

power of

By

air,

the proper conducting

the 524th.

recession, that

rent of air

is,

excited

Ax 21715th.
By pulsation,

is,

the perpendicular cur-

by the heated body, the

or radiation, the

2533rd part

from a metallic surface, and eight times as much,

or the 31 7th part from a surface of paper

should be observed, that Mr.

that air

which

is

is

(It

Leslie contends

instrumental in the radiation of heat,

contrary to the received opinion.)

ON THE MOTION OF HEAT.

106

11.

body

cools

than in air of the

more slowly

common

in rarefied air,

density

and the

dif-

ferent species of air have their respective refri-

Common

gerating powers.

air

and hydrogenous

According

gas exhibit remarkable differences.

Mr. Leslie, if the cooling power of common air

upon a vitreous surface be denoted by unity, that
of hydrogenous gas will be denoted by 2,2857
to

and upon a metallic surface the

In

1.7857.
surface

causes

.57

the loss from a vitreous

air

by radiation, and .43 by the other two

from a metallic surface, .07 and .43. In

hydrogenous gas the

is

to

.5

.57

is
:

common

ratio is

by

radiation,

loss

from a vitreous surface

and 1.71 by the other causes

from a metallic surface, .07 and 1.71.

He

finds

the radiation to be the same in the two gases, and

to

be very

the

little

diminished by rarefaction

but

other refrigerating powers

of the

effects

rapidly diminish with the density.

Those who wish

to see the experiments

and

reasonings from which these important conclusions are

Leslie's

to

Mr.

facts

and

derived, must have recourse

work

but as

opinions appear from

tionable, I shall

occurred to

my

some of the

experience to be ques-

now proceed

me on

reason to withhold

to state

what has

have no

these subjects.

my

assent from the

first

ON THE MOTION OF HEAT

articles

but the

3 are not equally

last

107
satisfac-

tory.

Before we enter upon a detail of experiments,

it

be proper to point out the correspondence

will

of the
in

new thermometric

the higher parts,

in the table,

it

scale with the old

one

being only given briefly

page 14.

CORRESPONDENCES OF THE THERMOMETRIC

SCALES.

Old

scale.

ON THE MOTION OF HEAT.

108

long from freezing to boiling mercury, was heated

to

442

new

scale,

and suffered

to cool in a hori-

The

zontal position in air of 42.

bulb in this and

every other instrument projected several inches

below the

The

scale.

times of cooling were the

same from 442 to 242, from 242 to 142, and

from 142

to 92,

namely, 2 minutes and 20

This was often repeated

conds each.

se-

the times

of cooling were always within 4 or 5 seconds of

that above,

and when any differences

in the suc-

cessive intervals took place, the times were

ways observed

more

in the

From

to

be rather

less in the higher,

al-

and

lower parts of the scale.

this

experiment

it

appears,

that

the

thermometer was raised 400 above the temperature of the air, or to 600 of the old scale
it
;

lost

200 of temperature in the

time, 100 in the second,

first

and 50

interval of

in the third.

This result goes to establish the principle an-

nounced

at

page 12,

that,

according to the

new

graduation, the temperature descends in geometrical jjr agression to equal

increments of time.

EXPERIMENT
According

to

Mr.

2.

Leslie, the

same law of

cooling does not take place from a metallic as

ON THE MOTION OF HEAT.

from a vitreous surface

me

always appeared to

this

109

very surprising, and I was anxious to

myself more particularly as to the

this view, I

satisfy

fact.

With

took another mercurial thermometer,

with a bulb of

.7

inch diameter, and scale of 12

to 300 old scale,

inches, having a range from

and corresponding new

scale attached to

it.

This

was heated, and the times of cooling through

every successive 10 degrees of the new scale

were noticed repeatedly

covered with

tinfoil,

bulb was then

the

pasted upon

it,

and the

sur-

made as smooth as well could be the thermometer was then heated, and the times of cooling

face

were again noticed as before, repeatedly.

mean results follow, and a column

The

of the differences

of the logarithms of the degrees expressing the

elevation of temperature above that of the sur-

rounding

air,

The temperature

which was 40.

of the thermometer was raised to 275 per scale

that

is,

235 above the

most convenient
of the

air

to

air,

and

it

is

obviously

reckon from the temperature

considered as zero

in

which case 19

represents the difference of the logarithms of 235

and 225, &c.

110

ON THE MOTION OF HEAT.

Thermom.

ON THE MOTION OF HEAT.

By

inspecting this table,

it

Ill

appears that the

whole time of cooling when the bulb was clear

was 851 seconds, and when covered

w^ith tinfoil

was 1206 seconds, which numbers are nearly as

17 to 24.

But the times

in cooling

from 175^ to

165 were 17 and 24 seconds respectively; and the

times in cooling from 95 to 85 were 34 and 48 respectively, which are exactly in the ratio of the

whole times

and by examining any two corres-

ponding times, they will be found to be as 17

nearly.

Whence

it

follows that the

to

24

same law of pro-

gressive cooling applies to a metallic as to a vitre-

ous surface, contrary to the results of Mr. Leslie's

experience.

It

must

not,

that this ratio for the

quite correct

however, be understood

two kinds of surfaces

thermometer may be coated with

face

is

is

however carefully the bulb of a

tinfoil,

necessarily enlarged, which

more quickly than

if

the sur-

makes

it

cool

the metallic surface were the

very same quantity as the vitreous.

The

differences of the logarithms happening

accidentally so nearly to coincide in magnitude

with the times of cooling of the metallic surface,

they require no reduction, and we have an opportunity of seeing

how

far the

progression in cooling

periment.

It

is

law of geometrical

supported by this ex-

appears that for 5 or 6 of the

ON THE MOTION OF HEAT.

112

highest intervals of temperature,

the times of

cooling were rather smaller, and for the two last

rather larger than required

by

EXPERIMENT

As Mr.

the law.

3.

Leslie found the times of cooling of

metallic surfaces considerably enlarged, in

rate elevations of temperature

more

mode-

especially, I

took another thermometer having a smaller bulb,

and a scale of an inch

for

10 degrees,

this

was

treated as in the last experiment, and the results

were

as

under

Thermom.

ON THE MOTION OF HEAT.

former

experiment

smaller,

it

ed

bulb

the

being

was more than proportionally increas-

in surface

by the

tinfoil,

which was pasted on

and consequently was twofold

in small slips,

many

because

113

in

places.

Being from these

results pretty well satisfied

that the surfaces of bodies do not disturb the law

of their

refrigeration,

affect the time,

though they materially

yet in consequence of the general

accuracy of Mr.

Leslie's

experiments, I was

desirous to ascertain the results in his

more

own way,

particularly because for the reason assigned

above,

my method

did not give the true rates of

cooling of equal surfaces.

EXPERIMENT
I took two

monly used

new

4.

tin canisters,

such as are com-

for tea, of a cylindrico-conical shape,

and each capable of holding 15

The
lid,

of water.

them was covered with

surface of one of

brown paper pasted on

oz.

it

instead of the usual

a cork of 1^ inch, diameter was adapted to

both,

and through a hole

in the centre of this, the

tube of a delicate thermometer was inserted, with

a scale of the
cork.

new graduation

above the

affixed

Both canisters were contrived

pended by small

strings

when

filled

to be sus-

with water,

ON THE MOTION OF HEAT.

114

and

to

have the thermometer with

They were

centres.

its

successively

bulb in their
filled

with

boiling water, and suspended in the middle of a

room of the temperature

40,

and the times of

cooling through each successive 10 degrees were

noticed as below.
Water

From 205

cooled.

to

195

ON THE xMOTION OF HEAT.

115

of being repeated by any one without the aid of

any expensive instrument or any extraordinary

dexterity

it

be unnecessary to

will

the accuracy of the above.

that the range of cooling

new

It will

upon

be understood

was from 205 of the

scale, to 65 of the same, the air

or 25 below

insist

being 40,

the extremity of the range, which

corresponds with 57 of the old scale.

It will

be proper now to enquire into the cause

of the difference in the times of cooling arising

from the variation of surface.

Mr. Leslie has

shewn the surface has no influence upon the time

when immersed

of cooling

seem then

in

water

it

should

that the difference of surfaces in the

expenditure of heat arises from their different

powers of radiation solely

indeed Leslie has

proved by direct experiments

that

the

radiating from a vitreous or paper surface

heat
is

times as great as that from a metallic surface.

Taking

this

for

granted,

we

the portions of heat dispersed

can

by

easily

radiation,

conducted away by the atmosphere.

find

and

For, let

denote the quantity of heat conducted away by

the atmosphere, from a vitreous or metallic surface in

any given small portion of time, and x the

quantity radiated from a metallic surface in the

same time

then 8

:v

will

be the quantity radiated

ON THE MOTION OF HEAT.

116

from a vitreous surface in that time

and from

we shall have,
l+.r: 1+8^; whence 2 + 16^= 3 + 3^',

the result of the last experiment

and

::

.t'=r3

this gives

l^,

for the

whole heat

dis-

charged by metal, and lj\ for that discharged by

glass in the

same time, where the unit expresses

the part conducted, and the fraction the

part

radiated.

That

is,

from a metallic surface 13 parts of

heat are conducted away by the air and

radiated

part

from a vitreous surface 13 parts are

conducted, and 8 parts radiated, in a given time.

The

quantity of heat discharged by radiation

from the most favourable surface, therefore,

probably not more than

.4

conducted away by the

air

of the whole, and that

not less than

.6.

Mr.

Leslie, however, deduces .57 for the former,

.43 for the latter

is

and

because he found the dispro-

portion in the times of cooling of vitreous and

metallic surfaces greater than I find

it

in the

lower

part of the scale.

The

obvious consequences of this doctrine in

a practical sense are,

1.

In every case where heat

is

required to be

retained as long as possible, the containing vessel

should be of metal, with a bright clear surface.

2,

Whenever

heat

is

required to be given out

ON THE MOTION OF HEAT.

by

body with

as

containing vessel,

much
if

vegetable matter

given out

will

celerity as possible, the

of metal, ought to be painted,

covered with paper,

or

11?

or

charcoal,
;

in

some animal

which case the heat

be 3 parts for 2 from a metallic

surface.

REFRIGERATION OF BODIES

VARIOUS KINDS OF

IN

ELASTIC FLUIDS.
Bodies cool in very different times
the

elastic fluids.

who

lieve,

in

Mr. Leslie was the

noticed this fact

some of

first, I

be-

and he has given us

common

air

and

hydrogenous gas, of the common density, and

also

the results of his experiments on

rarefied in various degrees.

made some expe-

riments with a view to determine the relative

cooling powers of the gases, the results of which
it

may be proper

to give.

My

apparatus was a

strong phial, containing about 15

inches

20 cubic

a perforated cork containing the stem of

a thermometer was adapted

tight

or

to

it,

so as to be air

two marks were made with a

file

on the

tube of the thermometer, comprizing an interval

of

15 or 20, about blood heat.

filled

The

bottle

with any proposed gas, and after

it

acquired the temperature of the surrounding

was
had
air,

the stopper was withdrawn, and the heated ther-

ON THE MOTION OF HEAT.

118

mometer with its cork was instantly inserted the

number of seconds which elapsed whilst the
;

mercury descended from the upper

mark were then noted

to the

The

as under.

under

surround-

ing air was of a constant temperature.

Thermometer immersed

In carbonic acid gas

/ 112

sulphuretted hydrogen,

cooled in
seconds.

nitrous 1

oxide, and olefiant gas

com.
and oxygen gas 100
nitrous gas
90
carburet, hyd. or gas
70
hydrogen
40
azotic

air,

coal

The

refrigerating effect of hydrogen

remarkable

I cooled the

is

truly

thermometer 10 times

successively in a bottle of hydrogen gas

at

each

experiment the instrument was taken out, and the

stopper put
restored

in, till

by

this,

the original temperature was

a portion of the hydrogen

escaped each time, and an equal portion of com-

mon

air

was admitted

the times of cooling regu-

larly increased

as

51, 53, 56, 58,

60 and 62 seconds, respectively

at this time the

mixture was examined, and found

half

follows

viz. 40, 43,

hydrogen and half common

air.

46, 48,
;

Equal mea-

ON THE MOTION OF HEAT.

common

sures of hydrogen and

mixed together, and put

to cool

and the

from mark

mark, in 62 seconds as before.

Condensed
air

were then

air

into the bottle,

heated thermometer was found

to

119

of common

pidly,

bodies more rapidly than

air cools

density

and rarefied

whatever be the kind.

own experience
Density of the

for

common

The
air

air less ra-

results of

were

as follows

Therm, cools

air.

my
:

in

85 seconds.

2
1

100

116

128

140

^
W

160
iVo

A small receiver of hydrogen gas, which cooled

the thermometer in 40 seconds,
or 8 times, took

But the exact

70 seconds

eff*ects

when

rarefied 7

to cool the same.

of rarefaction on this and

the other gases were not determined.

From Mr.

Leslie,

we

learn that in hydrogenous

gas, there is little difference

between the time of

cooling of a vitreous and metallic surface, the

former being as 2.28, and the latter as 1.78, from

which he justly infers " this inequality of effect

ON THE MOTION OF HEAT.

120

[between atmospheric
proves

its

influence to be exerted chiefly, if not

entirely, in

The
same

augmenting the abductive portion."

expenditure of heat by radiation being the

hydrogenous gas as

in

we may

infer

of gas

and therefore

it is

of the gas, and

as

and hydrogenous gas]

air

is

the

same
is

in

atmospheric

air,

every other species

performed independently

carried on the

same

in

vacuo

Indeed Mr. Leslie himself admits

in air.

the diminution of the

that

in

upon rarefaction

is

effects

consequent

extremely small, which can

scarcely be conceived

if air

were the medium of

radiation.

The

effect of radiation

being allowed constant,

that of the density of the air

and

will

may be

investigated,

be found, I believe, to vary nearly or

accurately as the cube root of the density.

In

order to compare this hypothesis with observation,

let

100

time of cooling in atmospheric

density being

air,

the

then from what has been said

above, .4 will represent the heat lost by a vitreous

by radiation, and .6 that lost by the conLet t = the time

ducting power of the medium.
of cooling in air of the density d; then if
surface

100

:: t

.004

t=

the heat lost

but the heat conducted away

the time

+ by

is,

by

radiation

by hypothesis,

the cube root of the density

as

.066

ON THE MOTION OF HEAT,

121

t^/d; whence .004

+ .006 tx/d=

1,

and

^=

3_
.004

+ 006 v/^/
.

Calculating from this formula,

times of cooling in

common

we

shall find the

air of

the several

densities as under
Density of the

air.

Times of cooling.

86.5 seconds.

100

114

129
143

T^

157

iV

170

182
193

250

iSflSit^

This table accords nearly with the preceding

one, the result of actual observation.

In

the

same way might the times of cooling of a metallic

surface in rarefied air be found,

by

substituting

.0007 for .004, and .0093 for .006 in the preceding formula.

The

cooling power of hydrogenous gas inde-

pendent of radiation, may be found thus

Q

if

ON THE MOTION OF HEAT.

122
100": .4

40": .16

::

by radiation
whence .84 = the

the heat lost

40 seconds

that gas in

in

heat conducted away by the air in 40", or .021

per second

common

but in

second by abduction

only .006

is

the loss per

air
;

from

this

it

appears that the refrigerating power of hydro-

genous gas

is

3 J times as great as that of

common

air.

It

may be asked what is

the cause

why

different

gases have such different cooling effects, especially

on the supposition of each atom of

same quantity of

different species possessing the

heat

differ

the

To

this

we may answer,

from each other

number of atoms

in

in a

the

all

that the gases

two essential points,

in

given volume, and in

the weight or inertia of their respective atoms.

Now

both

number

and weight tend

tard the motion of a current

that

is,

to
if

re-

two

gases possess the same number of particles in a

given volume,

it is

evident that one will disperse

heat most quickly which has

weight

and

if

its

atoms of the least

other two gases have particles of

the same weight, that one will most disperse heat

which has the

least

number

because the resistance

particles to be

moved,

will

in a

given volume

be as the number of

in like circumstances.

the gases that have nearly the same

number

Of
of

123

ON THE TEMPERATURE, &C.

same volume,

particles in the

are,

hydrogen, ear-

buretted hydrogen, sulphuretted hydrogen, nitrous

These conduct heat in

the order they are written, hydrogen best and
oxide, and carbonic acid.

carbonic acid worst

and the weights of their

ultimate particles increase in the same order (see

page 73).

