TIGHT BINDING BOOK
CD
00 ^
OU 160111 >m
OSMANIA UNIVERSITY LIBRARY
Call No. /2_'2> ,. ) Accession No.- / ^
*\l
Author
ucs SeKCjz*,, H
This book should be returned on or before the date
last marked below.
OSMANIA UNIVERSITY LIBRARY
WORLDS WITHOUT END
BY
H. SPENCER JONES
M.A., Sc.D., F.R.S.
ASTRONOMER ROYAL
HONORARY FELLOW, JESUS COLLEGE, CAMBRIDGE
Were a star quenched on high,
For ages would its light,
Still travelling downwards from the sky,
Shine on our mortal sight.
LONGFELLOW.
ii
THE ENGLISH UNIVERSITIES PRESS LTD
--trui *
LONDON
ALL RIGHTS RESERVED
FIRST PRINTED 1935
Reprinted December 1935
Printed in Great Britain by
Hazell, Watton 6- Viney, Ltd., London and Aylttbwy
TO
MY FATHER
CONTENTS
PAGE
PREFACE ...... xiii
CHAPTER
I. THE EARTH OUR HOME ... I
II. OUR NEAREST NEIGHBOUR THE MOON . 21
III. THE SUN'S FAMILY OF PLANETS . . 36
IV. LIFE IN OTHER WORLDS . . .82
V. COMETS AND SHOOTING STARS . . Q7
VI. THE NEAREST STAR THE SUN . . 113
r VII. GIANT AND DWARF STARS . . . 130
VIII. THE STARS OUR BLOOD RELATIONS . 15!
IX. TWIN STARS, PULSATING STARS AND NEW
STARS . . . . . .164
X. OUR STELLAR UNIVERSE . . . 1 82
XI. CELESTIAL CATHERINE-WHEELS . . 2OI
XII. THE AGE AND EVOLUTION OF THE STARS 215
XIII. WHAT WAS WHAT IS TO BE . . 239
INDEX 257
vii
LIST OF ILLUSTRATIONS
PLATE
I. THE ROYAL OBSERVATORY, GREENWICH
Frontispiece
FACING PAGE
II. THE MOON, AGED l6J DAYS . . . 22
III. PORTION OF MOON, INCLUDING THE SEA OF
SERENITY . . . . .26
IV. PORTION OF SOUTHERN HALF OF THE MOON 27
V. THE RANGE OF LUNAR MOUNTAINS THE
APENNINES ..... 30
VI. THE RING CRATER, COPERNICUS . -31
VII. (d) PHOTOGRAPHS OF MARS AND TERRESTRIAL
LANDSCAPE, WITH ULTRA-VIOLET AND
INFRA-RED LIGHT . . . -52
[b] DRAWING OF MARS, BY LOWELL . 52
VIII. PHOTOGRAPHS OF MARS IN LIGHT OF
DIFFERENT COLOURS .... 53
IX. (a) PHOTOGRAPHS OF JUPITER IN ULTRA-
VIOLET AND INFRA-RED LIGHT . . 66
(b) DIFFERENT ASPECTS OF JUPITER . 66
X. (a) SATURN WITH ITS SYSTEM OF RINGS . 67
(b) DIFFERENT ASPECTS OF THE RINGS OF
SATURN ...... 67
XI. (a) THREE MINOR PLANET TRAILS . . 78
(b) PHOTOGRAPHS OF PLUTO ... 78
xii. (a) HALLEY'S COMET IN 1910 . . . 102
(b) BROOKS'S COMET (igil) . . . IO2
xm. (a) MOREHOUSE'S COMET (1908) . -103
(b) BROOKS'S COMET (1903) . . -103
XIV. (d) LARGE SUN-SPOT, JULY 31, 1906 . 114
(b) LARGE SUN-SPOT, JANUARY 2O, 1926 . 114
ix
X LIST OF ILLUSTRATIONS
PLATE FACING PACK
XV. MOTION OF GROUP OF SPOTS ACROSS THE SUN 1 15
XVI. SUCCESSIVE STAGES IN THE DISSIPATION OF
A SOLAR PROMINENCE . . Il8
XVII. THE SUN PHOTOGRAPHED IN THE LIGHT OF
CALCIUM AND OF HYDROGEN . Iig
XVIII. MOTION OF A PROMINENCE FROM THE LIMB
OF THE SUN ON TO THE DISC . . 122
XIX. PHOTOGRAPHS OF SUN AT SUN-SPOT MINIMUM
AND AT SUN-SPOT MAXIMUM. . -123
XX. (a) SOLAR CORONA, MAY Q, IQ2Q . .126
(b) SOLAR CORONA, MAY 28, igOO . .126
XXI. THE CONSTELLATION OF CARINA, PHOTO-
GRAPHED WITH EXPOSURES OF I HOUR
AND 24 HOURS . . . . -133
XXII. (a) PHOTOGRAPHS OF KRUGER 60, SHOWING
ORBITAL MOTION . . . -158
(b) SPECTRUM OF SUNLIGHT AT TOTAL EC-
LIPSE, AUGUST 31, 1932 . . . 158
(c) PORTION OF SPECTRA OF SUN AND OF
ALPHA CENTAURI . . . -158
(d) SPECTRUM OF ZETA URS^ MAJORIS AT
TWO DIFFERENT DATES . . 158
XXIII. THE MILKY WAY IN THE CONSTELLATION OF
AQUILA . . . . . -177
XXIV. THE MILKY WAY IN THE CONSTELLATION OF
SAGITTARIUS . . . . 1 82
XXV. THE MILKY WAY IN THE CONSTELLATIONS OF
SCORPIO AND OPHIUCHUS . . .183
XXVI. (a) THE GREAT NEBULA IN ORION . . 1 88
(b) DIFFUSE NEBULA IN CYGNUS . . 1 88
XXVII. (a) ABSORBING CLOUD IN OPHIUCHUS. . 1 89
(b) ABSORBING CLOUD IN AQUILA . .189
LIST OF ILLUSTRATIONS XI
PLATS FACING PACK
XXVIII. NEBULOUS REGION IN OPHIUGHUS . IQ2
XXIX. (d) THE GLOBULAR CLUSTER, OMEGA CEN-
TAURI . . . . . 193
(b] SPIRAL NEBULA IN BERENICE'S HAIR . 193
XXX. DRAWINGS AND PHOTOGRAPH OF THE
" WHIRLPOOL " SPIRAL NEBULA . . 202
XXXI. (a) THE GREAT SPIRAL NEBULA IN ANDRO-
MEDA ...... 203
(b) ENLARGEMENT OF SOUTHERN PORTION
OF ANDROMEDA NEBULA . . . 203
XXXH. (a) THE SPIRAL NEBULA, MESSIER 8 1, IN THE
GREAT BEAR ..... 242
(i) THE SPIRAL NEBULA, MESSIER 10 1, IN
THE GREAT BEAR .... 242
LIST OF ILLUSTRATIONS xi
PLATE FACING PAGE
XXVIII. NEBULOUS REGION IN OPHIUCHUS . . IQ2
XXIX. (a) THE GLOBULAR CLUSTER, OMEGA CEN-
TAURI . J 93
(b) SPIRAL NEBULA IN BERENICE'S HAIR . 193
XXX. DRAWINGS AND PHOTOGRAPH OF THE
" WHIRLPOOL " SPIRAL NEBULA . . 2O2
XXXI. (a) THE GREAT SPIRAL NEBULA IN ANDRO-
MEDA - 203
(b) ENLARGEMENT OF SOUTHERN PORTION
OF ANDROMEDA NEBULA . . . 2O3
XXXH. (a) THE SPIRAL NEBULA, MESSIER 8 1, IN THE
GREAT BEAR ..... 242
(b) THE SPIRAL NEBULA, MESSIER IOI, IN
THE GREAT BEAR .... 242
PREFACE
IN this book I have endeavoured to give a picture of
the Universe and of the place that the Earth occu-
pies in it, as revealed by astronomical observation.
Being intended primarily for the general reader,
the use of technical terms has been avoided and no
account of the instruments and methods of observa-
tion has been included, though a general indication
has been given of the manner in which some of the
results have been obtained. Improvements in the
design and construction of telescopes and auxiliary
instruments, new methods of observation and the
continued refinement of technique have each played
an important part in obtaining the results which are
described.
The plan that has been followed is to start at
home, with our Earth and its satellite, the Moon,
and then to proceed outwards into space. After a
descriptive account of the members of the solar
system, we look at other stars to find how one differs
from another and to what extent the Sun can be
regarded as a typical star. We then pass first to an
account of what has been learnt about the stellar
universe to which we belong, and finally to consider
the other more or less similar universes with which
space is populated.
XIV PREFACE
These results of observation can be accepted as
reasonably well established, though undoubtedly
future observations will necessitate some changes
being made in the details of the picture here
presented.
The last two chapters are more speculative than
those which precede them. The theoretical as-
tronomer uses the material which has been pro-
vided by observation; from what is seen he attempts
to infer what is unseen, from the present to infer the
past and the future. These researches have proved
extraordinarily fruitful in many directions. It is
with some justification, for instance, that we can
claim to know more about the interior of a star than
about the interior of the Earth. But it is necessary
to emphasise that in many respects his conclusions
are neither final nor necessarily correct.
The originals of the photographs illustrating the
book were made at many different observatories
and by many different astronomers. The sources
of the plates are as follows: Royal Observatory,
Cape of Good Hope, xxi, XXIKT, xxmz; Lowell
Observatory, Flagstaff, Arizona, ixb, xib; Royal
Observatory, Greenwich, i, xm<2, xrwz and b,
xv, xxn; Helwan Observatory, Egypt, xmz
and b\ Heidelberg Observatory, Germany, xm;
PREFACE XV
Kodaikanal Observatory, India, xvi, xix; Lick
Observatory, Mount Hamilton, California, vna,
vma and b, ix#, xinb ; Meudon Observatory, France,
xvn, xvin ; Mount Wilson Observatory, California,
iv, xxviitf and , xxixi, xxx, xxxi, xxxn; Paris
Observatory, n; Sproul Observatory, Swarthmore,
Pennsylvania, xx#; Yerkes Observatory, Williams
Bay, Wisconsin, in, xxna, xxnrf, xxvia and b, xxxia;
Dr. F. E. Ross, xxm, xxiv, xxv; the late Prof. E.
E. Barnard, x<2, xx, xxvin. Plates v and vi are
from The Moon, by Nasmyth and Carpenter.
I am much indebted to the Directors of the
several observatories, and to Dr. F. E. Ross, for their
kind permission to reproduce these photographs.
H. SPENCER JONES.
ROYAL OBSERVATORY,
GREENWICH,
September izth y 1935.
CHAPTER I
THE EARTH OUR HOME
THE Greek astronomers and their followers believed
the Earth to be the centre of the Universe. Around
it moved the Sun, the Moon, the planets and the
stars, which were supposed to be carried on the sur-
faces of perfectly transparent crystal spheres. The
Greeks believed the order of increasing distance from
the Earth to be the Moon, Mercury, Venus, the Sun,
Mars, Jupiter and Saturn. The stars were supposed
to be attached to an outer crystal sphere. They as-
sumed that each of these spheres turned about a com-
mon axis in a period of one day, from east to west.
In this way they were able to account for the daily
rising and setting of the heavenly bodies. The dawn
of modern astronomy came in the sixteenth century
when the Polish astronomer Copernicus put forward j
in the year 1543 the revolutionary theory that in-
stead of the Sun moving round the Earth, it was the
Earth which moved round the Sun and that the
daily rising and setting of the heavenly bodies was
to be explained by the rotation of the Earth about
its axis in a period of one day. The theory of Co-
pernicus met with much opposition; it was seen that
if it was accepted the Earth must be displaced from
its proud position at the centre of the Universe. To
an age which was still bound by the views put for-
ward many centuries before by the Greek philoso-
phers this was a real difficulty. It was at least as
late as the time of Galileo, in the seventeenth cen-
2 WORLDS WITHOUT END
tury, when confidence in the old ideas had been
shaken by his discoveries of the satellites of Jupiter
and by his investigations of the spots on the surface
of the Sun, before the theory of Copernicus began
to be widely accepted.
But gradually the new ideas prevailed. Yet it
was not until the year 1851 that the rotation of the
Earth was actually demonstrated by means of a
famous experiment made by the French physicist
Foucault. He suspended a heavy iron ball from
the dome of the Pantheon in Paris by a wire more
than 200 feet in length. The ball was set slowly
swinging to and fro. A pin fixed to the lower side
of the ball marked the surface of a tray of sand on
the floor beneath the pendulum and indicated the
direction in which the pendulum was swinging. If
the Earth were not rotating, the pin would continue
to trace the same line on the surface of the sand.
But if the Earth is rotating, though the pendulum
would continue to swing backwards and forwards
parallel to the same plane, the tray of sand would be
slowly carried round beneath it. The trace marked
on the sand would then gradually change its direc-
tion. This was just what Foucault found. The ex-
periment has often been repeated, and it can be used
to demonstrate to a large audience the rotation of
the Earth in the course of a few minutes.
The Earth makes one complete rotation about
its axis in the course of a day. The day provides
our fundamental unit of time, and we subdivide
it for convenience into smaller units hours, min-
utes and seconds. For the purposes of everyday
life, the Earth serves as the natural timekeeper, and
THE EARTH OUR HOME 3
by means of clocks we keep count of the subdivision
into hours, minutes and seconds. The astronomer
can just as easily use the Sun, the Moon or one of
the planets as a clock. When he does so, he finds
that there are small discordances between the time
as derived from the rotation of the Earth and from,
say, the motion of the Moon. Whether he uses the
Moon, the Sun, Mercury or Venus as his clock, in
every instance the discordances from the clock pro-
vided by the rotation of the Earth are the same. If,
out of five clocks, four agree in showing the same
time whilst the fifth shows a different time, the
chances are very great that the fifth clock is a bad
timekeeper. For such reasons, it is concluded that
the Earth is not a perfect timekeeper and that the
length of the day is slightly variable. From the
records of ancient observations of eclipses of the Sun
and the Moon, it has been concluded that during
the last two thousand years the day has been gradu-
ally getting longer; the average increase in the
length of the day in the course of a century is about
one-thousandth of a second. In addition to this
gradual but progressive increase in the length of the
day, there are also irregular changes which some-
times occur with great abruptness. Thus in 1785
the rotation slowed down and in 1899 it speeded up
again. During the interval from 1785 to 1899, the
cumulative effect of the slowing down amounted to
nearly one minute. The effect of the irregular vari-
ations in the rate of rotation may amount in the
course of a year to about one second ; the correspond-
ing change in the length of the day is about three-
thousandths of a second. Quantities of this smallness
4 WORLDS WITHOUT END
are at present beyond, but only just beyond, the
possibility of detection by the most accurate clocks
which have yet been made. Hitherto it has been
necessary for the astronomer to check his clocks by
means of the rotation of the Earth; he now looks
forward to the time when, as the result of still further
improvement in precision clocks, he will be able to
use his clocks to check the constancy of the rotation
of the Earth.
The rotation of the Earth once in the course of a
day may not at first sight seem rapid. Yet the ro-
tation causes every point on the equator to travel
at a speed exceeding 1,000 miles an hour, or through
a distance of one mile in every 3^ seconds. It is
therefore not surprising that the rotation has an im-
portant influence on the great circulatory move-
ments in the atmosphere and in the oceans. In a
cyclonic disturbance or " depression " in the north-
ern hemisphere, the air streams spirally inwards in
an anti-clockwise direction towards the centre of the
depression. In a depression in the southern hemi-
sphere the air streams inwards in a clockwise direc-
tion. These motions, opposite in the two hemi-
spheres, are caused by the rotation of the Earth.
The north-east trade-winds in the northern hemi-
sphere, the south-east trade-winds in the southern
hemisphere as well as the anti-trades in the upper
atmosphere, flowing from the south-west in the
northern hemisphere and from the north-west in
the southern, are all controlled by the rotation of
the Earth.
The size and shape of the Earth are determined
by surveying operations. The measurement of the
THE EARTH OUR HOME 5
distance of any heavenly body is based ultimately
upon a knowledge of the distance apart of two
points on the surface of the Earth. Such knowledge
is therefore of fundamental importance for the as-
tronomer. The survey work depends upon the ac-
curate measurement of an initial base-line; from
each end of this base-line a distantwell-defined mark,
such as a church spire or a specially constructed
beacon, is observed with a theodolite and its distance
computed. From this mark and from one end of
the base-line another mark is observed, and so on.
By thus building up a chain of triangles the survey
is extended, every distance which is determined de-
pending upon the length of the initial base-line.
If the survey extends over a large area, it is usual
to have several base-lines in different parts of the
area, to obtain increased accuracy.
By such observations extended over large areas of
land in various parts of the world, it is found that
the Earth is approximately spherical in shape but
somewhat flattened at the poles. It is therefore
shaped somewhat like a Tangerine orange. The
radius measured from the north or the south pole
is about 3,950 miles; in the plane of the equator the
Radius is about 3,963 miles, a difference of 13
miles. From these dimensions it follows that the
area of the surface is nearly 200 million square miles.
The mass of the Earth is represented in tons
by 6 followed by twenty-one zeros, or is 6,000
millions of millions of millions of tons. The deter-
mination of the mass of the Earth is often spoken of
as " Weighing the Earth." Such an expression is
misleading. The weight of anything on the Earth
6 WORLDS WITHOUT END
is the force with which the Earth attracts it. If
there were no such force as gravity, the Earth would
not attract it and it would therefore have no weight.
The same body on the Sun would weigh about
twenty-eight times as much as on the Earth, because
the Sun attracts with a much greater force than the
Earth. It is the quantity of matter in the Earth,
or in other words its mass, which we really find when
we weigh the Earth.
All the methods which have been used to weigh
the Earth depend upon Newton's law of gravitation.
According to this law, any two pieces of matter
attract each other with a force which is proportional
to the product of their masses and inversely pro-
portional to the square of the distance between
them. If, for instance, the distance is halved, the
attractive force becomes four times as great; if the
mass of one of the bodies is doubled, the attractive
force is also doubled. It is the attraction of the
Earth on an apple which causes it to fall to the
ground ; it is the attraction of the Sun on the Earth
which causes it to move around the Sun instead of
flying away into space ; it is the attraction of the
Moon on the waters of the oceans, tending to heap
them up, which is responsible for the tides ; it is the
attraction of the Earth on the air which prevents the
atmosphere from rapidly dissipating away into
space. If a body is weighed with a pair of scales
and a large massive lump of lead is then placed
under the pan of the scales, the weight needed to
balance the scales will be slightly increased because
of the extra attraction of the mass of lead. The
increase in weight provides a means of comparing
THE EARTH OUR HOME 7
the mass of the lump of lead with the mass of the
Earth; the mass of the Earth can therefore be in-
ferred. This is the principle of the methods by
which the Earth is weighed.
The mean density of the Earth is found by divid-
ing the mass by the volume ; it is about 5^ times the
density of water. In other words, the weight of the
Earth is 5^ times greater than it would be if the
Earth consisted entirely of water. The surface
rocks are only about three times denser than water.
Volume for volume, therefore, the interior material
is considerably heavier than the surface material.
The higher density of the interior is no doubt due
to some extent to the great pressure to which it is
subjected by the weight of the overlying rocks. But
this does not provide a complete explanation, and
it is believed that the inner portion of the Earth is
really composed of heavier materials than the
outer portion.
We cannot hope to obtain much direct informa-
tion about the interior of the Earth. The deepest
mine-shafts go down to a depth of several thousand
feet only equivalent to a mere scratch in the sur-
face. Near the surface the temperature increases
rapidly inwards, about i F. for every 200 feet. It
may be inferred that the interior is hot. Active
volcanoes and warm springs provide direct evidence
of this internal heat.
Some information about the nature of the interior
can be obtained from the study of earthquake waves.
An earthquake is the result of a sudden displacement
of a portion of the outer solid crust of the Earth,
usually occurring at some distance beneath the sur-
8 WORLDS WITHOUT END
face. Just as plucking a violin string sets it into
vibration, so the sudden displacement starts vibra-
tions which travel outwards in all directions from
the region of the disturbance. By means of sensi-
tive instruments, called seismographs, these vibra-
tions can be detected thousands of miles away from
the earthquake. Some of the vibrations or waves
travel around the surface of the Earth ; others travel
through the interior. The waves are of two kinds,
which have been called "push" and "shake" waves.
In the " push " waves the vibrations take place to
and fro along the direction in which the wave
travels ; when an organ-pipe is sounded, the vibra-
tions of the air in the pipe are of this type. In the
" shake " waves, the vibrations take place in the
direction perpendicular to that in which the wave
travels; the waves on the ocean are of this type.
The investigation of the way in which the various
waves travel across or through the Earth, based on
the records of many seismographs distributed over
the Earth's surface, is difficult and complicated.
From such investigations it is concluded that the
Earth has a hot liquid core. The radius of this
liquid core is about 2,000 miles and its density is
about equal to that of iron. It is believed that it is
mainly composed of metals, principally iron, with
probably a certain amount of nickel. Outside the
liquid core is the solid crust of heavy rocks with a
density about four times that of water. The lighter
surface layer, composed mainly of granitic rocks, is
estimated to extend to a depth of only 40 or 50 miles.
The flattened shape of the Earth, with the bulge
around the equator, has an important consequence.
THE EARTH OUR HOME 9
We have mentioned that it is the gravitational pull
of the Sun which keeps the Earth in its orbit and
prevents it flying away into space. This same
gravitational pull has another result; the Sun pulls
the portion of the bulge which is nearest to it more
strongly than the portion which is farthest from it.
The Earth can be regarded as a huge gyroscopic top
spinning in space. The axis of a spinning gyroscope
will always point in the same direction in space un-
less some disturbing force causes a change. The
gyroscopic compass is based on this fact; in whatever
direction the ship may turn, the compass continues
to point in the true north and south directions.
The unequal pull of the Sun on the Earth's opposite
bulges acts as such a disturbing force and causes the
pole of the heavens the point in the sky near the
Pole Star to which the axis of the Earth points to
" precess " or, in other words, to move slowly round
'in a circle. The pole moves completely round this
circle once every 25,800 years. As a consequence of
this precession, the Pole Star, which is now only
slightly more than one degree distant from the pole,
has not always been and will not always continue
to be near to the pole. About 13,000 years ago,
the Pole Star was about 47 distant from the true
pole, which was then not far from the bright star
Vega. At present the Pole Star is slowly getting
nearer to the true pole, but in a few hundred years'
time it will begin to move away from it again.
It is interesting to recall that this precession or
movement of the pole was discovered by the Greek
astronomer Hipparchus about 1256.0. Hipparchus
, determined the length of the year in two different
IO WORLDS WITHOUT END
ways. His first method was to set up a vertical
pole, called a gnomon, and to determine when the
noonday shadow had its shortest length. This
occurred at midsummer when the Sun was at its
highest in the sky. By making observations in two
successive years, the length of the year can be found,
though for increased accuracy it is desirable to con-
tinue the observations over a number of years. His
second method was to observe what were termed the
" heliacal risings " of stars. The heliacal rising
occurs when the star rises above the horizon exactly
at sunrise. By recording the dates on which the
heliacal risings occur in successive years, the length
of the year can be obtained. Hipparchus found that
the year given by the second method was about 20
minutes longer than that given by the first method,
and this discordance led him to the discovery of
precession.
The Moon also exerts a gravitational pull on the
Earth; this pull causes a " nutation " or wobbling
of the axis of the Earth about the mean position.
A complete wobble to and fro takes place in about
19 years. The nutation was discovered by Bradley,
as the result of a long series of observations com-
menced in the year 1725. The telescope used by
Bradley for the observations which led him to the
discovery of the wobbling of the axis of the Earth
is one of the treasured possessions of the Royal
Observatory, Greenwich.
The Earth is surrounded by an atmosphere of air.
At the surface of the Earth dry air consists of about
J 78 per cent, of nitrogen, 2 1 per cent, of oxygen,
! nearly i per cent, of argon and smaller amounts
THE EARTH OUR HOME II
of carbon dioxide, hydrogen and the rare gases
helium, neon, krypton and zenon. The presence of
argon was discovered by Rayleigh and Kamsay in
1894; the other rare gases which are present in
much smaller quantities were discovered by Ramsay
shortly afterwards. It is believed that the oxygen
in the atmosphere is due largely to the action of
vegetation. Oxygen is chemically a very active
element, eager to form compounds with other,
elements. The rusting of iron is due to the affinity
of iron for oxygen. Combustion is caused by the
combination of oxygen with carbon; our coal fires
would not burn if there were no oxygen in the air. ;
By these processes, and by many others, oxygen is
continually being abstracted from the atmosphere.
Under the action of sunlight the green cells in plants
and the leaves of trees absorb carbon dioxide ; tho,
carbon is utilised for building up the cells of the
plant and the oxygen is given out to the atmosphere.
Oxygen is used up by animal life in the process of
respiration and carbon dioxide is given out as a
waste product. Vegetation therefore performs a
valuable function in preventing the continued ac-
cumulation of the carbon dioxide and in replenishing
the oxygen in the atmosphere.
The composition of the atmosphere does not
change much up to a height of several miles. With-,
in this region, the temperature decreases with height
and vertical convection mixes the air and keeps
the composition nearly uniform. At great heights,
where this mixing does not occur, the atmosphere
probably consists mainly of helium, or of helium and
hydrogen, which are the two lightest elements.
12 WORLDS WITHOUT END
The density at such heights is, however, so low that
the total amount of these gases is small. Of con-
siderable interest to the astronomer is a layer at a
height of about 20 miles containing ozone. The
ozone, though equivalent to a layer at the surface
only one-eighth of an inch thick, absorbs strongly
light of short wave-length in the far ultra-violet
region of the spectrum. The light which reaches
us from the Sun and the stars is therefore deficient
in these wave-lengths. Though this is a handicap
to the astronomer, the ozone layer is beneficial to
mankind, for it protects us from the intense actinic
rays which are injurious to human beings.
If the atmosphere had the same density through-
out as at the surface, it would extend to a height of
about 5 miles only. But as the density decreases
rapidly with height, the atmosphere extends to a
considerable height above the Earth's surface.
Shooting stars enter the Earth's atmosphere from
outside, and many become visible at heights as
great as 120 miles. A shooting star must travel
for a considerable distance through the rarefied
upper atmosphere before it is sufficiently heated by
friction to become incandescent. It seems probable,
therefore, that the atmosphere extends to heights
considerably greater than 120 rniles.
Water-vapour plays a part of great importance
in the atmosphere. It is present only in the lower
layers, clouds rarely being found at heights greater
than about 6 miles. If the atmosphere contained
no water-vapour, there would be neither clouds,
dew, rain, hail, snow nor thunderstorms, and neither
plant nor animal life would be possible. The water-
THE EARTH OUR HOME 13
vapour in the atmosphere also plays a role of great
importance in controlling the temperature of the
Earth. The Earth is warmed by the heat radiation
which it receives from the Sun; but the Earth
also emits radiation into space. There is a general
balance between the amount of heat which the
Earth as a whole receives from the Sun and the
amount which it, in turn, sends out into space.
Actually it is not the surface of the Earth which is
mainly concerned in re-emitting the radiation re-
ceived from the Sun ; much the greater part of it is
emitted by the clouds and the water-vapour in the
atmosphere.
If the heat received from the Sun were to be in-
creased, the Earth's surface would at first become
warmer, the effect at the equator being greater than
at the poles. The increased difference in tempera-
ture between the equator and the poles would result
in a more vigorous circulation of the atmosphere and
therefore in increased windiness. As the combined
result of the higher temperature of the surface and
of the stronger winds, there would be greater evapo-
ration and an increase both in cloudiness and in
rainfall. The increased cloud would reflect back
into space a greater proportion of the radiation from
the Sun, and this would bring about a fall in tem-
perature at the surface. It is probable that the out-
put of radiation from the Sun is not absolutely con-
stant, and there is some evidence in support of the
view that the Earth becomes cooler when the Sun's
output of heat becomes greater. Thus we have the
paradox of a hot Sun and a cool Earth, due entirely
to the important part played by the water-vapour
14 WORLDS WITHOUT END
contained in the atmosphere. Another consequence
of an increase in cloudiness may be mentioned.
