Baynes (Lessons of Thermodynamics) (1878)

EDWARD
LESSONS OF THERMODYNAMICS
BAYNES, ROBERT
OLD CLASS
PhyTB
c.l
PASC
THERMODYNAMICS
BAYNES
llontron
HENEY FROWDE
OXFORD UNIVERSITY PRESS WAREHOUSE

7
PATERNOSTER ROW
dMamitom
|jress
Strics
LESSONS
ON
THERMODYNAMICS
BY
ROBERT
E.
BAYNES, M.A.
SENIOR STUDENT OF CHRIST CHURCH, OXFORD
AND
LEE'S
READER IN PHYSICS
AT THE CLARENDON PRESS

M DCCC LXXVIII
[All rights reserved]
PREFACE.
THE
substance of the following Lessons formed a
Lectures on Thermodynamics which were
course of
delivered
the
at
Clarendon
Laboratory,
Oxford,
in
Hilary Term 1876. They have been revised, and are

now published in the hope of supplying a complete
English Mathematical Introduction to the Dynamical
Theory of Heat.
No
such treatise exists at present
Rankine's Manual of the Steam- Engine and

Prime Movers, being written for the use chiefly
Professor
other
of practical Engineers, contains only a short sketch of
the subject;
Professor
T ait's
Thermodynamics gives
very valuable
information
theory and
place in the general
but
its
as
too brief for the wants
is
Professor
Clerk Maxwell
in
to
the
history of the
Science of Energy,
of the
his
student;
Theory
and
of Heat
has not gone far into the Mathematical Development,
though
in
other respects the work
is
most valuable
and perhaps the most suggestive that has been written

on the subject. To this last book the present writer
owes
takes
his
this
obligations
first
knowledge
opportunity
to
it.
of
of
Thermodynamics, and
acknowledging
his
great
PREFACE.
vi
Many
memoirs have been studied
original
the
in
course of the preparation of these Lessons, and copious

reference
able
is
made
them
the student will thus be
more extended
find
to
to
investigation
der
Thtorie
mechanischen
de
Mecanique
Robert's Principes de
been
No
Wdrmetheorie,
la
many
Dr. Zeuner's Grund-
particular points of the theory.
zuge
of
Chaleur,
M. Verdet's
and
M. Saint-
Thermodynamique have
also
consulted.
account
because Mr,
is
given of the kinetic theory of gases,
Watson's
of Gases contains
will also find
Professor
all
treatise
that
is
The Kinetic Theory
required.
The
student
an elementary and lucid explanation
in
Clerk Maxwell's Theory of Heat^ Chapter
XXII.
CHRIST CHURCH,
May, 1878.
CONTENTS.
CHAPTER
I.
Introductory Mechanical Principles.
........
...........
...........
...........
...........
......
.....
..........
...........
PAGE
i.
Fundamental Units
2.
Force
3.
Work
4.
Energy
5. Stress
6. Strain
7.
Elasticity
8.
Mechanical work done by or upon a body during a
9.
Virial
CHAPTER
Introductory
io.
Measure of Heat
II.
Temperature
Choice of variables
14.
Thermal
9
9
9
II
II.
Thermal Principle
les.
14
14
13.
1 6.
.........
12. Characteristic equation of a
15.
strain
body
17
20
20
capacities
General equations of the effects produced by heat

Relations between the thermal capacities
.
.21
.22
17. Volume-elasticities
24
CHAPTER
III.
Conservation of Energy.
FIRST
18.
LAW OF THERMODYNAMICS.
Heat a form of energy
19. First
Law
of Thermodynamics
.....
......
.
25
27
CONTENTS.
viii
DR.
JOULE'S EXPERIMENTS.
PAOB
20. I.
23.
By magneto-electricity
II. By the friction of fluids
III. By the compression of gases
IV. By the thermal effects of an electric
24.
Mechanical equivalent of heat
25.
M.
21.
22.
.'
27
31
34
29
current
.
33
Him 's experiment
34
CONSTITUTION OF BODIES.
26. Intrinsic energy
36
27. Constitution of bodies
37
28. Sensible
heat
39
work
External work
29. Internal
39
30.
40
GENERAL EQUATION OF EQUIVALENCE.

31. General equation
32.
Thermal
40
lines
42
33. Adiabatics
CHAPTER
45
IV.
Transformation of Heat.
CARNOT'S
34.
Heat engines
35.
Carnot's
PRINCIPLE.
49
cycle
36. Closed cycles
49
53
37. Reversible engines
54
38.
Carnot's
39-
Work in reversible
54
principle
engines of finite range performing
Carnot's
57
cycle
ABSOLUTE SCALES OF TEMPERATURE.

40. First absolute scale of temperature
58
Second absolute scale of temperature

42. Comparison of the two absolute thermometric scales
43. Zero thermal lines
41.
44. Intersection of thermal lines
59
62
63
64
SECOND LAW OF THERMODYNAMICS.

45. Generalization of Carnot's principle
46. Equivalence of transformations
69
72
CONTENTS.
ix
PAGE
47. Reversible cycles not closed
48. Mathematical expression of C a mot's principle
49. Dr.
50.
...
Zeuner's method
Connection between the two laws of thermodynamics
7^
77
77
80
NON - REVERSIBLE PROCESSES.

83
51. Non-reversible cycles
CHAPTER
V.
Fundamental Equations.
52. Application of the First
53. Application of the
54.
Law
Law
85
Second
85
86
Values of the Specific Heats
55. Difference of the Elasticities
87
88
56. General
Equation
57. Determination of the Intrinsic Energy
88
58. Professor
Rankine's method
89
59. Professor
Clausius' method
91
60.
94
Entropy
6 1. General determination of intrinsic energy and entropy

62. Relations between partial differential coefficients
...
63. Application to perfect gases
95
97
98
64. Sensible heat
103
106
65. Real specific heat
CHAPTER
VI.
Degradation of Energy.
.no
66. Available energy

67. Dissipation of energy
115
68. Possibility of restoring lost availability
117
CHAPTER
VII.
Comparison of Theory with Experiment.

APPLICATION TO GASES.
69. Dr. Joule's experiments
7. M. Hirn's experiment
120
122
CONTENTS.
PAGE
72.
W. Thomson's method
Dr. Joule and Sir W. Thomson's
73.
M. Cazin's experiments
123
71. Sir
74. Specific
A.
B.
experiments
127
132
heat of gases at constant volume
By adiabatic expansion or compression

By the velocity of sound
34
134
139
APPLICATION TO SOLIDS AND LIQUIDS.

75.
General formulae
76. Application to
143
water
77. Application to other substances

78.
79.
80.
Adiabatic compression
149
Thermal
Thermal
150
effects
accompanying traction
of drawing out a film of liquid
effect
CHAPTER
Change of
81.
State.
158
Apparent discontinuity
159
Law
Second Law

155
VIII.
82. General equation
85. Determination of the
159
160
161
Entropy
86. Determination of the Intrinsic
161
Energy
87. Latent heat

88.
144
147
162
Expansion along an Adiabatic
162
APPLICATION TO THE TEMPERATURES OF FUSION.

89. Fusion of ice
90. Other experiments
163
1
70
APPLICATION TO THE TEMPERATURES OF VAPORIZATION.

91.
Maximum
tension
171
92. Latent heat: total heat

.
93. Table for water
94. Critical temperature
172
'
95. Specific heat of saturated
174
1
vapour
temperature of inversion
96. Adiabatic expansion

97. Density of saturated
vapour
76
176
178
180
APPLICATION TO THE TEMPERATURES OF SUBLIMATION.

98. Sublimation
.185
CONTENTS.
xi
THERMAL LINES OF THE SATURATED CONDITIONS.

PAGE
99.
Thermal
185
lines
i7
100. Triple point
CHAPTER
On
Flow of
the
IX.
Fluids.
101. General formula
190
102. Incompressible fluids
191
103. Compressible fluids
191
104. Saturated fluids
192
CHAPTER
X.
Heat- Engines.
105.
M. C a mot's
engine
Engines of maximum efficiency

107. Dr. Stirling's engine
1 06.
Ericsson's engine
Dr. Joule's engine
108. Captain
109.
no. Steam-engine
196
197
198
199
201
202
EXPLANATION OF REFERENCES.
Pogg. Ann.
C. R.
Conaptes rendus de
B. A. Rep.
British Association Report.
Phil. Trans.
Trans. R. S. E.
Phil.
Mag.
Proc. R. S.
=
=
=
=
Poggendorff's Annalen.
1'
Academic des
Sciences.
Philosophical Transactions of the Royal Society.
Transactions of the Royal Society of Edinburgh.

Philosophical Magazine.
Proceedings of the Royal Society.
The number in the square brackets [ ] is the series, that in Roman

characters the volume, that in ordinary brackets ( ) the year, and the final
number is the page.
ERRATA.
p. 23, line 15, for
77,
41-675
read 41-55.
20,
or suggestive a letter
with our notation.
14,
Idv
I8v.
In Fig. 19, page 67, a dotted curve should be drawn through

inclined to Ov than either of the other curves.
less
LESSONS ON THERMODYNAMICS.
ERRATA AND CORRIGENDA.
Page
8,
line
force read action
24.
Add
Add
X 3-
^r on the
the envelope being usually of glass
per square centimetre

For
conductibility read conductivity
29.
33- Delete per unit area
17,
J 8,
3>
For
For upon read per
16.
15,
2.
4.
9>
axis .... minimum raw? by a line on the /z/-plane

that does not pass through the origin, since the volume
cannot be reduced to nothing
23,
15.
For 41-675 read 41-55
20.
for or
2 5>
J 5-
For has
17.
Delete in nitrogen or
12.
In the experiment described the ends of the wire were connected inside the tube, the commutating arrangement being
28,
suggestive a letter read with our notation

.... as raz</ is
no heavier when hot than
used only in proving that the heat generated
is
propor-
tional to the square of the current-intensity

34>
33-
After plate insert bent at the bottom so as to form a narrow
36-
Before between add on the ledge
6.
Remove
ledge
35,
fig.
the rest and bend round the iron plate attached to
so as to support
,
line 34.
i.
40,
After cooling add per degree of this excess

For a state read all states
36-
After conduction add
42,
25.
Delete per unit area
43>
J 9-
Fr
22.
44,
4.
>
friction
similarly read therefore
Add These lines are also called isopleres

Add we shall however retain both terms,
since the former
not only denotes the latter but includes also the further
notion of entropy being a measurable quantity
))
31
Add On
a diagram with volume and entropy as coordinates

no thermal lines of the same type intersect each other
For shews read
points out
Page 45,
line 18.
50,
8.
After large add in comparison with the volume of the work-
63,
3.
For
8.
Delete
ing substance
fi
21.
the exponent
we
T read
T'
again on line 18
Delete they .... cold on line 23
67,
i.
Before lines insert real thermal
4.
For
fig. 19.
also read necessarily
Draw a
dotted curve through
less
inclined to
Ov
than
either of the other curves

68, line 34.
For
cycles read processes
24.
For
cycle read process
77>
1 4'
For Idv read I8v
76,
For
?8>
83,
,,
For OB' CO' read O'B'CO'
22.
~ and |f
ii.
For 5
V
3i.
For 25 read 24
85,
I4-)
86,
12.)
85,
15.)
-,
13-
For
87,
7.
For
For
^T_^L
d/
d^
di)
/^^\
dl|)
read
/^^\
(^
these equations raz^ this equation
After expression add (in terms of v and

In this case dS = (ft
90,
13.
92,
15.
Add
2 4-
^^
<
28
read 33
2
(5)' Multiply the second determinant by /
The expression in the 2nd row i9th column should be
98, table.
97> e 1'
that in 4th
row 4th column,
-r
dv
7f
that in 5th
row 5th column,
~-
7kp ov

that in 5th
row
that in 5th
row i8th column,
that in 5th
row 2oth column.
1 3th
column,
-f?;
kp ot
^
K Of
row 4th column,
that in 6th
row
that in 6th
100, line
2th column,
row I4th column,
that in 6th
row i6th column, -
For
ft
(p
V
ot'
kp
<(>-<$)
read
ft'
8.
Add
and ft'= ft-^f,
101,
2.
^r
25 read 24
102,
23.
k-b
that in 6th
2.
dp
~ (K^+
ot'
oV p -^)
kp
kp
Page
/^r megadynes
t'
being equal to
raz</ atmospheres,
or
E
else
add per square
centimetre
i.
104,
2.
For the temperature read temperature

After body add i.e. is a function of this particular temperature
,,
105,
10.
For
1 06,
15.
After
107,
24, 29.
conductibility read conductivity
T add
i.e.
a function of
For atoms read molecules
28.
For na\ and
29.
For a read
Fr
aft
read w/zft and
jtft
/*
12.
atomic weight read molecular mass

For weight read mass a
After unity add so that its molecular mass p
For aft=ioo read /*ft=2OO
29.
For 100 read
31
For weight read mass, and add and whose molecule
108,
10.
.
sists
no,
3.
J 7-
of
its
After
Add
Sir
200,
and for 2.n read 2.
is
ii.
For
insert extreme
W. Thomson
-- log, y
i
IJ 3>
8.
con-
atoms
calls
it
the thermodynamic motivity
of the body
112,
two
read
log f
^1
e
(Q
After energy insert or thermodynamic motivity
4
Page 115,
line 23.
For
work
the
.... to bottom of page substitute
while the
maximum work obtainable by cyclical processes from heatenergy / dH taken in by a body along a given path is, by
CdH
/*// 7-T
(i),
less
fdH
than
by
where 4
is
JT'
maximum work
the lowest
available temperature, the

that can be done
in a cycle partly formed of this path as a whole is less than
are the entropy of the body
<o), where <p,
by t (<
fdH
<
and
in the final
initial states defined
by
this path, so that
an amount
of available energy is lost owing to the nature of the path,

or this amount of energy is dissipated along the path. This
51, is always positive when the proexpression, by (20),
non -reversible, being zero
cess is
If the process
121,
17.
For
122,
10.
For the
>
,,Z Q>?>'1 '
129,
6.
140,
12.
read
is cyclical,
for 0-02 read
0-2
For volume read temperature
For
c. G.
s.
read our
After lose insert on each side

ii. For fraction read function
I 4'
6.
Delete
and substitute degree
32.
See however Nature,
169,
20.
1 86,
last.
For log read loge

For <f>+ read
187,
28.
191,
13.
For 0-0074 read

For liquids read
3 2>
Fr A
99>
Before and insert so that the equation pv = RT cannot be

taken to hold for air, as M. Regnault has otherwise shewn
148, line 2 1.
165,
tQ
TQ
For u l} uz read vl} v2
J 63,
is
read in the water
air
141, eq. (17).
156,
for reversible processes.
the dissipation
xxiii.
(1880) 211 (Rep. Chem. Soc.).
<f>
o-
0734
fluids
.... figure substitute
Two
receivers are required for
one coupled directly without any valves to each

end of the cylinder, and the plunger of one is always half a
this engine,
stroke before that of the other

203,
9.
Add
This engine is not perfect because condition (i) of

06 is not satisfied ; for maximum efficiency, either
must be an adiabatic as well as CD, or CD must be isoand not an adiabatic
diabatic with
AB
AB
204,
17.
For lowers and
increases read lower
and
increase
CHAPTER
I.
1.
Fundamental Units.
units
employ
quantities that
In Mechanics
it is necessary to
measurable
of
the
different
expression
all these may howunder consideration
for the
come
ever be derived from the three simple units of Length, Mass,

shall in this chapter briefly describe such of
and Time.
We
these mechanical quantities as concern us at present, employing

the Centimetre-Gram-Second or
c. G. s.
system of
exhaustive treatment of which the student

fessor
Everett's
Illustrations
is
units, for
an
referred to Pro-
of the Centimetre-Gramme-Second
System of Units (1875).

The unit of length is one centimetre it is the
Length.
hundredth part of the length of a certain platinum bar at Paris,
:
when
called the Metre,

1
Captain Clarke
temperature of melting ice.

0-3937043196 inch, so that
the
at
finds that
is
i c.
the British foot contains 30-47972654 c.

Mass. The unit of mass is one gram
this is
the thousandth
part of the mass of a certain piece of platinum, called the

Kilogramme des Archives, which is kept in Paris. Professor
Miller 2 has determined

a British
pound
is
to
i g.
be 15-43234874 grains, so that

He also concludes 3 from
g.
453-59265254
Professor Kupffer's comparison of standards that the mass of

i c.c. of water of maximum
density (i.e. at 4c.) is 1-000013 g.
and not
Time.
i g.
exactly.
The
unit of time
recent discussions
is
one
and the most
give
365-24222x24x60x60
1
solar second,
Proc. R. S. xv. (1866) 313.

Phil. Trans, cxlvi. (1856) 890.
or
3
*
31556928 seconds
Proc. R. S. xiv. (1865) 293, note.

Main's Astronomy (1863), 99.
length of the tropical year, that is, the time which

elapses between two successive transits of the sun through the
as the
vernal equinox.
It
be feared that
to
is
so slowly however
absolutely constant;
many hundred
observation through
this
is
it
interval is
not
lengthened, that
years can scarcely detect
the change.
The
Velocity.
Acceleration.
unit of velocity
The
second per second.

Force
2. Force.
is
and the unit of force

for unit time
is
it
that
is
produces in
called a dyne
is
one centimetre per second.

is one centimetre
per
unit of acceleration
which produces motion in a body,

which acting on unit mass
that force
it
In the
unit velocity.
c. G. s.
a million dynes, which for us

shall term a megadyne.
is
system
a more
convenient unit, we
Forces are usually measured by weights, that is, by the earth's
attraction upon masses ; they are then comparable only at the
same
attraction varies with the position
as the
spot,
earth's surface.
This
is
called the gravitation
on the
measure of force
reduced to absolute measure on multiplication with g,

which denotes the number of units of acceleration impressed
is
it
on a falling body at this spot by gravitation; for the weight

of a mass acting on the mass for one second generates in it,
if allowed freely to fall, a velocity of g centimetres per second,
The value of
or the weight of a gram at this place is g dynes.
should be determined by pendulum or other experiments at
1
each different spot it is well given however by the formula
g
where A
is
2
978-048 (1+0-00514491 sin A)(i
the latitude, h the height above sea-level, r the earth's

2
636669500 c.). The form of this formula

and
its constants are deduced from many
given by theory,
combined observations it may be written
mean
radius (about
is
g
which
is
= 980-5642-516 cos 2 A
more convenient
for
0-000002
calculation.
Mem. R.A.S.
Clark's Earth (Figure of the}, in the Ency. Brit.
vii.
h,
Thus
at
Oxford
(1833) 94, and Pratt's Figure of the Earth (1871), 51.

vii.

(lat.
at
51 45' 36' N.) the weight of a gram
Manchester
Much
(lat.
53 29'
it is
N.)
is
981-15 dynes, and
981-30 dynes.
is introduced into ordinary language by the

which such terms as gram and pound are used,
not only as masses or quantities of matter (which is their real
loose
confusion
way
in
but also as the weights of these masses, that is,

the force of a pound means
signification),
as forces.
This must be avoided
all, and should never be used, while the

absolutely nothing
a
force of
pound's weight or the force equal to the weight of a
and these latter forms of exis
pound
perfectly intelligible
at
though the longer, must always be employed.

This error has arisen from the conception of mass being
pression, even
more
difficult
portional
to
than the conception of weight, which

it,
and from the
compared by the balance,
in
is
pro-
masses are usually

which the weight of one mass is
fact
that
The weights of a
balance are of course not real weights or forces, but masses
whose weight is put against that of the bodies weighed.
made
3.
when
to equilibrate the weight of another.
Work.
The
unit force
c. G. s.
system
is
unit of work is that work which is done

overcome through unit distance this on the
We shall however find that
called an erg.
is
thermodynamics a million ergs is a more convenient unit,

and this we shall call a megalerg. Work may be done in comin
municating motion to a body as well as in overcoming force

work will be exhibited in the next article.
this aspect of
The
practical unit of
measure
work used by engineers
in British units
it
is
against the earth's attraction in raising one
height of one foot, and

units
it
is
work done
pound through
the
equal to 13825.38 g- ergs ; in metrical

a kilogrammetre, or the work required to raise one
kilogrammetre 98-115 megalergs

13-567 megalergs.
Energy. This
4.
a gravitation
is
kilogram through one metre, and

Thus at Oxford a foot-pound
is
is
a foot-pound, or the
is
the
equal to 100000^- ergs.
is
is
:
at
13-565 megalergs, and a
Manchester a foot-pound
power of doing work which a body
or system of bodies possesses by reason of its state

it is of
two kinds, kinetic and potential the former due to the motions
:
B 2
4
of
various parts, the latter to the mutual actions between
its
them.
of the system at time / is
Suppose that the small mass
the axial-components of
x,y, z, moving with the velocity v
are
the effective forces
px, py, \LZ, and, if the components of
the whole impressed force acting on it are X, Y, Z, the principle
JJL
at
of virtual velocities gives
= o,
2.{(X-px)&x + (Y-w)ty + (Z-iJLz)tz}

whence on writing dx
dy,
we have
dz for bx, by, bz
^.^(xdx+ydy + zdz) = ^.(Xdx + Ydy + Zdz),

= 2.(Xdx + Ydy + Zdz}.
or
dZ.^fJi.v'*
Now 2.(X dx + Ydy + Zdz) represents the work done by
forces
upon
2 (X dx + Ydy + Zdz)
the system, or
is
done by the system against the forces
which
to
the
it is
(i)
the
work
subject,
and, for the equation to be integrable, these expressions must
be exact
differentials
hence,
if
2.(Xdx + Ydy + Zdz) =

we have
2
as
first
.1^-2
u being the
initial
lp
.f(Xdx+ Ydy + Zdz)
velocity of the mass
=x
-X,
(3)
//.
suppose that the different parts of the system are

connected together then no work is done by the mutual
:
actions of the parts,
and the
holds good, by
D'Alembert's
system is brought to
in which case X,
becomes
zero,
rest
Y,
equation
even
principle,
are negative
total
if
Hence
by doing work against
and the
work done
integral represents the
the system by external forces only
reactions are not included in X, Y, Z.
part
(2)
Now
(i)
rigidly
upon
d\,
integral
(3), indeed,
the internal
if
the whole
resisting forces
the value of v for each
work done,
viz.
2.f(Xdx+ Ydy + Zdz),

2
The latter term, which depends only on
equal to 2. |jui&
the initial velocities of the different parts of the system, thus
represents the total work that can be done by the system before
is
is brought to rest, i. e. the work it can do in virtue of its

motion
it is therefore appropriately termed the kinetic energy
it
of the system in
its
initial state.
Hence
the general equation
for a rigid system (i) the work done against resistequal to its loss of kinetic energy, (2) the work done in
altering the motion of its parts is equal to the increase of kinetic
shews that
ance
is
energy.
The most complicated motion

duced
of a rigid system can be reand a rotation of the
to a translation of its mass-centre
system about an axis through this centre, and, the mass of the
whole being m, the velocity of translation u, the radius of
gyration about the instantaneous axis k, and the angular velocity
we
o>,
easily find that
mu2
is at
any moment the kinetic energy of translation, and
2
J zrfoo the kinetic energy of rotation. If at any moment there
is no rotation, the velocity of all the parts is the same and
i
equal to u, and the total kinetic energy \mu
energy of a mass m moving with velocity v is \
As energy
or the kinetic
mv 2
measured by the amount of work it represents,

the unit of energy is also the erg on the absolute system
hence
is
also the kinetic energy in the last case, expressed in gravitation
mv 2
measure,
is
*g
Suppose now that the system is not rigid and is subject
only to the mutual actions of its parts, there being no external
forces
then the action between two particles
//, whose dis(ii)
jut,
tance apart
rt being ^'^(r),
and
such forces being

replaceable by their several axial components, the principle of
is
all
virtual velocities gives
V.(Xdx+Ydy + Zdz)
= 2 ./*//<
= dU
(r)rfr
for S.J/xz; 2 and denoting the

Hence, writing
say.
values by suffixes, we have by equation (3),
K-K.= U-U
maximum value of u and we take P = II

P represents the maximum amount of work that the system
If II
then
initial
is
the
/,
capable of doing on starting from the state corresponding to

or the work that the system can do by reason of its conthat is, by reason of the mutual actions which
figuration
depend
is
7,
on the positions of
its
parts
P is
therefore called the potential
energy of the system in that state.

(iii) In the above case we have further by substitution
K-K. = (n-p)-(n-p
K+P=K + P
or
),
thus in a system of bodies subject only to the mutual actions

of its parts the sum of the kinetic and potential energies that
is a constant
This theorem is
is, its total energy
quantity.
called the Principle of the Conservation of Energy.
(iv) In the general case the impressed forces will consist of
both internal and external actions, and, if the axial-components

of the external impressed forces acting on the part p are X',
Y', z', the equation (i)
becomes
.
say,
where
hence
if
(X'dx + Y'dy + Z'dz)
is
the total energy of the system
represents the
work done by
dW'= dK+ dP =
the external forces
dE,
or for any system (i) the work done against external force is
equal to its total loss of energy, (2) the work done in altering
its
state
We
is
equal to
its
gain of energy.
have also from (3)
^-ô =
or
Jr+x
Xo-X>
= Jr + Xo
the suffixes denoting the initial values,

x is called the ergal of
all the forces, and thus the sum of the kinetic
energy of a system
and the
ergal of the forces to which
is
it
subject
is
a constant
quantity.
If
dW is
we have
the
work done by the body against external
obviously
d^
(v)
It is
dynamics
moves
= dP+dW.
an interesting problem
forces
(4)
in connection with
thermo-
to determine the total kinetic energy of a
as a
whole and has also
at
oscillations with respect to its parts.
the coordinates of the
mean
body which
same time very rapid
At time / let x,y, z be
the
position of an oscillating particle
IJL
whose
this
x+f y+
real position is
moment
is
z+f
?/,
then
its
velocity at
given by
the velocity at this moment of the mean position of the

Hence the mean
particle, and v the velocity of oscillation.
if
is
of
value
kinetic
its
energy during
T,
the
time
periodic
of
oscillation, is
'
since during this extremely short interval
may
as the particle returns to
energy of the body
that
x, y, y,
z, z,
be considered constant, and
all
as the
is,
may
sum
its initial
The
position.
total kinetic
therefore be taken as
of the kinetic energy of
its
motion
and the mean
kinetic energy of the oscillations of
We
in
shall
shew
its
ultimate atoms,
the total kinetic energy of a hot
and
space
parts.
Chapter III that the heat of a body consists
in the oscillations of
in space
in
its
that of
its
whence
body
is
it
appears that
that of its
motion
heat-oscillations together.
If the oscillations follow the
pendulum-law so that
/'
CT
v*d/=
then
=<zcos27r(-+a),...,
rr
27T 2
or the
mean
kinetic energy of oscillation

/?2
itself
as heat
is
7r
{*"+.. .-a* cos 4ir(
^m
i
r2
.
that
+ a)-
...}<//
which exhibits
The
on any part of a body in equilibrium

it all round
by the matter in contact
with it, whether part of the same body or not, and is measured
by the number of units of force that act SfSft unit area that
5. Stress.
is
stress
the -feice exerted
upon
measure by the number of dynes per square

centimetre, and in gravitation measure by the number of grams'
weight per square centimetre. We shall notice only two kinds
is,
in absolute
we
shall meet with hereafter

they are both
same way.
Longitudinal Stress or Tension acts in one direction only,
of stress that
measured
in the
producing compression or extension.

Hydrostatic Stress or Pressure
and
is
the
same
acts normally to the surface of bodies
in
all
directions
equivalent to
it is
three equal longitudinal stresses acting at right angles to each

other.
It is this
stress to
which we
shall generally consider
a body subjected, unless otherwise specified.

The units of pressure employed by engineers are gravitation
measures in Britain they are the weight of a pound per square
:
foot or per square inch, which are equal to 479-05
and 68983
dynes per square centimetre respectively at Oxford in metrical

measure the unit is the weight of one kilogram per square
metre, which at Oxford is equivalent to 98-115 in absolute units.
;
An unit much used in scientific tables is an atmosphere,

defined as the pressure due to a barometric height of 76 c. of
here g being
mercury at oc. in latitude 45 at sea-level
;
980-564 and the density of mercury 13-596
at
oc., this presequivalent to the weight of 1033-3 grams per square

centimetre or to 1013226 in absolute measure.
sure
is
more
after use
scientific unit,
and one which we
shall
under the name of an atmosphere,
of one megadyne per square centimetre. It

a barometric height in latitude 45 of 75-008
of 74-9 6 3
We
mon
always herethe pressure
is
equivalent to
c.,
or at Oxford
c.
must once more
inaccuracy of comof pressures in pounds per square

quite improper, the accurate phrase being pounds'
call attention to the
The measurement
talk.
inch, &c.
is
is
weight per square inch.
This is a general term to denote any alteration

form or dimensions of a body. We shall consider two
cases only, which correspond to the above varieties of stress.
Longitudinal Strain occurs when there is deformation in one
6. Strain.
in the
direction only
measured by the
is
it
of the extension
ratio
or compression of any line in the body in the direction of the

strain to
its
original length.
when
Isotropic Strain occurs

similar to
original form, that
its
deformation
the
body
when
is,
it
after
alteration
is
has undergone no
the result of three equal longitudinal strains

at right angles to each other, and is measured by the ratio of
it
the change of
strain,
is
volume
to the original volume.
being the ratio of two similar magnitudes,
is
the resistance
it
numerical quantity.
The
7. Elasticity.
elasticity
the stress required
to
of a body
is
dimensions, and
offers to alteration in figure or
produce a strain:
it
is
is
measured by
therefore the
quotient of an increment of stress by the resulting strain when

both are indefinitely diminished, and, being a magnitude of the
same kind
as stress,
We
notice the varieties
shall
and
stresses
is
/,
dynes per square centimetre.

only which correspond to the
in
strains described above.
change of tension
If a
length
measured
is
or a strain
also called Young's
dT
the Longitudinal Elasticity
change of hydrostatic
dv
in a
v,
is /
This
dl
Modulus of Elasticity.
If
volume
cause change of length dl in a
stress
dp cause change of volume
or a strain
the Volume-Elasticity
is
dp
V-T.
dv
8.
Mechanical work done by or upon a body during
a strain.
This is measured per unit volume by the product of

the strain into the average value of the component of the stress
in its direction.
We
section
dl,
<o
perpendicular to
the total
two particular cases.

same at every point of a
direction and causes deformation
shall consider only
If a longitudinal stress
work done
is
its
is
the
Twdl.
In the case of a rod or
io
a cord stretched by the weight of
work done
is
mgX
where A
grams, To*
= mg,
and the
the total change in length.

subject to a hydrostatic pressure />,
ergs,
is
body of volume v is
is the same at every point of the surface, the pressure
on a small area dA is pdA if the body increases in volume so
that this pressure is overcome normally through a distance dn,
the element of work done is pdAdn, and thus the element of
work done by the whole surface is pfdAdn or pdv hence the
If a
which
whole external work
the integral
fpdv
taken between the
limits.
proper
The work
is
motor or
or decrease, that
Mr.
is
is,
as dl
resisting according as /
and dv
and v increase
are positive or negative.
Watt
introduced a graphic mode of representing these

both of which are of the same form since T and p
integrals,
wdl and dv changes of volume. If p is measured

along the axis of ordinates and v along that of abscissae, then
the connection between the suc-
are stresses,
cessive states of a
body may be
represented by a curve, and
pdv
will represent the infinitesimal area
between the curve, the axis of

volumes, and two infmitesimally
-
tx
C;
JTJ
-,,
(/
distant ordinates.
Hence if ABC re-
presents the states through which

a body passes from that denoted
by C, the total work done by the body

Again, if CD A represents
represented by the area ABCcaA.
states
the successive
passed through by the body from that
C
that
to
denoted by A, work represented by the
denoted by
by
to that denoted
is
CDAacC is
done upon the body, for it has been compressed.

a body passes through the whole series of states
represented by the closed curve ABCDA, the useful work done
area
Hence
is
if
the excess of the area
ABCcaA
over the area
CDAacC
(A
and C being the points where the curve is touched by the extreme
this work is done by the body
ordinates) or the area ABCDA
if the expansion occurs under greater pressures than the con:
traction, as
when
the series follows the direction of the arrow,

that
is,
the direction of the hands of a watch
the contraction
if
1 1
occurs under greater pressures than the expansion so that the

curve is traversed in the direction opposite to the arrow, the
work is done upon the body.

These graphic representations of work are
called Indicator
an instrument called the Indicator which was

invented by Mr. Watt to determine practically the work done
by a steam-engine. M. Carnot gave the name cycle to such a
diagrams, after
series
of modifications as
When
body,
we
is represented by a closed curve.

a curve denotes the successive changes of state of a
shall say that the
body
moves, or that the modifications
occur, along that curve.

On the subjects of this chapter the student is recommended
to consult Professor Maxwell's Theory of Heat, chaps, iv, v, xxi :
he
will find therein also

(p.
105) a
full
description of
an im-
proved indicator.
9. Virial.
dynamical idea of great importance has re1

If the small
by Professor Clausius
acted upon by force F, whose axial com-
cently been introduced
mass
\L
at (x,
is
y, z]
ponents are X,
Y, Z,
we have
\ioc-X,
ny=Y, pz =
Z;
whence, since
we have
J/x^
=-\Xx + };~.nx\
(5)
The mean
value of each of these terms during any period /

found by multiplying it by dt, integrating for the time from
o to /, and then dividing by / thus, indicating mean values by
is
bars,
we have
Suppose now
strictly periodic,
last
term of (6)
returns to
1
its
SS' Ann
that the
motion of the parts of the system
and
any multiple of
is
/ is
zero, as at the
original position.
-
cxli
( l8
7)
their period
is
then the
end of a period each part

motion is not strictly
If the
I2 4> trans. Phil.
Mag.
[4] xl. (1870) 122.
12
periodic, but of the kind termed stationary
i.e.
when
the parts
do not continually move further and further from their original

positions and their velocities do not continuously increase or
decrease, the parts thus moving within a limited space and their
velocities only fluctuating within
efficient in
~.ftjr
then the co-
certain limits
square brackets (the difference between the values of
at the
beginning and end of the period
become zero
necessarily
for
any value of
does not
cannot however
it
/)
increase indefinitely with the time, but can only fluctuate between certain limits, so that, by taking / large enough, we may
render the
last
In both these cases we
term of (6) negligible.
have therefore
\^^-\Xx.
This mean value of the expression
component of
that
for any
kinetic
to the
As
Xx
xand the equation shews

stationary motion the mean
is
called the
the virial of the particle,
particle
energy relative
same
(7)
\
of a system
to
any
in
direction
is
equal
to the
virial relative
direction.
similar equations in j/
and
z hold,
we
have,
if
is
the
mean
velocity of the particle,

~
(8)
if
is
the radius-vector of the particle,
of the total force, and

that of the total force F.
it
and writing
K for
its
P the
resolved part along

its direction and
the angle between
Summing now over the whole system

mean kinetic energy, we have
(9)
these right-hand expressions are called the virial of the system,

so that the mean kinetic energy of a system in stationary
motion
The
is
equal to
its
virial.
general equation corresponding to (5)
is
obviously
\^.\*.?
(10)
good whenever
so that the above theorem holds
independent of
is
rigid
this is the case, for instance,
and moves with one point
fixed,
of homogeneous
consists
13
.
when
and
2 \L^
the system
again,
fluid
is
when
it
which always
incompressible
occupies the same portion of space though its molecules are in
motion, since as each molecule leaves a spot it is replaced by
another of equal mass.
In the general case there are both internal and external forces
acting upon a system, and hence therefore an internal and an
If R is the attraction between two particles
external virial.
at the points (x,y, z), (x',y
/u,'
r apart, then for the internal forces
/u,
z')
',
respectively which are
r
with similar equations in
and
r
therefore
-i2.(Xx+Y_y + Zz)
so that the internal virial
is
2 \
.
which innumerable atoms move
*.\Rr,
;
indeed, with a body in
irregularly but in essentially
like circumstances, so that all possible
phases of motion occur
simultaneously, the difference between this
value of 2
is
system
(n)
Rr
mean
value and the
\ Rr
corresponding to any configuration of the

inappreciable, so that the latter may be taken for the
.
internal virial.
The
external virial
is
easily
found in the case
wherein the external forces consist of an uniform hydrostatic

pressure p for then
;
pffx cos a da = -p ff x dy dz = -/>b,
Xx
where da
is
an element of area
inclination of the
normal
volume of the system
at this
y, z\ a the
point to the jt-axis, b the
at the point (x>
hence
-JS..(**+K>/ + Z*)
and the external
constant.
In
this
virial
is
f/>b,
= f^,
case therefore
^=1/6 + 2. J*r,
or, as
we may
(12)
both these quantities being
write
(13)
it,
>b
= S^-is.*r.
(14)
CHAPTER
II.
Introductory Thermal Principles.
Measure of Heat.
10.
able quantity;
amount spent
grams of ice
for
Whatever heat
much
thus twice as
is,
is
it
a measur-
effect is generally proportional
its
as to melt one,
to the
two
and two grams of coal give on
heat
is
required to melt
combustion twice as much heat as one.
The
usual thermal unit
defined as the
is
amount of heat
required to raise unit mass of water from the temperature of

melting ice through one degree c. of temperature measured on
the air-thermometer, and
is
it
it
is
In
called a calorie.
necessary to mention the
this definition
temperature, as
initial
M. Reg-
nault's experiments shew that

6
or,
according to
+ 0-00002
62
+ 0-0000003
later calculations
the heat required to raise one
perature of melting ice through

I
is
the
amount required
one degree.
formula
is
2
,
of water from the tem-
gram
degrees
c.,
and hence
that
+ O-OOO220
to raise the
same mass
real calculation gives
term independent of
for the
is
The
1
by M. Bosscha
+ o-ooo 1 1
is
this is
at 6 c. through
i-oooii instead of i
merely because the
empirical and thus not quite accurate
first
the difference
negligible.
As temperature
is a most important
of
heat, it may be well to
dynamical theory
review the principles and conventions of the ordinary scales
which depend on the properties of particular substances.
11.
Temperature.
quantity in the
Pogg. Ann. Jubelband (1874), 549.
When
15
two bodies are together and one gains while the other
loses heat,
we
say that they are at different temperatures
neither gains or loses,
we
if
say that they are at equal temperatures.
For the measurement of temperature we avail ourselves of the

volume or pressure as they gain or lose
fact that bodies vary in
we may thus take this change of volume or pressure,

reckoned with a standard substance from a standard condition,
heat
or any function of it to define our scale of temperatures, and

hence get an infinite number of systems. The following are
the necessary or usual conventions
(1) Definition of rising temperatures
a body that loses heat

by radiation or conduction is said to have the higher
temperature. This convention is opposite to that of M. Delisle
:
to another
and of M. Celsius'
original thermometer.
(2) Standard substance
temperatures, tKe
air for all
alcohol for low, mercury for mean,
W*Uf*
b&HC) v5/*ty oi^lass
those of melting ice and boiling

(3) Standard temperatures
water under standard pressure, as suggested by Newton.
(4) Standard pressure: for France 76 c. of mercury at the
:
temperature of melting ice at sea-level in Paris, or 1013573

dynes per square centimetre ; for England 29.905 in. of mercury
at
same temperature
the
dynes per square
A more
c.
at sea-level in
London, or 1013252
the ratio of these pressures
is
1-00032.
would be one megadyne, p^r orit

Number
of
degrees between the standard temperatures
(5)
on Reaumur's scale 80, on Fahrenheit's 180, on Celsius'
scientific pressure
or the Centigrade scale 100.

(6)
The
mometer
variable used to define successive states of the ther-
with alcohol and mercury this is the volume, with

under constant volume or the
air it is either the elastic force
volume under constant pressure.

(7) Definition of a degree: Galileo's is used with the
ordinary thermometer and with the air-thermometer when the
volume under constant pressure is the variable employed ; it is
if the thertnometric substance assumes volumes
the following
v, V Q at the higher and lower standard temperatures respectively
:
and n
is
the
number of degrees between
these temperatures,
one degree
is
defined by a change of volume equal to -
T when
and thus the temperature
the volume
is
is
>
given by
T7"
T
Dalton proposed
to define a degree as corresponding to
//
a change of volume V Q /\
T corresponding
to
volume
in this case the temperature
is
given by
L=L
y=
or
In neither of these cases
same
is
the heat-value of a degree the
it is more nearly so howDalton' s system, and hence the
in different parts of the scale
ever in Galileo's than in
preference given to the former.

In the air-thermometer, when the elastic force under constant
volume
the variable,
is
are the elastic forces at the
if p,
standard temperatures, a degree

pressure
hence for an
elastic force
the temperature
is
P-Po
this case the heat-value
of a degree
in all parts of the scale, as
ments on
defined by the change of
r=,^h n
given by
In
is
is
very nearly the
shewn by M. Regnault's
same
experi-
specific heat.
zero-point of the scale: Celsius and Reaumur

take for this the lower standard temperature; Fahrenheit
(8)
The
took the lowest temperature

by salt mixed with snow.
known
in his time, that
produced
we might
(9) Absolute temperature on the air-thermometer
represent zero temperature by the state of a gas entirely deprived of heat, in which case its pressure is zero ; if then -or
:
denotes the change of elastic force measuring one degree, absolute temperature corresponding to pressure
is
given by
Taking these degrees the same as above, so that
we
have,
The
if
absolute zero
quantity
is
is
what
is
coefficient of increase of elastic force at constant

its
or
>
degrees below ordinary zero,
the reciprocal of m,
generally denoted by a:
17
value
known
as the
volume, and
is
shewn by M. Regnault's
is
1
to be neither the same for all gases nor
exact experiments
constant for a given gas ; it varies with the density, and in the
case of dry air
may be
a
2
represented by the formula
= 0-00365343 + 0-00895
p,
Thus
the density in grams per cubic centimetre.
the so-called absolute temperature depends on the substance
where p
is
used and
When
not constant.
is
affect the coefficient,
we
find
the density is too small to

or absolute zero is
m= 273-72,
273. 72 c.
by a graphic construction from the
numbers
Professor
Rankine 8 finds this value
experimental
c.
-274-6
If we consider a constant, and measure temperature from the
zero thus obtained, Boyle and Charles' laws combined may
in this case
be written
pv
a simple form
A
in
first
= const.,
given by
M. Clapeyron.
perfectly absolute scale of temperature will
41
from thermodynamic principles
till
be deduced
then
we must
consider absolute temperature a function yet to be defined.

12. Characteristic equation of a body.
The action of
heat generally causes change in
the qualities of a body,

refractive index, emissive
elasticity,
density,^ stress, temperature,
&c.
two
of
these may however be
power,
any
eeflduc^ptfity,
all
taken in general as independent variables to characterize its

state
i.e. the state of unit mass
and each of the others will
1
Mem. de I'Acad. des Sciences, xxi. (1847) no.

3
Phil. Trans, cxliv. (1854) 357Trans. R. S. E. xx. (1853) 5^1.
be expressible as a definite function of the two selected ; thus

between every three of the qualities there exists a relation
f(x>y,
and such
relations are
z)
termed
o,
These
characteristic equations.
are only to be determined by experiment, and our knowledge

is at present so incomplete that we are in a
position to discuss
that relation only which connects stress, density, and temper-
and while
owing to the complexities of the

a simplification by considering
introduce
general question,
the density, stress, and temperature the same throughout the
ature;
herein,
we
body, and the stress a hydrostatic pressure, yet the problem of

determining the exact form of the function still surpasses far
the extent of our actual knowledge.
One of the chief difficulties
of the discussion arises from the experimental measurements
that are the most easily and accurately made not being those
most serviceable from a
We ought
not the stress
it is
on
theoretical point of view.

'
meant by the stress of a body'

exerted on it by external bodies, but depends
to explain
what
the actual state of the
is
body
itself only,
being the force
it
exerts at every point per unit area across any section in con-
sequence of the motions or

only equal to
is
relative positions of its
atoms
and
external stresses in the case of equilibrium.
This equilibrium we shall generally suppose to subsist.

It will be well here to introduce another quality, the entropy
of a body, which is the property that remains constant when
body undergoes any change but without receiving or losing

any heat a more precise mathematical definition of entropy
the
be given in
60.
We have thus to determine the form of the equation
will
/(/,
which
is
par
0, /)
o,
excellence called the characteristic equation or the
equation of elasticity,
p being
the stress (supposed a hydrostatic
pressure) per unit IIJCTI or the specific pressure, v the volume of

unit mass or the specific volume (called also very expressively
the bulkiness of the body by Professor Rankine and the rarity
by Professor Maxwell), and
may
the absolute temperature.
represent the equation graphically by
means of a
We
surface

if
we
axes
19
take /, v, t as coordinates referred to three rectangular

then the determination of the relation between p and v.
amounts
for a given value of /
surface
by a plane perpendicular
tion of these sections for

plete surface,
whence the
to finding the section of the
values of
all
and a determina-
to the /-axis,
sections
would give the comto
by planes perpendicular
The
would be known.
three possible series of

experiments here indicated are however so incomplete that it is
the other axes
necessary to make use of all, and not of one series only, to

determine the surface. The surface differs for all bodies, with
common
however three points of

in the octant where p v,
on
are
all
likeness
positive
(2)
it
exists only
is
it
terminated
oriin
the side nearest the
-re&uiec -o atin^
volume only approachcD a finite minimum
(3) on the
other side it extends to infinity with a hyperbolic paraboloid
to?
me
as asymptotic surface, experiment
shewing
that,
as the
tem-
bodies approach more and more to the perfect
perature rises,
state,
defined below, wherein
pv
= Rt.
The
equation assumes a different form in the saturated state

of a body, or that in which all its parts are not in the same
physical
state
thus
the
solid
and
liquid
forms
exist
may
and gaseous, or the solid

together simultaneously,
and gaseous, but experiment shews that in these cases the
pressure and the temperature cannot vary independently of each
the liquid
other,
and the
specific
remain constant
volume may vary while both
the equation therefore breaks
one
F(p,f)
and the other an expression

another variable.
Let
up
/ and
into two,
o,
for v in terms of either
p or
and
be m, the proportion of the

substance in the higher form in unit-mass of the mixture then
if /, s are the
specific volumes of the body in the higher and
this variable
lower forms respectively, we have

v
s'm + s(im}
= s + (s's) m;
(i)
and, the quantities /, s being functions of / which experiment

must determine, this relation gives v in terms of / and m, or
in terms of v
and
/.
c 2
2O
In contradistinction to the term saturated a body is called

when its temperature is higher than that corre-
superheated
sponding to the
state of saturation for the
same pressure
as
o.
Thus slight compressions
given by the equation F(p<, /)
or expansions of superheated liquids and vapours will not alter
the state of aggregation of the body, though they may tend
to bring
them nearer
to the state of saturation.
Experiment has shewn that as the permanent gases are more

and more rarefied they follow Boyle's law the more exactly,
and their increments of pressure at constant volume are the

more exactly proportional to their increments of temperature
on the air-thermometer scale. We will define as a perfect gas
a substance which exactly conforms to these laws, temperature
being measured absolutely: its equation is easily found, as it
is
of the form pv
of
we
or
x'W
= \(f) and
= -\'M
is
to
be independent
representing then this constant by
constant;
pv.
get
13.
-~
Choice of variables.
=&
It is
(a)
not a matter of indifference
which pair of the three variables p, v, t we take as independent

in the equations we are about to form, though all three have
been adopted.
(1)
M. Clapeyron chose p and
tion in this case
is
the
however when the substance behaves

(2)
as the graphic representa-
v,
most convenient
M. Saint-Robert
takes
is
an ambiguity
this pair
cannot how-
there
like water.
p and
/:
ever be used in the case of a saturated mixture, since
p and
are not then independent.
Clausius and Sir W. Thomson use v and

so far as we know, is applicable in all cases.
(3) Professor
which
pair,
Hence
it
is
/,
best to use the last pair in general, though the
always be employed when no ambiguity is thereby

introduced (as in the discussion of gases) the first will however
others
may
generally be used in diagrams.

14. Thermal capacities.
The amount
of heat which
is
required to change by unity any quality of unit-mass of a body
Thermal Principles.
Introductory
under given circumstances
responding to the given
capacities special
called the thermal capacity cor-
is
change
names
21
in three cases only
have these
the specific heat at constant volume,
the specific heat at constant pressure,
and the
latent heat of
expansion.
15. General equations of the effects produced by heat.
For an increase dt of temperature when the volume remains
constant, heat equal to k dt

at
constant volume
when
is
and
is
required,
similarly for
if
the temperature remains constant an
required,
if /
variations of /
dH is
amount of heat Idv
represents the latent heat of dilatation.
If the
and v take place together, then the heat imparted
dH =
given by
obvious that
It is
the specific heat
is
an increase dv of volume
with regard to v
/ is
when
therefore denotes
it
by
k dt+ldv.
(3)
the partial differential coefficient of
remains constant
(
)_
Professor
Clausius
and M. Saint-Robert by
(ff}) vt
symbols being used to express the other capacities.
similar
These are much more expressive than the
single
letter
we
employ, but present such a formidable appearance in the thermodynamic equations that the simpler symbols, even with the
trouble of
If a
remembering their meaning, are preferable.

body undergoes an increase dt of temperature when
pressure remains constant, heat

-)_
and
or
for
(ff^) t
Kdt
is
being the specific heat
an increase dp of pressure
amount of heat
I.
without specific
name which
dp
is
required,
is
absorbed,
at
at constant
is,
temperature an
expressed by (-7-)-
capacity
If these
dH required
dH= Kdt+Ldp.
is
(4)
represent the state of the body by the variables/ and v,

get in a similar manner some such equation as
If
we
we
dH=$dp + *dv,
jj,
the
constant pressure
L being a thermal
variations occur simultaneously the total heat
that
S>
being the partial differential coefficients of
(5)
H with
respect
22
to
p when
is
when p
constant and with respect to v
is
constant
respectively.
The
student should find no difficulty in remembering the

six symbols.
With regard to other
meaning of the above

differential notation,
the
symbol
<)
we
use the symbol
---
its
partial
used in
v and
this latter
/,
is
with
coefficient only
thus
respect to
if
case the variables are those
means
and
/,
no brackets
we have
are
adopted,
one being constant which does not occur in the

and if the variables are otherwise, that which is con-
that
expression
stant
differential
and
to signify total
to denote partial differentiation
the total differential coefficient of
c/
denoted by a subscript outside a bracket;
(-^
thus representing the partial differential coefficients of

respect to / when v and p respectively remain constant.
with
We
omit the bracket and subscript where /, v t alone

are concerned, the omitted variable being considered constant
shall further
thus in
constant.
and p are the variables considered, the
latter
In Chapter VIII the only variables used are
and no brackets are employed.