Of those

that have their atoms of the

their number in a given volume

and carburetted hydrogen
oxygen
are
has the greater cooling power and the

same weight and

different,

the latter

fewer particles in a given volume.

SECTION

8.

ON THE TEMPERATURE OF THE

ATMOSPHERE.
It is a

been
phere
in

remarkable

and has never,

fact,

I believe,

satisfactorily accounted for, that the atmosin all places

and seasons

is

found to decrease

temperature in proportion as we ascend, and

nearly in an arithmetical progression.

the fact

may have been

Sometimes

otherwise, namely, that

the air was colder at the surface of the earth than

above, particularly at the breaking of a

have observed

it

so

but

frost, I

this is evidently the

ON THE TEMPERATURE

124

of great and extraordinary commotion in

effect

the atmosphere, and

What

duration.

then

the occasion of this di-

is

minution of temperature
this question

most of a very short

at

is

ascending

in

can be solved,

may be proper
common solution.

it

consider the defects of the

Air,

it is

the sun

to

not heated by the direct rays of

said, is

which pass through

it

medium, without producing any

till

Before

as a transparent
calorific

effect,

The

they arrive at the surface of the earth.

earth being heated, communicates a portion to

the contiguous atmosphere, whilst the superior

strata in proportion as they are

more remote,

re-

ceive less heat, forming a gradation of tempera-

what takes place along a bar of

ture, similar to

iron

when one

The

its

ends

is

heated.

part of the above solution

first

correct

Air,

to heat

it

it

should seem,

is

is

probably

singular in regard

neither receives nor discharges

radiant state

through

of

air

if

so,

must be

the

propagation

effected

power, the same as in water.

by

its

rity,
it is

is

in a

of heat

conducting

Now we know

that heat applied to the under surface of a

of water

it

column

propagated upwards with great cele-

by the

actual ascent of the heated particles

equally certain, too, that heated air ascends.

From

these observations

it

should follow that the

OF THE ATMOSPHERE.

125

causes assigned above for the gradual change of

temperature in a perpendicular column of the

atmosphere, would apply directly to a state of

temperature the very reverse of the fact

to

one

in v^hich the

namely,

higher the ascent or the more

remote from the earth the higher should be the

temperature.

Whether
must

I think

reasoning be correct or not,

this

be universally allowed, that the

it

fact

has not hitherto received a satisfactory explanaI

tion.

conceive

that

it

be one involving a new

to

by which I mean a principle

no other phenomenon of nature presents us

principle of heat

with,

and which

such.

I shall

is

not at present recognized as

endeavour

in

what follows

to

make

out this position.

The

The natural equilibrium

an atmosphere, is when each atom of

principle

is this

of heat in
air in the same perpendicular column is possessed
of the same quantity of heat ; and consequently,
the natural equilibrium
is

when

of heat in an atmosphere

the temperature gradually diminishes in

ascending.

That
denied,

this

is

a just consequence

when we consider

capacity for heat by rarefaction

tity

of heat

is

cannot be

that air increases in

:

when

its

the quan-

given or limited, therefore the

temperature must be regulated by the density.

ON THE TEMPERATURE

126
It is

an established principle that any body on

the surface of the earth unequally heated

ob-

is

served constantly to tend towards an equality of

temperature

seems
if it

the

new

principle

announced above,

But

to suggest an exception to this law.

be examined,

it

can scarcely appear in that

Equality of heat and equality of temperalight.

ture^ when applied to the same body in the same
state,

are found so uniformly to be associated

together, that
distinction

we

making any

scarcely think of

No

between the two expressions.

one would object to the commonly observed law

being expressed in these terms
is

When any body

unequally heated^ the equilibrium

is

found

to

when each particle of the body becomes in possession of the same quantity of heat.
Now the law thus expressed is what I apprehend
be restored

to

be the true general law, which applies to the

atmosphere as well as to other bodies.

equality of heat, and not

an

equality

It is

an

of tempera-

ture that nature tends to restore.

The atmosphere indeed

presents

peculiarity to us in regard to heat

perpendicular column of

air,

we

striking

see in a

body without any

change of form, slowly and gradually changing

its

capacity for heat from a less to a greater

all

other bodies retain a uniform capacity through-

out their substance.

but

OF THE ATMOSPHERE.
If

it

12?

be asked why an equilibrium of heat should

turn upon the equality in quantity rather than in

temperature

I rest the proof of

lity

know

I answer that I do not

it

of temperature

upon

but

the fact of the inequa-

observed in ascending into

If the natural tendency of the

the atmosphere.

atmosphere was to an equality of temperature,

me any

there does not appear to

why

reason

the

superior regions of the air should not be at least

warm as the inferior.

The arguments already advanced on
the principle we are endeavouring to
as

corroborated by

powerfully

are
facts

sure,

By the

establish,

following

observations of Bouguer, Saus-

and Gay Lussac, we

find that the tempera-

ture of the air at an elevation

J that

the

behalf of

at the surface, is

where

its

weight

is

about 50 Fahrenheit less

than that at the surface

and from

my

experi-

ments (Manch.

Mem.

appears that

being suddenly rarefied from 2

to

air

vol.

5,

measure

525,)

it

Whence we may

produces 50 of cold.

infer, that a

page

of air at the earth's surface

being taken up to the height above-mentioned,

preserving
to expand,

its

original temperature,

and suffered

would become two measures, and be

reduced

to

rounding

air

the
;

same temperature

or vice versa,

if

as

the sur-

two measures of

ON THE TEMPERATURE

128
air at the

proposed height were condensed into

one measure, their temperature would be raised

50,

and they would become the same

and temperature,
earth's

volume of

summit

air

of the

in density

volume of

In like manner we

surface.

that if a
to the

as the like

air at the

may

infer,

from the earth's surface,

atmosphere were condensed

and brought into a horizontal position on the

earth's surface,
sity

it

would become of the same den-

and temperature

around

as the air

it,

without

receiving or parting with any heat whatever.

Another important argument

theory here proposed

in

favour of the

may be derived from

contemplation of an atmosphere of vapour.

pose the present

annihilated,

serial

atmosphere were to be

and one of steam or aqueous vapour

were substituted

in its place

that the temperature of this

earth's

the

Sup-

surface

weight equal to

and suppose

further,

atmosphere

at the

were every where 212 and

30 inches of mercury. Now

its

at

the elevation of about 6 miles the weight would

be 15 inches or ^ of that below, at 12 miles,

it

would be 7.5 inches, or | of that at the surface,

&c., and the temperature would probably diminish
25 at each of those intervals.

minish more
diminution

for

It

could not

we have seen (page 14)

of temperature of 25

di-

that a

reduces the

OF THE ATMOSPHERE.
force of vapour one half;

if

129

therefore a greater

reduction of temperature were to take place, the

weight of the incumbent atmosphere would con-

dense a portion of the vapour into water, and the

perpetually from

But

regions.

if

thus

condensations

we suppose on

be
in

disturbed
the

upper

the other hand,

diminution of temperature in each of

that the

these intervals

regions

would

equilibrium

general

could

condensation

is

less than 25,

admit of
but

then the upper

more vapour without

must take place

it

at

the

surface, because vapour at 212 cannot sustain

more than the weight of 30 inches of mercury.

These three supposed cases of an aqueous

vapour atmosphere may be otherwise stated thus
1.

The

specific gravity of

steam at the earth's

surface being supposed .6 of atmospheric

air,

and

the weight of the atmosphere of steam equal to

30 inches of mercury,

its

temperature

at the

surface would be 212*^; at 6 miles high, 18?;

at

12 miles,

162;

miles, 112, &c.

at

In

18 miles, 137;

this case the

at

24

density, not

only at the surface, but every where, would be

a

maximum,

or the greatest possible for the ex-

isting temperature

so that a perfect equilibrium

having once obtained, there could be neither

condensation
R

nor

evaporation

in

any region.

ON THE TEMPERATURE

130

For every 400 yards of

meter would descend
If the

2.

elevation, the thermo-

degree.

atmosphere were constituted just as

above, except that the temperature

ed more rapidly than

miles

at

now

diminish-

the rate of 25 for 6

then the temperature of the higher regions

not being sufficient to support the weight, a condensation must take place

the weight would

thus be diminished, but as the temperature at the

surface

is

always supposed to be kept at 212,

evaporation must go on there with the design to

keep up the pressure

would be perpetual

at

30

strife

Thus

inches.

there

between the recently

raised vapour ascending, and the condensed drops

of rain descending.

position

much

less likely

than the preceding one.

The same things being supposed as

now the temperature decreases more

3.

but

than at the rate of 25 for 6 miles

the density of the

else

so that

slowly

in this case

steam at the earth's surface

would be a maximum

where

before,

for the temperature, but

if

no

a quantity of water were

taken up to any elevation

it

would evaporate

but the increased weight of the atmosphere would

produce a condensation of steam into water on

the ground.

In this case then there would not

be that of equilibrium, which we see

in the 1st

OF THE ATMOSPHERE.

131

and which accords so much more with the

case,

regularity and simplicity generally observable in

the laws of nature.*

That an atmosphere of steam does actually

surround the earth, existing independently of the
*

owe

Mr. Ewart the

to

which

elastic fluids,

present section

hint of the idea respecting

first

have endeavoured to expand

me some

he suggested to

was probable steam of any low temperature,

maximum

if it

without losing any heat, that

pace with

maximum density

is,

if

the containing vessel kept

it is

sion

there

it

would con-

its elasticity.

fact the heat (1000),

In

when

could be gradually compressed

be any condensation of steam into water, but

stantly retain

it

32, of

increase of temperature, there would never

in

it

as

same quantity of absolute

density, contained the

heat as the like weight of steam of 212 of

and that consequently

in the

time ago, that

which

condensed into water,

is

no change

water for heat

is

is

given out by steam

merely heat of compres-

in the affinity of the molecules of

the expulsion

is

occasioned solely by the

approximation of the molecules, and would be precisely the

same whether

that approximation

was occasioned by external

compression or internal attraction.

Indeed,

if

we

estimate

the temperature that would be given out by the mechanical

compression of steam from a volume of 2048 to that of

1,

by

successively doubling the density, and supposing as above,

that at each time of doubling, 25 were given out,

it

would

be found that 12 successive operations would reduce the

volume
is

to

and that only 300 would be given

out.

But

it

not right to conclude, that the same quantit}' of tempera-

ON THE TEMPERATURE, &C.

132

atmospheres

other

necessarily most

with

which however

intimately mixed,

capable of demonstration.

and

to enforce

the

illustrate

have endeavoured

I
it

in several

Memoirs of the Manchester

Essays

in

and

in

Society,

Nicholson's Journal, to which I must refer.

an atmosphere of any elastic

is

it

think

is

fluid,

Now

whether of the

weight of 30 inches of mercury, or of half an inch,

must observe the same general laws

but

it

should seem that an atmosphere of vapour varies

its

temperature less rapidly in ascending than the

one we have of
ture

Something of an

air.

would be given out

tions,

though

it

may

at

effect simi-

each of the successive condensa-

be nearly so for most of them

by the

the conclusion, the space occupied

particle bears a considerable proportion to the

occupied by

it

and

its

atmosphere.

At

the

first

towards

atom or

solid

whole space
compression,

the atmosphere of heat might be said to be reduced into half

the

space

greater,

but at the

last,

the reduction

would be much

and therefore more heat given out than determined

by theory.
Since writing the above, Mr. Ewart informs
idea respecting steam, which I had from him,

Mr. Watt's.

In Black's Lectures, Vol.

thor, speaking of

steam

is

at least as
It is

is

that the

originally

page 190, the au-

Mr. Watt's experiments on steam

temperatures, observes, "

diminished."

1,

me

much

we

at

low

find that the latent heat of the

increased as the sensible heat

is

wonderful that so remarkable a fact should

have been so long known and so

little

noticed.

ON CONGELATION.
what

lar to

is

133

pointed out in the 2nd case above,

ought therefore to be observed

atmosphere
in the

namely, a

condensation of vapour

is

going on below.

the case almost every day, as

own

moment

higher regions, at the same

evaporation

observation

our mixed

in

This

all

is

actually

know from

a cloudy stratum of

that

their

air

fre-

quently exists above, whilst the region below

is

comparatively dry.

SECTION

9.

ON THE PHENOMENA OF THE CONGELATION OF WATER.

Several remarkable phenomena are attendant

upon the congelation of water, and some of them

are

so

different from

from analogy, that

what might be expected

believe no explanation ac-

cording with the principles of the

mechanical

philosophy has been attempted such as to account

for

all

the appearances.

object of the present Essay.

This attempt
It will

is

the

be expedient

previously to state the principal facts.

1.

The

specific gravity of ice

is

of water in the ratio of 92 to 100,

less than that

ON CONGELATION.

134
2.

When water is

jar to cool in

still

exposed in a large suspended

air of

20 or

cooled 2 or 3 below freezing

30,

but

if

it

may be

any tremu-

lous motion take place, there appear instantly a

multitude of shining hexangular spicules, floating,

and slowly ascending

3. It is

in the water.

observed that the shoots or ramifications

of ice at the

commencement, and

in the

early

stage of congelation are always at an angle of 60

or 120.

Heat is given out during congelation, as

much as would raise the temperature of water
4.

The same quantity is

new scale.
This
taken in when the ice is melted.

150 of the
again

quantity

may be ^

of the whole heat which water

of 32 contains.

Water is densest at 36 of the old scale, or

38 of the new from that point it gradually expands by cooling or by heating alike, according
5.

to the

law so often mentioned, that of the square

of the temperature.
6.

If water be

tation,

it

exposed

to the air,

cannot be cooled below 32

and
;

to agi-

the appli-

cation of cold freezes a part of the water,

and the

mixture of ice and water acquires the temperature

of 32.
7.

If the water bo kept

still,

and the cold be

135

ON CONGELATION.
not severe,

it

may be

cooled in large quantities

to 25 or below, without freezing

if

the water be

confined in the bulb of a thermometer,

to freeze

difficult

water

the

cool

without

as 7 or 8,

but

it is

much below

freezing.

its

that temperature

have obtained

it

it

it

as low

again without

being frozen.

In the last case of what

8.

equally difficult to

and gradually heated

any part of

very

by any cold mixture above

it

15 of the old scale

it is

may be caWed/orced

cooling, the law of expansion

is still

observed as

given above.

When

9.

bulb,

but

it

water

retains the

if it

cooled to 15 or below in a

is

most perfect transparency

accidentally freeze, the congelation

instantaneous, the bulb becoming in a

opake and white

like snow,

is

moment

and the water

is

pro-

jected up the stem.

10.

When

water

is

cooled below freezing, and

congelation suddenly takes place, the temperature

rises instantly to 32.

In order to explain these phenomena, let

it

be

conceived that the ultimate or smallest elements

of water are
size

let

all

globular,

and exactly of the same

the arrangement of these atoms be in

squares, as exhibited in Fig.

each

particle

touches

1,

Plate

3,

so that

four others in the same

ON CONGELATION.

136

Conceive a second stratum of

horizontal plane.
particles placed

squares, but

upon these

so that

each globule

concavity of four others on the

consequently rests upon

falls

first

stratum, and

perpendi-

exhibited in

is

Conceive a third stratum placed in like

Fig. 3.

manner upon the second, &c.

similar to a square pile of shot.

The whole being

The above con-

that of water

stitution is conceived to represent

To

into the

of such globule resting upon two

diagonal globules of the square

at the

of

four points, elevated 45

above the centres of the globules.

cular section

order

like

in

temperature of greatest density.

find the

number of globules

vessel, the side of

which

is

given

in
let

cubic

n =

the

number of particles in one line or side of the cube;

then n^ is the number in any horizontal stratum

and because a

line

joining the centres of two

contiguous particles in different strata makes an

angle of 45 with the horizontal plane, the num-

n -^ sine
number of

ber of strata in the given height will be

=71-^

Whence the
J\/2.
particles in the cubic vessel = ri' -r ^\/2= 7i^V2.
of 45

Now

let

instantly

2)

it

be supposed that the square

drawn

into the shape of a

then each horizontal stratum will

of the

same number of

pile

is

rhombus (Fig.
still

consist

particles as before, only

in

now

ON CONGELATION.

137

more condensed form, each

particle being

But

in contact with six others.

to counter-

act this condensation, the several successive strata

are

more elevated than

increased in

before, so that the pile

height.

whether a vessel of given capacity

number

greater

disposition
last case,

will

hold a

of particles in this or the former

It

is

then arises

question

must be observed, that

in the

each particle of a superior stratum rests

only on two particles of an inferior one, and

is

therefore elevated

by the sine of 60 as repre-

sented in Fig. 4.