Cloudy days and cloudy summers are relatively cool,
because the clouds reflect back a part of the radia-
tion from the Sun, so that the amount which reaches
the surface of the Earth is reduced. Cloudy nights
and cloudy winters are relatively warm, because
the escape of radiation from the Earth's surface
into space is impeded. Therefore an increase in
solar radiation, by causing an increase in cloudiness,
tends to produce smaller extremes of temperature
between summer and winter and between day and
night.
The blue colour of the sky and the beautiful
colour effects often seen at sunrise and sunset are
due to the Earth possessing an atmosphere. In its
passage through the atmosphere some of the light
from the Sun is scattered in all directions by the
molecules of air, but the blue light is scattered to a
much greater extent than the red light. When the
Sun is near the horizon, its light reaches us through
a much greater thickness of atmosphere than when
it is higher in the heavens ; the blue light is gradually
scattered sideways, leaving the light which is much
richer in the colours of longer wave-length the
reds, oranges and yellows to reach us. The blue
colour of the sky arises from sunlight passing through
the upper layers of the atmosphere; the scattered
blue light is thrown sideways and reaches the ob-
server. If the Earth, like the Moon, were devoid of
an atmosphere, the sky would appear black and the
stars would be seen shining steadily in the sky in
broad daylight but never twinkling. What a para-
THE EARTH OUR HOME 15
disc for astronomers if life were possible under such
conditions !
The study of the past history of the Earth as re-
vealed by igneous and sedimentary rocks and by
fossils is the concern of the geologist rather than of
the astronomer. The astronomer is, nevertheless,
interested in any evidence which the geologist can
provide as to the age of the Earth. Any conclusions
which the astronomer may reach about the age of
the Sun and of the stars in general must not be in
direct conflict with the conclusions of the geologist
as to the age of the Earth. It seems certain that the
Earth and the other planets were formed from the
Sun; the age of the Sun must necessarily therefore be
greater than the age of the Earth. About 250 years
ago, a former Astronomer Royal, Edmond Halley,
suggested that the amount of salt in the ocean would
provide a means of estimating the age of the Earth,
though sufficient data for the purpose were not
available to him. Every river carries down to the
sea soluble salts, which have been dissolved from
the soil as the rain-water percolated through it, as
well as solid matter in suspension. The solid par-
ticles are deposited on the ocean bottom ; the soluble
matter which is mainly common salt remains in
the sea. Thus the oceans are gradually becoming
more and more salt. It has been estimated that
the total amount of sodium contained in all the
oceans of the globe is about 12,600 million million
tons and that the total amount added by the rivers
in the form of soluble salts is about 35 million tons
per year. If it could be assumed that the salinity
of the oceans has been increasing at a uniform rate
l6 WORLDS WITHOUT END
since they were first formed, we could conclude that
the age of the Earth which cannot be less than the
age of the oceans must be at least 360 million
years.
But the assumption is certainly not correct. Geo-
logists have shown that the Earth has experienced
a succession of stages of mountain building or uplift,
followed by periods of denudation during which the
rivers and streams have slowly but surely acted
as levelling agents, carrying solid matter from the
higher regions and depositing it on the lower. Just
after a period of uplift this action must have pro-
gressed at a much greater rate than near the end of
such a period. It is believed by geologists that the
present rate of denudation is considerably greater
than the average rate during past geological history.
The age of the oceans must therefore be consider-
ably greater than 360 million years. Another argu-
ment of a somewhat similar nature is based upon
the rate at which sedimentary deposits are laid down.
The geological evidence indicates that the total
thickness of sedimentary rocks which have been de-
posited since sedimentation first started is some-
where about 500,000 feet. Since the reign of
Rameses II, over three thousand years ago, sediment
has been deposited in Lower Egypt at the rate of
one foot every five hundred years. If this repre-
sented the average rate in the past, the age of
the oldest sedimentary rocks would be about 250
million years. But again this is believed to be a
considerable underestimate.
Such estimates, rough approximations as they
were, sufficed to prove that the estimate of the age
THE EARTH OUR HOME 1J
of the Sun, about 22 million years, made by Lord
Kelvin, was seriously in error and must be dis-
carded. But we do not need to depend now on such
uncertain methods, for in the phenomenon of radio-
activity physics has placed at our disposal a clock
whose rate has been constant throughout past geo-
logical history. The atoms of the heaviest known
elements, such as uranium, thorium and radium, are
unstable and gradually disintegrate; they pass
through a series of successive stages, and finally end
up as lead. During these processes electrically
charged atoms of helium are shot out from the com-
plex radioactive atoms with enormous velocities of
several thousands of miles per second. This radio-
active disintegration proceeds at a perfectly definite
rate : though it is not possible to say that any par-
ticular atom will disintegrate during the next day or
year or even during the next century, we can say
with certainty that in any given time a certain
fraction of the atoms will have disintegrated. The
premiums charged by life assurance companies are
based on tables of mortality compiled from statis-
tics of the mortality at different ages, and it can be
inferred with reasonable probability that out of
every 100 persons now living and, say, 44 years old,
so many will die next year, so many will survive for
30 years and so on. But the mortality tables need
revision from time to time to take account of the
increasing expectation of life consequent upon im-
proved social conditions, the advance in medical
knowledge and other causes. The mortality tables
of the radioactive atoms need no such revision.
Even though the sedimentary rocks may have ex-
l8 WORLDS WITHOUT END
perienced great changes in the temperature and in
the pressure to which they have been subjected
since they were first deposited on the floor of the
ocean, the rate of break-up of the atoms of uranium,
thorium or radium which they may contain pro-
ceeds at a rate which is perfectly indifferent to all
such changes. If we have some radium or radium
salt, half oftheatomsof radium will have disintegrated
by the end of 1,580 years; in the case of uranium,
4., 500 million years must elapse before half of the
atoms have disintegrated; whereas for thorium 2 2,000
million years are required.
Although in every case the ultimate product of
radioactive disintegration is lead, which is chemi-
cally indistinguishable from common lead, there is
a difference in the weight of the lead. The lead
produced by the break-up of uranium atoms, which
we may call uranium lead, is lighter than common
lead; thorium lead is heavier than common lead.
If, in any uranium mineral, lead with the same
weight as uranium lead is found to be present, it
may reasonably be inferred to have been produced
by the break-up of uranium atoms, and from the
relative amounts of lead and uranium present in the
mineral its age can be deduced with considerable
i accuracy. It has been mentioned that the disinte-
gration of the atoms is accompanied by the produc-
tion of helium ; if the helium gas is occluded within
the pores of the mineral, the age of the mineral can
also be inferred from the amount of helium present.
But as some of the helium may have escaped, the
age determined from the amount of helium gas in
the rock can only be regarded as a lower limit and
THE EARTH OUR HOME ig
as providing a general check on the age deduced
from the lead content.
An upper limit to the age of the Earth can be ob-
tained if it is assumed that the whole of the lead
contained in the igneous rocks was of radioactive
origin. It has been calculated that all the lead
present in these rocks could have been produced in
something over 3,000 million years. It is unlikely
that the age of the Earth can exceed this figure, for
the radioactive elements may have existed in the
Sun before the Earth was born and may there have
become partially transmuted into lead; so it is
possible that some of the lead in the igneous rocks
may not have been produced by radioactive dis-
integration since the Earth was formed.
It is satisfactory to find that the relative ages
which had been assigned to various rocks from geo-
logical evidence are confirmed by the direct deter-
minations of age by the radioactive method.
The oldest rocks whose ages have been definitely
determined are found in Manitoba and South
Dakota, the age in each instance being about 1,700
million years. To this period in the Earth's history
belong the primitive marine invertebrates. In
rocks with an age of about 900 million years are
found the oldest known fossils. Five hundred million
years ago the inYerlebrate animals were beginning to
appear. They were followed by the age of fishes. The
earliest land floras developed some 80 million years
later. Some 280 million years after they appeared,
the first coal measures were laid down. Then fol-
lowed the age of the dinosaurs and the great flying
reptiles. The primitive mammals did not appear
2O WORLDS WITHOUT END
in the slow progress of evolution until about 60
million years ago ; still more recently came the man-
apes, 8 million years ago, and in the comparatively
near past came man, who has existed on the Earth for
only about i million years. The mountain systems of
the Earth are of very varied ages. The Himalayas
were formed comparatively recently, only 8 million
years ago; the Alps are about 21 million years old,
the Pyrenees about 30 million years, the Rocky
Mountains 105 million years and the Appalachian
Mountains about 240 million years. It therefore
seems probable that the age of the Earth from the time
when the crust formed is of the order of two or three
thousand million years. This conclusion is sup-
ported by some ratherindirectastronomical evidence.
If we suppose the history of the Earth since it was
born to be written in a book, each page of which
has to deal with the history of 5 million years, some
four or five hundred pages would already have been
written. The history of the time which has elapsed
since mankind first appeared on the Earth must be
summarised in the last eight lines, and to the whole
of the Christian era, unless it is to receive more than
its proper proportion of the space, only one letter
could be given. How many more such volumes
would need to be written before conditions on the
Earth have changed to such an extent that man is
no longer able to exist only the future can reveal.
So far as any conclusions can be drawn from the
results of modern astronomical investigation, it
seems possible that even when 500 volumes have
been written the conditions on the Earth will not
differ greatly from what they are at present.
CHAPTER II
OUR NEAREST NEIGHBOUR
THE MOON
THE nearest neighbour to the Earth in space is the
Moon. Its distance can be determined without any
great difficulty. The method is to observe the posi-
tion of the Moon relative to the stars from each of
two observatories which are several thousand miles
apart; observations could be made, for example,
at the Greenwich and Cape of Good Hope Observa-
tories. The distance apart of these observatories is
known because the shape and dimensions of the
Earth have been determined. We have thus a base-
line of known length, and if the direction of the Moon
is observed at the same instant from the two ends of
the line, the distance of the Moon can be calculated,
as in ordinary surveying operations.
The mean distance of the Moon from the Earth
is about a quarter of a million miles; the distance
varies within a rather wide range of about 30,000
miles. If we set out to reach the Moon by means
of an aeroplane with a cruising speed of 100 miles
an hour, we should take about one hundred days
to reach it. A quicker method would be to be pro-
jected from the Earth with a sufficiently high speed
in a large rocket, as imagined by Jules Verne in his
story entitled A Voyage to the Moon. The rocket
will rapidly slow down because the gravitational
pull of the Earth is tending to draw it back the
whole time. If the rocket is fired with a speed of
21
22 WORLDS WITHOUT END
less than 7 miles per second, it will slow down until
it comes to a stop, and it will then begin to fall
back to Earth again. When the initial speed is 7
miles per second, the rocket is just able to escape
from the pull of the Earth and to get out into space.
This critical velocity is called the " velocity of
escape." It is about fifteen times the muzzle velocity
of a high-velocity shell. If we succeeded in pro-
jecting a rocket in the direction of the Moon with
this velocity it would be able to reach the Moon in
about 50 hours.
The Moon is a much smaller body than the Earth,
for the Earth could contain fifty bodies of the size
of the Moon. The discrepancy in weight is even
greater than in volume, because volume for volume
the Moon is not so heavy as the Earth ; actually, the
weight of the Earth is nearly 82 times that of the
Moon. The material of which the Moon is com-
posed is about 3^ times heavier than water: it is
therefore comparable in density with granite and
the other surface rocks on the Earth. We have
already seen that the outer crust of the Earth is
made of much lighter material than the interior.
The force of gravity on the Moon is only about
one-sixth of its value on the Earth. A high jumper
who can scale 6 feet on the Earth could with the
same expenditure of effort jump to a height of 36
feet on the Moon. A moderate golfer who cannot
drive the ball for more than 150 yards would be
delighted, if he were on the Moon, to find that he
was driving it for half a mile. Golf courses on the
Moon would need to be on a correspondingly longer
scale than on the Earth.
PLATE II.- THE MOON, AGED iGi DAYS (SHORTLY AFTER FULL MOON).
The large crater in the lower portion of the photograph, with the
radiating streaks, is Tycho, 54 miles across and 17,000 feet deep. It
has a central peak 5,000 feet in height.
22]
OUR NEAREST NEIGHBOUR THE MOON 23
Due to this low value of gravity on the Moon, the
velocity of escape from the Moon is only aboujLj^
miles a second. It is thus very much easier for a
fast moving body to escape from the Moon thar^
from the Earth. One consequence is that the Moonj
has entirely lost its atmosphere. The Earth's atmo-*
sphere consists of billions upon billions of mole-
cules of oxygen, nitrogen and other substances,
flying backwards and forwards with very high
speeds, continually colliding with one another and
rebounding in different directions. The lighter the
molecule, the faster it moves. At the temperature
of the freezing-point of water the average speed of
a molecule of hydrogen is about one mile a second,
whereas that of a molecule of oxygen is only about
J mile a second. Individual molecules attain speeds
much higher than the average, so that at the outer
limits of the Earth's atmosphere molecules of
hydrogen from time to time rebound outwards with
speeds that are greater than the velocity of escape
and never return. But a molecule of oxygen or
nitrogen will very rarely attain a speed about 30
times greater than its average speed, which it must
do before it can escape. Thus though the Earth has
been able to retain an atmosphere, the hydrogen,
Ehich was probably the chief constituent of the
;mosphere when the Earth was v^ry young, has
most entirely been lost. The velocity of escape
from the Moon, on the other hand, is so low that the
hydrogen originally present in its atmosphere must
have been lost very quickly, whilst the oxygen,
nitrogen and other constituents, though escaping
more slowly, have long since been completely lost,
24 WORLDS WITHOUT END
with the result that now the Moon has no atmo-
sphere at all.
We can convince ourselves in a very simple way
that the Moon has no atmosphere. If, on any clear
night, the position of the Moon relative to nearby
bright stars be noted, it will be seen in the course
of an hour or two that the Moon is changing its
position relative to the stars. The Moon appears
to move eastwards amongst the stars, and it is a
consequence of this eastward motion that the Moon
rises on an average 50 minutes later each night than
.the preceding night. This apparent motion is a
consequence of the motion of the Moon in its orbit
round the Earth. From time to time the Moon
may be seen to pass in front of a star, or to " occult "
it, as the astronomer terms it. When this happens,
the star is seen to disappear instantaneously. When
the Moon is near first quarter, the advancing eastern
limb is dark and cannot be seen; if the star about
to be occulted is viewed through a telescope, its
disappearance at the moment of occultation is
startling in its suddenness. If the Moon had any
atmosphere, the rays of light from the star would
be bent or refracted in passing through the Moon's
atmosphere; the nearer the rays approached the
edge of the Moon, the greater would this bending
be, and instead of disappearing suddenly the star
would gradually fade from view.
The Moon, having no atmosphere, must also be
entirely devoid of water. If there were water on the
Moon, evaporation would take place when the surface
of the Moon was heated by the Sun and the mole-
cules of water-vapour would gradually escape into
OUR NEAREST NEIGHBOUR THE MOON 25
space. This process would continue until there was
no water left. This conclusion is in agreement with
what we observe on the Moon with the aid of a tele-
scope. In a powerful telescope objects on the Moon
the size of St. Paul's Cathedral can be distinguished.
If there were seas, rivers, lakes or even large ponds,
we should frequently see them shining brightly when
they were suitably placed to reflect the sunlight in
the direction towards the Earth. Nothing of this
sort has ever been observed.
The Moon moves round the Earth under the
influence of the Earth's gravitation, one complete
revolution being made in rather less than one month.
The average speed of its motion is not much greater
than half a mile a second but little more than that
of a high-explosive shell. This is much less than
the speed of about 20 miles a second with which the
Earth moves in its orbit around the Sun. The Moon,
like the Earth, is also rotating. One complete rota-
tion on its axis takes exactly the same time as one
revolution in its orbit round the Earth; in other
words, the length of a day on the Moon, as mea-
sured from, say, noon to the next noon, is about one
month. It is for this reason that we always see the
same face of the Moon ; the other side of the Moon
is completely and for ever hidden from our sight.
The Moon does not shine by its own light but
reflects some of the light from the Sun which falls
upon it. The average reflecting power of the sur-
face of the Moon is low, for only one part in fourteen
of the sunlight falling on the Moon is reflected, the
remainder being absorbed. It can be seen with
the naked eye that some portions of the surface
26 WORLDS WITHOUT END
appear brighter than others (Plate II). Different
substances have different reflecting powers; chalk
and white sand, for instance, reflect much more
than dark-coloured rocks, such as granite or basalt.
The brightest portions of the Moon have a reflect-
ing power about equal to that of white sand, but
the average for the whole surface corresponds to
that of darkish rocks. It may therefore be con-
cluded that the greater portion of the surface of the
Moon is of a brownish jcplour.
At any time only one half of the surface of the Moon
can be in sunlight, the other half being in darkness.
The amount of sunlit surface which we can see
depends upon the relative positions of the Earth,
Moon and Sun. When it is Full Moon, the Sun,
Earth and Moon are practically in a line, with the
Earth between the Sun and the Moon; the whole
of the bright surface of the Moon then faces the
Earth. When it is New Moon, the three bodies are
again in a line, but the Moon is then between the Sun
and the Earth, with its dark side towards the Earth.
At other times facing the Earth is a portion of the
! bright side of the Moon and a portion of the dark
side. Thus we have the varying appearance of
the phases of the Moon. Just after New Moon,
when the Moon is seen as a narrow bright crescent,
>the remainder of the disc of the Moon may be dimly
seen; this appearance is known as " the Old Moon
in the arms of the New." The cause of this appear-
ance will be readily understood if we remember
that the Moon is then between the Sun and the
Earth and nearly in a line with them. The dark
face of the Moon is turned towards the Earth, but
PLATE III. PORTION OF MOON, INCLUDING THE SEA OF SERENITY (THE
UPPER DARK AREA), AND THE SEA OF TRANQUILLITY (THE LOWER
DARK AREA).
Note the line of'lava flow. The large crater at the right-hand edge is
Posidonius, 62 miles in diameter.
PLATE IV. PORTION OF THE SOUTHERN HALF or THE MOON, SHOWING
NUMEROUS CRATERS, SOME WITH AND OTHERS WITHOUT CENTRAL PEAKS.
The dark area is called the Sea of Clouds.
OUR NEAREST NEIGHBOUR THE MOON 27
the bright side of the Earth (that is, the side which
faces the Sun) is towards the Moon. The Earth
reflects some of the sunlight falling upon it back
towards the Moon and the face of the Moon becomes
dimly visible by means of this " Earth-light."
We have stated above that at the times when the
Moon is new or full, the Sun, Earth and Moon are
in a straight line. This is not strictly correct, for
in general the Moon will be either above or below
the plane of the orbit which the Earth describes
round the Sun. When the three bodies are actu-
ally in line, we observe an eclipse. If this occurs
at Full Moon, the Earth is between the Sun and
the Moon, and cuts off the sunlight from the Moon,
we then have an eclipse of the Moon, and this is
visible over the whole hemisphere of the Earth
facing the Moon. The Moon does not become in-
visible when totally eclipsed in this way but appears
of a dull copper colour ; this is due to rays of light
from the Sun being refracted or bent in their passage
through the atmosphere of the Earth so that some
are thrown on to the Moon's surface.
If it is New Moon when the three bodies are in
a line, the Moon is between the Sun and the Earth
and we have an eclipse of the Sun. Though the
Moon is very much smaller than the Sun, it is so
much nearer that it is possible, but only just possible,
for the Moon completely to obscure the Sun. When
this happens we have a total eclipse of the Sun.
If, at the time of eclipse, the Moon happens to be
at its greatest distance from the Earth, it is unable
completely to obscure the Sun. We then see, at the
time of mid-eclipse, a narrow ring of sunlight all
28 WORLDS WITHOUT END
round the Moon; this is called an annular eclipse.
A total or an annular eclipse is visible only over a small
portion of the surface of the Earth; over a much
larger area, a partial eclipse, when a part only of
the Sun is obscured, will be seen. We learn there-
fore that an eclipse of the Moon can only happen
at Full Moon and an eclipse of the Sun can only
happen at New Moon. It can be shown that there
must be at least two eclipses (partial or total) in any
year; if there are only two, they must both be
eclipses of the Sun. There were only two eclipses
in 1 933 ; there will be only two in 1 940. There can-
not be more than seven eclipses in a year, of which
there may be five of the Sun and two of the Moon
(as in the year 1935) or four of the Sun and three
of the Moon (as in 1917). Though eclipses of the
Sun are more frequent than eclipses of the Moon,
eclipses of the Moon are more commonly seen,
because an eclipse of the Moon is visible from half
of the surface of the Earth, but an eclipse of the Sun
is visible from only a limited area.
If by means of our rocket we were able to land on
the Moon, we should find a world very different
from our Earth. There are no oceans or rivers or
water of any sort, nor are there any trees or other
vegetation. There are no signs of any habitation.
Complete stillness reigns everywhere; no breezes
blow, for there is no atmosphere. There is no dust
or haze to impair the visibility. Not a sound is heardj
to disturb this unnatural stillness. The Sun and
stars appear much bluer than they do to us on
Earth, and the stars shine brightly in broad daylight
in a sky of inky blackness. We should notice that
OUR NEAREST NEIGHBOUR THE MOON 2Q
they shine steadily without a twinkle. The shadows
cast by the Sun are black, with sharp and hard out-
lines. Large areas of the surface are flat desert
expanses ; but most of the surface is extremely moun-
tainous. The mountain walls are for the most part
sharp and jagged, for the weathering action of wind
and rain and the eroding effect of streams and
rivers have been absent. (Note the shadows of the
mountains in Plate V.) During the long lunar day,
equal to about 14 of our days, the Sun strikes down
with pitiless fury; at noonday on the equator the
rocks are heated to a temperature well above the
boiling-point of water. But as soon as noon has
passed the temperature begins to fall, for there is no
water-vapour to act as a blanket in retaining the
heat. Towards evening the fall is very rapid. By
sunset the temperature is well below the freezing-
point of water. Sunsets on the Moon are strangely
different from those to which we are accustomed
on the Earth. We should never see the Sun shining
red through clouds or haze as it approaches the
horizon, with the clouds reflecting its glory. Until
the Sun finally disappeared beneath the horizon it
would shine with the same steely blue light and
then total darkness would fall with bewildering sud-
denness. Of twilight there would be none. Even
in the daytime, in the mountain valleys sheltered
from the Sun, it would be pitch-dark. During the
long cold night the temperature falls far below any-
thing experienced in the coldest regions of the
Earth's surface, to about 150 F. below freezing-
point. The temperature would be even lower were
it not that the rocks had been heated to such a great
30 WORLDS WITHOUT END
extent during the daytime. The Moon is certainly
not a place to be recommended for a pleasure trip.
We have mentioned that the Moon is an extremely
mountainous body. There are mountain ranges on
the Moon which appear very similar to mountain
ranges on the Earth. One of the ranges, called the
Apennine Range, is shown in Plate V. But the
mountain ranges are few in number. Most of the
mountainous formations on the Moon are in the
form of mountain rings of Various sizes. The largest
of these have diameters exceeding 100 miles and
could contain within them two or three English
counties. There are others of all sizes down to
small ones which are only visible in very powerful
telescopes. The majority of these craters, as they are
called, are from 5 to 20 miles in diameter. There
are at least 30,000 of them on the side of the Moon
which faces us. The typical crater is practically
circular; the mountain ring wall rises to a height of
anything up to 20,000 feet above the surrounding
country. Yet if we were to stand near the centre of
one of the larger craters, no trace of this high en-
circling mountain wall would be visible. The
mountain wall is usually much more precipitous on
its inner slopes than on its outer slopes. The floor
within the ring is sometimes above and sometimes
below the level of the surrounding surface. At the
centre of the crater there is often, but not invariably,
a mountain peak or a group of peaks. In Plates III
and IV may be seen examples of craters both with
and without central peaks. On some portions of
the surface the craters are extremely numerous.
Such a portion is shown in Plate IV. It will be
PLATE V. THE RANGE OF LUNAR MOUNTAINS, CALLED THE APENNINES.
The range extends for about 450 miles and contains upwards of 3,000 peaks.
The highest peak, Mount Huyghens, is 21,000 feet high. The large crater is
called Archimedes. It has a diameter of 52 miles : the crater wall is at an
average height of about 4,300 feet above the surrounding plateau. There is
no centra] peak.
30]
m * JR
PLATE VI. THE RING CRATER, COPERNICUS, is ONE OF THE MOST IMPRESSIVE
CRATERS ON THE MOON.
Its diameter is 46 miles. Within it is a group of craters, three of which are
upwards of 2,400 feet high. The ring wall rises to 12,000 feet above the level
of the plateau.
OUR NEAREST NEIGHBOUR THE MOON 31
noticed that small craters are often to be found with-
in larger ones and that occasionally a newer crater
has encroached upon the ring wall of an older crater.
A typical crater, Copernicus, is shown in Plate VI.
Travel of any sort, other than by aeroplane, over
such portions of the surface of the Moon would be
extremely arduous and railway construction would
tax the ingenuity of the most skilled engineers.
The term crater and the general appearance of
these formations suggest a volcanic origin. The
name is unfortunate because it is misleading, for we
cannot be certain that the craters have been pro-
duced by volcanic action. It is difficult to believe
that a crater of i mile in diameter and another
crater exceeding 100 miles in diameter can both
have been produced by volcanic action. As there
are craters of all sizes between the smallest and
the largest, we are compelled to believe that they
have all had a common mode of formation. We
might argue that as there are no volcanic formations
on the Earth which at all resemble the craters on
the Moon, the lunar craters cannot have been formed
by volcanic action. This, however, is an argument
which can be used against any suggested explana-
tion of their formation. We must remember also
that the effects of erosion and denudation from
water, wind and blown sand have entirely changed
the appearance of terrestrial formations, and we can-
not assert that the Earth may not also when it was
young have had crater formations like those on the
Moon.
Another explanation which has been suggested
supposes that soon after the Moon was born and a
32 WORLDS WITHOUT END
solid crust had commenced to form, the generation
of gas or steam beneath the crust caused it to blow
out into vast bubbles at its weakest points. The
bubbles expanded as the pressure increased and at
last burst. The central portion of the bubble fell
back into the liquid interior and became molten
again, but all round the opening the circular rim
was left, forming the mountain wall. The crust
formed again inside the ring. The same process
may have occurred many times over ; once a
bubble had burst, the newly formed crust within it
would be relatively weak and a further bursting at
the same point would be liable to occur.
The extensive dark areas on the Moon have from
the time of Galileo been called maria or seas ; when
this name was given it was believed that they were
actually expanses of water. The name has been
retained, though it has long since been realised that
they are not seas. Plate III shows the Mare
Serenitatis and the Mare Tranquillitatis. It is pos-
sible that as the Moon cooled large areas of the crust
subsided and were overrun by molten lava. Some
of the seas show irregular wave-like markings which
appear to mark the limits of successive outflows of
lava. Such a marking across the Sea of Serenity is
to be seen in Plate III.
Other interesting features can be seen on the
Moon's surface. There are many deep, narrow
valleys which are called rills. More interesting are
the clefts which run for hundreds of miles across
mountains, valleys and plains. They are perfectly
straight and about half a mile in width, and are
evidently vast crevasses of unknown depth produced
OUR NEAREST NEIGHBOUR THE MOON 33
by the cracking of the Moon's surface. They would
offer serious obstacles to any lunar explorer.
Most interesting and puzzling, perhaps, of all
the lunar formations are the bright streaks which
radiate from some of the larger craters. The system
of rays from the large crater Tycho is shown in
Plate II. These streaks, which in a photograph
give the Moon somewhat the appearance of a peeled
orange, are best seen at the time of full Moon; from
5 to 10 miles in breadth, they may stretch for many
hundreds of miles, crossing mountains, valleys and
other craters without any change in width or colour.