16. Relations between the thermal capacities.
characteristic
and
/,
equation we can
being
and
By
/,
the
express v as a function of
and therefore
*-S*+g*;
hence we have from
(3)
Kdt + Ldp =
whence
it
and
(4)
kdl+ldv
follows that
or
= A--4,
(6)
methods we
similar
By
find
= K+L*$
or
and
= K-k,
-L^-f
c>/
&
all
(8)
d/
= A"
o)
dz;
d/
Thus
23
the capacities are expressed in terms of the specific
We have made no supposition as to the units in which

heat dH and therefore all the different thermal capacities are
heats.
the
measured ; as however we
heat
is
shew in the next chapter
shall
that
megalerg therefore being a conshall consider the capacities all expressed in
equivalent to work, the
venient unit,
we
megalergs.
M. Regnault has experimentally determined
in thermal units,
and denoted by C, the specific heat under constant pressure

for most substances
represents the value of the same in
work units, andâs shewn in the next chapter, is equal to JC
where J is ff-&r. Denoting by c the specific heat of a body
at constant volume in thermal units, we have also k = Jc.
:
The
fc
or
ratio of the specific heats, viz.
is
a very im-
*\>
portant quantity
It
K.
is
we
denote
shall
sometimes denoted by
convenient or ouggectivo a
From
we
equation (a)
it
y,
therefore
which
oor
letter .u/*tA
when v
have,
is
by a
is
single
symbol
not however so
oot**<ffl>
constant or dv zero,
or
and
similar relations
and therefore by
j
and
also
(<$)
\ /
we have
c
A -A
Ây
further
.
Awi
=.
^-^-r
c
c
o^
Z'
"
i.
24
Introductory
Thermal Principles.
Volume -elasticities.
17.
their ratio, are
The volume-elasticities
of a sub-
two variations of condition, or rather
stance with respect to
important quantities
these are the elasticities
when change of condition occurs without addition or loss of

heat and when it occurs at constant temperature, i.e. the elasticities at
Sir
of
constant entropy and at constant temperature (which
W.Thomson
elasticity),
see
60) and
We
and
also terms the kinetic
and they
static
are denoted respectively
bulk-moduli
by
(or e^,
e,.
have by
7,
dP-\
^P
= -r =
hence
t
(12)
*,
or the ratio of the kinetic and static bulk-moduli

ratio of the specific heats under constant stress
is
equal to the
and constant
strain.
The
reciprocal of the volume-elasticity
of compressibility and
,
is
is
called the coefficient
generally denoted by
/3.
CHAPTER
III.
LAW OF THERMODYNAMICS.
FIRST
18.
Heat a form of energy.
a system of bodies
is
It
has been shewn
subject only to the
that, if
mutual actions of
its
parts, the total energy of the system remains constant, whatever
changes occur within it. But in no machine does this condition seem to be exactly realized, even after allowance has
been made
energy spent in producing sound and the
for the
sensible vibrations of the

in all the cases
when
machine
heat however
is
produced
apparent loss of
energy occurs as in
the case of friction, percussion, &c., and this fact has led to the
conclusion that heat is a form of energy.
The chief arguments
for this
is
this
assumption and for the further hypothesis that heat
the energy of motion of the ultimate molecules of matter
may be
briefly stated as follows.
Heat is not materialjout

ialjou a form of energy*
u A ft** the same
v
u
bodyQias
weight when hot as)when cold,
(a)
unless there has been oxidation
this may be shewn by weighing
i
'i&*^'t&&.
AA
it
in Ritrogea eg- in a
(/3)
vacuum.
Heat may be generated in any amount and cannot therethus Count Rumford found himself able to
fore be material
any quantity of water by the heat produced in boring
boil
cannon.
(y)
The
disappearance of mechanical energy
is
always ac-
companied by the production of heat.

(8)
The
quantity of heat produced
to the quantity of
in
24.
work spent
in
is
exactly proportional
producing
it
this is
shewn
26
Heat
(e)
When
is kinetic
energy, at least partly.
heat passes by conduction between bodies in con-
changes occur in their volumes,
tact at different temperatures,
shewing that heat
is
accompanied by motion.
(f ) If the bodies are
separated interchange of heat is still

thus the energy of heat during its radiation between
the bodies must exist in the kinetic form.
observed
(77)
The
perfect identity of the laws of radiant heat with
those of light as to reflexion, refraction, interference, absorption,

polarization, double -refraction, &c. requires the same theory
as to their nature.
Heat
(0)
No
is
molecular energy.
motion of a body as a whole is observed as a conthe motion must therefore be molecular and,
sequence of heat
at least in solid bodies, oscillatory.
The argument
(rj)
may be expanded
radiation from a luminous
with advantage.
be received on a screen
slit
If the
after
passing through a prism or grating, it is found in general to

be decomposed into a multitude of rays of very varied refrangiNow
bility, some of which are visible and others invisible.
each ray of the luminous spectrum produces not only the sensation of light in the eye, but also the sensation of heat
when
on the nerves and chemical action in the haloid salts

of silver and other substances it is also found that any variait
falls
power of producing one of these effects entails a

proportionate variation in its power of producing the others,
whether the variation is caused by reflexion or refraction, interference or absorption, polarization or double-refraction, whence
we conclude that all three effects are the result of one and
tion in
the
its
same action according
and as the phenomena of
as
transversal vibrations of ether,
explanation for those of heat.
tended further to the
works
it
light
one way or another
by a lens
no difference
by concentrating the inon a solid body, or by passing
invisible radiation, as there is
whatever in kind between them

visible ultra-red rays
in
can only be explained by the

we can only accept the same
This explanation must be ex-
for
27
the invisible ultra-violet rays through quinine, we can transmute

them into radiation visible to the eye the phenomena of calorescence
19.
clusive
and fluorescence.
Law of Thermodynamics. The really conargument of those given above in proof of the dy-
First
namical nature of heat
and work
is
is
(8)
called the First
H.
this
Law
exact equivalence of heat

It
of Thermodynamics.
was
dropped out of sight

Dr. Mayer 2 restated it in 1842 on theoretical grounds and
determined the ratio between the units of heat and work
first
stated
Sir
by
Davy
in 1812, but
but by an assumption at that time unwarrantable ; it owes its

conclusive demonstration to Dr. Joule's experiments 3 in 1843,
which have been confirmed in their result by the widely differing methods of experiment he has since employed and which
are given below.
The ratio of the unit of heat to the unit of
work
commonly called the mechanical equivalent of heat is

therefore generally denoted by J\ its reciprocal or the ratio
of the unit of work to the unit of heat hence termed the heat
equivalent of
work
is
often expressed
depend upon
necessarily
by A.
These
ratios
the units adopted.
DR. JOULE'S EXPERIMENTS.

20.
I.
By
magneto-electricity.
It is
known
that,
if
wrapped with insulated wire forming a circuit be

rotated between the poles of a magnet, induced currents are
soft-iron core
set
up
the wire
in
on the successive magnetization and de-
magnetization of the core.
mined
that
energy
is
if
Dr. Joule, having already deter-
voltaic electricity pass along a resisting wire its
transformed into a heating-effect proportional to the

and to the square of the current intensity,
resistance of the wire
shewed
law,
in 1843 tnat the induced currents conform to the same

and further that the heating effect is in their case pro-
portional also to the
5
*
work spent
in causing the rotation
4
.
The
Elements of Chemical Philosophy. London, 1812.

Lieb. Ann. xlii. (1842) 233, trans. Phil.
Mag. [4] xxiv. (1862) 371.
Phil. Mag. [3] xxiii. (1843) 263, 347, 435.
Ibid.
28
core he used consisted of soft-iron plates insulated from each

other by oiled paper and bound together
this was wrapped
with silk-covered copper wire and enclosed with water in a
:
which was mounted horizontally between the poles
glass tube
of an electromagnet where it could be rotated by wheelwork

about a vertical axis through its centre.
The tube was covered
with
^p
^
^
tin-foil,
prevent
horizontally to
slit
external
induced
cur-
'
^^^
rents,
and with
flannel to pre-
vent radiation and convection of

heat.
The
dipped
into
terminals of the wire
commutator grooves,
containing mercury covered with

water, which communicated with
a galvanometer. In each experithe tube was rotated for 1 5
ment
^/^-
minutes and the

Fig.
rise in
tempera-
ture of the water measured.
2.
The
necessary corrections were

(1) For radiation, which was experimentally estimated at a
quarter of the difference between the mean temperatures of the
room and
the water.
(2) For the heating of the water due to its agitation by the
rotation and to possible conduction from the pivots
experiments made with the current of the electromagnet broken, when
:
corrected for radiation, gave this correction.

(3) For the heat developed by the induced currents set
up
was estimated by experimenting with the circuit of the wire round the core broken
so as to prevent induced currents being generated in it, and by
in the separate plates of the core
this
then applying the two former corrections.

The tube and core were made to revolve by the fall of
weights, a correction being made for the weight required to
move
Work
the whole with equal velocity without the inductive action.

being expressed in gravitation measure, the mean of the
equivalent to
y=
45976, or that one calorie of heat

45976 gram-centimetres of work.
experiments gave
is
29
the friction of fluids,
By
II.
21.
(a)
By working
holes in a glass cylinder

piston perforated by a number of small
as in I, Dr. Joule
work
the
water
and
measuring
containing
in
J=
1843 found
42464.
1849 Dr. Joule experimented by another

(t>)
method 2 with the apparatus constantly improved. He finally
used a cylindrical box of thin copper in which rotated a vertical
armed with
spindle
From 1845
to
eight brass paddles of

complicated form, as
shewn
four
the figure
in
fixed
vanes,
so
shaped as
barely to
allow passage to the
paddles, prevented the
with which
fluid
box was
carried
filled
the
being
round by the
motion, the energy of

rotation being
trans-
formed by the consefriction
quent
heat
this effect
be read to
^^
into
Fig.
3.
could
degree
c.
through an opening in the
by a
sensitive
thermometer inserted
The
paddles were set in motion

of
which,
by
weights W,
turning the axes of the pulleys
P supported on friction-wheels, communicated a rapid motion
the
lid.
fall
to the roller
by means of the cord C
the roller could be
detached from the spindle by removing the peg p, so that the

weights could be wound up without moving the paddles an
operation performed 20 times in each experiment, which lasted
about half-an-hour. The part b of the axis was of boxwood
to prevent loss
culation
1
Phil.
Phil.
is
The
of heat from the box by conduction.
as follows
cal-
[3] xxiii. (1843) 442, corrected in [3] xxvii. (1845) 206.

xxv ( l8 47)
[3] xxvii. (1845) 205, xxxi. (1847) J 73> C309, Phil. Trans, cxl. (1850) i. 61.
Mag.
Mag.
30
Let
the
be the observed
rise
of the
in temperature
fluid,
by radiation (estimated by experiment joined with

Newton's Law of Cooling), m' the reduced water-capacity of
the
loss
box and
its
contents
then the work done has raised the
equivalent of m' grams of water through

m' (0 + T) or pj grams through i degree c.
Let
H be the total
the
fall is
the
fall
+r
made during an
this velocity
is
number of times
mean velocity of
of the weights, n the

experiment, v the
fall
degrees, that
would be generated by a
fall
through
v2
and hence the whole height through which the weights
o
fallen to acquire the energy which is destroyed on
the weights reaching the ground without a corresponding thermal
must have
V*
box
effect in the
fall.
M be
Let
is
or h say:
thus
is
the effective
2g
the
mass of the moving weights,
the
mass
with equal velocity when the

paddles are unattached (this measurement being attended with
then the work done by the fall, expressed in
uncertainty)
The elasticity of the
gravitation measure, is (Hti)(Mni).
whose
fall
would turn the
roller
string
is
found to do work
called the resilience
after the
weights have reached the floor, nE being thus done altogether,

so that the total work gained by the fall of the weights is
(H-ti)(M-m) + nE:
but part of this is spent in producing vibration in the apparatus
and in prothe energy of which V is difficult to estimate
S
which
Dr.
of
measured
the
ducing sound,
Joule
energy
by
work required to obtain from a violoncello a note of the

same pitch and intensity hence the total work spent in friction
of the fluid is in gravitation measure
the
(H'ti)(M'ni) +
and the
ratio
of
Wi
nEVS
or
TO
say,
to $? gives y.
The experiments performed were upon
water,
sperm-oil,
mercury, and cast-iron working upon cast-iron in mercury, and

In one
in the last case only had -6" and V appreciable values.
of the
final
experiments on water the following were the values
of the above quantities expressed in grams, centimetres, and
degrees centigrade
Q
T
m>
f^
=
=
=
=
22.
=
h
1976-20;
y= 42445,
y= 42394III.
By
3201-001,
6315-97,
whence
made,
H=
0.3196,
-0-0067,
6-15,
26318-223,
m =
183-834,
20,
0-386;
when proper
or,
=11701-287,
021= 83880155;
correction for
the compression of gases,
(a)
vacuum
is
When work
spent in compressing a gaseous fluid, evolution of heat oc-
is
curs
its
and
no heat
if
absorbed or emitted by the gas during
is
variations in density, the
thermal effect
dition
is
may
be
work spent
in producing a given
The above
directly determined.
the principle unwarrantably assumed by Dr.
con-
Mayer;
Joule however shewed in 1844 that it is appreciably satisfied

by air, and he deduced 43781 as the mean value of J given by
Dr.
his experiments
1
.
Air was compressed by

a force pump into a copper
fV
Kf
reservoir of
2237 c.c. capahaving been drawn

through a Ca C1 2 drying
tube and a leaden spiral
city after
tube
immersed
which gave
it
in
water,
constant
temperature. To collect the

evolved heat the pump and
reservoir were
known mass
immersed
in
of water in a
tinned- iron jar, which served
as
a calorimeter and was
provided with double walls

to diminish the loss of heat
by
radiation.
After
300
Fig. 4.
strokes of the piston, when

the pressure had become about
1
Phil.
Mag.
22 atmospheres, the rise in
[3] xxvi. (1845) 369.
temperature of the water was noted by a thermometer reading
to ^\-Q degree
(1)
For
The
c.
radiation
Newton's Law
(2)
necessary corrections were

was estimated by experiment and
this
of Cooling.
For the heat due
to the friction of the piston in the
pump
was estimated by working the piston 300 times without

compressing air, and making an allowance for the difference
this
mean pressures during the two experiments a

much uncertainty.
As the pump was worked by the experimenter,
in the
of
correction
work
the
could not be directly measured: but assuming Boyle's Law,

pv
p Q v to hold in the case of air, we have by integration, as
the temperature
Work
is
nearly constant,
=/p<h= p v Jf"
^V = A p. log, ^v
V
a
the suffixes denoting
initial values.
ing the compressed air to expand

v is the capacity of the reservoir.
V Q was determined by allowpressure became
its
till
pQ
One of Dr. Joule's experiments gave 1/0 = 23482 c.c., and = 77- I 6s c. of
barometric pressure, or the weight of 77-165 x 13-596 = 1049-13 g. per sq. c.
The gain of heat was equivalent to a rise of 13 25-4 g. of water through l
degree
c.,
whence
for gravitation
measure
1049-13 x
34 82 X
ge
=
-fj
(b)
43702.^*
When
ex-
gases
pand doing work, they

experience a cooling and,
the above condition being
satisfied, the cooling effect
is
proportional to the
Dr. Joule
work done.
determined
the
relation
between these two

Fig.
in the following
5.
Compressed
effects
way
air
1
:
was
allowed to expand from the copper reservoir containing

1
Phil.
Mag.
[ 3 ] xxvi. (1845) 369.
it
33
through a long spiral tube into an inverted cylindrical glass

vase standing over water: as it chased out the water it did
work against the atmospheric pressure, and at the same time the
water in which the reservoir and spiral were placed experienced
The gas was made to issue so slowly that the

energy given to the water was negligible.
If p Q is the atmospheric pressure and V Q the volume of water
a cooling.
displaced, the
work done
is
equivalent to the raising of a
column
square centimetre in section through V Q centim being

metres, and is therefore p Q V Q in gravitation measure,
the water-equivalent of the calorimeter and all its contents and
of the atmosphere
mO
the cooling,
and the
ratio of/>
calories
to
z>
mO
is
the heat that has disappeared,
is J.
In one of the experiments /> = 76-529 c. of barometric pressure, or the

weight of 76-529 x 13-596 = 1040-5 g. per square c. 1/0 = 44621, 771 = 10682,
and 6 = 0-0966 whence J = 44992.
;
23. IV.
By
the thermal effects of an electric current.
In 1867 the method adopted 1 was to pass an electric current

through a resisting coil and measure the heat generated. The
coil consisted
of 366
c.
of platinum- silver wire, bent double to
on its parts and then coiled in a spiral round

a glass tube
it was
provided with stout copper terminals and
the whole was coated with shellac and mastic varnish.
The
calorimeter was of thin copper and, to diminish radiation as
much as possible, was coated with tin-foil, then with two layers
of silk net, and lastly with another layer of tin-foil in it were
inserted a thermometer and a stirrer driven by clock-work.
Radiation and conduction by air were calculated from a series
of interpolated experiments and allowed for.
If m is the reduced water-capacity of the calorimeter and all
its contents and r the corrected rise in temperature, mr is the
thermal effect during the time /.
If
is the horizontal component of the earth's magnetic intensity and a the angle of
prevent induction
:
deflection of the needle of a galvanometer placed in the circuit,
will
measure the intensity of the current, k being a

1
B. A. Rep. (1867),
i.
512.
34
constant that must be determined for each galvanometer by

r being the resistance of the coil and / the time
experiment,
of the experiment,
done
(^tana)
in absolute measure,
is
/-/
and the
(by Joule's Law) the work

ratio of this quantity to
mr
gives the corresponding value of y.
The numbers
H = 0-16957,
in c. G.
s.
units for the best experiment
were k
3-0426,
= 6081-96,
= 3600,
989,530,000, /
T
T = 1-312
for
when
corrected
X
IO
whence J is 4-216
vacuum,
or,
7
4-212 X io
Dividing by 981-3, the value of g at Manchester, we find
42923 for the value of J when work is expressed in gravitation measure.
tana
0-59563, r
24. Owing to the greater accuracy attainable in Series II

of these experiments Dr. Joule concluded from it in 1849 tnat
the integer 772 best represents the number of foot-pounds of
work at Manchester that are required to raise a pound of water
from 60
6iF., and a recent
to
of experiments 1, also on
set
the agitation of water, gives the number 772-55 for Greenwich,

or 771-9 foot-pounds to raise a pound of water from 32 to
54-8635, we obtain
Multiplying then by 1x30-4797
number of gram-centimetres of work required to
raise a gram of water from o to ic., or 42349 is the value
33F.
as the
42349
of
J when work is expressed in c.
G. s. gravitation units.
by 981-13, the value of g

o as the value of y when work
multiplication
4 I j55
lute units, or 41-55 if
megalerg
work,
will
we
work
measured
is
Further
Greenwich, gives
measured
in megalergs.
in abso-
As
the
pages as the unit of

take 41-55 as the value of y or i calorie as
be adopted
shall
is
at
in the following
equivalent to 41-55 megalergs.
25.
M. Hirn's experiment.
made by M.
formed
in
Him
to evaluate
1859 which gives a
result
above.
cylindrical block of iron
of stone
(faced
on the
Of
J we
the
many experiments
will
describe one per-
almost identical with the
A and a
end opposite A with
prismatic block
a wrought iron
were suspended horizontally with their axes collinear,

ends being each supported by two strings so as to allow
THate*)
<*
their
of their swinging only in a vertical plane through their axes

between the blocks when hanging in their lowest positions,
Phil Trans,
clxviii. (1878).
Theorie mecanique de la Chaleur, 3 eme ed. (1875),
i.
95.
35
axis collinear with theirs, and just touched by them

was
both,
placed with its mouth facing A a cylindrical piece
of lead C partially bored out so as to form a cup. A was
then raised through a known height and allowed to swing back
with
and
its
strike
maximum
went
ly
in
just as reaching its lowest position
velocity
it
attained
of the energy thus communicated to
its
part
it
deforming
a quantity ultimatetransformed into
heat, part
mitted to
was
trans-
making it
back, C also
swing
joining in this motion,
and
part
back
to
recoil.
must
was given
causing
fourth part
its
Fig. 6.
been
also have
absorbed in deforming the faces of A and B, ultimately taking

but this quantity is negligible from the hard;
the form of heat
ness of wrought iron.

Let m be the mass of A, m' the mass of B, \L the mass of C,
h the height through which A is raised and h" the height
through which
rise together
in
is
it
:
m (hh")
recoils, h' the height through which B and

then the energy that is transformed into heat
(m +i^h' or OT in gravitation measure.
The
heating effect of this energy was determined by suspending C vertically, pouring into it a known mass of water
oc. immediately
at
after the blow- (at
the
same time
start-
ing a metronome), and then reading the temperature of this

water every four minutes, noting also the temperature of the
As
air.
the temperatures in question were low
ences small,
and
their differ-
Newton's Law of Cooling was employed
whence,
the excess in temperature of the water and lead over the

air at time x after the blow, and v is the velocity of cooling, JXf*
if
is
=- OT
the suffix denoting the
or
initial
= r.r-,
value of
2
r,
could the water and
36
have immediately attained equilibrium of temperature.

of r after times x, zx, ^x, ... being TI} r 2 r3 ...,
lead
The values
we find
===
~~
Tl
rO
and therefore
or
TO
T2
= ^- =
...
r2
r *+l 1
-)"
T +l
(-
the temperature of the air being a, the temperatures of

T I} ... are / , /x ,...,
,
if,
the lead corresponding to the excesses r
we have
If
is
*+i
the
initial
temperature of the lead, and
/u'
the
mass of
the water, the energy given by the blow has raised the lead
6 and the water through / degrees, and is thus
through /
calories, C being the specific
equal to jxC"(/
0) + /// or
heat of lead J is thus the ratio of TO to |^.
In one of M. Hirn's experiments ra = 3-5 X io 5 m''= 9-41 X io ^ = 2948,

= 116-6, A'= 10-3, A"= 87, whence TO = 28042336; also o = 8-8,
= 12-1, / = 11-75, whence / = 12-49; and again /*'= 18-5, = 7-873,
C = 0-03145, and therefore p = 659-351 hence J = 42530, work being
5
/!
expressed in gravitation measure.
CONSTITUTION OF BODIES.
26. Intrinsic energy.
Heat thus being energy, the capaa
of
for
work
is altered as well by its loss or
city
system
doing
gain of heat as by the expenditure of mechanical work upon or
by it. The total work that a body can do in virtue of its
any form
is deand
energy
actual condition without
any supply of energy
whatever from without
called
noted by
is
(often also
by
its
^7)
intrinsic
it
in
appears also in Sir W.

and in Professor
Thomson's
papers as the mechanical energy
Kirchhof f s
as the work-function of the body, while Dr.
calls
it
the internal work.
It is
obviously impossible to
the intrinsic energy of a body, since for this purpose
rob
it
of
all
its
heat:
we
we must
however only
are at once
and
these
energy,
are really concerned
with the variations of the intrinsic
Zeuner
measure
37
in all forms
measured^^y determining the amount of energy
that enters and the amount that leaves the body during the
given change.
As
the energy of a system
is
presumably both kinetic and
potential, the intrinsic energy of a body has been broken up

into two corresponding parts, called the sensible heat and the
These and
internal work.
their variations are
the reach of experimental measurement,
and
however beyond
their consideration
belongs entirely to molecular science ; still, as they help us to

form a conception of thermal phenomena, we shall briefly
review them and the theories of the constitution of bodies on
which they are founded,

is
at the
same time remembering
that
all
pure assumption.
27. Constitution of bodies.
Matter is generally supposed

be made up of indivisible unalterable particles called atoms
which are, at ordinary temperatures at least, separated by dis-
to
tances very large in proportion to their size ; but they are so

small that an enormous number are contained in what would
appear to us an infinitely small mass in fact their estimated

There are
diameter is one hundred millionth of a centimetre.
supposed to be as many
different kinds of
elementary chemical substances, and these
atoms as there are
all attract
each other
and may therefore form groups which are called molecules :

thus, according to the different atoms grouped together is one
compound formed, and according to the

of grouping of similar atoms is one or another allotropic
form produced. The grouping together of two atoms of
or another chemical
mode
for instance is supposed to form the ordinary molecule

of oxygen,, while the anomalous grouping of the oxygen atoms
oxygen
in threes gives
four
isomeric
ozone; and again the differences between the

of hydrogen tartrate C^H6 6 one
conditions
that rotates the plane of polarization of light to the right

is
it
therefore called
hydrogen levro-tartrate, a third made

two and called hydrogen racemate which is
and a fourth which is also inactive are supposed to
to the left
up of
hydrogen
and
dextro-tartrate, another rotating
and
called
the other
inactive,
be due to differences in the grouping of the
C,
and
38
atoms.
The atoms
of a molecule so defined can obviously
have both vibratory and rotatory motions.

The interatomic spaces have been supposed to contain ether
atoms which are much smaller and which repel each other
while they are attracted by matter
they will thus form atmospheres around the material atoms, the density of which decreases from the centre outwards.
Professor Rankine uses
:
the term atom to signify the central nucleus together with its
atmosphere: M. Redtenbacher coins the special name dynamid. The movements of the particles of ether must obviously
be confined within the boundary of the atmosphere.

Thus when radiation falls on a body, the atoms, their atmospheres, or both
its
in
atoms
may be
set in
motion and the body is warmed

can put the external ether
:
too, or their atmospheres,
motion and the body then

Professor
Rankine
radiates.
in his hypothesis of molecular vortices
considers that a body is solid when the atmospheres are so

about the nuclei that their mutual repulsion
much condensed
is
very small compared with the attraction between the nuclei

it is
liquid when the atmospheres are less condensed, so
:
that
that the attractive action
a cohesive force
between the nuclei produces merely

balance the repulsion between the
sufficient to
and that in a gas the action between the nuclei

atmospheres
almost insensible.
;
is
Professors
Clausius and Maxwell, working out M. Ber-
noulli's idea, reject the notion of atmospheres, the ether atoms

not differing except in mass from those of matter, and consider
the difference of physical state to depend on the distances
between the molecules: in the solid state they are so close
together as not to be able to move far away from their mean
positions, so that their
motion
is oscillatory;
in the liquid state
one molecule can move out of the sphere of attraction of

another, but only to come into the sphere of attraction of a
third, thus moving with an irregular progressive motion ; in the
gaseous state the molecules are so far apart as to exercise

but slight actions on each other, and they therefore progress in straight lines until they collide against each other or
come
into
39
each other's sphere of attraction and take
new
directions.
Sensible heat.
28.
Sir
H. Davy,
Rankine
Professor
that the sensible heat
considers, with
s of a body
is
the kinetic
energy of the atmospheres the particles of which are in oscillatory or progressive motion, and to connect this hypothesis
with the undulatory theory of radiation he introduces the further
supposition that radiation is effected by the motion under their
mutual actions of the nuclei vibrating independently of their
atmospheres.
On
other theory the sensible heat is the total kinetic

of
the
this is a more complicated quantity, as
energy
system
the kinetic energy of a molecule is the sum of the kinetic
the
energies (i)'of the vibratory or rotatory movements of the

atoms themselves, (2) of the rotatory movement of the molecule
as a whole, (3) of its motion of translation, (4) of the motion
This is
of any molecules of ether that may be entangled in it.
the preferable assumption.
29. Internal work.
As
the atoms in any molecule attract
each other, and also one molecule attracts another, work must
be spent to increase the mean distances either of the atoms
or the molecules from each other
this
work
is
however not
the atoms or molecules by reason of their mutual actions are able to do just as much work in returning to their
lost, for
original configuration
the
work
is
therefore stored
up as po-
energy of molecular separation. The work that is done

against molecular forces only in increasing the mean distances
tential
of the atoms or molecules from each other and in overcoming

molecular forces when the motions are increased is called the
work and denoted by /.

body continuously receives heat, it finally assumes the
and as changes of density in this state
perfectly gaseous state
entail no internal work whatever the temperature
(as shewn
in
it follows that the same amount of work 3 must be
63),
done against molecular actions to bring a given body from the
state in which it is absolutely cold
in which its atoms and
molecules are absolutely motionless and in such positions that
internal
If a
40
,U
the attractions and repulsions balance each other
which
in
it
amount of
the
is
maximum
work
that
When
action of external forces

if
this
is
the
can be done in the body or
potential energy of molecular separation that
can possess.
30. External work.
them
and
perfectly gaseous,
internal
^ states
maximum
into
it
a body expands against the
does external work in overcoming
it
on the contrary
is
it
contracts
and the external
forces
are such as tend to cause compression, then these forces do
work upon it. We shall denote

body does work.
it
by
IV,
which
is
positive
when
the
GENERAL EQUATION OF EQUIVALENCE.
We
can now construct an equation expressing the First

of Thermodynamics, and it will be simpler if we measure
quantities of heat dynamically, that is, in megalergs, rather than
in calories, as by this means we avoid the introduction of either
31.
Law
or
into our expressions
Professor
direct use
this is the
method adopted by
obviously cannot make

of the tables of different heat quantities, such as
Rankine.
In
this
case
we
and latent heats, that have been constructed by M. Regnault and other experimenters; we must first multiply the
numbers by 41-55 to reduce them from calories to megalergs.
specific
We
shall consider the modifications in the
molecular motions
if change also is produced by heat in the

only of a body
motion of a body as a whole its visible or external motion,
:
as
M. Zeuner terms
we
it
which
shall leave out of consideration
concerned with such change,

except in special cases.
Consider heat dH communicated to unit-mass of a body
that part of the heat
is
there ensue in general

(1) An increase dE of the intrinsic energy of the body:
(2) External
(3)
bodies
(4)
work
dW done
communication
against external forces

energy dv to
of kinetic
external
An amount
of
heat
dR communicated
bodies by radiation, conduction, Y&c.
to
external
Hence
the conservation of energy gives
dH=
This
41
dE + dW+dV+dR.
(i)
most general equation, but we
the
is
shall generally
by supposing dR = o, or that the body is never

whose temperature is not infinitely near
own, and dv = o, or that all changes of volume occur with
it
simplify
in presence of others
its
exceeding slowness
stress exerted
this latter
condition requires also that the
by any part of the body should never
differ
more
than infinitesimally from the external pressure acting upon the

same part. We shall consider too that no part of the external
work
dw takes
the form of electric or magnetic energy.
dH
thus have
For
hand
dE
solid bodies
is
insensible.
Of
or negative ; thus
of gaining it, and
the
(2)
term on the rightbeing insignificant ; dw is some-
dw
but in gases it is dE that is nearly

course any of the above quantities may be zero
for liquids
if
We
far the larger
by
side of the equation,
what larger
negative
dE-^-dW.
dH is
o
it is
if
negative
if
there
body contracts
the
body
loses heat instead
is
neither gain nor loss
phenomenon which
occurs on the subtraction of heat, but also on
dW is
generally
the addition of
heat in the case of water between o and 4c., of certain aqueous

solutions, and of silver iodide between 10 and YOC.
dE
is
a perfect differential, since
depends only upon

its
attaining
between the
that
its
actual state,
state;
but
dw
the energy of
a body
and not upon the mode of

depends on the relations
and the expansion at the different points

of the body, and thus not only on the initial and final states
of the body but also on the whole series of states through
which it has passed
hence dH is not a perfect differential,
and for the integration of equation (2) a relation connecting
stress
the stress
is
on
this
and expansibility
point that
wrong from
and
their
ential
their
must be given. It
and Poisson have gone
at every point
MM. Laplace
acceptance of the material theory of heat

.that dH is a perfect differ-
consequent assumption
the true case
Given the necessary
was
first
relation,
by M. Glaus i us.
by integration
W.
(3)
distinctly stated
we
get
H=E-E.+
42
According to Professor Clausius' hypothesis,
made up
general
(1)
An
dE
in
is
of
increase of the molecular kinetic energy, or an
increase of sensible heat
ds
(2) Internal work dl done against molecular forces, partly

in augmenting the kinetic energy of the molecules and their
atoms (whether
this is done by an increase in their amplitudes

of oscillation or a superposition of new forms of vibration) and
partly in increasing their mean distances from each other
:
whence
becomes
dff=dS+df+dW,
(4)
wherein ds and dl would both be perfect differentials. dS

o
during fusion and vaporization under constant pressure, as the
temperature then remains constant.
1
Following Professor Clausius further
equation by writing
dD
for
we may
df+ dW, and we
simplify this
get
dHdS^-dD.
dD
(5)
work of disgregafion, and represents on this

the
whole
work done, both internal and external,
hypothesis
the
during
given change of volume it is obviously not a perfect differential.
We shall consider this method later in 59.
called the
is
We
shall in general suppose, unless express
otherwise,
the
that
stresses
mention
is
made
simply of a hydrostatic
points if then / is this pres-
consist
pressure which is the same at all

sure pef-ttnit area: and v the specific volume of the substance,
pdv, and the relation between simultaneous values of
:
dW
p and
v must be given.
We
W=
have then
Fpdv,
(6)
JVQ
32.
The loci of the points on the characlines.

surface which represent the successive conditions of a
that alters its state under given circumstances are called
Thermal
teristic
body
thermal
fessors
lines.
We
distinguish with special
Rankine and Gibbs)
names (due
to Pro-
the following which are the
most
important.
1
Pogg. Ann.
cxvi. (1862) 73, trans, Phil.
Mag.
[4] xxiv.
(1862) 81.
43
Adiabatic lines, or lines of no transmission, are the thermal
denoting the states of a body that
lines
communication or abstraction of
heat.
is
altered without
As no
vessel
can be
made
perfectly impermeable by heat, these lines can only apIn this case
o,
proximately be determined by experiment.
H=
and equation
E _ Ej
w=
(3) gives
(y)
or the substance loses intrinsic energy equal to the external

work done.
Isopiestic lines, or lines
represent the states of a
of equal pressure, are the curves that

body when its pressure is constant
:
orthogonal projections on the plane of pv are therefore

and
always parallel to the axis of v. In this case / = /
their
equation (6) gives
These
a
=A (_.).
lines are also called isobars.
Isometric lines, or lines of equal volume, denote the states of

in which its volume is constant
their projections on the
body
plane of pv are s*Wa!?y always parallel to the axis of/.

this case dv
o, or
all
the
In
w_
therefore internal. T*ese> &/>
work done being
Isothermal lines, or lines of equal temperature, connect the

body wherein its temperature is constant.
different states of the
As an
increase of pressure must produce a diminution of volume,

otherwise an unstable condition will be realized, Isothermals are
convex towards the origin of co-ordinates. On the above hypowe have in this case dS
o, and equation (4) reduces to
thesis
dH
dl+dW.
If the substance is in the saturated state the isothermals coin-
cide with the isopiestics.

Isenergic
lines,
or lines
M. Cazin isodynamic
in this case therefore
or
all
of equal
lines), are
the heat put into the
The
dE =
o;
body is spent in external work.

no energy, coincides with the
E = o, both 5 and / must be zero
zero isenergic, or line of

zero isothermal ; for, when
too, as neither
energy (called also by
given by the equation
can be negative.
44
Isentropic lines,
or lines of equal entropy, denote the suc-
cessive states of a
body wherein the entropy remains constant.
From
the definition of this property in
it
follows thatthe
isentropics coincide with, the adiabatics.t^^ss^^^tHr/onêT'

f <*&*rMi*c(&t a(*o fte- *rf<W *je*in>nbeM * -s*u/>likThe above names are generally given, not tcrthe actual lines
we have
described, but to their orthogonal projections on the

plane of pv ; for by these the work done during the corresponding transformations may be at once graphically represented
as in
jections
hence we
convenience denote the pro-
shall also for
by the above terms,
characteristic surface the real thermal lines.

line is defined
line
is
the
by
F (p, v) =
intersection
o, the
of the
F(p,
two
real thermal lines of the
other, as they imply
If then a thermal
corresponding real thermal
cylinder
generating lines are parallel to the axis of

teristic surface of the substance
f(p, v, /)
No
on the
calling the actual curves
/,
v)
= o,
whose
with the charac-
o.
same type can
intersect each
an essential difference in the states of the
hence, if we are given a model

of the characteristic surface of a body with its real thermal
body characterised by them

lines traced thereon,
we can always immediately determine
its
pressure, density, temperature, energy or entropy corresponding

to a state defined by any two of them
for the intersection of
:
two curves characterising this state determines a point

through which must pass all the curves corresponding to the
the
The same
generally true if we are given a

complete diagram of the projections of these curves on the
or, according to the above nomenclature, a complane ofpv
other qualities.
is
diagram of the thermal

however shewn that in certain
plete
lines
cases,
Professor Riicker has
which we
shall
examine
44, two isentropics may cut each other, and

therefore also two thermal lines of any. the same typeJ?" **"V*
cfloftttl***, no ther*L (W* i^tfrêtt c**h| W*r
tafVe*
i he ?;-axis is
asymptotic to all isothermals and adiabatics
at
length in
for indefinite expansion brings every
body approximately
into
the perfectly gaseous state, in which p is infinitely small for

infinite values of v
it is
asymptotic also to all isenergics, since
these in the case of a perfect gas coincide with the isothermals,
:
as shewn in
63.
Adiabatics.
33.
thermal
it
is
for
if
7,
volumes
Where an
adiabatic
a substance that expands
its
work spent
formed into
crosses an iso-
inclined at the greater angle to the axis of
temperature rises is enclosed in a

vessel and then slightly compressed, the
as
45
in the
heat,
compression is transwhich is retained if the
walls are perfectly impermeable
by heat
and, by tending to increase the volume of

the substance, increases
its
resistance to
Fig.
7.
compression and causes a greater increase

of pressure than if it were allowed to escape and the temperature
to remain constant ; and again, if the substance contracts as its
temperature rises, compression lowers the temperature, and the
heat that enters in consequence from without, if the walls are
permeable by it, tends to produce a greater compression than if

1
the walls were perfectly non-conducting.
Professor Maxwell
ntfc
<wtskews
for
that this
any
greater force
is
no
is
an
of the general principle that
illustration
alteration in the state of a

is
body
subject to constraint
when there
required than for the same alteration
constraint.
Rankine 2
has applied the properties of adiabatics

the graphic representations of the energy of a body, the
heat spent in any operation, &c.
we give the most important
Professor
to
of these, which depend on the following theorem.

Suppose a body initially in the state A (Fig. 8) to expand
without loss or gain of heat, that is, along the adiabatic A a,
doing work against external pressure that is at all times only

than its own. Now when the volume is
infinitesimally less
work that is done by further expansion

Ov is asymptotic to A a; hence the area
bounded by the ordinate A a, the adiabatic A a, and the zero
isopiestic Ov, has a finite value, and represents the maximum
amount of work that can be done by the body, or, by equation
(7), its maximum change of intrinsic energy, when it expands
infinitely
large the
is infinitely small,
as
Theory of Heal, 4th ed. (1875),

Phil. Trans, cxliv. (1854) JI 5-
I 2> 1
4-6
as above.
But, the expansion being indefinite, the body finally
assumes the perfectly gaseous state, in which
changes of density
entail no internal work (as shewn
in
hence the external
63);
work
is
ultimately
done
entirely
at the
expense of the molecular

kinetic energy, that is, the temperature of the body, which therefore
becomes
finally
infinitely
small, so that the intrinsic energy
of the body,
Fig. 8,
after infinite
sion adiabatically,
potential energy
that, if
is
29).
(see
Thus we
the intrinsic energy of the

area.
AavaA
=E
is
arrive at the
body
expan-
simply
its
theorem
in the state A,
3,
or the intrinsic energy of the body in the state A is represented

AavaA and a constant area equal
by the sum of the area

to 3.
assumed by a mistake that the intrinsic energy of

body in its infinitely extended state and cold is zero instead
of 3 this mistake does not however affect the following results
as they depend on differences only of the values of E, or on difIt is often
the
ferences of areas such as
AavaA,
so that
always consider areas to be positive

direction of motion of the hands of a watch.
shall
3 disappears.
if
We
traversed in the
(1) The change of intrinsic energy of a body in passing from

one state to another.
the intrinsic energy is AavaA + 3, and at B it is

BfivbB + Hb'. the change of intrinsic energy in passing from
At
to
area
is
therefore the difference between these areas, or the
AabBfiaA, which is independent of the path AB.

more than infinitesimally from the
results will not differ

if
we suppose
same
that
all
Our
truth
adiabatics touch the zero isopiestic at the
point.
The heat required to transform a body by any path

one state to another.
(2)
This
is
from
represented by the area included between the path-
47
curve and the adiabatics through the initial and final states
for in the path AB it is the sum of the increase of energy
:
AabBftaA and
A B fta A,
the
work done ABbaA, and
this area is positive, thus
diagram
is
thus the area
regard being paid to the signs of the areas.
In the
denoting heat absorbed (the
to a higher adiabatic), and it lies

of
the
direction
of change AB ; if the direction
right
body passing from a lower

to the
of the change were from
negative area BAafiB which

emitted: hence
to A, heat represented
lies
to the left of
BA
(3) To determine in any transformation whether heat

or emitted.
Draw
state
if
if
is
by the
must be
absorbed
the adiabatic through the point representing the initial
the path-curve lies above the adiabatic, heat
is
absorbed ;
under, emitted.
(4)
The heat supplied when

states are on the same
and final
This
the points representing the initial
adiabatic.
the area between the path-curve
is
and the
adiabatic,
as follows at once from (2) when the second adiabatic is made

to coincide with the first this area is positive if the body passes
:
from the curve
to the adiabatic in the direction
of the hands of
a watch.
The above method

has removed
this
requires an infinite diagram
M. Cazin 1
objection
by introducing isenergics a
slight complication which is
a convenience on
however
Let the isenergic
the whole.
through
the
initial
state
cut the adiabatic through the
final state
then the area
in the point
ABbaA
repre-
work done during

the passage of the body from
sents the
to B, the area
BCcbB
Fig.
9.
represents the change in
its intrinsic
Theorie elementaire des machines a air chaud. (1865).
48
energy, and the area ABCcaA the heat that has been supplied
along the path AB. For
E=
BCfivbB + 3,
E = C(3vcC+3, since C is on the isenergic through A
hence E E = BCcbB. We may further note that
is not
and
it
necessary to draw the particular adiabatic that passes through B

provided we can draw the isenergic that passes through it for
;
if
any other adiabatic B'c'
is
drawn, the area E'c'cb'tf equally
represents the intrinsic energy acquired by the
from
to B, or, indeed, in passing
isenergic to any point
on the
determine the real phenomena.
other.
body
in passing
from any point on the one
The
signs of these areas
CHAPTER
IV.
C A KNOT'S PRINCIPLE.
34.
Heat engines.
tinuously transformed into
machine by which heat can be conwork is called a heat engine. We
have seen that when heat
is given to a body an equivalent

amount of work is done, partly internal and partly external, but
work cannot be continuously performed without a reciprocating
motion of some kind in the engine air, for instance, enclosed
in a cylinder by a piston and heated will expand and press out
the piston, thereby doing work, but no more work can be done
when the piston has reached the end of its stroke till by some
means it has been brought back to its former position. For
this action work must be spent by the engine if it is selfacting, so that useful work is not done in amount corresponding to the whole amount of heat communicated
part of this
heat is therefore untransformed and must have been given up
as heat to other bodies if the air is brought back to its original
:
condition, or, as
been performed.
it
is
a complete
said, if
cycle
of operations has
The
bodies from which the working substance of the engine (in the above case air) receives heat are
called sources, and those to which it gives up heat are called
refrigerators ;
these terms are often replaced by furnace
and
condenser in the case of vapour engines.
Carnot's cycle. The simplest cycle for an engine

1
perform is that devised in 1824 by M. Carnot and though
cannot be practically realized, our reasonings are not thereby
35.
to
it
1
See Clapeyron's Memoire de la puissance motrice de la chaleur, Journ. de
VEcole Poly. xiv. (1834) J 53> or Thomson's An Account of Carnot's Theory,
<Sec.
Trans. R. S. E. xvi. (1849) 54 1
Carnot's original tract, Reflexions sur
la puissance motrice du feu et sur les machines propres a developer cette puissance,
is not procurable
it has however been reprinted in Ann. de VEcole Norm.
-
[2]
i.
(1872) 393.
invalidated.
Heat
is
supposed to be taken
by the working
in
substance of the engine always at one constant temperature

and to be given out always at another constant temperature:
thus there are one source and one refrigerator only whose
temperatures are not to vary with the losses or gains of heat

that they experience
a condition which is realized by a vapour
in contact with
its
liquid at constant pressure, or
when
the
source and refrigerator are very large; the engine moreover

must be so constructed and of such
materials as to prevent loss or gain
of heat except
these tempera-
at
s
M. Carnot's
tures.
hypothetical
engine consists of a cylinder the

piston and sides of which are perfect non-conductors while the bot.
tom
a perfect
is
conductor,
the
whole possessing no capacity for
Fig. 10.
heat
are formed of two stands
and
the source
and
refrigerator
and C
respectively,
is
third stand of perfectly non-conducting material which has no

If the cylinder stands upon A or B, the
capacity for heat.
temperature of the substance will be constant however its state

changes ; and if it stands on C, no gain or loss of heat is possible.
Let now unit mass of any substance, enclosed in the cylinder,
be
at the
temperature
of the re-
frigerator and be represented as to

pressure and volume in its initial
state by A in the diagram.
operations are as follows.
(i)
__
ae
it
Spend work
in
without loss of heat
The
compressing
till
its
tern-
I/
perature rises to T, that of the

Fig. IT.
source: during this process the cyIf Q and
-are the
linder rests on the non-conducting stand.
l
intrinsic
energy in the
spent, equation (3),
initial
and
31, gives
final states
and W\ the work
5T
and
AB.
the successive states of the
(2) Let
H from
it
body are given by the
expand and do external work
the source so that
its
temperature
W^
is
adiabatic
receiving heat
always constant
T: during this process the cylinder rests on the

source-stand, and the successive states of the body are represented by the isothermal BC.
If E 2 is the intrinsic energy in
and equal
to
the final state C,
we have
H = E -E +W
2
(3) Let
it
expand
receiving heat until

refrigerator
and do external work
further
again to
falls
its
without
that of the
t,
temperature
CD denotes the successive states of
the adiabatic
the body during this operation, in which the cylinder is again

upon the non-conducting stand. If the intrinsic energy in the
D is
final state
then
= E -E +W
o
(4)
Expend work
in
compressing the substance
at the
constant temperature / until it returns to its original condition

in this case the
after giving up heat h to the refrigerator
:
cylinder is placed on the refrigerator-stand, and the successive

have
states of the body are given by the isothermal DA.
We
_ = E -E - W.
therefore
Adding
together these four equations,
Hh
we have
W^W^ W - W
3
= area A BCDA =
w,
work done, no change of intrinsic energy on

the whole having occurred, since the body has performed a
the total external
complete
cycle.
M. Carnot, assuming, though not without doubts, the material theory of heat, took h = IT, since on the whole no heat is
gained or lost by the body in the cycle, and considered the work
done as due
fall of heat from a higher to a lower temwork may be done by the fall of water from
to the
perature, just as
a higher to a lower level
his directions therefore for the fourth
operation (in his order the third, as he started from B) were to

compress the body until as much heat has left it as entered.
E 2
52
Dr. Joule having conclusively established the dynamical theory

of heat, Professor Clausius 1 first rectified M. Carnot's mistake by directing the compression to be continued so far along
DA
the isothermal
until further
would bring the body
compression along an adiabatic

We have followed Professor
to state B.
Maxwell's order of the
operations, and thus avoided all chance

and difficulty of expression.
That h cannot be equal to
seems clear from the con-
of error
when
sideration that,
from a body
totally transferred
is
at
a higher to a body at a lower temperature by conduction, no

work is done; thus the circumstances ought to be altered if
work
done and again, if the engine were employed to agitate

heat
would be generated, as Dr. Joule's experiments
water,
shew but this heat would be created out of nothing, were h
is
equal to H.
direct
each stroke of the piston

at
temperature ec.
condenser
this inequality was first given

the steam-engine 2 in 1854.
For
boiler
of
leave
the
grams
vapour
proof of
by M. Hirn's experiments on
the boiler being fed with water from the
at 0c., this
vapour according to M. Regnault's exfrom the
m {606-5 + 0-305(9 0)} calories

H= Jm {6o6-5 + o-305(e 6)}. To
periments absorbs
furnace: thus
condenser
at
3c.
are injected at each stroke
hence h
= JM(Q
It
keep the
temperature 0c., M grams of water
at the constant
and absorb
M(Q
was always found
that
$) calories;
is
greater
Him
of the experiments M.
calculated the heat R lost by radiation and conduction (a very
uncertain determination) and the work
done in each stroke.
than h] but to
test the correctness
The
latter is the integral
JQ
(Pp)u>dl, where
oo is
the area of
the piston, / the length of its stroke, P, p the working and resisting pressures on the faces of the piston due respectively to the
vapour that has entered the cylinder and the vapour in the condenser, the latter being generally called the back pressure: this
work
is
however more conveniently measured by the area of the
indicator diagram, or the indicator card, as

'
Ann.
Ixxix. (1850) 380, trans. Phil.
it
is
Mag.
Theorie mecanique de la Chaleur, 3 eme ed. (1875),
technically called.
[4]
i.
ii.
54.
(1851) 9.
53
H=h+ W+R
he found 40-51 as the mean

the equation
value of J\ this agrees sufficiently well with Dr. Joule's number, especially considering the difficulty of the experiments.
From
36. Closed cycles. The general case, wherein the lines of

the diagram are any whatever and not
merely isothermals and adiabatics, admits
of the same treatment.
of state from
to
working substance
takes in heat
If
gives
during change
mass of the
unit
up heat
Hv
during the process rethe propresented by BC, and

3 during
cess CD, giving up ff4 in its return to A
2
along DA, the work done
u
pj^ I2
is
represented
by the area A BCDA, and the heat spent in this work
where
is
represents heat taken
in, being reckoned negative if

Thus the most general expression of the first
given
law of thermodynamics for closed cycles is
out.
fdH=
A
closed cycle
is
termed simple
two types only, and complex

thus
Carnot's
cycle
is
W.
if its
bounding curves are of
the curves are any whatever

that of Fig. 12 is comwhile
simple,
if
plex.
Being given any number of successive operations in each of

which the initial and final states are identical, we may always
consider them as forming part of one
For considering two such
closed cycle.
by the closed curves

DEF-D, we may join them by any
series represented
ABCA,
arbitrary
differ
line
CD',
then together they
from the closed cycle
ABCDEFDCA
only by the operations represented by
CD and DC
which are exactly inverse
to
Fig. 13-
each other in every respect, the difference being thus nothing.