The

as 1

v/f,

bases of the two piles are

and their heights as x/^

^/f, but

the capacities are as the products of the base and

height, or as \/^

that

is,

.750

as .707 to

94 to 100. Thus it appears that

arrangement contains more particles in a

nearly, or as

the

first

given space than the second by 6 per cent.

The

last or

rhomboidal arrangement

is

suppos-

ed to be that which the particles of water assume

upon congelation.

The

specific gravities of ice

and water should therefore be as 94

it

But

to 100.

should be remembered that water usually con-

tains 2 per cent, in

that this air

is

bulk of atmospheric

liberated

air

and

upon congelation, and

commonly entangled amongst the

ice in

is

such sort

as to increase its bulk without materially increass

ON CONGELATION.

138
ing

its

weight

this

reduces the specific gravity

of ice 2 per cent., or

makes

it

92, which agrees

Hence

exactly with observation.

the 1st fact

is

explained.

The

angle of a rhombus

ment 120

if

60,

is

and

its

supple-

any particular angles are manifested

we ought

in the act of congelation, therefore

to

expect these, agreeable to the 2nd and 3rd phe-

nomena.

Whenever

any remarkable

internal constitution

of any

change in the

body takes

place,

whether by the accession and junction of new

particles, or

by new arrangements of those already

existing in

it

pheres

of

heat must

though

it

tity,

some modification

may be

evidently

difficult to

in the atmos-

be

required;

estimate the quan-

and sometimes even the kind of change so

produced,

as

the

in

present

case.

So

far,

therefore, the theory proposed agrees with the

4th phenomenon.

In order to explain the other phenomena,

will

it

be requisite to consider more particularly the

mode by which bodies

ment of the

individual atmospheres of the com-

ponent particles
fluids,

by heat. Is
by the enlarge-

are expanded

the expansion occasioned simply

This

and perhaps with

is

the case with elastic

solids,

but certainly not

ON CONGELATION.
with liquids.

How is

it

139

possible that water should

be expanded a portion represented by

upon the

addition of a certain quantity of heat at one tem-

perature, and

340 upon the addition of a

quantity at another temperature,

like

when both tem-

peratures are remote from the absolute zero, the

one perhaps 6000 and the other 6l70

fact

The

cannot be accounted for on any other sup-

position than that of a change of arrangement in

the component particles

and a gradual change

from the square to the rhomboidal arrangement

is

in all probability effected both

and abstraction of

heat. It

is

to

by the addition

be supposed then

that water of the greatest possible density has its

particles arranged in the square

given quantity of heat be added

it,

the particles

commence

form

to,

but

if

or taken from

their approach to the

rhomboidal form, and consequently the whole

is

expanded, and that the same by the same change

of temperature, whether above

or

below

that

point.

If heat be taken

expansion

is

away from water of

38, then

the consequence, and a moderate

inclination of the particles towards the rhomboidal

form

but this only extends a small

the mass

is

way

whilst

subject to a tremulous motion, so as

to relieve the obstructions

occasioned by friction

ON CONGELATION.

140

by the energy of certain

is

affinities,

new form

the

completed in a moment, and a portion of ice

formed

heat

is

then given out which retards the

subsequent formation,

This

gealed.

But

tion.

if

is

till

at last the

whole

is

con-

the ordinary process of congela-

the mass of water cooled

is

kept in

a state of perfect tranquillity, the gradual approach

to

the rhomboidal form can be

farther

carried

much

the expansion goes on according to the

usual manner, and the slight friction or adhesion

of the particles

is

sufficient

to

counteract the

new forma-

balance of energies in favour of the

tion,

till

some accidental tremor contributes

adjust the equilibrium.

similar

operation

performed when we lay a piece of iron on a

and hold a magnet nearer and nearer

mity of the approach, without contact,
sisted

to
is

table,

the proxi-

is

much

as-

by guarding against any tremulous motion

of the table.

Hence

are accounted for.

the rest of the

phenomena

CHAPTER

II,

ON THE

CONSTITUTION OF BODIES.
There

are three distinctions in the kinds of

bodies, or three states, which have

more

especially

claimed the attention of philosophical chemists

namely, those which are marked by the terms

elastic fluids^ liquids^

instance

A very familiar

and solids.

exhibited to us in water, of a body,

is

which, in certain circumstances,

assuming

all

the three states.

is

capable

In steam

we

of

recog-

nise a perfectly elastic fluid, in water, a perfect

liquid,

and

in ice a

complete

solid.

These obser-

vations have tacitly led to the conclusion which

seems universally adopted, that

sible

all

magnitude, whether liquid or

bodies of sensolid,

are con-

number of extremely small

particles, or atoms of matter, bound together by a
force of attraction, which is more or less powerful
stituted of a vast

according to circumstances, and which, as

vours to prevent their separation,

is

it

endea-

very properly

ON THE CONSTITUTION OF BODIES.

142

called in that view, attraction

as

it

of cohesion ; but
them from a dispersed state (as from

collects

steam into water)

gation,

names

more

or,
it

may go

same power.

called, attraction

it is

simply,

by, they

It is

not

affinity.

and the

signify one

still

my

of aggreWhatever

design to

call in

ques-

which appears completely

tion this conclusion,

satisfactory,

but to shew that we have hitherto

made no use

of

the

it,

and that the consequence of

neglect has been a very obscure view of

chemical agency, which

in proportion to the

thrown upon

The

is

new

daily

growing more so

lights

attempted to be

it.

opinions I

more

particularly allude to, are

those of Berthollet on the

Affinity";

of Chemical

such as that chemical agency

tional to the mass,

there

Laws

and that in

exist insensible

all

propor-

is

chemical unions

gradations

in

portions of the constituent principles.

the

pro-

The

in-

consistence of these opinions, both with reason

and observation, cannot,

I think,

fail

to strike

every one who takes a proper view of the phe-

nomena.

Whether

the ultimate particles of a body, such

as water, are all alike, that

is,

of the

same

some importance.

weight, &c.,

is

From what

known, we have no reason

is

a question of

figure,

to appre-

ON THE CONSTITUTION OF BODIES.

hend a

diversity in these particulars

exist in water,

it

must equally

143
does

if it

exist in the ele-

ments constituting water, namely, hydrogen and

Now,

oxygen.

how

it is

scarcely possible to conceive

the aggregates of dissimilar particles should

be so uniformly the same.

ticles

If

some of the par-

of water were heavier than others

if

parcel of the liquid on any occasion were consti-

tuted principally of these heavier particles,

must be supposed

it

to affect the specific gravity of

the mass, a circumstance not known.

Similar

may be made on other substances

we may conclude that the ultimate

observations
therefore,

particles

of all homogeneous bodies are

alike in weighty figure^

every particle of water

is

In other words,

^c.
like

'perfectly

every other particle

of water; every particle of hydrogen

is

like

every

other particle of hydrogen, &c.

Besides the force of attraction, which, in one

character or another, belongs universally to pon-

derable bodies,

we

find another force that

wise universal, or acts upon

all

is

like-

matter which

comes under our cognizance, namely, a force of

repulsion.

This

is

now

generally, and I think

properly, ascribed to the agency of heat.

atmosphere of

this

rounds the atoms of

An

subtile fluid constantly surall

bodies, and prevents

them

ON THE CONSTITUTION OF BODIES.

144

This

from beino^ drawn into actual contact.

appears to be satisfactorily proved by the observation, that the bulk of a

by abstracting some of

body may be diminished

heat but from what

its

has been stated in the last section,

that enlargement

it

should seem

and diminution of bulk depend

perhaps more on the arrangement, than on the

size of the ultimate particles.

we cannot

avoid inferring,

Be

this as

it

may,

from the preceding

doctrine on heat, and particularly from the section

on the natural zero of temperature, that solid

bodies, such as ice, contain a large portion, per-

haps of the heat which the same are found to

contain in an elastic state, as steam.

We

now

are

to consider

antagonist powers

how

these two great

of attraction and

repulsion

are adjusted, so as to allow of the three different

of elastic fluids^

states

We

fluids
tic

and

solids,

divide the subject into four Sections

shall

namely,

liquids,

first,

on

the constitution of 'pure elastic

second, on the constitution of mixed elas-

fluids

third,

on the constitution of

and fourth, on the constitution of solids.

liquids,

145

ON PURE ELASTIC FLUIDS.

SECTION

1.

ON THE CONSTITUTION OF PURE

ELASTIC FLUIDS.

piire elastic fluid

particles of

which are

tinguishable.

is

one,

all alike,

the constituent

way

or in no

dis-

Steam, or aqueous vapour, hydro-

genous gas, oxygenous gas,

azotic

These

several others are of this kind.

constituted of particles possessing

gas,* and
fluids are

very diffuse

atmospheres of heat, the capacity or bulk of the

atmosphere being often one

or

two thousand

times that of the particle in a liquid or solid form.

Whatever therefore may be the shape

of the solid atom abstractedly,

by such an atmosphere
as all the globules in

it

or figure

when surrounded

must be globular

but

any small given volume are

subject to the same pressure, they must be equal

in bulk,

and

will therefore

be arranged

zontal strata, like a pile of shot.

* The novice will

all

in hori-

volume of

along understand that several che-

mical subjects are necessarily introduced before their general

history and character can be discussed.

ON PURE ELASTIC FLUIDS.

146

fluid

elastic

pressure

is

found to expand whenever the

taken

is

This proves that the re-

off*.

The

pulsion exceeds the attraction in such case.

absolute attraction and repulsion of the particles

of an elastic
ing,

fluid,

we have no means

though we can have

little

contemporary energy of both

of estimat-

doubt but that the

is

great

but the

excess of the repulsive energy above the attractive can

be estimated, and the law of increase and

diminution be ascertained in

many

may be

in steam, the density

cases.

Thus

taken at y^g^ that

consequently each particle of steam has

of water;

12 times the diameter that one of water has, and

must press upon 144

face

that of a

watery sur-

particles of a

but the pressure upon each

is

equivalent to

column of water of 34 feet

therefore

the excess of the elastic force in a particle of

steam

is

equal to the weight of a column of par-

ticles of water,

feet

And

whose height

is

34x144 = 4896

further, this elastic force decreases as

the distance of the particles increases.

With

re-

spect to steam and other elastic fluids then, the

force of cohesion

is

entirely counteracted

of repulsion, and the only force which

to

move

the particles

is

is

by

that

efficacious

the excess of the repulsion

above the attraction. Thus,

if

the attraction be as

10 and the repulsion as 12, the effective repulsive

ON PURE ELASTIC FLUIDS.

force

as

is

fluid, so far

from requiring any force

its particles, it

them

appears then, that an elastic

It

2.

147

to separate

always requires a force to retain

in their situation, or to

prevent their sepa-

ration.

A vessel full

of any pure elastic fluid presents

to the imagination a picture like

The globules

shot.

are

all

one

full

of small

of the same size

but

the particles of the fluid differ from those of the

shot, in that

small

they are constituted of an exceedingly

central

atom

of

solid matter,

which

is

surrounded by an atmosphere of heat, of great

density next the atom, but gradually growing
rarer according to

some power of the distance

whereas those of the shot are globules, uniformly

hard throughout, and surroundedwith atmospheres
of heat of no comparative magnitude.
It is

known from

experience, that the force of

a mass of elastic fluid

Whence

is

directly as the density.

derived the law already mentioned,

that the repulsive

versely as

is

its

power of each

diameter.

repulsive power,

if

That

we may

particle

is,

so speak

or absolute force of repulsion

is

is in-

the apparent
;

for the real

not known, as

long as we remain ignorant of the attractive force.

When we
its

expand any volume of

particles are enlarged, without

elastic

fluid,

any material

ON PURE ELASTIC FLUIDS.

148
change

in the quantity of their heat

it

follows

then, that the density of the atmospheres of heat

must

fluctuate with the pressure.

a measure of air

were expanded

Thus, suppose

into 8

measures

then, because the diameters of the elastic particles are as the

cube root of the space, the

dis-

tances of the particles would be twice as great as

before,

and the

elastic

atmospheres would occupy

nearly 8 times the space they did before, with nearly

the same quantity of heat

whence we see

that

these atmospheres must be diminished in density

in nearly the

Some
resist

them.

same

ratio as the

elastic fluids, as

mass of elastic

fluid.

hydrogen, oxygen, &c.

any pressure that has yet been applied

In such then

force of heat

is

it is

more than

to

evident the repulsive

a match for the affinity

of the particles, and the external pressure united.

To what
say
a

still

extent this would continue

we cannot

but from analogy we might apprehend that

greater pressure would succeed in giving

the attractive force the

superiority,

elastic fluid

would become a liquid or

other elastic

fluids, as

when

the

solid.

In

steam, upon the application

of compression to a certain degree, the elasticity

apparently ceases altogether, and the particles

collect in small drops of liquid,

and

fall

This phenomenon requires explanation.

down.

ON PURE ELASTIC FLUIDS.

From

149

the very abrupt transition of steam from

a volume of 1700 to that of

1,

without any mate-

one would be inclined

rial increase of pressure,

to think that the condensation of

it

was owing

the breaking of a spring, rather than to the

The

ing of one.

The

fact.

however

last

to

cwh-

I believe is the

condensation arises from the action

of affinity becoming superior to that of heat, by

which the
ened.

latter

As

but not weak-

overruled,

is

approximation of the particles

the

takes place, their repulsion increases from the

condensation of the heat, but their

creases,
till

it

should seem, in a

the approximation

degree,

when an

powers takes
result.

That

has

still

affinity in-

greater ratio,
a

attained

certain

equilibrium between those two

place,

and the

liquid, water, is the

this is the true explanation

we may

learn from what has been stated at page 131

wherein

it is

shewn

that the heat given off

condensation of steam,

is

in all probability

more than would be given

nently elastic

fluid,

densed into the

could

like

it

by the

off

no

by any perma-

be mechanically con-

volume, and

is

moreover a

small portion of the whole heat previously in

combination.

cerned

in

this

As

far

then as the heat

is

con-

phenomenon, the circumstances

would be the same, whether the approximation of

ON MIXED ELASTIC FLUIDS.

150

the particles was the eiFect of affinity, or of external mechanical force.

The

as water,

constitution of a liquid,

must

then be conceived to be that of an aggregate of

exercising in a most powerful

particles,

manner

the forces of attraction and repulsion, but nearly

in

an equal degree.

Of

this

SECTION

more

in the sequel.

2.

ON THE CONSTITUTION OF MIXED

ELASTIC FLUIDS.
When
cles

two or more

elastic fluids,

whose

parti-

do not unite chemically upon mixture, are

brought together, one measure of each,

they

occupy the space of two measures, but become

uniformly

difi'used

main

so,

The

fact admits

whatever

through each other, and re-

may be

their specific gravities.

of no doubt

but explanations

have been given in various ways, and none of

them completely

satisfactory.

As

the subject

is

one of primary importance in forming a system

of chemical principles,

more

we must

enter somewhat

fully into the discussion.

Dr. Priestley was one of the

earliest to notice

ON MIXED ELASTIC FLUIDS.

the fact
that

naturally struck

it

two

elastic

fluids,

him with

having apparently no

selves

according to their

liquids

do in like circumstances.

specific

was not the case

this

surprise,

each other, should not arrange them-

affinity for

found

151

gravities, as

Though he

after the elastic fluids

had once been thoroughly mixed, yet he suggests

it

as probable, that if

exposed

fluids

could be

each other without agitation, the one

to

specifically heavier

He

tion.

two of such

would retain

much

does not so

its

lower situa-

as hint at such

gases being retained in a mixed state by

With regard

to his suggestion of

affinity.

two gases being

carefully exposed to each other without agitation,

I

made

a series of experiments expressly to de-

termine the question, the results of which are

given in the Manchester Memoirs, Vol.

From

series.

these

it

1.

new

seems to be decided that

gases always intermingle and gradually diffuse

themselves amongst each other,

carefully

but

it

if exposed

ever so

requires a considerable time to

produce a complete intermixture, when the surface

of

communication

may vary from

is

small.

This

time

a minute to a day or more, ac-

cording to the quantity of the gases and the

freedom of communication.

When

or

by whom the notion of mixed gases

ON MIXED ELASTIC FLUIDS.