They cast no shadows and cannot therefore be much
elevated above the surface. They have the appear-
ance of deposits of fine volcanic ash, but why such
deposits should occur in the form of straight streaks
and not be uniformly spread around the parent
crater is a mystery. Another suggestion is that they
mark the tracks of fine rifts which are too narrow to
be visible and that the surface has been stained by
vapours which have arisen from these rifts.
It will be evident that the various problems of
lunar topography present a fertile subject for specu-
lation, but, since geological evidence for or against
any theory cannot be obtained, we can do no more
than form our own conclusions as to the probability
or improbability of the various theories which have
been put forward.
The Moon is of great economic importance to
mankind because it is largely responsible for the .
(ocean tides. The tides are due to a regular rise and
fall of the water, generally twice a day. This rise
and fall is due to the gravitational attraction of the
34 WORLDS WITHOUT END
Moon and the Sun. The Moon attracts the ocean
directly beneath it more powerfully than it attracts
the solid Earth and tends to pull the two apart, with
the result that the water becomes heaped up beneath
the Moon, On the far side, the Moon attracts the
ocean less powerfully than the solid Earth and again
tends to pull the two apart, causing another heaping
up of the water. As the Earth rotates, the oceans
move over its surface, the heaping up of the waters
tending to occur at the point beneath the Moon and
at the point diametrically opposite. The movement
of the oceans is impeded by various causes and is
affected by the local land configurations, so that the
actual phenomena are more complicated. The
Sun also exerts a tide-raising force, but the effect of
its greater distance more than counterbalances the
effect of its greater mass, with the result that the tide-
raising force of the Sun is less than half that of the
Moon. Bearing these facts in mind, some of the
main phenomena shown by the tides can be simply
explained. At any given place, the height of the
tide varies considerably ; the rise and fall of the tide
is greatest at spring tides and lowest at neap tides.
The spring tides occur when the tide-raising forces
of the Sun and Moon are acting together ; this hap-
pens when the Sun, Moon and Earth are in a line, in
other words, at New Moon or Full Moon. When the
Sun and Moon are in opposing positions, i.e. near first
and last quarters of the Moon, we have neap tides.
The retardation from day to day in the times of high
and low tides is well known ; this is a consequence of
the retardation from one day to the next in the time
of rising of the Moon, since the heaped-up waters
OUR NEAREST NEIGHBOUR THE MOON 35
follow the motion of the Moon. Spring tides at the
same place are not always of the same height; a
variation is caused by the change in the distance of
the Moon from the Earth, in consequence of the
Moon's orbit being not circular but elliptical. The
tidal effect due to the Moon is at its greatest when
the Moon is at its nearest to the Earth, the range in
the tides due to the change in distance of the Moon
being about 30 per cent. The greatest rise and fall
of the tide happens when New or Full Moon occurs
at the time when the Moon is at its nearest to the
Earth.
CHAPTER III
THE SUN'S FAMILY
OF PLANETS
FROM night to night we can detect no change in the
relative positions of the so-called "fixed stars";
even the span of a human lifetime would not be
sufficient for any changes to be apparent to the
naked eye, though refined observations with tele-
scopic aid reveal that the positions are slowly chang-
ing. It was known to the ancients that there were
a few bodies which moved about amongst the stars,
and these they termed planets or wanderers (Latin,
planeta, a wanderer). Under this term they in-
cluded the Moon, Mercury, Venus, the Sun, Mars,
Jupiter and Saturn; these bodies were supposed to
move around the Earth and to be in this order of
increasing distance from the Earth. The term
" planet " is now restricted to the bodies which are
revolving around the Sun in definite orbits. It in-
cludes the Earth ; the five bodies which have been
known from prehistoric times, viz. Mercury, Venus,
Mars, Jupiter and Saturn ; the three bodies which
have been discovered in comparatively recent times,
viz. Uranus, Neptune and Pluto ; and many hun-
dreds of small bodies which are termed minor
planets or asteroids. These bodies form the Sun's
family of planets.
Certain regularities in the arrangement of this
family may be noticed. The planets all move
around the Sun in the same direction and their
36
THE SUN'S FAMILY OF PLANETS 37
paths lie very nearly in the same plane. It is there-
fore only in certain regions of the sky that the
planets are to be found ; they are never very far dis-
tant from the " ecliptic/ 5 the path amongst the
stars which the Sun appears to describe in the course
of ayear as a result of the motion of the Earth around
the Sun. This path is marked in any good star
atlas. The planets also rotate about their axes in
the same direction as that in which the planets
themselves revolve around the Sun.
Similar regularities are also seen in the satellite
systems of the planets, with one or two trivial excep-
tions. The Earth has one satellite, the Moon,
which revolves around the Earth just as the Earth
revolves around the Sun. Jupiter and Saturn have
nine satellites each, Uranus has four, Mars has two
and Neptune has one. In general, the rotations of
these satellites about their axes and their revolutions
about the parent bodies are in the same direction as
that in which the planets themselves move around
the Sun. There is a definite " rule of the road " to
which the members of the Sun's family conform.
These regularities cannot be the result of mere
chance and undoubtedly point to some common
origin for all the members of the Sun's family.
What this origin is we shall consider in Chapter XIII.
It is desirable to have a general idea of the scale
and arrangement of the solar system. If we repre-
sent the diameter of the Earth by I inch, the Sun
must be placed at a distance of 320 yards and will
be 9 feet in diameter; the Moon will be only 2| feet
away and will be J inch in diameter, or about the
size of a pea. Jupiter, the largest of the planets,
38 WORLDS WITHOUT END
will then be about the size of a football, with a
diameter of n inches. Mercury is somewhat
larger than the Moon, being about f inch in
diameter and about equal in size to the largest of the
satellites of Jupiter. The minor planets would
hardly show on this scale; the largest of them would
be about the size of the letter " o " in this type, the
FIG. i . Relative sizes of Sun and planets.
smallest would be much smaller than the full-stop.
Saturn is large, but smaller than Jupiter. The next
largest are Uranus and Neptune, which are about
the size of a grape-fruit. Pluto is probably smaller
than Mercury. Jupiter would be at a distance of
nearly i mile, whilst Pluto would be more than
7 miles away. The sizes of the planets relative
to each other and to the Sun are illustrated in Fig. i .
The movements of the planets in their orbits take
THE SUN S FAMILY OF PLANETS 39
place in accordance with certain rules which were
formulated by Kepler between the years 1607 and
1620. The three famous laws of Kepler may be
stated as follows :
1 . Each planet moves round the Sun in an ellipse
and the Sun occupies one of the foci of the ellipse.
An ellipse can be simply drawn by placing two pins
in a piece of paper, running a loop of cotton around
them, drawing the loop tight with the point of a pen-
cil and moving the pencil round. The positions of
the two pins are called the foci of the ellipse. If the
two foci are far apart, the ellipse is very elongated in
shape; if they are brought nearer and nearer to-
gether, the ellipse becomes more and more nearly
similar to a circle, and, in the limiting case when the
two foci coincide, the ellipse becomes a circle.
None of the orbits of the planets, except those of
Mercury, Pluto and some of the asteroids, differs very
greatly from a circle.
2. Each planet moves in its orbit round the Sun
in such a way that the line joining it to the Sun
moves through equal areas in equal times. It fol-
lows that the motion of the planet in its orbit is
most rapid when the planet is at its nearest to the
Sun and slowest when it is at its greatest distance.
3. The time in which a planet travels once round
its orbit is related to the size of the orbit. The
actual relationship is that the squares of the times
of orbital revolution are proportional to the cubes
of the distances of the planets from the Sun. It
may be deduced that the actual velocity of the
planet in its orbit is inversely proportional to the
square root of its distance from the Sun. Thus if
40 WORLDS WITHOUT END
there were a planet whose distance from the Sun
were four times that of the Earth, its speed in its
orbit would be only one-half that of the Earth.
The further the planet from the Sun, the slower the
speed with which it moves. This is in direct con-
trast to what happens when a rigid body rotates.
If we think, for instance, of a wheel spinning round,
the speed of a point on the rim is greater than the
speed of a point on the hub ; the speed of any point
on a spoke is directly proportional to its distance
from the centre of the hub. If the solar system were
rotating like a rigid body, the most distant planets
would move the fastest.
Newton showed that these empirical laws of
Kepler could be simply explained by the universal
law of gravitation. It is the gravitational pull of
the Sun which prevents the planets flying away into
space and keeps them moving in their paths around
the Sun. The nearer the planet is to the Sun, the
greater is the pull of the Sun and the faster the
planet must move to prevent itself falling towards
the Sun. Thus we find that Mercury is hurrying
around with a speed of about 32 miles a second,
completing one revolution of its orbit in 88 days,
whereas Pluto, which is so far away that it takes
about 248 years to go once round its orbit, is a com-
parative laggard, with a speed of little more than
3 miles a second, a speed which though high
when judged by ordinary terrestrial standards is
low when compared with those of most other celes-
tial bodies.
The planets, like the Moon,, are not self-luminous,
but shine by the light from the Sun which they
THE SUN'S FAMILY OF PLANETS 41
reflect or scatter back into space. Unless a planet
has a store of heat in its interior as it would have,
for instance, if it were largely composed of radio-
active materials there must be a general balance
between the amount of energy which it receives in
the form of light and heat from the Sun and the
amount which it sends back into space. The
amount of solar energy falling on any planet can be
computed without difficulty, for the size of each
planet and its distance from the Sun is known. The
energy emitted by the planet can be detected and
measured with the aid of a sufficiently powerful
telescope arid a sensitive energy-detecting device
such as a thermopile or a bolometer. By placing a
small transparent cell containing water in front of
the thermopile, the radiation of long wave-length is
absorbed and only the short wave-length radiation
falls on the thermopile ; thus the radiations of long
and short wave-lengths can be separately measured
and it becomes possible to deduce the temperature
of the planet. The temperatures obtained in this
way are in close agreement with those deduced from
the amount of solar energy falling on the planet, and
it can be inferred that none of the planets has any
appreciable source of internal heat. These tem-
peratures are not necessarily the temperatures of the
actual surface of the planet, for if the planet has a
covering of cloud the cloud layer may play an im-
portant part in modifying the surface temperature,
as we have already seen in the case of the Earth.
Mercury, the planet nearest to the Sun, is the
smallest of the planets with the possible exception of
Pluto, the asteroids being left out of consideration.
42 WORLDS WITHOUT END
Its distance from the Sun is about 36 million miles.
It is too near to the Sun to be seen with the naked
eye, except in the evening shortly after sunset, or in
the morning shortly before sunrise; it is more
favourably placed for observation in the tropics than
in more northerly or southerly latitudes.
The diameter of Mercury is about 3,000 miles, or
about half as large again as the diameter of the
Moon. The Earth could contain about sixteen
bodies of the size of Mercury. Its weight is not
known with great accuracy, but it is probably only
about one-twenty-fifth of the weight of the Earth.
The velocity of escape from Mercury is not much
greater than the velocity of escape from the Moon,
and we may conclude with reasonable certainty
that Mercury has no atmosphere. Direct con-
firmation of this conclusion can be obtained, for we
find that Mercury reflects only about 7 per cent, of
the sunlight which falls upon it; the remainder is
absorbed by the surface rocks and emitted again as
heat radiation of long wave-length. The Moon
reflects the same proportion. If the surface were
cloud-covered the proportion would be much
higher.
The absence of any atmosphere on Mercury neces-
sarily implies also the complete absence of water.
Any water there may once have been will gradually
have been evaporated, and the water-vapour, like
the atmosphere, slowly dissipated away into space.
It may be concluded that Mercury is a dead, arid
world.
We cannot find out much about its surface, for
there are no well-defined markings to be seen on it,
THE SUN'S FAMILY OF PLANETS 43
as there are on the Moon. Careful observation has
shown little more than faint ill-defined shadings,
visible only under favourable conditions. The
study of these markings has shown that the planet
always turns the same face to the Sun, just as the
Moon does to the Earth. Thus the day and the
year on Mercury are of the same length, and each
equal to 88 of our days. By the word " day "
we here mean the period of rotation of the
planet about its axis. If, however, we use the terms
day and night to express the periods of light and
darkness, we can say that there is perpetual day
over one half of Mercury and perpetual night overl
the other half.
The temperature on the sunlit face of Mercury is
extremely high; by direct measurement it is found
that the temperature of the portion of the surface
which has the Sun overhead is about equal to that
of molten zinc. The opposite side of the planet
must be intensely cold. Near the border line be-
tween the two regions it would be possible to pass
in a comparatively short journey from a region of
intense cold, with a temperature lower than any
temperatures experienced on our Earth, to a region
of intense heat far surpassing that of our tropical
regions. The faint markings are occasionally ob-
scured in certain parts of the planet, as though by
clouds of dust. It is possible that there are active
volcanoes on Mercury, which from time to time are
in violent eruption and eject clouds of fine dust to
a considerable height above the surface. The dust
gradually falls to the ground but temporarily
obscures the surface of the planet beneath it.
44 WORLDS WITHOUT END
There are no winds to disperse the dust, because
there is no atmosphere.
The next planet in order of distance from the Sun
is Venus, which has special interest for us as being
the planet which most nearly resembles the Earth.
Venus may be regarded as the Earth's twin sister,
for in size, in weight, in density and in general con-
stitution it is not greatly different from the Earth.
It is a little smaller, a little less massive and a little
less dense. Its greatest angular distance from
the Sun is about 46, and so it can best be seen
within a few hours after sunset or before sunrise.
Its distance from the Earth varies widely, from 26
million miles to 160 million miles, and its apparent
brightness varies considerably in consequence.
Venus, like Mercury, shows phases in the telescope.
Doughty, in his Arabia Deserta, states that the
Arabs, who were possessed of great acuity of vision,
called Venus the " horned star," and he considered
that they were able to see it as a crescent with the
naked eye. I do not think that there is any con-
clusive evidence of Venus actually having been seen
as a crescent with the naked eye by a trustworthy
observer who was entirely free from bias or prior
knowledge.
It is when Venus is at her nearest to the Earth
that she appears as a thin crescent. As the distance
increases, more of the bright disc becomes visible;
the increasing distance tends to make the planet
appear less bright, but the change in phase acts in a
contrary direction; the resultant of the two effects
is that the brightness continues to increase for
about 36 days from the time when Venus is at her
THE SUN S FAMILY OF PLANETS 45
nearest. Thereafter the effect of the increasing
phase is more than counterbalanced by the greater
distance. When at her brightest, Venus is much
brighter than any star or any other planet and can
be seen without difficulty by the naked eye in broad
daylight. I have on more than one occasion seen
it without looking for it, and without realising at
first that I was actually looking at Venus.
The velocity of escape from Venus is slightly
lower than the velocity of escape from the Earth.
It is therefore to be anticipated that Venus has suc-
ceeded in retaining an atmosphere. This antici-
pation is fully confirmed by observation. Venus
reflects about 60 per cent, of the sunlight which falls
upon it. This is about equal to the reflecting power
of freshly fallen snow and is higher than the reflect-
ing power of any other planet. No known rocks
or soils have so high a reflecting power, and it may
be inferred that Venus is covered with a highly
reflecting layer of cloud. The appearance of the
planet in the telescope is consistent with this infer-
ence. No surface markings can be seen, though ill-
defined darkish shadings have been seen at times.
These do not persist for long and are undoubtedly
atmospheric objects of some sort.
Photographic plates which are sensitive to the
infra-red rays, whose wave-length is outside the
limit to which the human eye is sensitive, have been
used recently with much success for the photography
of distant landscapes. Such photographs show far
more detail than those taken on ordinary plates.
(See Plate VII.) The reason is that the infra-red
rays are of long wave-length and are therefore able
46 WORLDS WITHOUT END
to pass through a greater length of atmosphere than
the actinic rays of short wave-length, which are
rapidly scattered in all directions. Photographs of
Venus have been obtained using plates specially
sensitive to infra-red light, in the hope that these
rays would penetrate through the atmosphere of
Venus and tell us something about the surface of
the planet. But they show no surface details ; even
the infra-red rays cannot penetrate the planet's
atmosphere. This is not surprising, if the planet
has a permanent dense covering of cloud.
We can attempt to gain some information in
another way. The light reaching us from Venus
can be compared with the light which we receive
directly from the Sun. In neither case is the light
pure sunlight. The light which we receive from
the Sun directly has some wave-lengths weakened
or missing because of absorption in the atmosphere
of the Earth. Oxygen cuts out one set of wave-
lengths, water-vapour another set and so on. If
we find that light of a certain wave-length is missing
from sunlight, it may be either because this light
has been absorbed by the outer layers of the Sun or
because it has been absorbed by the atmosphere
of the Earth. We can easily distinguish between
these two possibilities by comparing the light re-
ceived from the Sun when it is high in the sky with
that received near sunset. As the Sun gets lower
in the sky, its rays have to pass through a greater
extent of the Earth's atmosphere in order to reach
us, and any effects produced by absorption by our
atmosphere become intensified. In a similar way,
if we compare the light coming to us directly from
THE SUN'S FAMILY OF PLANETS 47
the Sun with that which has penetrated to the cloud
layer of Venus, then out again and so through our
atmosphere to us, the differences must be due to
the absorption in the outer layers of the atmosphere
of Venus. The information which we may hope
to obtain in this way is necessarily limited, for a
group of elements which includes hydrogen, helium,
nitrogen, argon and neon does not absorb any of
the sunlight, and so we cannot detect these elements,
even though they may be present.
But once again Venus defeats our attempts to
wrest her secrets from her. No traces either of
oxygen or of water-vapour are to be found ; the test
for oxygen is much more sensitive than that for
water-vapour, and we can only draw the negative
conclusion that the amount of oxygen above the
cloud layer which covers the planet is less than the
one-thousandth part of the quantity in the atmo-
sphere of the Earth. The only positive information
that we have derived is a clear indication of the pre-
sence of carbon dioxide in the atmosphere. The
amount of carbon dioxide above the cloud layer
is surprisingly large; it is equivalent to a layer 2
miles thick at the Earth's surface. The total amount
above the surface of Venus must be considerably
greater. There is very little carbon dioxide in the
Earth's atmosphere. The whole atmosphere of the
Earth is equivalent to a layer 5 miles thick at the
surface; of this total, the carbon dioxide amounts
to a thickness of only 30 feet, so that there is several
thousand times more carbon dioxide in the atmo-
sphere of Venus than in our atmosphere.
Little more is known about the length of the day
48 WORLDS WITHOUT END
on Venus. It is reasonably certain that it cannot
be less than 2 or 3 of our weeks and equally certain
that the planet does not always turn the same
face to the Sun; if this were the case the dark side
would be very cold, whereas it has been found that
a considerable amount of heat is radiated from the
dark side of the planet. The day on Venus must
therefore be considerably less than 224 of our days.
A day equal to 2 or 3 of our weeks seems probable.
As the year on Venus is equal to about 32 of our
weeks, there are only some 10 or 15 days in the year.
In the face of all this negative evidence, any at-
tempt to describe the conditions existing on Venus
must be very speculative. Since the planet is nearer
than our Earth to the Sun, we may expect the sur-
face of Venus to be much warmer than that of the
Earth. Direct measurement of the temperature
shows that it is practically uniform over the planet;
the temperature of both day and night sides is about
15 F. below zero, but, as in the case of the Earth,
the high clouds are likely to be much cooler than the
surface. The failure to detect water-vapour, even
though the tests are not fully decisive, is perhaps not
surprising. It is difficult to conceive of what the
cloud layer over Venus is composed, if it is not con-
densed water-vapour; the region above the clouds
is probably too cold for water-vapour to exist with-
out condensing. The water-vapour in the Earth's
atmosphere is almost entirely confined to the lower
layers; clouds at a height greater than about 5
miles are rarely found, unless perhaps in tropical
regions. It seems probable that there are oceans
on Venus; the high temperature at the surface
THE SUN'S FAMILY OF PLANETS 49
would then result in extensive evaporation and the
formation of a dense cloud layer. We may con-
clude that Venus must have a hot, damp climate with
heavy precipitation.
The failure to detect any traces of oxygen is the
most surprising feature about Venus, for in other)
respects conditions are not greatly unlike those on'
the Earth. This may possibly point to a lack of
vegetation on the planet. Oxygen is chemically an
active substance, which unites readily with other
substances to form compounds. On our Earth
vegetation plays a great part in replenishing the
store of oxygen; if Venus has no vegetation, then the
absence of oxygen is perhaps explicable.
The last to be considered of the four inner planets
is Mars, whose average distance from the Sun is
about half as large again as the distance of the Earth
from the Sun. When at its closest to the Earth, it
is only about 35 million miles away and appears
brighter than any other planet except Venus.
Mars is considerably smaller than the Earth its
diameter being little more than half that of the Earth
and its weight is only about one-tenth of that of the
Earth. As the velocity of escape from Mars is
rather less than one-half of the velocity of escape
from the Earth, it is not to be expected that Mars
will have an extensive atmosphere; we may indeed
anticipate that the atmosphere has largely been
lost. The appearance of Mars in the telescope con-
firms this; it is a beautiful object with a strong orange
colour, on which dark and light surface markings
may be clearly seen. Around whichever pole is
visible there is a bright white cap, called the polar
5O WORLDS WITHOUT END
cap. The two polar caps show regular seasonal
changes in size; the northern cap, for instance,
shrinks during the northern summer, whilst the
southern cap grows. During the northern winter
the northern cap grows and the southern cap
shrinks. At midwinter either cap has a diameter
of somewhere about 3,000 miles and extends about
half-way from the pole to the equator. The south-
ern cap sometimes disappears altogether in the
summer, but the northern cap never shrinks to a
diameter less than about 200 miles. In contrast to
the polar caps, the reddish areas on the planet show
little or no change with the seasons, and in them we
probably see the actual surface of the planet. The
amount of light reflected by the surface of the planet
is what would be expected from moderately dark
rocks, and we may conclude that Mars has a rough
rocky surface.
There are other markings on Mars which are of a
bluish-grey or greenish shade ; they are found mainly
in the southern hemisphere. The early observers
of Mars thought that these were oceans and named
them accordingly. It is certain that there are no
large sheets of water on Mars, for, if there were, bril-
liant reflections would be seen from them when the
Sun is suitably placed; but such reflections have
never been observed. These markings, though
fairly permanent in form and position, show changes
of a seasonal nature and are sometimes much less
conspicuous than at others; changes in colour have
also been noticed. They may be in part an atmo-
spheric phenomenon and possibly in part the effect
of local precipitation of rain.
THE SUN S FAMILY OF PLANETS 5!
In addition to these broad general features, which
may be seen with a telescope of moderate size, some
observers have claimed to see a number of narrow,
dark, straight markings on Mars, extending in some
cases for thousands of miles. The existence of such
markingswas announced by Schiaparelliin 1877, and
they were interpreted by him as a network of chan-
nels for water. He designated them by the Italian
word canali. Other competent observers also claim
to have seen these canals ; in particular Lowell and
Slipher, observing in the clear air and at the high
altitude of the Flagstaff Observatory in Arizona,
have studied them. One of Lowell's drawings,
showing many canals, is reproduced in Plate VII.
Other skilled and equally competent observers have
failed to detect them. The canals have never been
shown on a photograph of the planet; but it has to
be remembered that with the largest telescope the
actual image of Mars is less than one-tenth of an
inch in diameter. In order to photograph Mars,
a larger image is necessary and so an enlarging lens
is used ; but in magnifying the image, the brightness
is reduced and the time of exposure must be length-
ened. Under the best conditions an exposure of
several seconds is necessary to secure a photograph.
Atmospheric tremors, even under the best conditions
at the most favourably placed observatory, are suf-
ficient to wash out such extremely fine detail.
It seems probable that though there is undoubtedly
much fine detail on the surface of Mars, the canals, as
described by Schiaparelli, do not actually exist and
that their appearance is due to a subjective pheno-
menon: the eye tends to connect up faint irregular
52 WORLDS WITHOUT END
markings, which are visible only with difficulty, by
straight lines. The experiment was made of getting
a class of schoolboys to copy a drawing of Mars on
which there were no canals marked. Most of the
boys at the back of the classroom put numerous
straight lines into their drawing. There was no
question here of any preconceived bias, for the boys
were not aware that the drawing they were copying
was a drawing of Mars.
When photographs of Mars are obtained, using
first ultra-violet light and then infra-red light, some
interesting differences appear. The infra-red photo-
graphs show the surface markings clearly ; the ultra-
violet photographs do not show them at all (Plates
VII and VIII). This provides direct proof of the
existence of an atmosphere on Mars. Sunlight of
long wave-length is able to get through the atmo-
sphere to the surface of Mars and out again ; but
the light of short wave-length is completely scattered
before it can reach the surface. There is also evi-
dence of clouds of two distinct types. Clouds of the
first type are seen as whitish areas on the ultra-violet
photographs, but do not show on the infra-red
photographs ; an example of a cloud of this type is
to be seen in Plate VIII. Such clouds (which are
not necessarily in any way similar to clouds in our
own atmosphere) must be high up in the atmo-
sphere of Mars, or they would not be seen in the
ultra-violet photographs, and they must be suffi-
ciently tenuous to allow the infra-red light to go
through. Clouds of the second type show just the
opposite effect; they are seen on the infra-red but
not on the ultra-violet photographs. These clouds
THE SUN'S FAMILY OF PLANETS 53
appear yellowish to the eye. They must be much
lower in the Martian atmosphere, because the ultra-
violet light does not penetrate to them, though
observations at the Flagstaff Observatory show that
they may be as high as 15,000 feet above the surface.
It is possible that they may be clouds of water-
vapour, seen through a yellowish atmosphere.
The two polar caps show an unexpected differ-
ence in photographs taken with light of long and
short wave-length. The most natural explanation
of these caps is that they are surface deposits of snow
or rime in the polar regions of Mars, analogous to
the snow- and ice-caps of the polar regions of the
Earth. If so, we should expect them to be most
clearly shown on the photographs taken with the
infra-red rays. Exactly the opposite is found; the
caps are seen most clearly on the photographs taken
with the ultra-violet rays and become less prominent
the longer the wave-length of the light in which the
planet is photographed. This is well illustrated by
the two photographs of Mars in Plate VII. The
polar caps must therefore be largely, though not
entirely, an atmospheric phenomenon. It is possible
that over the polar regions there are clouds of par-
ticles of snow or ice of no very great thickness, so
that light of long wave-length can get through them,
and that there is in addition a surface deposit of
snow or ice. This surface deposit cannot be more
than a few inches in thickness, or we should not
observe such rapid melting of the caps as is seen dur-
ing the summer months.
The image of Mars on a photograph in ultra-
violet light is considerably larger than the image on
54 WORLDS WITHOUT END
a photograph in infra-red light taken with the same
telescope. The latter image, on which the surface
details are clearly shown, must correspond to the
solid globe of the planet ; the ultra-violet image
represents the atmospheric shell which surrounds
Mars. The difference in the radii of the two images
is from 50 to 60 miles, so that the atmosphere,
though very thin, has a considerable depth. The
total atmospheric pressure on Mars is probably not
more than a few per cent, of that at the surface
of the Earth. Plate VIII shows photographs of
Mars obtained with ultra-violet and infra-red light;
beneath them are the same photographs with their
opposite halves in juxtaposition. The difference in
size is at once apparent.