We may also break up any given closed cycle into any
number of
others
formed both
directly
by introducing arbitrary operations perand inversely. Thus the cycle ABCDA is
54
equivalent to the two cycles
ABDA, BCDB together, the arbitrarily
introduced operation BD being performed

both directly and inversely and having
therefore
no influence on the whole.
37. Reversible engines.

said to be reversible
diagram
is
when
cycle is
the indicator
accurately retraced in the opif the operations are ex-
posite direction
Fig. 14.
actly reversed.
Carnot's
for considering the

allow
it to expand along the
(i)
adiabatic BA doing work without receiving heat while falling in
temperature from T to /, (2) by giving heat h and maintaining
body
cycle
initially at
B,
fulfils
this condition;
we may
its
AD,
along the isothermal

temperature constant bring it to
(3) by compression without loss of heat along the adiabatic
DC raise
its temperature again to T,

(4) by further compression
along the isothermal CB restore it to its initial state after taking
away heat H. Thus work W, measured by the area ABCD, has
been done on the body which has on the whole

these being equivalent, we have
- (HK) =
area
BADCB
lost heat
Hh\
-W,
the negative signs denoting loss of heat and work done on the
There is this difference only with respect to perfect
body.
the body is slightly cooler than the source when
reversibility
:
it
receives heat
from
it
as in the direct cycle, but slightly
warmer when it gives up heat to it as in the reversed cycle

and similarly it is slightly warmer than the refrigerator in the
if however
direct and slightly cooler in the reversed process
the expansions and compressions are performed slowly enough,
these inequalities may be indefinitely diminished and a perfectly
;
reversible cycle obtained.
We
commenced
the above process at
instead of A, so as
to avoid the difficulty of expression noticed in
35.
principle. We have seen that for an engine

to supply continuous work, there must be not only a transformation of heat into work but also a transfer of heat from a body
38.
Carnot's
at a higher to
one
at
a lower temperature.
If
is
the
work
done when heat
is
is
is
55
abstracted from the source, the ratio
termed the Efficiency of the engine ; and Carnot's Principle

that of all engines with one source and one refrigerator at
given constant temperatures the efficiency of a reversible engine

is
the greatest.
M. Carnot's proof
on the
rests
impossibility of the per-
of obtaining useful work at no expense of

petual motion,
on
the
caloric theory, if any engine A is more
for,
energy:
i.e.
efficient
than a reversible one
JR,
so that
If
>
or
W' >
W,
where W'
represent the work the engines do respectively
while transferring heat
from the same source to the same
f
will be obtained
refrigerator, a balance of useful work
,
W W
without on the whole any transfer of heat when the engines are
coupled inversely so that R restores to the source exactly the
H which
heat
abstracts
from
it
or the perpetual motion
is
realized.
Professor
ciple
still
rests
it
Clausius
holds
demonstrated in 1850 that this printhe dynamical theory of heat

he
good with
on the following axiom, not
distinctly
formulated
till
1854, which he calls the Second Law of Thermodynamics :

Heat can never pass from a colder to a warmer body without
some other change, connected therewith, occurring at the same
time ; or
Heat cannot by
itself
pass from a colder
to
a warmer
body.
Sir
form
W. Thomson 3
//
is impossible,
1851 put this into a more precise

means
by
of inanimate material agency,
in
mechanical effect from any portion of matter by cooling it

below the temperature of the coldest of the surrounding objects.
This latter axiom the impossibility of getting work by cooling
a condenser seems the safer one to assume on general exto derive
perimental evidence; the modification of M. Carnot's proof

of the above principle required in either case is very slight.
Professor Clausius' modification is the following: if
any
1
2
3
Ann. Ixxix. (1850) 500, trans. Phil. Mag. [4] ii. (1851) 102.
Pogg.
"*
TA
/_ O - .\ .OO
v*1
n-r
^ x-v ^ +
Pogg. Ann. xciii. (1854) 4 88 tf ans. Phil. Mag. [4] xii. (1856) 86.
Trans. R. S. E. xx. (1851) 265, or Phil. Mag. [4] iv. (1852) 13.
A.
56
engine
w w
H >H
more
is
#>#',
or
than a reversible one R, so that
efficient
H H represent the heat abstracted

f
where
from the same source by the respective engines to perform the

with the same refrigerator, then an amount of
same work
heat
HH'
be transferred from the refrigerator to the
will
source without on the whole any expenditure of work
engines are coupled together inversely so that

and this is contrary to the axiom.
W. Thomson's
Sir
efficient
modification
than R, so that
w'
--
H >H
or
this:
is
if
W >W
f
when
the
just drives
more
is
where W',
work done by the respective engines with the

same refrigerator on abstracting heat H from the source, then
f
will be done without any loss of
an amount of work
represent the
W W
heat to the source

that
by
R
;
when
this
work must
the engines are coupled inversely, so

source exactly the heat
abstracted
restores to the
therefore have been
done
at the
expense
of the heat of the refrigerator, which is contrary to the axiom.

Taken in its strictly literal sense Professor Clausius' axiom
from a colder to a hotter
that heat cannot of itself pass
body is perfectly true; thus in Carnot's reversed cycle the

working body is just colder than B when it takes in heat and
just hotter than
A when
it
gives
it
out
but
it
is
not
strictly
correct to call the transference of energy (though taken in by

the body as heat and given out as heat) from
to A a transfer-
ence of heat from a colder to a hotter body, as we know not

the form in which the energy exists during the body's adiabatic
change of condition.
The
assert the impossibility of
statement
compensation,
and
axiomatic
the theory of entropy
if
this,
as Professor
also points out that Sir

restrictions, as
is
however intended
to
any such transfer of energy without
Maxwell
is
so.
points out, is only

Professor Maxwell
W. Thomson's axiom
requires
some
in a cylinder at the
by allowing air, compressed

temperature of the room, to expand we could get work from
though its temperature would fall below that of the coldest

of surrounding objects, and that with such restrictions the
it
axiom would merely
re-assert
Carnot's
57
principle with greater
In fact the second law of thermodynamics is shewn

by the kinetic theory of gases not to be true for a group
of molecules unless they are sufficient in number to form a
generality.
sensible part of a body,
a mathematical truth,
axiomatic statement.
The
so
that
it
is
statistical
and not
and therefore not capable of
great importance of
Carnot's
principle lies in
strictly
its
shew-
engines have the same efficiency: but

reversibility is entirely independent of the working substance
of the engine hence the efficiency of a reversible engine is
independent of the working substance and depends only on
ing that
reversible
all
the temperatures at which heat is taken in

Reversible engines are therefore called perfect.
Thus
in
whence,
if
small, or
Carnot's
cycle
and given
out.
we have
the difference between the temperatures
is infinitely
= TdT,
w=
F(T, T-dT] = F(T, T}-F'(T, T)dT
si
since
F(T,T)
is
zero, as
no work can be done without a

F' (T, 7') is
quantity
differ-
The
ence of temperature.
Carnot's Function, and
is
denoted by
ft
it
is
called
a function of
the temperature of the source only.
39.
Work
in reversible engines of finite range per-
forming Carnot's cycle.

versible
engine of
finite
By 36 we may consider a rerange to be made up of an infinite
number
of reversible engines of infinitesimal range, the refrigerator of each being the source of the next and the heat
rejected by
one being that which
on the whole
there
is
limiting temperatures,
If in
HdH
taken in by the next
and the same amount of work
for
is
done.
taken in at temperature T
at
the first law
temperature
rejected
one of these engines heat
and heat
is
neither loss nor gain of heat except at the
is
Tdr,
5$
gives the expression dff,
pHdT,
and the second law the expression

amount of work done. Equating
for the infinitesimal
these values
we
get
,///
(!)
we extend
over the whole range 3

By integration
thus if T, t are the extreme temperatures, H, h the heat taken
ip
and given
this result
W the work done,
out,
(3)
This
is
Sir
done by a
W. Thomson's
reversible
work
general expression
Carnot's
of
finite
engine
range performing
for the
ju is a function of the temperature which can only be

determined by experiment the best method will be indicated
in
But the form of the function will vary with the ther71.
cycle,
mometric
scale employed, for, as
scales in actual use
ever has
is
we have
made Carnot's theorem
the
seen,
none of the
W. Thomson how-
Sir
really absolute.
basis
of two
scales
perfectly absolute in the sense of independence of the thermometric substance these we now give.
:
ABSOLUTE SCALES OF TEMPERATURE.

40. First absolute scale of temperature.
W.Thomson
proposed
In 1848 Sir
to consider the difference of the tem-
peratures of the source and refrigerator constant when the work

done by a perfect engine on abstracting unit heat from the
Now the work done is by (3) constant
source is constant.
+T
\Ldt is constant, where /+T and / are the
/t
.
and on the proposed scale this integral

;
be independent of / and a function of r only. This requires
to be constant
for the differential coefficient of the integral
limiting temperatures
will
/u.
with regard to
1
(which must be zero since the integral
Trans. R. S. E. xx. (1851) 273, or Phil

Mag. [3] xxxiii. (1848) 313.
^Phil.
Mag.
[4] iv. (1852) 106.
is
/
i
independent of
the difference between the values of
is
/)
corresponding to the temperatures

are therefore equal ; but r
ju must be the same for
that
In
/+T and
may be any
if
we
=a
also equation (2)
write b
//,
and these values
quantity whatever, so
temperatures.
equation (3) becomes
this case therefore
W = !-e~^=
if
/,
all
-a*,
fi
-M
59
= e* = -
As
becomes
therefore the difference in tempera-
ture of the source and refrigerator of a perfect engine changes

in arithmetical progression, the ratio of the amounts of heat
taken from the source and given to the refrigerator changes in

geometrical progression.
It is
found by experiment, indicated
in the next article, that
the efficiency of a perfect engine working between o

is
hence
if
we
consider these temperatures the arbitrary
on
points which coincide
we have
=
and therefore a
This scale
is
and iooc.
the absolute
100
0-99689,
found to
differ
and centigrade
scales,
=
1-00312,
//.
0-003114.
very greatly from that of the
air
thermometer, more and more degrees of the latter going to one

of the former as the temperature rises
and again, the absolute
zero of temperature on this scale is
oo ;
for, were it the
:
o (as all bodies absolutely

temperature of the refrigerator, h
cold are entirely deprived of heat), and therefore r is infinite,
or absolute zero
source
may
below any finite temperature that a

both accounts then this scale is very in-
is infinitely
have.
On
convenient and has therefore yielded to another absolute system.

41. Second absolute scale of temperature.
In 1845
M. Holtzmann,
that
jui
is
starting
from incorrect
inversely as the temperature
thermometer, and
in
principles,
concluded
from zero on the
air-
1850 Professor Clausius arrived at the
60
same conclusion by applying
his
amended form of Carnot's
principle to the case of gases in a hypothetical perfect state.
But
this scale is
not
strictly absolute, as
it
depends on the
state
of the gas used and does not rest entirely on thermodynamic

defining temperatures however with Sir W. Thomso that Carnot's function at any temperature is the
principles
son 1
reciprocal of that temperature

absolute.
we
obtain a scale that
is
perfectly
This definition moreover appears natural;
for in
equation (i), as temperature is a definite magnitude, homogeneity requires /* to be the reciprocal of a linear function of T.
Taking then
jut
we transform
equations (2) and (3) into
H= T
T 7'
W T-t
and
Hence
Sir
W. Thomson remarks 2
two temperatures are
that 'the absolute values of
one another
in the proportion of the

heat taken in to the heat rejected in a perfect thermodynamic
engine working with a source and refrigerator at the higher
to
and lower of the temperatures
respectively.'
This scale admits of a very elegant graphical expression.
Draw
the isothermal for any temperature
the
7 and
1
let
represent
mass
state of unit
initial
of the working substance of a

perfect engine,
C,
cessive states after
. .
cation of heat H, zH,

the
Through A, B,
if
any
abc ...
perfect engines
C, ...
other
is
as
Ccy,
isothermal
drawn, the
they
...
at
T.
draw the
A a a, Bbfi,
ABbaA, BCcbB,
equal,
work done by
,.
constant temperature
adiabatics
Fig. 15.
suc-
its
communi-
...
as
areas
will all
measure
be
the
working between the same tem-
Phil. Trans, cxliv. (1854) 35 1 Trans. R. S. E. xxi. (1854) 125, or Phil.
Mag.
[4] xi. (1856) 216.
61
peratures
source.
and abstracting the same amounts of heat from the
Draw
the isothermal a(3y
, . .
corresponding to
total
deprivation of heat or to absolute zero temperature if this

the temperature of the refrigerator, the whole of the heat
;
taken in
BCyfiB,
at
is
are
...
is
transformed into work, or the areas ABfiaA,

equal to H. Divide these areas into T equal
all
TT
parts
by a
series of isothermals,
each part being therefore
W. Thomson's scale of absolute temperatures is

If abc ... is the isogiven by these successive isothermals.
thermal corresponding to temperature /, there are T t of these
then Sir
small areas in
ABbaA, or ABbaA
work done by a
taking in heat
at the higher
therefore
now ABbaA
working between
perfect engine
=H
temperature;
its
is
and
the
and
efficiency is
the expression found before.
Rankine arrived in 1851 at a similar expression

terms of the absolute temperatures on the air-thermometer,
without assuming a
deriving it directly from the first law
1
Professor
in
second
on
his peculiar hypothesis of molecular vortices.
it is
degrees of the new scale are arbitrary
convenient to determine them by drawing the isothermals for
Thus
far the
oc. and iooc.,
dividing the area between these lines
and con100 equal parts, and taking one of these

or area between consecutive isothermals.
secutive adiabatics into
parts as unit area,
Then
if
AB
that for temperature T.
is
the area
ABfiaA
contains
such areas, the isothermal
Experiment shews that the absolute zero temperature
is
about
273-72 c. ; the reason of the approximate coincidence of

this zero with that of the air-thermometer, which
suggested the
scale, will
appear in
be described.
It is best to
71,
where the method of experiment
draw the adiabatics so

1
will
that the area included
Trans. R. S. E. xx. (1851) 208.
62,
between consecutive adiabatics and isothermals represents unit

TT
work.
to T,
This area being
H must be taken numerically equal
and then consecutive adiabatics are defined by a series of

... on the isothermal for T, which
represent the
points A, B, C,
mass of the working substance of a
successive states of unit
perfect engine after successive additions of

at the constant
temperature
T.
The
T megalergs
adiabatics
of heat
may be num-
o, i, 2, ..., (j>, beginning with the line of no heat: then

a perfect engine work between the isothermals /, / and the
bered
if
adiabatics $,
The
</>
the total
signification
of
work done
is
is (/
given in
4)
60;
W*"^)it
is
really the
entropy of the body.
In future, the next article excepted, we shall always denote

absolute temperature by /, temperature on the Centigrade scale
by d, and absolute temperature on the air-thermometer by r:
all three scales 100
degrees occur between the temperatures
of melting ice and boiling water.
42. Comparison of the two absolute thermometric
in
scales.
To shew
the disadvantage of the absolute scale
first
proposed we subjoin a table of simultaneous temperatures on

r-273-7
perature of melting ice
is
easily
and
made
at the
is
taken as
zero.
its
63
The comparison
for
b^
temperature of melting ice
T=
273-7,
f=
hence
>
273-7^',
= "9-37,5 log r- l8 o,, 5
43. Zero thermal lines.

We know absolutely nothing
about the properties of solids at very low temperatures, but/we
can easily see to what results
The stress
27, 28 will lead.
exerted by a body, being the result of the motion of its atoms,
is
zero only
when they
perature also
is
zero
isothermal coincide.
a single value of
v,
are at rest, which occurs
hence the zero
when
isopiestic
the tem-
and the zero
o can correspond only

But to p
o, /
which must be the minimum volume that
mass can occupy, since otherwise the atoms would begin

move under their mutual attractions and so to exert stress
unit
to
the zero isopiestic or isothermal thus reduce to the point which

characterizes this limiting state of the body.
Again,)as the temof
a
no
heat
is
in
which
there
is
perature
always zero, the
body
zero adiabatic coincides with the zero isothermal, and so also does
the zero isenergic by
by the point
32
they arc therefore also represented
that characta-iflOD the
absolutory culd.
ctatc
Zero isometric there
is
-^f-tfaeba
obviously none, for
body cannot be squeezed into nothing.

As no two real adiabatics cut each other, so no
the
cuts the real zero isothermal,
no
which
is
adiabatic cuts the zero isothermal,
real adiabatic
a real adiabatic
and thus
it
hence
would be
impossible to construct Fig. 15 exactly as is described: we

might however substitute for the zero isothermal that part of
the axis of v (infinitely distant from the axis of
co-ordinates)
which is asymptotic to the adiabatics.
Professor, Maxwell
given by
this
method
1
ingeniously avoids the infinite diagram

and at the same time all hypotheses
Theory of Heat, 4th ed. (1875), 164.
64
as to the state of the bodies at

temperatures
of experiment
by employing a
fictitious
a series of
Thus
if
beyond the reach
zero isothermal and

fictitious adiabatics.
the isothermal
//'
cor-
responding to
temperature /
bounds the part of the dia-
gram
that
known by
is
ex-
periment, and the adiabatics

that cut it are continued by
"
f|
Fig.
lines
.;:>--"''
that
do
not
intersect
each other, then the line ZZ',

which is drawn so that the
1 6.
area included between

isothermal
and
(/>',
is
it,
the
and the continuations of any two adiabatics (f>

ma7 be treated as the zero iso<#>')>
equal to /(</>
//',
thermal and the prolonged lines as adiabatics. For the area

included between the adiabatics <f) and $', the isothermal //',
and any other isothermal TT' corresponding to temperature

T in the known part of the diagram, is (r /)($ $'), and
the continuation of this area
is
$'),
/(</>
so that the whole area
and the adiabatics $, <'

T((f)(f/), and it therefore repre-
included between the lines TT*, ZZ'
',
with their continuations, is

sents the heat taken in by a body when it passes from the
adiabatic <' to the adiabatic </> at the constant temperature T.
44. Intersection of thermal lines.

tains that
two adiabatics cannot
intersect,
M. Verdet 1 mainand
offers as
proof
if AC, BC are two adiabatics

the following:
that intersect in C, a body, after travers-
the isothermal
AB
and receiving heat
can traverse the adiabatics, doing

work equal to ABCA without any accom-
\ing
therein,
Fig 17
is
panying transfer of heat; and this is imThis proof is fallacious; for

possible.
Professor Riicker 2 has pointed out that it
impossible for the body to pass at the point

1
Theorie mecanique de la Chaleur (1868),

Proc. R. S. xx.il (1874) 452.
i.
C from one
184.
65
adiabatic to the other without absorbing or emitting heat, i. e.

condition that it should not pass from
the
while
very
fulfilling
one adiabatic
proposition
when
make
He
has
is
that
this
perfectly defined
and density
striking case
shewn
further
true only if the temperature
the pressure
most
the
is
another.
to
are given.
single instance
of the intersection of adiabatics will
this clear.
At ordinary atmospheric pressure denoted by the dotted

of
isopiestic in the diagram let A and B represent the state
unit mass of water at the temperatures oc. and 4c. respectively:
starting
from any
between these two
ture below 4c., the
initial
states at a
body
condition
tempera-
will contract
on
absorbing heat at constant pressure till it

reaches B, where its density is maximum
further addition of heat will cause it to
:
repass through the same states of density

as before, but at different temperatures.
On
Yig. 18.
regaining the density represented by P, it will now possess

its initial stock of heat all that has been sup-
in addition to
plied; for none has been expended in work, the contraction

and expansion having exactly compensated each other hence
each point P between A and B may be regarded as on two
:
adiabatics,
which there
This case
still
is
intersect.
even more striking;
at the constant pressure, until
it
for if
freezes,
we
cool the water
in
which case
its
denoted by A, and then

continue the abstraction of heat, we may reduce its volume
again to that of the initial state P; this state must therefore
volume becomes greater than
that
correspond to a third adiabatic through P, as on the whole no
work
is
done while the heat
is
batics pass through every point
abstracted.
Thus
three adia-
between those corresponding
oc. and 4c., and, the case being similar on the neighbouring isopiestics, there is a region in the plane of pv each point
of which corresponds to three different states of the water-
to
substance
hence not only do three adiabatics intersect
each point in
this region,
but there also pass through
it
at
three
66
isothermals, three isenergics, in short, three thermal lines of
every type except isopiestics and isometrics.

Such points of intersection of thermal lines are not the projections of real intersections of the corresponding real thermal
lines,
other
but denote that these curves in space seem to cross each

when viewed from the plane of pv Professor Riicker
:
therefore proposes to reserve the usual terms (intersect, cut,

meet, &c.) for the projections of curves that actually meet on
the characteristic surface, and to use the term cross for the
which
projections,
intersect,
of curves that do not meet in
space.
There are some

region wherein
peculiarities
has also pointed out.
we
in
the thermal lines in this
Riicker
upon which
crossings occur, which Professor
all
It is
an experimental
fact,
more in 78, that a substance which would conon absorbing heat under constant pressure, such as water
between o and 4c., experiences a rise in temperature when
shall say
tract
allowed to expand without receiving or emitting heat ; this rise

however cannot go on indefinitely, as work is done at the
is finally due (see
63)
a
constant quantity (the
by
hence on the
potential energy of infinite molecular separation)
real adiabatic through the point representing the initial state
expense of the
intrinsic
energy which
to the temperature increased only
and
all
such points
there
must be a point of maximum tem-
perature, and the real isothermals through all other points of

the same curve which lie within this region must meet it twice.
Now
the condition
f=
T<
coupled with the equation to the characteristic surface

/(A v, t) o
gives the locus of the points of maximum temperature, and this

curve therefore bounds the region in which real adiabatics are
cut twice by the same real isothermal, and its projection on the
plane of pv bounds the region in which two isothermals or any
two thermal
lines of the
since
points
at all
same type can cross each
other.
Also,
on the curve of maximum temperature the
tangent planes to the surface are perpendicular to the plane of
pv, the projections
its
on
that plane of
all
lines
67
meeting
it
touch
projection.
Again, the projections of a real adiabatic and a real isothermal that meet twice also meet in two points, say A and B
they must further cross each other at some intermediate point

C, and thus form two loops, since
the adiabatic at a point
ways steeper
The
C
33) al-
is (
than the
isothermal.
isothermal that corresponds at
to this adiabatic will of course
be
represented by some such line as the

dotted one, which conforms to the
same
rule.
These three points coinwhere the
cide in the limiting case
adiabatic meets the isothermal cor-
responding to the
maximum
temperature above mentioned
point where it touches the bounding curve of the

region), and there is therefore contact of the second order between these lines.
(that
is,
at the
Consider (Fig. 20) a pair of isothermals
ABCD, D'C'&A',
each cutting twice a pair of adiabatics BC'tfC, AEfA'D, and

the different lines further cross each other in the points O,
P, Q,
^?, S,
as in the diagram
they
all
the second order with the line which
curve of
maximum
temperature.
have contact of
is
let
(/,
at least
the projection of the
Between the temperatures
corresponding to these isothermals we can make the substance

under consideration perform Carnot's cycle in twelve different
ways.
(i)
Giving or removing the same amount of heat along AB,

traverse either the cycle ABC'I/A or the
we can make the body

these
cycle ABB*A' A
areas are therefore equal, and their dif;

ference, the area C'OB'A'OD'C', is zero, or the two areas OA'B'O,
OZ/c'O, which are traversed in opposite directions, are equal.

again, giving or removing the same amount of heat along
f
A'B', we can allow either the cycle A'B*CDA or the cycle
Or
A'tfBAA'
to be traversed;
these areas (regard being paid to

and their difference, the area
their sign) are therefore equal,
F 2
68
CDOABOC,
is
OABO, ODCO, which are

Hence the areas
two adiabatics and an isothermal which
or the
zero,
areas
traversed in opposite directions, are equal.
of
the
two loops formed by
meets each of them twice are equal.

'(ii)
Similarly, if
we
consider either of the pairs of equal cycles

'A'A^
ACC'D'A}
ACB'A'A\
we
'
B'tfPC'OPB' is zero, or the areas Pd B''/>,

and these equal cycles, paired together dif-
find that the area
PC' OP are equal
ferently as
A cc'A'A
A CB'A'A
ACC'tfA\
ACB'D'A\'
I/OQA'(/Qf is zero, or the areas QOtfQ,
QA'O'Q equal. Or again, of the four equal cycles A'c'CAA',
A'C'CDA', A'C'BAA', A'C'BDA' either of the pairs
shew that
the* area
A'C'CAA'
gives the area
equal,
and
A'C'CDA'
A f c'BAA f \ A'C'BDA'\
BORCO'RB zero, or the areas ROBR, RCO'R
either of the pairs
A'C'CAA'^
A'C'CDA']
A'C'BAA'}
r
A'C'BDA'\
gives the area DCfSAOSD zero, or the areas SO AS, SDO's equal.
Thus the areas of the loops formed by any adiabatic and an iso-
thermal which meets
This
latter result
it
twice are equal.
may be deduced from
the former
by sup-
posing the two points of intersection of one of the adiabatics

with the isothermal to coincide.
see therefore that, if a
We
body perform a
adiabatic
cycle of operations
and an
isothermal,
it
which can
be represented by
an
will on the whole do no useful
work.
From
the above the areas
ROBR,
RCOR
are equal,
and by
adding to each the area O'PRCf we see that the areas o'POBtf,
tfPCtf are equal ; if now we subtract from these respectively
POC'P, PB'tfP, we find that the areas O'c'BO',

hence the further proposition: the
two loops formed by two isothermals and an adiabatic
the equal areas
OB* CO'
are also equal:
areas of the
which meets each of them twice are
equal.
69
Similar results are obtained on considering the shapes of the

thermal lines near their other intersections with the boundary
anomalous region where

maxi-
line of the
it
corresponds for any given
pressure to a local
mum
volume
case
of water the
but in the
siderations are
con-
more com-
by reason of
plicated
dis-
continuity occurring in the
thermal
Hence
lines.
the
real isothermals that pass
through all the points of

a real adiabatic that lie
two points
between the
where
touches
it
boundary
the
of this re-
line
the points of local
gion
maximum and minimum

temperatures
three points:
meet
the
it
same
in
is
true of the projections of
these
curves,
adiabatic
but, as
an
is
always steeper
than an isothermal at their
T7 .
rig. 20.
point of intersection, the
must further cross in two points intermediate to the points

of intersection, and there are thus formed four loops which
by the above proposition are equal, two and two.
lines
SECOND LAW OF THERMODYNAMICS.

45.
To Sir
Generalization of Carnot's principle.
is due the following extension and generaliza-
W. Thomson 1
if
tion of the results of Carnot's principle
represents the
heat taken in or given out at the absolute temperature T, \>eing
considered negative if given out, by a perfect engine with several
:
Trans. R. S. E. xxi. (1854) I26 > or Phil
MaS-
xi
'
( l8 5 6 )
2I 7-
70
sources and refrigerators at different temperatures, then the
sum
TT
of
all
the quantities
This
have
is
is
zero, or
--- =
o.
Consider now, with Professor
TI} T
cycle with two sources at temperatures
of heat fflt
we
obviously true for Carnot's simple cycle; for
are taken,
which heat h
is
and a
Zeuner 1
wherefrom amounts
refrigerator at temperature / to
given out, the diagram being composed of isothermals and adiabatics.

Let
the initial state of unit
mass of
the working substance of the
engine at temperature / be denoted by A

(i) compression
:
without loss
temperature
of heat
till
the
rises to 7\ brings
mass along the adiabatic
the
B\ (2) heat ff1 given at the

constant temperature 7\ takes
it
along the isothermal to C;
to
Fig. 21.
(3) further expansion without
gain of heat
adiabatic to
till
the temperature
(4)
falls
to
communication of heat
carries
it
along the
at the constant
temperature Tz moves it along the isothermal to E ; (5) a final

expansion without gain of heat till the temperature falls to /
takes it along the adiabatic to F\ (6) a final compression at the
constant temperature / with loss of heat h brings it back to its
This process is obviously reversible. If the
initial state A.
adiabatic CD intersects the isothermal FA in G, and h is com-
posed of heat h" emitted along FG and K emitted along GA,

the complete cycle may be considered as made up of the two
simple cycles
ABCGA, DEFGD
ff1
h'
7T7
1
and
hence we have
h"
^-T :=0
Grundziige der mechaniscken Wdrmetheorie,
2 te
'
Aufl. (1877), 62.
71
whence on addition
applied whatever the number of

sources and refrigerators and however complicated the diagram
provided that it is composed of isothermals and adiabatics.
The same method may be
The
Suppose the
general for this case.
are taken in at temperatures
... Tn respectively, a negative value of
denoting emisadd a perfect engine emitting H^ at 7^ and taking in
following proof
7\,
T2
sion
&
is
H^H^...Hn
quantities of heat
H
at E,
another emitting
at
and taking
in
7\
rf
at ,E,
and so on: these n perfect engines together with the given

one emit on the whole no heat, but take in
this quantity
must however be
zero, for,
whenever heat
is
taken
in by an engine, part of it is emitted since the whole of

cannot be converted into work ; hence
TT
/7 1
^r
If the
comes
TJ
TJ
/J
2Jn
T_T
fl
= 2- =
+ ^r + ---+^
adiabatics
integration,
come
it
Q.
close together, the
summation be-
and
o.
In the case of a continuous cycle-curve, divide the diagram

an infinite number of cycles
into
by adiabatics infinitely near each

other and consider one of them
PQRSP,
along
in
which
dH
is
taken in
PQ at T and dh given
RS at /: draw the
out
isoalong
thermals that pass through P and
Then we may consider the
R.
cycle
PQRS
three
s'RSS'
as
cycles
\
but
composed of the
PQ&P, P(^RS'P,
in the cycle PQQ'P
Fig> 22
the difference between the
7.2-
taken
heat
in
and the
heat
emitted
is
negligible,
being
measured by the area PQQ'P which is an infinitesimal of the

second order, and the same is the case with S'RSS'-, hence
Carnot's
cycle
order only)
giving out dh at
first
or
PQfRS'P (whose area
may be
/,
an infinitesimal of the
is
dH
considered as taking in
at
and
and therefore
dH dh
on integration over the whole cycle
we take
If
the integrals between any two points
and B, we
rs
rA dH
dH
~ +
=
/
T
T
J
B ~^F
JA
have
where the
first
and the second
integral refers to the part

to the part
BRSA
if
constant but vary the second in any way,

cycle reversible, the value of the integral
same
the
or
is
APQB
we keep
then
still
JB
of the curve
the
first
part
keeping the whole
the integral therefore depends only
remains always
upon
its
limits,
a perfect differential in reversible cycles.
The above generalization of Carnot's principle we may

Thus the two
consider as the second law of thermodynamics.
laws with respect to the absorption and transformation of heat
in closed reversible cycles are expressed
by
I dH = W
'
Cdff
J ~T
46.
= o.
Equivalence of transformations.
Professor Clau-
has established the above equation (6) by a very different

course of reasoning, and has given it a physical meaning.
sius
M. Carnot's theorem may be
stated thus:
Whenever a
quantity of heat is converted into work by a reversible process

and the substance effecting this transformation passes through
1
Pogg. Ann.
xciii.
(1854) 4 8l trans Phil

>
Ma g-
xii - ( l8 5 6 )
8l
73
a complete cycle of changes, another quantity of heat must

be transmitted from a warmer to' a colder body; and the relation between these quantities depends only on the temperatures.
The
real law of dependence of this relation on the temperatures

not brought into view by the simple cycle proposed by
Carnot; for the heat transmitted is taken therein at the
is
M.
same temperature as that transformed. The following

suggested by Professor Clausius is more general.
cycle
Let unit-mass of any substance, enclosed in a piston-fitted

be at temperature T, and be represented as to pressure
cylinder,
and volume
in
its initial
state
by the position A on the diagram
(1) expansion without gain of heat
till
the temperature
falls
to 7\
brings it along the adiabatic to B

(2) then communication of heat h
:
constant
the
at
moves
it
temperature 7\
along the isothermal to
C\ (3) a further expansion without gain of heat
becomes
the
till its
Tz
brings
adiabatic to
pression
now
temperature
hence along
it
(4)
com-
constant tern-
at the
Fig- 23.
perature T2 until all the commu:

nicated heat h is given up moves it along the isothermal to
a
heat
till
its temfurther
of
without
loss
(5)
compression
perature rises to
where
its
volume
T
is
changes
its
the isothermal AF, the line
along the adiabatic to F,
state
smaller than at
EF
F is
for otherwise, as
would cut one of the
on
lines
already traced, and work represented by the area so enclosed

would have been done without on the whole any expenditure
of heat: (6) communication of heat
at the constant tem-
it
back along the isothermal to
perature
brings
condition A.
In this cycle, then, heat
transformed into work, while also heat h
body
at the
at
is
its
is
from a
transferred
higher temperature 7^ to one at the lower

is reversible, and the relation between
This cycle
original
temperature
H and
depends only on the temperatures; for were two engines performing such cycles in which ff, Tt 7\, T2 are the same but
74
h
coupled together with that one reversed wherein

on the whole, without any other change,
heat equal to the excess of the larger value of h over the smaller
would be transformed from the lower to the higher temperature
different,
is
the greater, then
which contradicts the fundamental axiom.

Heat naturally tends to equalize differences of temperature,
that is, to pass from warmer to colder bodies; and Clausius'
cycle shews that
a passage in the contrary direction can only

take place under circumstances where simultaneously another
quantity of heat passes from a warmer to a colder body, or
'
when some change occurs which has

being reversible without causing on
from a warmer to a colder body
the peculiarity of not
its
part such a transfer
V Two transformations,
one
of which must accompany the passage of heat from a body at

one temperature to a body at another, may be considered as
equivalent
tion of
thus from the above cycle reversed the transformaat temperature T is equivalent to the
into heat
work
transformation of heat h from the higher temperature 7\ to the

lower T2 and these are considered negative transformations;
,
similarly the transformation into
work of heat
equivalent to the transformation of heat
is
temperature
tions.
We
H at
temperature
h from the lower
T2 to the higher 7\ these being positive transformahave here reversed Professor Clausius' convention
f
of signs (as he himself did

already assumed.
later )
for
agreement with that
Every transformation must therefore be measurable, or have

TT
an
equivalence- value.
If
that of a transformation into
is
work
-x^
temperature T, % being a certain function of T,

// from tem(7\, T^ that of a transformation of heat
of
the
T
above cycle
consideration
to
perature 7\
temperature 2
of heat
at
and H-fy
Another similar
and
T replaced
cycle,
wherein
by H' and
Pogg. Ann. cxx. (1863) 430.
',
h,
T^ T2
are
unchanged but If
will give
Pogg. Ann. cxxv. (1865) 356.
Suppose
greater than 7
be coupled to the first

to T2 and from T2 to
while heat //
i.e.
is
',
and
lt
is
transformed both from 7\
the total result
taken in at
the transformation into
the second cycle reversed
let
then, as h
work of heat
HH'
together with the transformation of heat

perature T to the temperature T* ; therefore
eliminating H,
Hence
H' by
work performed
is
and heat H' given out
On
75
at
at
T'',
temperature
the tem-
H* from
the above equations,
we
the second fundamental theorem, which
find
is
in- this
fitly called the theorem of the equivalence of transformations,

be thus enunciated: If two transformations, one of which
form
may
must
accompany a third transformation, are called equivalent, then a

at the temperature T has the
transformation into work of heat
equivalence-value
perature T^
to
>
H from
tem-
T2 has the equivalence-value H(^--
^-),
and a transformation of heat
temperature
*i
*a
a function of the temperature entirely independent of
the process by which the transformation is effected.
Thus the equivalence-value of a transformation of the second
wherein
kind
is
is
equal to the difference between those of two transformafirst kind, and we can therefore prove in the most
tions of the
general case that the algebraic

all the transformations
the equivalence-values
of
occurring in a reversible cycle is zero.
For the transformations may be

transformations of the
sum of
first
all
considered as composed of
is, of heat into work, thus
kind, that
ff
ff
giving a series of equivalence-values ~-y -='> ... ; and each of

2
Sj
these latter is equivalent to a transformation of the second kind
between two given temperatures
T, 7*
so that
hence the transformations are together equivalent to a trans-
76
=$
^ + ^3 +
formation of heat
fr
the lower temperature
Q
higher T\ now if ty is positive, this gives a passage of
heat from a lower to a higher temperature without any other
compensating change which is contrary to the fundamental
to the
axiom
it
ty
and
positive
is
is
negative, a reversal of the cycle will
make
foregoing impossible case:
this to the
therefore zero.
Hence
cycles
or
if
and reduce
the mathematical expression of this law for reversible
is
more generally
where d
r^
H represents the
reckoned negative
By assuming
if it is
that
heat taken in at temperature T, being
given out.
no heat
is
work when changing
in internal
absorbed by permanent gases

state,
Professor
Clausius
finds
proportional to the absolute temperature on the airthermometer. This suggests taking Z as absolute temperature
that
is
without reference to any thermometric scale or any substance,

and the above results then become those deduced somewhat
earlier
Sir
by
W. Thomson.
47. Reversible cycles not closed.
Let A, B, C represent the successive states
of a body in an open cycle
differential
perfect
dH
as
(d$>
say),
<{>c
<t>A
its
is
integral
AC
along the path
=
-^r
The physical meaning of is given in 60.

The value of this integral may be exhibited
(7)
(j)
graphically.
Let
the isothermal through A meet the adiabatic through C in

then, the path ABC being reversible, the whole cycle ABCDA
reversible, and the integral over the whole cycle
CdH
f
rr*
_
*~"~"
[c
/
JA
dH
T
/T-I
CD
*
Jc
CA
dH
T
i-rt
'
JD
dff_
T
/TT?
""
is
CD dff
-
but
Jc
if
zero, as
is
CD
is
77
an adiabatic, and
f A dff
the heat required to bring the body from

which is the isothermal for /: therefore
is
AD
is
JD
D along
to
L ~?=r
by
But,
33,
is
ADEeaA, where E
the area
is
the intersection
of the isenergic through A with the adiabatic through C; thus

the value of the integral is the / th part of this area.
48. Mathematical expression of

Consider an infmitesimally small cycle
adiabatics
DA
AB CD
t
/,
Draw
S/.
DA to
isothermal BC
and produce
the ordinates Bb, Cc,
heat given along the
the notation of
principle.
formed of the
and the isothermals BC,
for the temperatures
The
Car net's
ABCDA
15
is
Idv
be
F.
in
and
ABCDA, which measures the work

done, may be considered a parallelogram
and therefore equal to BCEFB or BFxbc'
the area
but
BF
stant
is
the increase of pressure at con~&p
volume due to the increase 6/ of temperature, or
hence by (i)
^p
^1
In
this
form
it
convention of
was
first
given by
1
:
with the
4 1 this becomes
*l
d/
49. Dr.
M. Clapeyron
=L
(8)
Zeuner's method. From the
difficulty
of giving a
precise mathematical definition of temperature and from a desire

to derive, if possible, all thermodynamics from the first law only,
Dr.
Zeuner 2 has proceeded by
state of a
1
body
is
another method.
completely defined in general by
Journ. de VEcole Poly. xiv. (1834) 153.

Grundziige der mechanischen W'drmetheorie (1877), 39
Since the
its
density
et seq.
78
and pressure, E may be expressed as a function of p and

and the general equation (2) of 31 may therefore be written
dH
where
dp +
c^
that
may be
dH is
called the
=-
ft
^v
\-
p:
differentiation
perfect differential,
differentiation
This equation, which
shews
^v v
and
i?
then - - being a
it
differential.
Let - therefore be an integrating factor
whence on
and subtraction
fundamental equation, and
first
not a perfect
dv,
<)
and
-r
^p
from these we have on partial
This
t>
v,
by
simplification
ZT
ZT
d/>
dv
-- *
may be
called the
(9)
we
find
(10)
second fundamental
given by M. Clapeyron , whose reasoning

and method were of entirely different character.
equation,
was
first
Using the value of
given by equation (10),
we have
Yp
Hence
may be
written
dH =
according to the variables

1
we wish
to take.
Journ. de VEcole Poly. xiv. (1834)
53-
79
now we
If
and two
consider a cycle represented by two adiabatics

const, respectively,
defined by r
const.,
so defined that for all closed reversible cycles
lines
then, as r
is
/=
"dff
we have
(12)
W = H-ff=^(r-r \
work done when heat H
f
where
T'
the
is
the line r
if
to a level
a mass
the
if
be
lifted
mass
work done
is
is
done
the
level
weight
A
is
through a height h
w, the work done
^ = ^^
is
re
if
its
raised only through a height A' to the level C,
whence
Again,
from a
where
at a spot
against gravitation
the
is taken in along
out along the line
const.
Now
and
H* given
const., and heat
(13)
mass
is
raised
is
TO'
(14)
^.
from
level
to level ^, the
work
n,
this process is reversible; for, if the mass is allowed to

sink slowly from the higher level to the lower and do work, the
work it does is given by the same equation.
and
As
ff,
being heat,
equations (12),
is
a perfect likeness between

(14), (15): r is thus analogous to a
work, there
(13) and
is
TT
height and
to a weight, whence- Dr.
Zeuner
calls
them a
thermic height and thermic weight respectively; and the above

cyclical process may be looked on as the transference of a
TT
from the
thermic weight
(15) however gives the
level r to the level
r'.
Equation
minimum amount of work required
to
through the height h h', or the maximum

amount of work that the mass can do in sinking through
raise the
mass
80
this
i.e.
height
when
the
levels is infinitely small;
mass
on reaching the new
velocity
furthermore
it
to rise without the expenditure of
impossible for the

work hence, if the
is
above analogy holds, (i) the work done in the cycle when the
TT
thermic weight
a
maximum;
work may
is let
down from
the level T to the level r'
is
under other conditions
(2) during this transfer
be done; (3) the transfer from the lower to the

level
cannot
occur without the expenditure of work,
higher
less
rr
which
at least equal
is
to
(r
Now we know
/).
by ex-
perience that we can gain work when heat is allowed to pass

from a body at a higher temperature to one at a lower (e.g. in
the steam-engine from the furnace to the condenser), that this
work
the less the
is
occurs, and
more
radiation
and conduction of heat
that heat cannot of itself pass
from a body
at
lower temperature to one at a higher we are then justified in

assuming r to be absolute temperature, which thus obtains a
:
perfectly precise mathematical definition, as the reciprocal of
integrating factor of
Thomson's
dH, and we are brought back
to Sir
an
W.
absolute scale.
Connection between the two laws of thermodynamics. Attempts have been made to deduce the second
law from the first, by the aid of mechanical principles or me50.
chanical assumptions only, but in general these assumptions

no more obvious than the thermal axioms of 38 thus
are
Rankine, Boltzmann, aud Clausius have given

demonstrations, the first on the hypotheses of molecular vor-
Professors
tices
and
motions
3
,
circulating streams
and the
third
on
2
,
the second
on
that of periodic
that of quasi-periodic motions
Hr. Szily has given the following demonstration

of
31 are respectively the K, P,
W of
5
.
4
.
The S, 7, W
hence
Trans. R. S. E. xx. (1851) 205.

x. (1855) 416.
Phil. Mag.
[4]
See Mr. Burbury's simplified proof in Phil. Mag. [5] i. (1876) 61, or
Mr. Watson's Kinetic Theory of Gases (1876), 46.
4
p gg> A nn. cxlii. (1871) 433.
5
Muegyetemi Lapok, i. (1876) 165, trans. Phil. Mag. [5] ii. (1876) 254.
3
ii
81
But,
r denotes time,
if
d(xbx)
and
we have
also
= xdlx + dxlx =
similar equations in
and
hence
(16)
(xbx+_yby + zby) are

and if 8/7, ^ represent the mean energy communicated
system per unit of time and the mean kinetic energy
If the initial
equal,
to the
and
final
values of S
./u
during the interval r respectively, then
T^=8.2r^
Now we
system
for
it
is
is
(17)
cannot assume that the thermal state of any body or
ever absolutely the same at two successive instants ;

composed of numberless molecules which are in con-
and only those molecules

whose velocity
There will therefore be con-
tinual vibratory or progressive motion,
are in the
happens
same
positions in successive instants
for the time to
be zero.
tinually varying rarefactions
and condensations, and,
as the
chance of these rarefactions and condensations exactly compensating each other is infinitely small, it follows that the volume
of the body must be subject to a continual variation which is
When therefore we
periodic and of very small magnitude.
the
of
a
volume
of
we
mean
speak
specific
body
really the mean
volume occupied by unit-mass during the period of this pulsation,

it, and when the
specific volume remains
as Hr. Szily calls
constant, the period of the pulsation
must be constant
too.
83
During these changes of volume work
will also be done, which

zero in a complete pulsation there are similarly variations of
the kinetic energy S, the potential energy /, the intrinsic energy
is
E, and
is
it
pulsation that
the
mean
we
take to characterize the thermal state of the
values of these functions during the
Further these changes occur adiabatically, for
body.
dH = dS+(dI-\-dW}
Hence
if
d instead
of
="2./m
{xdx+ .,.xdx..'.}
{x x
./x
xx
...
...}
dr
o.
we take r the period of the pulsation, and write

6, we have on integration through any closed cycle
dffdr
= o,
and
since r
-5"
Or we may
at the close
resume
their initial values.
write (17) as
dff_
S
and therefore on integrating over a complete cycle we get
JH =
-=-
o.
S
This
is
the second law of thermodynamics, since
tional to the absolute temperature (see
The
at the
values of 2
//,
(x
when no energy
pulsation
is
interval r
(x$x+jby + z.bz)\
is
added
constant and the
is
propor-
64).
bx + y fy + 2 Sz) have been
beginning and end of the

.ju
since,
for
2TS
taken equal
to a system, the period of the
mean
kinetic energy
undergoes no
change.
If however what we say in
38 is true, then no proof of the
second law can be deduced directly from known dynamical
and any such deduction, wherein probability has not
principles
:
been taken into account, must be imperfect, however

it
may
appear.
satisfactory
83
NON-REVERSIBLE PROCESSES.
Non-reversible cycles.
51.
In the cycles hitherto con-
we have supposed
sidered
equation (2)
following conditions to be satisfied
presence of others whose temperature
:
31 to hold, that is, the

the body is never in
(i)
is
not
near
infinitely
its
pressure never differs more than in(ii)

from
the
exerted by the body, (iii) the changes
stress
finitesimally
of density occur with exceeding slowness, (iv) no heat is spent
the external
own,
overcoming friction or transformed into magnetic or electric

energy in all such cases the cycles are reversible.
In other cases we cannot perform the process in exactly the
in
reversed order, and these are called non-reversible cycles.

their case the equation
In
must hold, as
we
fdH = W
expresses the Conservation of Energy
it
but
have always
for
we have shewn
in
46 that the transformations of any cycle
may be reduced
temperature
positive, but
TQ
to the transformation of heat ^ from the lower

to the higher T, and that f) cannot therefore be
we cannot
in this case reverse the cycle
and shew
cannot be negative. Thus for non-reversible cycles ty is

negative, or the sum of the transformations is a negative trans-
that
f)
whence, calling a transformation uncompensated which

unoccurs without any accompanying change, we may say:
work
can
be
that
is, of
compensated transformations
only
negative,
formation
and of heatfrom higher to lower temperatures.

M. Verde t 1 has shewn in detail that neglect of any of
into heat
above conditions
entails a negative value for the integral
the
no difficulty in shewing this for himself.

In the case of a non-reversible process that does not form
the student should have
ABC in Fig. 25, we may, following M. Claucomplete the cycle as therein done by the adiabatic
a closed cycle, as
sius
1
2
,
Theorie mecanique de la Chaleur (1868), i. 188.

xcvii. (1856) 452, trans. Phil. Mag. [4]
Pogg. Ann.
G 2
xii.
(1856) 248.
84
through C and the isothermal through

the whole cycle ABC DA
Ada
but
is
by
47 equal to (^4
then the integral over
<c, whence
t>C~<t>A.
(20)
CHAPTER
V.
Law.

tions of
15, 31,
Recalling the equa-
we have
= dH=kdt+ldv
=
dE
kdt+ (/-/>) dv.
dE+pdv
or
dE
But
is
an exact
differential
hence
a*
s/
57-^ =
~isp
(2)
j7'
1
but
This equation was first given by Professor Clausius
by a somewhat tedious method the above is Sir W. Thomson's
2
But for this relation we should
very precise demonstration
,
have had no reason for thinking that

independent of each other.
By
similar
and k are not entirely
methods we may get the corresponding equations
the last being already given in
by taking
an exact
49.
Second Law.
differential
This
is
expressed
hence the function
Pogg. Ann. Ixxix. (1850) 384, trans. Phil. Mag. [4] ii. (1851) u.
Trans, R. S. E. xx. (1851) 270, or Phil. Mag. [4] iv. (1852) 19.
86
is
an exact
differential,
so that
*- A Z
A
d'/~~d/V
which becomes on simplification
whence by
(2)
<^
_^_^_
d/~d/
This
is
_^
Sz>~/"
the result found directly in

48 from
also obtain from (5) and (6)
Car not' s
prin-
We
ciple.
By
using the other expressions for
wefind
*v
dH
in a similar
manner
az
*jr
(8)
fc
and
j)
^z;
c)/>
t^^p
dzr
&
54. Values of the Specific Heats.

If ft,
represent
values of which will be found in
/, the probable
functions of
65, equations (7)
and
(9) give respectively
=,/
A-=lt
//
^^
(12)
which is obvious, since the

being indefinite
a body can depend only on its actual state. If
the integrals are taken definitely, then ft, It are not functions
As / is to
at the lower limit.
of / only but are the values of k,
the
integrals
specific heat of
be taken constant in evaluating these integrals, it is convenient

to introduce two quantities w w' to represent respectively the
indefinite integrals fpdv, fvdp wherein / is constant, these
t
quantities being obviously connected
by the
pv = w + w' + const.
'
relation
The above equations then become,
87
with the notation of
15,
(I3)
(H)
From
16 and (6)
(6)
we have
or
16 and (8), or
These equations may also be got from (8)
is
that
remark
from (9)
1 6 and
Hence
we
may
(10).
always greater than k; for

of
is
always negative and the sign
without influence.
is
ct
The
difference
characteristic equations corresponding to the bodies for
is
always a constant quantity
partial differential equation
it
gives
this difference
is
a function of
t
i
is
a function of
primitives,
A particular case of the
p-b = a^-a
where
this
Rt,
wherein a and b are arbitrary constants.

the equation to a perfect gas.
When
which
are easily found from the
two complete
(-)(-&) =
only,
we
find
former
is
one complete primitive
related to the given function, a
and
6 being
arbitrary constants.
55. Difference of the Elasticities.
hence
e
e,
k
V
k
and
similarly
^p
<>v
From
vt ,^p. z
(
)%
k x d/ x
.^>v^
17
we have
(16)
88
56. General Equation.
This
may now by
(13) be written
dv
(18)
~.
And by
(14)
we have
similarly,
(19)
when
and/
are the variables,
(20)
57. Determination of the Intrinsic Energy.

we have
dE
= dH-pdv = k#+/<//+(/
By
-p) dv
t-V>),
the complete integral of which
(21)
is
-.)-(/
the suffixes
(18)
-a>.),
denoting the values corresponding to the
(22)
initial
condition.
If the intrinsic
energy
use of which variables
is
is
required in terms of
limited, as
we have
/ and
the
already noticed
equation (20) gives

7
__
grtt
Jj
whence
EE^\
ftdt-ifivpj)^
We
can determine the
equation
dE
the integral of which
K(-<}.
(23)
energy directly from the
intrinsic
^p
kdt+(tf
p)dv,
(24)
is
E-E. = t\dt + Jf* {&-})&.