152

being held together by chemical

know

propagated, I do not

affinity

but

it

was

first

seems proba-

ble that the notion of water being dissolved in

led

air,

being dissolved in

air.

found that water gradually

dis-

that of air

to

Philosophers

appeared or evaporated
its elasticity

to

ance of the

air,

be unable to overcome the

was necessary

resist-

therefore the agency of affinity

In the

to account for the effect.

elastic

fluids

did not seem to be so

all

and increased

air,

but steam at a low temperature

was known

permanently

in

indeed,

much wanted,

able to support themselves

agency

this

as they are

but the diffusion

through each other was a circumstance which did

not admit of an easy solution any other way.
reg'ard to the

been

satisfied

nay,

for

air,

it

was

one might almost have

without the aid of experiment,

that the different gases

affinities

of water in

solution

natural to suppose,

In

water,

would have had

different

and that the quantities of

water dissolved in like circumstances, would have

varied according to the nature of the gas.

Saus-

sure found, however, that there was no difference

in this respect in the solvent
acid,

hydrogen

gas,

be expected that

powers of carbonic

and common

at least the

air.

It

might

density of the gas

would have some influence upon

its

solvent pow-

ON MIXED ELASTIC FLUIDS.

153

would take half the

ers, that air of half density

water, or the quantity of water would diminish in

some proportion
again

we

density

to the

are disappointed

faction, if

but even here

whatever be the rare-

water be present, the vapour produces

the same elasticity, and the hygrometer finally

settles at

extreme moisture, as

These

density in like circumstances.

sufficient to create

ception

how any

between

air

extreme

common

in air of

difficulty in

fiicts

are

the con-

principle of affinity or cohesion

and water can be the agent.

It is

truly astonishing that the same quantity of vapour

should cohere to one particle of

space, as to one

thousand

in the

dissolves water

and

a torricellian

in this instance

have vapour existing independently of

temperatures
able

is,

the

what makes

vapour

it

in such

given

same space. But

the wonder does not cease here

vacuum

air in a

we

air at all

still more remarkvacuum is precisely

the same in quantity and force as in the like

volume of any kind of

to

air of

extreme moisture.

These and other considerations which occurred

me some years ago, were sufficient to make me

altogether abandon the hypothesis of air dissolving

water, and to explain the

phenomena some other

way, or to acknowledge they were inexplicable.

In the autumn of 1801, I hit upon an idea which

ON MIXED ELASTIC FLUIDS.

154

seemed to be exactly
phenomena of vapour

calculated to explain the

;

it

gave

rise to a great

variety of experiments upon which a series of

essays were founded, which were read before the

Literary and Philosophical Society of Manchester,

and published

in the 5th Vol. of their

memoirs,

1802.

The

distinguishing feature of the

new theory

was, that the particles of one gas are not elastic

or repulsive in regard to the particles of another
gas, but only to the particles of their

own

kind.

Consequently when a vessel contains a mixture

of two such elastic fluids, each acts independently

upon the
if

vessel, with

its

proper

elasticity, just as

the other were absent, whilst no mutual action

between the

fluids

themselves

is

observed.

This

position most efl'ectually provided for the existence

of vapour of any temperature in the atmosphere,

because
to

it

and
more nor

support

neither

could have nothing but

;

it

was perfectly

less

vapour could exist in

extreme moisture, than

temperature.

So

far

theory was attained.

own weight
obvious why

its

in a

air

of

vacuum of the same

then the great object of the

The

law of the condensation

of vapour in the atmosphere by cold, was evidently

the same on this scheme, as that of the condensation

of pure steam, and experience was found to con-

ON MIXED ELASTIC FLUIDS.

conclusion at

firm the

only thing

all

now wanting

155

The

temperatures.

to completely establish

the independent existence of aqueous vapour in

the

atmosphere,

liquids

to

was the conformity of other

water, in regard to the diffusion and

This was found

condensation of their vapour.

to

take place in several liquids, and particularly in

sulphuric ether, one which was most likely to

shew any anomaly

to

advantage

if it

existed,

account of the great change of expansibility in

vapour

The

at ordinary temperatures.

of vapour in the atmosphere and

on
its

existence
occasional

its

condensation were thus accounted for; but another

question remained,
face

of water

atmosphere

how does

subject

The

it

to the

rise

from a sur-

pressure

consideration of this

no part of the essays above mentioned,

apprehended, that

if

From

it

made
being

the other two points could

be obtained by any theory,

in the sequel,

of the

this third too,

would,

be accomplished.

the novelty, both in the theory and the

experiments, and their importance, provided they

were
both

correct, the essays

at

were soon

home and abroad.

circulated,

The new

facts

and

experiments were highly valued, some of the latter

were repeated, and found correct, and none of

the results, as far as 1 know, have been contro-

ON MIXED ELASTIC FLUIDS.

156
verted

but the theory was almost universally

and

misunderstood,

consequently

This must have arisen partly

reprobated.

at least

from

being too concise, and not sufficiently clear

my

in its

exposition.

Dr. Thomson was the

who

publicly animadverted

gentleman, so well

known

know,

upon the theory

this

for his excellent Sys-

tem of Chemistry, observed

that work, that the theory

as far as I

first,

in the first edition of

would not account

the equal distribution of gases

but

that,

for

granting

the supposition of one gas neither attracting nor

repelling another, the two must

still

arrange them-

selves according to their specific gravity.

the most general objection to

different kind

was adapted so

permanent

it

it

was admitted, that the theory

most uniform and

as to obtain the

diffusion of gases

that as one gas

But

was quite of a

but

was as a vacuum

measure of any gas being put

it

was urged,

to another, a

to a

measure of

another, the two measures ought to occupy the

space of one measure only.

Finding that

my

views on the subject were thus misapprehended,

I

wrote an illustration of the theory, which was

published in the 3rd Vol. of Nicholson's Journal,

for

November, 1802.

voured

In that paper I endea-

to point out the conditions of

mixed gases

ON MIXED ELASTIC FLUIDS.

more

my

at large, according to

particularly

touched

feature of

that of

it,

upon

more

or

were interposed
affecting the
particles

of

casually to

A.

particles of another gas B,

Or,

come

stated law,

in a direct line, without at all

action of the

any

if

it,

this

said

particle of

in contact with

two

were

one of A, and

pressure did not preclude

the cotemporary action of

particles of

and

discriminating

the

by the known

reciprocal

press against

hypothesis

two particles of any gas A,

repelling each other

whilst one

15?

all

upon the one

in

the surrounding
contact with B.

In this respect the mutual action of particles of

the

same gas was represented

magnetic action, which

is

resembling

as

not disturbed by the

intervention of a body not magnetic.

As

the subject has since received the animad-

versions of several authors, which

to notice

more

or less,

it

will

it is

out the order intended to be pursued.

shall

consider the objections to the

made by

expedient

be proper to point
First, I

new theory
own views

the several authors, with their

on the subject

and then

shall give w^hat modifi-

cations of the theory, the experience

and

reflection

of succeeding time has suggested to me.

The

authors are Berthollet, Dr. Thomson, Mr. Murray,

Dr. Henry, and Mr. Gough.

ON MIXED ELASTIC FLUIDS.

158

Berthollet in his Chemical Statics (1804) has

given a chapter on the constitution of the atmosphere, in which he has entered largely into a

new

discussion of the

chemist,

upon comparing the

ments made by
sier,

Watt,

De

&c.,

This celebrated

theory.

results of experi-

Luc, Saussure, Volta, Lavoi-

together

with

Lussac, and his own, gives his

fact,

that vapours

elasticity of

much

those

assent to the

full

kind increase the

of every

each species of gas

alike,

and just as

as the force of the said vapours in

and not only

so,

vapour in

and vapour

air

the same (Vol.

Gay

of

vacuo

but that the specific gravity of

1.

in

Sect.

vacuo

4.)

is

in all cases

Consequently he

adopts the theorem for finding the quantity of

vapour which a given volume of

which

have

laid

down

can dissolve,

air

namely,

P
where p represents the pressure upon a given
volume (1) of dry air, expressed in inches of
mercury, /= the force of the vapour in vacuo

at

the temperature, in inches of mercury, and s =

the space which the mixture of air and vapour

occupies under the given pressure,

tion.

So

he objects

far therefore

we

to the theory

J9,

after satura-

perfectly agree

by which

but

attempt to

ON MIXED ELASTIC FLUIDS.

159

explain these phenomena, and substitutes another

of his own.

The

first

objection I shall notice

is

one that

clearly shews Berthollet either does not under-

he

stand, or does not rightly apply the theory

opposes

interstices

he says, " If one gas

another,

of

vacancies, there

of volume

they were

would not be any augmentation

when aqueous

combined with

occupied the

though

as

the air

or ethereal vapour was

nevertheless there

is

one

proportional to the quantity of vapour added

humidity should increase the specific gravity of

the

air,

as has
is

whereas

it

renders

it

specifically lighter,

been already noticed by Newton."

This

the objection which has been so frequently

urged
if I

it

has even been stated by Mr. Gough,

understand him aright, in almost the same

words (Nicholson's Journal, Vol.

yet this last

gentleman

is

9?

page 162)

profoundly skilled in

the mechanical action of fluids.

drical glass vessel containing

Let

dry

air

tall cylin-

be inverted

over mercury, and a portion of the air drawn out

by a syphon,

till

an equilibrium of pressure

established within and without

is

let a small portion

of water, ether, &c., be then thrown up into the

vessel

the vapour rises and

stices of the air as a void

occupies the inter-

but what

is

the obvious

ON MIXED ELASTIC FLUIDS.

160

consequence

Why,

the surface of the mercury

being now pressed both by the dry

new

raised vapour,

and

without,

of air

is

and by the

air,

more pressed within than

enlargement of the volume

an

restore

unavoidable, in order to

Again, in the open

equilibrium.

there

is

air

the

suppose

were no aqueous atmosphere around the

earth, only an azotic

one

and an oxygenous one

= 23 inches of mercury,
= 6 inches. The air

being thus perfectly dry, evaporation would com-

mence with great speed.

The vapour

being constantly urged to ascend by

by the

first

formed

that below,

must, in the

and

as constantly resisted

first

instance, dilate the other two atmospheres

the ascending steam adds

(for,

upward

elasticity of the

quence of which

its

is dilitation.)

force to the

two gases, and

pressure, the

alleviates their

air,

in part

necessary conse-

At

last

when

all

the vapour has ascended, that the temperature

will

admit

of,

equilibrium

the aqueous atmosphere attains an

it

no longer presses upon the other

two, but upon the earth

their

original

In this case

the others

return to

density and pressure throughout.

it

is

true, there

would not be any

augmentation of volume when aqueous vapour

was combined with the air humidity would increase the weight of the congregated atmospheres,
;

ON MIXED ELASTIC FLUIDS.

I6l

but diminish their specific gravity under a given

One would have

pressure.

solution of the

this

phenomenon upon my hypothesis

was too obvious

in

thought that

any one

to escape the notice of

any degree conversant with pneumatic chemisBerthollet indeed

try.

division of the

**

enquires,

Is

such a

same pressure of the atmosphere

analogous with any physical property yet known?

Can

be conceived that an

it

which adds

exists,

volume

its

to that of another,

and which nevertheless does not

expansive force ?"
conceive

it,

Certainly

is,

act

and

Two

reduced

to
air

atoms

may

act

as inelastic bodies

So two

all,

unless

if

when

inelastically,

and therefore

Hydrogen gas and oxy-

gen gas form water in a given circumstance;

gas,

and oxygen gas, can

the

reciprocal

combinations

same

elastically,

in contact.

Berthollet observes, "

but

they were

particles of the

upon each other

and upon other bodies

not at

not only

magnets repel each

no doubt would be the same

kind of

its

do not act upon other bodies

same way, but merely

this

we can

by

it

upon each other with an expan-

sive force, yet they

in the

on

act

but bring an instance that must be

allowed to be in point.
other, that

substance

elastic

also

action

produce

which

azotic

nitric acid

decides

the

cannot be considered as a force

ON MIXED ELASTIC FLUIDS.

162

commencing

at the precise

it

ducing

effect,

its

moment

at

which

it is

must have existed long before pro-

manifested,

and increases gradually

becomes preponderant."

It is

till

it

no doubt true that

the opposite powers of attraction and repulsion

are frequently, perhaps constantly, energetic at

the same instant; but the effect produced in those

cases arises from the difference of the two powers.

When

the excess of the repulsive power above

the attractive in different gases

small and insignificant,

comparatively

which may be denominated neutral, and which

ter
I

is

constitutes that charac-

it

supposed to exist in the class of mixed gases

which are not observed

chemical union.

to manifest

any sign of

would not be understood

to

deny an energetic

affinity

between oxygen and

hydrogen,

mixed

state

is

&c., in a

but that

affinity

more than counterbalanced by the repulsion of

the heat, except in circumstances which

it is

not

necessary at present to consider.

Again, "Azotic gas comports

gas, in

the

Is

it

with oxygen

changes occasioned by temperature

and pressure, precisely

gas

itself

necessary to

like

one and the same

have recourse

to a sup-

position which obliges us to admit so great a dif-

ference of action without an ostensible cause ?" It

is

possible this

may appear an

objection to a person

ON MIXED ELASTIC FLUIDS.

who does

163

not understand the theory, but

can be any to one

who

such as atmospheric

air,

it

never

If a mixture of gas,

does.

containing azote pressing

with a force equal to 24 inches of mercury, and

oxygen with a force equal

to

6 inches, were sud-

denly condensed into half the compass, the azotic

gas would then evidently, on
with a force
force

just the

change

my hypothesis,

press

48 inches, and the oxygen with a

12 inches, making together 60 inches,

same

as

any simple gas.

a similar

would take place

in the elasticity of each

by heat and

And

Will the opposite theory of

cold.

Berthollet be equally free from this objection

We

shall presently

examine

Another objection

is

it.

derived from the very

considerable time requisite for a body of hydrogen

to

descend into one of carbonic acid

were as a vacuum

for another,

brium not instantly established

is

certainly plausible

we

why
?

is

if

one gas

the equili-

This objection

shall consider

it

more

at large hereafter.

In speaking of the pressure of the atmosphere

retaining water in a liquid state, which I deny,

Berthollet adopts the idea of Lavoisier, " that

without

it

the moleculse would be infinitely dis-

persed, and that nothing would limit their separation, unless their

own weight should

collect

them

ON MIXED ELASTIC FLUIDS.

164
to

form an atmosphere."

This, I

may remark,

is

not the language dictated by a correct notion on

the subject.
hilated,

Suppose our atmosphere were anni-

and the waters on the surface of the globe

were instantly expanded into steam

surely the

action of gravity would collect the moleculse into

an atmosphere of similar constitution to the one

we now

possess

but suppose the whole mass of

water evaporated amounted in weight to 30 inches

how could it support its own weight

common temperature ? It would in a short

of mercury,
at the

time be condensed into water merely by its weight,

leaving a small portion, such as the temperature

could support, amounting perhaps to half an inch

of mercury in weight, as a permanent atmosphere,

which would effectually prevent any more vapour

from

rising, unless

temperature.

there were an

increase

Does not every one know

of

that

water and other liquids can exist in a Torricellian

vacuum

at

low temperatures solely by the pressure

of vapour arising from them

What need

then

of the pressure of the atmosphere in order to pre-

vent an excess of vapourisation

After having concluded that

pressure of the

serial

" without the

atmosphere, liquids would

pass to the elastic state," Berthollet proceeds in

the very next paragraph to shew that the quantity

ON MIXED ELASTIC FLUIDS.

may in

of vapour in the atmosphere

165
be much

fact

more than would exist if the atmosphere were

suppressed, and hence infers, " that the variations
of the barometer occasioned

midity of the atmosphere

by those of the hu-

may be much

greater

than was believed by Saussure and Deluc."

cannot see

how

the author reconciles the opposite

conclusions.

The experiments

of Fontana on the distillation

of water and ether in close vessels containing

air,

are adduced to prove, that vapours do not penetrate

air

doubt

without resistance.

vapour cannot make

This
its

way

true no

is

in such cir-

cumstances through a long and circuitous route

without time, and

keep the vessel

densed by

its

form as

fast as

trating

in

the

if

cool,

and

sides,
it is

external

atmosphere

may be condown in a liquid

the vapour
fall

generated, without ever pene-

any sensible quantity

to

its

remote

extremity.

We

come now

the consideration of that

to

theory which Berthollet adopts in his explanation

of the

phenomena

ing to his
affinity.

of gaseous mixtures.

there

theory,

The one

is

are

strong,

Accord-

two degrees of

makes the

particles

of bodies approach nearer to each other,

generally expels heat

the effect of this

and

may be

ON MIXED ELASTIC FLUIDS.