Until recently it was believed that the presence of
both oxygen and water-vapour in the atmosphere
of Mars had been definitely established, but the
amount of each of these constituents was surprisingly
small. Recent more delicate observations made at
the Mount Wilson Observatory have failed to detect
either oxygen or water-vapour, though an amount
of oxygen equal to a thousandth part of that above
an equal area of the Earth could have been detected
with certainty. The tests for water-vapour are less
sensitive, but it is certain that the total amount must
be small. Yet there seems little reason to doubt
that there must be both oxygen and water-vapour
in the Martian atmosphere. The melting of the
polar caps must produce water-vapour. The red-
dish colour of the surface of Mars is probably due to
the oxidation of iron-bearing ores by atmospheric
oxygen. It is possible that most of the oxygen on
THE SUN'S FAMILY OF PLANETS 55
Mars has been converted by the ultra-violet light of
the Sun into ozone ; the presence of ozone at the
surface of Mars would greatly accelerate the oxida-
tion of the rocks. The amount of ozone in the
Earth's atmosphere is extremely small, its equivalent
thickness being not more than one-eighth of an inch.
Yet this thin layer completely prevents any of the
extreme ultra-violet light from the Sun reaching the
surface of the Earth; the completeness of this ab-
sorption makes it impossible to detect the presence
of ozone in the atmosphere of Mars.
Observations of the temperature at the surface of
Mars confirm the low amount of water-vapour in
the atmosphere. Water-vapour absorbs strongly
radiations of long wave-length and is therefore very
effective in preventing the escape of the heat radia-
tion from the surface of a planet. It is common
experience that in damp tropical climates there is
very little fall of temperature at night, whereas at
places where the air is very dry the temperature falls
rapidly after sunset. The maximum temperature
on the Earth is usually reached during the afternoon
and not at noon, when the Sun is highest in the sky
and the solar radiation received is at its maximum.
This is a consequence of the blanketing action of the
water-vapour in our atmosphere preventing the
rapid escape of the radiation from the heated sur-
face of the Earth. But on Mars the maximum tem-
perature occurs at Martian noon. This is proof that
there is not sufficient water-vapour in its atmosphere
to exert any appreciable blanketing effect.
Mars is naturally cooler than the Earth, being at
a greater distance from the Sun. In the equatorial
56 WORLDS WITHOUT END
regions the temperature may be as high as 50 F.
during the daytime. The temperature falls very
rapidly after noon, and even at the equator must
be well below freezing-point by sunset. It is
very low at night, probably reaching 130 F.
below zero in the equatorial regions. The enor-
mous daily range in temperature, of about 180 F.
'(equal to the difference between the boiling-point
of water and the freezing-point), must make con-
ditions extremely uncomfortable for any life that
may exist on Mars. The temperature of the polar
cap is somewhere about 100 F. below zero, but in
the late summer, after the cap has disappeared,
there seems to be little, if any, difference in tempera-
ture between the pole and the equator.
The length of the day on Mars does not differ
very much from the length of our own day, being
about 40 minutes longer. None of the other planets
has a day which is so nearly equal to our own day.
The seasonal changes on Mars are much more
pronounced in the southern hemisphere than in the
northern. This is because the orbit of Mars round
the Sun is much more elliptic than the orbit of the
Earth. Whereas the distance of the Earth from the
Sun does not vary by as much as 3 million miles, the
distance of Mars from the Sun varies by 26 million
miles. Mars is nearest to the Sun at the time when
it is winter in the northern hemisphere and summer
in the southern hemisphere ; it is at its farthest when
it is summer in the northern hemisphere and winter
in the southern. The southern hemisphere, there-
fore, has a warmer summer but a colder winter than
the northern.
THE SUN'S FAMILY OF PLANETS 57
Mars is the first planet to be met, as we travel out-
wards from the Sun, with more than one moon. It
has two moons or satellites, but they are insignificant
little bodies compared with our own Moon. The
one which is nearer to Mars, called Phobos, is less
than 40 miles across ; the outer one, called Deimos,
is even smaller, being only about 8 miles across.
They are, however, so much nearer to Mars than
the Moon is to the Earth that to a Martian observer
Phobos would appear about the same size as the
Moon appears to us. Phobos is only 5,800 miles
and Deimos about 14,600 miles from the centre of
Mars. Being so near to Mars, they move round it
very rapidly, Phobos in about y| hours and Deimos
in about 30 hours. The rapid revolutions of the
two moons round Mars have some very curious
consequences: thus, though both satellites move
round Mars in the same direction as that in which
Mars moves round the Sun, Phobos would be seen
by an observer on Mars to rise in the west and to set
in the east about 4^ hours later. Deimos, on the
other hand, rises in the east and sets in the west
66 hours later; between rising and setting Deimos
would be seen to go through all its phases (new to
full and back to new again) twice.
It is of interest to recall a remarkable prediction
made by Dean Swift in Gulliver's Travels, published
in 1726, a century and a half before the satellites of
Mars were discovered. He relates that the astro-
nomers of Laputa " have discovered two lesser stars,
or satellites, which revolve about Mars, whereof the
innermost is distant from the centre of the primary
planet exactly three of his diameters, and the outer-
58 WORLDS WITHOUT END
most five ; the former revolves in the space of ten
hours and the latter in twenty-one and a half."
Considering the nearness to Mars of its satellites and
their rapid revolution round it, the prediction was
extraordinarily near the truth.
A curious empirical relationship between the dis-
tances of the planets from the Sun was formulated
by Bode in 1772 and is known as Bode's Law. The
law is as follows: to the numbers o, 3, 6, 12, 24, 48,
etc., add the number 4. The resulting numbers
divided by 10 express, approximately, the dis-
tances of the planets from the Sun, in terms of the
distance of the Earth as unity. The numbers ob-
tained in this way are given in the first column
of the following table. The second column gives
the actual mean distances of the planets whose
names are contained in the third column.
Bode's Law.
True Distance.
Planet.
0-4
o-39
Mercury
0-7
0-72
Venus
I'O
I'OO
Earth
1-6
1-52
Mars
2-8
(Minor Planets)
5'2
5-20
Jupiter
10-0
9-54
Saturn
19-6
19-18
Uranus
38-8
30-07
Neptune
77'2
39-50
Pluto
At the time Bode formulated this law, Mercury,
Venus, the Earth, Mars, Jupiter and Saturn were
the only planets known. The distances of these
planets agree closely with the distances given by the
law, provided that a gap is left at the distance 2 '8.
In 1781 Uranus was discovered by William Herschel,
THE SUN'S FAMILY OF PLANETS 59
and it will be noted that the distance of Uranus is in
good agreement with the value to be expected from
Bode's Law. The more recently discovered planets,
Neptune and Pluto, show an increasing deviation
from the values required by Bode's Law. The gap
at the distance of 2*8 times the Earth's distance can
be regarded as filled by the discovery of a zone of
small planetary bodies lying between Mars and
Jupiter. The first of these bodies to be discovered,
known as Ceres, was found accidentally by the
Sicilian astronomer Piazzi. On January i, 1801,
in the course of observations for his star-catalogue,
he observed a seventh-magnitude star where there
had been no star a few days previously. The next
night he found it had perceptibly moved with
respect to the adjacent stars, and further observations
showed that it continued to move. He named the
new planet Ceres, after the tutelary divinity of the
island of Sicily. In 1802, a second planet, Pallas,
was discovered ; a third, Juno, was found in 1 804,
and Vesta, the brightest of all the minor planets and
the only one ever visible to the naked eye, was dis-
covered in 1807. The fifth asteroid, Astraea, was
not discovered until 1845, but since that date the
list has grown rapidly. More than 1,200 have been
observed sufficiently well for their orbits to have
been computed; several hundred others have been
discovered and lost again. They are for the most
part very faint objects, usually discovered photo-
graphically. During the exposure, which may be
several minutes or hours, the telescope is slowly
moved by clockwork so that the effect of the rota-
tion of the Earth is counteracted, and it continues
6O WORLDS WITHOUT END
to point to the same stars as they move across the
sky. The image of a star on the photographic plate
is perfectly round, but the image of a minor planet
is drawn out by its motion relative to the stars into
a streak. The asteroids are easily detected in this
way. Plate XI (a) shows a number of round images
of stars together with three trails of asteroids.
Compared with the planets, the asteroids are all
small bodies, whence the description of them as
" minor planets." The largest of them, Ceres, has
a diameter of about 480 miles; the diameter of
Pallas is 306 miles, of Vesta 241 miles and of Juno
121 miles. Most of them, however, are not more
than a few miles in diameter. It is quite certain
that they are all devoid of any atmosphere; the
velocity of escape from Ceres is, for instance, only
about \ mile per second.
The asteroid zone extends from Mars to Jupiter,
though in 1932 a tiny asteroid was discovered with
a very elliptical orbit which actually passes within
the orbit of Venus. This small body can approach
to within about 3 million miles of the Earth and
therefore comes nearer to us than any other known
asteroid. In May 1932 it was only 6| million miles
from the Earth. No closer approach of any planet-
ary body has ever been observed.
The total weight of all the asteroids must be con-
siderably less than the weight of Mars, for otherwise
their gravitational attractions would produce dis-
turbances in the orbit of Mars. It seems probable
that the total weight of all the known asteroids is less
than the one-thousandth part of the weight of the
Earth.
THE SUN'S FAMILY OF PLANETS 6l
We do not know what was the origin of this swarm
of tiny planetary bodies. They may represent the
debris of a planet which exploded long ago, or they
may be the remnants of matter out of which the
planets were formed, that has never aggregated
into a single body. There does not seem to be any
way by which either of the possibilities can be
definitely proved or disproved, and we must there-
fore leave the question of their origin as an unsolved
problem.
Passing through the zone of the minor planets we
come to the largest and most massive of all the
planets, Jupiter. As we have seen above, when
considering Bode's Law, its distance from the Sun
exceeds five times the Earth's distance. Jupiter's
year is nearly twelve of our years, and its velocity in
its orbit round the Sun is rather less than half the
velocity of the Earth, or about 9 miles per second.
Both in volume and in weight Jupiter far exceeds
all the other planets combined. It is so large that
it could contain more than 1,300 bodies of the size
of the Earth. In proportion to its size, it is not so
massive as the Earth. Jupiter is 317 times heavier
than the Earth; but if it were composed of the same
materials as the Earth, it would weigh about four
times as much as it actually does. The average
density of the material of which it is composed is
only about 30 per cent, greater than the density of
water. The reason for this low mean density we
shall see presently.
The velocity of escape from Jupiter is very high,
about 40 miles per second. We should con-
sequently expect to find that Jupiter has an exten-
62 WORLDS WITHOUT END
sive atmosphere ; even the fast-moving molecules of
hydrogen are not able to escape from the strong
binding power of Jupiter's gravitation. The tele-
scope confirms our anticipations, for we can see
nothing of Jupiter's surface, but merely atmospheric
formations that are continually changing. In the
telescope, Jupiter is seen to be distinctly flattened.
This is evidence of rapid rotation, for, if it were not
rotating, its gravitation would have pulled it into a
spherical shape. Jupiter rotates more rapidly than
any other planet, the length of the day being rather
less than 10 hours. The year on Jupiter therefore
contains about 10,000 Jovian days. If the mark-
ings revealed by the telescope are observed for a few
hours, they will be seen to be moving across the
planet's disc, giving direct proof of the rotation.
More exact observation shows that different zones
of the planet rotate at different speeds, the equa-
torial portion of the planet having the most rapid
rotation. This provides confirmation that we do
not see the solid surface of the planet.
The markings occur in zones which are approxi-
mately parallel to the equator (Plate IX). 'The
equatorial zone is usually very bright and is bor-
dered to the north and south by two darker brown-
ish zones. Numerous small, well-defined, bright
and dark spots are frequently observed in these two
zones; these spots, which must be some sort of cloud
phenomenon, though their nature is still a mystery,
sometimes last for several months. There is one
remarkable marking which has persisted for many
years. It is known as the Red Spot. When first
seen in 1878, it appeared as a pale pinkish oval spot.
THE SUN'S FAMILY OF PLANETS 63
It rapidly developed in size, changing to a brick-red
colour, until it stretched over a length of about
30,000 miles and had a breadth of about 7,000 miles.
It was then about equal in area to the whole of the
surface of the Earth. Though it faded considerably
in subsequent years, it remained a conspicuous
object until about 1919, when it gradually became
less conspicuous. It can still be seen, though it is
no longer a prominent object. The spot does not
remain in a fixed position; during the last 30
years it has drifted to as great a distance as 20,000
miles on each side of its mean position and through
several thousand miles in latitude. It has been sug-
gested that the spot was produced by an eruption
of some sort, possibly volcanic in origin, and that
the ejected gases poured out over the highest cloud
zone and remained in practically a stationary
position relative to it. The drifting of the spot to
and fro could then be explained by currents in the
atmosphere of Jupiter. But in the light of new
knowledge about the constitution of Jupiter, this ex-
planation does not seem probable. It is noticeable
that the red spot is less conspicuous on photographs
taken in infra-red light than in photographs taken
in ultra-violet light, and this seems to point to a posi-
tion in the upper regions of Jupiter's atmosphere.
The cloud-layer on Jupiter is so thick that infra-
red photographs completely fail to penetrate it.
Planetary observers were formerly of the opinion
that the rapidity of the changes which occur on
Jupiter was an indication that the planet was hot.
On account of the great distance of Jupiter from the
Sun, the amount of heat which falls on each square
64 WORLDS WITHOUT END
foot of Jupiter is only one-twenty-seventh of that
which falls on each square foot of the Earth, so that
Jupiter could only be hot if it had a large internal
store of heat. There were various theoretical con-
siderations which made it difficult to accept this.
When direct measurement of the temperature of
Jupiter became possible, it was found that it was
extremely cold; the observed temperature is about
200 F. below zero, far lower than any experienced
in the Earth.
Bearing this extremely low temperature in mind,
it will be evident that the clouds in the atmosphere
of Jupiter cannot be composed of water-vapour or
even of particles of snow or ice. By the analysis of
the light from Jupiter, which is, of course, reflected
sunlight, two constituents of its atmosphere have
been detected ammonia, the pungent gas which is
used in many refrigerating plants, and methane, or
marsh-gas, the gas produced by vegetable matter
decomposing under water, and known to the coal-
miner as the dangerous fire-damp.
This provides the key to the interpretation of the
atmosphere of Jupiter. We believe that Jupiter, to-
gether with the other planets, was formed from
matter ejected by the Sun. It is known that the
Sun consists largely of hydrogen. We may suppose,
therefore, that Jupiter, when it was first formed, con-
tained a large proportion of hydrogen. Unlike the
Earth, it was able to retain the hydrogen because of
its large gravitational attraction. As it cooled down
from its initial hot state, the oxygen in its atmosphere
combined with some of the hydrogen and formed
water-vapour which, when the planet had cooled
THE SUN'S FAMILY OF PLANETS 65
still further, was deposited as a thick layer of ice.
Nitrogen and carbon also combined with some of
the excess hydrogen to form saturated compounds.
The most volatile compounds of hydrogen with
nitrogen and carbon are ammonia and marsh-gas
respectively; these are therefore the gases that we
should expect to find in the atmosphere of Jupiter.
It may be remarked that the presence of these
substances in the atmosphere of Jupiter is in itself
conclusive proof that Jupiter cannot be hot, for, if
it were, they would be dissociated into their consti-
tuent elements. Even at low temperatures the
ultra-violet radiation from the Sun gradually breaks
up the molecules both of ammonia and of marsh-
gas ; but in the absence of oxygen the break-up is
followed by a natural recombination. We thus have
a further proof of the absence of oxygen. From the
quantity of ammonia observed to be present in the
atmosphere of Jupiter, Dunham has deduced that
the temperature cannot be lower than about 185 F.
below zero, if it can be assumed that the atmosphere
consists mainly of hydrogen. This is in good agree-
ment with the measured temperature, when the
difficulties of the observation are considered.
Our conclusion that the solid surface of Jupiter
must be covered with a thick coating of ice was sug-
gested several years ago by Dr. Jeffreys, before we
had any positive information about the constitution
of the atmosphere of Jupiter. From various theo-
retical considerations, it has been concluded that the
real Jupiter which is not what we see in the tele-
scope is a solid body with an average density about
equal to that of the Earth. This rocky core occu-
66 WORLDS WITHOUT END
pies about one-eighth of the whole volume of Jupiter.
It is covered with a thick ice layer measuring some
16,000 miles (twice the diameter of the Earth) in
thickness. Outside this glacier coating there is a
very extensive atmospheric layer about 6,000 miles
deep. It is only the outer layer of this atmosphere
that we see in the telescope and from which we
draw our conclusions as to the size of Jupiter. The
great depth of this atmosphere has some interesting
consequences. The pressure at the bottom of it is
so enormous many thousands of tons to the square
inch that the gases are entirely liquefied. Higher
in the atmosphere the substances which are the most
refractory to liquefy, such as helium and hydrogen,
will exist as gases ; the more easily liquefied consti-
tuents will still be liquid. In this atmosphere at
these great pressures peculiar things can happen.
If, for instance, a mixture of hydrogen and helium
is subjected to great pressure the hydrogen will
liquefy while the helium still remains as a gas; but
the compressed helium is heavier than the liquid
hydrogen, and so it sinks below it, and we have the
strange phenomenon of the liquid hydrogen floating
on the helium gas. Probably there are analogous
conditions for other gases in other regions of the
atmosphere. It will be readily understood that an
atmosphere consisting of a mixture of gases and
liquids will be extremely unstable and violent dis-
turbances will occur in it. The changes that we
see occurring on Jupiter are probably due to these
disturbances. The Red Spot may be caused by
some commotion of this sort rather than by vol-
canic action. The immense thickness of the glacier
(a) SATURN WITH ITS SYSTEM OF RINGS.
) DIFFERENT ASPECTS OF THE RINGS 01 bA'iuRN, DEPENDING UPON
THE POSITION OF THE EARTH WITH RESPECT 1O THE RlNGS.
PLATE X.
[67
THE SUN'S FAMILY OF PLANETS 67
coating would seem to make volcanic action im-
possible.
The outer gaseous atmosphere of Jupiter above
the cloud layer, containing hydrogen, ammonia,
marsh-gas and other gases, is probably only a few
hundred miles in thickness. The low mean density
of Jupiter, not much greater than that of water and
less than that of any rock, now appears in its true
perspective, for it is not the true mean density of the
solid rock core but an average value for the solid
core, the ice layer and the deep atmosphere.
Jupiter appears to be a singularly uninviting
world, with its glacier-covered surface, its biting
cold and its particularly unpleasant atmosphere of
hydrogen, pungent ammonia and pestilential marsh-
gas. Its cloudy layer probably consists largely of
minute particles of frozen ammonia; the rays of the
Sun never penetrate through it, so that the glacier
surface of Jupiter is in perpetual darkness. Even
though conditions were otherwise suitable for man-
kind, its strong gravitational pull would have un-
pleasant consequences. A man who weighs 13
stone on the Earth would weigh more than 34
stone if he were transported to Jupiter; he could
move about only with great effort and would be in
serious danger of collapsing under his own weight.
Jupiter has a wealth of satellites : nine are known
and a tenth has recently been suspected. The four
brightest were discovered by Galileo in 1610 one
of the first-fruits of the invention of the telescope^
These four major satellites, which are named Io^
Europa, Ganymede and Callisto, are easily seen with
the aid of a pair of field-glasses. Their positions
68 WORLDS WITHOUT END
MOEDICEORVM. P/.A/STETARVM
ad inuuem, ft ad IOVM Conffitattonti t jutur# hi Mtruibuf Afuffi*
<t Jptile Sn MDCKIU * GALILEO G L farun
H, 'nee nan 'Ptrioettforwn attar urn mo1tavn,
/ r*_-
e t .jior.j tlOfttru Q.
^. f . Q.
X-r O
Q
O
O
O
3. Q
!Z>>// -^0. Q
>f:*. O
//+. o
JC^. O
Fio. 2. Observations of Jupiter's satellites by Galileo. (From Istoriae
Dimostrazioniintorno alle Macchie Solari, 1613.)
THE SUN'S FAMILY OF PLANETS 69
with respect to Jupiter change from night to night;
the changes in position can easily be followed with
field-glasses, the configurations each evening being
given in Whitaker's Almanack. Some of Galileo's
observations of the movements of the satellites are
reproduced in Fig. 2. Ganymede and Callisto
are the largest satellites in the solar system ; they are
rather larger than Mercury but weigh appreciably
less. Callisto has only one-quarter of the weight of
Ganymede and its average density is only six-
tenths that of water ; it probably consists largely of
ice and solid carbon dioxide. lo and Europa are
comparable to the Moon in size, and both they and
Ganymede appear to be masses of rock, like the
Moon. It is unlikely that any of these satellites has
an atmosphere. The remaining five satellites are
small bodies with diameters probably ranging from
about 15 to 100 miles.
The other planets can be passed over rather
briefly. Saturn is Jupiter's smaller brother, the
second largest planet in our system. Its volume is
more than 700 times that of the Earth, or somewhat
more than half that of Jupiter; for its size its weight
is low, being only 95 times that of the Earth, so
that its average density is only seven-tenths that
of water, less than that of any other planet. The
velocity of escape from Saturn is 24 miles a second,
so that Saturn has not lost its atmosphere. The low
average density can be accounted for by supposing
that the atmosphere is very extensive. As in the
case of Jupiter, the disc of the planet shows markings
in the form of belts which run in zones more or less
parallel to the equator (Plate X). They are less
JO WORLDS WITHOUT END
sharply defined and less variable than those of
Jupiter, but they are certainly atmospheric and not
surface markings. Occasionally well-defined spots
are seen. A large bright spot appeared in August
1933, which rapidly grew until it extended over
1 5,000 miles. It remained visible for several weeks ;
its appearance suggested that a mass of dust had
been thrown up by an eruption on the planet and
that the ejected matter was carried forward by winds
in the upper atmosphere, while still being fed from
below.
Saturn is the most flattened of all the planets.
The equatorial diameter is 76,000 miles; the polar
diameter is only 68,000 miles. This suggests rapid
rotation. As in the case of Jupiter, the rotation is
more rapid near the equator than in higher latitudes.
A complete rotation at the equator takes about 10 J
hours, so that Saturn rotates rather more slowly than
Jupiter. The planet is even colder than Jupiter
because of its greater distance from the Sun, and the
composition of its outer atmosphere appears to be
generally similar to that of Jupiter, except that am-
monia is less in evidence and marsh-gas is relatively
more prominent in the atmosphere of Saturn than
in that of Jupiter. As ammonia is more easily
condensed than marsh-gas, the low temperature of
Saturn has caused most of the ammonia to condense
out of its atmosphere. Without making any ob-
servations, we could have predicted that there would
be less ammonia in the atmosphere of Saturn than
in that of Jupiter.
Saturn has the same general constitution as
Jupiter a solid core of rock, overlaid by a thick
THE SUN S FAMILY OF PLANETS Jl
glacier coating and surrounded to a great height by a
dense atmosphere of compressed gases. But the at-
mosphere of Saturn is relatively much more extensive
than that of Jupiter. The depth of the atmosphere
of Jupiter is only about one-seventh of the whole
radius of the planet, whereas the depth of the atmo-
sphere of Saturn is nearly half the whole radius.
The solid core is about 28,000 miles in diameter,
or approximately 3^ times the diameter of the Earth.
This is covered with a coating of ice some 6,000
miles in depth and then above this the atmosphere
extends for a further 16,000 miles. The total
weight of this atmosphere is about equal to the
weight of the rocky core. It is because the atmo-
sphere of Saturn is relatively much more extensive
than the atmosphere of any of the l( other planets that
the average density of Saturn is so low.
The system of rings which surround Saturn makes
it unique and one of the most beautiful objects to
be seen in a telescope. The rings lie in the plane of
the equator and extend to a distance of 85,000 miles
from the centre of the planet. Their thickness is ex-
tremely small in comparison with their width, and
is probably not more than about 10 miles. Because
of this small thickness, the rings become quite in-
visible when the Earth passes through the plane of
the rings and they are edgewise on to us. The dis-
appearance of the rings was a puzzle to Galileo. In
1610 he had noticed an appendage on either side of
the planet, but the poor definition of his small tele-
scope did not permit him to recognise the true nature
of the rings. After a while he noticed that the append-
ages had disappeared, but a few years later they
72 WORLDS WITHOUT END
were again visible. The true explanation of the
appendages was given by Huyghens in 1655. Dif-
ferent aspects of the rings are illustrated in Plate X.
The rings are composed of a swarm of small frag-
ments, which move round Saturn in nearly circular
orbits. Clerk-Maxwell proved mathematically in
1856 that the rings could not be either solid or
liquid, because in either case they would break up
into a large number of small portions forming, as
it were, a swarm of little moons each moving round
Saturn under the influence of its gravitational pull.
There is no doubt that the rings actually do consist
of such a swarm of small moons. As in all cases of
bodies moving round a parent body under the com-
pelling force of its gravitation, the rule of the road is
that the fastest moving parts take the innermost
traffic lane and the slowest moving take the outer-
most lane. In 1895 Keeler obtained direct observa-
tional proof that the ring system consists of such a
swarm. By means of the spectroscope he was able
to detect the rotation of the rings and to show that
the inner edge rotates more rapidly than the outer.
If the rings were solid we should find exactly the
opposite, for the outer portion would then be moving
more rapidly than the inner. It occasionally hap-
pens that Saturn passes in front of a bright star;
when this occurs the star can be seen through the
rings, dimmed in brightness but never disappearing
completely.
The rings were produced by the disruption of a
satellite of Saturn which came too near to Saturn
and paid the penalty. It came so close that the
mighty pull of Saturn split it into pieces. These
THE SUN'S FAMILY OF PLANETS 73
pieces, colliding with one another, were still further
broken up until at length the fragments were scat-
tered in a ring all round Saturn. If any satellite
comes within a certain distance of its parent body,
the tidal forces exerted on the satellite are so great
that it must break up. This can be proved mathe-
matically, and it is of interest to note that the rings
of Saturn lie within the danger limit, but that in no
other case do any of the satellites of the planets in
the solar system come inside the danger zone within
which disruption is likely to occur.
Collisions between the separate fragments of
which the rings are composed must be numerous.
When two fragments collide, their velocities are
changed and they have to seek the traffic lanes
appropriate to their new speeds. In doing so, still
more collisions take place. But the swarm as a
whole goes moving on. The numerous collisions
must result in a continuous shower of fragments
falling on to Saturn. Thus Saturn is not only
extremely cold, with an unpleasant atmosphere, but
it is also subject to a continual bombardment from
these fragments : it is certainly a world which we
should avoid if we were desirous of seeking for a newi
habitation.
In addition to the rings, Saturn has a wealth of
satellites ; nine are known, equal in number to Jupi-
ter's system. The largest is appropriately called
Titan and is intermediate in size between our own
Moon and Mercury. The others are considerably
smaller, though larger than the five small satellites
of Jupiter. Little is known about the physical con-
ditions of any of these bodies; they are almost
74 WORLDS WITHOUT END
certainly devoid of any atmosphere and probably all
keep the same face continually towards Saturn.
The Sun, as seen from Saturn, would appear much
smaller than it does to us, but it would be in-
tensely bright. Most of the moons would appear
considerably larger than the Sun, though they would
not have its brilliancy. The view from Saturn, of
the nine moons of different sizes, circling round at
varied speeds, and of the rings, stretching as a vast
silvery arch across the sky, must be a fascinating
spectacle.
The next two planets, Uranus and Neptune, are
very similar to one another. Much smaller than
either Jupiter or Saturn, they are considerably
larger and heavier than the Earth, having about
sixty times its volume and sixteen times its weight.
Volume for volume, they are therefore much less
massive than the Earth. They are too far away for
the telescope to reveal very much; we should expect
such massive planets to possess considerable atmo-
spheres, and the spectroscope confirms this. We
again find very strong evidence of marsh-gas which,
apart from hydrogen, is probably the principal con-
stituent in their outer atmospheres. The Sun's
radiation at such great distances is very feeble,
and both planets are consequently extremely cold.