Of
J&
vn
(25)
since (v. Professor
Price's
du
be an exact
Int. Calc.
372)
89
if
= f(x, y}dx + $ (x, y) dy
differential, its integral is
-o =
/(*OO dx +
+ (*o, y} dy
(26)
/(*, *)<**
/"
JX
Similarly, with /
and
/ as the variables, the
equation
-ir^g)*^
gives the integral
These equations were
by
W. Thomson
Sir
1
,
first
who
given in a slightly different form

thus shews that the variation of
may be determined without experiment

connecting p,
we know
and the thermal capacity
v, /
constant density or pressure and
We
if
all
for
the equation
any particular
temperatures.
have also
which gives on integration
where
5J represents
finite
if
we
Hence
a function of
for the intrinsic
we must have
the integral being inde-
becomes
energy to be a function of the temperature only

p
in this case the internal
pressure.
v,
take the integral definite, the equation
work must be independent of both
density and
These are therefore properties that belong to a perfect gas.
58. Professor
have obtained so
Rankine's method.
far are derived entirely
The
equations
we
from the two laws of
thermodynamics, and are perfectly independent of all hypotheses

as to the molecular constitution and physical properties of
1
Trans. R. S. E. xx. (1851) 479, or Phil.
Mag.
[4] ix. (1855) 527.
they contain such quantities only as are capable of

For further development of the theory
bodies;
measurement.
direct
in
26-29) certain hypotheses have been introduced; and though these take us out of the domain of pure
(indicated
thermodynamics into
that of molecular dynamics, yet
briefly discuss the results so obtained,
we
shall
mindful however of their
hypothetical nature.
Rankine 1 assumes
Professor
the whole of the
first
term
of the right-hand expressions of equations (18) and (19) to

represent the increment of the sensible heat in the body: this
is
assumption
double
really
(i) that the
sensible heat in a
independent of its density or the arrangement of its

molecules (2) that the expression for the internal work during
body
is
any change contains no term depending on the temperature

On these suppositions equation (18) may thus be anaonly.
lysed
(a) the variation
\dt or ds of the sensible heat of unit-
mass;
(/3)
the
work
dl
done against molecular action
small increase dt of temperature

(y) the
work
t-^jdv
in the
done against molecular action and also
against external resistance in the small increase dv of volume,
so that the work done
only
is
The
(/
--
p)
its
this
account against molecular action
dv.
total internal
whence
on
work done
therefore
is
absolute value for any state of a body
is
/= /*?_, + *,
as
we might have
inferred
from
(22),
(28)
3 being the constant of
29
1
Manual of the Steam Engine,
&c., 4th ed. (1869), 312.
Rankine
Professor
91
defines also another function
$ such
dH=td\
we
that
(29)
therefore have
on
and,
where
integration,
is
<
the value of the function in the
initial
state cor-
is called the thermodynamic function

responding to /> ^,-/
</>
for the kind of work under consideration, and the quantity
.
is
called the heat-potential or melamorphic function of the
substance for the work in question.
The further assumption that ft is a constant quantity for

each substance whatever its state (see
65) leads to the
equation
59. Professor Clausius* method.

this
method
is
'
that
The assumption
in
work which can be done
the mechanical
by heat during any change of the arrangement of a body is

proportional to the absolute temperature at which this change
occurs
1
.'
The
justification
'internal probability/
which
of this assumption
is
is
referred to
however scarcely obvious: we

results, and it seems to be true
must rather judge of it by its

with perfect gases and for thermo-electricity 2
new mag-
nitude disgregation is introduced to express the degree in which

Now
the molecules of a body are separated from each other.
to alter the disgregation, and it is by this
done whether internal or external hence,
according to the assumed law, if in any reversible change a
total amount of work dl+dW is done, the disgregation Z must
the action of heat

action that
work
is
is
so be measured that
tdz=dl+dw,
1
2
Pogg. Ann. cxvi. (1862) 82,

Pogg. Ann. xc. (1853) 513.
trans. Phil.
Mag.
(32)
[4] xxiv. (1862) 88.
Fundamental
92
Eqtiations.
and the general equation becomes,
as in (5),
31,
dH=dS+tdZ.
Now dz
given states, and
an exact
(33)
depends only on the arrangement of the molecules in

therefore
is
an exact
and therefore
differential,
differential
is
must be so
too, or
also
is
This is what we have termed the first

Rankine's assumption.
If now we make the second part of Professor Rankine's
assumption, we get
a function of
/ only.
part of Professor
Professor Clausius does not however admit

and therefore
*~*"J4**
ZQ
this
generally,
<
lf
(35)
17-~ljf
being the disgregation in the
initial state,
and
&
a function
of/.
Hence Hr. Clausius
arrives at
an extension of
his
theorem
of the equivalence of transformations. As an increase of disgregation is the conversion of heat into work, it is a positive
transformation, and
definition
or
its
now
we may say
dH
external objects, and
temperature
hence
equivalence-value
dH is
is
dS
is
is
dZ, since this
the heat received
is
by
by the body,
the heat given up by the body to

the heat that goes to increase the
dSdH represents the heat converted into

dSdH also
.
work, and
its
gives for
any reversible process, which
equivalence-value
is
is
equation (28)
not necessarily a
cycle, the relation
therefore the algebraic
sum of
all the transformations
in any
reversible process is zero.
Omitting the condition of
reversibility,
and remarking
that
no
93
can ever occur, as before shewn,

accompanying negative one whose equivalenceleast as great, while a negative transformation can
positive transformation
without an
value
is
at
occur with a smaller positive one or even entirely uncomcan never

pensated, we see that the change of disgregation
be greater than that which corresponds

though it may be less, or
dZ
the
to
work done,
dl+dW
<
hence the general equation gives
CdSdH
J
or the theorem takes
its
+J<tZ <
most general form
all the transformations occurring
of
dition
whatever can only be negative,
(36)
o,
the algebraic
sum
during any change of conor, in the limiting case
of
reversibility, zero.
C*
/T
is
called
by Hr. Clausius the transforma-
tion-value of the heat actually present in a
from a given
initial
condition
body when calculated
the important integral
which we have before represented by fd<$> under the name

of thermodynamic function, is called the transformationalcontent of the body, or its entropy 1
We shall adopt this last
.
name, the term
thermodynamic function not being sufficiently
characteristic.
Hr. Clausius suggests
the
body
also the following nomenclature:

the internal ergon or ergonal-content of
the thermal-content of the body : thus the intrinsic
W the external ergon ;
energy E is the sum of the thermal and ergonal contents.

Latent heats of expansion, fusion, evaporation, &c. are all
varieties of ergonized heat, and are generally both internal and
external.
1
Pogg. Ann. cxxv. (1865) 390.

Abhandlungen uber die mechanische Wdrmetheorie.
Memoir.
2
Appendix to 6th
94
60. Entropy. The entropy of a body is its property, expressed as a measurable quantity, which remains constant
when
heat
is
by the body, and which
neither gained nor lost
creases or diminishes according as heat enters or leaves
The
it
in-
for
and adiabatics of a body are therefore coin32, and their equations are $ = const.
isentropics
cident, as stated in
Just as
we cannot determine
the absolute value of the in-
energy of a body, since we cannot entirely deprive it

of heat, so we cannot find the absolute value of its entropy
trinsic
convenient therefore to reckon entropy from a definite

If then this state corstate within the range of experiment.
it
is
and it is the limit from which

responds to /> V Q /
then
measured,
equation (31) gives the entropy
,
~&w
As
,
I
is
<)z;
*=* 'H + '57"Sf

.,
(37)
a relative quantity and may be negative

the absolute entropy of a body however is always positive.
The best method of determining the entropy of a body by
thus defined
which
is
it is
meant the entropy of unit-mass
is by performing two
without addition or removal of
its state
operations: (i) altering

heat till the temperature becomes
the entropy remaining
this
unchanged during
process; (2) bringing its pressure to/
the temperature being kept constantly /
if // is the heat emitted
the
latter
during
process, absorption being reckoned negative,
,
the entropy
is
lowered during
this
process by
and
this
quan-
'o
tity is
The
the relative entropy of the
entropy of mass
operation heat
is
is
body
therefore
in
original condition.
its
m-
or,
if in
the above
4
emitted by a mass
of the body, the
TT
entropy of unit-mass
If a
body
is
mtQ
loses heat
at
the constant temperature
its
TT
entropy
is
lessened by
and
if
another body at constant
/
temperature
this
gains
TT
hence when heat
temperature
is
its
heat,
entropy
95
increased
is
from a body
transferred
to another at constant temperature
/,
by
at constant
there
is
an
increase in the entropy of the system equal to
H H - T~ H
7~T--~^T
t
'
Now
the natural tendency of heat is to pass from hotter to

hence Professor Clausius' theorem that the
colder bodies:
entropy of a system tends to a
maximum 1
General determination of intrinsic energy and

2
entropy. Hr. Kirchhoff
wishing to obtain more general
61.
expressions, took as variables the absolute temperature / and

an undefined magnitude x which may be the density, or pres-
Hr. Clausius 3 , by taking any two

as variables, has introduced the utmost
sure, or refractive index, &c.
and
and
this
method we
magnitudes
generality,
In the most general case

td(f>
dH
hence, employing no brackets
and
if
therefore, as
we
dE
is
will exhibit.
we have
dE + dW',
as there
an exact
difference corresponding
ferential,
xy.
no ambiguity,
differential,
write <&xy for this function of

to
is
As
W which we
further
is
call the
ergonal
an exact
dif-
we have
^
gg-
Pgg-
Phys.
xi.
Ann cxxv
2
(1865) 400.
Pogg. Ann. ciii. (1858) 177.
Ann. cxxv. (1865) 359, and Schlomilch's Zeitschrifc fur Math, und
<
(1866) 31, trans. Phil. Mag. [4] xxxii. (1866)
i.
96
or
whence, by (39),
(40)
On
writing in (40)
for
<)x
^x
~&y
we &et
vT.
(41)
also, since
v (p
and
d(t> is
<)
whence
we
&xy
.1 ~&E
/_
W~\
^x
vlV
(42)
-|-
we have
.1 <)-"
c)
c)
^x
vE
o(p ~~
differential,
(^-.4
this
~&W^\
^y
^x
^-
function
&xy for
&xy being obviously
write
and
oTV
-J-
an exact
^y
if
oE
^n
^y
- ^--- ^- );
of
(43)
the relation between
(44)
If
we had taken
f,
r]
we should
as variables instead of x,y,
have found similar expressions, and since