166

called combination; for instance,

gas

is

when oxygen

put to nitrous gas, the two combine, give

out heat, are condensed in volume, and become

possessed of properties different from what they

The

weak it does not

sensibly condense the volume of any mixture,

had previously.

other

is

nor give out heat, nor change the properties of

the ingredients
or dissolution

may be called solution

instance, when oxygen gas

its effect

for

and azotic gas are mixed

constitute

atmospheric

in

due proportion, they

which they retain

air, in

their distinguishing properties.

It is

upon

tic fluid in

this

supposed solution of one

elas-

another that I intend to make a few

observations.

That

have not misrepresented

the author's ideas, will, I think, appear from the

"

following quotations.

mixed, whose action

no change

is

is

observed

When different gases are

confined to this solution,

in the

temperature, or in

the volume resulting from the mixture

may be

hence

it

concluded, that this mutual action of two

gases does not produce any condensation, and that

it

cannot surmount the effort of the

elasticity, or

the affinityfor caloric, so that the properties of each

gas are not sensibly changed

."

"Although both

the solution and combination of two gases are the

effect of a chemical action,

which only

differs in its

ON MIXED ELASTIC FLUIDS.

intensity, a real difference

maybe

tween them, because there

difference

established be-

very material

is

between the results

167

the combination

of two gases always leads to a condensation of

their volume,

and gives

new

rise to

properties

on their solution, the gases share in common the

changes arising from compression and temperature,

and preserve their individual properties, which are

only diminished in the ratio of the slight action

which holds them united." (Page 198.)

"The mu-

tual affinity of the gases can, therefore,

between them an

effect

which

is

difference of specific gravity, but which

to the elastic tension

which belongs

cule of both, so that the volume

this action

is

(Page 218.)

inferior

to each mole-

not changed by

like the gases."

Solution must be distinguished from

combination, not only because in the

the substances
it

is

the liquids which take the elastic state,

comport themselves afterwards

that

produce

greater than their

is

preserves

retained

by an

first,

each of

affinity so

weak,

" (Page 219.)

dimensions.

its

Again, ''It cannot be doubted, that the parts of

elastic fluids

cohesion, as

undergo

an

not happen

endued with the force of

the substances dissolved by them
are not

equal

distribution,

which

could

but by the means of a reciprocal

chemical attraction

that which

constitutes the

ON MIXED ELASTIC FLUIDS.

168

(Researches into the Laws

force of cohesion."

of chemical

apprehend, mistaken the

The

author means to say, that

has,

the translator

English idiom.

Here

Eng. Trans, page 57.)

affinity,

the parts of elastic fluids are endued with the

force of cohesion

terogeneous

but

this

He

particles.

he applies only to he-

certainly does not

mean

homogeneous

that the particles of

elastic fluids

possess the force of cohesion.

Newton has demonstrated from

the

phenomena

of condensation and rarefaction that elastic fluids

are constituted of particles, which repel one an-

other by forces w^hich increase in proportion as

distance

the

of their

centres

diminishes

in

other words, the forces are reciprocally as the

This deduction

distances.

the
are.

Laws
What

of elastic fluids

reason,

a pity

is

it

will stand as

long as

continue to be what they

that all

who attempt

or to theorise respecting

to

the constitu-

make themselves
tion of
thoroughly acquainted with this immutable Law,
elastic fluids,

and constantly hold

start

should not

it

any new project

in their
!

view whenever they

When we

contemplate a

mixture of oxygenous and hydrogenous gas, what

does BerthoUet conceive, are the particles that
repel each other according to the Newtonian Law?

The mixture must

consist of such

and he ought

ON MIXED ELASTIC FLUIDS.

in the

very

first

l69

instance to have informed us what

constitutes the unity of a particle in his solution.

If he grants that each particle of

its

oxygen

and each particle of hydrogen does the

unity,

same, then

we must conclude

action of two particles of

that of a particle of

mutual

that the

oxygen

is

the same as

oxygen and one of hydrogen,

namely, a repulsion according to the

stated,

retains

Law

above

which effectually destroys the supposed

But

by chemical agency.

solution

be sup-

if it

posed that each particle of hydrogen attaches

itself to

a particle of oxygen, and the two particles

so united form 07ie^ from which the repulsive

energy emanates, then the new

perfectly conform to the

elastic

case a true saturation will take place

number of

particles of

fluid

Law

Newtonian

may

in this

when

the

hydrogen and oxygen

in

a mixture happen to be equal, or at least in the

ratio of

some simple numbers, such

3, &c.

Now

take place

something like

when

as 1 to 2, 1 to

this

a real combination

for instance, steam,

and

nitric acid

does actually
is

formed, as

formed of a

mixture of oxygen and nitrous gas.

have new

elastic fluids, the

atoms of which repel

one another by the common Law, heat

out, a great condensation of

the

new

fluids

differ

Here we
is

given

volume ensues, and

from their constituents in

ON MIXED ELASTIC FLUIDS.

170

their chemical relations.

It

remains then to de-

termine whether, in the instance of solution, all

these effects take place in a " slight" degree ;
that

is,

degree as not to be cognis-

in so small a

able to any of the senses.

It certainly requires

an extraordinary stretch of the imagination to

admit the affirmative.

One

great reason for the adoption of

any other theory on the subject,

phenomena of the evaporation

or

this,

from the

arises

How

of water.

water taken up and retained in the atmosphere

It

cannot be in the state of vapour,

because the pressure

too great

is

it

consider that the surface of water

said,

is

there must

But when

therefore be a true chemical solution.

we

is

is

subject to

a pressure equal to 30 inches of mercury, and

besides this pressure, there

between the

is

a sensible affinity

particles of water themselves

does the insensible

affinity

water overcome both these powers

quite explicable

object of which

menon.

upon
is

Further,

why

how

It is to

me

this hypothesis, the leading

to account for this very phenoif

a particle of air has attached

a particle of water to

assigned

of the atmosphere for

it,

what reason can be

a superior particle of air should rob

an inferior one of

its

property,

possesses the same power

when each

particle

If a portion of

com-

l7l

ON MIXED ELASTIC FLUIDS.

mon

be dissolved in water and a

salt

acid added,

muriatic

little

there any reason to suppose the

is

combined

additional acid displaces that already

with the soda, and that upon evaporation the salt

is

not obtained with the identical acid

ously had
water,

Or,

if

oxygen gas be confined by

hydrogen of the water

to the air

from the same

is

be said in the

will

it

of one particle for one,

affection

mass of particles for another mass

ted action of

all

the atoms in

these energies are

it is

it

not the

that of a

is

the uni-

the atmosphere upon

reciprocal,

on the

is

it

But

the water, which raises up a particle.

like action

its

and receiving an equal quantity

Perhaps

constantly giving

case of air dissolving water, that

have a

previ-

there any reason to suppose that the

is

oxygen

it

air,

as all

the water must

and then an atmos-

phere over water would press downward by a

force greater than

its

weight, which

is

contradicted

by experience.

When

two measures of hydrogen and one of

oxygen gas are mixed, and

spark,

the whole

is

fired

by the

electric

converted into steam, and

if

the pressure be great, this steam becomes water.

It is

most probable then that there

number
as in

is

the same

of particles in two measures of hydrogen

one of oxygen.

Suppose then three mea-

ON MIXED ELASTIC FLUIDS.

172

sures of hydrogen are

mixed with one

of oxygen,

and this slight affinity operates as usual

the union effected

According

how

is

to the principle

of equal division, each atom of oxygen ought to

have one atom and a half of hydrogen attached

one half of the atoms
to it, but this is impossible
;

oxygen must then take two of hydrogen, and

But the former would
the other half, one each.
of

be specifically lighter than the

to

be found

like this is

at the

and ought

latter,

top of the solution

nothing

however observed on any occasion.

Much more

might be advanced

to

shew the ab-

surdity of this doctrine of the solution of one gas

in another,

and the insufficiency of

it

to explain

any of the phenomena indeed I should not have

dwelt so long upon it, had I not apprehended
;

was

that respectable authority

credit,

likely to give

more than any arguments

in

its

it

behalf

derived from physical principles.

Dr. Thomson,
of Chemistry,
the subject of

in the

3rd Edition of

his

System

has entered into a discussion on

mixed gases

he seems to compre-

my

notions on

these subjects, with great acuteness.

He does
my hypo-

hend the excellence and defects of

not conclude with Berthollet, that on

thesis, ''there

would not be any augmentation of

volume when aqueous and ethereal vapour was

ON MIXED ELASTIC FLUIDS.

combined with the

mon an

air,"

tion which this

which has been so com-

There

objection.

173

is

however one objec-

gentleman urges, that shews he

does not completely understand the mechanism

of

my

At page

hypothesis.

448,

Vol.

he

3.

observes that from the principles of hydrostatics,

" each particle of a

Nor can

sure.

fluid sustains the

whole pres-

any reason why

I perceive

this

principle should not hold, even on the supposition

well founded."

Upon

would observe, that when once an

equili-

that Dalton's hypothesis

this

brium

is

established in any mixture of gases, each

particle of gas

particles

is

of

is

its

pressed as

own kind

if

by the surrounding
It is in the re-

only.

nunciation of that hydrostatical principle that the

The

leading feature of the theory consists.

est particle of

the weight of

oxygen

all

in the

the particles of

and the weight of no other.

maxim with me,

low-

atmosphere sustains

oxygen above

It

it,

was therefore a

that every particle

of gas

is

equally pressed in every direction, but the pressure arises from the particles of

its

own kind

Indeed when a measure of oxygen

measure

come

of azote, at the

in contact, the

against those of the

moment

is

only.

put to a

the two surfaces

particles of each gas press

other with their

full

force

but the two gases get gradually intermingled, and

ON MIXED ELASTIC FLUIDS.

174

the force which each particle has to sustain proportionally diminishes,

same
its

at last

till

The

volume.

just

becomes the

as that of the original gas dilated to twice

.26

cube

ratio of the forces is as the

root of the spaces inversely

or as

it

nearly.

that

is,

as ^\/2

1,

In such a mixture as has

been mentioned, then, the common hypothesis

supposes the pressure of each particle of gas to

be 1.26

whereas mine supposes

it

only to be

but the sum of the pressure of both gases on the

containing vessel, or any other surface,

is

exactly

the same on both hypotheses.

Excepting the above objection,

all

the rest

which Dr. Thomson has made, are of a nature

not so easily to be obviated

he takes

notice of

the considerable time which elapses before two

gases are completely diffused through each other,

as Berthollet has done,

makes against the

a

vacuum

and conceives

this

supposition, that one gas

to another.

He

fact
is

as

further objects, that if

the particles of different gases are inelastic to each

other, then

a particle of

actual contact with one

unite with

it,

and form a

oxygen coming

into

of hydrogen, ought to
particle of water

but,

on the other hand, he properly observes, that the

great facility with which such combinations are
effected in such instances as mixture of nitrous

ON MIXED ELASTIC FLUIDS.

and oxygen gas,
hypothesis.
jection

state

is,
;

an argument in favour of the

Dr. Thomson founds another

upon the

when one
that

is

175

facility of certain

of the ingredients

is

in a

ob-

combinations,

nascent form,

just upon the point of assuming the elastic

he observes,

this,

*'

seems incompatible

with the hypothesis, that gases are not mutually

Upon

elastic."

the whole. Dr.

Thomson

property of dissolving each other

" however problematical

it

inclines

have the

to the opinion of Berthollet, that gases

and admits,

may appear

at first

view, that the gases not only mutually repel each

other, but likewise mutually attract."

doubt

if

have no

he had taken due time to consider

this

conclusion, he would, with me, have pronounced

it

absurd

but of this again in the sequel.

With regard to the objection,

makes a more durable resistance to
of another, than

it

This occurred to
speculations

ought to do on

me

in a

that one gas

the entrance

my

hypothesis

very early period of

I devised the train

my

of reasoning

which appeared to obviate the objection

but

it

being necessarily of a mathematical nature, I did

not wish to obtrude

it

upon the notice of chemical

philosophers, but rather to wait

for.

The

to the

resistance which

till it

was called

any medium makes

motion of a body, depends upon the surface

ON MIXED ELASTIC FLUIDS.

176

of that body,
greater,

all

and

greater as the surface

is

is

other circumstances being the same.

diameter meets with a

ball of lead 1 inch in

certain resistance in falling through the air

made

the same ball, being

but

into a thousand smaller

ones of To of an inch diameter, and falling with

the same velocity, meets with 10 times the resis-

tance

it

did before

because the force of gravity

increases as the cube of the diameter of any particle,

and the resistance only

Hence

diameter.

as the square of the

appears, that in order to

it

increase the resistance of particles

medium,

it is

that the resistance will be a

division

is

maximum.

consider particles
their

own

moving

in

any

only necessary to divide them, and

gravity,

We

maximum when

the

have only then to

of lead falling through air

by

and we may have an idea of

the resistance of one gas entering another, only

the ^articles of lead
nitely small^ if I

Here we

must he conceived

may be

shall find

to he infi-

allowed the expression.

great resistance, and yet no

one, I should suppose, will say, that the air and

the lead are mutually elastic.

The
shall

other two objections of Dr. Thomson, I

waive the consideration of

Mr. Murray has

lately

at present.

edited a system

of

chemistry, in which he has given a very clear

ON MIXED ELASTIC FLUIDS.

description of the

and of other

He

phenomena

177

of the atmosphere,

mixtures of elastic

similar

fluids.

has ably discussed the diff'erent theories that

have been proposed on the subject, and given a

perspicuous view of mine, which he

thinks

is

ingenious, and calculated to explain several of the

phenomena

well, but

satisfactory with that

upon the whole, not equally

He

which he adopts.

not object to the mechanism, of

my

does

hypothesis in

regard to the independent elasticity of the several

gases entering into any mixture, but argues that

phenomena do not require so extraordinary a

postulatum
and more particularly disapproves
the

my

theory to account for

principal feature in

Mr. Murray's theory,

of the application of
evaporation.

The

and which he thinks distinguishes

is

it

from mine,

" that between mixed gases, which are capable,

under any circumstances of combining, an

tion

must always be exerted."

to recount the

clusion, because

It is

unnecessary

arguments on behalf of
it

Murray announces

will

attrac-

this con-

not be controverted. Mr.

his views of the constitution

of the atmosphere, as follows

" Perhaps that

chemical attraction which subsists between the

solid bases of these gases, but which,

are merely
z

when they

mixed together, cannot, from the

dis-

ON MIXED ELASTIC FLUIDS.

178

tance at which their particles are placed

repulsive power of caloric, bring

union,
their

may

still

separation

be so

them into intimate

far exerted, as to

they

or,

by the

may be

prevent

retained in

mixture by that force of adhesion, which, exerted

at the surfaces of

many

these notions at length

them

bodies, retains

He

contact with considerable force."

by various

in

supports

observations,

and repeats some of the observations of Berthollet,

whose doctrine on
is

this subject, as

has been seen,

nearly the same.

Before we animadvert on these principles,

may be
farther,

convenient to extend the

and

to adopt as a

first

it

little

maxim, " that between

the particles of ;;wr^ gases, which are capable

under any circumstances of combining, an

tion

must always be exerted."

ray cannot certainly object

steam, a pure elastic

fluid,

to,

in the

distinction
relative,

for

Mur-

case of

the particles of which

known in certain circumstances

Nor will it be said that steam and

are

gas are different

attrac-

This, Mr.

to combine.

a permanent

he justly observes, "

(between gases and vapours)

is

this

merely

and arises from the difference of tem-

perature at which they are formed

the state

with regard to each, while they exist in

precisely the same."

Is

it,

is

steam then constituted

ON MIXED ELASTIC FLUIDS.

of particles in which the attraction

erted as to prevent their separation

exhibit

no traces of

number

What

is this

oxygen

then

notwithstanding

all bodies^ at all times,

it

is

is

so far ex-

No

they

more than the

attraction,

of particles of

gaseous form.

179

when

do,

like

in the

the conclusion

It

must he allowed^ that

and

in every situation^

attract one another; yet in certain circumstances,

they are likewise actuated hy a repulsive power;

the only efficient motive force

is

then the

differ-

ence of these two powers.

From the

circumstance of gases mixing together

without experiencing any sensible diminution of

volume, the advocates for the agency of chemical

affinity, characterise it as a " slight action," and
**

a weak reciprocal action

are consistent

being so

far

but

:"

So

far I think

when we hear

they

of this affinity

exerted as to prevent the separation

of elastic particles, I do not conceive with what

propriety

it

can be called weak.