It is because they are so cold that no traces can be
found of ammonia, which is so prominent a con-
stituent of the outer atmosphere of Jupiter. There
is probably plenty of ammonia on both Uranus and
Neptune,but it is condensed out of their atmospheres
and exists only in the liquid or solid form. Their
low mean densities suggest that their atmospheres
THE SUN'S FAMILY OF PLANETS 75
must be very extensive; the depth of the atmosphere
of Uranus is probably about 3,000 miles, and of that
of Uranus is about 2,000 miles. As in \he case of
Jupiter and Saturn, both planets are covered with
a glacier coating which is some 6,000 miles in thick-
ness.
Uranus was the first planet to be discovered, for
Mercury, Venus, Jupiter and Saturn have been
known from time immemorial. It was found by
William Her^chel musician by profession, but
astronomer by inclination on March 13, 1781, with
a 7-inch reflecting telescope which he had con-
structed himself. Herschel found it by chance in
the course of a systematic survey of the entire visible
heavens in which he was engaged. He noticed
that it had a different appearance from a normal
star. A star appears only as a point of light in a
telescope, but this object showed a definite disc.
That he had found a new planet did not occur to
him; he described it in his astronomical journal as
a " curious either nebulous star or perhaps a
comet." When it was found to be moving relatively
to the stars it was thought to be a new species of
comet, without any tail. Herschel named it the
Georgium Sidus, in honour of the King, George the
Third. Further observations proved it to be a new
planet, nearly twice as far from the Sun as Saturn.
The name Uranus finally adopted for it was sug-
gested by Bode.
After the path of Uranus in the sky had been cal-
culated, it was found on looking back through earlier
observations that Uranus had been observed many
times previously, the earliest observations being one
76 WORLDS WITHOUT END
by Flamsteed, first Astronomer Royal, in the year
1690. But the quality of the early telescopes was
so poor that the difference in the appearance of
Uranus and of a star had not been noticed. These
earlier observations served a valuable purpose.
They not only helped to provide material from
which a more accurate orbit of Uranus could be
computed than would otherwise have been possible
until after an interval of many years, but they also
revealed unexpected discordances between the ob-
served positions and the computed positions at these
times.
It occurred at about the same time to two young
astronomers, J. C. Adams in England and Leverrier
in France, that there might be an unknown planet
which was exerting a gravitational pull on Uranus
and causing it to deviate from the expected path.
On July 3, 1841, Adams, whilst still an under-
graduate at Cambridge, made this note: " Formed
a design, in the beginning of this week, of investigat-
ing, as soon as possible after taking my degree, the
irregularities in the motion of Uranus, which are
yet unaccounted for, in order to find whether they
may be attributed to the action of an undiscovered
planet beyond it, and, if possible, thence to deter-
mine approximately the elements of its orbit, etc.,
which would probably lead to its discovery. 35 A
first solution of this problem Adams attempted in
1843, and in September 1845 he communicated his
final results to Professor Challis, at the Cambridge
Observatory. Challis set out to find the unknown
planet by determining the positions of all the stars
in the suspected region, and investigating whether
THE SUN'S FAMILY OF PLANETS 77
any one of these showed evidences of motion.
Challis made his observations faster than he could
map them, and had already observed the unknown
planet twice without realising it when it was dis-
covered by Galle at Berlin. Leverrier had com-
pleted his investigations a few months later than
Adams, and he wrote to Galle, at the Berlin Obser-
vatory, requesting him to search for a new planet in
exactly the same region of the sky in which Adams
had asked Challis to search. The Berlin astrono-
mers had the advantage of a new star chart covering
this particular region of the sky ; all they had to do
was to compare the chart with the sky and see
whether there was an interloper in the sky which
was not shown on the chart. Within half an hour
of commencing the search, on September 23, 1846,
Galle found such an interloper and further observa-
tions the next night showed that it was moving
amongst the stars and confirmed its planetary
character. Thus Adams and Leverrier can justly
share the honour of the discovery of Neptune.
Both Adams and Leverrier had assumed, for the
purpose of their calculations, that the distance of
the unknown planet was the distance to be expected
according to Bode's Law. This proved later to be
incorrect, Neptune being the first planet to show a
serious departure from Bode's Law. Consequently,
the shape of the orbit and the mass of Uranus, which
they predicted, were considerably in error. Never-
theless, the investigations gave, with sufficient
accuracy, the direction of the new planet from the
Sun during the time covered by the bulk of the
observations of Uranus. The direction in which
78 WORLDS WITHOUT END
telescopes should be pointed to find the new planet
was correctly indicated, and this enabled the planet
to be found, which was the main purpose of their
investigations.
Uranus possesses four rather small satellites; two
of these were discovered by William Herschel, a few
years after he had discovered Uranus; the other
two were found by Lassell in 1851. Lassell had
already in 1846 discovered the single satellite of
Neptune; this is brighter than any of the satellites
of Uranus, though it is at a considerably greater
distance from the Earth. It must be a body of some
size, probably larger than the Moon and about
equal in size to Mercury or to the two largest
satellites of Jupiter.
The most distant member of the solar system,
Pluto, was discovered in January 1930 by Tom-
baugh at the Flagstaff Observatory. The history
of the events which led to this discovery is of some
interest. There remain small discordances be-
tween the observed and the computed positions of
both Uranus and Neptune, after the disturbing
effects of the other known planets are allowed for.
These are no doubt due, in part at least, to the
shapes of the orbits of these two planets not being
known with absolute accuracy; it must be remem-
bered that Neptune was discovered only in 1846,
and, as Neptune takes nearly 165 years to go once
round its orbit, it has not completed one revolution
since discovery. It was, however, possible that the
discordances might be due in part to the disturbing
action of an unknown planet more distant than
Neptune. A detailed mathematical investigation
PHOTOGRAPH SHOWING STARS AND THREE MINOR
PLANETS.
The images of the stars are round. The jimages of the
minor planets are drawn out by their motion into short
streaks or trails.
(&) PHOTOGRAPHS OF PLUTO, MARCH 2 AND MARCH 5, 1930.
The two arrows point to the image of Pluto. The change
in the position of Pluto, arising from its orbital motion
round, the Sun, will be noticed.
PLATE XT.
78]
THE SUN'S FAMILY OF PLANETS 79
was made by Lowell in the year 1915. He derived
an orbit for the hypothetical unknown planet and
concluded that its weight was about 6| times that
of the Earth, or rather less than half the weight of
Uranus.
The planet was anticipated to be faint, and a
search was made by photographic methods in the
region of the sky in which Lowell's calculations in-
dicated that it should be. The new planet was dis-
covered as the direct result of this search. In
Plate XI, two photographs, obtained at Flagstaff
in March 1930, at an interval of 3 days, are
reproduced. The bright star is Delta Geminorum,
and the motion of Pluto, relative to the stars, in the
3 days is easily apparent. Further investiga-
tions have shown that the weight of Pluto is much
smaller than Lowell had estimated. Instead of
weighing 6^ times as much as the Earth, it has
probably less than one-tenth the weight of the
Earth. Further observations will be required
before any more definite statement can be made.
It is certain that a planet so small in weight will pro-
duce such minute deviations in the paths of Uranus
and Neptune that they will be swallowed up in the
errors of observation. In other words, Pluto could
not have been predicted. That a planet should be
found where Lowell's investigation indicated that a
planet should be looked for is one of those happy
coincidences which sometimes occur. Some time
ago a new faint comet was discovered ; the large tele-
scope at the Yerkes Observatory was set to observe
the comet, and there it was, right in the centre of the
field. But it was then discovered that an error had
8O WORLDS WITHOUT END
been made in setting the telescope and that a new,
hitherto unknown comet had been found ! Prob-
abilities are entirely against such happenings:
nevertheless, they sometimes occur.
Pluto must be a small planet, probably smaller
than the Moon. It is therefore unlikely to have any
atmosphere. Its temperature must be so low (prob-
ably about 230 G.) that all but the most refrac-
tory gases would be liquefied or even solidified. If
it is not mere barren rock, it must be covered with
a layer of ice, solid carbon dioxide, ammonia,
nitrogen and other substances. The mean distance
of Pluto from the Sun is about 3,675 million miles,
but its orbit is more elongated than that of any other
major planet, so that the actual distance can vary
by as much as 920 million miles. The distance to
be expected according to Bode's Law is about twice
the mean distance, so that this empirical law which
was much in error for Neptune is completely
wrong for Pluto. The time which Pluto takes to
travel once round its orbit is about 248 years. The
mean velocity in the orbit is about 3 miles a
second, less than one-tenth of that of Mercury.
The small size of Pluto suggests that it may really
be more akin to the minor planets than to the
planets proper, and it is possible that other bodies
will be found at about the same distance, forming
an outer zone of small planets. The discovery of
these small distant worlds will not be easy, because
of their faintness. Until such other bodies may be
found or until we have more accurate knowledge of
the size and weight of Pluto, it is rather idle to
speculate.
THE SUN'S FAMILY OF PLANETS 8l
It is of interest to enquire what the Sun and other
planets would look like from Pluto, the most remote
known member of the solar system. The Sun
would appear smaller than Jupiter appears to us
when at its nearest, and would therefore not show a
visible disc. The intensity of its light would be
about 300 times that of the r ull Moon on the Earth.
The Sun would therefore appear very much brighter
than any star and would look somewhat like an in-
tensely brilliant arc-light in the sky; the intensity
of the sunlight on Pluto would be about equal to
that of a 500 candle-power lamp at a distance of 10
feet, and so there would be plenty of light for the
purposes of everyday life if life could exist on such
a world. Neither Mercury, Mars, Uranus nor Nep-
tune could be seen from Pluto with the naked eye;
Venus and the Earth would be visible in the
absence of the Sun, but as they would always be
within about i from the Sun, they could not
be seen. Jupiter and Saturn would never be more
than 7 and 14 respectively from the Sun
and could therefore only be seen very shortly after
sunset or before sunrise : Jupiter would appear as a
star of about the fourth magnitude and Saturn as
a star of about the sixth magnitude, so that neither
would be very conspicuous in the sky of Pluto.
Pluto is thus not a desirable observation post for the
investigation of the solar system.
CHAPTER IV
LIFE IN OTHER
WORLDS
A QUESTION that I am often asked is whether life
exists in other worlds, either on one of the other
planets belonging to our own solar system, or pos-
sibly on a planet belonging to some other Sun.
Before attempting to give an answer to this question,
some assumptions must be made. The forms of life
which have evolved on our Earth are forms which
are presumably adapted to the varied conditions
which are found on the Earth or beneath the
ground or in rivers and seas. It is conceivable that
life may exist under conditions that differ widely
from any of which we have experience. Yet it is
difficult to believe that there can be life of any sort
on a world that is entirely devoid of an atmosphere
or on one that has an atmosphere containing no
oxygen. Water is probably also essential. Nor
does it seem probable that life can exist under great
extremes of temperature either intense heat or in-
tense cold, though we should be rash if we attempted
to state limits of temperature outside which life
could not exist.
Given that conditions on any world are such that
life is possible, can it be assumed that life must neces-
sarily exist there ? The conditions under which life
can exist may not necessarily be conditions under
which it is possible for it to originate. We really
know nothing about when or how life began on our
82
LIFE IN OTHER WORLDS 83
Earth. It is believed that the Earth and the other
planets condensed out of a long filament or tongue
of matter drawn out of the Sun by another star
passing near it. Whether or not this theory of the
birth of the planets is correct in detail, it seems
reasonably certain that the planets of the solar
system were formed in some way out of the Sun.
In the beginning there could have been no life on
the Earth or on any of the other planets, for no life
is possible on the Sun. The temperature of the Sun
is so high that all but a few of the simplest chemical
compounds are broken up into the constituent
atoms. A single living cell is a complicated struc-
ture from the chemical point of view and could not
exist on the Sun.
At some time, then, in the past history of the
Earth a history which we have already seen ex-
tends back somewhere about three thousand million
years life first made its appearance. Geologists
have found primitive marine invertebrates in rocks
of the Archean period, of an age of about 1,300
million years. Life may therefore have appeared
comparatively shortly after a solid crust formed on
the Earth. Are we to suppose that life developed
spontaneously, because conditions were suitable for
it, or that its appearance was due to a special act of
creation on the part of the Creator ? If we adopt
the latter viewpoint, further discussion becomes
superfluous, and for the sake of argument we assume
that life developed spontaneously. As to the con-
ditions which may be necessary for such spon-
taneous development we can perhaps argue that
since no biologist has ever succeeded, even with all
84 WORLDS WITHOUT END
the resources of modern laboratory technique, in
producing life of any sort a delicate balance
between various factors is essential before life will
appear.
If this is so, the way to further discussion is again
blocked because we are in ignorance as to the con-
ditions in which life will originate. All we are able
to do is to enquire whether conditions elsewhere in
the Universe are such that life could exist, recognis-
ing the limitation that even though conditions on
any world arc suitable for the support of life, it
does not necessarily follow that life exists on that
world.
Considering the solar system, it seems probable
that we can at once rule out of consideration most of
the planets and all of their satellites. Mercury
appears to be entirely devoid of any atmosphere;
moreover, the temperature of the face which is
turned to the Sun is about the temperature of
molten zinc, whilst the night face of the planet, in
the absence of any atmosphere, must be extremely
cold. Jupiter and Saturn have very extensive atmo-
spheres, but both observation and theory agree in
indicating a complete absence of oxygen and water-
vapour. The outer atmospheres of these planets
apparently consist predominantly of hydrogen,
ammonia and marsh-gas. The atmospheres are
many thousands of miles in depth and the pressures
in their lower portions are so great that the gases
are mostly liquefied. We cannot conceive that life
can be supported under such conditions, even if the
intense cold on the planets did not by itself make
life impossible. Conditions are even more extreme
LIFE IN OTHER WORLDS 85
on Uranus and Neptune; there is a greater degree
of cold, and in the outer atmospheres marsh-gas is
more prominent than ammonia merely because the
ammonia cannot remain gaseous at such low
temperatures. Pluto is a small body, probably
without any atmosphere and with a temperature
so low that neither nitrogen nor oxygen could
exist on it in a gaseous condition. We cannot
regard it as at all probable that any form of life,
either vegetable or animal, could exist on any of
these worlds.
Is it possible that life may exist on one or other of
the larger satellites in the solar system ? We can
at once disregard all the small satellites, because
they must long ago have totally lost their atmo-
spheres. We have seen that our own Moon is a
dead world, completely devoid of any atmosphere.
The only other satellites which perhaps demand
some consideration are the two largest satellites of
Jupiter, Callisto and Ganymede, and the largest
satellite of Saturn, Titan. The two satellites of
Jupiter arc not very different in size from Mercury,
but they are considerably less massive. Ganymede
weighs rather more than half as much as Mercury.
Callisto, though about equal in size to Ganymede,
has less than one-third of its weight. The pull of
gravitation at the surface of either of these satellites
is therefore less than at the surface of Mercury. It
follows that the velocity of escape from each of these
satellites is lower than the velocity of escape from
Mercury. We should not therefore expect to find
any atmosphere on either of them. Direct observa-
tional evidence seems to support this; the surfaces
86 WORLDS WITHOUT END
of both satellites can be seen, and spectroscopic ob-
servations give no evidence of any atmosphere on
either. Similar remarks apply to Titan ; it is smaller
and not so massive as Mercury, and there is no in-
dication of any atmosphere.
Venus and Mars remain for consideration. Venus
keeps her secrets well guarded. Her face is per-
manently hidden from us by a thick layer of cloud
which even the haze-penetrating infra-red rays can-
not pierce. Neither oxygen nor water-vapour has
yet been detected in her atmosphere. But we can-
not at present regard the failure to detect them as
absolutely conclusive proof of their absence. It
may be that the clouds on Venus extend to such a
great height in her atmosphere that the sunlight is
reflected back from them before it has had the
opportunity to pass through a thickness of atmo-
sphere great enough for the absorption by oxygen or
water-vapour to be of sufficient amount to be cap-
able of detection by the tests which are at our dis-
posal. These tests are not capable of detecting
very small amounts of either oxygen or water-
vapour.
The clouds presumably provide direct evidence
of the presence of water-vapour; the failure to
detect water-vapour above the cloud-layer is prob-
ably due to all the water-vapour having been
condensed out of the upper atmosphere.
The only positive information as yet obtained
about the atmosphere of Venus is a definite proof
of the presence of carbon dioxide. This is present
in much larger amount in the atmosphere of Venus
than in our own atmosphere. The amount of car-
LIFE IN OTHER WORLDS 87
bon dioxide above the cloud-layer is so considerable
that it is possible with some justification to argue
that the failure to detect oxygen is really due to its
total or almost total absence. We should expect to
find both water-vapour and carbon dioxide present
on a cooling planet of the size of Venus. As the
molten rocky mass cooled and began to solidify,
large quantities of water-vapour, carbon dioxide
and other gases would be evolved from it. When
the surface had cooled sufficiently, the water- vapour
would condense and oceans would cover much of
the surface ; the atmosphere would contain the inert
gases, nitrogen, argon and neon, and a large amount
of carbon dioxide. It is unlikely that oxygen would
be present at this stage in any appreciable amount;
most terrestrial volcanic rocks are incompletely
oxidised and, if free oxygen had been present in the
Earth's atmosphere when the crust solidified, these
rocks would have greedily laid hold upon it. The
present abundance of oxygen in the Earth's atmo-
sphere is probably due to the action of vegetation,
which steadily extracts carbon dioxide from the air
and returns oxygen in its stead. The carbon dioxide
supply is in turn replenished by processes such as
respiration, combustion and the decay of vegetable
matter. Life on the Earth probably started at a
time when the atmosphere contained comparatively
little oxygen and was in the form of primitive
vegetable life. The vegetable life, as it developed,
gradually increased the amount of oxygen in the
atmosphere, and conditions at length became suit-
able for animal life. Venus is apparently at a stage
which the Earth passed through millions of years
88 WORLDS WITHOUT END
ago, having an abundance of carbon dioxide but
little, if any, oxygen.
So far as the temperature of Venus is concerned,
conditions are probably favourable. Though
warmer than the Earth, it is not excessively warmer ;
there is probably a hot, humid climate which, pro-
vided only that free oxygen is present in sufficient
quantity, should be quite capable of supporting life.
There is nevertheless the possibility that the
blanketing effect of the carbon dioxide in the atmo-
sphere may raise the temperature at the surface
very considerably, and in the equatorial regions
even to the boiling-point of water. Conditions suit-
able for life might therefore develop first in the
polar regions.
What type of life there may be if it exists we can
only hazard a guess. On the Earth we find the
extremes of climate within which it seems possible
for the human race to exist. Is this the result of
chance, or is it not rather the result of evolution,
conditioned by the prevailing physical conditions ?
The conditions on Venus now are possibly not
greatly unlike the conditions which prevailed on the
Earth at an earlier stage in its history, when the
climate was warmer and more humid than it now
is and when vast swamps covered the Earth. If so,
it is possible that a form of life adapted to such con-
ditions, somewhat akin to the swamp-dwelling
mammoths which once lived on our Earth, may
have evolved on Venus.
The balance of evidence would seem, however,
to be in favour of the view that the amount of
oxygen in the atmosphere of Venus is not at present
LIFE IN OTHER WORLDS 89
sufficient for the support of any life, except possibly
primitive vegetable life.
We know more about Mars than about Venus
because we can see its surface. We have found that
Mars possesses a thin atmosphere, in which both
oxygen and water-vapour may be assumed to be
present. The atmosphere is scanty, it is true, the
amount of oxygen being considerably less than on
the Earth. We know that man can become ac-
climatised in a short time to live at considerable
heights, in a very thin atmosphere, and it is quite
conceivable that, by the process of evolution, forms
of life may have developed on Mars which can exist
in the attenuated atmosphere of that planet. A
great impetus to the belief that Mars may be in-
habited by a race of intelligent beings was given by
the publication of Professor Lowell's book entitled
Mars as the Abode of Life. Lowell studied the surface
of Mars under the exceptionally favourable climatic
conditions, and at the high altitude of the Flagstaff
Observatory in Arizona. He observed a large
number of the fine straight markings which had
been discovered by Schiaparelli and named
" canals " by him. He found that the canals cov-
ered the surface of the planet in a complex network,
forming a fairly regular geometrical design. Many
of them extended for thousands of miles. At vari-
ous points, called oases by Lowell, several canals
intersected. The oases are darker regions which
may cover a large area of several thousand square
miles. There is no doubt as to the existence of these
darker patches or oases. Many of the canals were
believed by Lowell to be double, appearing as two
go WORLDS WITHOUT END
fine equidistant lines, somewhere about one or two
hundred miles apart. From a study of many draw-
ings of the planet, Lowell concluded that the canals
changed their appearance with the Martian seasons.
Usually faint or invisible during the spring, they
gradually become more prominent as the polar cap
shrinks, appearing to grow towards the equator at
a rate of some fifty >miles a day, and often extending
beyond the equator well into the opposite hemi-
sphere. This appearance gradually fades away
and then, half a Martian year later, the canals
are again seen to be growing towards the equator,
but this time they are extending from the opposite
polar cap.
Lowell interpreted these observations as providing
evidence of the activities of a race of intelligent
beings who were engaged in a desperate struggle for
existence on an arid world, where there was little
or no rainfall. He supposed that the canals were
artificial channels constructed to bring water from
the melting polar caps towards the equator, for pur-
poses of irrigation, in regions where there was little
or no rainfall, and he believed that the darkening
of the canals was due to water flowing along them.
The oases were interpreted as large irrigated areas ;
here were to be found the centres of population, the
inhabitants being driven from the arid regions by
the scarcity of water.
Lowell asserted that some of the canals dis-
appeared for several years at a time and new canals
appeared where none had previously been seen.
This was apparently due to some of the canals silting
up from time to time and after a lapse of years being
LIFE IN OTHER WORLDS Ql
opened up again, or even to new canals being con-
structed.
To cover a planet with a network of " canals," or
water channels, each of which must be at least fifty
miles wide, extending for hundreds or thousands
of miles, would be an immense undertaking, which
might not unfairly be attributed to a race of beings
in a higher state of civilisation than our own. Such
was Lowell's opinion. The theory is a fascinating
one, if only we could be certain that the canals are
really there !
But another equally skilled observer, Professor
Barnard, possessed of remarkable acuity of vision,
who had, moreover, the advantage of years of obser-
vations with some of the largest telescopes in
America, was unable to see the fine geometrical net-
work of lines described and drawn by Professor
Lowell. At times Barnard could see " short,
diffused, hazy lines, running between several of the
small very black spots which abound in this region
of the planet." Observing Mars with the great
Go-inch reflector at the Mount Wilson Observatory
in California, Barnard said that it gave him " the
impression of a globe whose entire surface had been
tinted a slight pink colour, on which the dark details
had been painted with a greyish coloured paint,
applied with a very poor brush, producing a
shredded or streaky and wispy effect in the darker
regions."
Such differences between competent observers are
Erobably to be attributed to complex psychological
ictors. The observer gazes intently at a highly
magnified image of the planet; the detail which he
Q2 WOKLDS WITHOUT END
strives to see is at the limit of vision. The slightest
tremor in the atmosphere is magnified by the tele-
scope, with the result that the image becomes some-
what unsteady. For a moment it settles down, if
conditions are of the best, to an almost perfect
image; but before the eye can see the fine detail,
another tremor has spoilt the image. Faint details
which the eye has seen momentarily (and the sur-
face of Mars is certainly rich in fine detail) the mind
tends subconsciously to connect by straight lines.
The final picture, though as faithful a representation
as the observer can draw, may bear but a faint re-
semblance to the real object or even to the picture
drawn by another observer. We may compare the
early drawings of some of the spiral nebulae by Lord
Rosse with modern photographs of the same nebulae
(Plate XXX). The spiral structure of these nebulae
was discovered by Lord Rosse with his great 6-foot
reflector about 1850. Though the spiral structure
is certainly there, a comparison of the details of the
drawings and photographs shows little resemblance;
the spiral formation is not nearly so much in evidence
in the photographs as in the drawings.
The balance of astronomical opinion is against
the objective reality of the system of canals depicted
by Lowell. If we do not accept the objective reality
of the canals, we must also abandon his graphic pic-
ture of the race of intelligent beings struggling
desperately for existence. In answer to the ques-
tion whether Mars is inhabited, we can only point
to probabilities. We have seen that the atmosphere
of Mars, though rather rare, could probably sup-
port life. The conditions as regards temperature,
LIFE IN OTHER WORLDS 93
though not very pleasant, are not such as to make
life impossible. Near noon, in the equatorial re-
gions, the temperature rises to about 50 F.; but in
the afternoon, as the Sun gets lower in the heavens,
it rapidly falls and after sunset the cold becomes in-
tense, the minimum temperature at night probably
reaching 130 F. below zero. This enormous daily
range of temperature and the rapidity of the changes
must prove very trying for any form of life.
It would seem probable that such life as may exist
must protect itself from the bitter cold at night by
taking refuge in caves or holes in the ground.
Vegetable life is most likely to be in the form of
lichens or mosses.
Our conclusion is that life of any sort is unlikely
to exist anywhere in the solar system, apart from
our own Earth, except possibly on Venus or Mars.
Conditions on neither of these planets appear par-
ticularly favourable, but, in view of our present in-
complete knowledge of these conditions, the possi-
bility of life cannot be ruled out. The conditions
on Venus may be compared with those on our Earth
many million years ago ; those on Mars are perhaps
similar to those that will prevail on the Earth
many million years hence, when the Sun is cooler
than it is now and when our oxygen supplies may
have been largely used up.
The possibility of life elsewhere in the Universe
than in our solar system remains for brief con-
sideration. We can only deal with probabilities,
for we have no means of detecting whether any of
the stars, other than the Sun, has a system of
planets; if such systems exist, we can never know
94 WORLDS WITHOUT END
anything about the actual physical conditions which
prevail on them. It seems probable that planets
are only born when two stars pass so close to one
another that enormous tides are raised on their sur-
faces; great jets of matter are then drawn out of
each star, which break up and finally condense into
a system of planets.
This theory of the origin of a planetary system is
not without its difficulties, but for want of any more
plausible theory we may accept it provisionally. It
would seem to imply that the birth of planets must
be an event of extremely rare occurrence. On the
basis of the random wandering of the stars, as we
find them distributed in the neighbourhood of the
Sun, a sufficiently close approach of two stars can
happen on the average only about once in every five
thousand million years.
This argument was a valid one before the expan-
sion of the Cosmos had been discovered. But it is
now known that the Cosmos is expanding at such
a rate that in about 1,300 million years the distance
of any one stellar universe from any other universe
is doubled. It may be thought that this is an ex-
tremely slow rate of expansion. But when we con-
sider that our Earth is some 3,000 million years old,
it will be seen that when it was born distances were
less than one-quarter of what they are now, pro-
vided that we can assume the expansion to have
been uniform in the meantime. It is therefore more
than probable that close approaches of two stars
were formerly far more frequent than they are now.
In that case, there must have been a progressive
diminution in the birth-rate of planets. In our
LIFE IN OTHER WORLDS 95
Universe, which contains about 200 thousand mil-
lion stars, it may consequently be regarded as not
improbable that there are many stars even though
an extremely small percentage of the total number
which have planetary systems.
But even when this possibility is granted, life is
not likely to develop on any planet unless there
be a suitable combination of various factors. The
planet must be at such a distance from its parent
Sun whose output of radiation in the form of heat
and light may be either very much greater or much
smaller than the output from our Sun that its
temperature is neither too high nor too low. It
must be of sufficient size and weight to be able to
retain its atmosphere; it is probably necessary that
it should not be so large and massive as to retain the
hydrogen which would initially predominate, for
then the whole of the oxygen would be likely to
combine with some of the hydrogen to form water,
leaving an atmosphere entirely devoid of oxygen.
Conditions suitable for the support of life are not
likely therefore to prevail in more than a small pro-
portion of planetary bodies.