-.
v.
rf*
"T*
~\
r*
~~~
~\
~\
'
~\~,
we
97
easily find
(45)
Hence, replacing only one of the variables
*>
y?
The
a~~
~ Oj
and b y
by
77,
we have
( 4 5)
equations (41) and (43) are the most general for the
entropy and intrinsic energy: they reduce to Hr. Kirchhoff's

if /
be put
tor
for then
= o,
i,
and
(47)
whence
t
9T,
dx,
_i
d.fl
dx,
(48)
we have
E' being functions of/; or

r
whence, by (26),
"
i
1
(49)
In the particular case where
dw = pdv, we have
(50)
if TT is
written for P.
-:
and the
latter
equation of (49) then gives
(25) or (27) according as v or
p is put for x.
between partial differential coefficients.
and E known as well as
foregoing we may suppose
H
62. Relations
From
the
<
98
hence we
p, v, t;
and
variables
take any two of these five quantities as
may
express the other three in terms of
them
we can
then find the partial differential coefficient of any one with

have
regard to any other while a third remains constant.
We
subjoined a table of the values of all such coefficients in terms

of the specific heats and the simpler differential coefficients:
the relations between them may then be deduced by inspection.
A single example will

}
to find (-
shew how the table has been constructed
thus,
we have
dE = tdj-pdv
whence
but, since
d<t>
we
= K
have
(-r
-:
&c.,
last
<>t
,<>v
T-
or (T^
dv
therefore
We
(~ST~)
thus find by inspection
which relations might indeed be found by geometry: the

may be expressed as
of them by (41) and (50)
63. Application to perfect

equation being
we have
%p
_R
pv
yp
hence by (u) and (12)
gases.
The
characteristic
= Rt,
,
%v_
___
R_
Wv
__
FTS.
Cons!
dv
/sr
[To face
p. 98.]
Fundamental Equations,
99
or the specific heats are independent of the pressure or density.
By
(15)
or the
we have
between the
difference
specific heats
is
a constant
quantity.
In
this
case
t
d/>
~-
O/
hence by (22) the
=p
it
is
c/
intrinsic
energy
=[
JS-JE
or
<)w
e
and therefore
is
=w
given by
\dt,
(53)
a function of the temperature only and entirely in-
dependent of the pressure and density.
On
there
Professor
is
Rankine's
no heat spent
quantity for
all states
hypothesis, equation (28) shews that
in internal
work,
this
being a constant
of the gas,
7=3.
(54)
Hr. Clausius' equation (35) gives us instead

(55)
,-'
4)
JtQ
whence
and
therefore
7-7 =
rt
/
(56)
&/<//,
/Q
which vanishes only if E is zero, being in any case independent

of the pressure and density.
Thus no internal work is done in
altering the
volume of a perfect gas when
temperature remolecules are too far
mains constant
its
but this implies that its

;
apart to act on each other, and therefore there cannot be any
internal
work
at all
however
its
state alters
hence
for a per-
It must be zero therefore for all

gas & must be zero.
bodies that can be brought into the perfectly gaseous state, as
it is a function of the
temperature only and not of the density
fect
to
be as general as possible, however, Professor
Rankine
ioo
writes equation (31)
neglecting the constant
as
(57)
where
'a function of the temperature alone which is o for

temperatures at which the substance is capable of approximating indefinitely to the perfectly gaseous state, and is int
is
all
troduced into the formula solely to provide for the possible

existence of substances which at some temperatures are incapable of approximating to the perfectly gaseous state V
M. Regnault has shewn that

is constant for all rarefied
at
all
and
permanent gases
pressures
temperatures on the airthermometer scale;
same
scale
this
as our absolute one,
gas K, and therefore
being very approximately the

that for a perfect
we may assume
k by (52),
also
constant.
is
On
this
supposition (53) gives
=kt+3,
where 3
(58)
29 representing the molecular

in
the
case
of
indefinite rarefaction, which is
potential energy
is
the constant of
the whole intrinsic energy
when
the temperature
is
infinitely
small.
Since
K and k are constant, their ratio K
so too, and there-
is
17 the ratio of the elasticity at which we may

by
to the elasticity at
call the elasticity at constant entropy e^
fore also
We may now
perfect gas.
e,
and we have
find the equations to the thermal lines of a
For the isothermals we have
at
once from the
characteristic equation
pv
and thus
= Rt = c
nst.,
(60)
the isothermals are equilateral hyperbolae.
Again, on the above assumption the isenergics are given by
pv
= Rt=
(E
3)
const.,
J>
Phil. Trans, clx. (1870) 281.
(61)
101
In
and are therefore coincident with the isothermals.
therefore the points D,
whole
lines
The
this
47, coincide, as
of Fig. 25,
case
do the
AD, AE.
isentropics or adiabatics are also easily

/
_ ^p _ R _
1 ~~~Yt~l)~
found
for
Kk
v
and therefore
which on integration gives
/,-!
Multiplying this
first
= const.
and secondly by
by
both
which quantities are constant, we get

pv*
const,
t K p*~ K
= const.,
These equations may be
the former giving the adiabatics.

written
^(jr,
fo
Po
r$>"? r
'o
= <>"'
PQ
the suffixes denoting initial values.

These relations were found
1
2
also by
Laplace and Poisson on the caloric theory
MM.
of heat, as they too admitted that K is constant.

We might have found the isenergics from the property of
isentropics given in
and/
VQ
ing to
may be
33
for
taken as the current coordinates correspond-
The work done by

a perfect gas during any change at
is
dv
=t\og
=pvloge
Mecaniqne celeste, xii. 127.

Ann. de ch. et de ph. [2] xxiii. (1823) 15, or Traite de
Mecanique,
(1833),
ii.
647.
2 de ed.
IO2
and the work done along any isenergic
given by the same
is
expressions.
The work done along an
isentropic or adiabatic
is
by
32,
(7),
(64)
were the more general one p m v n
If the curve
the
work done
In
= pv n
Q
then
is
this case the
change of
intrinsic
energy being k(t
),
we
have to add heat equal to
the quantity
mn
called the
is
change of
relative to the
specific
For
state in question.
formulae are indeterminate and
heat of the gas
=n
we must proceed ab
these
initio
as
above for the isothermals.
numerical example or two will clear up all difficulties. We will

a very approximately true assumption if we
suppose air to be a perfect gas
--
take
or
-^-
perature.
2^2-8
pressure of 76 c. in lat.
square centimetre (see
air is
0-00129279
we have
and
instead of
specific heat
5), is
fav
is
i c.c.
273-7
of air at o
c.
as absolute zero tem-
and under a barometric
c.
is, under a pressure of 1013226 dynes per

o 001292 79 g., or the specific volume of this
45, that
773-521
R=
The
c.
Further, the mass of
c.c.
hence, measuring pressures in megadynes,
1-013226x773.521
0-2389 (Hr.
Wiedemann's
^
1
value) or
K = 0-2389 x 41-55 = 9-9263,

jr
= 1-4072.
k = K-R = 7-0538, K =
1
gg- Ann.
clvii.
(1876) 21.
103
Consider unit-mass of this air enclosed in a cylinder at 44

1
1 atmospheres of pressure
i| megadynes per sq.
(i.e.
-- = 606-672
X 316-8
-
2.8721;
= -
c,c.
c.
and under
so that
c.)
"0
let its
fall
pressure
While
(i)
it
to
atmosphere
neither gains nor loses heat
l--.I.-,
v
W=
and
W measures the
While
(2)
*('o-0
7-0538
x(-
3
.-.
2'
P
e
-=
x 316-8
281-6
8-8c.
316-8)
248-598 megalergs,
loss of intrinsic energy.
- = h- =
W = Rt \og
kept at constant temperature: here
it is
V.
...
3 3
here
2-8725
X.
X 606-672
910-008
316-8 x (2-3026 X log 1-5)
c.c.
368-976 megalergs,
V*
and
measures the amount of heat that must be
While
(3)
kept at constant volume
it is
_..*
/.
= -x
316-8
3
A 3
E-E = k (t-( = - 7-0538 x
supplied.
here
211-2
or
-6i-6c.
A,
and
this loss
Let
(4)
i.e.
to
-v
it
megalergs,
expand under constant pressure to the same volume as
in (i),
here
^,
/.
= -X
316-8
422-4
or 149-6
(v-v } =
1-5 x
E-E = k(t-t =
(O
606-672)
7-0538 X (^ x 316-8)
3
t)
and the heat that must be supplied
EEQ+ W =
64. Sensible heat.
303-336 megalergs.
745-795 megalergs,
is
1049-131 megalergs.
We
have presented in
in favour of supposing sensible heat to
kinetic energy,
the density,
c.
ô
W=p
ments
= - 745-795
of intrinsic energy measures the heat that must be removed.
-= ^=
.4>
105-6
and therefore
and we have seen
pothesis leads to the
same
in
all
in
result
18 the argu-
be molecular
probability independent of
59 how Hr. Clausius' hyviz. that it
is
a function of
IO4
the temperature only.

The question now arises whether all the
heat contained in a body is at the temperature of the body, and
Hr. Clausius concludes

the
same temperature
that
'we must not ascribe one and

we must
to the whole quantity of heat
rather conceive this whole quantity to be divided into an infinite
number of elements, and
to each element
we must
consider that
temperature to correspond which the body on cooling would

have at the moment when it parted with this element, or that
which on being heated it would have at the moment when
it
received this element of heat/
For the transformation-value
of the heat put into unit-mass to raise
of
its initial
condition
is
its
temperature from that
so that each increment of heat
spent in raising the temperature is a function of the temperature

which it is taken. And unless we accept this conclusion
Hr. Clausius' fundamental axiom is contradicted in the fol-
at
lowing process which was suggested by M. Him.

M.
considered a theoretical machine consisting of two
2
Him
A and B
cylinders
equal in section and communicating with
each other by the tube O which opens into

their bases
they are provided with pistons
which move without any
and the
friction,
piston rods are indented and work into the

teeth of the same wheel, so that one piston
always move up as much as the other

moves down, the volume between them being
will
The cylinders contain perand, the pressure being therefore

the same throughout, the pistons are in equi-
therefore constant.
fect
gas,
hence no work will

librium in any position
be done in any small motion of the pistons
:
and therefore in any motion so long as it is

slow enough. The whole apparatus is supposed impermeable
by heat except the tube of communication which is kept
at
the constant temperature 3.
Consider
Pogg. Ann. cxx. (1863) 449.
Theorie mecanique de la Chaletir,
now /
ed. (1875),
i.
z>
255.
to
characterise unit-mass of the gas initially
in the cylinder
(the piston of
105
when
it is all
collected
lowest position),
and the constant volume
being in
and suppose the total mass is

between the pistons V. The piston of
its
then depressed
is
and each successive portion of gas is heated to 3 as it passes

through the tube 0; it therefore expands or tends to expand,
and thus increases the pressure, causing in consequence an
When
increase in temperature in every other part of the vessel.
all the gas is transferred to B, we shall, by reason of the extremely low conductibility of gases, have a series of layers at
3 and that
much above S, and it thus appears that without
expenditure of work we have caused heat to pass from the
different temperatures, that of the lowest layer being
of the highest
the
tube
which
to gas at a higher temperature
is
contrary to our
axiom.
We will
at
mean temperature
calculate the
any moment of a layer of gas
the vessel being
p and
T in B. If t is the temperature
A or B, the pressure throughout
pv
ft dm = MT
V,
= Rt
p fvdm
always
';
subsists,
= fpvdm = Rftdm,
pV=RMT:
or
but
in
the specific volume of this particular layer v, then
fvdm =
and, since the relation
dm either
we have
further
rn
and therefore
t,
Now
the temperature of the gas in A has become t

pressure without the communication of heat; hence
K
K
}
-"
t'
2\
(\
by the alteration of the

by (62)
I
Again, as the mass dm of gas passes through the tube, its temperature
is raised from / to S, and it
slightly expands since it modifies the total
pressure only infinitely little : hence it takes in heat at constant pressure
and not
at constant volume,
dm
and the heat thus taken
in
is,
by
/*
)dm:
15,
this causes
a variation
dT
in the
mean temperature
of the whole mass at
constant volume, whence
which from o to
integration of
all
(i.e.
corresponding to the passage of
the gas into 5) gives
dT
T
and the solution of

1
If
we put
tQ
<c "~
L-
-r
== ^J
this equation will give T, the final
mean
temperature.
=$x,
so that
xa
= A,
the equation becomes
I-X
Taking
or
rarefied air to represent perfect gas
2-456:
then, if
we may assume
1-4072,
0-5 so that the value of the constant
is
= 0-6306, whence if f be oc.

0-0742, an approximate calculation gives x
or 272-8 (see
63) and therefore 3 = 545-6 or 272-8c., we find 7*= 607-6
or 334-8 c., which is 62 degrees above the temperature of the tube.
If the heat in the gas
were
all at
the temperature T, then the
axiom could no longer hold; as without any expenditure of

work a body would have been raised to a higher temperature
than that of the source of heat: Hr. Clausius however points
out that for any part of the gas in B to be raised above 3 heat
must be given to some part of the gas in A at a temperature
lower than 3, and this is the compensation that renders the
phenomenon possible. We shall shew in 66 that no work is
gained by the process.

65. Real specific heat.
The
coefficient
or
fc
is
termed
indifferently the real specific heat, the capacity
absolute thermal capacity of a body, these
to signify that
it is
for heat, or the

names being intended
independent of the density, i. e. of the moleNow in one series of arrange-
cular arrangement of the body.
ments
those of the perfectly gaseous state
of the temperature
in
all
arrangements.
and
it is
independent
therefore probably constant

Professor Rankine considers it to be
also,
it
is
Fundamental Eqtiations.
107
constant only throughout each of the different states of aggregation, solid, liquid, or gaseous, in which a body may exist, and
to differ in the different states
2
out
1
;
Hr. Clausius however points
going on just as much,

any one state as in the passage from
that molecular alterations are

in a less degree, in
though
one state to another, and that therefore it is rather arbitrary
to assume for the smaller changes what is denied for the larger.
Hence we assume
to determine
it
to be constant for each substance, and,
fe
therefore,
we ought
to determine the specific
heat at constant volume of the substance
when
in the state
of a
of highly superheated vapour.

This operation
perfect gas,
for the conversion into perfect gas
is however doubly difficult
of most substances requires temperatures not readily at our
i.e.
the specific heat at constant volume of even an
command, and
ordinary gas has as yet been only approximately determined

MM. Dulong and Petit' s law however, that the
74).
(see
heat of unit-mass of all bodies is inversely proportional to the
atomic weight, or that the atomic heat of all bodies is the same,
comes to our aid
this law cannot indeed be true for the
specific
specific heats at constant pressure, as its discoverers imagined,
by reason of the varying external and internal work; but Avogadro's law, that in equal volumes of all substances in the
perfect state at the same pressure and temperature there is the
same number of atoms, coupled with MM. Delaroche and
Be'rard's experimental conclusion that equal volumes of substances in the perfect state require equal
an equal
rise in
where n
is
amounts of heat
for
temperature, gives
na\
const,
or
a\
= const.,
number of atoms
in a given volume and a the

thus the specific heat in the above law is the real
3
specific heat, as has been proved otherwise by M. Moutier
the
atomic weight
To
determine
this
we will
now since
constant
take the case of the most
the mass of i c.c. at oc.

perfect gas, hydrogen
under a pressure of ?6c. of mercury at Paris (i.e. of 1013573
:
A Manual
of the Steam-engine, &c., 4th ed. (1869), 307.

cxvi (1862) 99, trans. Phil. Mag. [4] xxiv. (1862) 205.
Bull, de la Soc. Philomat. [7] i. (1877) 3.
gg'
Ann
io8
dynes per
sq.
is
c.)
stant pressure being

c.c.
the volume
g.,
of
i g.
is
and, the coefficient of expansion under con-
c.c.,
0-003661
0-000089578
0-003661, the expansion per degree
is
and the external work done
0-000089578
0-003661 x 1-013573
-
= 41-329
0-000089578
but the specific heat at constant pressure
3-409x41-55
megalergs:
is
141-644,
so that the specific heat at constant volume
141-644-41-329
is
100-315.
Taking then as usual the atomic weight of hydrogen
we have
ok
in
as unity,
for all substances
=.
100
round numbers.
Thus
for iron,
whence
55-9
ft
1-7889
but
# = 0-1 1379x41-55
and therefore the heat spent
in internal
4-72801-7889 =
now
the density of iron

,
is
so that the external
about
18000
7-75
its coefficient
work done by a
rise in
of expansion about
temperature of
X 1-013573 =0-000007 megalergs, which
work
thus the internal
is
2-9391 megalergs:
about 7-75 and
18000
is
=4-7280,
and external work
is
degree
insensible
requires 2-9391 megalergs of heat or 1-643 times
mere change of temperature.

206-4 an d & = 0-4845 also
that required for the
For lead a
K = 0-03140
and the external
work done
x 41-55
1-3047,
is
equally negligible
is 0-8202 megalergs and requires 1-693 times the
merely raising the temperature.
Similarly Hr.
Wcestyn's law
ft
where n
is
the
- na\2
na
gives in the case of
loo2
number of atoms
element whose atomic weight
hence the internal work

amount of heat spent in
is
a.
compounds
n
>
na
in the
compound of
that
Fundamen tal Equations.

Thus
and, since
for
CO
carbon dioxide
K=
109
a,
0-2169 X 41.55 = 9-0122, the internal and external work is

but, the normal density at Paris of CO 2 being
2-1769 megalergs
0-001977414 and
its coefficient
0-003710. the external
of expansion under constant pressure being
work done
is
-- =
1-0017, so that
0-001977414
the internal
work
is
only 0-2752 megalergs, and requires
th of the heat
that merely increases the temperature.
For water
H O we have similarly
2
ft
and
100 x
(
V
2X
therefore, as the external
work when a gram of water

41-55
(i
+0-00022
0}
+ IX
work
at
is
16-704
16-704,
negligible, the heat spent in internal
c. is raised
15-96'
one degree
is
by
10
24-846 + 0-009140 megalergs.
CHAPTER
VI.
66. Available energy. We cannot convert into work the
whole of an amount dH of heat at temperature /, as shewn
in
34
its
practical value for useful transformation is
*-^dH
where
dH-t*,
or
the lowest available temperature, being that of a

of
dimensions and capacity for heat large enough
refrigerator
to preserve its temperature sensibly constant when heat is given
up
to
it
is
thus the
work
that
can be got out of an amount fdH
of heat by any process whatever
When
which
is
only
the process is reversible, this work has a maximum value

is called the available part of the heat-energy
fdH, as
represents the utmost amount that is capable under the given

circumstances of being transformed into something more useful.
it
It depends on temperature only, and is obviously the greater

the higher the temperature of the body in which it is and the
lower that of the refrigerator.
If this heat has been supplied at the expense of the intrinsic

energy of a body which is surrounded by the refrigerator, and
this energy has changed from E to EQ in the process, we have
fdH
also
we have
for reversible processes,

initial
and
final states;
where
and
</>,
(j>
^>
are the entropy in the

are perfectly de terminable
in
when
given
the final temperature /
and the
final pressure pQ are

of a body, enclosed in a
and constant temperature / ,
hence the part of the energy
medium
of constant pressure
pQ
that can be usefully transformed, or the available energy of the
body under the given circumstances, is

2?-2Z -/ (<-&).
(2)
Thus
the greater the original entropy or the higher the final

temperature, the less the available energy.
Professor
Maxwell
Let A,
gram.
body: through
B
A
has shewn
this very elegantly by a diarepresent the initial and final states of the
draw the isometric A a and the isentropic A a.
meeting the axis of v at an

infinite
draw
and
through
distance;
similar
lines
Bb, Bfi,
also the isothermal
then the path-line
AB
BC\
repre-
senting the successive states

of the body must lie below A a
no body of higher
temperature from which heat
may be received, and it must
as there
lie
as
is
is
above the isothermal
BC
the lowest available temperature.

The limiting path
therefore ACS, formed by the isentropic and isothermal:
/
is
now by
(2),
33, the heat given out along this path, /
(<
</>
),
represented by the area BCafiB and is obviously a minimum under the given conditions, and by (i), 33, the loss
is
of intrinsic energy
area BbaAafiB-,
difference
ACBbaA,
is
EE
is constant, being represented

Q
by the
hence the work done, which is equal to the
^((f>
a maximum.
(f)
and is represented by the area

)
Therefore the area ACBbaA repre-
sents the available part under the given circumstances of the
energy corresponding to the
state
which
is
represented by
AavaA + 3.
Applying this result to our case in
64, we shall obtain the greatest
amount of work, while bringing back the heated gas to its initial condition,
1
Theory of Heat, 4th ed. (1875), 187.
if
we
(i) allow
to
it
from
temperature
falls
and
compress
(2) then
initial
and
value pQ
it
at constant temperature
in the first process
second
in the
expand without receiving or losing heat

to / and its pressure from
we
till
its
we
shall spend an
shall gain
pressure rises to the
till its
an amount of work k (T
/),
amount
R
hence the
T K_i
to
log*
(y )
maximum amount
*/ log*
T
-
(since
k)
of work which can be got out of the heated
T
gas as
it
of heat
returns to
k(Tt^
its initial state is
log e
),
and as an amount
has been communicated to the gas, the efficiency of the
whole
(T
cyclical process is
=^
log^
The maximum
efficiency of a
closed cycle in which the temperatures are those of the tube and of the gas
in
its
initial
state is
now
while this in our given case
is 0-5,
the
\7
efficiency of the other process is only 0-3475, or there is
a greater
sacrifice
of heat to the refrigerator hence if without expenditure of work we can

continuously bring a body to a higher temperature than that of the source,
it is simply because in place of sacrificing work we offer to the refrigerator
:
a continuous and equivalent sacrifice of heat.
We
may consider the more general case of an irregularly

heated body in a perfectly non-conducting vessel of no capacity
if all transferences of heat between the different parts
for heat
:
by perfect engines till the whole assumes an uniform

temperature T, the work so done measures the available energy.
Here is no infinite refrigerator for use with the perfect engines,
are effected
we may
but, as its only requisite is a constant temperature,
employ instead any part of the substance whose temperature

is T, such
part on the whole receiving no heat during the
it serves also as source with regard to .parts at
as
processes
lower temperatures.
change
Let
be the
specific heat of the
mass dm
(x y, z) for the particular circumstances of the

of temperature, being K, for example, if the change
at the point
takes place under constant
pressure
robbed of heat by the perfect engine
then as
it
this
gives
up
mass
to
is
the
rt
refrigerator the heat

rt
being
c^/(i
dm\
JT
ceives
no
heat,
dm
idl-i the work done by the engine
--T
t
113
On
the whole the refrigerator re-
and therefore
'dt,
the integration extending
The work done by
all
over the body
all
the engines
is
(3)
hence we find
W= jdmj* tdt(i - y) ^jdmTtdt,

by
(3),
and
(4)
energy of the body in
this represents the available
condition.
its initial
If
T.
similarly
is
body and thus a function

=ftdm\ hence (3) becomes
the thermal capacity of the
of the temperature,
we have
pe~dt=o
JT
Cf
or
Jt
~dt=
7
CTQ
\
JtQ
-<//,
t
being a conveniently low temperature. The right-hand expression might be tabulated, the whole body being supposed
for its calculation to be of uniform temperature:
then T
would be
that value of / for
which
the left-hand side of the equation.
W=
Ifc
is
this
expression is equal to
should also have
We
f e dt.
JT
independent of the temperature, (3) and (4) become
f C log t dm
=o
T=
or
10
W=fdmt(t-T).
If the
of
it,
body is also homogeneous, c is the same

and we get the further simplifications
M being the whole mass.
If the
i
body
for every part
consists of masses
m 1 at
ii4
temperature
/1?
at
temperature
geneous material for which

tions
is
/2 ,
...
of the same homo-
independent of
/,
these equa-
become
2. wt log/
r=
*-
10
= (/^/^
W=t(2.mt
i7
...)
**,
TZ.m).
This method was first given by Sir W. Thomson in 1852:

the above equations are his, as slightly simplified in expression
by Professor Tait.
1
The
still
more general case we might consider of
several
pressures in the nonthe final uniform state as to pressure
irregularly heated bodies
at
different
conducting enclosure
to which the system may be brought when all
transferences of heat are effected by perfect engines is the same
:
and temperature
in whatever way or order the engines act

hence we may bring
each body to a state of uniform temperature, and then the
system to an uniform temperature, whence the theorem the
:
of a system is the sum of the available energies of

each of its parts together with that of the system consisting of these
parts each with its individual available energy exhausted.
available energy
The
above
equations in this case are of the same form as the

now equation (3) indicates that the entropy of the
system must remain unchanged if the maximum amount of

is to be obtained, and thus Professor Maxwell's pro-
work
posed method
is
suggested
(a) Bring each of the bodies without communication of
heat to what will be the final uniform temperature of the
:
the entropy
system, this temperature being calculated by (3)
of each is unchanged by this process.
com(b) Let those bodies that exert the greatest pressures
:
press the others, the process being conducted so slowly that

is no sensible variation of temperature in any part: as
there
communication of heat occurs only between bodies of the same

temperature, the total entropy is unchanged.
The entropy of the system being thus constant throughout
1
v.
Proc. R. S. E.
(1853) 102.
iii.
(1852) 139, or Phil. Mag. [4]
iv.
(1852) 304, and [4]
the whole process, the total
change
measure the work obtained, and
in
115
energy of the system
will
will
therefore represent the
These operations might have been suggested

a
by consideration of the above graphical method.
know energy under many
67. Dissipation of energy.
available energy.
We
different
us
forms which are of different orders of usefulness to
thus
is
it
most
useful to us as mechanical effect,
and as
heat of high temperature it is more useful than as heat of low

temperature we may therefore class one form as higher than
another. One form may in general be transformed into another
:
either wholly or in part
Law
1
,
but subject to what
of the Degradation of Energy,
are accompanied by a degradation of energy,

tion
to
may be
called the
all transformations
viz.
by a transforma-
i.e.
less
of energy
useful form.
This was first recognized by Sir W. Thomson 2 in the case
of heat and mechanical effect, and he called it the Law of the
The former term is better,

Dissipation of Mechanical Energy.
both as indicating what has become of the energy that it is
not destroyed but present in a lower form and also as more
His reasoning, which depends on Carnot's principle,
can always transform mechanical
thus be exhibited.
general.
may
We
energy entirely into heat, the transformation being to a lower

form but the work that is obtained from an amount of heat;
fdH
energy
by any process
quantity
is,
by
than
(i), less
fdH
by the
fdff
'J'
where
is
the lowest available temperature,
and
therefore represents the energy dissipated.

reversible, this dissipation is a
where $,
/o($ </><>)>
(j> Q
the heat is taken in its
minimum, and
are the entropy of the

initial
and
not reversible, the dissipation

actual waste of availability equal to
is
greater,
expression
then equal to
body from which

If the process
and there
is
an
See Professor Balfour Stewart's Conservation of Energy.

Proc. R.S.E. iii. (1852) 139, or Phil. Mag. [4] iv. (1852) 304, and [4]
(1853) 102.
I 2
v.
it is
final states.
is
this
If the process is
n6
From
the conditions of reversibility in
reversible
we
51
see that a
impracticable ; for it is impossible to

prevent either friction in the parts of a machine or thermal
is
process
communication between parts

conduction
or
radiation
at
different
therefore
temperatures by
waste is not pre-
this
ventible.
It is found that all other forms of energy tend to be transformed into the energy of diffused heat, but that a re-transformation cannot occur except on the expenditure of work or
a sacrifice of heat to a refrigerator
in either case a loss
of
hence dissipation accompanies every transformation

of energy, and as every operation in nature consists in such
:
availability
transformations,
follows
it
energy of the universe,
the
that
while remaining constant in quantity, is gradually being dissipated or degraded; its final state must therefore be one of
uniformly diffused heat.
we cannot
thus energy that

energy: could we
is
Dissipated energy
direct at will, such as molecular
atom
follow each
would be no
in
its
course and seize
it
at pleasure, there
dissipation.
In the general case of
many
irregularly heated bodies in a
non-conducting enclosure of no capacity for heat, the effect of

radiation and conduction is to diminish differences of tem-
and thus the
available energy of the system

but
not
the
does
when
the
temgive
dissipation
equation (4)
perature has become uniform, for the final temperature will
perature
be greater than
in this case
T, since
no heat has been removed
form of work done, as there supposed. The real conif T' is the final uniform
ditions are these
temperature, the
f
heat gained by a particle at temperature T from all the others
in the
by radiation and conduction
is
JdmJ
whence
transferring
has
left
also the
by
tdt
that
zero, or
= Q>
(5)
could have been gained in
perfect engines to this particle
the others
W*
work
on the whole
dm
all
the heat that
is
tdt(i
T'
dm
dt,
(6)
by the preceding equation, and
this
117
represents the
dissipa-
tion.
If the
is
system
homogeneous and
become
is
independent of the
temperature, these equations
IV*
68. Possibility of restoring lost availability. Professor

Rankine suggested 1 that it might be possible to reconcentrate,
as
it
were, dissipated energy or to restore to it availability for

for as diffused heat ultimately takes the
useful transformation
radiant form (the ether not being thereby raised in temperature

as it is perfectly diathermanous), this radiant energy on reach-
ing the confines of space

ultimately reconcentrated
may
possibly be totally reflected
and
'
of such intensity
that,
should a star (being at that period an extinct mass of inert
compounds) in the course of its motions arrive at that part of
into
foci
it will be
vaporized and resolved into its elements, a
of chemical power being thus reproduced at the exThis
pense of a corresponding amount of radiant heat.'
hypothesis really assumes that at a focus of concentrated rays
space,
store
a body may be heated to a temperature higher than that of the

bodies emitting the rays, and is therefore contrary to our
fundamental axiom that heat cannot of itself pass from a
warmer body.
Hr. Clausius has further refuted
colder to a
tailed discussion
lowing summary
this hypothesis by a deof the theory of radiation 2 he gives the folof his principal results
:
'To
bring the effects of ordinary radiation, without concentration, into harmony with the principle that heat cannot of
i.
pass from a colder to a warmer body, it is necessary to

that the intensity of emission from any body depends
not only upon its own constitution and temperature, but also
itself
assume
upon
the nature of the surrounding
tensities
1
medium
in fact the in-
of emission of one and the same body in different

B. A. Rep. (1852)
nn cxxi
-
ii.
12, or Phil.
Mag.
[4] iv. (1852) 358.
1 1
media must be inversely proportional to the squares of the

velocities with which rays are therein propagated, or, in other
words, directly proportional to the squares of the coefficients
of refraction for these media.
assumption as to the influence of the surrounding

medium be correct, the above principle must obtain, not only
2.
'If this
when the rays become conmanner

whatever
through reflexions and
any
for ordinary radiation, but also
centrated
in
for
refractions;
although concentration
the ab-
may change
solute magnitudes of the quantities of heat which two bodies
radiate to each other,
cannot
it
of these mag-
alter the ratio
nitudes/
We
can
easily
shew the
case of reflexion.
latter in the
normal
fall at
area da of
at
at incidence i'
da cos
i'
and, e being the normal
of A, the
radiation
then the
cone of rays
solid angle of the incident

is
Let
an angle i to the
a distance r on the small
radiation leaving the small area da of
total
on
radiation falling
da' cos
ffe cos
da
amount of
from
is
z''
;
Fig. 28.
similarly, if
of B, the total radiation falling on

r>
/*
ffe
and the
ratio of these
suppose the rays from

falling
on
falling at
is
cos
*'
two
da
is
from
aa COS
/
.
reflected
is
I
>
quantities
first
the normal radiation
is
obviously
Now
by the surface C before
then the radiation from the small area da of
a distance p upon a small area da of
at incidence
by the foregoing
.
where da
da cos
da cos
or
is
a co a a cos
t,
the solid angle subtended at the ele-
119
ment da by the small area da; and similarly the radiation from
the small area da' that falls upon this same element da is
/ dto da cos i
if then da is the element of C at which the
from
da
are
reflected to da', so that one is the image of
rays
the other, the laws of reflexion give da) = du>' and t = t', and
therefore the ratio of the radiations from da to da' and from
:
da to da
is
,>
which, being constant,
whole radiation from
Thus
to
to the
is
also the ratio of the
whole radiation from
to
unaltered by the reflexion.

It is to be
noticed that the radiation which falls upon B in the two cases
A.
the ratio
is
does not necessarily proceed from the same parts of A, and

vice versd.
CHAPTER
VII.

APPLICATION TO GASES.
OUR results have hitherto been expressed in terms of temperature on Sir W. Thomson's absolute scale: in order to
make use of actual experiments we must determine the relation
between
this scale
and those
This relation has
in ordinary use.
been deduced from the behaviour of gases during expansion.

Two copper vessels, of
69. Dr. Joule's experiments 1
.
respectively, the first exhausted of

capacity 2196 and 2237
air and the second containing air at about 22 atmospheres
c.c.
pressure, were connected together by a brass pipe provided with

a stopcock and- placed together in a tinned-iron can with double
7484 grams of water at 14-9 c. the stopcock

then
was
opened so that the air expanded into the empty vessel,
of the temperature of the water was observed to
no
variation
but
walls containing
result (even
^^
degree c.)
though the thermometer used read to
were made for radiation and the
after the necessary corrections
warmth produced by
stirring the water to equalize its
tempera-
Joule therefore concluded that 'no change

of temperature occurs when air is allowed to expand without
developing mechanical power (i. e. doing external work).
ture throughout
Dr.
'
To
analyze the above experiment Dr.

Joule placed the vessels in separate cans
of water and the stopcock in a third the
;
air,
initially
under
pressure, caused a
per kilogram of water in A which contained all the air at first and a rise of
Fig. 29.
0-600 degree
fall
20-7
atmospheres
of 0-595 degree c.
c.
per kilogram in B, a rise of 0-078 degree

1
Phil.
Mag.
[3] xxvi. (1845) 376.
c.
per
21
the above
kilogram further occurring in C. This confirms
due to the partial
result, the slight redundance of heat being
A to C through
heating of the cooled air as it passed from
which could not be immersed in water.
that part of the pipe

If this
zero, or
is so,
the internal
58
by
and by
i^*
is
_
=
d/'7
~'
F represents a function of
w =fpdv = /F,
p = ttf,
if
whence,
the expansion of air
work during
z>,
(22),
57,
E-E = \(t-t,l
Q
k being also
by
54, equal to
(13),
M. Regnault's experiments shew

referred to the c. scale
EE
so that
but
if
/,
T refer to
normal pressure and
therefore
K=
-^-
k'
= o.
But
that the specific heat of air
the air -thermometer
the
/
/-/
(T
is
constant,
temperature of boiling water under

to the temperature of melting ice,
= T-T =
O
100
and we have generally
ft,
But TO
on
since
ft
=/ =o
= /+(T
-/ ).
O
corresponding to absolute
or absolute temperature is the
same as that measured on the air-thermometer.
at the temperature
deprivation of heat, so that T
/,
The degree of exactness in the above experiments

be determined. In the first the mass of air used was
274
----x 0-0012932
2237 x 22 x
may
thus
60.355 grams,
and the water-equivalent of the calorimeter-vessels and water

was about 9750 grams: now a variation of i degree c. in
975
S-
tion of
water requires the same amount of heat as a varia-
degree
c.
= 408 12
in
g. of air, or
a variation
0-2389
through
the
= 676
degrees
c.
of the amount of
air
used in
experiment; thus, as the thermometer immersed in the

water read to only
-^
degree
c., it
would not be affected by a
variation in the temperature of the air less than
1-88
Dr. Joule's conclusion is not therefore absolutely

established by these experiments in fact we are going to shew
c.
degree
that
work
internal
to a
slight extent does occur during the

variations in the density of real gases.
The absolute and airthermometer scales are not therefore identical, but differ very
slightly
from each other.
We
should not get any better result by placing the thermometer in the vessel instead of the air, for before an accurate
reading of the temperature of the air could be made, the air
would have taken the temperature of the walls of its enclosure.
Such an experiment was performed by M. Gay Lussac

found a fall of temperature in the one vessel and a rise
other, but could
70.
make no
who
in the
quantitative measures.
M. Hirn's experiment. A more
was made by M.
Him
delicate
experiment
in 1865.
long copper cylinder was
divided into two equal compartments by a membrane, and an
open-air oil-manometer, provided with a stopcock, was fitted to
one of them both compartments were then filled with gas at
:
atmospheric pressure and the manometer stopcock turned, the

oil being at the same level in both arms.
The gas was then
pumped from one compartment
into the other, and,
when
the
became \ and i \ atmospheres respectively, the membrane was broken by the fall of a leaden bullet and the
manometer simultaneously opened. No displacement in the
manometer levels was observed and therefore no variation of
pressures
temperature revealed.
Now a depression of
mm.
of the
oil
in
responds to a difference in pressure of 0-02

_.
_ 0-02
density of oil being 0-8) of - -
76
and
since,
when
the volume
is
0-8
13-596
constant,
the guage corof oil or (the
c.
i
-r
-?
atmosphere
6458
/ oc
or 6/
/>,
the change of temperature corresponding to this change of

1
Mem.
d'Arciieil, i. (1807) 180.

Theorie mecanique de la Chaleur (3 me ed.),
i.
298.
when p
pressure
is
experiment,
of o-o i
LAI
and
285-7
or I2
= 0-044
could easily
c.
/=
c -)
as
if it
m ê
degree.
0458
be observed, even
123
a^ ove
depression
lasted scarcely a
second, and therefore a variation of 0-044 degree would have

itself, if it had occurred.
shewn
The
delicacy of this arrangement
is
of
M.
little avail,
Him
being convinced by numerous experiments that 'even when

the gas
had
fallen
or 2 degrees below the
through
initial
temperature after the membrane was broken, it

so quickly at the expense of the walls of the cylinder that the
manometer would not have time to shew an appreciable fall of
would be heated
temperature/
71. Sir W.
method 1
is
Thomson's method.
due to
Sir
still
W. Thomson, who
to replace the vessels in Dr.
more
delicate
in
1851
Joule's experiments by long
spiral
proposed
still
retaining the stopcock or its equivalent, a partition
with a small orifice, and to force air continuously but slowly
If there is any
through the whole at constant temperature.
tubes,
variation of temperature
orifice, the
of the air in passing through
the
perhaps be again attained
temperature may
only after the passage through a considerable length of the
second spiral, which must therefore be long enough to ensure
initial
this.
The first spiral is required merely to fix the temperature
of the entering air, and the process is to be so managed that
the pressures of the entering and issuing air do not vary.
The
velocity of issue
is
to differ very
that the pressure is
sistance encountered
employed
by the
little
from that of entrance, so
entirely in
air,
and not
overcoming the
in giving
it
re-
kinetic
energy.
Let
when
/>,
v characterize the entering and p', v' the issuing air

motion has become steady that is, when equal
the
amounts of
air enter
and leave the
friction in the rapids (the orifice
the
In general the fluid
neighbourhood) will
the same time heat may be required
and at
work of expansion;
give rise to heat,

for
tube.
and
Trans.
R.S.E.
the
its
balance therefore
xx. (1851) 294.
will
be
124
absorbed by the water in which the second spiral is immersed,

and may thus be measured. The heat that results from the
friction of the particles against each other and the
edges of
the orifice is equal to the mechanical work that would have
been done if the expansion had occurred without
any friction
it can thus be measured
by our supposing the pierced partition
removed and a series of moveable pistons without inertia sub;
stituted,
between each successive pair of which unit-mass of air

is enclosed, their motion being
such that each piston A moves
past the position of the orifice
^ till the next piston B reaches
0, the volume between them
A
I
~T
being constantly v, that B then

remains at rest till by the con-
Fig 30
tinued motion of
between them has become

this
z>',
that they then
the volume
move forward
During these two processes
volume unaltered.
with
for each
/v' (pp'}dv
.
respectively, the
in the integral being variable
sum
hence
pfo+pvprf
equal to the heat caused by the fluid
friction.
Again, during
the expansion at constant temperature heat equal to
be absorbed
their
/v
JV
is
but by
48 the general expression for
will
fldv
/ is
where p is Garnet's function and / is the temperature

measured on any scale whatever; supposing therefore the
scale to be that of the air-thermometer and consequently replacing / by T, we have for the heat absorbed
-F
ttj.
being placed outside the integral as it is a function of the
The heat taken up by the calorimeter is thus
temperature.
fj,
+H
state
12$
denoting as usual the heat taken in by the air as it changes

and whatever changes of temperature there may actually
;
'
air in or near the orifice, this expression will give

total quantity of heat emitted by that portion of
the
rigorously
the tube which contains the orifice and the whole of the second
be of the
spiral
during the passage of a volume v through the
first spiral,
or if through any portion of the second spiral where the temNow absolute temperature is defined
perature is sensibly r.'
as the reciprocal of
Car not' s
C
I
We
know
function,
and therefore
pdv+pvp'v' + H
not the true relation between />,
v,
and
or
r,
we could
equation the exact relation between temperature on

but the approximate
the absolute and air-thermometer scales
formula
RT
^v
find
by
this
gives the result
__ j
(a)
log,f
which
Sir
and
is
very approximately true.
W. Thomson proposed 1
also to
modify
this
experiment
to determine
by a thermopile whether cooling or heating

accompanies expansion, one set of the solderings being fixed
within the tube on one side of the orifice and the other set
within the tube
the rush of
air.
on
the other side, both sets being placed out of

Supposing as before that the air is flowing
gently even though the pressure may be very high (the pressure
being thus employed merely in overcoming
^___
the resistance introduced into the tube by
?B
!-Ax *&
'
the pierced partition), there will be a section A of the tube not far distant from the
|y
rig. 31.
orifice,
everywhere behind which (/. e. in a

away from the orifice) the pressure, specific volume
direction
and temperature are constant and represented by

1
Trans. R. S. E. xx. (1851) 481.
p, v, T,
and
12,6
there will also be
on
the other side of the orifice and near to
it
a section B, everywhere behind which the pressure, specific

volume and temperature are constant and represented by />',
&', r',
the rapids being between these sections now in forcing

air across the section A we spend work equal
:
unit-mass of
to pVj but, as unit-mass simultaneously crosses the section
p if we communicate to it energy only

equal to pvp'v'. It may also absorb heat H from surrounding bodies, and it may lose part of its energy 2 in producing
doing work equal to
sound as it passes through the rapids

increase of intrinsic energy
thus
H^+pvp'v\
its
intrinsic
and behind the sections A and B,
at
energy
experiences an
f
E, E denote
it
or, if
E -E = H-V+pv-p'v'.
f
(3)
In the case wherein loss and gain of heat are prevented and the
energy of sound vibrations is negligible the equation becomes
= E'+p'v',
d.(E+pv) = o:
E+pv
F
(4)
the thermal capacities to shew that they are

of temperature on the air-thermometer,
in
terms
expressed
but, accenting
we have
generally
= (dE+pdv) + vdp = K' dr + (L + v) dp

f
'
d.(E+pv)
since
by
L'
= /'^,
and
/'=-^,
dT
also
^p
Hence
in this case,
Cp CT
fJ.
when
is
--V
dT
i
^v
or, as absolute
16,
(7),
temperature
is
= -^
zero,
A^//^
V
T\
d/>
I,
the reciprocal of
/*,
CT
Now
the approximate formula
pv
dz>
K*
and thus
127
= R T gives
v
<)
"
or substituting differences for
we
differentials, since
they are small,
get
Ji
log,/- log,/
(6)
5 being
the resulting increase of temperature (measured
on the
from p to p'.
air-thermometer) during the change of pressure
This equation is approximately the same as (2) ; the signs of
the second terms of the right-hand expressions are the
since, if heat is absorbed by the air or
positive, there
same
must
5 negative.
Joule and Sir W. Thomson's experiments.
have been a cooling or

72. Dr.
latter of these proposed methods has been extensively

1
used by Dr. Joule and Sir W. Thomson working together
but the pierced partition was replaced by a porous plug, as with
a mere orifice a thermometer placed in the issuing air underwent different actions according to its position and shape.
The
The nozzle finally adopted was attached to the upright end

of the long spiral pipe aa of copper through which the air
was forced it consisted of a perforated metallic disc b resting
:
on a shoulder within
the pipe, on which stood a short piece

of india-rubber tube cc enclosing a silk plug d, that was kept
in a compressed state by the upper perforated metallic plate e,
against which the metallic tube ff was pressed down by means
of the screw
1
gg
a tube of cork hh was placed within
ff
to
Phil Mag. [4] iv. (1852) 481; Phil. Trans, cxliii. (1853) 357, cxliv.
(1854) 3 21 J Proc.R.S.x.(iS6o) 502 ; B.A.Rep. (1861) ii. 83; Phil. Trans.
clii. (1862) 579.
128
protect the bulb of the thermometer from the effects of a too
rapid conduction of heat from the bath, and cotton wool was
packed loosely round the bulb so as to distribute the flowing
air as
the tube zt, attached to the nozzle by

evenly as possible
india-rubber tubing, was of glass, in order to
:
permit readings of the thermometer. This nozzle

obviated the rapids and caused the work done
air to be immediately
spent in
without generation, even temporarily, of
The difordinary kinetic energy or of sound.
by the expanding
friction
ference between the pressures on each side of

the nozzle was varied by altering the amount of
silk or cotton compressed in the plug.
The
Fig 32.
was always found
issuing air
to be cooled
by the expansion, but several precautions were

found necessary for quantitative measurement:
thus (i) the cooling was greater at first owing to

necessary drying of the material (silk or cotton) of the plug
of temperature were caused by the intermittent
(2) oscillations
:
action of the
pump, from which
of dust were carried over
also particles of sperm-oil or
the air was therefore
made
to pass
through a box with perforated caps and stuffed with cotton

wool before entering the spiral (3) a series of fluctuations in
:
the temperature occurred on opening the stopcock which allowed the air to pass through the plug; the cooling effect
arising from the instantaneous increase of pressure caused a
depression at first which was soon overtaken by a larger heating effect arising from the compression of the air in the spiral,
and though these disturbances soon ceased owing to the air
reaching the temperature of the bath by contact with the good

conducting copper pipe, yet other and longer continued
fluctuations were caused by its contact with the badly conducting walls of the plug, the temperature of which was long in
becoming steady ; the pump was therefore allowed to work for
a considerable time
The
before
any observations were recorded.
correction for conduction of heat through the plug was

determined by experiments in which the difference between the
129
temperatures of the bath and the air was made very great, and
was considered directly proportional to the difference of
it
temperature and inversely to the quantity of fluid transmitted

in a given time.
o.
of experiments shewed that
first series
increases
i& 7
with
^7-,
and we must therefore use the more complete formula

<
=T+
-"'-->
(7)
obtained from (i) by evaluating the integrals only on the supposition that
Rr and
pv
substituting
K'
for
H.
second series then shewed that for a constant temperature

of the bath the cooling
S varies directly as the difference of
the pressures
the values of
p'
-PP
^
,>
mospheres, being
Air
and decreases as the temperature
when
pressure
is
measured
in c. G.
rises,
s.
at-
130
that this equation holds
from the
truth)
and
substituting
2-8725 x(272-8
pQ v Q
for
pv
830-628
in the
= 2-8725,
i6c. (which cannot be
^TO
far
or
+ i6) =
829-578
63) we have
(see
and thus
at
good
K'
above case
i-4266/ + o-0339i2/
pvp'v'=
9-9263, so that
K*$
3-4015
also
63)
-6-7102,
3-7345;
289-69, and, as the value of
*log,-^
therefore
/ at
/=
288-8
+-
3'7345
oc. must be, within a very small fraction,
less
by 16 than
value at i6c., the absolute temperature corresponding to

oc. is 273-69. The result of all the experiments leads to the
its
conclusion that absolute zero temperature is

273-7 c., and
gives the following comparison of the absolute scale with that
of an air-thermometer in which
air
is
employed of
density d, the standard density being that of air at o
pressure of one c. G. s. atmosphere (which

per cubic centimetre).
Temperature
by absolute
scale measured from oc.
'-273-7
is
c.
relative
under a
0-0012759 grams
third series of experiments
megadyne of
is
difference
shewed
between the
131
that the cooling per
initial
and
final
pressures
inversely proportional to the square of the initial tempera-
ture, or
where a
and
for
a constant, the value of which for
is
CO
Hence
air is
about 20337
about 102571.
the characteristic equation for
permanent gases may
be determined from
an equation obtained in the same manner as (5);

substituting from (8) we have
for
on
whence on integration
since
JB
is
constant.
The
function
must be of the form
R is constant, as Boyle's law is very approximately

followed by gases, especially when / is great ; thus we obtain
where
This formula was also found by Professor Rankine by another

method 1 and shewn to give results agreeing most satisfactorily
with experiment.
The
further results of these experiments are the following

o.
values of
for different gases:
132
It will be noticed that whereas a cooling was observed before

during the expansion of hydrogen, the real effect is shewn by
these more exact experiments to be a heating, and this is
doubtless connected with the compressibility of

being less than
Boyle's law would give: and further, the cooling effects for
CO
N, air, are in descending order of magnitude, as we should

expect from their coefficients of dilatation being in the same
2
order.
The experiments
also
shewed
variably gives a smaller cooling
that a mixture of gases in-
effect
than the average
effect
of
the pure gases.
Professor
Rankine found 1
that the Centrifugal
Theory of
a generalization of the theory of molecular vortices

Elasticity
indicates that the characteristic equation of an imperfect gas
may be
represented by the formula
pv
where #
a lt
...
_i__ fa
are functions of the density
determined by experiment.
- which must be
Dr. Joule and Sir
W. Thomson
have used 2 the simpler formula

ft
where
of
air
y =
tf, a, /3,
7xi
y are constants, the values of which

be R
2-8659, a = 777'3 8 6, ft
they find to
in the case
8 445^>
214325840.
M. Cazin's experiments.
M. Cazin in 1867 emhim

which
allowed
to examine the state
an
apparatus
ployed
of a gas at any moment of its passage from one reservoir to
73.
another at approximately constant temperature.

A and B, of capacity 8923 and 33805
reservoirs
Two
c.c.
metal
respect-
were connected together by a large brass stopcock C ;

connected also to a force pump D and to separate
were
they
ively,
Trans. R. S. E. xx. (1853) 581.
Phil. Trans, cxliv. (1854) 360.

Ann. de ch. et de phys. [4] xx. (1870) 251, trans. Phil.
(1870) 81, 197, 268.
3
Mag.
[4] xl.
133
F\ to preserve their temperature

open-air manometers E and
constant they were provided with double walls between which
a constant circulation of
Air
water was maintained.
having been
pumped from
into A the stopcock was

opened, and the variation
of pressure in B during the
motion of the gas was measured by a special sensitive

open-air
which
manometer
hydrogen
G,
in
sulphate
was the indicating liquid

manometer was put in
Fig. 33.
this
connection with
the opening of C,
at
known
interval (which
and the times
at
which
was
varied) after
different observations
were made or operations performed were registered by
electric
on a chronograph. The inertia of the liquid prevented

the manometer G from indicating the exact pressure at each
moment, but a large series of experiments enabled M. Cazin
action
to
draw an approximate curve of pressures. His results are

a reservoir containing compressed gas is placed
When
communication, by a large
orifice,
in
with a reservoir containing
same gas rarefied, equality of pressure establishes itself very

When equality of
quickly in the two reservoirs (o-i second).
pressure commences, the velocities acquired have not comAt this instant the gas in the first repletely disappeared.
the
is cooler than at first, as it has done mechanical work

on the contrary, the gas in the second reservoir has been
heated, mechanical work having been done in compressing it;
servoir
but a part 6nly of the kinetic energy produced during the

efflux has been transformed into heat.
While however this
transformation continues the pressure increases, and when the
motion has ceased the heat that has disappeared from the first
reservoir is greater than that which has appeared in the second
equilibrium of temperature is quickly brought about by conduction from or through the walls, and thus, on the whole, heat
134
absorbed from the outside. If the walls were impermeable

by heat] the equilibrium would be established by exchanges between the hot and cold parts, but the final temperature would
be less than the initial.
is
In
Specific Heat of gases at constant volume.
65 we have shewn that ft, which is constant for each sub74.
stance in
of aggregation, is equal to the specific heat

volume of the substance when it is in the perfectly
all states
at constant
state, that is, a highly superheated vapour: hence a determination of this specific heat of gases is necessary. Hitherto
no direct determinations have been made, and indeed they
gaseous
scarcely appear possible, since the mass of the gas enclosed in

any vessel is so small relatively to that of the vessel that there
no appreciable
difference in the heat necessary to produce a

in the vessel, whether it is vacuous or full
thermal
effect
given
of gas. The determination can however be made indirectly by
is
finding the value of K, the ratio of the specific heat at constant

pressure to the specific heat at constant volume, and two
methods have been employed, (A) the observation of the

thermal effects that accompany the adiabatic expansion and
compression of gases, (B) the comparison of the observed
velocity of
sound with the value given by a
theoretical formula.
confined in a vessel which is incapable either of

heat
to the gas or receiving heat from it, then,
communicating
A. If
a gas
is
when compression
cific
or expansion takes place,

volume, and temperature change from />
so that
by
(62),
its
,
pressure, spe-
V Q)
63,
\ogv
log/>-log/
log/>-log/>
we can determine
quantities, we can at once
If then
known which
will
-log/+log/
the change of any two of the three

find K
as however no substance is
:
not conduct heat the experiment
is
in this

form impossible, unless the compression or expansion
135
is
effected
so rapidly that the heat lost or gained by the gas is negligible ;

and, though in this case the time is too short for us to measure
the thermal effect or the changes of pressure
and temperature, yet
we may estimate them.

Cle'ment and De'sormes 1 used a 20-litre glass
observing the final state of the gas
by
(i)
MM.
globe closed by a large stopcock A and connected with a

vertical glass tube B dipping into water
which acted as a manometer. The air with-
was