Suppose

this

affinity

should be exercised in the case of steam

of 212

then the attraction becoming equal to

the repulsion, the force which any one particle

would exercise must be equal

column of water of 4896

to the

weight of a

feet high.

(See page

146.)
It

is

somewhat remarkable

that

those gases

ON MIXED ELASTIC FLUIDS.

180

which are known to combine occasionally, as

azote and oxygen, and those which are never

known

combine, as hydrogen and carbonic

to

acid, should dissolve

lity

one another with equal

faci-

nay, these last exercise this solvent power

with more effect than the former

for,

hydrogen

can draw up carbonic acid from the bottom to the

top of any vessel, notwithstanding the latter

times the specific gravity of the former.

is

20

One

would have thought that a force of adhesion was

more

to

be expected in the particles of steam,

than in a mixture of hydrogen and carbonic acid.

But

it

is

the business of those

who adopt

the the-

ory of the mutual solution of gases to explain these

difficulties.

In a mixture where are 8 particles of oxygen

for

1 of

hydrogen,

it

is

demonstrable that the

central distances of the particles of hydrogen are

at a

medium

Now

twice as great as those of oxygen.

supposing the central distance of two adja-

cent particles of hydrogen to be denoted by 12,

query, what

is

supposed to be the central distance

of any one particle of hydrogen from that one

this

oxygen with which

weak chemical union ? It

those

who understand and main-

particle, or those particles of

it is

connected by

would be well
tain

the

if

doctrine

of chemical

solution

would

ON MIXED ELASTIC FLUIDS.

how they conceive this

enable those who are desirous to
represent

be

to

it,

who

would

are dis-

to point out its defects with

The greatest possible

precision.

it

learn, to obtain

a clear idea of the system, and those

satisfied with

181

more

central distance

would be 8j in the above instance, and the least

might perhaps be 1. Berthollet, who decries the
diagram by which I endeavoured

to illustrate

ideas on this subject, has not given us

cise

information,

my

any pre-

either verbally or otherwise,

relative to the collocation of the heterogeneous

particles, unless

it is

sideration

that the

mixture of

fluids

to

be gathered from the con-

affinity is

preserves

can this

weak

repulsive

power of

its

affinity do,

so

weak

that the

What

dimensions.

when opposed by a

infinite superiority ?

In discussing the doctrines of elastic fluids

mixed with vapour, Mr. Murray seems disposed

to

question the accuracy of the

quantity of vapour

is

it

more abounds.

touchstone
theories

of

and

the
I

that the

the same in vacuo as in air,

though he has not attempted

case

fact,

to ascertain in

This

mechanical

is

which

certainly the

and

chemical

had thought whoever admitted

the truth of the fact, must unavoidably adopt the

mechanical theory.

Berthollet, however, con-

vinced from his own experience, that the fact was

ON MIXED ELASTIC FLUIDS.

182

incontrovertible, attempts to reconcile

as

it is,

it is

to the chemical theory

left to others to judge.

Berthollet

inimical

with what success

Mr. Murray joins with

condemning

in

it,

as

extravagant the

position which I maintain, that if the atmosphere

were annihilated, we should have

aqueous vapour than

Upon which

at

presents

little

exists

more
in

it.

I shall only remark, that if either of

those gentlemen will calculate, or give a rough

estimate upon their hypothesis, of the quantity of

aqueous vapour that would be collected around

the earth, on the said supposition, I will engage
to discuss the subject with

In 1802, Dr.

them more

Henry announced

at large.

a very curious

and important discovery, which was afterwards

published

the

in

Philosophical

Transactions

namely, that the quantity of any gas absorbed by

water

is

increased in direct ^proportion

to the

pressure of the gas on the surface of the water.

Previously to this, I was engaged in an investigation of the quantity of carbonic acid in the

atmosphere

it

was matter of surprise

to

me

that

lime water should so readily manifest the presence

of carbonic acid in the

air,

whilst pure water

by

exposure for any length of time, gave not the

least

traces of that acid.

of time ought to

I thought that length

compensate

for

weakness of

ON MIXED ELASTIC FLUIDS.

affinity.

183

In pursuing the subject I found that the

quantity of this acid taken up

or less in proportion to
in the gaseous mixture,

and therefore ceased

its

by water was greater

greater or less density

incumbent upon the surface,

to

be surprised

at

water ab-

sorbing so insensible aportion from the atmosphere.

I

had not however entertained any suspicion that

this

law was generally applicable to the gases

ately upon this,

till

Immedi-

Dr. Henry's discovery was announced.

struck me as essentially necessary

it

in ascertaining the quantity of

any gas which a

given volume of water will absorb, that we must

be careful the gas

with

is

perfectly pure or

any other gas whatever

maximum

effect for

otherwise

Henry, and found

it

to

be correct

became expedient

his experiments relating to

in

consideration of

all

became convinced,
elastic fluids

consequence

to repeat

some of

the quantity of gas

absorbed under a given pressure.

telligible

the

any given pressure cannot be

This thought was suggested to Dr.

produced.

of which,

unmixed

Upon due

these phenomena, Dr. Henry

that there

was no system of

which gave so simple, easy and

in-

a solution of them, as the one I adopt,

namely, that each gas in any mixture exercises a

distinct pressure,

which continues the same

other gases are withdrawn.

if

the

In the 8th Vol. of

ON MIXED ELASTIC FLUIDS.

184

Nicholson's Journal,

may be

seen a letter ad-

Henry

dressed to me, in which Dr.

pointed out his reasons for giving

In the 9th Vol.

preference.

is

has clearly

my

theory a

a letter from

Mr.

Gough, containing some animadversions, which

were followed by an appropriate reply from Dr.

Henry.
In the 8th, 9th, and 10th Volumes of Nicholson's Journal,

chester

and

in the first

Memoirs (new

series)

Vol. of the

may be

seen some

my

doctrine of

animadversions of Mr. Gough, on

mixed

gases, with

some of

Man-

his

own

opinions on

Mr. Gough conceives the

the same subject.

atmosphere to be a chemical compound of gases,

vapour, &c., and he rests his belief chiefly upon

observance of certain hygrometrical pheno-

the

mena, such as that

air

absorbs moisture from

bodies in certain cases, and in others restores

to them,

shewing that

air

it

has an affinity for water,

which may be overcome by another more powerful one.

is

This opinion, as Mr. Murray observes,

the one

we have from Dr. Halley

supported be

Le Roy, Hamilton and

and might be considered

till

it

was

Franklin,

as the prevailing opinion,

Saussure, in his celebrated Essays on Hygro-

metry, published in 1783, suggested that water

was

first

changed

into vapour,

and was

in that

ON MIXED ELASTIC FLUIDS.

state

dissolved by the

185

This amphibious

air.

theory of Saussure does not seem to have gained

any converts

to

though

it,

it

pointed out the

in-

Finally, the theory of the

stability of the other.

chemical solution of water in

received

air,

its

death blow in 1791, by the publication of Pictet's

Essay on Fire, and more particularly by

paper on evaporation, published

all

in the Philoso-

These gentlemen

phical Transactions for 1792.

demonstrated, that

De Luc's

the train of hygrometrical

phenomena takes place just as well, indeed rather

quicker, in a vacuum than in air, provided the
same quantity of moisture

is

All the

present.

influence that any kind or density of air has,

retard the effect

but in the end

is

to

becomes the

it

same.

The

only objection which Mr.

sented that appears to

is

me

to raise

any

that in regard to the propagation of

the atmosphere consist

elastic

to

Gough

media,

it is

chiefly

difficulty,

sound

be heard double

that

is,

same sound

as

was brought

it

By calculation

I find that if

sound move

per second

an atmosphere of azotic gas,

to

move

at the rate of

in the other gases as follows

A a

If

the

or other of the atmospheres.

in

of two distinct

urged that distant sounds ought

would be heard twice, according

by one

has pre-

1000
it

feet

ought

namely,

ON MIXED ELASTIC FLUIDS.

186

Feet.

Sound moves

in azotic

1000 per second.

gas

oxygen gas

930

carb. acid

804

aqueous vap.

According

to

this

1175

table, if a strong

sound were produced 13 miles

be a weak impression of

it

off,

the

and loud

would

first

brought by the atmos-

phere of aqueous vapour,

in

59 seconds

second would be the strongest of

brought by

all,

the atmosphere of azotic gas, in 68^ seconds

third

would be much

inferior

to

the

the

the second,

by the oxygenous atmosphere, in 74

the fourth and last brought by the carseconds
bonic acid atmosphere would be extremely weak
Now, though observation does
in 85 seconds.
brought

not perfectly accord with the theory in this respect,

it

comes

that of the

as near

more simple

it,

perhaps, as

it

does to

constitution of the atmos-

phere which Mr. Gough maintains.

Derham,
who has perhaps made the greatest number of
accurate observations on distant sounds, remarked
that the report of a

cannon

fired at the distance

of 13 miles from him, did not strike his ear with

a single sound, but that

times close to each other.

was repeated 5 or 6

it

"

were louder than the third

The two

first

cracks

but the

last

cracks

ON MIXED ELASTIC FLUIDS.

were louder than any of the

rest."

18?
Cavallo, in

experimental philosophy, after quoting the

above observations, proceeds, " this repetition of
his

the sound probably originated from the reflection

of a single sound, from

objects, not
it

much

houses, or other

hills,

distant from the cannon.

But

appears from general observation, and where

no echo can be suspected, that the sound of a

cannon, at the distance of
ferent from the sound
case, the crack

we cannot

is

or 20 miles,

when

near.

appreciate the height.

may be compared

musical sound
ous,

it

begins

softly, swells to its

may be observed

it is

in

a grave

to a determinate

Nearly the

are likewise

by the distance."

now proceed

greatest loudness,

same

with respect to a clap of

thunder, other sounds

quality

Whereas

and instead of being instantane-

and then dies away growling.

thing

In the latter

loud and instantaneous, of which

the former case, viz. at a distance,

sound, which

dif-

is

(Vol.

2.

altered in

page 331.)

my present views
on the subject of mixed gases, which are somewhat different from what they were when the
I shall

to give

theory was announced, in consequence of the

fresh

lights

diffused.

which succeeding experience has

In prosecuting

nature of elastic

fluids, I

my

enquiries into the

soon perceived

it

wa>

ON MIXED ELASTIC FLUIDS.

188
necessary,

if possible,

whether the

to ascertain

atoms or ultimate particles of the different gases

are of the same size or volume in like circum-

By

stances of temperature and pressure.

or volume of an ultimate particle, I
place, the space
elastic fluid

it

bulk of the particle

bulk of the supposed impenetrable

signifies the

nucleus, together with that of

repulsive atmosphere of heat.

fused idea, as

many

surrounding

its

At

formed the theory of mixed gases,

the time I

had a con-

have, I suppose, at this time,

that the particles of elastic fluids are

same

size

that a given

in this

occupies in the state of a pure

in this sense the

mean

the size

gas contains just as

many

volume of hydrogenous

all

of the

volume of oxygenous
particles as the

same

or if not, that we had

no data from which the question could be solved.

But from a

train

of reasoning, similar to that

exhibited at page 71, I became convinced that

diff'erent

same
as a

trary

size

gases have not their particles of the

:

maxim,
:

and that the following may be adopted

till

some reason appears

to the con-

namely,

That every species of pure elastic fluid has its

particles globular and all of a size ; hut that no
two species agree in the size of their

pai'ticles^

the pressure and temperature being the same.

ON MIXED ELASTIC FLUIDS.

189

There was another thing concerning which

was dubious
repulsion.

whether heat was the cause of

was rather inclined

to

ascribe re-

pulsion to a force resembling magnetism, which

on one kind of matter, and has no

acts

For,

another.

if

effect

on

heat were the cause of repulsion,

why

there seemed no reason

a particle of

oxygen

should not repel one of hydrogen with the same

force as

one of

its

were both of a

size.

deration, I see

no

the

common
and

heat

may

still

own

kind, especially if they

Upon more mature

consi-

reason for discarding

sufficient

opinion, which ascribes repulsion to

I think the

phenomena of mixed gases

be accounted

for,

by

repulsion, without

the postulatum, that their particles are mutually

inelastic,

and free from such of the preceding ob-

jections as I have left unanswered.

When we

contemplate upon the disposition of

the globular particles in a volume of pure elastic

fluid,

we perceive

must be analogous to that of

it

the particles must be dis-

a square pile of shot

posed into horizontal

strata,

forming a square
particle rests

of

its

in

upon four

contact with

all

each four particles

a superior stratum, each

particles below, the points

four being 45 above the ho-

rizontal plane, or that plane which passes through

the centres of the four particles.

On

this

account

ON MIXED ELASTIC FLUIDS.

190

the pressure

is

steady and uniform throughout.

But when a measure of one gas

measure of another

in

any

is

vessel,

presented to a

we have then

a surface of elastic globular particles of one size

an equal surface of particles of

in contact with

another

in such case the points of contact of the

heterogeneous particles must vary

from 40
from

to 90

all

the

way

an intestine motion must arise

this inequality,

and the

particles of

one kind

The

be propelled amongst those of the other.

same cause which prevented the two

elastic sur-

faces from maintaining an equilibrium, will always

subsist, the particles of
size,

one kind being, from their

unable to apply properly to the other, so

no equilibrium can ever take place amongst

The intestine mothe heterogeneous particles.

that

tion

must therefore continue

till

the

particles

arrive at the opposite surface of the vessel against

any point of which they can rest with

and the equilibrium

each gas

is

stability,

at length is acquired

when

uniformly diffused through the other.

In the open atmosphere no equilibrium can take

place in such case

till

the particles have ascended

so far as to be restrained
that

is,

It is

till

by

their

own weight

they constitute a distinct atmosphere.

remarkable that when two equal measures

of different gases are thus diffused, and

sustain

ON MIXED ELASTIC FLUIDS.

IQl

an invaried pressure, as that of the atmosphere,

the pressure upon each particle after the mixture
is

This points out the active

less than before.

principle of diffusion

for, particles

of fluids are

always disposed to move to that situation where

the pressure

least.

is

Thus, in a mixture of

equal measures of oxygen and hydrogen,

common

being denoted by

1,

that after the mixture

the gas becomes of half

noted by

the

pressure on each particle before mixture

Vi=

its

density, will

when

be de-

.794.

This view of the constitution of mixed gases

agrees with that which I have given before, in
the two following particulars, w^hich I consider as
essential to
it

plausibility.

The

1st.
is

every theory on the subject to give

effected

to the

diffusion of gases

by means of the repulsion belonging

homogenous

which

is

through each other

particles

or to that principle

always energetic to produce the dilata-

tion of the gas.

2nd.

When

any two or more mixed gases

ac-

quire an equilibrium, the elastic energy of each

against the surface of the vessel or of any liquid,
is

precisely the same as

if it

were the only gas

present occupying the whole space, and

rest

were withdrawn.

all

the

ON MIXED ELASTIC FLUIDS.

192

In other respects I think the last view accords

better with the phenomena, and obviates the objections which Dr.

the former

why mixed

Thomson has brought

against

particularly in regard to the query,

gases that are

known on

certain oc-

casions to combine, do not always combine

why any

more disposed

to

combination than when

already assumed the elastic form.

more

and

gaseous particle in its nascent state

It

it

is

has

will also

clearly explain the reason of one gas making

so powerful

and durable a resistance to the en-

trance of another.

One

difficulty still

remains respecting vapour,

which neither view of the subject altogether

removes

though vapour may subsist

in the

at-

mosphere upon either supposition, as far as the

temperature will admit, not being subject to any

more pressure than would arise from its own

particles, were the others removed, yet it may
be enquired, how does

it

rise

from the surface of

water subject to the pressure of the atmosphere?

how does

vapour, which ascends with an elastic

force of only half an

itself

inch

from water when

30 inches of mercury
This

difficulty applies

theories

of the

it

to

of mercury, detach

has

oppose

nearly the

solution

weight of

the

of water

its

ascent?

same
in

to

air,

all

and

ON MIXED ELASTIC FLUIDS.

it is

193

therefore of consequence for every one, let

him adopt what opinion he may, to remove it.

Chemical solution but ill explains it; for, the
of air for vapour

affinity

weak, and yet

always described as

sufficient to

is

it

is

overcome the

pressure of a powerful force equal to the weight

of the atmosphere.

page 284) what

ject are.