Our own Universe is one of many millions. It is
estimated that, up to the distance to which the most
powerful telescope yet made can probe, there are
some 75 million universes, generally comparable in
size and in other respects. There must then in the
Cosmos as a whole be many millions of stars with
families of planets belonging to them. Though we
cannot make any definite assertion, it is reasonable
to suppose that conditions on some at least of these
must be such that life is possible. If we can assume
96 WORLDS WITHOUT END
that life will appear when conditions are suitable for
it, the balance of evidence must be in favour of the
conclusion that the probability is that there are
many other worlds scattered through the Cosmos
on which life exists.
CHAPTER V
COMETS AND
SHOOTING STARS
THE motions of the p'anets are known with such
accuracy that we can predict in what part of the
sky Jupiter or Mars or any other planet will be
found in a hundred years' or in a thousand years'
time. But there are other members of the solar
system whose appearance cannot generally be pre-
dicted; these members are comets and shooting
stars. We shall see that there is a reason for con-
sidering the two types of object together.
From the very earliest ages, comets and shooting
stars have attracted widespread attention. Shoot-
ing stars were thought to be stars which had dropped
from the sphere on which all the stars were supposed
to be fixed and had fallen to Earth. Comets were
also believed by many primitive people to be stars
with a long hairy tail: the word " comet " means
hairy star. In the old Chinese records the appear-
ance of comets and new stars are mixed up together,
and it is not possible in many cases now to decide
whether the record refers to a comet or to a new
star. But Aristotle and his followers thought that
comets were exhalations from the Earth, which had
caught fire in the upper regions of the atmosphere.
This view is not really so surprising as it may now
seem to us. The tail of a comet is longest and the
comet itself is at its brightest when it is nearest to
the Sun. The comet can then be seen only shortly
7 97
98 WORLDS WITHOUT END
after sunset or before sunrise and, since the tail of a
comet always points in the direction away from the
Sun, the comet will then be seen with its tail point-
ing upwards and appearing like a rising flame, the
head of the comet possibly being below the horizon
and therefore invisible.
It is not surprising that these strange apparitions,
appearing suddenly and at rare intervals, were re-
garded as omens of misfortune, the harbingers of
famine, pestilence, wars or the impending " death
of Princes."
" When beggars die, there are no comets seen,
The Heavens themselves blaze forth the death of
Princes"
says Calpurnia in Julius Casar. For a great comet,
with its tail blazing forth like a great fiery arch
across the sky, is a magnificent spectacle, and it
is easy to understand how people, ignorant of the
true nature of a comet, could be terrified by so
unusual a sight. Many instances of the association,
in the popular mind, of the appearance of a comet
with a tragedy or misfortune could be mentioned.
One will suffice. Daniel Defoe in his Journal of the
Plague Tear mentions how much the alarm in London
was increased, when the great plague of the year
1665 was in its early stages, because a blazing comet
had appeared some weeks earlier. This comet was
of a " dull, languid colour " and its " motion very
heavy, solemn and slow, 55 the interpretation being
that " a heavy judgement, slow but severe, terrible
and frightful, was already begun. 53 In 1666, a little
COMETS AND SHOOTING STARS QQ
before the Great Fire, another comet appeared
" bright and sparkling or, as others said, flaming,
and its motion swift and furious," portending a
judgement " sudden, swift and fiery." He men-
tions that many remarked " that those two comets
passed directly over the city, and so very near the
houses that it was plain they imported something
peculiar to the city alone."
Comets, like the planets, move round the Sun
under the controlling force of the Sun's gravita-
tional attraction. But whereas the planets move
in orbits which do not in general differ very much
from circles, the orbits of comets are usually ex-
tremely elongated. A comet is in general only visible
during the portion of its path in which it is nearest
to the Sun. The speed of the comet is greatest when
it is at its nearest to the Sun because the comet, like
the planets, obeys Kepler's Law of Equal Areas
(P- 39) J it therefore rushes very quickly round this
portion of its orbit, and as it moves away from the
Sun its speed becomes slower and slower. So
though a comet may take several hundred years to
go once round its orbit, the time during which the
comet is visible may be only a few weeks or months.
It used to be thought that a comet appeared,
moved past the Sun, and disappeared, never to
return again. The proof that a comet may return
and be seen again was due to Halley. He found
that the paths of the bright comets which had ap-
peared in the years 1531, 1607 and 1682 were
almost identical. He concluded that it was the
same comet which had appeared in these three
years, and that the bright comets seen in the years
IOO WORLDS WITHOUT END
1456 and 1305 were also the same comet. At the
end of his account he added, " Hence I think I may
venture to foretel, that it will return again in the
year 1758." Halley died in the year 1742. Little
attention had been paid to his prediction at the time,
but it gradually began to be realised that the return
of the comet, in accordance with this prediction,
would provide a remarkable vindication of Newton's
law of universal gravitation. As the time for the
return drew near there was great excitement among
the philosophers of the age. On Christmas Eve,
1 758, the comet was seen by a farmer in Saxony and
Halley's prediction was triumphantly verified.
Halley's comet is the most famous of all comets.
Records of its appearance at every return back to
the year A.D. 989 have been found and, with a few
gaps, it has been traced back to the second century
B.C. Josephus recorded " a fiery sword hanging
over Jerusalem " at the great siege in 170 B.C., an
omen foretelling the destruction of Jerusalem. This
was Halley's comet. The comet appeared in March
1066, and men thought it predicted the success of
the Norman invaders; the English were frightened
and the Normans encouraged by this sign in the
heavens. In the famous Bayeux tapestry, made by
Queen Matilda to celebrate the victory of her hus-
band, there appears a representation of the comet
and of the wonder and dismay of the people (Fig. 3).
Over the picture is the legend " Isti mirant Stella."
In 1456 the comet was of extraordinary brilliancy,
its tail stretching half-way 'across the sky; in that
year Constantinople was taken by Mahomed II,
who proceeded to advance westward into Europe
COMETS AND SHOOTING STARS IOI
and to spread terror throughout Christendom.
The most recent appearance of the comet was in
the year 1910; it was not then favourably placed
for observation in England, but was a magnificent
spectacle in the southern sky.
Many other periodic comets, as a comet which
returns again is called, are known. Perhaps the
most famous is Encke's comet, which was first seen
in 1786 and returns every 3^ years. Not a single
return since its discovery has been missed, but the
comet is not a striking object in the sky and can only
be seen with the aid of a telescope. Sometimes a
comet in the course of its wanderings happens to
pass very near to Jupiter or Saturn; these massive
planets then pull the comet out of its path and may
entirely change the shape and size of its orbit. A
comet, known as Brooks's Comet, which used to take
30 years to travel round its orbit, passed very near
to Jupiter in 1886, and the gravitational attraction
of Jupiter was so great that the direction of motion
of the comet was changed to such an extent that it
now takes only 7 years to complete its orbit.
Not all comets are periodic comets in the sense
that they have been observed at more than one
return. It is nevertheless probable that every comet
will return in time, though the orbits of many are so
extremely elongated that several hundreds or even
thousands of years may be taken by the comet in
going once round its path. We might perhaps think
that there may be comets wandering about in space
and having no connection with the solar system ; if
such is the case, occasionally one of these comets
may come under the influence of the Sun's gravita-
IO2 WORLDS WITHOUT END
tion and be drawn towards the Sun. The comet
will then pass near the Sun and may or may not be
captured, depending upon its speed and the close-
ness of approach. If not captured, it will go off
again into space, never to return. We have no cer-
tain evidence, however, that any comet of which we
have record has ever come from outside the solar
system in this way. They all appear to be per-
manent members of the system.
Before we attempt to answer the question " What
is a comet ? " we must describe the appearance of a
typical comet. When at a considerable distance
from the Sun, and visible only in a telescope, a
comet usually appears as a faint nebulous or hazy
cloud in which a central brighter condensation or
nucleus may sometimes be seen. No tail is visible
at this stage. As the comet approaches the Sun,
this head or coma takes on a more clearly defined
outline, but it never becomes absolutely sharp.
The nucleus usually appears as a bright star-like
point near the centre of the coma, though sometimes
no nucleus can be seen. The photograph of Hal-
ley's comet reproduced in Plate XII shows the coma
and its nucleus. From the coma there now begins
to stream out a nebulous-looking tail (or tails, for
some comets have several distinct tails) ; the tail
seems to spring from the nucleus and its brightness
decreases rapidly with distance from the nucleus. As
the comet gets still nearer to the Sun, the tail grows
in length and becomes brighter. The structure of
the tail is often very complicated and liable to rapid
changes. The tail is always directed away from the
Sun, as though it is repelled by some force emanat-
^
(a) HALEY'S COMET (1910), SHOWING NUCLEUS AND HEAD.
In taking the photograph, the telescope has ioilowed the
movement "of the comet and the star images have been drawn
out into streaks,
102]
() BROOKS'S COMET (1911).
The structure of the tail is very complex.
PLATE XII.
MOREIIOTJSB'S COMET ( 1 908) .
BROOKS'S COMET (1903).
Two photographs, taken with an interval of 24 hours,
have been superposed to show that the tail does not
stream behind the comet along the direction in which
the comet is moving.
PLATE XIII.
[103
COMETS AND SHOOTING STARS 103
ing from the Sun. It does not therefore follow the
comet, like smoke from an engine, and the popular
idea of a comet as a star with a tail of fire trailing
behind it is entirely erroneous. Some photographs
taken by Professor Barnard show the effect well.
He obtained two photographs of a comet at an in-
terval of 24 hours; during this interval the comet
had time to move appreciably. He then combined
the two pictures so that the stars on the one fell
exactly on the corresponding stars on the other.
The combined picture is reproduced in Plate XIII ;
the movement of the head of the comet in the in-
terval between the two photographs is clearly shown,
but it will be noticed that the tail does not point
along the direction in which the comet is moving.
In 1843 a bright comet appeared which came to
within 80,000 miles from the Sun. It was then
moving with a tremendous speed of several hundreds
of miles per second and swung half-way round the
Sun in about a couple of hours. The tail was at
least 300 million miles in length and as the comet
swung round the Sun the end of the tail moved with
a speed considerably greater than half the speed of
light; such a speed is almost in itself sufficient to
prove that the tail could not be a rigid appendage
of the comet.
The tail of a comet consists of minute particles
of matter which are blown out from the nucleus by
the pressure of the radiation from the Sun. When
light falls on any body, it exerts a pressure upon it
just as the waves of the sea press against any obstacle
on which they impinge. The more intense the light
the greater the pressure. The pressure of light was
IO4 WORLDS WITHOUT END
first predicted theoretically by Clerk-Maxwell and
was subsequently demonstrated by delicate experi-
ments in the laboratory. We shall see later that the
pressure of radiation plays an extremely important
role in the interiors of stars, where the intensity of
the radiation is very high. The nearer a comet
approaches to the Sun, the stronger is the radiation
falling upon it and the greater the pressure. Every
small particle in the head of the comet is acted upon
by two opposing forces the pressure of radiation
which is trying to blow the particle away and the
gravitational pull of the comet which is trying to hold
it back.
Suppose that for a particle of a certain size these
forces are just balancing one another; now let us
consider what will happen in the case of a particle
of half the linear dimensions. The pressure of
radiation has only one-quarter of the surface to act
on and is therefore one-fourth of the pressure on the
larger particle. But as the smaller particle has only
one-eighth of the weight of the larger, the gravita-
tional pull of the comet is less in the same propor-
tion. It is clear that for this particle, therefore, the
pressure of radiation will be twice as great as the
pull of gravitation and the particle will therefore be
blown out of the head, in the direction away from
the Sun. There will consequently be a continuous
stream of minute particles blown out from the
comet.
What happens as the comet moves round the Sun
can be imagined by thinking- of the analogy of a
spray of water coming out of the nozzle of a hose as
the hose is swung round. The particles of water
COMETS AND SHOOTING STARS 105
are continuously moving along the spray and the
spray is being continuously fed from the nozzle.
We see also why the tail of a comet grows as the
comet nears the Sun and becomes smaller again as
the comet recedes; this is a direct consequence of
the greater intensity of the Sun's radiation when the
comet is nearest to the Sun.
Comets differ very much in the size of their heads
and of their nuclei. The heads themselves may be
very large. The head of the celebrated comet of
1 81 1 was larger than the Sun. The nucleus is
usually not more than a few hundreds of miles, or
in exceptional cases a few thousand miles across.
The tail often extends over many millions of miles,
with a volume many times greater than that of the
Sun. It is surprising therefore to find that the
weights of comets are insignificant judged, of
course, not by terrestrial standards but in compari-
son with the weights of the planets. We have men-
tioned that sometimes a comet will pass so near to
a planet that the pull of the planet on the comet
entirely alters its path; yet it has never been possible
to detect any effect of the pull of a comet on a
planet. It is probable that the weight of every
comet is less than a one-millionth part of the
Earth's weight; even so, the comet may weigh many
millions of tons.
The weight being so small and the head fre-
quently large, the average density of matter in the
head must be very small probably of the order of
the density of the residual air in a chamber exhausted
by a good air-pump. The average density in the
tail must be much smaller even than this. Ten
IO6 WORLDS WITHOUT END
thousand cubic miles of tail will not contain more
matter than one cubic inch of ordinary air. It is
not surprising, therefore, that when a comet passes
in front of a faint star, the star may be seen shining
through the head without any perceptible diminu-
tion in brightness. We must not suppose that the
head of the comet actually consists of a very attenu-
ated gas. If a comet were gaseous, its gravita-
tional pull would not be sufficient to hold it together
against the tendency of a gas to diffuse outwards in
all directions, and the comet would be rapidly dissi-
pated away into space. The head is a loose collec-
tion of rocks, stones and smaller particles which
may range in size from large masses weighing mil-
lions of tons to the very finest dust, the larger
masses being widely separated from one another.
In May 1910, Halley's comet passed directly be-
tween us and the Sun at a distance of about 15 mil-
lion miles. The tail was at least 20 million miles
long and was pointing directly towards us, so that
we probably passed through it. Some alarm was
caused by the publicity given to this passage, because
the spectra of comets show the presence of the very
poisonous gas, cyanogen. But so attenuated is a
comet's tail that we had no indication at all that
the Earth was passing through the tail. Astrono-
mers examined the Sun carefully to detect the pas-
sage of the head of the comet across its face, but no
trace of it could be seen. As the head is mainly
empty space, this was not at all surprising.
A few years ago, in June 1921, the Earth escaped
collision with Pons-Winnecke's comet by a few days
only. Such a collision, if the comet were a large
COMETS AND SHOOTING STARS IOJ
one, would be serious for the region of the Earth
where the impact occurred. But for a comet of an
average small size the effects would be relatively
local. In Arizona there is a cup-shaped crater
about a mile across and 600 feet deep which was
formed by the collision of either a small comet or
of a very large meteor with the Earth. Many pieces
of meteoric iron, of all sizes up to lumps weighing a
few tons, have been found within an area extending
to a distance of several miles from the crater.
Occasionally a comet is seen to split up into two
or more fragments. In 1889 Brooks's comet was
observed to split up into two portions, which slowly
separated ; the disruption was apparently a result of
the comet passing very close to Jupiter. Biela's
comet provides another example. This comet was
discovered in 1826 ; it proved to be a periodic comet
with a period of about 6f years. At the return in
1846, it had the appearance of a normal comet when
it was first seen, but shortly afterwards it divided
into two parts which gradually separated from each
other. In 1852 the twin comets appeared again,
but by then their separation had greatly increased.
Neither comet was ever seen afterwards, though
they were looked for at successive returns. But in
1872, when the Earth passed the track of the lost
comet, there was a fine display of shooting stars.
Whenever the Earth passes through the track of the
vanished comets, this display is observed, but not
always with the same brilliancy. What has hap-
pened is that the comet has completely disrupted
and the fragments have become scattered along its
orbit.
108 WORLDS WITHOUT END
The most famous of all meteor showers is the
Leonid shower, which appears about the middle of
November. On November 13 and 14, 1866, there
was a particularly brilliant display when many
thousands of shooting stars flashed across the sky,
radiating out in all directions from a definite centre.
Frequently, several could be seen at the same in-
stant, and many left bright luminous trails which
lingered in the sky for several minutes. These
meteors follow the track of the great comet of 1 866,
which has disintegrated. It takes this swarm about
3 years to pass any given point, but 33^ years
to go completely round its orbit, so that the frag-
ments of this comet have not yet become by any
means uniformly spread round the orbit.
The appearance of the ordinary shooting star is
caused by a solid fragment of matter entering the
Earth's atmosphere at a high speed. As the meteor
moves rapidly through the atmosphere, a cap of air
forms in front of it and becomes heated by friction
with the surrounding air through which it passes.
The heat is communicated to the surface layers of
the meteor itself, which become vaporised. The
envelope of heated air and vaporised material may
trail behind the meteor and leave a luminous streak
persisting for several seconds. A meteor is usually
first seen when it is at a height of about 80 miles
above the surface of the Earth, and remains visible
until the height has decreased to about 50 miles.
The length of the path may be as great as several
hundred miles, but the apparent length of the path
in the sky depends upon the direction in which the
meteor is moving. If it is moving towards us, the
COMETS AND SHOOTING STARS
path will appear short and the motion slow, but if
moving across the line of sight, the path will appear
long and the motion rapid.
The average shooting star is very small ; the
range in size is probably from a pea to a grain of
sand. These small bodies become completely
vaporised in their passage through the atmosphere
and never reach the ground. It is fortunate that
the Earth has a protective atmosphere, for other-
wise the effects produced by small meteors moving
with a speed many times greater than that of a rifle
bullet would be serious. It has been estimated that
the total number of meteors which enter the Earth's
atmosphere in the course of a day is several mil-
lions ; many of these would pass harmlessly by, if
our Earth had no atmosphere, but a sufficient num-
ber would reach the surface to be distinctly un-
pleasant for us.
When the Earth crosses the track of a disrupted
comet, each fragment entering the atmosphere ap-
pears as a shooting star and we have what is called
a meteor shower. Most of the isolated shooting
stars are caused by particles which do not belong to
the solar system, but come from outer space and
enter the Earth's atmosphere with velocities fre-
quently greater than 60 miles a second.
Some of the fragments may be larger than the
small particles which appear as the typical shooting
star. We may then have the appearance of a fire-
ball a brilliant ball of light, easily visible in broad
daylight, which is usually dissipated in an explosion
or a series of explosions of considerable violence, to
the accompaniment of a loud report. Fragments
IIO WORLDS WITHOUT END
may then fall to Earth as meteorites. Meteorites
frequently weigh many tons and usually consist of
masses of limestone, magnesia or siliceous stone,
generally mixed with globules of iron. A small
percentage of meteorites consist of nearly pure
iron.
The greatest meteor fall of recent years occurred
on June 30, 1908, in a remote part of Siberia, near
Vanovara on the Stony Tunguska River, about 700
miles north of Lake Baikal. There was little in-
formation available about the details of the fall
until an expedition under Professor Kulik was sent
by the U.S.S.R. Academy of Sciences in 1927 to
investigate the locality. The actual place in which
the meteorite fell was a crater-like valley. It was
found that in this valley and on the surrounding
hills all the vegetation had been burned. The trees
in the surrounding forests to a distance of about
20 miles had been blown down by the blast of
wind when the meteor fell ; they had all fallen out-
wards from the place of fall so that, when seen from
a height, the forest presented a fan-like appearance.
The trees had been stripped of their bark and most
of their branches were scorched as though by the
heat from an enormous furnace. In the valley were
found dozens of deep funnel-shaped holes, marshy
at the bottom, varying in diameter from several
yards to about 50 yards. Each of these probably
contains a meteorite, though the nature of the
ground made it impossible to recover any. About
one thousand reindeer are said to have been killed,
but the loss of human life was slight, the fall occur-
ring in a thinly populated region. Reports collected
COMETS AND SHOOTING STARS III
by the expedition from eye-witnesses of the fall tell
of the appearance of a fiery flame, considerably
forwrhter than the Sun, giving off great heat, and of
a violent explosion followed by a tremendous can-
nonade, like loud thunder and guns firing, which
lasted for about five minutes. One native, whose
hut was said to be at three days' march from the
place, had his hut knocked down, the top blown
away by the wind, his brother stunned and his
reindeer scattered. Professor Kulik considers that
the total weight of the meteorite was about 130 tons
and that the meteor, in falling, carried the cap of
heated incandescent gas in front of it right down to
the ground. When the meteor struck the ground,
the heated air was driven violently outwards in all
directions, felling the trees over a wide area and
blasting the nearer ones. The aerial waves pro-
duced b\ y aie fall travelled outwards at a speed of
about ii miles a second and were recorded by
microbarographs in England, nearly 4,000 miles
away from the fall, and in America, at a distance of
about 6,000 miles. Brilliant green, gold and crim-
son twilight hues were seen in England on the night
of the fall and for several subsequent nights ; they
were caused by fine dust carried high up into the
atmosphere by these waves. Seismic waves of the
earthquake type were produced by the impact and,
travelling through the Earth, were recorded on
seismographs up to distances of about 3,000 miles
from the fall.
It is perhaps surprising that records of meteor
falls are not more numerous, and that there are very
few known instances of damage caused by such falls.
112 WORLDS WITHOUT END
On February 9, 1913, a slow procession of some
hundreds of bright meteors was seen to pass across
Canada and over the Atlantic Ocean to near Ber-
muda, where they struck the sea. The meteors
passed in groups, some containing 20 to 40 meteors,
others only 3 or 4, with their tails streaming out
behind. One observer, awed by the sight, re-
marked that they must be " souls going to heaven. 5 '
This stream of meteors was possibly the final display
of some worn-out comet of past ages.
On July 19, 1912, a shower of meteoric stones fell
near Holbrook in Arizona. The meteor before fall-
ing made a loud booming noise that was heard
some 40 miles away, though in broad daylight
the meteor itself was not visible. The falling stones
were seen, however, raising puffs of dust from the
dry sand of the desert. Upwards of 14,000 frag-
ments were subsequently collected over an area
about 3 miles long by half a mile wide, some
buried to a depth of several inches ; the total weight
of these fragments was about a quarter of a ton.
CHAPTER VI
THE NEAREST STAR
THE SUN
THE Earth and the Moon, the planets and their
satellites and the comets the bodies which we have
so far considered are the children of the Sun.
They are all under his tutelage, constrained by him
to follow their allotted courses and dependent upon
him for such light and heat as they receive. The
Sun is a much larger body than any of the planets.
We have seen that the largest of the planets, Jupiter,
when its extensive atmosphere is taken into account,
has a volume about 1,300 times that of the Earth;
the Sun, however, could contain about 1,000
bodies of the size of Jupiter, or about 1,300,000
bodies the size of the Earth. If the Earth were
placed at the centre of the Sun, the orbit of the
Moon, which is at a distance of nearly one-quarter
of a million miles from us, would be contained en-
tirely within the Sun and would be little more than
half-way out to the surface. In proportion to its size
the Sun is much less massive than the Earth ; it is on
the average less than one and a half times as dense as
water, or about one-quarter as dense as the Earth.
This low average density suggests that the Sun
must be largely gaseous. If we look at the Sun
through a suitably darkened glass, its surface appears
uniformly bright; but when the Sun is photographed
we find that the brightness falls off rapidly as the
edge or limb of the Sun is reached. This is a proof
8 "3
114 WORLDS WITHOUT END
that the surface of the Sun is gaseous and not solid
or liquid ; the light reaching us from the central por-
tion of the disc passes through a lesser thickness of
the cooler outer layers of the Sun's atmosphere than
the light from near the limb. The actinic light of
short wave-length is scattered in the Sun's atmo-
sphere much more than the light of longer wave-
length in the visual region of the spectrum ; that is
why the falling off in brightness towards the limb
is much more noticeable in photographs than it is
visually.
If we could get sufficiently near to the Sun, we
should probably see one or more dark spots on its
surface. These are called "sun-spots." The largest
sun-spots can be seen without difficulty with the
naked eye. The best time to look for a large spot,
is when the Sun is near the horizon, as its brightness
is then much reduced ; at other times, the Sun must
be looked at through dark or smoked glass, to
protect the eyes from injury. There are many old
Chinese records of sun-spots, but the real interest in
them dates from the year 1610, when Galileo studied
them with his primitive telescope. He found that
the spots changed their positions on the disc of the
Sun from day to day, and at first thought that they
were planets, seen projected on the Sun. But con-
tinuing his observations, he noticed that some of the
spots were of irregular shape and that the shape
changed from day to day. A reproduction of one
of Galileo's drawings (from his Istoria e Dimostrazioni
intorno die Macchie Solari, published in 1613) is given
in Fig. 4; in this drawing, B, C are developments of
the spot A, and E, F, G, H, L are developments of the
.aw
rt oi ** * '
< 0> ^
S O ^ 5 ' !
S W 'S '
? ^ rt iu w i
^ &^
(jj SH T
O ^S
D u **
|S a,
X
:i " ; '''-i
<u ^
H|;
NH" i
S S 2
Hi
O
111
^ w c
h 8 S
rt CW
PLATK XV.
SERIES OF PHOTOGRAPHS SHOWING MOTION OF GROUP OF SPOTS ACROSS
TIIK DISC OF THE SUN, CAUSED BY THE SUN'S ROTATION.
Note the changes from day to day in the structure of the group
(March 19-25, 1920). The centre of the visible disc of the Sun is
approximately dehned by the intersection of the two black lines. Each
photograph shows only a small portion of the disc.
["5
THE NEAREST STAR THE SUN
spot D. Galileo therefore A *f
came to the conclusion ^
that the spots were really
markings of some sort on
the surface of the Sun.
The apparent movement
of the spots across the face
of the Sun was correctly
interpreted by Galileo as a
consequence of the rota-
tion of the Sun about its
axis; the rotation causes
the spots to appear to
move along tracks parallel
to the Sun's equator. The
movement of a large group
of spots across the Sun is
illustrated in Plate XV.
As with Jupiter and Saturn,
the period of rotation is not
the same for all parts of
the surface. One com-
plete rotation takes about
25 days at the Sun's equa-
tor, but the period increases
steadily as we go towards
either pole, where it is
about 34 days. This affords
further proof that the Sun FIG? 4. Observations ofsun-
ic nrkf erJirl spots by Galileo. (From Istoria
JLO JLlwl* Ov/J.LV_l r\- ... || , i .
A T e Dimostra&om intorno alle Macchte
A spot may appear and solan, 1613.)
grow with great rapidity
and may last for several months. But most spots
Il6 WORLDS WITHOUT END
have a life of only a few days. A normal fully
developed spot consists of a dark central region,
called the umbra, with a lighter penumbra., in which a
radial fibrous structure may be seen. The umbra
appears black only by contrast with the much
brighter surface of the Sun around it. If we could
shut off the light from the rest of the Sun, we should
find the spot to be intensely bright. A very large
spot may have an umbra which is 50,000 miles
across and a penumbra which is 150,000 miles
across. The area of the umbra of such a spot is
about 10 times the area of the whole of the surface
of the Earth and the area of the penumbra is about
80 times the Earth's surface. Photographs of
typical large spots are reproduced in Plate XIV.
The spots on the Sun were assiduously studied for
nearly 50 years, from 1826 onwards, by Schwabe,
an apothecary of Dessau. Every day that the Sun
was visible at Dessau, Schwabe observed it with his
telescope and made a record of the spots that he
saw. In 1843 he was able to announce that there
was a periodicity in the appearance of the spots;
during the years 1828 to 1831 there were only a few
days on which no spots were visible, but in 1839
no spots were to be seen on 139 days. From 1836
to 1840 there were in all only 3 days on which
he saw no spots, but in 1843 there were 149 days.
His further observations confirmed his first an-
nouncement of a periodicity. The frequency of
appearance of the spots fluctuates in a period of
about ii years; this period is not, however,
absolutely constant. During the present century
sun-spots were numerous in the years 1905, 1917
THE NEAREST STAR THE SUN IIJ
and 19285 and infrequent in the years 1901, 1913,
1923 and 1933. At the present time the number of
spots is increasing and the next maximum may be
expected in 1938 or 1939.