in the vessel
partially rarefied
by an
air-
could be connected with C; A

was then opened for half a second, in which
pump which
time the pressure was raised to that of the
atmosphere by the entrance of air; the incoming air had however by compressing
the air already there raised its temperature,
and as this cooled again to the surrounding temperature the
Let n be the atmospheric pressure, v the
pressure diminished.
volume of the
temperature, / its original and

the
adiabatic
p
pressure by
compression the pressure
of the air is increased from pQ to n ; also the specific volume
specific
its final
is
air, / its
diminished in the proportion of

log
since
to /, or
= log/>-log/
the pressure of the air after regaining
is
z>
- logy
its
initial
temperature: hence
log/-log/>
In one of the experiments
n= 1-0136,
whence
(2)
by
A>
allowing
it
'9953>
/=
1-0088,
1-3524.
MM. Gay Lussac
initially
and Welter 2
altered this experiment
introducing compressed air into the vessel and then

to escape adiabatically till the
pressure in the vessel
became equal
to that of the
1
atmosphere
the expansion having
Journal de Physique, Ixxxix. (1819) 428.

Ann. de ch. et de phys. [i] xix. (1821) 436.
136
cooled the
air,
its
atmosphere when
have in this case
is
pressure
it
finally greater
has regained
its initial
than that of the
temperature.
We
- log/
log pQ
In one of the experiments
= 1-0096,
K = 1-3745.
n
whence
method M.
this
By
1-0314,
Him
1-0155,
has found 1-3845, M. Dupre*
1-399, Hr. Weisbach 1-4025, M. Masson 1-419, as the value

of K for air: M. Masson found also that K
1-30 for carbon
dioxide.
The
discrepancy between these numbers is explained by the

compression or expansion does not occur adia-
fact that the
batically as the formula implies
thus part of the heat generated

to the in-rushing air and
lost to the walls of the vessel ; and during the
by the compression
another part
is
expansion heat
is
is
communicated
communicated by the
Hence the experiments wirl give

amount of air in proportion to the
the larger the vessel:
longer than
is
walls to the cooled
air.
better results the larger the
surface of the vessel, that
the stopcock too should not be
is,
opened
required to equalize the internal and external
pressures.
(3)
But here M. Cazin
has pointed out another difficulty.

pan of a spring-balance, on which
Just as a weight depresses the
placed, below the final position of equilibrium by reason of

the velocity acquired, equilibrium being only restored after a
series of oscillations, so on the opening of the stopcock more
it is
gas rushes in or out than is necessary to produce equilibrium of

pressure and a series of oscillations occurs the final pressure
:
in the
globe therefore depends
on
the phase of the oscillation
M. Cazin found that by graduthe stopcock is closed.

time
the
which
the stopcock was opened
ally increasing
during
he obtained very fluctuating values of K, tending at first to a
when
limit
from which they afterwards deviated more and more as

1
Ann. de
ch. et
de phys. [3] Ixvi. (1862) 206.
137
became less adiabatic this limiting value is thus

the nearest to the true value of K that the experiment is capable
the process
of giving.
connecting the globe with another vessel of the same size
By
and containing the same gas instead of with the outer air
M. Cazin was able to determine K for different gases: he
found the same value 1-41 for air, oxygen, nitrogen, hydrogen,
carbon monoxide, and 1-291 for carbon dioxide.
(4) As the manometers in the above experiments were
scarcely sensitive enough, Hr.
Kohlrausch 1 proposed
to re-
place them by an aneroid, which is a convenient instrument as

the deviations of its needle (counted from the position in a
vacuum) are proportional to the pressures, so that each division
of the scale corresponds to a given change of pressure. When

placed in the receiver of an air-pump the air of which was
suddenly rarefied by a single stroke of the piston, an aneroid
when the time t is measured from the beginning

of the stroke, the rate of change of pressure at any moment
due to the heat conducted or radiated from the outside was
indicated that,
proportional to the difference

at that
moment and
the
between the actual pressure
final
pressure
original temperature being regained, or
dx
=
Tt
whence
~ mx
that ensues
on
the
>
x9 e~m *.
These equations cannot be used for the period of the piston's

motion, but, as the period was small, Hr. Kohlrausch assumed
that the
mean rate
mean
equal to the
now
at
of change of pressure during the stroke was

of the rates at the beginning and end of it
the beginning the
rate
is
o,
and
at
the end
it
is
m ^,
if t x is the
length of the period, so that the variation
of pressure during the stroke is \i^mxQ e-m^\ as further the
m
pressure at the end of the stroke is xQ e~ ^ less than that
mxê~
finally
assumed by the
the process
is
(i
the total change of pressure during

e- mi i.
Then if pQ is the original
air,
+ %mt )x
1
pressure of the air before rarefaction,
its final
nn cxxxvi. (1869) 6l8

-
pressure,
and
138
m
(i-\-\mt^)xQ e- t-Lp' the lowest pressure caused by the
we
have
rarefaction,
In Hr. Kohlrausch's experiment,
= 0-01522, m = 0-1877, = o-75>

= 0-01415, p = 1-00290,
e-'
p = 0-95356, p
0-93941, whence K = 1-296.
.#
(i
ti
+ $mt )x
ii
This
and
is
not a good
result, partly
partly because the air
from the uncertainty about
was contained
/>',
in a very small re-
still the method is more perfect

ceiver, of capacity 6 litres only
than the former ones, as allowance is made for the heat com:
municated during the rarefaction.

(5) In 1870 Dr. Rontgen shewed
that the pressure-indica-
by an ordinary aneroid are not sufficiently reliable,

and, by employing instead a circular corrugated plate of German
tions given
silver (as
used in aneroids)
he obtained much better
let
into the side of a glass globe,
about 1-41 as the value
results, finding
In 1872 he made 2 a more complete series of

experiments with a more perfect apparatus, consisting of a
7o-litre glass globe A over an opening in the side of which a
of K for
air.
circular corrugated plate of
German
silver
was cemented
to
was glued a bit of

reed which was also atthis
tached to a small mirror

c
moveable about a fixed
axis, so that the
motion
of the plate could be exactly measured by the
of
reflexion
vertical
scale
distant
viewed
a
telescope.
through
This was protected from
Fig. 35-
variations of atmospheric
pressure during the half-hour occupied by each experiment

by an air-tight cover C with a plate-glass window, which
1
Pogg. Ann.
cxli.
(1870) 552.
Pogg. Ann.
cxlviii.
(1873) 580.
Comparison of Theory
was
^v^tk
139
Experiment.
further connected by a tube with another
yo-litre
globe
containing air at atmospheric pressure, so that any variations of temperature during the experiment (and these were
prevented as much as possible by paste-board covers and layers

of hay) might have equal effects on the two sides of the plate
two small mirrors clt cz were also fixed on the outside of the
:
window
to
throw comparison-reflexions of the
The globe A was

stream into
and
its
it
vertical scale.
by allowing the gas to be tested to

for several days so that all air might be removed,
filled
stopcock was shut when the pressure of the gas was
somewhat greater than that of the atmosphere; the stopcock

was then quickly opened and shut and the maximum change of
pressure (which occurs at the
moment
of shutting) observed
was observed when the initial temperature

had been regained: then the heat communicated during the
lastly,
the pressure
expansion (which was almost negligible except in the case of

hydrogen) was allowed for by Hr. Kohlrausch's method, and
K was calculated by the ordinary formula.
The means of TO
1-4053 as the value of
carbon dioxide, and 1-3852 for hydrogen:
excellently agreeing experiments gave

K for air, 1-3052 for
number is much too small, for the original temperature

was regained so quickly that the change of pressure with the
time could not be determined, and thus no correction applied,
though much more required than in the former cases. These
are the most trustworthy of the experiments yet made.
the last
B.
Newton
1
gave the formula
as the relation between the velocity of wave-propagation in a

medium, its elasticity, and density ; he considered also the tem-
perature to remain constant, so that by (59),

63, the elasticity
is equal to the pressure
thus in air at oc. and under a baro:
metric pressure of 76
c. at
dynes per square centimetre),

stances being 0-0012932
equivalent to 1013573
density under these circum-
Paris (which
its
grams per
Principia,
ii.
is
c.c.,
8.
the velocity of
sound
.140
A / --
should be
is
= 279960.
per
a result which
sec.,
0-0012932
too small by over 15 per cent., the values determined by
MM. Bravais
and 33066 (M. Le Roux' determination).
however pointed out that the rarefactions and
experiment ranging between 33237 (as found by
and Martins)
M. Laplace 1
air during wave-propagations cause an

or
a
absorption
production of heat, and he shewed that these
constant changes of temperature in the parts of the air have the
condensations of the
same
as
effect
if
its
were increased while
elasticity
perature remained unaltered, the
coefficient
its
tem-
of increase being
the ratio of the elasticity at constant entropy to the elasticity at

constant volume, or K, by (12),
17 thus
:
determined by experiment, K may be calculated.

Rankine is due the following elementary
method of demonstrating this relation 2 a method applicable
whence,
to
if
is
To
Professor
all
waves of
permanent
finite
type, the
longitudinal disturbance which are of
word
denoting the relation between
type
the extent of disturbance at a given instant of the particles and

their respective undisturbed positions.
Let v be the linear
volume of the substance

state, m the mass
velocity of the wave, v the specific
which transmits the wave in

that
is
disturbed
in
(called the mass-velocity)
time, the
is
unit-time
:
now two imaginary
in the direction of the wave,
tube
wave
in a tube
of
unit-section
traverses v in unit-
of unit-section that
or
is v,
Consider
in
then, as the
volume of substance
disturbed in unit-time
undisturbed
its
mv.
planes
u^
(13)
moving with the
velocity v
being the velocities of longi-
tudinal disturbance of the particles that are at any moment

in the planes
these particles pass through them with the rela:
tive velocities
Mecanique
u^v,
celeste.
u2
v,
and,
if
the specific volumes of the

Phil. Trans, clx. (1870) 277.

substance in
its
141
disturbed state at these planes are v 1} v2
the
mass of substance which passes in unit- time through unit-section

of each is as
but, as the wave is
;
permanent,
much must
leave the space between the planes as enters
Z>
"l
so
must equal -
since these fractions
in the undisturbed state being zero.

,
and
it,
vu = vu = m.
that
Thus
= -**)
= vmv^
3
either of these equations
Now the mass

manency.
the planes with velocity
the velocity of the particles
the kinematical condition of perhaving entered the space between

is
m
^ and
an equal mass
having left
with the velocity u2 the gain of momentum m(u^
u^) can
only be due to the difference p1 pz of the external pressures
it
at the planes, as the
other cannot alter the
mutual actions of the particles on each

of the whole thus
momentum
A -A = (i-a),
= m*(Vt- Vl
= - m\
(15)
by (14); or
This
is
the dynamical condition of permanency:
written in the
is
may
l6 )
also
be
form
A+
where
it
X=A + X=A
the pressure in the undisturbed state.
(17)
If e repre-
sents the elasticity of the substance,

e
dp =
m
=v-fdv
or
v*
v*
5
=-
(18)
which
is
Newton's
formula.
Now in
sound-waves the com-
pressions and rarefactions are very sudden and succeed each

other exceedingly rapidly, so that the heat which is
developed
cannot pass from particle to particle by conduction ; hence in

calculating e we must suppose that no heat is lost or take e
142
as the elasticity at constant entropy,

elasticity at
have
and as
this is K times the
constant temperature, or, in a perfect gas,
finally
v"
Kpv
Experiment shews that there
is
K/>,
we
p
-
K.
(19)
no permanency
waves
in
if
the tem-
perature of the medium is always the same: hence Professor Rank in e

points out that change of temperature and conduction of heat are necessary
permanence; but, as the transfers of heat must be from particle to

and not from other bodies, the thermodynamic condition for per-
for
particle
manency
is
that the total heat received
by the medium
is
zero,
or, rf/x
representing an element of mass,
fft d(j> dfj. =

wave
If a certain type of
o.
(20)
capable of permanence, a relation must exist
is
between the temperatures of the particles and their

conduction
may
effect the
relative positions so that
proper transfers of heat.
If
is
the rate of
variation of temperature with the distance, and c the conductivity, the

quantity of heat received by unit^mass in the time during which the dis-
turbance traverses
t d<j>,
that
it
is,
seconds
is
rf.c
dx
but this
is
also
and thus
*,
,^_JLrf.
m t dx
dt
whence,
and
These formulae are general
a + mftd(f>,
-.
b+f
a + mftd(j>
J
dx
(21)
they cannot in general be evaluated except for
perfect gases.
(1)
The
sound has been determined by
velocity of
direct
oc. MM. Bravais and Martins found

experiment.
33237, Hrn. Moll, van Beek, and Kuytenbrouwer 33226,
the French Academy 33200 in 1738 and 33120 in 1822, M.
Regnault 33070, M. Le Roux 33066.
In
(2)
air at
This velocity has also been determined
indirectly.
If
A.
the wave-length of a particular note and n its frequency of

n\ thus a note may be produced by an
vibration, then v
organ-pipe, and n measured by a siren, A being
is
=
2
(i
a)
or
4(1
2n
+ 1)^
i

according as the pipe
open or stopped,
is
its
being
143
length,
i the correction for the mouth-piece, g the correction for the
open end, n the number of the note
in the series
of tones which
the pipe can sound.
For the
velocity of
sound
method found 33300, Hr.
this
33206,
oc. M. Dulong has by

Seebeck 33277, M. Schneebeli
in air at
Hr.Wertheim
(3) Hr.
33133.
has lately experimented on the velocity of
Kayser
sound by means of Professor Kundt's

very exactly the wave-length
of vibration-frequency
of v at oc. for air.
We may hence
of v for
air,
and
finds
which give
to a given tone
33250 as the value
take 33200 as not very far from the true value
in this case
/=
VK
As a final
number and
and he thus
n,
dust-figures,
A corresponding
result
33200
- or
27996
we
then
that determined
the value of K for
1-4063.
shall take the
by Dr.
Rontgen,
and thus k
air,
mean between
this
or 1-4058, as
= 7-0610. We
1-4058
have throughout the calculations considered air a perfect gas,
but the error we have introduced thereby cannot be very great ;
and
if
(52),
we proceed to find the value of J on this supposition by

63, we obtain a result almost identical with Dr. Joule's;
thus
K
K
1-4058
x 2-8725 =
0-4058
0-2389
very small error in the value of K would cause a large one

in 3) since
6/c
dK
-F^c^'l^rr" 738
on
substituting 41-55 for
yand
1-4058 for
"'
K.
APPLICATION TO SOLIDS AND LIQUIDS.

75. Our experimental data are much fewer and much more
uncertain for solid and liquid bodies than for gases yet such as
they are they shew a very fair agreement with the theory.
:
144
The best-known
coefficient
is
quantity for such substances is the real

of expansion under constant pressure, when the pressure
one atmosphere
denoted by
is
it
with the pressure,
little
we may
and, as
a,
consider
it
it changes very
a function of the
temperature and write
= av.
(22)
<)/
Other experiments give the volume-elasticity

perature, or
its
at constant
tem-
reciprocal the true coefficient of compressibility at

is denoted by /3, so that
constant temperature which
9
(23)
^=-Pv.
The
variation of (3 with the pressure
is
not
known
be negligible its variation with the temperature

that it must be considered a function of /.
:
These two
but seems to
not
slight,
so
results give us the variation of the pressure with
the temperature
when
the volume remains
^p
The
is
equations of
unchanged
for
_^_ a
16, 53,
54 therefore become
c)/>
a2
<)z>
(25)
*
B
= k,<)/ = K-avt,
^v
av
M. Rossetti discussed 1
experiments on the density
Application to water.
76.
1868 the
results
water between o
of the best
he concluded that the
these temperatures
Ann. de
cJi.
et
in
of
and iooc., and compared them with an
elaborate series of observations
meter
= K d/ = K
is
made by himself with
specific
best represented
de phys. [4]
x.
a dilato-
volume of water between
by the formula
(1867) 461,
xvii.
(1869) 370.
145
where
^ = 2-24329x10-
^=3-78702x10
6 representing /
273-7 or centigrade temperature; hence
The
and
indices 1-6
must be considered odd and even
2-6
respectively.
For temperatures between 100 and i82c. with a pressure of

1
about 15 atmospheres M.
gives the formula
Him
where
a2
<r
4
1-08679 x io~
9
2-87304 x io~
<5
dz
=
=
6
3-00737 x io~
12
6-64570 x io~
and
His apparatus consisted of a copper vase

a vertical pipe B with a cast-iron vessel
was led
C, from which an iron tube
communicating by
ioj metres
vertically
upwards,
its
upper
part terminating in a wide glass tube E,
graduated in c.c., and with a short branchpipe F\ water filled the vase A and the
upper part of B, the rest of the apparatus

being filled with mercury up to E. A
was heated by two gas-burners enclosed
with
double cover
in the
it
G,
whose
inner walls were of tinned iron and the

outer of
wood
the water was continu-
ously agitated by the stirrer H, the axis

of which passed through B and C so as
be worked
frftm outside, and the exwas

measured
pansion
by the mercury
that overflowed into E or was drawn off
to
at F.
The
best experiments
on the com-
pressibility of water are those of M.
Grassi
1
from
Ann. de
his first series
ch. et de phys. [4] x.
Fig. 36.
Professor
(1867) 32.
C.
R.
xxvii. (1848) 153.
146
Rankine shewed 1
that the compressibility
is
inversely propor-
and absolute temperature jointly, but the

second and more complete series 2 gives no countenance to this
law.
If the results of this latter series are plotted on a diagram,
are
found to disagree very considerably we may however
they
tional to the density
deduce from them the formula
= a -& e + c
3
2
3
where
as
5
5-0357 x io~
33
7
2-8922 x io-
cs
9
3-0743 x io~
The range of
pressure being measured in c. G. s. atmospheres.
the experiments was between o and 53c.
We get also from M. Bosscha's calculation of M. Regnault's experiments
K= 41-55(1+0-000220)
= +
<*t
where
t>tO,
41 '55,
34
x io~3
9-141
In these experiments a mass m z of water at 3 was forced by its

steam pressure from the boiler to the calorimeter, raising the
mass
of water contained therein from
to
but the kinetic
energy thus converted into heat was not taken into account.
If F is the pressure of saturated steam at 3 n the atmospheric
,
pressure, v 3 the specific

specific heat of
heat between
and
it is
but
this
water between
and
0^
and
at
JT1)2 the
A"2>3 the
mean
mean
specific
then
M. Regnault neglected. Professor

the
applied
necessary correction and proposed the
the last term that
Rankine
formula 3
volume of water
K _ ^ {i +0-000001 (0- 4
2
)
does not represent so well the specific heats at low
temperatures, while at high temperatures it differs very slightly

from Hr. Bosscha's, which we therefore prefer.
From these data we have made the following tables, which
will give
1
an idea of the magnitudes of the quantities
Proc. R. S. E. in. (1851) 58, or Phil. Mag. [4]

Ann. de ch. et de phys. [3] xxxi. (1851) 477.
Trans. R. S. E. xx. (1851) 441.
i.
in question.
(1851) 548.
147
148
We
hence deduce
periments
for the temperatures of
M. Grassi's
ex-
149
K= 41-55x0-0559 = 2-3226, and thus k 2-2692

= 1-0235. By an entirely different method Hr. Edlund
found K = 1-0203.
but
and
l
78. Adiabatic
When
compression.
pressed without losing or gaining any

the relation
Kdt+Ldp
(**\
v
or
d/>
w when
no heat
heat,
are
there
com-
subsists
_..
L
avt'
hence the small increase of temperature

of pressure
bodies
has
is
3-
due to an increase
gained or lost
is
very approxi-
mately given by
*=!if
Sir
is
the density of the substance.
W. Thomson, who
< 26)
This formula
is
due
to
also earlier gave the formula
approximately, for the small increase of temperature 3- that

occurs on a slight adiabatic expansion v, as may be found by
a method similar to that above.
Dr. Joule has verified* the former formula by compressing
water in a copper vessel connected with a cylinder on the

piston of which weights could be placed, a copper-iron thermo-
(which was found to be uninfluenced in action

to which it was subjected) being placed in the
centre of the vessel to measure the thermal effect
each exelectric junction
by the pressures
periment lasted 40" (the time required by the galvanometerneedle to assume a new position), and it was found that in this
short time the cooling due to the expansion of the vessel did
not affect the thermopile. The experimental results were
:
2
3
*
gg- Ann cxxvi. (1865) 566.

Proc. R. S. viii. (1857) 568, or Phil. Mag. [4] xv. (1858) 541.
-
Phil. Trans, cxlix. (1859) 133.
150
151
If T be the stress acting as a tension, x the

f
its specific heat under conof
unit-mass
of the rod,
length
f
stant tension, L the latent heat for an increase of tension at
to the above.
constant temperature, then
dETdx = dH= K'dl+L'dT,

the negative sign being used in the first expression as during
hence we
elongation work is done by the tension on the body
:
find exactly as in
52, 53,
z ' =/
^7
= tax,
if
is
stress.
(28)
[
J
the real coefficient of linear expansion under longitudinal
We may
also
the rod to perform a
shew this by considering unit-mass of

Carnot's cycle ABCDA by undergoing
(i) a diminution of tension without loss
or gain of heat till the temperature falls

by 6/, (2) an increase of tension T, the
temperature being kept constant by the

addition of heat L'bT, (3) a further increase
of tension without loss or gain of heat
till
(4)
the initial temperature is attained,

a diminution of the tension to its initial
value, the temperature
being maintained
constant by the withdrawal of heat
then
37-
by Carnot's
principle,
area
ABCDA
__" 8/
'
Now the area ABCDA may be considered a parallelogram and

BCEFB or CExGC', but CE is the
therefore equal to the area
elongation due to the change of temperature 8/ under constant

tension or ^
8/,
and GC
is
the change of tension 5
L'bT
whence equation
(28).
7*;
hence
152
If,
therefore, the rod
losing heat,
we have
is
,
d/
suddenly stretched without gaining or

if
axt
or the small change of temperature

.
tension <C adiabatically is
a'xt
$=
5-
due to an increase of
'
(29)
~jp-E.
The
negative sign indicates a cooling on extension in those

bodies which expand when heated, and similarly a heating on
extension in those bodies which contract as their temperature

No experiments have been made to determine f the
rises.
specific heat under a constant longitudinal tension, but in calculations it is assumed equal to K, the specific heat under a
constant hydrostatical pressure, from which it cannot very much

similarly, as the dependence of a', the coefficient of
differ:
linear expansion
under longitudinal
stress,
upon
the stress has
not been experimentally determined, it is generally assumed

equal to the real coefficient of linear expansion under ordinary
atmospheric pressure.
Dr. Joule has verified
this
formula also by experimenting
on
cylindrical bars about 30-48 c. long and 0-635 c. in diameter,

the upper ends of which were screwed into a fixed piece of
metal and the lower ends attached to a lever on whose ex-
tremity weights could be hung without the operator approaching

the apparatus.
The change of temperature was measured
a
thermo-electric
by
junction of fine copper and iron wires
inserted into a hole, 0-063
centre of the bar.
He
c.
in diameter, bored through the
obtained the following
153
Similarly satisfactory results were obtained with gutta-percha

different kinds of wood, and also when the rods were com-
and
pressed instead of being elongated.

The behaviour of india-rubber, the peculiarities of which were
1
first pointed out
by Mr. Gough was also carefully examined.
,
When
suddenly stretched
cooling
effect,
and
is
W. Thomson
it
undergoes a heating instead of a
immediately cooled
if
allowed to contract
negative, or
that a stretched piece of india-rubber will contract when its
temperature is raised a fact demonstrated by Dr. Joule, who
Sir
therefore concluded that a'
is
found that when
initially stretched to double its length it conof

that length on rising through 50 degrees.
by TVth
The elasticity of india-rubber is very imperfect and at low
it then acts like a metal, being
temperatures it becomes rigid
tracts
cooled on extension and
warmed on compression.
Better ex-
periments can be made with vulcanized india-rubber, the elasticity

of which is more permanent, and the general facts observed
are
i) the effects
of adding or removing tension are sensibly
equal but opposite, (2) when the tension is small but continuously increased, there is at first a cooling effect which
reaches a
maximum and
that increases
Hr.
more
ultimately
becomes a heating
effect
rapidly than the tension.
Schmulewitsch 2 has
also
shewn
that the
effect
of
heating a stretched piece of india-rubber is to lengthen it if the

tension is small, and to shorten it if the tension is large thus
for a certain tension there will be neither elongation nor con:
traction.
He
determined
this
tension and the corresponding
elongation in the following cases
J54
This example shews that the values of a should in all cases

be determined by special experiment, for the vulcanized indiarubber used by Dr. Joule dilated when heated in an unstretched
its coefficient of cubical
expansion 5-26 x io~ being
than
that of any other solid.
greater
Hr. Edlund 1 has made similar
experiments on wires: these
4
condition,
were fixed by their

upper end A, the
lower end being so
fastened to a moveable lever at
as to
be instantly loosened on a pin being

drawn.
On the lever,
the weight of which
was
by
C,
counterpoised
was a roller
that supported a
1
weight.fi
which could
thus easily be
moved
backwards and
for-
wards, and the elon-
gation of the wire
caused thereby was

estimated from the
Fi S- 33.
angular
displace-
ment of the
lever de-
termined by means
of the mirror F.
The temperature of the wire was determined
by a small thermo-electric element G consisting of a crystal
of antimony and another of bismuth separated by the wire
An experiment
against which they were pressed by springs.
included three processes: (i) the weight E was rapidly pulled
to the far
end of the
lever
always a cooling, observed

temperature (known by
1
gg'
Ann
and the thermal

;
(2)
on
effect
^, which was
the wire regaining
its initial
the return of the galvanometer needle

cx i v (1861)
i,
cxxvi. (1865) 539.
155
was pulled back to its first

a warming, determined;
was
(3) when the weight was at the far end of the lever, the pin
suddenly withdrawn and the wire loosened its contraction was
to
its initial
position,
position) the weight
and the thermal
effect
$2
unaccompanied by external work and caused the thermal effect

$3
With a steel pianoforte wire, 59 c. long and o-ii4C. in
.
diameter, the following results were obtained:
56
the corresponding pressure

is
spherical,
so that
-\
r\
7\
is
rt
=T
T2
when
or
the surface
and rl
=r =
z
r.
In the
case of a thin spherical soap-bubble, the values of T and r may

be supposed the same on each side of the film, and hence the
4.
inside pressure will be -
T
-
+ n, when n
is
the atmospheric
Then the work done per unit of augmentation of

pressure.
the area of one side of the film will be 2 T.
Now for a given liquid

the temperature rises, as
T varies, and
in general decreases, as
shewn by the
is
different heights to
which the liquid
rises in a capillary
tures, the heights
being in general the less the higher the tem-
Thus
peratures.
/,
/',
the
T' being the tensions at the temperatures

(TT') will be gained if a bubble on the end
allowed to collapse so as to lose unit-area at /, and
work
of a tube
is
tube at different tempera-
T,
then, after being heated to
/',
is
blown up again
to its original
of temperature is small and

at the higher
to
an
will
be
there
bt,
equal
absorption of heat
temperature and at the lower an evolution of heat which is
dimensions;
or, if the difference
infinitesimally less
Carnot's
by
2 -T-^/,
the
work done, whence by
principle
7'
Hence if a film such as a soap-bubble be enlarged, its area

being augmented in the ratio of i to m, it experiences a cooling
effect to an amount calculable by finding the lowering of tem1
perature produced by removing a quantity of heat equal to
mt-
from an equal mass of liquid unchanged in form/
For water
T=
81 dynes per centimetre at 20
centimetre in drawing out a film
162 ergs
c.,
or the
also
work per square
dT =
--T
about.

Hence
at 20
where
c.,
duce the cooling

ergs.
Thus
if
293.7, the heat that
work
gram
spent in drawing
energy of the water

its
or
is
its
sq. c.
energy
is
it
valent to a heating effect of
centimetres at
must be removed to pro-
per square centimetre of surface-extension is 86-5

of water be drawn out to a film of 100 square centi-
effect
out
is
= -= degree c., and

,
6
4803
41.55 X io
=
100 x 162
16200 ergs, and is equi-
metres, the cooling effect will be 100 X
the
157
c
6
41.55 X io
as much increased
increased
by
?- degree
2565
when
original temperature as if
it
it is
c.
hence the total
drawn out
to 100 square
were raised through
megalerg when
it
is
drawn out
to
4036
CHAPTER
Change of
81.
we
VIII.
State.
Apparent discontinuity. Under
find that heat given to a
increase
body
certain circumstances
at constant pressure does not
temperature as in the general case, but produces
its
molecular changes, altering
its
state
from the
solid to the liquid
form, from the liquid to the gaseous, or from the
solid to the
gaseous these are the processes of fusion, vaporization, and

When a higher form exists simultaneously and
sublimation.
:
in contact with
a lower form of the same body,
it is
said to be
saturated ; and for this case the general thermodynamic equations require modification.
(the letter generally used to express vapour

the
pressure of a body in a saturated condition,
pressures)
s
the
specific volumes (i.e. the volumes of unit-mass) of
by /,
Denoting by
the
body
and lower form, these quantities being
in the higher
functions of
higher form
and by
t,
in
the proportion of the substance in the

12
unit-mass of the mixture, we have by
= (s's)m + s
= mu +
(i)
s,
if
we
u for
write
The form
Rankine
s,
so that u
a function of
/:
also
of the function F(i) has been found by Professor

on a certain hypothesis as to the molecular con-
stitution of bodies to
be
is
--...,
(2)
denoting absolute temperature, and a,b,c... being constant

1
Edin.
New Phil
Journ.
xlvii.
(1849) 28
Change of
State.
159
The
'
hypothesis referred to is that the

in
a
of
presence of its liquid or solid is the
vapour
pressure
result of an expansive elasticity in the vapour balanced by an
for a given substance.
attractive force
tending to condense
on
it
the liquid or solid
surface/
The
heat absorbed by unit-mass in changing state is called

: it is a function of the temperature, and will be
denoted by A megalergs. Now the total volume is increased
the latent heat
during this change of state by s's or u under the constant

pressure f, so that external work fu is done in addition to
the
work against molecular forces
concerned in the molecular change
y=
82. General equation.
hence the heat A'
is
really
given by
A -fu.
Let
s',
(3)
z
be the
specific heats of
a body in the higher and lower form respectively when saturated, that is, the heat requisite to raise the temperature by
one degree when the pressure varies so that equation (2)

always
satisfied
these quantities are functions of
temperature and pressure alter simultaneously
/,
is
and as the
we have
''
L having the ordinary meanings and referring to the lower

f
L f referring to the higher.
form of the body,
K,
When
a saturated mixture takes in heat in the general case,

the temperature of the whole is raised and a portion of the
lower form
heat
dH will
is
transformed into the higher;
raise the
temperature by
dt, the
thus addition of
lower form taking
'
and the higher mx dt, and cause dm

and render latent \dm\ hence
(\m)%dt
dff
to
change
{(*'-*)m+*}dt+Xdm.
Law. Since
(5)
dH
but this
is
dE +fdv
a perfect differential
/
hence
^
_ s)OT + s _ /^_J = _
_
(X /
x
/x
state
dz\
160
Change of State.
dX
+ -* = df
,
^v
ii^
Tt
= TVU,
by
(i),
on
writing, as
is
(6)
for
very convenient,
This
was first given by Professor Clausius

but by a
somewhat cumbrous method; the above exact demonstration
1
relation
is
due to
Sir
W. Thomson 2
Second Law.
Since
is
perfect differential,
"
'
dX
Hence by
(6),
Equation
(8)
demonstrating
application
was
(6).
= tsut.
ofCarnot's
ubm;
increased by
cycle being
bt,
Xbm
thus, the
is
change of pressure
in the small
we have
whence the above
relation.
W. Thomson 3 who
,
when bm changes state

absorbed and the volume
principle: for
at constant temperature, heat

is
(9)
given by Hr. Clausius in the paper

Equation (9) may be obtained by the direct
first
used a
It
was thus
first
different notation
found by Sir
and denoted
Carnot's Function by the general symbol p instead of - as

later
assumed: Professor
virtually
gave
it
4
James Thomson
tending the fusion of ice as detailed in

1
2
3
*
at the
same time
from consideration of the circumstances

89.
Pogg. Ann. Ixxix. (1850) 389, trans. Phil. Mag. [4] ii. (1851) 16.
Trans. R. S. E. xvi. (1849) 55 6 equation (6).
Trans. R. S. E. xvi. (1849) 575.
at-
Change of
161
State.
Relation (9) shews that

/ is greater or less than s
is positive or negative according

thus according as a body expands
or contracts in changing state from a lower to a higher form,
so does an increase of pressure on the body in a saturated state
as
raise or lower
its
This
temperature.
result
might have been
expected: if fusion, for instance, is accompanied by increase

of volume, that is, by further separation of the molecules, compression must reconvert
some of
the liquid into the solid form,
and by rendering sensible the corresponding latent heat must

increase the temperature of the whole ; and vice versa, if fusion
diminishes the bulk of a body, compression will liquefy part
of the solid and the necessary absorption of heat will diminish
the temperature of the whole.
This relation also gives the

v
= *+
specific
volume
a very convenient form.
whence
*-fc
(5)
(10)
may be
The entropy
^_
^
^0
From
(7)
and
(10),
we have
written
is
therefore given
['
by
dl.
( I3 )
*^0
86. Determination of the Intrinsic

Energy.
(6), (i),
form
85. Determination of the Entropy.
hence the general equation
in the
m\
From
(5),
62
Change of
Hence by
State.
integration,
(14)
f^dt,
-/J)
These equations and that of the preceding article may
deduced from the general relations (49),
61.
The
87. Latent heat.
We may
by
is
(3)
integration of
also express the real internal
and
Thus
(9).
A'
is
zero
when
(15)
easily
be
determines
(n)
work
<ff.
A' as
-\oge
/= -
88. Expansion along an Adiabatic. In this case there

no change of entropy, and equation (13) therefore gives
*_. +
^0
ri
<0=0>
(I8)
A)
We
m can be determined for any period of the process.

have here assumed not only that there is neither gain nor
loss
of heat, but also that
whence
no
kinetic energy
is
acquired or
communicated this supposition requires the external pressure

to vary with and be always equal to that of the mixture.
:
The work done

by
during a finite change of volume is shewn

32 to be the right-hand expression of (14) or (15) with
every sign changed.
From
(10) and (18),
and according as
(which
is
found by differentiating
this
C/uuige of State.
163
expression with respect to /) is positive or negative will comIt is necessarily

pression lower or raise the temperature.
negative
On
when both
& and
differentiating (18)
- increase with
we
/.
obtain
X,
which we might get at once from
as the
(5) ; hence according
expressions within the brackets are positive or negative will an
increase of temperature diminish or increase the value of m,
that is, cause condensation or
vaporization.
Finally, since / is a fraction of
by
(i 8),
we have
d
and according
as this
is
A.
positive or negative will vaporization or
condensation accompany expansion.

This case was first treated by Professor Clausius
1
.
APPLICATION TO THE TEMPERATURES OF FUSION.

89. Fusion of ice.
As
ice contracts in the act
the melting-point will be lowered
The
of melting
by an increase of pressure.
reasoning which led Professor

conclusion may be thus exhibited.
2
James Thomson
to this
o c.
compressed
be confined in a cylinder, all of non-conducting material except
the bottom which is perfectly conducting and of no capacity
for heat,
and
water at o
air at
If
this cylinder be immersed in a small mass of

the air will do work when allowed to expand at
c.,
constant temperature and will at the same time absorb heat

from the water, part of which will consequently be frozen. If
1
2
gg- Ann xcvii. (1856) 458, trans. Phil.

Trans. R. S. E. xvi. (1849) 575.
-
Mag.
[4] xii. (1856) 253.
164
Change of
the water
enclosed
is
formation of
State.
another piston-fitted cylinder, the
in
ice, tending to increase the volume, will increase
the pressure and

till the pressure
work can be obtained

falls
to
its
in allowing expansion
If
original value.
now work be
compressing the air to its original volume, the heat

absorbed in the first operation will be restored to the water and
spent in
reconvert into water the ice that was previously formed, work
at the same time being done by the external pressure on the
water during the accompanying contraction.
now
in
its
is
original state,
hence the excess of the work done by the water

expanding over that done on it during contraction must
expended
in
Everything
and on the whole no heat has been
be equal to the excess of the work done on over
The
excess in the
that
done
case necessarily exists, as

by
the pressure is least during the contraction ; the excess in the
second case can only exist if the air has a lower temperature
the
air.
first
during expansion than during compression it must therefore

at a temperature below oc. when in contact with water
:
be
freezing under
greater than atmospheric

that of the water.
pressure
and
this
temperature must be
calculation in this case of equation (9) is simple.

The
at
of
water
oc.
relative
to
that
at
is
4c.
density
0-999871, as
The
deduced by M. Rosetti 1 from
all
the best observations;
and
absolute density in grams per cubic centimetre

i
is 0-999884, as that of water at 4c. is 1-000013 by
2
1-000116.
Professor
Bunsen's
hence
By
experiments
therefore
its
/=
with his ice -calorimeter the relative density of ice at

0-91674 or its absolute density is 0-916752, whence
s
1-090808
and
u=
The same
oc.
is
0-09069.
oc.
experiments give 80-025
which are equal to the hundredth part of the heat
required to raise one gram of water from o to iooc., that is,
to i -oi i calories; hence
f r tne latent heat at
in units
A
1
80-025 x i-oi
x 41-55
3361-6 megalergs.
Ann. de ch. et de phys. [4] xvii. (1869) 375.

p gg- Ann. cxli. (1870) 7, trans. Phil. Mag. [4]
xli.
(1871) 165.
State.
Change of
When
is
heat (that
expressed in
is,
is
work)
megadynes
expressed in megalergs, pressure

denotes pressure in mega-
thus
dynes per square centimetre or in
oc.
is
c. G. s.
in absolute temperature 273-7
*f_ -~~
dt
A
~
tu
-
165
33 61 6
atmospheres.
Also
hence
__
0-00738
273-7x0-09069
or the temperature of fusion is lowered by 0-00738

addition of one atmosphere to the pressure.
by every
W. Thomson 1
confirmed this calculation by

a piezometer of stout glass in which were placed an airmanometer formed of a straight glass tube and a delicate
Professor Sir
thermometer of sulphuric ether enclosed

tube
for
phuric
protection against compression.

ether was chosen from having a
in
a sealed glass
Sul-
small
and a large expansion-coefficient

0-001513, so as not to deform the thermometer
by its weight and yet give large readings for small
differences of temperature.
The thermometer
density
0-736
scale contained 15! divisions to the centimetre,

each equivalent to 0-0078 degree. The piezometer was filled with ice and water with a ring
of lead to keep clear from ice that part of the

thermometer where readings were expected. In
series of experiments the pressure when increased by 8-1 and 16-8 atmospheres, each of
which corresponded to 29-922 inches of mercury
two
at
Glasgow or
to
1-0142
lowering in temperature of
c.
G.
s.
atmospheres, produced a
7j and i6J
divisions respectively,
of 0-0586 and 0-1289 degrees, or a lowering of

0-00713
and 0-00756 degrees (with a mean of 0-00735 or ri~5 degree)
per additional c. G. s. atmosphere.
From the above then it is obvious that the highest tempera-
that
is,
ture at
for
which
by
1
Proc.
ice
ii
its
R.
S. E.
can
0-00738 x 1-0136
o-oo748c.;
melting-point is oc. under the pressure of
ii.
exist is
(1850) 267, or Phil. Mag. [3] xxxvii. (1850) 123.
1 66
Change of State.
1-0-136 atmospheres, and

a vacuum.
possible temperature
is
highest in
W. Thomson
Sir
ment under
with
its
suggested the repetition of this experigreat pressures as obtained by a Bramah press
minimum thermometer
melting-point.
Professor
to
Mousson
mark
1
the
corresponding
carried out this idea in
a different way, directly shewing that by compression water may

be maintained fluid at low temperatures. He first used two
similar prisms of iron which when screwed together
their
surfaces of contact being carefully worked plane and covered
with caoutchouc to prevent leakage had down their centre a
cavity whose lower end was firmly closed by a
At the top was another screw, the turning of which
pressed down a piston-rod and compressed water with which
the cavity was filled.
The state of the water was examined
two
windows
of
thick glass let into the lower part,
through
cylindrical
screw.
the test of fluidity being the mobility of a copper index in

the bore.
After compression the whole was placed in the cold
and the temperature of freezing determined

this
to be between
and
the
3
5c.,
corresponding
above theoretical number,
pressure, calculated from the
For higher presbeing about 540 atmospheres.
winter
air
was found
sures he used a prism of steel, as in the figure, with a cylindrical

bore of 0-71 c. diameter; the lower part of the bore was closed
by a copper cone pressed in by a strong screw;

the upper part was slightly conical and was closed'
by a long plug a of copper pressed down by a steel
piston by means of a strong screw and long lever.
yielding to pressure made both ends
while
the steel allowed any pressures to
water-tight,
The upper part was closed, and the
be used.
The copper by
whole reversed;
dropped
FiT.o
in,
a small brass
and the bore
filled
rod b was then
with water.
When
9-5 c. the water

exposed to the cold winter air at
the cylindrical
from
the
ice
protruding
congealed;
bore was then removed, the copper cone put in and screwed
Change of
State.
67
and the whole again reversed and put into a freezing

1 8 c.
mixture at
The brass rod was obviously then uppermost, but by sufficient compression the ice was liquefied and
the rod fell, as was shewn on removal of the lower screw and
up,
cone
for the
rod issued
first,
followed by a cylinder of ice
Mo
Hr.
us son roughly
estimated the compression at 0-134 and the pressure employed
as not less than 13250 atmospheres:
now the empirical
formulae for the compression and expansion of water and ice
frozen
on
the reduction of the pressure.
1185 atmospheres only as the pressure corresponding

i c.c. of ice at
9-5 c. is 0-99851 c.c.
at oc., and this becomes 0-91548 c.c. of water at oc.
indicate
to this compression; for
of ice
under one atmosphere, or 0-8619 c.c. under 1185 atmospheres,

or 0-866 c.c. of water at
18 c. under the same pressure and
:
again, the pressure required according to the theory is 2410

atmospheres only hence though we are not perhaps justified in
:
applying the above empirical formulae and theoretical number

so far below oc., we must regard the pressure in the above
experiment as indefinite. It may of course have been much
greater than
was necessary
to liquefy the ice.
Older experiments made by Major Williams
seem
shew
to
that the possibility of expansion
is
at
Quebec
a necessary
condition of congelation.
strong i3-in. iron shell, the
thickness in different parts varying from 5 J to 4^ c., was filled
with water and closed by an iron plug, of about 1-2 kilograms
by a sledge-hammer, the
roughened by a file to secure greater
was then exposed to the cold. The most striking
in mass, driven into the fuze-hole

sides of the plug being
firmness
it
result was obtained with an out-door temperature of

28Jc.
the plug was suddenly shot out at an elevation of 45 and
thrown to a distance of 126^ metres (indicating an initial
:
velocity of at least
35^ metres per second or 7450 megalergs

ice, 2 2 c. long, was forced out of
of work), while a cylinder of
the hole, the expansion being about TVth of the original volume.
On
another occasion with an out-door temperature of
the plug being fastened internally

1
Trans. R. S. E.
ii.
by springs, the
(1786) 23.
24^
c.,
shell burst
i68
Change of State.
and the
came out of
ice
the cracks in thin plates like
fins, in
The temperatures given are not necessarily

places 5 c. high.
those of the water, and we cannot therefore estimate the internal pressures.
Clausius 1
first used equation

(8) to determine the
rate of increase with the temperature of the latent heat of ice
Professor
normal melting-point.
at its
now
hence, though
K=
K = 0-5037 X 41*55 =
We
s.
'
have
equations (4)
and so
L, L' are both very small,
ence
*'
By
is
a fortiori their
differ-
K K
r
about 135, we may write

for
from
and
M.
measures
Person's
41-55,
is
'
r </A_
we have
20-929; hence
J^K= 20-621, and
3361-6
or less than one per cent, of the value of A at this temperature.

This result shews that A diminishes with / when the pressure
varies so as to preserve saturation.
M. Person 2 had concluded,
Before
this
demonstration
since the specific heat of ice
is
than that of water, that the latent heat diminishes when

the freezing-point is lowered without increase of pressure by
less
the fluid being kept perfectly

calculate
If
this
still:
we cool a gram of water
effect
we may
thus
at constant pressure
'
/27S.7
.
dt] if
we could
freeze the water at this temperature
by absorbing the latent heat

becomes sensible a process really impracticable,
the failure of which however does not invalidate the reasoning
the heat given out is A if the ice so formed is cooled to r, the
as fast as
it
heat given out
is
Kdt', thus in these three processes heat

rt
T273-7
iequal
to
change of
quantity,
K' dt + \+
intrinsic
is
Kdt
is
given out;
but the total
energy and external work, represented by
this
entirely independent of the intermediate temperature
p gg- Ann
R. xxx. (1850) 526.

Clausius' Abhandlungen uber
Ixxxi. (1850) 168, trans. Phil.
Mag.
[4]
ii.
(1851) 548.
C.
die mechanische
Warmetheorie (1864),
i.
98.
Change of
/,
as the pressure
to /
constant
is
we have
d\
dt
State.
hence
169
differentiating with respect
= K'K.
similar method, wherein the pressure is supposed to vary

so as to preserve saturation, results in equation (6), thus justiIn fact, in raising a gram of ice from r
fying the procedure.
always at saturation-point, in liquefying it at /, and in

raising the water so formed to 273-7 always at saturation-point,
to /
/*273-7
/t
external
work
sdy+A +
fds-\-f(ss)-\-\ fds'
initial
and
is
done,
if
are
o-,
Js'
J<r
the
during these processes
s dt\
final
volumes and
s,
correspond to the tem-
now
the total change of intrinsic energy is independent of the intermediate temperature /; hence differentiating
with respect to / the difference between the above quantities
perature
(which
/:
is this
change of
<t\
intrinsic energy)
we
\*
+ s-s -(':~(s
~*)^/
'
If instead of considering equation (8)

in (16)
and consider
get
s constant
we
take
and equal
to
its
integral
K'K,
we
have directly
/
= 12*282 + 20-621 loer3

273-7
= 47-481 (log/ 2-1786).
On
this supposition A is zero at temperature 150-9, that is,

122-8 degrees below the normal melting-point: to this temperature will correspond a pressure of at least
122-8
16632 atmospheres.
0-00738
It
would seem
distinction
that at
between the
The phenomena
and below
solid
and
temperature there
liquid forms of water.
this
is
no
of regelation (a better term for which would
be congelation, as Professor, James Thomson suggests) and

glacier motion are easily explained by the lowering of the
melting-point of ice by pressure.
;o
Change of State.
90. Other experiments. Sir W. Thomson's

prediction
an increase of pressure will raise the
melting-point of those
that
bodies that expand in melting was verified

by the following experiments made independently by Professor Bunsen 1
tube
s
AB
of stout glass, 30 c. long, was drawn fine at both

ends the capillary part AD, about 4 c. long, was bent
:
and the
as in the figure,
46
c.
long, was
capillary part DC, about
carefully calibrated.
Both ends being
open, mercury was sucked in and the end C sealed

when the tube was full. Some mercury was driven
out at D by heat the space left on cooling was then
;
by spermaceti, &c., and the end D sealed up.

The end C having been now opened, mercury was
driven out by heat ; on the subsequent cooling air
entered and the end C was again sealed so that BC
filled
could act as a manometer.
mersed
in hot water
was allowed
Fig. 41.
The end D was
then im-
and the spermaceti liquefied it

and the temperature of solidifi:
to cool,
was noted together with the corresponding

by the manometer, which increased with
The temperature of
the length of the tube AB immersed.
it is
fusion was found to increase with the pressure
approxication
pressure, as indicated
mately given for spermaceti in centigrade degrees by the formula

(up to 158 atmospheres)
47-68
0-0207
Similar experiments
denoting the pressure in atmospheres.
with paraffin led to similar conclusions and a corresponding
formula (up to 101 atmospheres)
=
Mr.
Hopkins
46*2 7
+ 0-03 2 6 /.
extended these
results
by employing great
in
pressures the substance to be experimented on was placed
B
full
of
a
brass
a \ in. brass tube A connected to
oil,
cylinder
Glass could not be used bein which worked a piston c.
:
cause of the great pressure, and even the brass had to be

case-hardened to prevent escape of the liquid through its pores.
1
Pogg. Ann.
Ixxxi. (1850) 562.
Rep. &. A. (1854),
ii.
56.
Change of
The
State.
171
liquefaction of the substance could not therefore be di-
rectly observed, but a small iron ball or, better, a
magnetic needle in the tube
at
D, resting upon the
substance while solid, deflected, an external

netic needle;
fell
on
ture of
this deflection
liquefaction of the solid,
an oil-bath
in
when
ceased
mag-
the ball
and the tempera-
which the whole instrument

Fig. 42.
was immersed gave the temperature of fusion
cor-
responding to the pressure during the experiment. The pressure was estimated by finding the weight W\ which pushes in
the piston against the external pressure through a certain small
distance and the less weight W^ which allows the piston to
return through this distance, the atmospheric pressure and
weight of the piston being included
PW^ +
F,
whence
if
P is
F the friction, we
The
P=\(W^ + W
pressure on the piston and
z ).
the total internal
have
centigrade degrees thus found are the following
Pressure in
atmospheres.
P+ F,
melting-points in
:
Change of
172
State.
of several vapours, and the results conform very exactly to the
law represented by
(2)
the annexed table gives the constants

Rankine 1 only so far altered that
as calculated by Professor
f may
from
Vapour.
be expressed in
274c.
c. G. s.
atmospheres,
being measured
Change of State.
173
should properly be called the total heat of the vapour.

integral is however inappreciable for all liquids, so that
be taken as the real
summed up
total heat.
M. Regnault's
may
results are
in the formula
the constants of which are given in the following table
Vapour.
This
ft'
Change of State.
74
From
these tables
we can by equation
the constants being thus tabulated
Vapour.
(20) at once find
State.
Change of
VO
OM"*. ON
CO to J^
to * >O
HI
O to
ON M
OO OO
i-i
175
f^ COVO VO
^t-00
r^_ r!
r^w> -^vo
<* <^^0
M-OO
VO G-. ro cOVO
f^
HI
<*
VO
00
f-. "!f
i>
"+
CO
rfs\
1^
ONOO ON CO ON 10
~.OOVO'*-cON<Ni-ii-ii-H
i>.
<$-
M M
oc o
OOOniHif<McOTt-'^-
O
^VC
pppppppppppppppMMMMH
io o;
to
HI
"*
<N
HI
" O<$ O
M
^
VD iOCO.l~-ONtOu~.OO
Tj-
O
O
cOVO VO c*
ON co co ON
HI
ci
JO Cfl
-4}-
ON ^vo >O
oo co "^* c*
o
H*
10 r^co ON
ONtOMt^rOO
I^.'rt-M
O M f
-t~-
iO
-^-
.*
c^
vo
-^-vo
CO
c<
HI
O'OMVOrp^M
<>
r5<NOOVO
ON
CN.OO
ON^
moo
rO to CNVO
ON CO *O
OO VO CO
HI
Is
t^
ON *o to
ON ONVO
co
<*
ONO *
t--oo vo
ONOO J>.VO
-^-
O
M
co
co co
c*
i->.o6
ON
HI
cq
00 VO
"it-
c^
ONOC
c*
ONOO
ei
o ^f^'O O
HI
ro
r>.
tOTJ-POOOVOCOONtOHiiO
.
M OO CO ON
O ^* 1*0 O VO
"^*
ON to CO ci
CO ONVO CO O ^*
"^"00
<s
(N
OOOOOHivocpcii
c< ^^ co 6 vo co 6
**f*
HI
OO to
<N
O^.VO
c*
i>- !>.
-^-O
C^
ON
r^
ONM
ptpcspONONOHiThrv,MVO
to co c* o
to c* o ON
"^" ^ ON
Ovo tototototo ^--^-'^--^-'^-'Î
t->*
ON
t^
cpvp"
> i>.vo
ONOO oo oo *
O O O O M
<N
ts
ooo O MVO cot^Tt-OHiVOVO ONN IOQVO

î-v,
r<^
M O O O
M r>
*H co
ov6
ON to
OsVO
ONVO M ONVO
ONVO
ONOO 00 00
COCOCONMISMMMOOOONON
ON to
10
"
^*
>C
co
HI
ts -<J-VO ONiNVOO^hCN-ÔN-^-OVOcôOiO
iOTj-cOtS<SHiMpON ONOO OO CO r^. X^VO VO
J^ fO iovo
Tl-00
^-oo
ONOO I-.VO
O O
t->.
f^P^^S^^^
tovo l-ÔO ON
00
co rt-vo J^-OO O M co ^-vo

JÔNO HI co^->Ot>.ONO M
i~-VO \OtOTj-cOCJHiOO ONOO i--O to -^- T*- CO 01 M O
tOVO ^
O HI C4 CO ^t" ti"iVO t^ t>00 ON O M IN CO
0606 ONONONONONONONONONONOxO 6 6 6 6 O 6 6
OO ON
ON c<
o oo "^ HI vo " "^ t^* ON ON o* oo to co ON
<>
rr: ro TJ- to tOVO VOVOVOVOVOVOVOvO 'Oto-^--<4ONVO cOOl^' TJ"MGCto<MOVOcOO*>' '^-HtOO'OC*
co co co c*<sc**-iM'-'i-HOOO ON ON ON
^t* "^ "^
Q M
O HI
10
"^t"
VO
M COVO OOOO N <SVO COONON ^-^t f^"^HiOO M.t~- tOO

f^ to "^ O HI O O O cOVO co ^ ci COOO M
O O O CO
co ON O to covo i^~ HI ~^- M ON cooo ON
co c^ co N
TJ-1--.N ONpVOt^t^-tO tOOO 1^-. tp CO U-. <N ONVO t^
HI co ThVO ON coco Tj- o ci rt-OO
M
TJ- t^, co
6 6 6 6
HI
o ro -4- tOVO 00 O <M to
-
r-*
r<
HI
r~-
t->
C;!
tjri
i-i
^M
O O
CO
ca
M M
HI
<s
rOVO ON CO ONVO
f-l
OO
OO VO ON tOOO
-t^-sO
t**
COOO I>VO
VOCOCOM MOOi^M
HI
iO"4-
pppOpOOOOOOpOOHii-iHiMisc^cp
666066666666666666666
S^
VONrOMcOctOOOcÔcO
OMM-q-l^c<ONHirÔHi
^VO co O "^
O N ^ O OVO
-^-OO r^.
COOO
OOOOOOOOOOMMHIC^CO
TJ-VO r>.
COCO
ONOO
c<
to
rO CO CO CO CO rO CO CO
CO cO CO ro CO CO CO CO CO CO CO CO
ON O HI d ro * tOVO i~~-OO ON O HI c* co < vovo *"-
l^-OO
ooooooooooooooooooooo
M
O
O
cs
ro
Tj-
io\o t~ÔO ON
HI
c*
CO
^t"
>OVO 1--OO ON
crj
ij6
Change of
State.
94. Critical temperature. Examination of the numbers

column 9 shews that the real latent heat of steam may be
in
very well expressed by the formula
X'
23924-33.20;
thus X' diminishes as the temperature rises and is zero when

Hence above this temperature
720-6 c. or /
994-3.
there
is
no
difference
between water in the liquid and the

is for water what Dr. Andrews calls
vapour forms, or 72o-6c.

the critical temperature.
The
formulae for other substances lead to a similar theoretical
determination of their
critical temperatures, but these as a rule

so far beyond the limits within which the formulae have
been justified that the results cannot be relied on; the cal-
lie
culated temperatures however agree very well with those ex-
perimentally determined.
Andrews
3
,
To MM. de
and Avenarius
la
do we owe
Tour 1 Drion 2
,
direct determina-
the last enclosed the liquid in greater or less quantities

hermetically-sealed glass tubes, and in all cases (even
though the different masses used caused the pressures to be
very different) it was found that over a certain temperature
tions
in
fixed for each substance there

liquid
and vapour
is
no
distinction
between the
so that pressure alone will not cause

following Dr. Andrews, we reserve the
states,
a gas to liquefy if,

term gas for a substance in the third state of aggregation and
above the critical temperature, using the word vapour when
Hr. Avenariu|s thus finds

temperature is below this.
292-5, 276-1, 246-1 c. as the critical temperatures for carbon
tetrachloride, carbon bisulphide, and acetone respectively, while
the
the temperatures calculated from
M. Regnault's formulae
are
346, 333> 272.
95. Specific heat of saturated vapour
The
of inversion.
1
Ann. de
ch. et
values of
s' given in
Ann. de ch. et de phys. [3] Ivi. (1859) 5.

Phil. Trans, clix. (1869) 575.
Pogg. Ann.
de phys. [i] xxi. (1822) 127, 178; [i] xxii. (1823) 410;
[i] xxiii. (1823) 267.

3
:
temperature
of the
column
cli.
(1874) 303.
Change of
table
preceding
formula
thus
%' is
zero
summed up
are
when
State.
177
exceedingly
well
in
the
789-8 or at the temperature 5i6-ic.,
all lower, and

was
discovered alordinary, temperatures.
most simultaneously by Professors Rankine 1 and Clausius 2
being positive
thus
for all
higher and negative for
This
all
fact
1850. The interpretation of the negative sign is that in

order to keep saturated a dry vapour which is rising both in
temperature and pressure we must withdraw heat, otherwise the
in
vapour
will
so that
its
become superheated thus if the vapour expands

pressure diminishes we must communicate heat in
:
order to keep
condense