It

have endeavoured

(Manch. Memoirs,

in another place
series^

my own

to

shew

vol. 1.

new

ideas on the sub-

appears to me, that

it is

not

till

the

depth of 10 or 12 strata of particles of any liquid,

that the pressure

becomes uniform
cles

in

the

uppermost stratum are

subject to but

Bb

upon each perpendicular column

and that several of the parti-

little

pressure.

in

reality

ON LIQUIDS.

194

SECTION

3.

ON THE

CONSTITUTION OF LIQUIDS,
AND THE MECHANICAL RELATIONS BETWIXT
LIQUIDS AND ELASTIC FLUIDS.

A liquid

may be

or inelastic fluid

defined to

be a body, the parts of which yield to a very

small impression, and are easily

another.

This definition

moved one upon

may

for the

suffice

consideration of liquids in an hydrostatical sense,

but not in a chemical sense.

there

is

no substance

cause of elasticity,

the

word

elastic to

inelastic

if

heat be the

bodies containing

all

necessarily be elastic

Strictly speaking,

but

such

it

must

we commonly apply

fluids

only as have the

property of condensation in a very conspicuous

degree.
if it is
little,

Water

is

a liquid or inelastic fluid

compressed by a great

and again recovers

the pressure

is

removed.

its

force,

original

We

it

but

yields a

bulk when

are indebted to

ON LIQUIDS.

Mr. Canton

195

is

demon-

about

aTrroth

the compressibility of several liquids

Water, he found,

strated.

part of

its

by which

for a set of experiments,

lost

bulk by the pressure of the atmos-

phere.

When we
steam,

consider the origin of water from

we have no reason

pressibility,

and that

would be wonderful

The

in a

if

to

sure of the atmosphere

it

times

very small degree

it

is

equal to the pres-

what a prodigious force

have when condensed 15 or 18 hundred

We

reduced

to

know

that the particles of steam,

the state of water,

greatest part of their heat.

still

What

resistance then ought they not to

compressing force

The

analogy, other liquids,

dies,

com-

at its

water had not this quality.

force of steam at 212

must

wonder

truth

retain the

a powerful

make

is,

against a

water, and

must be considered

by

as bo-

under the controul of two most powerful and

energetic agents, attraction and

tween which there

compressing force

is
is

repulsion, be-

an equilibrium.
applied

it

If

any

yields, indeed,

but in such a manner, as a strong spring would

yield,
pitch.

when wound up almost to the highest

When we attempt to separate one por-

tion of liquid

from another, the case

here the attraction

is

is

different

the antagonist force, and

ON LIQUIDS.

196
that

being balanced

heat, a

by the repulsion of the

moderate force

capable of producing

we

But even here

the' separation.

the attractive

is

force

to prevail,

perceive

does this arise

particles

It

being a

there

Whence

manifest cohesion of the particles.

should seem that

when two

of steam coalesce to form water, they

take their station so as to effect a perfect equili-

brium between the two opposite powers

any foreign force intervene, so

but

if

as to separate the

two molecules an evanescent space, the repulsion

decreases faster than the attraction, and conse-

quently this last acquires a superiority or excess,

which the foreign force has

were not the

case,

why do

to

overcome.

they at

first,

If this

upon

or

the formation of water, pass from the greater to

the less distance

With regard
ment of particles

and arrange-

to the collocation
in

an aggregate of water or any

other liquid, I have already observed (page 139)

that this
air.

It

is

not, in all probability, the

same

as in

seems highly improbable from the phe-

nomena

of the expansion of liquids

law of

expansion

fine liquids to

is

by

unaccountable

heat.

The

we

con-

if

one and the same arrangement of

their ultimate particles in all temperatures

we cannot avoid concluding,

if that

were the

for,

case,

MUTUAL ACTION OF FLUIDS.

197

the expansion would go on in a progressive

with the heat, like as in air

way

and there would be

no such thing observed as a point of temperature, at which the expansion was stationary.
;

RECIPROCAL PRESSURE OF LIQUIDS AND ELASTIC

FLUIDS.

When

an elastic

fluid is

confined by a vessel

of certain materials, such as wood, earthenw^are,

&c.

it

found slowly to communicate with the

is

external
till

is

air,

to

give and receive successively,

a complete intermixture takes place.

no doubt but

this is

There

occasioned by those vessels

being porous, so as to transmit the

fluids.

Other

vessels, as those of metal, glass, &c. confine air

These therefore cannot be

most completely.
porous

or rather, their pores are too small to

admit of the passage of

air.

I believe

no

sort of

vessel has yet been found to transmit one gas

and confine another

such a one

in practical chemistry.

is

a desideratum

All the gases appear to

be completely porous, as might be expected, and

therefore operate very temporarily in confining

each other.

Do

they

gases,

in

How

are liquids in this respect

resemble
regard

to

glass,

their

or

earthenware,

or

power of confining

MUTUAL ACTION OF FLUIDS.

198

elastic fluids

Do

they confine

do

they treat

some,

gases alike, or

all

and

transmit

others.

These are important questions they are not to

We must patiently
be answered in a moment.
:

examine the

facts.

Before we can proceed,

down

lay

a rule,

if

it

will

possible,

be necessary to

by which

to

dis-

tinguish the chemical from the mechanical action

of

I think the

a liquid upon an elastic fluid.

following cannot well be objected to

When an

kept in contact with a liquid, if

elastic fluid is

any change is ^perceived, either in the elasticity

or any other property of the elastic fluid, so far
the mutual action must he pronounced chemical hut if NO change is perceived, either in
the elasticity or any other property of the
elastic fluid, then the mutual action of the two
:

must he pronounced wholly mechanical.

If a quantity of lime be
agitated,

lime

falls

upon standing a

kept in water and

sufficient

time,

the

down, and leaves the water transparent

but the water takes a small portion of the lime

which

Laws

it

permanently retains, contrary

of specific gravity.

portion of lime

is

Why ?

dissolved

to the

Because that

by the water.

If a

quantity of air be put to water and agitated,

upon standing

a sufficient time,

the air rises up

MUTUAL ACTION OF FLUIDS.

and leaves

to the surface of the water

parent

199
trans-

it

but the water permanently retains a

portion of

air,

Why?

gravity.

Laws

contrary to the

air is dissolved

of specific

Because that small portion of

by the water.

So

two

far the

explanations are equally satisfactory.

But

we

if

place the two portions of water under the receiver of an air pump, and exhaust the incumbent
air,

the whole portion of air absorbed by the

water ascends, and

drawn out of the receiver

is

whereas the lime remains

before.

If

now

the air retained in the water

water

is

an

upon

none on that

for lime

its

is

The water

in.

it

the pressure on the surface

effect

why

on the surface

elastic force

which holds

as

The answer must

appears passive in the business.

have some

solution

the question be repeated,

be, because there

of the

in

still

But, perhaps,

of the water

may

for air,

and

affinity

Let the

air

be drawn

off

from the surface of the two portions of water,

and another species induced without alleviating

the

pressure.

changed

The

lime

water

the air escapes from the other

the same as in vacuo.

The

lation of water to air appears

still

more

remains un-

difficult

much

question of the re-

by

at first the air

this fact to

seemed

retained by the attraction of the water

to
in

be
be
the

MUTUAL ACTION OF FLUIDS.

200

second case, the water seemed indifferent

the third,

yet in

all

it

appears as

three,

it is

repulsive to the air

From

these

facts,

namely,

One

it,

but we must not

that the clay of a porous earthen retort,

whatever

which

air.

Dr. Priestley once imagined,

decide hastily.

*'

and

water

and that water repels

of these must be true

red hot,

there

air

that water attracts air, that

does not attract

for maintaining three opinions

on the subject of the mutual action of

water

in

the same air that has to act

on the same water.

seems reason then

if

when

destroys for a time the aerial form of

air

aerial

is

exposed to the outside of

form

it

recovers, after

it

it

has been

transmitted in combination from one part of the

clay to another,

the

retort."

till it

But

has reached the inside of

he soon discarded so ex-

travagant an opinion.

From

the recent experiments of Dr. Henry,

with those of

my

own, there appears reason to

conclude, that a given volume of water absorbs

the following parts of
gases.

its

bulk of the several

MUTUAL ACTION OF FLUIDS.

Bulk of

201

MUTUAL ACTION OF FLUIDS.

202

degree, the gas absorbed

in the

same degree

is

condensed or rarefied

so that the proportions ab-

sorbed given above are absolute.

One remarkable
at

another in water
like as in a

which has been hinted

fact,

no one gas

that

is,

it

is

capable of retaining

escapes, not indeed instantly,

vacuum, but gradually,

like as car-

bonic acid escapes into the atmosphere from the

bottom of a cavity communicating with

It

it.

remains now to decide whether the relation

between water and the above mentioned gases

of a chemical or mechanical nature.
facts

just

stated,

it

appears

elasticity of carbonic acid

of the
It

first class is

not at

From

is

the

evident that the

and the other two gases

all

affected

by the water.

remains exactly of the same energy whether

the water

is

All the other

present or absent.

properties of those gases continue just the same

as far as I

know, whether they are alone or blended

with water

abide by the

we must

Law

mutual action

therefore, I conceive, if

just laid down,

we

pronounce the

between these gases and water

to

be mechanical.

very curious and instructive phenomenon

takes place

three gases

when
is

a portion of

any of the above

thrown up into an eudiometer

MUTUAL ACTION OF FLUIDS.

tube of

3^0

203

of an inch diameter over water

the

water ascends and absorbs the gas with considerable speed

if

a small portion of

suddenly thrown up,

common

air is

ascends to the other, and

it

commonly separated by a fine

That instant the two
for a time.
is

film
airs

of water

come

into

the above situation, the water suddenly ceases to

ascend

in the tube, but the film of

with great

water runs up

enlarging the space

speed,

below,

and proportionally diminishing that above,

This seems to shew that the film

finally bursts.
is

till it

a kind of sieve through which those gases can

common

easily pass, but not

In the other gases

it is

air.

very remarkable their

density within the water should be such as to require the distance of the particles to be just 2, 3,
or 4 times what

it is

without.

In olefiant gas, the

distance of the particles within

without, as

is

is

just twice that

inferred Trom the density being ^.

In oxygenous gas, &c., the distance

great,
is

and

in

is

3 times as

hydrogenous, &c., 4 times.

This

certainly curious, and deserves further investi-

gation

whether the
tion to

we have only to decide

general phenomena denote the rela-

but at present

be of a chemical or mechanical nature. In

no case w^hatever does

it

of any of these gases

affected

is

appear that the elasticity

;

if

water takes Vt

MUTUAL ACTION OF FLUIDS.

204
of

bulk of any gas,

its

exerts -^ of the

absorbed

gas so

the

elasticity, that the exterior

does, and of course

it

gas

escapes from the water

when the pressure is withdrawn from its surface,

or when a foreign one is induced, against which
As far as is known too,
it is not a proper match.
all

the other properties of the gases continue the

same

containing oxygenous gas

thus, if water

be admitted to nitrous gas, the union of the two

which the water takes up

gases

is

of

bulk of nitrous gas, as

its

if this

certain

after

circumstance had not occurred.

clear then that the relation

Dr.

is

aV

would have done,

it

It

seems

a mechanical one.*

Thomson and Mr. Murray have

bo-th

largely in defence of the notion that all gases are

written

combined

with water, that a real union by means of a chemical affinity

which water exercises

gases, takes place

in a greater or less

this

affinity is

slight kind, or of that kind

degree towards

all

supposed to be of the

which holds

all

gases in a state

of solution, one amongst another, without any distinction.

The
the

opposite doctrine was

first

stated in a paper of mine, on

absorption of gases by water.

series,

Vol.

Dr. Henry,

(Maneh. Memoirs, 7iew

Previously to the publication of that paper,

1.)

who was convinced from

that the connection of gases with water

nature, wTote

two essays

son's Journal, in
clearly, and,

which

think,

in the
tlie

his

own

experience,

was of a mechanical

8th and 9th Vol. of Nichol-

arguments for thai opinion are

unanswerably stated.

do not intend

MUTUAL ACTION OF FLUIDS.

'205

Carbonic acid gas then presses upon water

the first instance with

time

it

its

whole force

in

in a short

partly enters the water, and then the re-

action of the part entered, contributes to support

largely into a discussion

to enter

to be, that "

that

the

of the arguments these

Dr. Thomson's leading argument seems

gentlemen adopt.

water will absorb such a portion of each gas,

repulsion

between the

absorbed,

particles

He

balances the affinity of water for them."

just

then proceeds

to infer, that the affinity of carbonic acid for water

such

is

as nearly to balance the elasticity, that the affinity of oleliant

gas for water

and of

is

equal to half its elasticity, that of oxygen, ^,

azote, ^, &c.

Now

of water attract one

if a particle

of carbonic acid by a force analagous to that of repulsion,

must increase directly as the distance decreases

such particles must be in equilibrium at any distance

it

two

if so,

and

applied to the particle of gas propelling

if

any other force

it

towards the water, the two particles must unite or come

into

is

most intimate contact.

Hence,

should

infer,

from

Dr. Thomson's principle, that each particle of water would

take one of acid,

and consequently

lib.

of water would

Mr. Murray

combine with 2^1bs. of carbonic acid nearly.

mentions a great

make

many circumstances which he

against the mechanical hypothesis

of the acid and

largely

alkaline gases are

for instance,

known

by water, and undoubtedly by

conceives

some

to be absorbed

affinity

therefore

the less absorbable gases must be under the same influence,

only in an inferior degree, and that "it would be impossible

to point out the line of distinction

between those where the

^absorption might be conceived to be purely mechanicah

and

MUTUAL ACTION OF FLUIDS.

206

the incumbent atmosphere.

gas

the

Finally,

gets completely diffused through the water, so as

to be of the

same density within as without

the

gas within the water then presses on the containing vessel only, and reacts upon the incumbent

The water

gas.

then sustains no pressure either

from the gas within or without.

In olefiant gas

the surface of the water supports | of tha pres-

oxygenous, &c., ff and in hydrogenous,

sure, in
CVC,

4.

When

any gas

those where the

operate."

is

confined in a vessel over

exertion of affinity must

conceive nothing

the exact line of distinction

is
;

be allowed to

more easy than

whe^'ever water

diminish or destroy the elasticity of any gas,

agent

wherever

it

does neither of these,

Whoever undertakes

agent.

to

to point out
is

it is

it is

fou7id

to

a chemical

a mechanical

maintain the

chemical

theory of the absorption of gases by water, should in the

outset overturn the following

Henry

"

argument preferred by Dr.

The quantity of every

absorbed by water,

gas,

follows exactly the ratio of the pressure

rule in philosophizing, that effects of the
differing in

and since

it is

degree, are produced by the same cause,

perfectly safe to

same kind, though

it is

conclude, that every, even the minutest

portion of any gas, in a state of absorption

tained entirely by incumbent pressure.

by water,

There

is

is

re-

no occasion,

therefore, to call in the aid of the law of chemical affinity,

when

a mechanical law fully and satisfactorily explains the

appearances."

;;

MUTUAL ACTION OF FLUIDS.

water

in the

207

pneumatic trough, so as to commu-

medium

nicate with the atmosphere through the

of water, that gas must constantly be filtering

through the water into the atmosphere, whilst the

atmospheric

air is filtering

contrary way, to supply

through the water the

place in the vessel

its

becomes

so that in due time the air in the vessel

atmospheric, as various.chemists have experienced.

Water
tort

in this respect is like

it

an earthenware re-

admits the gases to go both ways at the

same time.
It

is

not easy to assign a reason

should be so permeable to

and not

to the other gases

and why there should

be those differences observable

The

densities

|,

^^^

^V)

a reference to a

wi^

equilibrium could take place

No
if

the gases should not

one of these forms,

Upon

the whole

earthenware,

is

meable

in

mechanical

the densities of

by

this

law

agree in some

all

do not see any reason.

it

appears that water, like

incapable of forming a perfect

barrier to any kind of air

earthenware

others.

mechanical origin, but none

the gases within were not regulated

why

the

in

^^ve most evidently

whatever to a chemical one.

but

why water

carbonic acid, &c.,

one respect

but

it

the last

to all the gases, but water

is

differs
is

from

alike per-

much more

ON THE CONSTITUTION OF SOLIDS.

208

permeable

to

some gases than

have not been

liquids

Other

to others.

sufficiently

examined

in

this respect.

The mutual
number

action of water, and the greater

of acid gases and alkaline gas partaking

most evidently of a chemical nature

will

be best

considered under the heads of the respective acids

and

alkalis.

SECTION

4.