A similar periodicity is shown by certain terres-
trial phenomena, amongst which we may mention
the aurora borealis or australis and magnetic storms.
The aurora does not appear with equal frequency
each year but shows a regular fluctuation with a
period of about 1 1 years, the same period as
that of the sun-spots. Not only is the period the
same; there is a direct correlation between aurorae
and sun-spots. Aurorae are frequent when sun-
spots are numerous and infrequent when the spots
are scarce. The same direct correlation is shown
by magnetic storms ; when a magnetic storm occurs,
a compass needle, instead of pointing steadily in one
direction, oscillates wildly to and fro and telephone
and telegraph circuits may be put out of action.
It does not necessarily follow that either the aurora
or the magnetic storm has any direct connection
with a particular sun-spot or even with sun-spots in
general. All that we can assert is that periodic
changes of some sort are occurring within the Sun
and that the fluctuation in the numbers of sun-spots
is one manifestation of these changes ; the appear-
ance of aurorae and the occurrence of terrestrial
magnetic storms must be dependent upon and
caused in some way by these changes in the Sun.
A sun-spot is a gigantic funnel-shaped vortex in
the outer regions of the Sun. Around the vortex
intensely hot gas from within the Sun is whirling
spirally upwards. We can compare a sun-spot
Il8 WORLDS WITHOUT END
vortex with the hollow vortex formed by water
emptying out of a bath, if we imagine the water to
be made to run in the opposite direction and to
stream upwards from the outlet into the bath. As
the gases stream out of the funnel-shaped mouth of
the vortex, the pressure which has urged them
upwards is released; they expand rapidly and are
considerably cooled as a result; the emitted gases
then stream more or less radially outwards from the
spot along the surface, producing the radial fibrous
structure seen in photographs of the penumbra of
spots. The vortical motion of the heated gas gives
rise to intense magnetic fields, comparable in
strength to the magnetic field between the pole-
pieces of a fair-sized dynamo. Spots frequently
occur in pairs and the magnetic fields associated
with the two spots of a pair are of opposite polarity,
like the two poles of a magnet. This is a conse-
quence of the vortical motions around the two spots
being in opposite directions.
Suppose the leading spot of a pair in the Sun's
northern hemisphere has positive polarity; then we
find that the leading spot of a pair in the southern
hemisphere has negative polarity, the circulations
around the leading spots in the two hemispheres
being in opposite directions. We may compare
with this the atmospheric circulation around
cyclonic disturbances on the Earth, which is in the
anti-clockwise direction in the northern hemisphere,
and in the clockwise direction in the southern hemi-
sphere. The leading spots in the two hemispheres
keep their respective polarities for one complete
sun-spot period, or, in other words, for about
PLATE XVI. PHOTOGRAPHS SHOWING SUCCESSIVE STAGES IN THE
DISSIPATION OF A PROMINENCE ON MAY 26, 1916.
The interval of time from the first to the last photograph is 64 minutes.
The top of the prominence rose 250,000 miles in half an hour.
THE NEAREST STAR THE SUN
ii years; but when the next period commences,
the leading spots in each hemisphere show the
polarity opposite to that previously shown. We
do not know what the real explanation of this
regular reversal of the polarities of the spots may be,
but it would seem to indicate that the fundamental
period for the Sun is twice the normal n-year
sun-spot period.
When the face of the Sun is entirely hidden by the
Moon passing in front of it at a total eclipse of the
Sun, gigantic red tongues of flame are usually to be
seen standing out from the Sun's limb to distances
of many thousands of miles. These jets of flame are
called " prominences." When a total eclipse of the
Sun takes place, the period of totality, during which
the Sun is completely hidden, is brief, never lasting
for longer than about y-| minutes, and the total
phase is visible only from a very limited portion of
the Earth's surface. Total eclipses are therefore
rare occurrences in any given region of the E^rth.
There have been total eclipses visible from some
part or other of the British Isles on the following
dates since the beginning of the fifteenth century :
1424, June 26; 1433, June 17; 1598, March 6; 1652,
April 8; 1715, May 2; 1724, May 22; and 1927,
June 29. The next total eclipse visible in Great
Britain will occur in 1999, August n, and will be
visible from near Land's End. It is therefore for-
tunate that by means of an instrument called a
spectroheliograph it is possible to photograph the
prominences at any time when the Sun is visible.
The prominences show an enormous variety of
shapes and sizes and vary greatly in behaviour. They
I2O WORLDS WITHOUT END
may persist for many days or even weeks, without
violent changes, some appearing as fountains of
flame, others as pyramids of fire and others as
vividly bright columns. Many prominences, on the
contrary, show extremely rapid and violent changes,
evidences of tremendous disturbances occurring at
or beneath the surface of the Sun. At the total
eclipse of the Sun on May 29, 1919, a great pro-
minence was observed, appearing like a gigantic
prehistoric mammoth, about 300,000 miles in length
and about 20,000 miles in height. We shall briefly
describe the history of this prominence, as a typical
example of one of the larger prominences. It was
first seen on March 22, more than two months before
the eclipse; it was then already about 100,000 miles
in length. During the following weeks it gradually
grew in brightness, in length and in height. By
May 27 it had grown very considerably, arid it then
appeared as an enormous body of interlacing
streamers which were continually shifting. On May
29, at its southern end, there was a brilliant column
which had a clawlike structure, as though it was
attached to the Sun by roots. At the north end
there was a more slender column reaching down to
the Sun's surface, and in between there were faint
streamers connecting the body of the prominence
with the Sun. The prominence now began to show
signs of rapid changes. The column at the north
end broke away from the Sun's surface and the main
body of the prominence started to rise rapidly. By
3 a.m. (G.M.T.) it had risen to a height of 140,000
miles and by 5.30 a.m. to 200,000 miles. The main
body of the prominence had now become completely
THE NEAREST STAR THE SUN 121
detached from the Sun, leaving merely the stump
with its claws rooted to the Sun's surface ; it had a
spiral structure as though coiled into a gigantic
spring. When last seen, this enormous cloud of
heated gas, about 250,000 miles in length, was at
a height of more than 500,000 miles above the sur-
face of the Sun and was rapidly receding from it, in
the process of being blown away into outer space.
It had risen 375,000 miles in about 5 hours. The
two ends of the initial prominence remained visible
until August 5.
When the eruptive stage of a prominence sets in,
it is not uncommon for the prominence, or at least
for the main portion of it, to be hurled upwards at
the rate of tens of thousands, or even of hundreds
of thousands, of miles per hour. In Plate XVI are
reproduced some photographs showing the remark-
ably rapid changes observed in a prominence which
appeared on May 1 6, 1 9 1 6. In about one half-hour
the top of this prominence rose by fully 250,000
miles.
By means of the spectroheliograph, sunlight of a
single wave-length can be separated and the Sun
can be photographed in light of that wave-length.
This enables us to photograph the Sun in the light
of hydrogen or calcium or some other element, such
as iron. Such photographs give a representation of
the distribution of the substance, by whose light the
photograph was taken, over the Sun. In Plate
XVII are shown photographs of the Sun taken on
the same day in the light of calcium and of hydro-
gen. The photograph in calcium light shows a
coarse mottled structure; the bright patches are
122 WORLDS WITHOUT END
regions of intense radiation; the darker patches re-
present cooler clouds of calcium vapour, each cloud
being much larger than the Earth. The photograph
in hydrogen light shows much finer structure. A
number of long dark markings are seen on both
photographs, which will be found to correspond
closely. These are prominences seen in projection
on the Sun's disc; being much cooler than the main
body of the Sun, a prominence appears dark when
seen against the hotter and brighter disc. In Plate
XVIII we see a prominence at first at the limb of
the Sun; it is gradually carried round by the Sun
as it rotates, and in the last photograph the promin-
ence appears as a dark marking on the disc.
The number of prominences on the Sun fluctu-
ates with the sun-spot cycle, the prominences being
most numerous when sun-spots are most numer-
ous and vice versa. The two photographs of the
Sun, taken with calcium light, in Plate XIX illustrate
the contrast between the quiescent state of the Sun
at sun-spot minimum and its disturbed state at sun-
spot maximum. The photograph taken at sun-spot
minimum shows no spots and only a few small pro-
minences. That taken at sun-spot maximum shows
several groups of spots and many prominences.
We are now in a position to understand why there
seems to be a connection between sun-spots and
various happenings on our Earth, such as magnetic
storms and aurorae. We have seen that matter can
be ejected with great speeds from the Sun; this
matter comprises atoms, many of which have had
some of their electrons knocked off them and are there-
fore electrically charged, together with free electrons
PLATE XIX.- THE SUN, PHOTOGRAPHED IN CALCIUM LIGHT, AT SUN-
SPOT MINIMUM (ABOVE) AND AT SUN-SPOT MAXIMUM (BELOW).
There are no spots and few prominences in the upper photograph :
in the lower there are several spots and many prominences. The
position of the Sun's axis of rotation is shown in each photograph.
[123
THE NEAREST STAR THE SUN 123
which carry a negative charge. From time to time
streams of these electrically charged particles will
approach near to the Earth ; they are then influenced
by the Earth's magnetic field, which causes them to
move in a spiral motion towards the two magnetic
poles. On entering our atmosphere, electrical
effects are produced which give rise to displays of
aurorae; these are most frequent and most vivid
in the regions around the magnetic poles. The
charged particles also ionize the upper layers of the
atmosphere and make them a conductor for elec-
tricity; it is the magnetic effects produced by the
electrical currents circulating in these ionized layers
that we call magnetic storms. The ejection of
charged particles from the Sun in the way we have
described occurs most frequently near the time of sun-
spot maximum, when sun-spots and prominences are
most numerous. The general connection between
sun-spot activity and the appearance of aurorae
and magnetic storms is therefore accounted for.
It is possible to trace effects due to the sun-spot
cycle in other directions. The facility with which
radio waves can travel round the Earth depends upon
the state of ionization in the ionosphere or ionized
region of the atmosphere the " radio roof of the
world/ 3 as it has been called. This is so intimately
connected with the frequency of appearance of sun-
spots that it is not surprising that the ease with
which distant radio stations can be detected is
closely related to the sun-spot cycle. A more
obscure phenomenon is the relationship between
sun-spots and weather. That there is a connection
is undoubted, but the weather is such a complicated
124 WORLDS WITHOUT END
phenomenon that the dependance upon sun-spots is
usually smothered by a host of other factors. It is
curious that the connection is most clearly shown
in the rings of trees. If a cross-section of the trunk
of a tree is examined, it will be noticed that the
annual rings are not uniformly spaced. The con-
secutive rings in a group may be crowded close to
one another; but those in an adjacent group may be
comparatively wide apart. Each ring marks the
growth of the tree in the course of a year, the growth
depending upon a variety of factors such as the
amount of rainfall and its distribution during the
year, the temperature changes and the amount of
sunshine throughout the year. Some factors are
more important than others, and the relative im-
portance of the several factors varies from one region
to another. The tree ring gives an integrated effect
of these factors acting throughout the whole year.
Dr. Douglass has found that in the arid regions of
Arizona and New Mexico the rings show a clear
dependance upon sun-spots, the width being great-
est when sun-spots are most numerous. He suc-
ceeded in tracing the eleven-year sun-spot cycle in
the Arizona pine trees for a period of 500 years, with
the exception of an interval from 1650 to 1 725, when
the evidence of the sun-spot effect vanished. Some
years later Dr. Douglass heard from Mr. Maunder,
who was in charge of the solar department of the
Greenwich Observatory, that he had found that
there were practically no sun-spots between 1645
and 1715 unexpected confirmation of the results
derived from the study of the rings. By using the
evidence afforded by the tree rings, Dr. Douglass
THE NEAREST STAR THE SUN 125
has succeeded in fixing the date of the ruins of
Pueblo Bonito, the oldest and largest of the great
Indian communities in Chaco Canyon, New Mexico,
which had for long proved a stumbling-block to
archaeologists.
At the time of a total solar eclipse, as the last thin
crescent of the Sun is hidden by the Moon, a bright
aureole surrounding the Sun flashes out. This is
called the corona. It has about one-half the bright-
ness of the Full Moon or about one-millionth of the
brightness of the Sun. The light of the corona is
thus so much weaker than sunlight that it is not sur-
prising that we can see the corona only at the time
of a total eclipse of the Sun. The inner portion of
the corona has a slightly yellowish tinge; the outer
portion is of a pearly white colour. The brightness
falls off very rapidly with distance from the limb of
the Sun. The corona can usually be traced to a
distance of two or three diameters from the Sun. It
thus appears that the Sun is surrounded up to a dis-
tance of a couple of million miles by a very tenuous
atmosphere. The light of the corona appears to be
mainly sunlight which is scattered by this atmo-
sphere.
The general shape of the corona varies during the
sun-spot cycle in a very marked way. When sun-
spots are most numerous the corona is fairly com-
pact, without very long streamers, and is distributed
more or less uniformly around the Sun's disc. At
sun-spot minimum, long curved streamers stretch
out from the equatorial zones, whilst around each
pole of the Sun's axis is a tuft consisting of a large
number of short streamers, suggesting the lines of
126 WORLDS WITHOUT END
force near the two poles of a bar magnet (see
Plate XX) . At other times the shape of the corona
is intermediate between the two extremes. We do
not know the explanation of these changes of shape,
which recur with such regularity that we can pre-
dict with considerable accuracy the shape of the
corona to be expected at a future eclipse.
The structure of the inner corona is very compli-
cated, showing numerous filaments and curved
arches, particularly in the neighbourhood of spots
or prominences.
The Sun is continuously pouring out a dense
stream of radiation in the form of light and heat;
its surface is therefore intensely bright and extremely
i hot. Langley performed an experiment to compare
[ the brightness of the Sun's surface with the bright-
ness of molten steel. Fifteen tons of molten iron
were placed in a converter; half a ton of silicon
and carbon was then added and air was blown
through the glowing mass to raise its temperature.
A further 15 tons of molten iron were poured
into the converter, appearing, as Langley said, like
chocolate poured into a white cup. After thorough
mixing, the cataract of liquid steel was discharged,
scattering showers of brilliant scintillations all
around. With special instruments Langley made
measurements of the brightness of the cataract of
fire, which was so bright that dark glasses were
needed to protect the eyes. He found that, so far
as the light-rays which affect the eye are concerned,
the surface of the Sun was five thousand times
brighter per square foot than the molten steel, and
taking all the radiations light and heat together
(a) SOLAR CORONA, May 9, 1929 (NEAR SUN-SPOT MAXIMUM).
(b) SOLAR CORONA, May 28, 1900 (NEAR SUN-SPOT MINIMUM).
PLATE XX.
THE NEAREST STAR THE SUN 127
the radiation from the Sun was 87 times more
intense than that from the molten metal.
What do we mean when we speak of the tempera-
ture of the Sun ? The Sun has no solid boundary
but is entirely gaseous. If we were able to travel
into its interior, we should find the temperature
rapidly mounting up: it would measure at first
several thousands of degrees, then tens and hundreds
of thousands, and finally, when we were well on the
way into the interior, millions of degrees. The
radiation which we receive from the Sun is a mixture
of radiations which originated at different depths
in the Sun and therefore at different temperatures.
It would seem, then, that we are not justified in using
the term " the temperature of the Sun."
Suppose we conceive a solid body, the size of the
Sun, made of a material which will not melt or
vaporise and imagine it to be gradually heated
uniformly. As it is heated it will after a time begin
to glow with a dull red heat, then it will become
successively red hot, yellow hot, white hot and so on.
At a certain stage in this heating process the body
will be giving out the same amount of light and heafl
as the Sun. We should find that, when this is so,
the colour is the same yellow colour as the Sun.
The " effective temperature " of the Sun is the tem-
perature of the hot body when it has the same colour
as the Sun and the same output of heat and light.
When we speak of the temperature of the Sun, we
must use the term in this sense.
The effective temperature of the Sun, defined in this
way, is about 6,000 C. This is much higher than
the melting-point of carbon. We can gain a better
128 WORLDS WITHOUT END
appreciation of what such a temperature implies by
stating that every square foot of the Sun's surface
is radiating energy at a rate of a g,ooo-h.p. engine
or every square inch at the rate of a Gs-h.p. engine.
The proportion of this radiation which the Earth
receives is only one part in 2,200 millions. Yet the
energy received by the Earth in the form of solar
radiation amounts to nearly 5,000,000 h.p. per square
mile of surface. If we were able to utilise all of this
energy and valued it at the rate of -Jd. per Board of
Trade unit, the monetary value of the solar energy
striking the Earth every second would be about
200,000,000. In comparison with the enormous
potential value of the energy which the Earth is
receiving, our National Debt seems a mere trifle ; if
Dnly we could convert the Sun's energy into cash,
we could pay it off within one minute. Various
mechanisms for utilising some of the Sun's energy
lave been proposed, but none has yet been successful
3n a commercial scale. Probably in the years to
:ome, when our supplies of coal and of oil will have
Deen exhausted, some means by which a portion of
:his energy can be captured and harnessed into the
;ervice of man will be discovered.
For how long the Sun has been radiating energy
it this rate we do not know. The evidence of geo-
ogy points to considerable changes of climate on
Dur Earth during geological time alternations of
ce-ages and warm periods. The only way in which
t seems possible to account for such variations is on
;he supposition that there have been fluctuations in
che output of the Sun's radiation. Yet, on the other
hand, it does not seem possible that the variations can
THE NEAREST STAR THE SUN I2Q
have been considerable at least since the time when
man appeared upon the Earth. If the stellar mag-
nitude of the Sun were changed in either direction
by half a magnitude not a large variation for a vari-
able star conditions on the Earth would be such
that human life would become impossible.
From various considerations it seems probable
that the Sun's output of radiation has not differed
greatly from its present value throughout the life
history of the Earth; in other words, for several thou-
sand million years. How does the Sun maintain
so great an output for so long a time ? If the energy
were derived solely from the store of heat within the
Sun, the radiation could only continue for a few
thousand years at most. Lord Kelvin supposed that
the Sun was gradually shrinking, the matter of
which it is composed slowly falling towards its
centre. A decrease in the radius of the Sun of about
75 yards in the course of a year would release
enough energy to maintain the Sun's radiation;
such a slow decrease in the radius could not be
detected by observation. But on this theory the
Sun's energy would all have been expended in a
mere 25 million years or so. Even if we take into
account the energy that can be provided by the
disintegration of the radioactive bodies, such as
uranium or radium, and suppose that the Sun was
; originally composed entirely of uranium, we can add
only a few million years to the possible lifetime of
the Sun, as a giver of heat and light. We must look
for some other explanation, and we shall return to
this question again when we come to consider the
ages of the stars (Chapter XII).
CHAPTER VII
GIANT AND DWARF STARS
WHEN we look at the sky on a clear moonless night
we cannot see more than some 2,000 or 2,500 stars.
In the whole of the heavens the number of stars
visible to the average eye is between six and seven
thousand, but at any one place and time one-half of
these are below the horizon and near the horizon
the fainter ones cease to be visible. It is therefore
somewhat surprising that from time immemorial the
number of stars in the sky has been used, equally
with the number of grains of sand on the seashore,
to denote a number inconceivably large. " As the
host of heaven cannot be numbered, neither the
sand of the sea measured" (Jeremiah). But even a
small telescope reveals an enormous increase in the
number of stars; with a 3-inch telescope it is
possible to observe about one million stars, and
every increase in telescopic power brings more
stars into view.
The stars were divided by the ancients into con-
stellations, as a convenient method of describing the
position of any star and identifying it. The names
of many of the constellations, particularly those of
the zodiac the zone of the sky bordering the path of
the Sun in the heavens are of very great antiquity.
The origin of these names is not known with cer-
tainty, but it is believed that they originated in Meso-
potamia, for the animals are those of the Bible.
There is no constellation named after the tiger, the
130
GIANT AND DWARF STARS 131
elephant, the hippopotamus or the crocodile; it is
therefore unlikely that the names originated in
either India or Egypt. Many other names have
been drawn from Greek and Roman mythology,
whilst the names of about forty constellations have
been given since 1600, to include stars which did not
belong to any of the older constellations.
The stars were originally described by their posi-
tions in the constellations ; thus, for instance, refer-
ence was made to the star at the end of the tail of
the Little Bear or to the star at the tip of the right
horn of the Bull. In 1603 Bayer, when publishing
his Uranometria, containing a series of star maps with
descriptions, adopted the plan which has since been
followed of denoting the stars in a constellation by
the letters of the Greek alphabet, usually, but not
always, assigned in order of brightness. Some
sixty or so of the brighter stars have proper names,
of Greek, Latin or Arabic origin, in common use.
Thus Sirius, the brightest star in Canis Major, is
Alpha Canis Majoris; Betelgeuse, the brightest star
in Orion, is Alpha Orionis. Other very bright
stars, such as Alpha Centauri, have never been
given proper names.
Other stars are named by the numbers assigned
by Flams teed, the first Astronomer Royal, together
with the name of the constellation, for example, 61
Cygni ; or by the number of the star in the earliest
star catalogue in which it is to be found, e.g. Lalande
21185 is the star of this number in the catalogue of
Lalande (1790). Names such as this may appear
very prosaic, but they achieve their purpose of iden-
tifying the stars.
132 WORLDS WITHOUT END
Whether we look at the stars with the naked eye
or with a telescope, we notice that some appear
much brighter than others. " One star differeth
from another in glory. 55 Hipparchus, who lived in
the second century before Christ, graded the stars
visible to the naked eye into six classes according to
their brightness. The brightest stars he placed in
class i, the faintest in class 6. A star in class i is
called a first magnitude star; a star in class 6 is
called a sixth magnitude star. To the eye, the
difference in brightness between, say, a first magni-
tude and a second magnitude star appears equal to
the difference in brightness between a second mag-
nitude and a third magnitude star or between a fifth
magnitude and a sixth magnitude star. But owing
to psychological factors connected with vision, the
true differences in brightness measured by com-
paring the energy received in the form of light from
the stars are far from equal. If we take the
brightness of a star of the sixth magnitude as unity,
the relative brightness of stars of other magnitudes
is as follows :
Magnitude ..654 32 I
Brightness . . i ai 6J 16 40 100
We see that the difference in brightness between the
fifth and sixth magnitudes is ij units; but between
the second and first magnitudes it is 60 units. It
will be noticed that a difference of one magnitude
corresponds to a ratio of brightness of about 2|, and
a difference of five magnitudes to a ratio of exactly
100.
S 1
CO
c
8
I
c
a
t
in
C
PLATE XXI. PORTION OF THE CONSTELLATION OF CARINA.
The upper photograph has an exposure of I hour, the lower
of 24 hours.
The long exposure reveals extensive faint nebulosity, not
seen in the short-exposure photograph.
GIANT AND DWARF STARS 133
After the invention of the telescope, when fainter
stars began to be observed, the same system of classi-
fication was extended to these stars. Gradually the
system was made more precise and the magnitude
classes were subdivided, so that the magnitude as
now used provides an exact indication of the ap-
parent brightness of the star. For the very bright
stars, the magnitudes are denoted by o, or i or
intermediate values. Thus Alpha Tauri (Aldebaran)
has magnitude i-o ; Alpha Centauri has mag-
nitude o-o and Sirius (the brightest star in the
sky) has magnitude 1-6. A negative magnitude
appears incongruous, even though logical. What
these figures imply is that Alpha Gentauri is 2%
times and Sirius n times as bright as Aldebaran.
On the same system, the stellar magnitude of the
Sun is 26-7 and of the Full Moon is 12-6. Ex-
pressed in units of brightness we may write:
Brightness of Sun
Moon
first magnitude star
sixth magnitude star
eleventh magnitude star
sixteenth magnitude star
twenty-first magnitude star
120,000,000,000
275,000
I
01
000 1
oooooi
oooooooi
The above figures refer to the apparent brightness
as seen by us and have no relationship to the true
brightness or intrinsic candle-power.
It is of interest to compare the numbers of stars
in the whole sky brighter than certain limits of
magnitude. The following figures are based on
counts of stars and are necessarily rather uncertain
for the fainter magnitudes.
134
Magnitude
Limit.
2
3
4
8
9
10
ii
WORLDS WITHOUT END
No. of stars.
Magnitude
Limit.
41
I 3 8
12
13
530
1,620
4,850
14
15
16
14,300
41,000
i?
18
117,000
19
324,000
20
870,000
No. of stars.
2,270,000
5,700,000
13,800,000
32,OOO,OOO
71,000,000
150,000,000
296,000,000
560,000,000
I,OOO,OOO,OOO
As the limiting apparent brightness is decreased,
there is a rapid increase in the number of stars.
This is illustrated by the two photographs of a part
of the constellation of Carina in Plate XXI. These
were taken with the same telescope, but with ex-
posures of i hour and 24 hours respectively. Fain-
ter stars are photographed with the longer exposure
and many more stars are therefore shown.
i The total light from all the stars is equal of the light
of 1,440 stars of the first magnitude. It follows that
the Full Moon gives about 200 times as much light
as all the stars together.
The relative brightness of the stars as they appear
to the naked eye does not give any information
about their true relative brightness. The apparent
brightness of a star depends upon two factors its
intrinsic brightness or candle-power and its distance.
If it were true, as the Greek philosophers supposed,
that the stars were all fixed to a sphere, they would
all be at the same distance and the apparent bright-
ness would then give a measure of the true or in-
trinsic brightness . But the stars are not all at the sam e
distance ; we cannot therefore learn anything about
their true luminosities or candle-powers until their
GIANT AND DWARF STARS 135
distances have been determined. A single candle
at a distance of 100 yards would appear of the same
brightness as four candles at a distance of 200 yards
or nine candles at a distance of 300 yards and so on.
If the same star were moved towards us to one-
tenth of its previous distance it would appear 100
times brighter; if it appeared at first as of magnitude
6, it would now appear of magnitude I.
The first astronomer to arrive at conclusions as to
the distances of the stars which are anywhere near
correct was Sir Isaac Newton. He argued that, as
the stars do not move in orbits round the Sun, like
the planets and comets, they must be so far away
that they are not affected by the gravitational pull
of the Sun. They must be, at the very least, many
hundreds of times as far away as Saturn. The stars
must therefore be self-luminous bodies like the Sun
or we should not be able to see them. He suggested
that they are probably comparable with the Sun
in brightness. He estimated that the Sun would
have to be moved to about 100,000 times its dis-
tance if it were to appear like Sirius, the brightest
star. If, then, Sirius is comparable to the Sun in
candle-power, it must be at a distance of about 10
million million miles. This is actually an under-
estimate of the distance of Sirius, but the order of
magnitude is correct.
The method used to determine the distances of the
nearer stars is exactly analogous to the method used
by a surveyor to determine the distance of a distant
object, S (Fig. 6), on the Earth. The surveyor first
carefully measures off a suitable base-line AB. He
places his theodolite at A and measures the angle
WORLDS WITHOUT END
136
SAB and then moves his theodolite to B and
measures the angle SBA. If the line AB is drawn^
to scale and the lines AS, BS are drawn so that the
angles SAB, SBA are equal to the measured angles,
the lines intersect in S and by measuring SA and SB
the distance of S from A or B
can be found ; the more accurate
method used by the surveyor is to
calculate the distances by means
of trigonometry.
The astronomer proceeds in
essentially the same way when he
determines the distance of a
celestial object. If he wishes to
measure the distance of the Moon
or of a planet, observations can be
made from two observatories, well
separated in distance, such as
Greenwich and the Cape. The
length of the base-line joining
these two observatories can be
computed from the data as to the shape and size of
the Earth which have been derived from surveying
operations. The measurement of the distance of
any one member of the solar system is sufficient
to fix the distances of all the other members, because
the whole system can be plotted to scale with-
out the measurement of a single distance ; the
measurement of any one distance enables the scale
of this plot to be determined.
For the distances of the stars the same method is
again employed, but on account of the great dis-
tances of the stars it is necessary to use the longest
A B
FIG. 6. Measure-
ment of distance by
triangulation.