steam for
it
saturated, otherwise part of the
to supply the requisite heat,

all
and
vapour
this is the case
temperatures below 5i6-ic., which
is
will
with
called the
temperature of inversion ; above this temperature compression

will cause condensation and expansion vaporization.
M.
Him
gave an experimental demonstration in 1862 of

the condensation accompanying a sudden adiabatic expansion
of dry saturated steam, and thus proved the negative sign of s'
low temperatures he allowed steam to pass gently from a

where it was generated under 5 atmospheres pressure,
through a copper cylinder, 200 c. long and 15 c. in diameter,
at
boiler,
the ends of which were closed by plates of glass, until all air
and condensed water had been driven out and the sides had
attained the temperature of the steam
the exit-stopcock of the
cylinder was then shut, and, the cylinder being full of dry
saturated steam, the connection with the boiler was cut off and
the exit-stopcock suddenly opened; the pressure at once fell,

and the cylinder, which had previously appeared perfectly
an observer looking along its axis, became

perfectly opaque from the formation of a cloud; this cloud
transparent
1
2
to
Trans. R. S. E. xx. (1850) 171.

Ixxix. (1850) 391, trans. Phil. Mag. [4] ii. (1851) 14.
Bulletin de la Soc. industr. de Mulhouse, cxxxiii. (1862) 129; Cosmos, xxii.
Pgg- Ann.
(1863) 413.
Change of State.
however soon disappeared, heat being supplied by the vessel
it cooled from
i52c. (the temperature of saturated steam
under 5 ordinary atmospheres) to iooc. (the temperature under
as
atmosphere).
M. Cazin 1 improved
apparatus by connecting the cy-
this
which a piston was moveable, and placing

the whole in an oil-bath the temperature of which could be
linder with another in
varied at will
to suddenly
saturated vapour in the one could then be allowed

expand into the other, or when filling both could
:
be suddenly compressed by the motion of the piston. A cloud

was always formed when steam expanded but never on its
compression, and with carbon bisulphide the same occurred;
contrary, as noticed earlier by M. Him, ether vapour
on the
never condensed during expansion but always on compression,

shewing that its temperature of inversion is below ordinary
temperatures; further, the temperature of inversion appeared
to be
between 125 and I2p
and i30c.
between 115
and
for chloroform,
c.
These
results are in
theory,
M. Regnault's formulae giving 790-2,
iooc.
for the
for benzine
keeping with
113, 123-5,
temperatures of inversion of carbon bisulphide,
ether, chloroform, and benzine respectively.
96. Adiabatic expansion. In the above experiments the
vapours expanded nearly adiabatically
by a
single
example
mixture at
50
it
may
therefore be well
half water and half steam,
c.,
in a non-conducting cylinder
out a piston ;
to illustrate the general action.
and
then allowed to
enclosed
what happens ?
For water both
TV
and - increase with
and therefore
.c)w.
is
is
expand by pressing
always negative; hence
is
)^
(^-
(
v
d/'*
/^^Z\
of opposite sign to
(-JTT)^
and vaporization or condensation accompany expansion ac-
88 m
cording as
i
8
is
positive
or negative.
1
C.
In the given case

R.
Ixii.
(1866) 56.
this
binomial
is
Change of State.
i
--
= +0-075,
tinues only
till
and vaporization
the binomial
becomes
m=
7-
\ ~
= o-5, Y
T423-7
49-010.
4)
so that by
on
273-7
7
/
#=
18-432,
8)
/*/
r+/
J
-dt=
273-7 *
(0-5x49-010) +18-432
42-937,
substitution of the above value of m,
A
7
/
*
:>
The
sum
zero, or
initially
mo
or,
at first occurs, but con-
-s
Now
179
/"'
J273-7
273-7
^
7*=
42-937-
'
93 gives 42-897 and 43-607 for the values of this

120 and i3oc., between which temperatures therefore
table in
at
lies; and, as the difference
to
sponds
43-607
42-897 =0.710 correan increase of 10 degrees, the difference
corresponds to
42-937-42-897 = 0-040
0-6 degree, whence
i2o-6c.
value of
at this
1-994
and
therefore
mu + s,
volume
temperature
0-6.
(1-994
which was
is
1-943)
'99 I >
= 0-5023:
becomes
Further, the
the specific
also
initially
(0-5x382-12)+ 1-09 = 192-15 c.c.,

becomes (0-5023 x 855-61)+ 1-06 = 430-83 c.c.
and the work that has been done during the expansion per
gram of the mixture is
;
191-06
x 49'567o-429'77 X 23-2607
+ ^(42-860 + 42.629) x (150

t>7
5)j
if
the
mean
120-6)'= 730-2 megalergs
value of s between 150
used in the integral.
and 120-6
c.
is
i8o
Change of
As
State.
on the temperature will still fall, but

be continuously condensed, and the mixture will
again become half water half steam when again
the expansion goes
will
vapour
'273-7 /
Now
and 90
100
at
spectively,
sum
this
c.
42-870 and 42-946 re-
is
so that the required temperature
volume
also
429-77 x 23-2607
the
c.c.,
c.
1108 x 9-4057
+ 4(42-629 + 42-377) X (120-6

97.
1108
9 1 -3
is
becomes (0-5 x 2213-6) + 1-04

and the external work is increased by
specific
91-3)
Density of saturated vapour.
820-6 megalergs.
was formerly
It
as-
sumed, for want of experimental data, that vapours conform

to the gaseous laws even up to their points of saturation
thus
:
the theoretical density of steam (H 2 O) under one

pressure at o c. (if it could so exist) being
| {(2
x 0-0000884) + 0-0014107}
megadyne
= 0-0007937
grams per cubic centimetre by M. Gay Lus sac's law, since

0-0000884, 0-0014107 are the densities of hydrogen and oxygen
which agrees very
respectively under the same circumstances
well with
M. Regnault's experimental
value
0-623 f r its
above
density in a superheated state relative to air, or in the
hypothetical state 0-623 x 0-00127592 =0-0007949 g. per c.c.

the specific volume of steam was calculated from the formula
pv '"
_
"
0-0007937
273-7
which was extended even to the

/-
state
of saturation, so that
^-
4-603.
(2l)
shewed that grave error was introduced by

and,
having deduced the values of / for steam
assumption,
at different temperatures by the method we have followed, gave
Hr. Clausius
first
this
the empirical formula
Fogg. Ann.
Ixxix. (1850) 520, trans. Phil.
Mag.
[4]
ii.
(1851) 116.
Change of
steam
for the density p' of saturated
to
air, a', b, c
being constants
under the pressure
at
temperature 6
c.
relative
as the density of air at 6
/ megadynes
density p of saturated steam
181
State.
is
is
OOI2 759 2 /
and
c.
ab so i ute
given by
=7=
where the constants, expressed
=
Hr.
6-2899,
in
c
30-087,
Zeuner has suggested
our units, are
more simple and exact, and

of the numbers we have given
as at once
ciprocals
table in
/3
93 extremely well
This gives
O'942.
if
Or
is
it
in
(23)
represents the re-
column 14 of the
the constants are a
f*t
where y
= a/ft,
^=272-48.
1-0072,
the formula
(24)
v,
a constant, which in the case of steam
again, since s
equation (3)
may
is
= 0-000597,
is
1675.
in general very small compared with /,
be written
// = A-V,
(25)
or in the case of steam
//=
In 1860 Sir
I276
+ 4-3230-0-004570 2
W. Fairbairn and Mr. Tate 2 made
direct ex-
periments on the density of saturated steam by means of the

saturation -gauge of the latter.
This in its
theoretical
B
its
form consists of two globes
A and
some
liquid and
and
a
bent
tube conconnected
vapour,
by
void of air but containing
taining mercury ;
if
B, and the whole
is
contains less liquid than

placed in a bath the tern-
Fig 43<
perature of which is continuously raised, the
level of the mercury in the tube will remain
unchanged until
the liquid in
is
the vapour in A,
vaporized
will rise at a, as at the
of increase of pressure
is
same
greater
Grundzuge der mechanischen Warmetheorie, 2 te Auflage (1877),

cl. (1860) 185; clii.
(1862) 591.
Phil. Trans,
all
further heating will then superheat
and the mercury
temperature the rate

1
2 ^6.
for
Change of
State.
saturated than for superheated vapours.

liquid introduced into A, and
mass of
then
If
we
we know
the
note the volume
occupied by its vapour and the temperature when the mercury

at a begins to rise, we can determine by division the corre-
sponding density. By using a bath of the vapour experimented

upon there can be no danger of bursting the globes, as the
pressures will be the
saturation continues.
The
same within and without so long
actual apparatus used
Torricellian
vacuum was made
as
represented in the figure: a

A and a small glass
is
in the globe
globule a inserted, the globule having previously been filled

with warm water and allowed to cool with
its
stem dipping in mercury, so that
it
was
safely closed without any explosion being
M. GayLussac's method:
was heated in a water-bath B
provided with a manometer and a safety
valve, and the lower part of the stem was
necessary, as in
the globe
heated as nearly as possible to the same

temperature in an oil-bath standing on a
sand-bath, the mercury levels being ob-
The vapour was
served by a cathetometer.
superheated so as to overcome
always
the force of cohesion between the water and
first
the glass, and this force did not again come

into play until the glass was again wetted by
Fig. 44.
The exthe vapour becoming saturated.

to
c
an
d
led
to the
from
144-7
58-2
periments ranged
->
formula
where the constants (expressed

<z
from
this
Dr.
25-62,
we deduce
a
where
Herwig
p gg- Ann
1
7+<'
in our units) are
=1677-61,
p-a-
0-039032,
has lately
(3
made
0-02440:
ft
(27)
2-556,
65-505.
elaborate experiments
cxxxvii. (1869) 19, 592
cxli.
(1870) 83.
on
the
Change of
i8 3
State.
behaviour of vapours with regard to the gaseous laws, and

arrives at the curious conclusion that if P, V represent the
pressure and specific volume of a vapour at temperature / in the
/ its pressure and specific volume

and
same temperature when saturated, then in all cases
perfect state
at the
PV
yy= 0-0595
whence,
(28)
A//,
the density of the vapour in the perfect state
if p' is
relative to air,
3-87*5
//=
P
pV/
48-28
// _
~7~
V7
o-o595
or
The
substances experimented on were alcohol, chloroform,

carbon bisulphide, ether, ethyl bromide, and water: for the
= 77'49^=
v
so that
latter />'= 0-623,
The
apparatus used consisted of a graduated glass cylinder

connected with a wider one B
which communicated with a ma-
nometer C through a drying tube

D a branch tube E led to an air;
pump when required.

were taken
Precautions
to exclude
damp from
by using warm mercury
in
fill-
amount
the small
ing it initially
of air also that entered
:
when a
glass globule containing the liquid
was introduced was
carefully de-
termined by varying the pressure and temperature before exploding the globule. The range of temperatures was not great
;
in the case of water they varied only
Hr.
Her wig
from 40 to 95
arrived also at another curious result.
c.
M. Reg-
nault 1 found an apparent exception to the law/~= F(f] in the

case of ether-vapour if air be present in any quantity, and
1
Mem.
de
I'lnst.
xxvi. (1862) 679.
Change of State.
184
when condensation begins, even in the

manometer indicates a less tension than
given by M. Regnault's formula, and that as the volume
Hr.
Herwig shewed
absence of
is
is
air,
that
the
lessened condensation continues, the pressure gradually in-
creasing up to M. Regnault's maximum; he explains it 'by

the assumption of a force of adhesion acting strongly between
the walls of the apparatus (sc. both the glass and the mercury)
on the one hand and the ether on the other, by means of which
from a highly superheated
in passing
state to a smaller
the vapour begins to condense before reaching the
volume
volume that
would correspond to purely saturated vapour were no such

force existing
on the contrary, at a very great distance from
the ideal limits of saturation up to the superheated state such
:
an influence can no longer be assumed.'

If/"/ is calculated from the equation
(28),
the
maximum
tension as determined by M. Regnault being then substituted

for/ a value of / is obtained which is exactly the volume which
the vapour
is
observed to occupy just before condensation
begins whence the real tension of the vapour is most probably

constant throughout the condensation, and the force of adhesion
:
that acts at any moment is equal to the difference between the

maximum and the indicated tensions.
The same action occurs with water and with ethyl bromide,
being also slightly traceable in carbon bisulphide.

*
M.
has experimented on superheated steam, and
2
Hr. Ritter shews that his results can be summed up in the
Him
formula (suggested by certain theoretical considerations)
'-+'.
K
p
(30)
v?
S=
^ 4-653,
where
1043800.
This also represents the numbers of 93 much better than any
of the foregoing formulae, as
is
shewn by the subjoined
table
wherein the specific volumes of saturated steam, calculated by

the different methods, are compared
it
may therefore be taken
:
as the characteristic equation of steam.

1
Theorie mecanique de la Chaleur, 3^ me ed. (1875)

[2] iii. (1878) 447.
Pogg. Ann.
ii.
Change of
State.
State.
Change of
between the
work
that
is
solid
done
and
-v
J fdv =/(v
2
The
1)
an isothermal
isenergics (or isodynamics) are,
J.if
external
is
=fu(m^-m^.
(v- s ) (vt-f) + [' (s
by
(15), given
by
-/ at )<// = const,
.
be expressed in terms of/! As we only

s, /,
the relations between these quantities empirically in a few
wherein
know
The
forms of water.
liquid
in passing along
must
all
cases and under limited conditions,
expression for
all
this function as
Zeuner 1
substances
we cannot find a general

but examination of the values of
deduced from the
to the conclusion
93 has led Hr.
table in
that the isenergics, at least for
steam in contact with water, are given by
fv
= const,
the value of n for water being 1-0456.

The isenergics must be
limited by the curves that bound the portion of the diagram
which correspond to the saturated state their prolongations
:
over these lines represent the intrinsic energy of the body as a

superheated vapour or as a vapourless liquid.
These bounding curves may be called by the names given to
their water-types
by Professor Maxwell;
viz. (i) the ice line
separating the solid from the liquid state, (2) the steam line
separating the liquid from the gaseous state, (3) the hoar-frost
Theory does
separating the gaseous from the solid state.
not indicate the form of these lines, but we gather from (23)
line
that the steam line for water at least
fv
is
given by
= const.,
v being 1-0616 for water.
curves representing the states wherein m is constant

by Hr. Zeuner the lines of constant vapour -mass for one
The
called
species of saturation
<f>
are,
/**
Jf
1
by
(13), given
dtm -t
Grundzuge
by
A.
const.
&c., p. 306.
State.
Change of
these are
more or
87
last-mentioned lines which
less similar to the
are their limiting cases.
The
iseniropics or adiabatics are

"*
by (13)
-dt=.
as in
88.
them
Professor
where
jx is
tion of m.
Rankine
attempted to express
vn
a constant; but Hr.

It is
has
f = const,
form
in the
const.,
Zeuner shews
that
ju.
is
a func-
easy to see that for such
vapours as steam which condense on adia-
expansion the isentropics cut the

steam line S as in i, since as the volume
batic
increases
the
vapour changes
superheated to the
saturated
that for vapours like ether,
in the reverse
as in
from
the
and
state,
which behave
p--
manner, the isentropics cut
2.
Chapters VI and VII in Professor Maxwell's 'Theory of

should be studied on the subjects of this article.
100. Triple point. We are led to very interesting results
Heat
if
we
'
consider the projections of the real
frost lines of water- substance
hoff 1 shewed
in
1858
on
ice,
steam, and hoar-
the plane of pt.
Hr. Kirch-
that these projections of the steam
hoar-frost lines are not continuous, as
but intersect at an angle which
is
and
M. Regnault
thought,
re-entrant downwards.
The
assumption being made that there is some temperature / at

which both ice and water can exist together and at which the
steam formed over
ice is exactly identical
formed over water
in
condition with
temperature being about 0-0074

c., as the pressure will be only about 0-00614 instead of i atmosphere it follows that, when unit-mass of ice at this tem-
that
this
perature is directly transformed into vapour, its intrinsic energy

increased by exactly the same amount as if the transformation
is
were
first
into the liquid
and then
during vaporization at temperature
into the vapour state.

/
from the
Now
solid state, the
i88
Change of
State.
'
(/
/
increase of intrinsic energy is #
cents denoting the process of sublimation
/" ) by (9), the ac-
but during the lique-
is A
", ^ \ * s
"
same pressure _/ and
the
faction the increase of intrinsic energy
the latent heat of fusion at the
change of volume, and during the subsequent vaporization the

is
hence
Q)
^(iz Q tQ
increase of intrinsic energy
o''o-/o) = \>-/o<' + oK'o-/o)

equal to UQ + U ", and it may be taken equal
:
'
but
is
"
is
Here
less
than
its
93); whence
hoff remarks:
as
= 3361-6 (see 89), U = 209905

= 0-0000585. Hr. Kirchfir/
273-7 sensibly, A
(see table,
to
2oo,ooo th part; thus the equation becomes
fir
'This difference
too small to be safely in-
is
It is interesting howsame sign and order of

furnished by the numbers
ferred from
Regnault's experiments.
ever to remark that a difference of the
magnitude as the theory requires is

which Regnault gives as the results of
the empirical formulae give
fir/
whence -sr/ fir = 0-000043. The

other, and the angle of intersection
w ^
as
"cr
lines
is
For
his experiments.'
= 0-000482,
fir
= 0-000439,
therefore cut
re-entrant
each
downwards
Professor James Thomson
has pointed out that the ice line
must
also pass
through
point P, which he
therefore calls the triple
this
point:
thus in the dia-
gram, which is of course

not drawn to scale, PC
may be
taken
as
the
steam line which ends

the critical point C,
as the hoar-frost line,
Fig. 47.
at
PA
PB
as the ice line, the length
OE
1
representing
B. A. Rep. (1871)
atmosphere.
ii.
30; (1872)
Following Sir
ii.
24
W. Thomson's
Proc. R. S. xxii. (1873) 27.
Change of State.
189
suggestion he has further applied equation (9) in the form
Ao
with our former notation and considering as before #
Now
at
oc. A
25200
(see
sum
vaporize the ice: thus
=U
and
93),
V= 3361-6+25200 =
being equal to the
'
28561-6,
of the heat required to melt and then
or
formulae giving the ratio

Kirchhoffs theorem.
1-133,
M. Regnault's
empirical
Hence a new proof
i-io.
of Hr.
Joining together these two theoretical results
ô'-ô
we
find
w =
0-0000585,
0-000498,
tzr
^=
OT o
1-133,
= 0-000440,
which are nearly the values given by experiment.

This reasoning applies to substances in general.
interesting discussion of the
istic
A full and
form and sections of the character-
surface representing the different states of water has lately
been made by Hr. Ritter 1
a general representation of the

properties of a substance by means of a surface, the coordinates
of which are the volume, entropy, and intrinsic energy of the
:
2
body, has been given by Professor Gibbs , and should be
in
Professor Maxwell's Theory of Heat/ p. 195.
studied
'
Pogg. Ann. [2] ii. (1877) 2 73> trans. Phil Mag. [5] v. (1878) 202.
Trans. A cad. of Sciences of Connecticut, ii. (1871) 309.
CHAPTER
On
the
Flow of
IX.
Fluids,
In 1856 Sir W. Thomson and

extended the reasoning of the former, which has
been given in
71, to the case in which fluids issue from an
orifice with a sensible velocity.
If the mean velocity of unit-
General formula.
101.
Dr. Joule
mass of the
pressure
changed from VQ at the section where the

where the pressure is p, kinetic
fluid is
to v at the section
is />
zy ) has been imparted to it; and if moreover it

rise or fall through a height h in reaching the
2
energy \(v*
has had to
its velocity is v, energy equal to

+ gh (where
the acceleration of gravity) has been absorbed in the process thus equation (3),
71, becomes more generally, if H' is
position where
is
the energy turned into heat
or
?- v
Now
*)
by means of
= (E-E} + (p
v Q -pv)
friction,
+ (H-H')-V$gh.
(i)
turned into heat by friction is always

of
fluid that escapes, so that H* may
small
unit-mass
per
very
be neglected also 2 is either zero or very small, and as a rule
the energy that
is
the efflux occurs approximately adiabatically, so that

zero : hence the equation becomes
the last
is
also
2
2
(2)
i(* -i>o ) = (E -JB) + (p v -fiv) + gh,
term being comparatively negligible in the case of
gases, but not in the case of liquids.

stitute the
value of
E given
Proc. R. S.
in (25),
viii.
We may
57,
(1856) 178.
further sub-
and obtain
On
102.
the
Flow of Fluids.
191
fluids.
Looking upon liquids as

and remembering that there
will be no change of temperature if no compression is effected
and no heat added, or that dt = o, we see that the equation
Incompressible
so that dv
incompressible,
becomes
in this case
(tf
*)
= o,
(p Q -p)
v,
+ gh,
(4)
the upper or lower sign of the last term being taken according
as the liquid rises or falls to the position where its velocity is v
:
if
then there
is
no
initial
velocity
we have
vz
(A>
which
is
M.
p) v o
h,
(5)
Bernoulli's formula for the velocity of efflux of liquids.
We
103.
shall consider a perfect

Compressible fluids.
of
a
as
and
examine two cases.
type
compressible liquids,
gas
at
constant
In
this
case
temperature.
(1) Efflux
f
hence,
if
is
the heat added to maintain this constancy of

if the initial velocity of the gas is zero,
temperature and
pQ
v*
2
Supposing that no heat

constant or
H = o, we
required to keep the temperature

obtain M. Navier's formula which was
is
deduced from the improbable hypothesis of
parallel sections.
In this case,
(2) Efflux without receiving or losing heat.
k
pv
Q
p. V,
63,
(/
/), by (53),
E -E =
hence equation
(2)
becomes,
if
the initial velocity
is
zero,
(7)
On
192
the
Flow of Fluids.
The second formula in (7) was

63.
by (62) and (64),
deduced in another way by Hr. Weisbach 1 in 1855.
first
If so much friction is caused against the orifice or walls of

the containing tube or vessel that H* cannot be neglected, / will
be greater and therefore u smaller than in the other case. The
variation in the velocity of efflux with the
due to the variation
is
variation of the
The
/>
form of the ajutage
by reason of the
form and dimensions of the vena contracta.
in the frictional effects
velocity increases as
p diminishes
in
comparison with
but the formula gives
2K
as the limiting velocity which cannot be surpassed.
EE
Q from (14),
Substituting for
get from equation (2), if the initial velocity is zero
104. Saturated fluids.
we
86,
and the
fluid
remains always saturated,
^2 = (/.".--*(A-
.(A
-/..)- Jf
t (S-/JK
(8)
with the further equation of condition
by
(18),
If
ing
88.
we
it
consider the efflux of saturated vapour only, considerto remain saturated, so that m Q
i, V Q
sj we have
(10)
if
condensation occurs during the
Ingenieur-
efflux,
and
und Maschinenmechanik, 3 te
Aufl.
i.
821.
On
the
Flow of Fluids.
no condensation occurs.
more simple equation
In the
if
7
by
92, if
(20),
and
initial
jj
latter
we
obtain the
= ?r.-&
denote the
()
total heat
fy
final states,
case
193
(^f^s^fs] and
reason of their smallness and
of the vapour in
its
being neglected by
being replaced by
its
approxi-
mate value K.
For all known liquids fy increases with / and the temperature

diminishes during adiabatic expansion hence fj
fj is a posiStill the
tive quantity, and
contains
no
contradiction.
(12)
:
issuing jet enlarges rapidly
consequently
same
beyond the contracted
vein,
and
velocity diminishes equally rapidly from the

hence v becomes negligible before the vapour
its
section;
mixes with the
air,
and
thus,
fj
fj
being neither zero nor
negative, the hypothesis of the persistence of saturation is untrue: and, further, there cannot be condensation, since,
being a fraction, the right-hand expression of (10) is greater

than that of (n) which we have just shewn is itself too great.
Hence
at a short distance from the orifice the issuing vapour

must be superheated and dry, though close to the orifice where
v is great it is saturated and partly condensed
the superheating
and re-vaporization of the condensed droplets being due to the

of the particles among each other and the transformation into heat of their energy of motion.
This is Sir -W.
friction
Thomson's 1
explanation of the well-known facts that, while

scalded if placed close to an orifice from which
high-pressure steam is escaping, it experiences only a pleasant
hand
the
is
sensation of
warmth
if
held a
little
distance off in the
jet,
and
that if the energy of motion has been utilized in actual work

as
in driving the piston of a
the
steam
afterwards
steam-engine
issues
1
saturated
Phil.
and therefore
scalds.
[3] xxxvii. (1850) 386: see also Clausius' memoir, Pogg.

(1851) 263, trans. Phil. Mag. [4] i. (1851) 398, and the notes
Mag.
Ann.
Ixxxii.
Phil.
Mag.
This explanation was
[4]
i.
(1851) 474,
ii.
(1851) 139, 273.
On
194
the
Flow of Fluids.
given afresh a little later by Hr. Clausius, who had misunderstood Professor Thomson's expression 'fluid friction/
The temperature of the vapour /' before mixing with the air
may thus be approximately determined. We may suppose
and dry,
becomes equal to the external pressure, its
temperature falling from / to / corresponding to this last
pressure and its velocity becoming v, and that its kinetic
energy is then immediately and totally transformed into heat,
bringing it at constant pressure into the superheated and dry
that the vapour expands, always remaining saturated
till
tension
its
state at
/'
temperature
K* being
with no velocity of motion.
r?
its
specific heat in the superheated state,
absorbed in
K* dt
is
the heat
Jt
superheating, and therefore from (12)
this
i?
r*
&-& = -=/ J?dt.
In the case of steam
and
fj
hence taking
to /
sponding
/,
0,
/',
K'
(13)
Jf
0-4805
x 41-55
19-965,
= 25200 +12-673 0;
as the temperatures centigrade corre-
Q'
we have
i2-67 3 (0
-0)=
19-965 (0'-0),
1 1
is generated at i5oc. where the pressure

4-7761 atmospheres and escapes into the open air where the
Thus,
is
the steam
if
pressure
is
1-0136 atmospheres, so that 0=- iooc., we find

0'
100
If the issuing fluid

orifice is
(150100)
-= f
(8)
and
c.
132
consists of liquid only
below the surface of the liquid
and equations
(9)
then
mQ
as
when
= o,
become
- fv - m (\-fu\- r
*t
(s
_/^fw/-
VQ
the
=J
On
the
Flow of Fluids.
195
the upper or lower sign of the ambiguous term being taken

according as the liquid is forced upwards or downwards.
In the case of water very little

s
may be taken as zero or v
is
;
vaporized and therefore

the temperature is further
unchanged (as shewn by the second of the equations) so

s = sn and therefore
that
(15)
the formula
we obtained
in
102.
o 2
CHAPTER
X.
Heat- Engines.
WE
and
and
shall briefly discuss certain heat-engines
their theory,
complete account of them to Professor Rankine's 'Manual of the Steam-Engine and other
refer the student for a
Prime Movers/
M. Garnet's engine.
35 we w iH now merely find
105.
the operations
on
a perfect gas.
/,
v',
D by
Denoting
p\,
v'
and therefore
v=
V
K ~~
this fully
(fig.
n) by />
we have by
or
V
AD
(62),
v,
C by
for the
tem-
by /,
being
63, since
= ^,
AB, CD
(I)
V '
Vn
also
Further the areas

n
equal to
have described
the isothermals BC,
peratures /, / respectively,
are adiabatics,
and therefore
We
the changes of volume during

the supposition that the working substance is
(/
/ )
BAabB, CDdcC
by
(64),
63:
are equal, being each

also the area
which represents the heat

given by the boiler
isenergics and isothermals of a perfect gas coincide
similarly the heat
h given out
to the refrigerator
is
BCcbB,
since the
is
Heat- Engines.
whence
the efficiency
197
is
by the above relation, as we should expect.

106. Engines of maximum efficiency. If /, / are the
highest and lowest temperatures between which an engine
maximum
the
works,
as
shewn
in
attainable
efficiency
and
41,
it
by the engine
can only be obtained
if
heat
is
is
taken in only at / and given out only at / as in M. Carnot's

But with air as the working substance this engine is
engine.
really impracticable by reason of the enormous cylinders that
,
would be required, since the rise in temperature from / to /

would have to be obtained entirely by compression and the fall
from / to / entirely by expansion there are however an infinite
:
number of other engines possible which possess maximum

efficiency when they work under the following conditions, which
are equivalent to M. Carnot's:
(i) that heat is taken in from
and given out
to external bodies only at
two fixed temperatures,
(2) that in the passage of the working substance from a higher

temperature to a lower during any stroke all the heat is given
out that
is
necessary to bring the substance from the lower
temperature to the higher in the next stroke, being in the

time stored up within the engine
performed
BC
are
is
the
itself.
Thus,
if
mean-
the cycle
ABCDA, where AD,

isothermals
corre-
to the extreme temper-
spending
atures at which alone the working
substance is in communication
with the exterior, and
if in passing from Qf to Q the substance

emits as much heat as it absorbs
in passing
from
P to X,
PQ,
G
Fig. 48.
/>'(/
being any pair of isothermals, the

engine will have maximum efficiency.
The
part of the engine
Heat- Engines.
198
wherein heat
is
stored between successive strokes of the piston
and
called the economizer or regenerator,
is
to Dr. Stirling:
in
its
latest
form
it
its
invention
consists of a
is
due
chamber
containing thin plates of metal or other solid conductors or

a network of wires ; these become heated as they cool the fluid
passing through them, and restore the heat when the cooled
fluid repasses in the contrary direction.
Practically this process

not perfect, as heat is lost by conduction, &c. ; but the waste
can be reduced below five per cent, of the heat alternately
is
abstracted from and restored to the
We
fluid.
easily determine the relation between the curves AB

so that ABCDA shall be a cycle of maximum efficiency.
can
and CD
=F
AB
If the equation of
bep
(v), we may eliminate v between
this and the characteristic equation of the working substance of
the engine
= o and express the result as

> = X,
_/"(/>, v, /)
and corresponding
to
CD we
?= *
Now
(3)
shall get a similar equation

(/>.
_
PQ, P'Q' are consecutive isothermals, unit-mass of

the substance will absorb Kdt -\-Ldp heat in passing from -Pto
if
and give out Kdt+L' dp' in passing from (/to Q; these

and we therefore have, by (8), 53,
quantities are to be equal,
For working substance we may use a perfect

Rl, and the equation becomes
po
which
gas, for
= dp'
7'
p = cp',
dp
7
whence
c being a constant;
and
(4)
this is the relation that
must
subsist
between the pressures at the points P, Q where the paths AB,

CD cut any isothermal. Professor Rankine calls two such
AB, CD isodiabatic to each other.

107. Dr. Stirling's engine. In this case
where a is a constant, so that if air is used
lines as
const,
isometrics.
is
(3)
is
at,
in the engine
the equation to the lines AB, CD, or they are

air thus (i) is raised in the regenerator from
The
Heat- Engines.
/
to
ature
199
/at constant volume, (2) expands at this constant temperon receiving heat from the furnace, (3) is cooled to / in
passing through the regenerator at constant volume, (4) conup heat to the re-
tracts at this constant temperature, giving

frigerator.
In
this engine,
receiver
invented in 1816, the air is enclosed in a

consisting of two similar vessels one
DCABACD,
within the other, which communicate by holes pierced through

one ; the annular space
the hemispherical bottom of the inner
AC between
them containing conducting materials and forming
the regenerator.
hollow and
filled
plunger E, which is
with brick-dust or other
badly-conducting matter, moves up and
down
within the internal vessel, and by
driving or drawing the air through the
regenerator causes it to be cooled or

heated at constant volume the refrigera:
tor consists of a fine
copper tube coiled
horizontally in the space
CD and traversed
by a current of cold water

linder
and
H the
is
the cy-
working piston. Conbottom of its
sider the plunger at the
stroke; as it rises, the cold air passes from above through

the regenerator and through the holes into the space vacated,
being thus heated at constant volume ; receiving heat from the
furnace on which the receiver stands the air expands at constant
temperature, lifting the piston H\ as the plunger descends, the

air passes through the regenerator, leaving its heat there and
being cooled at constant volume, and then through the refrigerator to the vacated space, being thus brought to its lowest
temperature; the refrigerator absorbs also the heat produced
air in G as
descends.
slide-valve
by the compression of the

is
required for this engine which
108. Captain Ericsson's
const., or the lines
AB, CD
not shewn in the figure.

engine. In this cycle (3)
is
are isopiestics.
The
is
air there-
fore (i) passes through the regenerator at constant

pressure,
being heated from / to /, (2) receives heat from the furnace
200
Heat- Engines.
and expands
at
constant temperature,
this
(3)
regenerator at constant pressure, falling again to /
pressed
at
repasses the
,
(4)
is"
com-
constant temperature, giving up heat to the
this
refrigerator.
This engine was invented 1 in 1833, and it was much im2

In the improved form it
in 1850.
proved by Mr. Dunn
consists of a cylinder A into which air is drawn through the
valve
f as
air is
chased through the valve
the piston a descends
passes at constant pressure
when
the piston rises, the
into the receiver
Z>,
through the regenerator
whence it
where
,
\\d
Fig. 50.
temperature is raised to /, into the cylinder B\ here it is

and raises the piston C, which is hollow
heated by the furnace
its
and
full
of brick- dust, and which
is
connected to the piston
C descends, the air passes out through

the regenerator, giving up its heat, and is caused by the slidevalve G (which is moved by the eccentric H] to go into the
a by the braces
as
Mechanic's Magazine, xx. (1833) II.

Mechanic's Magazine, Iv. (1851) 41.
Heat- Engines.
201
waste-pipe- Z communicating with the atmosphere.
gas than
air
is
it
used,
is
passed
by Z
If
any other
through the tube
M,
a refrigerator, to the valve t where it again

enters the engine,
k, k are air-holes, and e the stuffing-box
of the piston-rod d.
around which
is
In 1851 Dr. Joule 1 proposed

an engine without regenerator or refrigerator, which though
not theoretically of maximum efficiency approaches this limit
109. Dr. Joule's engine.
more
nearly in practice than the theoretically perfect engines

consists of isopiestics
its
and
diagram
from the atmosphere into a cylinder,
by /> z>
becomes
,
it
/, v,
adiabatics.
Air
is
drawn
its state being represented

tnen compressed adiabatically till its state
and is then forced into an indefinitely large
*s
/,
is kept constantly at a
high temhere it absorbs
perature f and at the pressure p by a furnace
heat K(t't} per unit-mass as it rises in temperature to /,
meantime expanding at the constant pressure p into a cylinder
reservoir containing air that
and driving a piston before

the reservoir as entered
is
it,
it
when
as
much
air
has
left
communication with the reservoir
shut off and the air in the cylinder
is
allowed to expand
pressure falls to its initial value, it is then

rejected into the air as the engine returns to its initial state, the
adiabatically
till
its
piston in this last process experiencing a back pressure p Q

have then
.
We
and between these

properties
quantities
of adiabatics
and
the
iso-
piestics give the relations
Fig.
"o
1
'o
Phil. Trans, cxlii. (1852) pt.
i.
65.
Heat- Engines.
thus the work done along
CD
done along
DA
is
along
k(f
is
BC
)
is
p(v'
v)
= k - (/
= R (/'/),
/ ),
the
the
work spent along
PQ(V\V^ = ^(/ -/ ) = R -1 (//), the work

AB is k(t / ); hence the total work done is
K(t'i)
and the
is
R,
in
!_<>,
Qr
maximum.
not the
We shall
110. Steam-engine.
engine
spent
efficiency is therefore
!_4>
which
J?k
-, since
work
which none of the
in practice are found.
consider a theoretical steam-
irreversible processes that occur
Thus, in the diagram unit-mass of water

at /
the temperature of
the condenser,
is
heated
/, the temperature of
the boiler, under a pres-
to
sure
which
is
at
each
instant equal to the maxi-
mum
~F\
~~e
"""72
tension of steam at
the
corresponding ternperature; it is then va-
Fig. 52.
porized at
and expands
into a cylinder the piston of which it drives out; it is then

allowed still further to expand adiabatically until its temperature
falls
to /
it is
The
put in communication with the condenser
finally
and compressed
till it
heat given
is
regains
/
/
proportion
along
is
AB
%dt along
and A along BC\ now a
^0
the adiabatic expansion along
DA
original liquid state.
of the steam will have been condensed during
tn
its
which
is
CD
so that the heat given out
given by
- dt=
Heat- Engines.
from
88
(18),
hence the
203
work done
total
is
r
J/o.
and the
efficiency is this quantity divided
%dt
If the temperatures of the boiler
50
c.,
/=
so that
ciency of 0-236,
= 0-824, m
423-7, /
we
\Q
find,
on
by
A.
and condenser are 150 and
323-7, giving a maximum

calculating from the table in
19562,
dt+ A
effi-
93,
25009; whence the
Jt*
is 0-218, or yfths of the maximum.
In the actual engine we cannot allow the complete expansion
denoted by CD because of the very large values of / thus
efficiency
with the temperatures 150 and 50 c.,

corresponds to a
volume 26-14 times greater than the volume at C, whereas in
practice a four-fold expansion (which lowers the temperature to
99s c -) is ordinarily employed, the most extreme case being

a ten-fold expansion which lowers the temperature to 72'9c.
Thus if the adiabatic expansion CE only is allowed and the
steam
work
is
then put into communication with the condenser, the

If ^ is
is lost, and this may be thus calculated.
EDFE
the temperature at
by
(14),
86,
EDdeE
is
and the area FDdeFvs
but at the same time
and the work
the area
we have
EDFE
is
equal to the excess of
EDdeE
over
FD de F.
If in the above case
ffz^
= 20308,
whence the area
we
area
EDFE
allow a four-fold expansion,
EDdeE
2539,
1498, and the
area
^=0-911,
FDdeF=
1041,
efficiency of the engine
Heat -Engines.
204
is
of the
diminished by 0-060, being therefore only |rds
maximum.
efficiency
the adiabatic expansion been
ten-fold,
the
would have been diminished by 0-022 and therefore
fths of the
We
Had
maximum.
ought further to allow for clearance (the space within
not traversed by the piston, including also

the ports) and for the steam reaching the cylinder at a temfor these details howperature lower than that of the boiler
the cylinder that
is
ever
we must
refer to Professor
Rankine's book.
Another imperfection of an ordinary steam-engine
arises
from the condensation of steam during expansion; the heat

thus emitted certainly does work, but at the same time it turns
the badly-conducting saturated but dry vapour into a fairlyconducting mixture of steam and water, so that the condensed
heat from the entering steam and comthe

cooled steam as it leaves the cylinder,
heat
to
municating
lowers the working pressure and increases the back pressure,
Mr.
thereby seriously affecting the efficiency of the engine.
particles, abstracting
Watt remedied
this by enveloping the cylinder in an annular

a
steam-jacket, which communicates with the
casing,
thus
and
boiler,
being filled with steam supplies sufficient heat
called
to prevent condensation within the cylinder.

may apply Fig. 52 to this case, the only difference being
We
that the
same ordinates of CD correspond
than in the ordinary engine.

to occur,
we have
s'
Jt
heat supplied
Considering the
/*4>
heat
to greater abcissae
dt
full
expansion
added along CD, so that the
is
=A +
by (8), 83, and as X
turned into useful work
is
is
f'-di,
*
Jto
the heat emitted along
>
If the temperatures of the boiler
and
DA,
the heat
refrigerator are
150
205
Heat-Engines.
and 50 c., we find 6011 for the value of the integral or the
work done during the stroke, and the efficiency is therefore
This is less than that given
0-202 or fths of the maximum.
at
the
same
time 563 megalergs more
by the ordinary engine;
work is done per stroke, but only by absorbing 4747 megalergs

more of heat and by expanding |th as much again. A steamjacket is thus no improvement to an engine so perfect as we
have supposed:
advantage
We
might
until the
supplied
it
is
found however practically to be of great
in the case of actual steam-engines.

easily
apply
temperature
of partial expansion
by E-, the heat
this to the case
falls
to
/x
as denoted
is
Ch
Cf
sV/
J tQ
= A +/Ch *dt+ C*X-<//;

J
J
1
tQ
/!
EDdeE is equal to the excess of the heat supplied

over the change of intrinsic energy, or
and the work

along
ED
'dt
{(A
-/.
- (A,-/, Ul
dt
=
J*
work
FDdeF
and
the
the
difference
is
f (V
vj, and the lost work EDFE is

the efficiency may
between these quantities
therefore be calculated.
above case a four-fold expansion were allowed, the

be cooled to iO2c.
thus the heat supplied
would
vapour
would be 27948, the area EDdeE 3201, and the area FDdeF
1285, so that the area EDFE would be 1916 and the total work
If in the
done 4095, whence the
efficiency
is
0-147 or
tns
f tne
maximum.
Other engines have been employed with superheated vapours
and with combined vapours: these are fully discussed by
Professor
Rankine.
(Klarcntron
xfottr,
$tess,
BOOKS LATELY PUBLISHED.
The
Scholar's Arithmetic
with Answers to the
Hensley, M.A., formerly Fellow and
By Lewis
Examples.
Assistant Tutor of Trinity College, Cambridge.
Crown
8vo.
cloth, 4s. 6d.
The
An
Scholar's Algebra.
on Algebra.
By
the
same Author.
Introductory work
Crown
8vo.
cloth, 4$. 6d.
Treatise on the Kinetic Theory of Gases.

By Henry William Watson, M.A., formerly Fellow of Trinity
College, Cambridge.
1876. 8vo. cloth, $s. 6d.
Treatise on Infinitesimal Calculus.

lomew
By
Bartho-
Price, M.A., F.R.S., Professor of Natural Philosophy,
Oxford.
Vol.
Vol.
I.
Second Edition, 8vo.
Differential Calculus.
cloth,
14^ 6d.
Integral Calculus, Calculus of Variations, and DifferSecond Edition, 1865. 8vo. cloth, i&s.
ential Equations.
II.
Vol.111. Statics, including Attractions Dynamics of a Material

Second Edition, 1868. 8vo. cloth, \6s.
Particle.
Vol. IV. Dynamics of Material Systems together with a Chapter
on Theoretical Dynamics, by W. F. Donkin, M.A., F.R.S.
;
1862. 8vo. cloth, i6s.
Treatise on Electricity and Magnetism.
By
Clerk Maxwell, M.A., F.R.S., Professor of Experimental

Physics, Cambridge. 2 vols. 8vo. cloth, il. us. 6d.
J.
Acoustics.
Crown 8vo.
By W.
F.
Donkin,
By A. W.
Chemistry for Students.

Phil. Doc., F.R.S., Professor of
London.
Exercises
M.A.,
F.R.S.
cloth, js. 6d.
Extra fcap. 8vo.
Williamson,
Chemistry, University College,
cloth, 8s. 6d.
Practical Chemistry (Laboratory

A. G. Vernon Harcourt, M.A., F.R.S., and
H. G. Madan, M.A. Crown 8vo. cloth, ?s. 6d.
Practice).
in
By
Treatise on Heat, with numerous Woodcuts

and Diagrams. By Balfour Stewart, LL.D., F.R.S., Professor
of Natural Philosophy in Owens College, Manchester.
Edition. Extra fcap. 8vo. cloth, js. 6d.
Macmillan
&
Co., London.
Third
(KlatenDon
BOOKS LATELY PUBLISHED
Handbook
F.
of Descriptive Astronomy.
Chambers, F.R.A.S., Barrister-at-Law.
Demy
By G.
Third Edition.
8vo. cloth, 2&s.
Geology of Oxford and the Valley of the Thames.

By John
Phillips,
M.A., F.R.S., Professor of Geology, Oxford.
8vo. cloth, 21 s.
Forms of Animal
By G.
Life.
Rolleston, M.D.,
F.R.S., Linacre Professor of Physiology, Oxford.

by Descriptions and Drawings of Dissections.
Illustrated
8vo.
Demy
cloth, i6s.
Text-Book of Botany, Morphological and PhyBy Dr. Julius Sachs, Professor of Botany in the
University of Wiirzburg. Translated by A. W. Bennett, M.A.,
assisted by W. T. Thiselton Dyer, M.A.
Royal 8vo. half
morocco, \l. us. 6d.
siological.
Selections from Berkeley, with an

and Notes.
For the use of Students in the

Alexander Campbell Fraser, LL.D. Crown
Novum Organum.
Bacon.
Introduction
Universities.
By
8vo. cloth, js. 6d.
Edited, with Intro-
duction, Notes, &c., by T. Fowler, M.A., Professor of Logic in

the University of Oxford. 8vo. cloth, 14^.
The Elements
of Inductive Logic, designed mainly
for the use of Students in the Universities.

By T. Fowler,
M.A., Professor of Logic in the University of Oxford. Third
Edition. Extra fcap. 8vo. cloth, 6s.
The Elements
mainly
the
same Author.
amples.
The
of
Deductive
Logic,
designed
for the use of Junior Students in the Universities.
By
Fifth Edition, with a Collection of Ex-
Extra fcap. 8vo.
Logic of Hegel
cloth,
y.
6d.
translated from the Ency-
clopaedia of the Philosophical Sciences. With Prolegomena.

By William Wallace, M.A. 8vo. cloth, i^s.
Genealogical Tables illustrative of Modern History.

By H. B. George, M.A. Second Edition. Revised and
Corrected.
Small
4to. cloth, I2s.
Macmillan
&
Co.,
London.
Sept. 1886.
xfotfc
A SELECTION OF
BOOKS
PUBLISHED FOR THE UNIVERSITY BY
HENRY FROWDE,
AT THE OXFORD UNIVERSITY PRESS WAREHOUSE,
AMEN CORNER, LONDON.
ALSO TO BE HAD AT THE
CLARENDON PRESS DEPOSITORY, OXFORD.

[Every book
bound in
is
cloth,
unless otherwise described^
LEXICONS, GRAMMARS, ORIENTAL WORKS,
An
ANGLO-SAXON.
&c.
Anglo-Saxon Dictionary based on the

',
MS.
Collections of the late Joseph Bosworth, D.D., Professor of Anglo-Saxon,

Oxford. Edited and enlarged by Prof. T. N. Toller, M. A. (To be completed
in four parts.) Parts I and II.
HWISTLIAN. 4to. 15-$-. each.
CHINESE.
Summers.
Handbook of the Chinese Language.
1863. 8vo. half bound,
By James
I/. Ss.
Record of Buddhistic Kingdoms by the Chinese

',
FA-HIEN.
Translated and annotated by James Legge, M.A., LL.D.

4to. cloth back, lew. 6d.
A New
ENGLISH.
Monk
Crown
English Dictionary, on Historical Prin-
founded mainly on the materials collected by the Philological Society.

Edited by James A. H. Murray, LL.D., with the assistance of many Scholars
ciples
and men of Science.
An
Part
I.
A ANT.
Part II.
ANT BATTEN.
Im-
i2s. 6d. each.
perial 410.
Etymological Dictionary of the English Language.
By W. W.
Skeat,
M.A.
Second Edition.
1884. 4to.
2/. 4^.
Supplement to the First Edition of the above. 4to. 2s. 6d.

A Concise Etymological Dictionary of the English Language.
GREEK.
By W. W.
Skeat,
M.A.
Second Edition.
Greek-English
Lexicon,
1885.
by
Crown
8vo. 5*. 6d.
Henry
George
and Robert Scott, D.D. Seventh Edition, Revised and Augmented throughout. 1883. 4to. i/. i6s.
Liddell, D.D.,
Greek-English Lexicon, abridged from Liddell and
Scott's 4to. edition, chiefly for the use of Schools.

1884. Square 1 2mo. 7*. 6d.
copious Greek-English
the best authorities.
1850.
24mo.
Twenty-first Edition.
Vocabulary, compiled from
3^.
Practical Introduction to Greek Accenttiation,
W. Chandler, M.A.
[9]
Second Edition.
1881.
8vo. ior. 6d.
by H.
CLARENDON PRESS, OXFORD.

The Book of Hebrezv Roots, by Abu '1-Walid
HEBREW.
ibn Janah, otherwise called Rabbi Yonah. Now

Appendix, by Ad. Neubauer. 1875. 4to. zL^s.dd.
Marwan
A
S.
edited, with an
first
Treatise on the use of the Tenses in Hebrew.
R. Driver, D.D.
Second Edition. 1881. Extra fcap. 8vo.
Hebrew Accentuation of Psalms, Proverbs,

By William Wickes, D.D. 1881. Demy 8vo. stiff covers, $s.
An
ICELANDIC.
By
?s. 6d.
and Job.
Icelandic- English Dictionary, based
on the
collections of the late Richard Cleasby.

Enlarged, and completed
Vigfusson, M.A. With an Introduction, and Life of Richard Cleasby,
MS.
G.
G.
Webbe
Dasent, D.C.L.
1874. 4to.
3/.
by
by
<js.
List of English Words the Etymology of which
illustrated by comparison with Icelandic.

APPENDIX to the above. By W. W. Skeat,
An
Glossary.
An
is
Prepared in the form of an
M.A.
1876. stitched,
Icelandic
rimer with Grammar,
By Henry Sweet, M.A. Extra fcap. 8vo. 3-r. 6d.
^
25.
Notes,
and
Icelandic Prose Reader^ with Notes, Grammar and

York Powell, M.A. 1879.
Glossary, by Dr. Gudbrand Vigfusson and F.

Extra fcap. 8vo. los. 6d.
LATIN.
Latin Dictionary, founded on Andrews' edition
of Freund's Latin Dictionary, revised, enlarged, and in great part rewritten

by Charlton T. Lewis, Ph.D., and Charles Short, LL.D. 1879. 4to. i/. 5*.
The Melanesian
MELANESIAN.
Languages.
Codrington, D.D., of the Melanesian Mission.
SANSKRIT.
By
R.
H.
8vo. i8s.
A Practical Grammar of the
Sanskrit Language*
arranged with reference to the Classical Languages of Europe, for the use of
on ier- Williams, M.A. Fourth Edition. 8vo. 15*.
English Students, by Sir M.
Sanskrit-English Dictionary^
Etymologically
and
Philologically arranged, with special reference to Greek, Latin, German, AngloSaxon, English, and other cognate Indo-European Languages. By Sir M.
Monier-Williams, M.A. 1872. 4to. 4/. l^s. 6d.
Nalopdkhydnam.
Story of Nala, an Episode of the
Maha-Bharata: the Sanskrit text, with a copious Vocabulary, and an improved

version of Dean Milman's Translation, by Sir M. Monier-Williams, M.A.
Second Edition, Revised and Improved. 1879. 8vo. 15^.
Sakuntala.
by
Sir
SYRIAC.
Sanskrit Drama, in Seven Acts.
M. Monier-Williams, M.A.
Edited
Second Edition, 1876. Svo. us.
Thesaurus Syriacus : collegerunt Quatremere, Bern-
Lorsbach, Arnoldi, Agrell, Field, Roediger: edidit R. Payne Smith,

S.T.P. Fasc.I-VI. 1868-83. sm. fol. each, i/. is. Fasc. VII. I/, us. 6d.
Vol. I, containing Fasc. I-V, sm. fol. 5/. 5-r.
stein,
The Book ofKalllah and Dimnah. Translated from Arabic

into Syriac.
Edited by
W.
Wright, LL.D.
1884. 8vo. 2U.

GREEK
Aristophanes:
and Fragments.
Aristotle:
CLASSICS, &c.
Complete Concordance to the Comedies
By Henry Dunbar, M.D.
i/. is.
4to.
The
Politics, translated into English, with IntroMarginal Analysis, Notes, and Indices, by B. Jowett, M.A. Medium
duction,
8VO. 2 Vols. 2IJ.
Graecorum
Codicum
Catalogus
Gardthausen Lipsiensis.
With
six
Sinaiticorttm.
pages of Facsimiles.
Recensuit
Heracliti Ephesii Reliquiae.
I.
V.
Scripsit
8vo. linen, 25^.
Bywater, M.A.
Appendicis loco additae sunt Diogenis Laertii Vita Heracliti, Particulae Hippocratei De Diaeta Libri Primi, Epistolae Heracliteae. 1877. 8vo. 6s.
Herculanensium Voluminum Partes
Fragmenta Herculanensia.
II.
1824. 8vo. los.
Descriptive Catalogue of the
Oxford copies of the Herculanean Rolls, together with the texts of several
Edited by Walter Scott, M.A., Fellow
papyri, accompanied by facsimiles.
of Merton College, Oxford. Royal 8vo. cloth, 2is.
Homer:
Hymns
in the Iliad,
Complete Concordance
Homer
of
to
the Odyssey
and
which is added a Concordance to the Parallel Passages

Odyssey, and Hymns. By Henry Dunbar, M.D. 1880. 4to. I/, is.
;
to
Edited by Professor
Scholia Graeca in Iliadem.
W.
Dindorf, after a new collation of the Venetian MSS. by D. B. Monro, M.A.,

and VI. In the Press.
Provost of Oriel College. 4 vols. 8vo. 2/. IQJ. Vols.
Scholia Graeca in Odysseam.

Tomi II. 1855. 8vo. i$s.(>d.
Plato
Apology, with a revised
Edidit Guil. Dindorfius.
Text and English Notes, and
a Digest of Platonic Idioms, by James Riddell, M.A.
Edward
-
1878. 8vo. 8s. 6d.
Philebus, with a revised Text and English Notes, by

M.A.
Poste,
Sophistes
1860.
8vo.
7-f.
6d.
and Politicus, with a
Notes, by L. Campbell,
M.A.
revised
TheaetetuS) with a revised
Text and English Notes,
by L. Campbell, M.A. Second Edition. 8vo.
The Dialogues, translated

and Introductions, by B. Jowett, M.A.
8vo.
1875.
Text and English
8vo. iSs.
1867.
IGJ. 6d.
into English, with Analyses

A new Edition in 5 volumes, medium
3/. los.
The
Republic, translated into English, with an Analysis

and Introduction, by B. Jowett, M.A. Medium 8vo. 12*. 6d.
Thucydides
Translated
into
English, with Introduction,

By B. Jowett, M.A. 2 vols. 1881.
Marginal Analysis, Notes, and Indices.
Medium
8vo.
\l. 12s.
B 2

THE HOLY SCRIPTURES,
STUDIA BIBLICA.
Essays
in Biblical
By Members
cism, and kindred subjects.
&c.
Archaeology and
Criti-
of the University of Oxford.
8vo.
los. 6d.
The Holy Bible in
ENGLISH.
the earliest
English Versions,
made from
the Latin Vulgate by John Wycliffe and his followers

the Rev. J. Forshall and Sir F. Madden. 4 vols. 1850. Royal 4to.
edited by
3/. 3^.
[Also reprinted from the above, with Introduction and Glossary

by W. W. Skeat, M.A.
Ecclesiastes, and the

made by Nicholas
de Hereford, about A.D. 1381, and Revised by John Purvey, about A.D. 1388.
The Books of Job, Psalms Proverbs,

',
Song of Solomon : according

Extra fcap. 8vo.
to
the Wyclimte Version
3^. 6d.
The New Testament
in English, according to the Version
by John Wycliffe, about A.D. 1380, and Revised by John Purvey, about A.D.
Extra fcap. 8vo.
1388.
6s.]
The Holy Bible : an exact
reprint,
page
for page, of the
Demy 4to. half bound, i/. u.

The Psalter, or Psalms of David^ and certain Canticles,
Authorised Version publish ed
in the year 16 1 1.
with a Translation and Exposition in English, by Richard Rolle of Hampole.

Edited by H. R. Bramley, M.A., Fellow of S. M. Magdalen College, Oxford.
With an Introduction and Glossary. Demy 8vo. iL is.
Lecttires
Abbey by
minster.
GOTHIC.
on
Ecclesiastes.
Delivered
in
the Very Rev. George Granville Bradley, D.D.,

Crown 8vo. ^s. 6d.
The Gospel of
St.
Mark
Westminster
Dean of West-
in Gothic, according to
the translation made by Wulfila in the Fourth Century.

Edited with a
Grammatical Introduction and Glossarial Index by W. W. Skeat, M.A.
Extra fcap. 8vo. 4^.
GREEK.
Vetus Testamentum ex Versione Septuaginta Inter-
pretum secundum exemplar Vaticanum Romae editum. Accedit potior varietas

Codicis Alexandrini. Tomi III. Editio Altera. i8mo. i8j.
Origenis Hexaplorum quae supersunt; sive, Veterum

Interpretum Graecorum in totum Vetus Testamentum Fragmenta. Edidit
2 vols.
Fridericus Field, A.M.
1875. 4 to 5^- 5 S
>
The Book of Wisdom: the Greek Text, the Latin

Vulgate, and the Authorised English Version with an Introduction, Critical
Apparatus, and a Commentary. By William J. Deane, M.A. Small 4to. i2j. 6d.
;
Novum Testamentum
Textus
E.
tO
H. Hansell, S.T.B.
2S.
Graece. Antiquissimorum
Codicum
dispositi. Accedit collatio Codicis Sinaitici. Edidit

Tomi III. 1864. 8vo. half morocco. Price reduced
in ordine parallel

Novum Testamentum
GREEK.
Accedunt
Graece.
Edidit Carolus Lloyd, S.T.P.R.

writing paper, with wide margin, los.
S. Scripturae loca, etc.
On
Novum Testamentum
i8mo.
parallela
3^.
Graece juxta Exemplar Millianum.
On writing paper, with wide margin, 9^.

Evangelia Sacra Graece. Fcap. 8vo. limp, is. 6d.
The Greek Testament, with the Readings adopted by
i8mo.
6d.
is.
the Revisers of the Authorised Version
(1) Pica type, with Marginal References. Demy 8vo. los. 6d.
(2) Long Primer type.
Fcap. 8vo. 45. 6d.
(3)
The same, on
writing paper, with wide margin, 155.
The Parallel New Testament, Greek and English
being
the Authorised Version, 1611

the Revised Version, 1881; and the Greek
in the Revised Version.
8vo. \2s. 6d.
The Revised Version is the joint property of the Universities of Oxford and Cambridge
;
Text followed
Canon Mttratorianus : the

Books of the New Testament. Edited
MS.
in
earliest Catalogue of the

with Notes and a Facsimile of the
the Ambrosian Library at Milan, by S. P. Tregelles, LL.D. 1867.
4to. ioj. 6d.
Outlines of Textual Criticism applied to the New TestaBy C. E. Hammond, M.A. Fourth Edition. Extra fcap. 8vo. 3-r. 6d.
HEBREW, etc. The Psalms in Hebrew without points. 1879.
Crown 8vo. 3^. 6d.
ment.
to
S.
Commentary on
Book of Proverbs.
the
Abraham Ibn Ezra. Edited from a MS. in

R. Driver, M.A. Crown 8vo. paper covers, %s.
The Book of
MS.
Tobit.
Attributed
the Bodleian Library by

6d.
Chaldee Text, from a unique
Bodleian Library with other Rabbinical Texts, English Translations, and the Itala. Edited by Ad. Neubauer, M.A. 1878. Crown 8vo. 6s.
in the
Horae Hebraicae
Edition, by R. Gandell,
LATIN.
Talmudicae^ a
4
vols.
J.
1859. 8vo.
Lightfoot.
I/,
Edidit B. Thorpe, F.A.S.
Old- Latin Biblical Texts : No.
7.
A new
u.
Libri Psalmorum Versio antiqua Latina,
phrasi Anglo-Saxonica.
-
et
M.A.
1835.
cum
Para-
8vo. los. 6d.
The Gospel
according
Matthew from the St. Germain MS. (g,). Edited with Introduction
and Appendices by John Wordsworth, D.D. Small 410., stiff covers, 6s.
- Old-Latin Biblical Texts: No. 77. Portions of the
Gospels
according to St. Mark and St. Matthew, from the Bobbio MS. (k), &c.
Edited by John Wordsworth, D.D., W. Sanday, M. A., D.D., and H. J. White,
M.A. Small 4to., stiff covers, 2is.
to St.
OLD-FRENCH.
Cod. MS.
Monumentis
Phil.
Doc.
Libri Psalmorum Versio antiqua Gallica e
in Bibl.
cum Versione Metrica aiiisque

descripsit et edidit Franciscus Michel,
Bodleiana adservato, una
pervetustis.
Nunc primum
1860. 8vo. ioj. 6d.
CLARENDON
PRESS, OXFORD.
FATHERS OP THE CHURCH,
&c.
St.Athanasius: Historical Writings, according to the BeneWith an Introduction by William
dictine Text.
8vo. los. 6d.
St. Augustine
D.D.
Crown
1873.
D.D.
Crown
1881.
With an Account of
Orations against the Arians.

Life by William Bright,
Bright,
Select Anti-Pelagian Treatises, and the

With an Introduction by William
of the Second Council of Orange.
Crown
D.D.
his
8vo. 9^.
Acts
Bright,
8vo. 9*.
Canons of the First Four General Councils of Nicaea, Constantinople, Ephesus,
and Chalcedon. 1877. Crown
8 vo. 2s.6d.
Notes on the Canons of the First Four General Councils.

By William
Bright,
D.D.
Crown
1882.
8vo. 5*. 6d.
Cyrilli Archiepiscopi Alexandrini in

P. E. Pusey,
in
A.M.
Tomi
II.
XII
1868. 8vo. cloth,
D. Joannis Evangelium.
Edidit
Prophetas.
2/. 2s.
Accedunt Fragmenta varia
necnon Tractatus ad Tiberium Diaconum duo.

P. E. Pusey, A.M. Tomi III.
1872. 8vo. 2/. 5^.
Edidit post
Aubertum
Commentarii in Lucae Evangelium quae supersunt

E MSS. apud Mus. Britan. edidit R. Payne Smith, A.M. 1858.
Syriace.
4to. i/.
2J-.
Translated by R. Payne Smith, M.A.
2 vols.
1859.
8vo. 14^.
Ephraemi
Syri, Rabulae Episcopi Edesseni, Balaei, aliorum-
que Opera Selecta. E Codd. Syriacis MSS. in Museo Britannico