ON THE

CONSTITUTION OF SOLIDS.

solid

body

is

one, the particles of which are

in a state of equilibrium betwixt

attraction

that

and repulsion, but

in

no change can be made

without considerable force.

of the particles

heat resists
resists

it.

it

The

is

if

two great powers,

such a manner,

in their distances

If an approximation

attempted by force, then the

a separation, then the attraction

notion of Boscovich of alternat-

ing planes of attraction and repulsion seems unnecessary, except that upon forcibly breaking the

cohesion of any body, the newly exposed surface

must receive such a modification

in its

atmosphere

ON THE CONSTITUTION OF SOLIDS.

of heat, as

may prevent the

209

future junction of the

parts, without great force.

The
solids,

between

essential distinction

perhaps consists in

liquids

and

heat changes

this, that

the figure of arrangement of the ultimate particles

of the former continually and gradually, whilst

they retain their liquid form

ter, it

is

whereas

in the lat-

probable, that change of temperature

does no more than change the

arrangement of the ultimate

size,

and not the

particles.

Notwithstanding the hardness of solid bodies,

or the difficulty

of

moving the

particles

one

amongst another, there are several that admit of

such motion without fracture, by the application
of proper force, especially

The

ductility

only to be mentioned.
ticles glide

if assisted

It

along each others surface, somewhat

without

being at

The

cohesion.

all

end of a mag-

weakened

of great practical importance.

by experiment,

It

is

an enquiry

has been found

of an inch in diameter,

broken by the annexed weights.

Dd

their

that wires of the several metals

beneath, being each

just

in

absolute force of cohesion, which

constitutes the strength of bodies,

were

heat.

should seem the par-

like a piece of polished iron at the

net,

by

and malleability of the metals need

ON THE CONSTITUTION OF SOLIDS.

210

Lead

29i

Tin

49i

Copper

299J
360

Brass

370
450
500

Silver

Iron

Gold

Pound

piece of good oak, an inch square and a

yard long,

will just

bear in the middle 330lbs.

But such a piece of wood should not in practice

be trusted, for any length of time, with above J
or I of that weight.
strong as oak, of the

One would be

Iron

is

about 10 times as

same dimensions.

apt to suppose that strength and

hardness ought to be found proportionate to each

other

but this

is

Glass

not the case.

than iron, yet the latter is

much

is

harder

the stronger of

the two.
Crystallization exhibits to us the effects of the

natural arrangement of the ultimate particles of

various

compound bodies

but

we

are scarcely

yet sufficiently acquainted with chemical synthesis

and analysis
process.

to

understand the rationale of this

The rhomboidal form may

the proper

position

of 4, 6,

8,

arise

from

or 9 globular

ON CHEMICAL SYNTHESIS.
particles, the

cubic form from 8 particles, the

triangular form from 3, 6, or

10

particles, the

hexahedral prism from 7 particles, &c.

in

211

due time, we may be enabled

Perhaps,

to ascertain the

number and order of elementary particles, constituting any given compound element, and from
that determine the figure which
crystallization,

mature

to

and vice versa

form any theory on

it

will prefer

but

it

on

seems pre-

this subject,

till

we

have discovered from other principles the number and order of the primary elements which

combine

to

form some of the compound elements

of most frequent occurrence

which we

shall

the

endeavour

to

CHAPTER

III.

method

for

point out in the

ensuing chapter.

ON CHEMICAL SYNTHESIS.
When

any body

exists

in the

elastic state, its

ultimate particles are separated from each other

to

state;

much

greater distance than in any other

each particle

occupies

the

centre

of a

ON CHEMICAL SYNTHESIS.

212

comparatively large sphere, and supports

nity by keeping

all

its

or otherwise are disposed to encroach upon

a respectful distance.

ceive the

number

dig-

the rest, which by their gravity,

When we

it,

at

attempt to con-

of particles in an atmosphere,

it

somewhat like attempting to conceive the

number of stars in the universe we are confounded with the thought. But if we limit the
subject, by taking a given volume of any gas, we
seem persuaded that, let the divisions be ever
is

so minute, the

number of particles must be

finite;

just as in a given space of the universe, the

ber of stars and planets cannot be

num-

infinite.

Chemical analysis and synthesis go no farther

than

to

the

separation of particles

another, and to their reunion.

or destruction of matter

We

chemical agency.
introduce a
GO annihilate

new

is

one from

No new

creation

within the reach of

might as well attempt

to

planet into the solar system, or

one already in existence, as to create

or destroy a particle of hydrogen.

we can produce,

All the changes

consist in separating particles

that are in a state of cohesion or combination,

and joiningthose thatwere previously at a distance.

In
i)een

all

chemical investigations,

it

has justly

considered an important object to ascertain

:he relative iveights of the simples which consti-

213

ON CHEMICAL SYNTHESIS.
tute a

compound.

But unfortunately the enquiry

has terminated here

whereas, from the relative

weights in the mass, the relative weights of the

ultimate particles or atoms of the bodies might

have been inferred, from which their number

and weight

compounds would

other

in various

appear, in order to assist and to guide future

and to correct

investigations,
it is

one great object of

importance

Now

work, to show the

and advantage of ascertaining the

relative weights

simple

their results.

this

of the ultimate

and compound

particles^ both

bodies, the

of
number of sim-

which constitute one

ple elementary particles

compound particle, and the number of less compound particles which enter into the formation
of one more compound particle.
If there are

two bodies,

and B, which arc

disposed to combine, the following

in

2
1

the order

which the combinations may take place, begin-

ning with the most simple,

1

is

A
atom of A
atoms of A
atom of A
atoms of A
atom of

+ atom of B =
+ 2 atoms of B =
1

atom of

-{-

3 atoms of

-f-

atom of

B
B
B

=
=
==

viz.

atom of C, binary.

atom of D, ternary.

atom of E, ternary.

atom of

atom of G, quaternary.

F, quaternary,

&c. &c.

The

following general rules

may be adopted

ON CHEMICAL SYNTHESIS.

214

guides in

as

our

all

investigations

respecting

chemical synthesis.
1st.

When

only one combination of two bodies

can be obtained,
binary one,

must be presumed

to

be a

unless some cause appear to

the

it

contrary.

When

2nd.

two combinations are observed,

they must be presumed to be a hinary and a

ternary,

When

3rd.

three combinations are obtained,

we may expect one

to be a binary^

and the other

two ternary.
4th.

we

When

four combinations

observed,

are

should expect one binary^ two ternary^ and

one quaternary^ &c.

5th.

compound should always be

heavier than the mere mixture of its

binary

specifically

two ingredients.
6th.

ternary compound should be specifi-

cally heavier than the mixture of a binary

which would,

simple,
it

combined,

if

and a

constitute

&c.

apply,

The above
when two

D and

E, &c. are combined.

7th.

From
chemical

rules and observations equally

the application
facts

such

bodies,

already

as

of these rules,
well

and D,

to

ascertained,

the

we

ON CHEMICAL SYNTHESIS.
deduce the following conclusions
water

215

1st.

That

a binary compound of hydrogen and

and the relative weights of the two
elementary atoms are as 1:7, nearly
2nd.
is

oxygen,

That ammonia is a binary compound of hydrogen

and azote, and the relative weights of the two
atoms are as 1 5, nearly ; 3rd. That nitrous
:

gas
.

is

compound of azote and oxygen,

a binary

the atoms of which weigh 5 and 7 respectively

that nitric acid

is

a binary or ternary

compound

according as it is derived, and consists of one

atom of azote and two of oxygen, together
weighing 19; that nitrous oxide is a compound
similar to nitric acid,

and consists of one atom of

oxygen and two of

azote,

nitrous acid

is

a binary

and nitrous gas, weighing 31

is

a binary

; 4th.

compound,

charcoal,

nearly 12

that

nitric acid

that oxynitric acid

compound of

weighing 26
binary

weighing 17

compound of

nitric acid and oxygen,

That carbonic oxide is a

consisting

of

one atom of

and one of oxygen, together weighing

that carbonic acid

is a ternary compound, (but sometimes binary) consisting of one

;

atom of charcoal, and two of oxygen, weighing

19

&c. &c.

In

all these cases the weights are

expressed in atoms of hydrogen, each of which is
;

denoted by unity.

ON CHEMICAL SYNTHESIS.

216

In the sequel, the facts and experiments from

which these

conclusions

detailed

well as a great variety of others

as

from which are

derived,

are

inferred the

will

constitution

be

and

weight of the ultimate particles of the principal

the alkalis,

acids,

metallic oxides

neutral

salts,

the

earths,

the metals, the

and sulphurets, the long

and

train of

in short, all the chemical

com-

pounds which have hitherto obtained a tolerably

good

analysis.

Several of the conclusions will

be supported by original experiments.

From

the novelty as well as importance of the

deemed
expedient to give plates, exhibiting the mode of
combination in some of the more simple cases.

ideas suggested in this chapter,

it

this first part.

specimen of these accompanies

The elements

is

or atoms of such bodies as are

conceived at present to be simple, are denoted

by a small

circle,

with some distinctive mark

and the combinations consist

of two or

more of these

when

particles of elastic fluids are

in one,

it is

to

in the juxta-position

three or

more

combined together

be supposed that the particles of

the same kind repel

each other, and therefore

take their stations accordingly.

END OF PART THE

FIRST.

217

EXPLANATION OF THE PLATES.

PLATE L Fig. is intended to illustrate the

author's
1,
the subject of the capacities of bodies for heatThere are three cylindrical vessels placed one
See page 3.
within another, having no communication but over their
margins the innermost is connected with a lateral and

ideas on

tube graduated, and supposed to represent the

degrees of a thermometer, the scale of which commences at
if a liquid (supposed to represent heat) be
absolute cold
poured into the tube, it will flow into the inner vessel,
through an aperture at the bottom, and rise to the same
Equal increments of heat
level in the vessel and the tube.
in this case are supposed to produce equal increments of
temperature. When the temperature has arrived at a certain
point (suppose 6000) the body may be supposed to change
parallel

solid form to the liquid, as from ice to water, in which

case its capacity for heat is increased, and is to be repreconsiderable portion of
sented by the second vessel.
liquid must then be poured into the tube before any rise
will be perceived, because it flows over the margin of the
innermost vessel into the lateral cavity of the second ; at
length it reaches the level, and then a proportional rise will
its

till the body becomes converted into an elastic fluid,

whilst a
the thermometer again becomes stationar^^
great portion of heat is entering into the body, now assuming a new capacity.
Fig. 2, is a comparative view of the old and new divisions
See Table,
of the scale of the mercurial thermometer.
page 14. The interval from freezing to boiling water is
180 on both scales, and the extremes are numbered 32"
and 212 respectively. There are no other points of temperature in which the two scales can agree.
Fig. 3, is a view of the divisions of a water thermometer,
conformably to the new scale of the mercurial ; the lowest
point is at 45; the intervals from 45 upwards, to 55, 65,
75, &c. are as the numbers 1, 4, 9, &c.
Also, 30" and 60^

ensue,

when

coincide, as do 20

PLATE

TL

and

70, &c.

represents an air thermometer, or

the expansion of air by heat ; the numbers are Fahrenheit's
and the intervals are such as represented in the 7th column
of the table, at page 14.

2E

Fig.

1,

218

EXPLANATION OF THE PLATES.

Fig. 2, is the logarithmic curve, the ordinates of which

are erected at equal intervals, and diminish progressively
by the ratio i. The intervals of the absciss or base of the
curve, represent equal intervals of temperature (25 for
or aqueous vapour, and 34^ for ethereal vapour)
the ordinates represent inches of mercury, the weight of
which is equal to the force of steam at the temperature.
See the 8th and 9th columns of table, at page 14. Thus
the force of steam at 212 and of ethereal vapour at 110,
new scale, is equal to thirty inches of mercury ; at 187 the
force of steam is half as much, or 15 inches, and at 76,

steam

that of ethereal vapour

is

also 15 inches, &c.

Fig. 3, is a device suggested by Mr. Ewart, to illustrate

the idea which I have developed in the section on the temIt is a cylindrical vessel close
perature of the atmosphere.
at one end and open at the other, having a moveable piston
the vessel is supposed to contain air, and
sliding within it
a weight is connected with the piston as a counterpoise to
it.
There is also a thermometer supposed to pass through
:

Now if
the side of the vessel, and to be cemented into it.
we may suppose the piston to move without friction, and
the vessel to be taken up into the atmosphere, the piston
will gradually ascend, and suffer the air within to dilate, so
as to correspond every where with the exterior air in density.
This dilitation tends to diminish the temperature of the air
within (provided no heat is acquired from the vessel). Such
an instrument would show what the theory requires, namely,
that the temperature of the air w ithin would every where in
the same vertical column agree with that without, though
the former would not receive or part with any heat absolutely, or in any manner communicate with the external air.
III.
See page 135. The balls in Fig. 1 and 2,
represent particles of water : in the former, the square form
denotes the arrangement in water, the rhomboidal form in
the latter, denotes the arrangement in ice.
The angle is
always 60 or 120.

PLATE

Fig. 3, represents the perpendicular section of a ball

resting upon two others as 4 and 8, Fig. 1.
Fig. 4, represents the perpendicular section of a ball
resting upon two balls, as 7 and 5, Fig. 2.
The perpendiculars of the triangles shew the heights of the strata in the

two arrangements.

EXPLANATION OF THE PLATES.

219

Fig. 5, represents one of the small spiculae of ice formed

upon the sudden congelation of water cooled below the
freezing point.
See page 134.

Fig. 6, represents the shoots or ramifications of ice at the

congelation. The angles are 60 and 120.
IV. This plate contains the arbitrary marks or
signs chosen to represent the several chemical elements or

commencement of

PLATE

ultimate particles.
Fig.

Fig.

Hydrog. its rel. weight 1

5
2 Azote
3 Carbone or charcoal... 5
1

11

Strontites

Barytes

4 Oxygen
5 Phosphorus
6 Sulphur
7 Magnesia

12
13
14
15
16
17
18
19

8 Lime
9 Soda
10 Potash
21

22

An

9
13

20
23
28
42

Iron

Zinc

Copper

Lead
Silver

Platina

Gold
20 Mercury

46
68
38
56
56
95
100
100
140
167

EXPLANATION OF THE PLATES.

220
33
34

An atom
An atom

35
36

An atom of acetous acid, 2 carbone + 2 water

An atom of nitrate of ammonia, 1 nitric acid +

37

An atom

of alcohol, 3 carbone, -f 1 hydrogen

of nitrous acid, 1 nitric acid + 1 nitrous

16
31

gas

ammonia

26
1

water

of sugar,

alcohol

-\-

carbonic acid

...

33
35

Enough has been given to shew the method ; it will be

quite unnecessary to devise characters and combinations of
them to exhibit to view in this way all the subjects that
come under investigation ; nor is it necessary to insist upon
the accuracy of all these compounds, both in number and
weight ; the principle will be entered into more particularly
It is not
hereafter, as far as respects the individual results.
be understood that all those articles marked as simple
substances, are necessarily such by the theory ; they are
Soda and Potash, such
only necessarily of such weights.
as they are found in combination with acids, are 28 and 42
respectively in weight ; but according to Mr. Davy's very
important discoveries, they are metallic oxides ; the former
then must be considered as composed of an atom of metal,
21, and one of oxygen, 7 ; and the latter, of an atom of
Or, soda contains 75
metal, 35, and one of oxygen, 7.
to

potash, 83.3 metal and

per cent, metal and 25 oxygen
It is particularly remarkable, that according
16.7 oxygen.
to the above-mentioned gentleman's Essay on the Decomposition and Composition of the fixed alkalies, in the Philosophical Transactions (a copy of which Essay he has just
favoured me with) it appears that " the largest quantity of
oxygen indicated by these experiments was, for potash 17,
and for soda, 26 parts in 100, and the smallest 13 and 19"
;

DIRECTIONS TO THE BINDER.

Plate

to face page 217.

2 to face page 218.

3 to follow plate

2.

4 to face page 219.

PRINTED BY SIMPSON AND GILLETT, MARKET

ST.,

MANCHESTER.

r2L0

oKr
Scale

A'^apoTxr

SoaEe

-7000
-200

ASateL-

190
100 --

500

I80

400

3oo

300

-B

Zee

'F

20
30

-40

Steam
t/,, Ya/.

ria.te ^
.

O0O0
10

VI

il

14

\A

15

'.fj

so
iH

17

-'O

I^/

JBifffiry

23

2^

21

O 0 O O
26

28

21

25

24.

I
29

o#o
30

O^

33

32

31

CTO

D
35

34-

O^DO
Se/Jtrjiffrjr

36

37

%
f\

iM