Bodleiana asservatis primus edidit J. J. Overbeck. 1865. 8vo.
Eusebius
Ecclesiastical History
',
et Bibliotheca
I/, is.
according to the text of
Burton, with an Introduction by William Bright, D.D.
1881.
Crown
8vo.
8^. 6d.
Irenaeus
The Third Book of
St. Irenaeus, Bishop of Lyons,

H. Deane, B.D.
against Heresies. With short Notes and a Glossary by

1874. Crown 8vo. $s. 6d.
Patrum
Apostolicorum,
S. Polycarpi,
S.
quae supersunt.
Fourth Edition, 1863. 8vo.
I/,
Clementis
Romani,
-,
II.
according to the Text of
Hussey, with an Introduction by William Bright, D.D.

6d.
Tomi
u.
Socrates' Ecclesiastical History

7-r.
S. Ignatii,
Edidit Guil. Jacobson, S.T.P.R.
1878.
Crown
8vo.

ECCLESIASTICAL HISTORY, BIOGRAPHY,
&c.
Ancient Liturgy of the Church of England, according to the

uses of Sarum, York, Hereford, and Bangor, and the Roman Liturgy arranged
in parallel
columns, with preface and notes.

1882. 8vo. 15^.
By William Maskell, M.A.
Third Edition.
Baedae Historia
Ecclesiastica.
Edited, with
by G. H. Moberly, M.A. 1881. Crown 8vo. los. 6d.
Bright
Chapters of Early English Church History.
W.).
English Notes,
1878. 8vo. I2s.
Burnefs History of the Reformation of the Church of England.
A new Edition.
Carefully revised, and the Records collated with the originals,

7 vols. 1865. 8vo. Price reduced to il. los.
by N. Pocock, M.A.
Councils
and
and
and Ecclesiastical Documents
relating to Great Britain
Edited, after Spelman and Wilkins, by A. W. Haddan, B.D.,

Stubbs, M.A. Vols. I. and III. 1869-71. Medium 8vo. each il. is.
Ireland.
W.
Vol. II. Part
I.
Vol.11. Part
II.
1873.
Medium
1878.
Stiff covers, 3^. 6d.
8vo. los. 6d.
Church of Ireland; Memorials of
Hamilton (John, Archbishop of
St.
Andrews), The Catechism
Edited, with Introduction and Glossary, by Thomas Graves

a Preface by the Right Hon. W. E. Gladstone. 8vo. i2s. 6d.
of.
Hammond (C.
E.).
St. Patrick.
Liturgies, Eastern
Law.
With
and Western. Edited,
with Introduction, Notes, and Liturgical Glossary. 1878. Crown 8vo. IQJ. 6d.
An Appendix to the above. 1879. CVown 8vo. paper covers, is. 6d.
John, Bishop of Ephesus.

siastical History.
[In Syriac.]
1853. 4tO. I/. I2S.
The Third Part of his EccleNow first edited by William Cureton, M.A.
Translated by R. Payne Smith,
M.A.
1860. 8vo. los.
Leofric Missal, The, as used in the Cathedral of Exeter

during the Episcopate of its first Bishop, A.D. 1050-1072 ; together with some
Account of the Red Book of Derby, the Missal of Robert of Jumieges, and a
few other early MS. Service Books of the English Church. Edited, with Introduction and Notes, by F. E. Warren, B.D. 4to. half morocco, 35^.
Monumenta Ritualia
Ecclesiae Anglicanae.
The
occasional
Offices of the Church of England according to the old use of Salisbury, the
Prymer in English, and other prayers and forms, with dissertations and notes.
By William
Maskell,
M.A.
Second Edition.
Records of the Reformation.
The
1882. 3 vols. 8vo.
2/. loj.
Divorce, 1527-1533. Mostly

now for the first time printed from MSS. in the British Museum and other libraries. Collected and arranged by N. Pocock, M.A. 1870. 2 vols. 8vo. il. i6s.
Shirley
(W.
Stubbs
Some Account of the Church
W.).
Second Edition, 1874. Fcap. 8vo.
Age.
(
in the Apostolic
35. 6d.
An
Registrum Sacrum Anglican-urn.
W.).
Warren
attempt
1858. Small 4to.
to exhibit the course of Episcopal Succession in England.

8*. 6d.
Liturgy and Ritual of the Celtic Church.
(F. E.).
1881. 8vo. 14*.
ENGLISH THEOLOGY.
Butler's Works, with
8vo
an Index to the Analogy.
IIS '
Sermons
$s. 6d.
$s. 6d.
Analogy of Religion,
Greswell s Harmonia Evangelica.

9-y.
2 vols. 1874.
Also separately,
Fifth Edition.
8vo.
1855.
6d.
Heurtlefs Harmonia Symbolical

Church.
Homilies appointed
J. Griffiths,
to
be
1859. 8vo.
M.A.
read in
7*.
6d.
Hooker s Works, with

Keble, M.A.
Creeds
the Western
of
8vo. 6s. 6d.
1858.
Edited
Churches.
by
his life
Walton, arranged
vo. i/. us. 6d.
Sixth Edition, 1874. 3 vols
by
by John
M.A.
the text as arranged by John Keble,
2 vols.
1875. 8vo. us.
8vo.
W.
Edited by R.
Jewel's Works.
Jelf,
D.D.
8 vols.
1848.
i/. los.
Revised and corrected by
Pearson* s Exposition of the Creed.
E. Burton, D.D. Sixth Edition, 1877. 8vo. los. 6d.
Waterland's Review of the Doctrine of the Eucharist, with

a Preface by the
late
Bishop of London.
Works, with Life, by

with copious Indexes. 6 vols.
Bp.
Crown
Van
1856.
Mildert.
8vo.
Wheatly's Illustration of the Book of

Edition, 1846. 8vo.
Wyclif.
by W.
A
W.
2!.
D.D.
Select English
1869-1871. 8vo.
A new Edition,
us.
Prayer.
A new
Works of John
Wyclif,
Common
5-r.
Catalogue of the Original

Shirley,
8vo. 6s. 6d.
1865. 8vo. 3*. 6d.
Works.
By
T. Arnold, M.A. 3 vols.
i/. is.
Trialogus. With the

By Gotthard Lechler. 1869. 8vo.
Supplement now
fs.
first
edited.
CLARENDON PRESS, OXFORD

HISTORICAL AND DOCUMENTARY WORKS.
British Barrows, a Record of the Examination of Sepulchral
Mounds in various parts of England. By William Greenwell, M.A., F.S.A.
Together with Description of Figures of Skulls, General Remarks on Prehistoric Crania, and an Appendix by George Rolleston, M.D., F.R.S. 1877.
Medium
8vo. 25^.
Britton.
Treatise upon the
composed by order of King Edward
Common Law of England,

The French Text
carefully revised,
with an English Translation, Introduction, and Notes, by F.

2 vols.
1865. Royal 8vo. i/. i6s.
M. Nichols, M.A.
Rebellion
Clarendon's History of the

England.
1839. i8mo.
7 vols.
I.
and
Civil
Wars
in
I/, is.
and Civil Wars in

by himself, in which is included a ConHistory of the Grand Rebellion. With copious Indexes.
Clarendons History of the Rebellion

England.
Also his
tinuation of his
Life, written
In one volume, royal 8vo. 1842.
i/. is.
Clintons Epitome of the Fasti Hellenici. 1851. 8vo. 6s. 6d.

- Epitome of the Fasti Romani. 1854. 8vo. Js.
The Poetry
Corpvs Poeticvm Boreale.
Old Northern
of the
Tongue, from the Earliest Times to the Thirteenth Century. Edited, classified, and translated, with Introduction, Excursus, and Notes, by Gudbrand
Vigfusson, M.A., and F. York Powell, M.A. 2 vols. 1883. 8vo. 42 s.
Freeman (E.
land;
-
its
History of the
A.).
Causes and Results.
Norman
Conquest of
In Six Volumes. 8vo.
The Reign of William Rufus and
Henry the
Gascoigne's
First.
5/. $s.
Eng-
6d.
the Accession
of
2 vols. 8vo. i/. i6s.
Theological
Dictionary
Veritatum ")
("Liber
Selected Passages, illustrating the condition of Church and State, 1403-1458.

With an Introduction by James E. Thorold Rogers, M.A. Small 4to. los. 6d.
Magna
Carta, a careful Reprint.
1879. 4to. stitched,
Edited by
W.
Stubbs, D.D.
is.
Passio et Miracula Beati Olaui.
Edited from a Twelfth-
Century MS. in the Library of Corpus Christi College, Oxford, with an Introduction and Notes, by Frederick Metcalfe,
M.A. Small
4to.
stiff
covers, 6s.
Protests of the Lords, including those which have been expunged, from 1624101874; with Historical Introductions. Edited by James
E. Thorold Rogers, M.A. 1875. 3 vols. 8vo. 2/. is.
T.).
History of Agriculture and
England, A.D. 1259-1793.
Vols. I and II (1259-1400). 1866. 8vo. 2/. 2s.
Vols. Ill and IV (1401-1582). 1882. 8vo. 2/. ior.
Rogers (J. E.
Prices in
io
Saxon Chronicles (Two of
the] parallel,
with Supplementary
Extracts from the Others.
Edited, with Introduction, Notes, and a Glossarial Index, by J. Earle, M.A.

1865. 8vo. i6s.
Sturlunga Saga, including the Islendinga Saga of

Sturla Thordsson and other works.
In 2 vols. 1878. 8vo. 2/. 2j.
York Plays.
Lawman
Edited by Dr. Gudbrand Vigfusson.
The Plays performed by the
Crafts or Mysteries
York on the day of Corpus
Christi in the I4th, I5th, and i6th centuries.

first printed from the unique MS. in the
Library of Lord Ashburnham.
Edited with Introduction and Glossary by Lucy Toulmin Smith. 8vo. 2is.
of
Now
Statutes
made for the
and Halls
us. 6d.
therein,
University of Oxford, and for the Colleges

by the University of Oxford Commissioners. 1882. 8vo.
Statuta Universitatis Oxoniensis.
8vo. $s.
1885.
The Examination Statutes for

B.C.L., and B.M.
Revised to
The Students Handbook

Oxford.
Extra fcap. 8vo.
2s.
the Degrees of B.A., B. Mus.,

Trinity Term, 1885. 8vo. sewed, is.
to
the University
The Oxford University Calendar for

The
and
Colleges of
6d.
the
year 1886.
Crown
8vo. 4*. 6d.

present Edition includes all Class Lists and other University distinctions for
the five years ending with 1885.
Also, supplementary to the above, price 5s. (pp. 606),
The Honours Register of the University of Oxford.

'
A complete
Record of University Honours, Officers, Distinctions, and Class Lists

Heads of Colleges, &c., &c., from the Thirteenth Century to 1883.
MATHEMATICS, PHYSICAL SCIENCE,

Acland(H. W., M.D., F.R.S.\
Series in the
Oxford Museum.
1867.
of the
&c.
Synopsis of the Pathological

8vo. 2s. 6d.
De Bary
(Dr. A.). Comparative Anatomy of the Vegetative

Organs of the Phanerogams and Ferns. Translated and Annotated by F. O.
Bower, M.A., F.L.S., and D. H. Scott, M.A., Ph.D., F.L.S. With two
hundred and forty-one woodcuts and an Index. Royal 8vo., half morocco,
i/. 2s.
6d.
Miiller (y.).
On
certain Variations in the Vocal
Organs of
the Passeres that have hitherto escaped notice. Translated by F. J. Bell, B.A.,
and edited, with an Appendix, by A. H. Garrod, M.A., F.R.S. With Plates.
1878. 4to. paper covers, ^s. 6d.
Treatise on Infinitesimal
Price (Bartholomew, M.A., F.R.S.).

Calculus.
Vol.
I. Differential
Calculus.
Second Edition.
8vo. 14*. 6d.
and Differential Equations.

Integral Calculus, Calculus of Variations,
Second Edition, 1865. 8vo. iSs.
Vol. III. Statics, including Attractions ; Dynamics of a Material Particle.
Second Edition, 1868. 8vo. i6s.
Vol.
II.
Vol. IV. Dynamics of Material Systems; together with a chapter on Theo1862. 8vo. 16*.
retical Dynamics, by W. F. Donkin, M.A., F.R.S.
Pritchard
(,
D.D., F.R.S.}.
Uranometria Nova Oxoniensis.
Photometric determination of the magnitudes of all Stars visible to the

naked eye, from the Pole to ten degrees south of the Equator. 1885. Royal
Svo.
Ss. 6d.
Astronomical Observations
made
at
the
University
Observatory, Oxford, under the direction of C. Pritchard, D.D.

1878. Royal Svo. paper covers,^. 6</.
Rigaud's Correspondency of Scientific
Men
No.
I.
of the ^^th Century,
with Table of Contents by A. de Morgan, and Index by the Rev.

M.A. 2 vols. 1841-1862. 8vo. i8s. 6d.
J.
Rigaud,
Scientific Papers and AdWith a

Arranged and Edited by William Turner, M.B., F.R.S.
Biographical Sketch by Edward Tylor, F.R.S. With Portrait, Plates, and
Woodcuts. 2 vols. Svo. I/. 4^.
Rolleston (George, M.D., F.R.S.).

dresses.
Thesaurus Entomologicus
O., M.A., F.R.S.).
Hopeianus, or a Description of the rarest Insects in the Collection given to
the University by the Rev. William Hope. With 40 Plates. 1874. Small
folio, half morocco, 7/. icw.
Westwood (J.
feacrrtr afoofcs
of
tfie
lEast.
TRANSLATED BY VARIOUS ORIENTAL SCHOLARS, AND EDITED BY

F. MAX MULLER.
[Demy
Svo. cloth.]
Translated by F. Max Miiller.

The A7zandogya-upanishad, The Talavakara-upanishad, The Aitareyaarawyaka, The Kaushitaki-brahmawa-upanishad, and The Vaâsaneyi-sawzhita-
Vol.
The Upanishads.
I.
Part
I.
upanishad.
Vol. II.
los. 6d.
The Sacred Laws
of the Aryas, as taught in the
Schools of Apastamba, Gautama, VasishMa, and Baudhslyana. Translated by

Prof. Georg Biihler. Part I. Apastamba and Gautama,
los. 6d.
12
The Sacred Books
Vol. III.
The Texts
of China.
Translated by James Legge. Part I. The Shu King,

gious portions of the Shih King, and The Hsiao King. 1 2s. 6d.
fucianism.
The Zend-Avesta.
Vol. IV.
Parti.
steter.
The
Vol. V.
Part
The
I.
los. 6<t.
Pahlavi Texts.
Translated by E.
Bahman Yajt, and Shayast la-shayast.
Bundahi-y,
by E. H. Palmer.
The
Vol. VII.
Translated by James Darme-
TheVendidad.
VI and IX. The
Vols.
of ConThe Reli-
Qur'an.
Parts I and II.
W.
1 2s.
West.
6d.
Translated
2is.
Institutes of Vishnu.
Translated by Julius
los. 6d.
Jolly.
Vol. VIII. The Bhagavadgita, with The Sanatsuâtiya, and

The Anugita. Translated by Kashinath Trimbak Telang, IQS. 6d.
The Dhammapada,
Vol. X.
translated from Pali
by F.
and The Sutta-Nipata, translated from Pali by V. Fausboll

Canonical Books of the Buddhists. los. 6d.
Miiller ;
Max
being
Translated from Pali by T. W.

The Mahaparinibbana Suttanta 2. The Dhamma-âkka3. The Teviggu Suttanta;
4. The Akankheyya Sutta;
ppavattana Sutta
5. The ^Tetokhila Sutta; 6. The Maha-sudassana Suttanta
7. The Sabbasava
Vol. XI. Buddhist Suttas.

Rhys Davids.
I.
ioj. 6d.
Sutta.
Vol. XII.
The 5atapatha-Brahmaâ, according
of the Madhyandina School.
Books
and
II.
12.?,
Text
Part
I.
6d.
Vol. XIII. Vinaya Texts.

Translated
T. W. Rhys Davids and Hermann Oldenberg.
The Mahavagga, I-IV. los. 6d.
Vol.
to the
Translated by Julius Eggeling.
XIV. The Sacred Laws
from
Part
I.
the Pali
by
The Patimokkha.
of the Aryas, as taught in the
Schools of Apastamba, Gautama, VasishMa and Baudhayana. Translated

by Georg Biihler. Part II. Vasish/â and Baudhayana. los, 6d.
The Upanishads. Translated by F. Max Miiller.

The Kaâ-upanishad, The Mu/zo'aka-upanishad, The Taittiriyakaupanishad, The BrzTiadarawyaka-upanishad, The ^"vetaj-vatara-upanishad, The
los, 6d.
PrajTza-upanishad, and The Maitrayawa-Brahmawa-upanishad.
Vol.
XV.
Part
Vol.
II.
XVI. The Sacred Books
fucianism.
Vol.
XVII. Vinaya Texts.
T.
W. Rhys
V-X.
The Texts
of China.
Translated by James Legge.
Part
II.
The Yi King.
of ConIQJ. 6d.
Translated from the Pali by
Davids and Hermann Oldenberg.

The Aullavagga, I-III. los. 6d
Part
II.
The Mahavagga,
13
Translated by E. W. West.
Vol. XVIII. Pahlavi Texts.
Part II. The Da^/istan-i Dinik and The Epistles of Manujihar. 1 2s. 6d.
XIX. The
Vol.
Fo-sho-hing-tsan-king.
Buddha
Life of
by A-rvaghosha Bodhisattva, translated from Sanskrit into Chinese by Dharmaraksha, A.D. 420, and from Chinese into English by Samuel Beal. IQS. 6d.
Vol.
XX. Vinaya
Translated from the Pali by T.
Texts.
Rhys Davids and Hermann Oldenberg.
Part III.
W.
The Aullavagga, IV-XII.
IQS. 6d.
Vol.
XXI.
The Saddharma-puwârika
Translated by H. Kern.
True Law.
the Lotus of the
or,
12s. 6d.
aina-Sutras. Translated from Prakrit

Vol. XXII.
mann Jacobi. Part I. The AMranga-Sutra. The Kalpa-Sutra.
Vol.
The Zend-Avesta.
XXIII.
mesteter.
Part II.
XXIV.
Vol.
Part III.
The
by James Dar-
IQJ. 6d.
Translated by E.
Pahlavi Texts.
los. 6d.
W.
Dina-i Mainog-i Khirad, ^Sikand-gumanik, and Sad-Dar.
Second
XXVI.
Vol.
Translated
and Nyayi^.
Sirozahs, Yarts,
by Her-
Part II.
IQJ. 6d.
Series.
The Satapatha-Brahma/za.
Julius Eggeling.
West.
Translated by
Just Published.
i2S.6d.
Vols. XXVII and XXVIII. The Sacred Books of China.

The Texts of Confucianism. Translated by James Legge. Parts III and IV.
The Li K\, or Collection of Treatises on the Rules of Propriety, or" Ceremonial
Usages.
2$s.
Just Published.
The following Volumes
Vol.
XXV.
Vols.
Manu.
XXIX
and
XXX. The
Domestic Ceremonies.
Vol.
XXXI.
Hermann Oldenberg.
The Zend-Avesta.
Part III.
Translated by the Rev. L.
XXXII. Vedic Hymns.
Vol.
I.
Grzhya-Sutras, Rules of Vedic
Translated by
Visparad, Afrinagan, and Gahs.
Vol.
are in the Press
Translated by Georg Buhler.
Translated by F.
Parts I and II.
The Yasna,
H.
Mills.
Max
Mullen
Parti.
Vol.
XXXIII.
Narada,
Translated by Julius Jolly.
Vol.
XXXIV. The
mentary.
and
some Minor
Law-books.
[Preparing.'}
Vedanta-Sutras, with Sankara's
Translated by G. Thibaut.
\_PreparingI\
* * The Second Series will consist

#
of Twenty-Four Volumes.
Com-
mes
s
I.
First Reading Book.

edited
by Anne
J.
Clough.
ENGLISH,
By Marie
Extra fcap. 8vo.
Oxford Reading Book, Part

fcap. 8vo.
stiff
An
stiff
covers,
An
W.
Tancock, M.A.
English
3-r.
For Junior
II.
ductory Notices.
Grammar and
Second Edition.
By O. W. Tancock, M.A.
is.
I.
Latimer to Berkeley.
Vol.
C., LL.D.).
Aspects of
delivered at Oxford. Crown 8vo. los. 6d.
Book for
M.A.
Lower Forms
for
Fourth Edition.
II.
Pope
Extra
Poetry ; being Lectures
the Beginner in Anglo-Saxon.
Third Edition.
Extra fcap. 8vo.
Anglo-Saxon Reader.
to Macaulay.
By John
Earle,
2s. 6d.
In Prose and Verse.
With Gram-
matical Introduction, Notes, and Glossary. By Henry Sweet, M.A.

Extra fcap. 8vo. 8s. 6d.
Edition, Revised and Enlarged.
An
By
6d.
from the best English Writers, with IntroSecond Edition. In Two Volumes.
Extra fcap. 8vo.
Shairp (J.
An
Exercise Book*
Extra fcap. 8vo.
6d. each.
Vol.
Extra
Classes.
6d.
Typical Selections
3-r.
Extra
Little Children.
Grammar and Reading Book,
in Classical Schools.
leap. 8vo.
and
covers, \d.
60?.
Elementary English
O.
For
I.
stiff
covers, 6d.
Oxford Reading Book, Part

fcap. 8vo.
&c.
Eichens of Berlin
Fourth
Anglo-Saxon Primer, with Grammar, Notes, and Glossary.

By
the same Author.
Second Edition.
Extra fcap. 8vo.
2s.
6d.
Old English Reading Primers ; edited by Henry Sweet, M.A.

Extra fcap. 8vo., stiff covers, is. 6d.
Extra fcap. 8vo., stiff covers, is. 6d.
I.
Selected Homilies of JElfric.
II.
Extracts from Alfred's Orosius.
First Middle English Primer, with Grammar and Glossary.

By the same Author. Extra fcap. 8vo. 2s.
The Philology of the English Tongtie. By J. Earle, M.A.
Third Edition.
An
Extra fcap. 8vo.
7.?.
6d.
Icelandic Primer, with Grammar, Notes,

By the same Author. Extra fcap. 8vo. %s. 6d.
and Glossary.
An Icelandic Prose Reader, with Notes, Grammar, and Glossary.

By G.
Vigfiisson, M.A.,
and F. York Powell, M.A.
Ext. fcap. 8vo. los. 6d.
Handbook of Phonetics, including a Popular Exposition

the Principles of Spelling Reform.
By H. Sweet, M.A. Extra fcap.
of
8vo. 4*. 6d.

Elementarbuch
des Gesprochenen Englisch.

Von Henry Sweet. Extra fcap. 8vo.,
Texte und Glossar.
15
Gram mat ik,

stiff
covers, 2s. 6d.
The Ormulum; with the Notes and Glossary of Dr. R. M.

Edited by R. Holt, M.A. 1878.
White.
Specimens of Early English.
2 vols.
New
Extra fcap. 8vo. 21 s.
and Revised Edition.
Introduction, Notes, and Glossarial Index. By R. Morris, LL.D., and

Skeat, M.A.
Part I. From Old English Homilies to King Horn (A. D. 1150 to A.D. 1300).
Second Edition. Extra fcap. 8vo. 9^.
With
W. W.
From Robert of Gloucester to Gower (A.D. 1298 to A.D. 1393).

Second Edition. Extra fcap. 8vo. 7-r. 6d.
Part II.
Specimens of English Literature, from
'
the
Ploughmans
'
Crede' to the Shepheardes Calender' (A.D. 1394 to A.D. 1579). Witl1 Introduction, Notes, and Glossarial Index. By W. W. Skeat, M.A. Extra fcap.
8vo.
7-r.
6d.
The Vision of William concerning Piers

William Langland.
Edited, with Notes, by
Edition. Extra fcap. 8vo. 4^. 6d.
the
W. W.
Plowman, by
Skeat,
M.A.
Third
I. The Prologue
to the Canterbury Tales;
the
The Nonne Prestes Tale. Edited by R. Morris, Editor of
Knightes Tale
Specimens of Early English, &c., &c. Fifty-first Thousand. Extra fcap. 8vo.
Chaiicer.
2s.
6d.
II.
Tale ;
M.A.
The Prioresses Tale ; Sir Thopas ; The Monkes

The
The Squieres Tale, &c.

Clerkes Tale
Second Edition. Extra fcap. 8vo. 4^. 6d>
;
Edited by
W. W.
Skeat,
Man
III. The Tale of the

of Laive ; The Pardoneres
Tale The Second Nonnes Tale The Chanouns Yemannes Tale. By the
same Editor. Second Edition. Extra fcap. 8vo. 4^. 6d.
;
Gamelyn, The Tale

W. W. Skeat, M.A.
of.
Edited with Notes, Glossary, &c., by
Extra fcap. 8vo.
for the use of Schools.
Kitchin,
Stiff covers, is. 6d.
Books
Spenser s Faery Queene.

With
and
II.
Designed chiefly
By G. W.
Introduction, Notes, and Glossary.
D.D.
Book
Book
I.
II.
Tenth Edition.
Sixth Edition.
Extra fcap. 8vo.

Extra fcap. 8vo.
2s.
6d.
2s.
6d.
Ecclesiastical Polity, Book f.

Edited
Church, M.A. Second Edition. Extra fcap. 8vo. 2s.
Hooker.
Marlowe and
Greene.
by R. W.
Marlowe^s Tragical History of Dr.
and Greene's Honourable History of Friar Bacon and Friar Bungay.

Edited by A. W. Ward, M.A. 1878. Extra fcap. 8vo. 5*. 6d. In white
Parchment, 6s.
Fattstus,
Marlowe.
O.
Edward
W. Tancock, M.A.
II.
With
Introduction, Notes, &c.
Extra fcap. 8vo.
3*.
By
and
W.
Aldis Wright, M.A.
The Merchant
Extra fcap. 8vo.
u.
of Venice,
Richard the Second,
W.
Edited by
The Tempest,
As You Like It.
Julius Caesar.
King Lear.
is.6d.
2s.
Aldis Wright, M.A.
Midsummer
is. 6d.
Night's
Coriolanus.
is. 6d.
Henry the
2s.
Richard the Third.
G. Clark, M.A.,
covers.
stiff
Macbeth.
Hamlet.
6d.
is.
W.
Edited by
Select Plays.
Shakespeare.
Fifth,
is. 6d.
zs.
Twelfth Night,
2s. 6d.
Dream,
2J. 6d.
is.
6d.
6d.
King John. is. 6d.

Shakespeare as a Dramatic Artist ; a popular Illustration of
the Principles of Scientific Criticism. By R. G. Moulton, M.A. Crown 8vo. 5^.
Bacon.
is.
Advancement of Learning.
I.
Wright, M.A.
II.
S.
Second Edition.
Milton.
I.
late
II.
Fellow of Brasenose College.
Areopagitica.
John W. Hales, M.A.
By
In Preparation.
Extra fcap. 8vo.
6.?.
Samson
2 vols.
Sold separately, Vol. 1.4*.; Vol.
6d.
In paper covers
L'Allegro, id.
By
$s.
Edited by R. C. Browne, M.A.
Poems.
Aldis
With Introduction and Notes.
Third Edition.
Fifth Edition. Extra fcap. 8vo.
Lycidas, 3^.
W.
4^. 6d.
The Essays.
H. Reynolds, M.A.,
Edited by
Extra fcap. 8vo.
II
Comus,
Penseroso, $d.
6d.
Agonistes, 6d.
III. Samson Agonistes.

Edited with Introduction
Notes by John Churton Collins. Extra fcap. 8vo. stiff covers, is.
Bunyan.
I.
II. 3*.
and
The Pilgrim's Progress^ Grace Abounding, Rela-
Imprisonment of Mr. John Bunyan. Edited, with Biographical

Introduction and Notes, by E. Venables, M.A. 1879. Extra fcap. 8vo.
tion of the
"
In ornamental Parchment,
t,6s.
II. Holy
In the Press.
Clarendon.
War,
frc.
Edited
by E. Venables, M.A.
Book VI.
History of the Rebellion.
by T. Arnold, M.A.
Extra fcap. 8vo.
Edited
4?. 6d.
Select Poems.
Stanzas on the Death of Oliver
Cromwell Astrsea Redux Annus Mirabilis Absalom and Achitophel
Religio Laici The Hind and the Panther. Edited by W. D. Christie, M.A.
Second Edition. Extra fcap. 8vo. 3-r. 6d.
Dryden.
Locke's Conduct of the Understanding.
Edited, with Intro-
duction, Notes, &c., by T. Fowler,
Addison.
Notes.
Selections
By T.
Parchment,
6s.
M.A. Second Edition. Extra fcap.
from Papers
Arnold, M.A.
in
the
Extra fcap. 8vo.
Spectator.
4*.
6d.
8vo.
2s.
With
In ornamental

Selections
Steele.
the Tatler^ Spectator;
from
Edited by Austin Dobson. Extra fcap. 8vo.
4^. 6d.
17
and Guardian.
In white Parchment,
*js.
6</.
With Introduction and Notes. By Mark Pattison, B.D.

Essay on Man. Extra fcap. 8vo. is. 6d.
Extra fcap. 8vo. 2s.
II. Satires and Epistles.
Pope.
I.
The Hermit.
Parnell.
Johnson.
I.
Paper covers, id.
Rasselas ; Lives of
Dryden and
by Alfred Milnes, M.A. (London). Extra fcap. 8vo.

Dryden and Pope only, stiff covers, 2s. 6d.
Human
II. Vanity of
Payne, M.A. Paper covers, 4^.
Selected Poems.
Gray.
fcap. 8vo.
or Lives of
With Notes, by E.
Wishes.
Edited by
Stiff covers, I s. 6d.
Edited
Pope.
^s. 6d.,
Edmund
In white Parchment,
Gosse.
J.
Extra
%s.
Elegy and Ode on Eton College. Paper covers, %d.

The Deserted Village. Paper covers, id.
Goldsmith.
Edited, with Life, Introductions, and Notes,
Cowper.
H. T.
-
I.
Minor
Griffith,
The Didactic Poems of 1782, with Selections from the

Pieces, A.D. 1779-1783.
Extra fcap. 8vo.
3^.
The Task) with Tirocinium, and Selections from the
II.
Minor Poems, A.D. 1784-1799.
Select Works.
by E. J. Payne, M.A.
Burke.
I.
by
B.A.
Second Edition.
Extra fcap. 8vo.
3^.
Edited, with Introduction and Notes,
Thoughts on the Present Discontents ; the two Speeches
on America.
Second Edition.
II. Reflections
Extra fcap. 8vo. <-,s.
III.
Four Letters on
Regicide Directory of France.
Keats. Hyperion,
Extra fcap. 8vo.
^s. 6d.
on the French Revolution. Second Edition.
Book
the Proposals for Peace with the

Second Edition. Extra fcap. 8vo. 5*.
With Notes by W. T. Arnold, B.A.
I.
Paper covers, 4 d.
Byron.
Childe Harold.
by H. F. Tozer, M.A. Extra
fcap. 8vo.
3^. 6d.
Scott.
Lay of the Last Minstrel.
Notes by W. Minto, M.A. With Map.
Ornamental Parchment,
Lay of
the
$s.
5.?.
Extra fcap. 8vo.
Stiff covers,
2s.
6d.
Last Minstrel.
Introduction and Canto
with Preface and Notes, by the same Editor.
[9]
In white Parchment,
Edited with Preface and
6d.
I.,
LATIN.
II.
Rudimenta Latina.
Comprising Accidence, and Exercises of
a very Elementary Character, for the use of Beginners.

Allen, M.A. Extra fcap. 8vo. 2s.
An
Elementary Latin Grammar.
First Latin Exercise Book.

Extra fcap. 8vo.
Edition.
By
31$-.
2s.
the
same Author.
6d.
the same Author.
Fourth
2s. 6d.
Second Latin Exercise Book.

fcap. 8vo.
By
Extra fcap. 8vo.
Forty-second Thousand.
By John Barrow
By
the same Author.
Extra
6d.
Reddenda Minora, or Easy Passages, Latin and Greek,

Unseen Translation. For the use of Lower Forms.
by C. S. Jerram, M.A. Extra fcap. 8vo. is. 6d.
Composed and
for
selected
Anglice Reddenda, or Easy Extracts, Latin and Greek, for

Unseen Translation. By C.
Enlarged. Extra fcap. 8vo.
S. Jerram,
M.A.
For the use of Passmen
Passages for Translation into Latin.

and
others.
8vo. 2s. 6d.
Selected
by
J.
Third Edition, Revised and
2s. 6d.
Y. Sargent, M.A.
Fifth Edition.
Extra fcap.
Exercises in Latin Prose Composition
with Introduction,
;
Notes, and Passages of Graduated Difficulty for Translation into Latin. By
G. G. Ramsay, M.A., LL.D. Second Edition. Extra fcap. 8vo. 4^. 6d.
Hints and Helps for Latin Elegiacs.
By H. Lee- Warner, M.A.,
Fellow of St. John's College, Cambridge, Assistant Master

School. Extra fcap. 8vo. 3^. 6d.
late
First Latin Reader.

Extra fcap. 8vo.
By
T.
By Charles
Maps.
Nunns, M.A.
Rugby
Third Edition.
2s.
The Commentaries
Caesar.
J.
at
E. Moberly,
(for Schools).
With Notes. and
M.A.
The Gallic War. Second Edition. Extra fcap. 8vo. 4*. 6d.
The Civil War. Extra fcap. 8vo. 35. 6d.
The Civil War. Book I. Second Edition. Extra fcap. 8vo. 2s.
Part
I.
Part
II.
Cicero.
Selection of interesting and descriptive passages. With

By Henry Walford, M.A. In three Parts. Extra fcap. 8vo. 4*. 6d.
Each Part separately, limp, is. 6d.
Notes.
Part
Roman History. Third Edition.

Beauties of Nature. Third Edition.
Anecdotes from Grecian and
I.
Part II.
Omens and Dreams
Part III. Rome's Rule of her Provinces.
Cicero.
Third Edition.
Selected Letters (for Schools).
late C. E. Prichard,
Extra fcap. 8vo.
3-r.
With Notes.
M.A., and E. R. Bernard, .M.A.
By
the
Second Edition.
19
In Verrem
De
Select Orations (for Schools).

Imperio Gn. Pompeii. Pro Archia. Philippica IX. With Introduction and
Notes by J. R. King, M.A. Second Edition. Extra fcap. 8vo. is. 6d.
Cicero.
With Notes.
Cornelius Nepos.
Extra fcap. 8vo.
Second Edition.
I.
Browning, M.A.
By Oscar
is. 6d.
With Notes and Maps.

Selections (for Schools).
H. Lee- Warner, M.A. Extra fcap. 8vo. In Parts, limp, each is. 6d.
The Caudine Disaster.
Part I.
By
Livy.
Part
II.
Hannibal's Campaign in Italy.
Part III. The Macedonian War.
Books
Livy.
VII.
Books XXI, XXII, and XXIII.
Livy.
By
Extra fcap. 8vo. y. 6d.
A. R. Cluer, B.A.
With Introduction
By M. T. Tatham, M.A. Extra fcap. 8vo. 45- ^ d

Ovid. Selections for the use of Schools. With Introductions
and Notes, and an Appendix on the Roman Calendar. By W. Ramsay, M.A.
Edited by G. G. Ramsay, M.A. Third Edition. Extra fcap. 8vo. $s. 6d.
and Notes.
Ovid.
Tristia.
Book
duction and Notes.
Plautus.
By
The Text
I.
S.
G. Owen, B.A.
The Trinummus.
revised, with
Extra fcap. 8vo.
With Notes and
Intended for the Higher Forms of Public Schools.

and A. Sloman, M.A. Extra fcap. 8vo. 3-r.
Pliny.
Selected Letters (for Schools).
late C. E. Prichard,
fcap. 8vo.
Sallust.
M.A.
Tacitus.
By
3-r.
an Intro6d.
Introductions.
M. A.,
C. E. Freeman,
With Notes.
M.A., and E. R. Bernard, M.A.
By
Second Edition.
the
Extra
3-r.
By W. W.

Extra fcap. 8vo.
Capes,
4^. 6d.
The Annals.
Books I-IV.
Edited, with Introduc-
and Notes for the use of Schools and Junior Students, by H. Furneaux,
M.A. Extra fcap. 8vo. $s.
tion
Terence.
Andria.
With Notes and
E. Freeman, M.A., and A. Sloman, M.A.
Virgil.
M.A.
Introductions.
Extra fcap. 8vo.

Two
vols.
Crown
8vo. lew. 6d.
Catulli Veronensis Liber.
Commentary on
Demy
Catullus.
8vo. i6j.
C 2
C.
T. L. Papillon,
separately, 4*. 6d.
Iterum recognovit, apparaturn
ticum prolegomena appendices addidit, Robinson
1876.
By
The Text
By
3^.
Ellis,
A.M.
By Robinson
cri-
1878.
Demy
Ellis,
M.A.
CLARENDON
20
PRESS, OXFORD.
Catulli Veronensis Carmina Selecta, secundum recognitionem

Robinson Ellis, A.M. Extra fcap. 8vo. $s. 6d.
Cicero de Oratore. With Introduction and Notes.

Wilkins, M.A,
Book II. 1881. 8vo. gj.
Book I. 1879. 8vo. 6s.
- Philippic Orations.
Second Edition.
With Notes.
By
J.
By A.
S.
R. King, M.A.
8vo. los. 6d.
1879.
Select Letters. With English Introductions, Notes, and

By Albert Watson, M.A. Third Edition. 1881. Demy 8vo. 185.
Appendices.
Text.
Select Letters.
Extra fcap. 8vo.
Edition.
same
the
By
Second
Editor.
4^.
pro Cluentio. With Introduction and Notes. By W.

Ramsay, M.A. Edited by G. G. Ramsay, M.A. Second Edition. Extra fcap.
8vo.
6d.
3-r.
Horace. With a Commentary. Volume I. The Odes, Carmen

Seculare, and Epodes.
By Edward C. Wickham, M.A. Second Edition.
Demy
1877.
reprint of the above, in a size suitable for the use
of Schools.
Livy,
Book
Extra fcap. 8vo.
By
J.
tjs.
Ellis,
A.M.
6d.
Introduction, Historical Examination,
R. Seeley, M.A.
P. Ovidii Nasonis
Scholia Vetera
R.
With
I.
and Notes.
Ovid.
8vo. I2J.
Second Edition.
Ex
Ibis.
1881.
8vo. 6s.
Novis Codicibus
Commentarium cum Prolegomenis Appendice Indice
edidit,
addidit,
8vo. los. 6d.
Persius. The Satires. With a Translation and Commentary.

Edited by Henry Nettleship, M.A.
Second
By John Conington, M.A.
Edition.
Tacitus.
1874.
8vo.
fs. 6d.
The Annals.
Books
I- VI.
duction and Notes, by H. Furneaux, M.A.
Nettleship (H., M.A.}.
Edited, with
Lectures and Essays on Subjects con-
nected with Latin Scholarship and Literature.
The Roman Satura :

its
literary
development.
its original
8vo. sewed, is.
Ancient Lives of Vergil.
Crown
form
Poets.
1869.
8vo. 15*.
'js.
8vo. sewed,
Manual of Comparative
Third Edition, Revised and Corrected.
Pinder (North^ M.A.).
8vo.
6d.
in connection
with
With an Essay on the Poems
of Vergil, in connection with his Life and Times.
Papillon (T. L., M.A.).
Intro-
8vo. i8s.
Selections
1882.
Crown
from
2s.
Philology.
8vo. 6s.
the less
known Latin

Sellar
(W.
VIRGIL.
Y.,
New
Roman
Edition.
8vo. QS.
iSSi.
New
(J.,
Edition, Revised
8vo. 14^.
M.A.).
Fragments and Specimens of Early
With Introductions and Notes. 1874. 8vo.
Latin.
III.
Poets of the Augustan Age.
Crown
1883.
Poets of the Republic.
and Enlarged.
Wordsworth
Roman
M.A.).
21
i8s.
GREEK.
Greek Primer, for the use of beginners
in that
Language.
the Right Rev. Charles Wordsworth, D.C.L. Seventh Edition. Extra fcap.
8vo. is. 6d.
By
Graecae Grammaticae Rudimenta in usum Scholarum.

tore Carolo
Wordsworth, D.C.L.
I2mo.
Nineteenth Edition, 1882.
Auc4^.
Greek-English Lexicon, abridged from Liddell and Scott's

4to. edition, chiefly for the use of Schools,
fs. 6d.
Twenty-first Edition.
1884.
Square I2mo.
Greek Verbs, Irregular and Defective ; their forms, meaning,

and quantity; embracing all the Tenses used by Greek writers, with references
to the passages in which they are found. By W. Veitch.
Fourth Edition.
Crown Svo. IQJ. 6d.
The Elements of Greek Accentuation (for Schools) abridged

from his larger work by H.W. Chandler, M.A. Extra fcap. Svo. is.Gd.
:
GREEK READERS:
W.
G. Rushbrooke, M.L. Second
By
SERIES OF GRADUATED
First Greek Reader.
Edition.
Extra fcap. Svo.
Second Greek Reader.

8vo.
3J-.
2s. 6d.
By A. M.
Bell,
M.A.
Extra
fcap.
6d.
Fourth Greek Reader ; being Specimens of Greek Dialects.

With Introductions and Notes. By W. W. Merry, M.A. Extra fcap. Svo.
4-r.
6d.
Fifth Greek Reader.
Selections from
Dramatic Poetry, with Introductions and Notes.

Extra fcap. Svo. 4.?. 6d.
Greek Epic and

By Evelyn Abbott, M.A.
The Golden Treasury of Ancient Greek Poetry: being a Collection of the finest passages in the Greek Classic Poets, with Introductory
Notices and Notes. By R. S. Wright, M.A. Extra fcap. Svo. 8.r. 6d.
Golden Treasury of Greek Prose, being a Collection of the

finest
passages in the principal Greek Prose Writers, with Introductory Notices

By R. S. Wright, M.A., and J. E. L. Shadwell, M.A. Extra fcap.
and Notes.
Svo.
4-f.
6d.
22
Prometheus Bound
Aeschylus.
With Introduc-
(for Schools).
and Notes, by A. O. Prickard, M.A. Second Edition. Extra fcap. 8vo.
tion
Agamemnon.
Sidgwick, M.A.
Second Edition.
With
Choephoroi.
Extra fcap. "Svo.
Editor.
Introductions, &c., by
I.
II.
Extra fcap. 8vo.
The
The
3^.
Introduction and Notes
by the same
3-y.
In Single Plays.
Aristophanes.
W. W.
Edited, with English Notes,
Merry, M.A.
Clouds, Second Edition,

2s.
Acharnians,
Extra fcap. 8vo.
is.
III.
The
Frogs,
2s.
Tabula. With Introduction and Notes.

Jerram, M.A. Extra fcap. 8vo. is. 6d.
By
Cebes.
Alcestis (for Schools).
Euripides.
Extra fcap. 8vo.
Helena.
2s.
With Introduction and Notes, by Arthur
By
C. S. Jerram,
C. S.
M.A.
2s. 6d.
Edited, with Introduction, Notes, and Critical

By C. S. Jerram, M.A. Extra
Appendix, for Upper and Middle Forms.

fcap. 8vo. 3^.
Iphigenia in Taiiris.
and
Critical
Edited, with Introduction, Notes,

By C. S. Jerram, M.A.
Appendix, for Upper and Middle Forms.
Extra fcap. 8vo. cloth,
3^.
Herodotus, Selections from. Edited, with Introduction, Notes,

and a Map, by W. W. Merry, M.A. Extra fcap. 8vo. 2s. 6d.
Homer.
By W. W.
Books I-XII (for Schools).

Twenty-seventh Thousand. Extra fcap. 8vo. 4-r. 6d.
Odyssey,
Merry, M.A.
Book
II, separately, is. 6d.
Odyssey, Books
same Editor.
XIII-XXIV
Second Edition.
(for Schools).
Extra fcap. 8vo. 5-r.
Iliad, Book I (for Schools).

Second Edition. Extra fcap. 8vo. 2s.
By D.
By
B. Monro,
the
M.A.
Iliad) Books I-XII (for Schools). With an Introduction,

a brief Homeric Grammar, and Notes. By D. B. Monro, M.A. Extra fcap.
8vo. 6s.
Iliad,
Notes.
Lucian.
Books VI and XXI.
M.A. Second
Plato.
With Introduction and
By Herbert Hailstone, M.A. Extra fcap.

Vera Historia (for Schools).
Edition.
Selections
is. 6d.
By
C.
each.
S.
Jerram,
the Dialogues [including the
whole of
Extra fcap. 8vo.
from
8vo.
is. 6d.
the Apology and Crito\. With Introduction and Notes by John Purves, M.A.,
and a Preface by the Rev. B. Jowett, M.A. Extra fcap. 8vo. 6s. 6d.
Edited with Intro-
For the use of Schools.
Sophocles.
23
ductions and English Notes. By Lewis Campbell, M.A., and Evelyn Abbott,
M.A. New and Revised Edition. 2 Vols. Extra fcap. 8vo. IQJ. 6d.
Sold .separately, Vol.
Text, 4*. 6d.
I,
Vol.
II,
Explanatory Notes,
6s.
In Single Plays, with English Notes, &c.
Sophocles.
By
Lewis Campbell, M.A., and Evelyn Abbott, M.A. Extra fcap. 8vo. limp.
Oedipus Tyrannus, Philoctetes. New and Revised Edition, 2 j. each.
Oedipus Coloneus,
Antigone,
Oedipus Rex:
Dindorfs
present Bishop of St. David's.
gd. each.
is.
2s.
Trachiniae,
Electra,
Ajax,
each.
Text,
with Notes
Extra fcap. 8vo. limp,
is.
by the
6d.
With Notes. By H.
Theocrit2is (for Schools).
D.D. (late Snow). Third Edition. Extra fcap. 8vo. ^s. 6d.
Kynaston,
Easy Selections (for Junior Classes). With a

Vocabulary, Notes, and Map. By J. S. Phillpotts, B.C.L., and C. S. Jerram,
M.A. Third Edition. Extra fcap. 8vo. 3*. 6oT.
Xenophon.
With Notes and Maps.
Selections (for Schools).

J. S. Phillpotts,
Extra fcap. 8vo.
Fourth Edition.
B.C.L.
By
3*. 6d.
Anabasis, Book I. Edited for the use of Junior Classes

and Private Students. With Introduction, Notes, and Index. By J. MarExtra fcap. 8vo.
shall, M.A., Rector of the Royal High School, Edinburgh.
2S. 6d.
Anabasis,
Jerram, M.A.
Book II. With

Extra fcap. 8vo. 2s.
Cyropaedia, Books
Notes by C. Bigg, D.D.
Notes and Map.
IV and V.
Extra fcap. 8vo.
Aristotelian
Book of the Nicomachean
Aristotelis
Bekkeri.
Ethics.
Demostlienes
8vo.
and
[In the Press.]

of the Seventh
By J. C. Wilson, M.A. 8vo. stiff, 5^.

ex recensione Immanuelis
Ethica Nicomachea
Crown
C. S.
2s. 6d.
By W. L. Newman, M.A.
I. On the Structure
Studies.
Aristotle's Politics.
By
5^.
Aeschines.
The
Orations of Demosthenes
By
and Êschines on the Crown. With Introductory Essays and Notes.

G. A. Simcox, M.A., and W. H. Simcox, M.A. 1872. 8vo. I2s.
Hicks (E. L.,M.A.).

tions.
Demy 8vo. los.
Homer.
A Manual of Greek Historical Inscrip6d.
Odyssey, Books I-XII.
Appendices, etc. By W.
1886. Second Edition.
Edited with English Notes,

W. Merry, M.A., and the late James Riddell, M.A.
Demy 8vo. i6s.
24
Homer.
M.A.
A Grammar of the Homeric Dialect. By D. B. Monro,

8vo. loj. 6d.
Demy
The
Sophocles.
Plays and Fragments.
With English Notes
and Introductions, by Lewis Campbell, M.A. 2 vols.

Vol. I.
Oedipus Tyrannus.
Oedipus Coloneus.
1879. 8vo. i6s.
Vol. II. Ajax. Electra. Trachiniae.
8vo. i6s.
Antigone.
Second
Edition.
IV.
%
Philoctetes.
Fragments.
1881.
FRENCH AND ITALIAN.
Brackets Etymological Dictionary of the French Language
',
with a Preface on the Principles of French Etymology.

Translated into
English by G. W. Kitchin, D.D. Third Edition. Crown 8vo. 7^. 6d.
'
Historical
Grammar of
lated into English

8vo. 3-r. 6d.
by G.
"Works by
W.
the
French Language.
Kitchin, D.D.
Fourth Edition.
GEORGE SAINTS BURY,
Primer of French Literattire.
Extra
TransExtra fcap.
M.A.
fcap. 8vo. 2s.
Short History of French Literature. Crown 8vo. ios.6d.

Specimens of French Literattire^ from Villon to Hugo. Crown
8vo.
s.
Corneiltts Horace.
George Saintsbury, M.A.
Extra fcap. 8vo.
by
2s. 6d.
Moliere's Les Precieuses Ridicules.

Edited, with Introduction
and Notes, by Andrew Lang, M.A. Extra fcap. 8vo. is. 6d.
Racine's Esther.
Edited, &c.
by George Saintsbury, M.A.
{Nearly ready.)
Beaumarchais* Le Bar bier de

and Notes, by Austin Dobson.
VoltairJs Merope.
George Saintsbury.
Seville. Edited, with Introduction

Extra fcap. 8vo. 2s. 6d.
Edited, with Introduction and Notes, by

Extra fcap. 8vo. cloth,
is.
Mussefs On ne badine pas avec V Amour, and Fantasio. Edited,

with Prolegomena, Notes,
8vo.
etc.,
by Walter Henries Pollock.
Extra fcap.
2s.
Sainte-Beuve.
Selections from the Causeries dti

by George Saintsbury. Extra fcap. 8vo. 2s.
Quinet's Lettres a sa Mere.

Saintsbury.
Extra fcap. 8vo.
2s.
Lundi. Edited
Selected and edited
by George
25
U Eloquence de la Chaire et de la Tribune Frangaises.

by Paul Blouet, B.A. (Univ.
Extra fcap. 8vo.
2s.
Louis
Edited
French Sacred Oratory
I.
6d.
GUSTAVE MASSOW,
Edited by
With Notes, Glossary,
CorneiHJs Cinna.
cloth, 2s.
Vol.
Gallic.).
B.A.
Extra fcap. 8vo.
etc.
XIV and his
Contemporaries ; as described in Extracts
from the best Memoirs of the Seventeenth Century.

Genealogical Tables, &c. Extra fcap. 8vo. 2s. 6d.
With English Notes,
Maistre, Xavier de. Voyage autour de ma Chambre. Ourika,

by Madame de Duras; Le Vieux Tailleur, by MM. Erckmann-Chatrian ;
La Veillee de Vincennes, by Alfred de Vigny ; Les Jumeaux de 1'Hotel
Comeille,by Edmond Abotit ; Mesaventures d'un Ecolier, by Rodolphe Topffer.
Third Edition, Revised and Corrected.
MoliMs
With
Extra fcap. 8vo.
Voltaire's Life of Moliere.
Extra fcap. 8vo.
Moliere's Les Fourberies de Scapin.

Moliere.
MoliMs
2s.
6d.
Les Fourberies de Scapin^ and Racine s Athalie.

Extra fcap. 8vo.
stiff
covers,
i s.
6d.
Voltaire's Life of
6d.
Les Femmes Savanfes.
Extra fcap. 8vo.
2s.
With
With Notes,
Glossary, etc.
cloth, 2s.
Racine's Andromaque^ and
Corneilles
Louis Racine's Life of his Father.
Le Menteur.
Extra fcap. 8vo.
With
6d.
2s.
RegnaroTs Le Joueur^ and Brueys and Palaprafs Le Grondeur.

Extra fcap. 8vo.
6d.
2s.
Madame
de>
and her
the Correspondence of.

Extra fcap. 8vo. 3-r.
from
Dante.
Notes.
Tasso.
Selections from the Inferno.

By H. B. Cotterill, B.A. Extra fcap.
La Gemsalemme
troduction and Notes.
By
the
V.
ScJierer
chief Contemporaries^ Selections
Intended more especially for Girls' Schools.
W.}.

8vo. 4^. 6d.
Cantos
same Editor.
Extra fcap. 8vo.
is. 6d.
Literature.
Translated
Max
i,
ii.
In-
GERMAN.
History of
German
from the Third German Edition by Mrs. F. Conybeare.

Miiller.
With
Liberata.
Edited by F.
2 vols. 8vo. 2 is.
Max Miiller.
The German
Classics, from the Fourth to the

With Biographical Notices, Translations into Modern
By F. Max Miiller, M.A. A New Edition, Revised,
Enlarged, and Adapted to Wilhelm Scherer's History of German Literature,'
Nineteenth Century.
German, and Notes.
'
by F. Lichtenstein.
2 vols.
crown 8vo. 2U.
26
GERMAN COURSE. By HERMANN LANG-E.

The Germans at Home ; a Practical Introduction to German
Conversation, with an Appendix containing the Essentials of German Grammar.
Second Edition. 8vo. 2s. 6d.
The German Manual ; a German Grammar, Reading Book,

and a Handbook of German Conversation.
8vo.
Grammar of the German Language.
7-r.
8vo.
German
Theoretical and
Composition ;
the Art of Translating English Prose into German.
6d.
y.
6d.
Guide to
Practical
8vo. 4s. 6d.
Lessing's Laokoon. With Introduction, English Notes, etc.

By A. Hamann, Phil. Doc., M.A. Extra fcap. 8vo. 43. 6d.
Schiller s
Translated into English Verse by
Wilhelm Tell
E. Massie, M.A.
Extra fcap. 8vo.
5^.
Also, Edited
by C. A. BTTCHHEIM, Phil. Doc.

Goethe's Egmont. With a Life of Goethe, &c. Third Edition.
Extra fcap. 8vo.
Iphigenie
3^.
A Drama.
auf Tauris.
troduction and Notes.
Second Edition.
With a
Extra fcap. 8vo.
Critical In-
3-r.
Heine's Prosa^ being Selections from his Prose Works.

English Notes,
etc.
Extra fcap. 8vo.
With
Heine's Harzreise.
of the Harz, and Index.
Lessing^s
Life of Heine, Descriptive Sketch
Extra fcap. 8vo. paper covers,
Minna von Barnhelm.
is.
6d.
Comedy.
of Lessing, Critical Analysis, Complete Commentary, &c.

Extra fcap. 8vo. $s. 6d.
Nathan der
Extra fcap. 8vo.
4-r.
With
Weise.
cloth, 2s. 6d.
With a Life
Fourth Edition.
Introduction, Notes, etc.
6d.
Egmonfs Leben und
Schiller's Historische Skizzen ;
Tod, and
Belagerung von Antwerpen. Third Edition, Revised and Enlarged.

a Map. Extra fcap. 8vo. 2s. 6d.
Wilhelm
torical
and
and Map.
Sixth Edition.
Tell.
With
With a
Tell.
critical Introduction,
Wilhelm
With
4^. 6d.
Life of Schiller; an hisArguments, and a complete Commentary,
Extra fcap. 8vo.
3^.
6d.
School Edition. With Map.
Modern German Reader.
2s.
Graduated Collection of Ex-
and Poetry from Modern German writers

Part I. With English Notes, a Grammatical Appendix, and a complete
Vocabulary. Fourth Edition. Extra fcap. 8vo. 2s. 6d.
Part II. With English Notes and an Index. Extra fcap. 8vo. is. 6d.
tracts in Prose
CLARENDON
PRESS, OXFORD.
27
MATHEMATICS, PHYSICAL SCIENCE,
VI.
&c.
By LEWIS HENSLEY, M.A.

Figures made Easy : a first Arithmetic Book. Crown 8vo. 6d.
Answers to the Examples in Figiires made Easy, together
with,
two thousand additional Examples, with Answers.
Crown
8vo.
The Scholar's Arithmetic : with Answers. Crown 8vo.

The Scholar s Algebra. Crown 8vo. 4s. 6d.
Lessons on Thermodynamics.
Baynes (R. E., M.A.).

Crown
8vo.
7-y.
F., F.R.A.S.).
Third Edition. 1877.
Clarke (Col. A. R.,C.B.,R.E.).
Cremona
1878.
6d.
Chambers (G.
Astronomy.
is.
4^. 6d.
(Luigi).
Handbook of Descriptive
Demy
8vo. 28^.
8vo.
us. 6d.
Elements of Projective Geometry.
Trans-
lated by C. Leudesdorf,
M.A.
1880.
Geodesy.
I2s.6d.
8vo.
Donkin. Acoustics. Second Edition. Crown 8vo. Js. 6d.

Euclid Revised.
Containing the Essentials of the Elements
of Plane Geometry as given by Euclid in his
R. C. J. Nixon, M.A. Crown 8vo. is. 6d.
Books I-IV.
Books
By
I, II.
By
!.
Demy
8vo.
Harcourt
the same Editor,
The Construction of Healthy
C.),
and J.
Maclaren (Archibald).
Theoretical and Practical.
Large
(H.
Ball.
G.,
Book-keeping.
Extra fcap. 8vo. limp cloth,
Exercises in Practical Chemistry.

Edition. Crown Svo. gs.
Madan
is. 6d.
is.
Vernon, M.A.), and
(A. G.
8vo. $s. 6d.
los. 6d.
Hamilton (Sir R. G.
and enlarged Edition.
Edited by
the same Editor,
Galton (Douglas, C.B., F.R.S.).

Dwellings.
Six Books.
Crown
the same Editor.
By
Also, nearly ready
Book
first
Vol.
Third
System of Physical Education
Extra fcap. Svo.
Tables
M.A.).
H. G. Madan, M.A.
Elementary Exercises.
I.
New
2s.
7-r.
6d.
of Qualitative Analysis.
4to. paper, 4^. 6d.
Maxwell
Treatise on Electricity
(J. Clerk, M.A., F.R.S.).
and Magnetism. Second Edition. 2 vols. Demy Svo. i/. us. 6d.
An Elementary Treatise on Electricity. Edited by
William Garnett, M.A. Demy Svo. 7-r. 6d.
Minchin (G. M., M.A.).
Treatise on Statics with Applica-
Third Edition, Corrected and Enlarged.
tions to Physics.
brium of Coplanar
Forces. Svo. gs.
Vol.
II. Statics.
Svo. i6s.
Vol.
I.
Equili-
28
Mine/tin (G.
Uniplanar Kinematics of Solids and
M.A.).
Jlf.,
Crown
Fluids.
8vo. JS. 6d.
Phillips (John, M.A., F.R.S.).

Valley of the Thames.
Vesuvius.
Geology of Oxford
Crown
1869.
Stratigraphical.
Vol.
the
los. 6d.
8vo.
Prestwich (Joseph, M.A., F.R.S.).

and
and
8vo. 2is.
1871.
Geology, Chemical, Physical,
Chemical and Physical.
I.
Rollestons Forms of Animal Life.
New
and Drawings of Dissections.
Illustrated
Edition.
Royal 8vo.
25^.
by Descriptions
(Nearly ready.}
Observed, Reduced,
Cycle of Celestial Objects.
and Discussed by Admiral W. H Smyth, R. N. Revised, condensed, and greatly
enlarged by G. F. Chambers, F.R.A.S. 1881. 8vo. Price reduced to \is.
Smyth.
Stewart (Balfour, LL.D., F.R.S.}.
Treatise on Heat, with
numerous Woodcuts and Diagrams. Fourth Edition. Extra
Vernon-Harcourt (L.
F.,
M.A.).
fcap. 8vo. ^s.6d.
Treatise on Rivers
and
Canals, relating to the Control and Improvement of Rivers, and the Design,
2 vols. (Vol. I, Text.
Vol. II,
Construction, and Development of Canals.
Plates.) 8vo. 2U.
and
Harbours
Docks ; their Physical Features, History,
Construction, Equipment, and Maintenance ; with Statistics as to their Com2 vols. 8vo. 25^.
mercial Development.
Watson (H. W., M.A.).

of Gases.
1876. 8vo.
Watson (H. W., D.
Treatise on the Kinetic Theory
3^. bd.
Sc., F.R.S.},
and
S.
H. Burbury, M.A.
Treatise on the Application of Generalised Coordinates to the Kinetics of

a Material System. 1879. 8vo. 6s.
Vol. I. ElectroII. The Mathematical Theory of Electricity and Magnetism.
statics.
8vo. los. 6d.
I.
Williamson (A.
Students.
W., Phil. Doc., F.R.S.).
A new Edition,
with Solutions.
VII.
Bluntschli (J. K.}.
Chemistry for
1873. Extra fcap. 8vo.
8j-.
6d.
HISTORY.
The Theory of
the
State.
By
J.
K.
Bluntschli, late Professor of Political Sciences in the University of HeidelAuthorised English Translation from the Sixth German Edition.
berg.
Demy 8vo. half bound, 12s. 6d.
History of Greece from its Conquest by the Romans to the present time, B.C. 146 to A.D. 1864. A new
Edition, revised throughout, and in part re-written, with considerable additions, by the Author, and edited by H. F. Tbzer, M.A. 7 vols. 8vo. 3/. IOT.
Finlay (George, LL.D.}.
29
Fortescue (Sir John, Kt.).

The Governance of England:
otherwise called The Difference between an Absolute and a Limited Mon-
archy.
Revised Text.
by Charles Plummer, M.A.
Edited, with Introduction, Notes, and Appendices,

8vo. half bound, \2s. 6d.
Freeman (E.A., D.C.L.}.

Conquest of England.
Short History of the
Second Edition.
Extra fcap. Svo.
Norman
2s. 6d.
George (H. B., M.A .). Genealogical Tables illustrative of Modern

Third Edition, Revised and Enlarged.
History.
Hodgkin
Italy
(T.).
Maps. Vols.
Plates and
and her Invaders.
Maps, Plans, and Tables.

each IQS. 6d.
i.
Down
Vol.
2.
From 1453-1624.
J.,
America.
1 2 s.
with
Illustrated
3/. Ss.
History of France. With numerous
In Three Volumes.
Vol.
Payne (E.
Ranke
IV., A.D. 376-553. 8vo.
Kitchin (G. W.,D.D.}.

to the
Small 4to.
Second Edition.
Crown
8vo.
Year 1453.
M.A.}.
Vol.
3.
From 1624-1793.
United States of
History of the
In the Press.
(L. von}.
History of England, principally in the
Seventeenth Century. Translated by Resident Members of the University of

Oxford, under the superintendence of G. W. Kitchin, D.D., and C. W. Boase,
M.A. 1875. 6 vols. 8vo. 3/. $s.
A Mamial
Rawlinson (George, M.A.).

Second Edition. Demy 8vo. 14^.
Select Charters
of Ancient History.
and other Illustrations of English
Constitutional
History, from the Earliest Times to the Reign of Edward I.

Arranged and
edited by W. Stubbs, D.D. Fifth Edition.
1883. Crown 8vo. 8s. 6d.
Stubbs
W., D.D.}.
in its Origin
Also
in 3 vols.
Seventeen
Modern Histpry,
loj-.
The Constitutional History of England,
and Development.
Library Edition.
3 vols.
demy
8vo.
2/. Ss.
crown 8vo. price iw. each.
Lectures
on
the
Stiidy
&c., delivered at Oxford 1867-1884.
of Medieval and
Demy 8vo. half-bound,
6d.
Wellesley.
Selection
from
the Despatches,
Treaties,
and
other Papers of the Marquess Wellesley, K.G.,

during his Government
of India. Edited by S. J. Owen, M.A. 1877. 8vo. i/.
4*.
Wellington.
Selection
from
the Despatches, Treaties,
other Papers relating to India of Field-Marshal the

Edited by S. J. Owen, M.A. 1880. 8vo. 24*.
History of British India.
By
S. J.
in Indian History in the

University of Oxford.
Duke of Wellington,
and
KG
Owen, M.A., Reader

In preparation.
3o
VIII.
LAW.
De
lure Belli Libri Tres.

Alberi'ci Gentilis, LCD., J.C.,
EdiditT. E. Holland, I.C.D. 1877. Small 4-10. half morocco, 2\s.
Anson (Sir William
Bart., D.C.L.).
fi.,
Principles of the
Law of Contract, and of Agency in its Relation, to Contract. Fourth

Edition. Demy 8vo. IO.T. 6d.
Law and Custom of the Constitution. Part I. Parliament. Demy 8vo. IO.T. 6d.
An Introduction to the Principles of
Bentham (Jeremy).
English
Morals and Legislation.
Crown
8 vo. 6s. 6d.
Digby (Kejielm~E., M.A.). An Introduction to the History of

the Law of Re-al Property. Third Edition. Demy 8vo. Jos. 6d.
Gait Institutionum Juris Civilis Commentarii Quattuor ; or,
Elements of
Roman Law by
by Edward Poste, M.A.
Gaius.
With a Translation and Commentary
Second Edition.
1875.
8vo. i8j.
Hall ( IV. E., M.A.}. International Law. Second Ed. 8vo. zis.
Holland (T. E., D.C.L.). The Elements of Jurisprudence.
Third Edition. Demy 8vo. IO.T. 6d.
The European Concert in the Eastern Question a Col,
lection of Treaties
and other Public Acts.
Edited, with Introductions and
Notes, by Thomas Erskine Holland, D.C.L. 8vo. 12s. 6d.

Imperatoris lustiniani Institutionum Libri Quatttwr ; with
Introductions, Commentary. Excursus and Translation. By J. E. Moyle, B.C.L.,
M.A. 2 vols. Demy 8vo. zis.
Justinian^ The Institutes of, edited as a recension of the
Institutes of Gaius, by Thomas Erskine Holland, D.C.L. Second Edition,
1881.
Extra fcap. 8vo.
5.?.'
Justinian^ Select Titles from the Digest

D.C.L., and C. L. Shadwell, B.C.L.
of.
By T.
E. Holland,
8vo. 14*.
Also sold in Parts, in paper covers, as follows

Part II. Family Law.
Introductory Titles. 2s. 6d.
Part III. Property Law. 2s. 6d.
Part IV. Law of Obligations (No. i).
Part IV. Law of Obligations (No. 2). 45. 6d.
:
Part
Lex
I.
Aquilia.
The Roman Law
of
Damage
to
is.
35. 6d.
Property
being a Commentary on the Title of the Digest Ad Legem Aquiliam (ix. 2).
With an Introduction to the Study of the Corpus luris Civilis. By Erwin
Grueber, Dr. Jur., M.A. Demy 8vo. IQS. 6d.
'
'
Markby (
W., D. C.L.). Elements of Law considered with refer-
ence to Principles of General Jurisprudence. Third Edition.
Twiss (Sir Tr avers, D.C.L.).
The Law
Demy 8vo. I2s.6d.

of Nations considered
as Independent Political Communities.

Part I. On the Rights and Duties of Nations in time of Peace.
Revised and Enlarged. 1884. Demy 8vo. 15^.
Part II. On the Rights and Duties of Nations in Time of War.
Revised.
1875.
Demy
8vo.
2U.
A new Edition,
Second Edition
CLARENDON PRESS, OXfORD.

IX.
MENTAL AND MORAL PHILOSOPHY,
Bacons Novum Organum.

G.
31
&c.
Edited, with English Notes,
6d.
W. Kitchin, D.D. 1855. 8vo.
Translated by G. W. Kitchin, D.D.
by
9-$-.
1855. 8vo.
9.9. 6'd.
The Works of George Berkeley, D.D., formerly
Berkeley.
Bishop of Cloyne
including
of his writings hitherto unpublished.

Life and Philosophy,
Svo. 2/. iSs.
many
With Prefaces, Annotations, and an Account of his

by Alexander Campbell Fraser, M.A. 4 vols. 1871.
The Life, Letters, &c. i vol. i6s.
With an Introduction and Notes.

Selections from.
For the use of Students in the Universities. By Alexander Campbell Fraser,
LL.D. Second Edition. Crown Svo. 7*. 6d.
The Elements of Dediictive Logic, designed
( T., D.D.}.
mainly for the use of Junior Students in the Universities. Eighth Edition,
with a Collection of Examples. Extra fcap. Svo. %s. 6d.
Fowler
The Elements of Inductive Logic, designed mainly
for
the use of Students in the Universities. Fourth Edition. Extra fcap. Svo. 6s.
Edited by T.
Bacon.
1878.
Locke's
FOWLEK,
D.D.
Introduction, Notes, &c.
Novum Organum. With

8vo. 147.
Conduct of the
Extra fcap. Svo.
Understanding.
Second
Edition.
2s.
The Wealth of Households. Crown Svo. $s.

(J. T.).
Edited by
Green (T. H., M.A.). Prolegomena to Ethics.
A. C. Bradley, M.A. Demy Svo. 12s. 6d.
The Logic of Hegel ; translated from the EncycloHegel.
Danson
With Prolegomena by William
paedia of the Philosophical Sciences.

Wallace, M.A. 1874. 8vo 14*-
Lotze*s Logic., in
Three Books
of Thought, of Investigation,
and of Knowledge. English Translation; Edited by B. Bosanquet, M.A.,

Fellow of University College, Oxford. Svo. cloth, I2s. 6d.
Metaphysicy in Three Books; Ontology, Cosmology,
Edited by B. Bosanquet, M.A.
and Psychology. English Translation
;
Svo. cloth,
1 2J.
6d.
Martineau (James D.D. ).
Types of Ethical Theory.
Edition.
2 vols.
Crown
Rogers (J.E. Thorold, M.A .).

for the use of Schools.
A Manual of Political Economy,
Third Edition.
SmitJis Wealth of Nations.

J.
Second
Svo. 15^.
E. Thorold Rogers. M.A.
Extra fcap. Svo. 4 s. 6d.
A new Edition, with
2 vols.
Svo.
Wilson (J. M., B.D.\ and T. Fowler, D.D.

of Morals (Introductory Chapters).
Notes,
by
1880. zis.
8vo. boards, 3*. 6d.
The Principles
32
X. ART, &c.
The Cultivation of
Hullah (John}.
Second Edition.
Extra fcap. 8vo.
2s.
Ouseley (Sir F. A. Gore, Bart.}.

Third Edition.
A
upon
the
Treatise on
Harmony.
4to. los.
Treatise on Counterpoint, Canon,

that of Cherubini.
Voice.
Speaking
6d.
Second Edition,
Treatise on Miisical
Second Edition. 4to.
^.to.
Form and
and Fugue, based
i6s.
General Composition.
los.
Robinson (J.
Critical Account of the

C., F.S.A.).
by Michel Angela and Raffaello in the University Galleries, Oxford.
Crown 8vo. 4^.
M. A.). A
Ruskin (John,
Drawings
Course of Lectures on Art, delivered
before the University of Oxford in Hilary Term, 1870.
Troutbeck (J.,
M. A.) and R. F. Dale, M.A.
(for Schools).
Second Edition.
Tyrwhitt (R. St.

With coloured
A. Macdonald.
Vanx ( W.
J.,
M.A.}.
Illustrations,
1870.
Crown
8vo.
8vo. 6s.
A Music Primer
is. 6d.
Handbook of Pictorial Art.
Photographs, and a chapter on Perspective by

1875. 8vo. half morocco, iSs.
Second Edition.
S. W., M.A.).
Catalogue of the Castdlani Collec-
tion of Antiquities in the University Galleries, Oxford.
The Oxford Bible for
Crown
8vo.
is.
Teachers, containing supplemen-
tary HELPS TO THE STUDY OF THE BIBLE, including Summaries

of the several Books, with copious Explanatory Notes and Tables
illustrative of Scripture History and the characteristics of Bible
Lands; with a complete Index of Subjects, a Concordance, a Dictionary of Proper Names, and a series of Maps. Prices in various sizes
and bindings from 3^. to 2.1. $s.
to the Study of the Bible, taken from the

OXFORD BIBLE FOR TEACHERS, comprising Summaries of the
Helps
several Books, with copious Explanatory Notes and

trative of Scripture History and the Characteristics of
with a complete Index of Subjects, a Concordance,
of Proper Names, and a series of Maps. Crown 8vo.
l6mo.
Tables
illus-
Bible Lands
a Dictionary
cloth,
y.
6d.
cloth) \s.
LONDON: HENRY FROWDE,
OXFORD UNIVERSITY PRESS WAREHOUSE, AMEN CORNER,
OXFORD: CLARENDON PRESS DEPOSITORY,

n6 HIGH
STREET.
DELEGATES OF THE PRESS invite suggestions and advice from all persons
education; and will be thankfiil for hints, &c, addressed to the
SECRETARY TO THE DELEGATES, Clarendon Press, Oxford.
ggr The
interested in
'nva'ana
IO
03
cC
II
0)
CD
CO
0)
03
CO
ilHffl
OINOH01 JO 1IISHHAIHQ
j
at the
OXFor, D University

Baynes (Lessons of Thermodynamics) (1878)

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Baynes (Lessons of Thermodynamics) (1878)

Hochgeladen von

Copyright:

Verfügbare Formate

EDWARD

OXFORD UNIVERSITY PRESS WAREHOUSE

SENIOR STUDENT OF CHRIST CHURCH, OXFORD

AT THE CLARENDON PRESS

[All rights reserved]

substance of the following Lessons formed a

Lectures on Thermodynamics which were

Hilary Term 1876. They have been revised, and are

such treatise exists at present

Rankine's Manual of the Steam- Engine and

of practical Engineers, contains only a short sketch of

place in the general

too brief for the wants

has not gone far into the Mathematical Development,

other respects the work

and perhaps the most suggestive that has been written

memoirs have been studied

course of the preparation of these Lessons, and copious

the student will thus be

Dr. Zeuner's Grund-

particular points of the theory.

given of the kinetic theory of gases,

The Kinetic Theory

an elementary and lucid explanation

Clerk Maxwell's Theory of Heat^ Chapter

Introductory Mechanical Principles.

Mechanical work done by or upon a body during a

12. Characteristic equation of a

General equations of the effects produced by heat

Heat a form of energy

Mechanical equivalent of heat

Him 's experiment

27. Constitution of bodies

GENERAL EQUATION OF EQUIVALENCE.

36. Closed cycles

37. Reversible engines

engines of finite range performing

ABSOLUTE SCALES OF TEMPERATURE.

Second absolute scale of temperature

44. Intersection of thermal lines

SECOND LAW OF THERMODYNAMICS.

Connection between the two laws of thermodynamics

NON - REVERSIBLE PROCESSES.

51. Non-reversible cycles

Values of the Specific Heats

55. Difference of the Elasticities

6 1. General determination of intrinsic energy and entropy

63. Application to perfect gases

64. Sensible heat

65. Real specific heat

66. Available energy

68. Possibility of restoring lost availability

Comparison of Theory with Experiment.

heat of gases at constant volume

By adiabatic expansion or compression

APPLICATION TO SOLIDS AND LIQUIDS.

77. Application to other substances

83. Application of the First

82. General equation

85. Determination of the

86. Determination of the Intrinsic

87. Latent heat

Expansion along an Adiabatic

APPLICATION TO THE TEMPERATURES OF FUSION.