GIANT AND DWARF STARS 137
possible base-line. The longest base-line available
is the diameter of the orbit of the Earth around the
Sun, which is about 186,000,000 miles. To illus-
trate the method, suppose S 15 S 2 in Fig. 7 are two
stars in a line with the Sun, S, and that observations
are made when the Earth is at
A and then, six months later,
when it has reached the op-
posite end of its orbit at B.
The near star S l will appear
first to one side of the more
distant star S 2 and then to
the other side. As the Earth
moves round its orbit, S x will
appear to swing backwards
and forwards with respect to S 2 .
When Copernicus put for-
ward his theory that the Earth
moves round the Sun, it was
objected by his opponents that,
if the theory were true, relative
shifts of the positions of the
near and distant stars, such as
we have just described, should be seen. They
argued that since such shifts were not observed,
the theory of Copernicus could not be correct.
The explanation is that the distances of even the
nearest stars are so great in comparison with the
diameter of the Earth's orbit that the shifts are ex-
tremely small and can only be detected by the most
refined observations. The nearest star is so far
away that if we were to attempt to draw the
triangle AS^B to scale and represented the distance
FIG. 7. Method of
determining stellar dis-
tances.
138 WORLDS WITHOUT END
AS of the Sun from the Earth (93,000,000 miles) by
i inch, we should need a strip of paper more than
four miles long ! Imagine a surveyor attempting to
measure the distance of an object four miles away
by making observations from the ends of a base-line
only 2 inches in length. It is not surprising that
repeated attempts to determine the distances of the
stars failed. It was not until 1835 that the first
stellar distances were determined and then three
different astronomers, working independently and
using methods which though in principle the same
were different in detail, each succeeded in measur-
ing the .distance of a star.
The nearest known star is about 25 million million
miles distant. Such a great distance does not convey
very much to the mind because the unit in which it
is expressed is too small in comparison with the dis-
tance itself. We need a unit which is more com-
parable with the distance. If I say that the dis-
tance from London to Edinburgh is 25 million
inches, I do not convey a very clear idea of how far
apart these two cities are. In dealing with stellar
distances it is more convenient to express them in
terms of the time which light takes to travel to us
from the star, just as we might express a distance on
the Earth as so many hours' walk or so many hours
by train. Light travels 186,000 miles in a second
and so could girdle the Earth several times in a
single second. Using this method of expressing dis-
tances, we can say that the distance of the Moon is
about i J light-seconds, the distance of the Sun is 8
light-minutes and the distance of Pluto is about 5^
light-hours. This mode of expressing the distance
GIANT AND DWARF STARS 139
of a celestial body has the advantage of reminding
us that we never see the particular body where it
actually is. We do not see the Sun in its true posi-
tion but in the position it occupied eight minutes
previously, when the light by which we see it set
out on its journey towards the Earth.
Expressed in light-time, the nearest star (Alpha
Centauri) is at a distance of about four light-years,
the light-year being approximately 6 million million
miles. We see this particular star where it was
about four years previously; it is actually some-
where about 200 million miles away from the posi-
tion in which we see it.
By such means we can determine with reasonably
good accuracy the distances of the nearer stars, up
to distances of about 500 light-years. For greater
distances, the results become rather uncertain; our
base-line, although 186,000,000 miles in length, has
become hopelessly inadequate. The measurement
of a distance of 500 light-years, by making observa-
tions at the two ends of this base-line, is equivalent
to the measurement of the distance of an object
3,000 miles away by making observations from two
points i foot apart. The exploration of space to
greater distances must be based upon indirect
methods, which we shall refer to later. In the
meantime, it is well to recall the steps that are
necessary to measure how far away are the nearer
stars. We start with the surveyor who measures a
base-line on the Earth with an accurate inch-tape;
from this measured base-line he proceeds by geo-
detic triangulation to determine the shape and size
of the Earth. From two stations on the Earth's sur-
I4O WORLDS WITHOUT END
face the astronomer observes one of the planets, and
determines its distance from the Sun and is able to
infer the distance of the Earth from the Sun. Hav-
ing found the size of the Earth's orbit, he makes
observations of the stars from opposite ends of the
orbit and so derives their distances.
When the distances of the stars have been deter-
mined we can allow for the differences in distance
and compare their true brightness or total candle-
power with the brightness. or candle-power of the
Sun.
The total candle-power of a star being known,
we can learn something about its size, provided we
know the colour or, what is really equivalent, the
temperature of the star. Some stars are yellow like
the Sun ; others are red, orange, white or blue. The
red stars are comparatively cool, the blue stars are
very hot. To each colour there corresponds a de-
finite temperature. We have seen that the tem-
perature of the Sun is 6000 C. ; a red star has a
lower temperature of about 3000 C., whilst a blue
star may have a temperature of 15,000 G. to
30,000 G. By means of an accurate measurement
of the colour of the star, we can therefore conclude
that the star is red-hot, or yellow-hot, or white-hot,
and we can estimate the candle-power of each
square foot of its surface. Combining this informa-
tion with our knowledge of the total candle-power
of the star, we can deduce the total area of its surface
and therefore its size.
We shall look first at the brightest naked-eye stars
and then at the stars nearest to the Sun and see to
what extent they differ from the Sun as regards both
GIANT AND DWARF STARS 141
true brightness and size. The table gives the
names of the twelve brightest stars in the first
column, and the stellar magnitudes in the second
column. The distances in light-years are contained
in the third column. The candle-powers of these
stars, in terms of the candle-power of the Sun as a
unit, are given in the fourth column. The last
column gives the radius of each star in terms of the
radius of the Sun as a unit. Some of these stars
are twin stars; in suchxases the data refer to the
brighter of the components.
Name.
StH1ar
Magnitude.
Distance in
light-years.
Candle-power
(Sun i).
Radius
(Sun i).
Sirius
- 1-6
9
26
2
Canopus
- 0-9
650
8o,OOO
1 80
Vega .
+ o-i
26
50
2
Capclla
0-2
47
I 5
12
Arcturus
O'2
4 1
IOO
3
Alpha Gentauri
0-3
4
I
I
Rigcl .
0-3
540
18,000
38
Procyon
0-5
10
5
2
Achcrnar
0-6
66
200
4
Beta Gentauri
0-9
300
3,100
ii
Altai r .
0-9
16
9
i
Betelgeuse
0-9
190
1,200
290
This table shows several features of interest. In
the first place we notice that the dozen brightest
stars include stars which are near and stars which
are fairly distant. The nearest known star. Alpha
Centauri, is one of the twelve. Sirius, the brightest
star in the sky, is relatively near to us ; but Canopus,
the second brightest star, is the most distant of this
group. The fourth column shows that there is a
142
WORLDS WITHOUT END
wide range in candle-power of the stars. Alpha
Centauri is comparable with the Sun, but Canopus
is a much more brilliant object. It is about 80,000
times as bright as the Sun. If Canopus were as
near as Alpha Centauri it would be a magnificent
object in the sky, for it would appear nearly half as
bright as the Full Moon and would cast shadows on
ANTARES
Fro. 8. Relative size of Antares,
Betelgeuse and the Sun. (The Sun is the
small black dot at the centre.)
the Earth. Rigel is another star in this group
which has high candle-power.
The last column gives a comparison of the sizes
of the stars, and a wide range in size will be noticed.
Several of the stars, such as Vega, Alpha Centauri,
Procyon, Achernar and Altair, are not very different
from the Sun in size; it may be remarked that this
group of the brightest naked-eye stars does not con-
tain any which are smaller than the Sun. But two
GIANT AND DWARF STARS 143
of the twelve stars, Betelgeuse and Canopus, are
giants compared with the Sun. Betelgeuse, for in-
stance, is so large that it could contain 24 million
bodies of the size of the Sun.
We see, therefore, that stars vary greatly both in
actual dimensions and in candle-power. The stars
which appear to us as the brightest stars in the sky
may be either near and of moderate candle-power,
or distant and of high candle-power. We may expect
that any stars which are of candle-power much in-
ferior to the Sun, even though fairly near to us, will
not appear amongst the brightest stars in the sky.
It is of interest to examine next the nearest known
stars and to compare them with the Sun both as
regards size and candle-power. The data are given
in the following table, which is arranged similarly
to the preceding table. Both components of twin
systems are given.
Name.
Stellar
Magnitude.
Distance in
light-years.
Candle-power
(Sun = i).
Radius
(Sun =* i).
Proxima Ccntauri
10-5
4'2
O'OOOI
0-05
Alpha Ccntauri A
0-3
4'3
I'2
i-o
99 B
1-7
4'3
0-3
1-2
Barnard's star
9'7
6-0
0-0004
o-ii
Wolf 359 .
I3-5
8-0
0-00002
0-025
Lalande 21185
7-6
8-3
0-005
0- 3 8
Sirius A
- 1-6
8-6
26-0
1-8
B
8-4
8-6
0-0026
0-032
Innes's Star
9'5
9'6
O'OOOI
0-06
Procyon A
0'5
10-4
5'6
i'9
B .
13-0
10-4
0-00006
0-004
Epsilon Eridani
3-8
107
0-28
0-8
We may note first the comparative emptiness of
space. Within a distance of 10 light-years from the
144 WORLDS WITHOUT END
Sun, we know of only seven stars (counting twin
systems as single stars). If we take an average star
and represent it by a tennis ball, a sphere of 10
light-years radius is about equal in size, on the same
scale, to our Earth. If we imagine our Earth,
therefore, as a hollow globe 8,000 miles in diameter,
containing some half-dozen or so tennis balls, we
have a fairly accurate picture of how empty space
in the near neighbourhood of the Sun actually is.
This group of the nearest known stars includes
the brightest star in the sky, Sirius, and three other
bright naked-eye stars, Alpha Centauri, Procyon and
Epsilon Eridani. But some of the stars have very
low apparent brightness; the apparent brightness of
Wolf 359 is only one-millionth that of Sirius. When
we look at the column headed candle-power we see
that the majority of the nearest stars have a very
much lower candle-power than the Sun; Wolf 359
has only i/5O,oooth of the candle-power of the
Sun. As might be expected, the stars of very low
candle-power are also small in size; three stars in the
above list the companions of Procyon and Sirius
and Wolf 359 are actually only of planetary dimen-
sions, being smaller than Neptune. The companion
of Procyon is comparable in size with the small
planet Mercury. We might be inclined to think
that the companions of Procyon and Sirius should
be regarded as planets moving round their parent
Suns, rather than as Suns. We shall see later that
this view is not tenable.
We have already compared Betelgeuse with the
Sun. It is interesting to have the corresponding
comparison between the Sun and the companion
GIANT AND DWARF STARS 145
of Procyon. The latter is so small that the Sun
could contain 15 million bodies equal to it in size.
From this examination of two groups of stars, the
brightest naked-eye stars and the nearest stars, we
have learnt that there is a very wide range in the
candle-power of the stars. If we represent the Sun
FIG. 9. Relative sizes of the Sun, Jupiter,
Sir! us B and Procyon B. (Procyon B is the
small black dot at the centre.)
by a single candle there are some stars which can be
compared toapowerful lighthouse lantern and others
which can be compared to a will-o'-the-wisp; the
range in candle-power from one extreme to the other
is of the order of one thousand million to one. The
disparity in size, though not so great, is considerable.
If we draw the circles to scale to represent the sizes
of the various stars and make the companion of
Procyon one inch in diameter, the Sun would have
10
146 WORLDS WITHOUT END
a diameter of about 7 yards and Betelgeuse would
have a diameter greater than one mile.
The stars which are large in size and of very high
candle-power are called " giants. 53 Those which
are small in size and of low candle-power are called
" dwarfs." There is no clear line of separation
between the two classes, except for the red stars.
When we examine tliese stars we find that they fall
into two distinct groups; the stars in the one group are
large and of high candle-power, those in the other
group are small and of low candle-power. The
orange stars can be fairly well divided in thesameway,
but for the yellow and white stars the two groups
merge into one another. The Sun is to be regarded
as a dwarf star, being a member of the fainter group
of yellow stars.
The smallness of the companions of Sirius and
Procyon suggested that they might be planets rather
than stars. We have hitherto said nothing about
the masses of the stars; we have seen, however, that
the planets belonging to our Sun are all much less
massive than the Sun itself. The mass of the Sun
is about 20,000 times that of Neptune, with which
we compared the small dwarf stars. If we can find
a method by which we can weigh the companions of
Sirius and Procyon we may hope to be able to
decide whether they are planets or stars.
We have no hope of ever gaining any information
about the weights of the great majority of the stars.
The only means we have of finding how much mat-
ter a star contains is to measure its gravitational
pull on another star; we really weigh the star. In
everyday life we find how much matter any object
GIANT AND DWARF STARS
147
contains by measuring the gravitational pull of the
Earth on it. In the case of a star alone in space,
millions of millions of miles from its nearest neigh-
bour, we cannot possibly detect any gravitational
influence. But many stars are twin systems, con-
FIG. 10. Relative orbits of stars in a
system, showing corresponding positions.
twin
sisting of two stars relatively close together, the one
held fast in the gravitational pull of the other. The
two stars must revolve round each other ; if they
ceased to do so, gravitation would draw them to-
gether and they would collide with one another.
An example of such a twin system is the star Kriiger
60 ; photographs of this system in the years 1908,
1915 and 1920 are reproduced in Plate XXII. A
complete revolution of the one star about the other
is completed in 44 years. When we make careful
148 WORLDS WITHOUT END
observations of a system such as this we find that
each star is moving around some point between the
two stars. In Fig. 10, ABCD represents the orbit
of one star, abed that of the other star, A, a etc.
being corresponding positions at the same instant.
The stars move so that the line joining them always
passes through the point G. The star which has the
smaller orbit will be the more massive ; the gravita-
tional pull of the heavier star on the lighter star is
exactly equal to the pull of the lighter star on the
heavier, but equal pulls will disturb the lighter star
more than the heavier. In the same way, the Earth
pulls a stone with a force equal to that with which
the stone pulls the Earth. But the stone is so much
less massive than the Earth that it is the stone which
falls towards the Earth. In Fig. 10, if AG is three
times Gfl, the star at a which describes the small
orbit has three times the mass of the star at A which
describes the large orbit.
It is therefore possible in the case of a twin system
to find the ratio of the weights of the two stars by
comparing the sizes of their two orbits. If we now
measure the distance of the system, we know the
actual size in miles of the orbit of each star. We are
then able to compare the weight of each of the stars
with the weight of the Sun: The reason is that for
any particular size of orbit, the period of one revolu-
tion is determined by the weights ; the heavier the
stars, the faster they must move if the orbit is to
remain of the same size.
The outcome of investigations such as these is
that we find that most stars contain about the same
quantity of matter. It is very exceptional in a twin
GIANT AND DWARF STARS 149
system to find that one of the stars weighs more than
three times the other or to find that the weight of
either of them differs greatly from the weight of the
Sun. A large majority of the stars have weights
which lie between one-quarter and ten times the
weight of the Sun. There are a few stars known
which are exceptionally heavy ; Plaskett's star is a
twin system with two stars which do not differ
greatly in weight, but their combined weight is more
than 1 60 times greater than the weight of the Sun.
The comparatively small range in weight amongst
the stars is a matter for surprise when we recall how
widely stars differ from one another both in size and
candle-power. Still more surprising is the compari-
son between the two components of some of the twin
systems. We recall that the brightest star in the
sky, Sirius, has a faint companion (called Sirius B
in the table on p. 143) which sends out only one
ten-thousandth part as much light as Sirius ; the
fainter star is so much smaller than the brighter that
the larger star could contain about 180,000 stars of
the size of the smaller. The weights of these two
stars are not very unequal. The faint star has about
the same weight as the Sun ; the bright star has a
weight 2 1 times as great. Procyon provides an even
more striking contrast ; the companion to Procyon
is an extremely faint star which can be seen only
with difficulty under the most favourable circum-
stances with powerful telescopes. It is only one-
hundred-thousandth as bright as Procyon. The size
of the faint star is not known with very great accur-
acy, but it is certainly so small that many millions
of stars of the same size could be contained in
150 WORLDS WITHOUT END
Procyon itself, although Procyon is not much larger
than the Sun. The weight of Procyon is about i J
times that of the Sun ; the weight of the com-
panion is about two-fifths of the Sun's weight.
The faint companions of Sirius and Procyon are
thus comparable to the planets in size, but compar-
able to the Sun in mass. For the amount of matter in
them to be packed into so limited a space, they must
have extremely high mean densities. We recall
that the mean density of the Sun is rather less than
1 1 times the density of water; the companion of
Sirius is on the average about 50,000 times denser
than water. The companion of Procyon is a still more
strange star. A match-box full of the material of
which the companion of Sirius is composed would
weigh a couple of tons ; but if filled with the material
of the companion of Procyon it would weigh about
400 tons.
The largest star known is Antares, with a radius
about 430 times the radius of the Sun (Fig. 8).
Its size is so great that if the Sun were placed at its
centre, the Earth would be only about half-way from
the centre to the surface, and even Mars would be
well inside the surface. The relative sizes of An-
tares and the Sun are shown in the diagram. We
do not know the weight of Antares. But if we sup-
pose it to be twenty times the weight of the Sun, it
follows that the average density of the substance of
Antares is about equal to that of the air in a fairly
well-exhausted vacuum ; in other words, sufficient
of the material to fill an average-sized room would
weigh only an ounce or two.
CHAPTER VIII
THE STARS OUR
BLOOD RELATIONS
IN the preceding chapter we have learnt some-
thing about the sizes, the weights and the candle-
powers of the stars, but we have not yet considered
of what they are made. To gain this information
we must employ an instrument called a spectro-
scope, which can break up the light from a star into
its constituent parts. If a beam of sunlight falls on
a glass prism, the light after passing through the
prism is found to be spread out into a coloured
band in which we can recognise the sequence of
colours in the rainbow : red, orange, yellow, green,
blue, indigo and violet. The rainbow itself is
formed by sunlight that has been broken up in
a similar way by the drops of rain.
A beam of sunlight contains light-waves of many
different wave-lengths; we may compare it to the
complex sound that we hear when we listen to an
orchestra, composed of many different notes some
high, which are of short wave-length, and others low,
which are of longer wave-length. What the prism
does is to bend the path of the beam of light so that
it travels in a different direction; but the light-waves
of short wave-length are bent through a bigger angle
than the light-waves of long wave-length. The
prism therefore sorts out the various wave-lengths
and rearranges them in a definite sequence. The
stellar spectroscope is an instrument in which one
152 WORLDS WITHOUT END
or more prisms are used to break up the light from
a star and to rearrange the various wave-lengths in
their proper sequence.
The violet light, which is bent through the great-
est angle, has the shortest wave-length, about 70,000
wave-lengths going to an inch ; the red light, which
is bent through the least angle, has the longest wave-
length, about 35,000 wave-lengths going to an inch.
But just as the human ear can only detect sounds
that fall within a certain range of wave-length and
fails to hear a note if its pitch is either extremely low
or extremely high, so it is with the human eye.
Light of wave-length shorter than the violet waves
or longer than the red waves is not perceived by the
eye. But our eyes are much more limited in their
range of vision than are our ears in their range of
hearing. For we can perceive sounds over a range
of about eleven octaves corresponding to a ratio
of about 2,000 to i in wave-length; but our eyes can
only perceive light-waves over a range of one
octave, corresponding to a ratio of 2 to i in wave-
length.
The radiations in a beam of sunlight which are
of longer wave-length than the red light can be
detected by their heating effect. If we allow the
prismatic band of sunlight or the spectrum, as it is
called to fall on a screen and place the bulb of a
thermometer just outside the red end of the band,
we shall find that a rise in temperature is recorded
by the thermometer, showing that infra-red rays
are present and that they have heating power. The
radiations of shorter wave-length than the violet
light can be detected by their actinic effect. If we
THE STARS OUR BLOOD RELATIONS 153
place a photographic plate (suitably screened from
direct light) just outside the violet end of the band
and develop it, we shall find that it has been heavily
fogged, showing the presence of ultra-violet waves
with great actinic power.
In 1814, Fraunhofer found that in the spectrum
of sunlight every wave-length was not represented.
When sunlight was passed through a fine slit before
passing through the prism and was then brought to
a focus, he found that the bright spectral band was
crossed by a large number of narrow dark lines.
These lines are generally known as Fraunhofer's
lines. To understand how these are produced, I
must say a few words about the spectrum of a simple
chemical substance.
If we place a little common salt in a hot flame of
coal gas and examine its spectrum, we shall not see
a continuous band of prismatic colours, but just a
few bright lines. This kind of spectrum called a
bright line or emission spectrum is produced by a
glowing gaseous vapour. If, for instance, we pass
an electric spark between two iron terminals, there
is incandescent vapour of iron in the hot region
where the spark occurs ; when we form its spectrum
we see a series of bright lines.
The series of lines given by any particular sub-
stance shows that the substance is sending out light
of certain wave-lengths only. Furthermore, just as
each of us has a different finger-print and each
finger-print is characteristic of the individual to
whom it belongs, so the group of lines forming the
spectrum of any element is characteristic of that
element and serves to identify it. Some substances,
154 WORLDS WITHOUT END
such as hydrogen, have a fairly simple spectrum
containing only a few lines; others, such as iron,
have a very complex spectrum, containing many
hundreds of lines.
An incandescent solid sends out radiations of all
wave-lengths and therefore gives a continuous band
of prismatic colours for its spectrum. If the light
from such a source is passed through a cool gas or
vapour, which by itself would give a spectrum of
bright lines, the spectrum now consists of a continuous
bright background crossed by a number of dark
lines that exactly correspond in position to the
bright lines previously obtained. These dark lines
are caused by radiations of certain wave-lengths
being absorbed by the cool vapour. Such a spec-
trum is therefore called a dark line or absorption
spectrum. The correspondence in position be-
tween the bright lines in the emission spectrum and
the dark lines in the absorption spectrum indicates
that the atoms of a gas can absorb exactly those
radiations that they can also emit and no others.
It should perhaps be mentioned that the darkness
of the lines is purely relative to the adjacent bright
background, just as the sun-spots appear dark in
comparison with the brighter portions of the Sun
surrounding them.
Let us now consider what is happening in the Sun,
which we may regard as an average star. Radia-
tion is welling up from the intensely hot interior to-
wards the comparatively cool surface layers. Deep
down in the interior the temperature is so high that
the whole of the radiation is of extremely short wave-
length, comparable to X-rays. As it works its way
THE STARS OUR BLOOD RELATIONS 155
outwards towards the surface, it is continually ab-
sorbed and re-emitted by atoms of various ele-
ments. As the temperature falls continuously from
the centre to the surface, the radiation is gradually
transformed into radiation of longer and longer
wave-length, until at last we get the infra-red, visual
and ultra-violet radiations that are emitted from
the surface. Before reaching the surface, all wave-
lengths in this range are present. The atoms of
iron in the relatively cool outer atmosphere absorb
from this radiation all those wave-lengths which
are present in the spectrum of iron; similarly, the
atoms of every other element present absorb each
their own particular series of wave-lengths. The
atoms, having absorbed these radiations, emit them
again in all directions. The light that finally
emerges has thus been robbed to a large extent of
the wave-lengths associated with all these different
sorts of atoms, because the radiations of these wave-
lengths have mostly been scattered in other direc-
tions. The. spectrum of sunlight consists therefore
of the prismatic band, crossed by a multitude
of dark lines corresponding in position to each one
of these wave-lengths.
If we could screen off the direct light coming from
the interior of the Sun, we should see the spectrum
of the hot gaseous atmosphere. This spectrum
should consist of a series of bright lines agreeing in
wave-lengths with those of the dark Fraunhofer
lines. A total eclipse of the Sun provides a suitable
opportunity. The Moon, as it moves in front of the
Sun, gradually cuts off its light. Just before the
Sun is totally eclipsed a narrow bright crescent of
156 WORLDS WITHOUT END
light remains around one edge of the Moon. We
see then the Sun's outer atmosphere. The atoms in
this atmosphere are absorbing radiations that
come from the interior of the Sun and are emitting
them again in all directions. The light that
reaches us consists only of the radiations emitted by
this atmosphere. The spectrum is found then to
consist of bright lines on a dark background the
typical spectrum of a hot gas. This bright-line
spectrum lasts for a few seconds only, until the ad-
vancing Moon has completely hidden the Sun's
atmosphere. A portion of the Sun's spectrum,
obtained in this way at the total eclipse of 1932, is
reproduced in Plate XXII. The spectrum consists
of a number of circular lines. Each one of these is
an image of the bright crescent of the Sun formed
by light of a definite wave-length. It will be
noticed that some prominences were projecting from
the Sun's limb. The stronger radiations in this
portion of the spectrum are due to hydrogen,
sodium, calcium and helium.
The spectrum of the Sun contains a large number
of lines; somewhere about 16,000 were catalogued
by Rowland. Many of the elements with which we
are familiar on the Earth have been detected in this
spectrum by their finger-prints. Fifty-seven known
elements have been definitely detected in the Sun.
The remaining thirty-five elements, which have not
been definitely detected, include two elements
which have not yet been found on the Earth;
several elements whose spectra contain no strong
lines in the range of wave-lengths which the Earth's
atmosphere transmits; and several other elements
THE STARS OUR BLOOD RELATIONS 157
whose spectra have not yet been sufficiently investi-
gated in the laboratory for identification to be pos-
sible. There remains the group of heavy elements,
osmium, iridium, platinum, tantalum, gold, mer-
cury, bismuth and polonium, none of which has
been detected, though we should expect to find
some of them unless they are present in very minute
proportions. It is probable that the weight of the
atoms of these heavy elements causes them to sink
to a considerable depth in the Sun's atmosphere;
this would make their detection impossible and
would account for their apparent absence.
We conclude that we are not justified in asserting
that any element found on the Earth is absent from
the Sun.
There is one element, helium, which was found
in the Sun before it had been discovered on the
Earth. At the total eclipse of the Sun in 1 868 a
strong line in the yellow was observed in the bright-
line spectrum of the Sun's atmosphere; this strong
line could not at that time be assigned to the spec-
trum of any known element. It was concluded that
it must be produced by an unknown element, which
was accordingly named helium after the Greek word
for the Sun. It was not until 1895 that helium was
discovered on the Earth. In that year the chemist
Ramsay found that it was present in small quantity
in the air we breathe and also in the mineral called
uraninite. Helium was regarded as an extremely
rare gas until some twenty years ago, when it was
found to be a constituent of the natural gas from the
oil wells in certain regions in the United States. It
was possible to obtain it from this source in quanti-
158 WORLDS WITHOUT END
ties sufficiently great to enable it to be used to inflate
airships.
There are many stars whose spectra are exact
counterparts of the spectrum of the Sun. Corre-
sponding portions of the spectra of the Sun and of
Alpha Centauri are shown in Plate XXII. Many
other stars have spectra that bear no resemblance
to the spectrum of the Sun. We find, however, that
we can select a series of spectra each one of which
shows only slight differences from the spectra that
precede and follow it and such that with few excep-
tions the spectrum of any star can be matched
against one of the spectra in this series. We can
compare such a series with a moving picture; each
picture shows but slight differences from the next,
but there may be no correspondence at all between
the first picture and the last.
When we have arranged the different types of
stellar spectra in such a series, we find that we have
arranged the stars in a sequence of progressively
changing colour and temperature. If we could take
a cool star and heat it up gradually, we should find
that its spectrum would pass continuously through
this series. We actually see such changes taking
place in certain stars that are variable in bright-
ness, these variations in brightness being accom-
panied by changes in temperature.
A giant red star, many thousands of times larger
than the Sun in volume and of extremely low
density, has a spectrum that is almost identical
with the spectrum of a dwarf red star of small size
and high density. The reason is that the nature of
the spectrum is determined mainly by the tempera-
(a) PHOTOGRAPHS OF THE TWIN STAR, KRUOER 60 (TOP UIFT-IIANB
CORNER), IN 1908, 1915 AND 1920, SHOWING ORBITAL MOTION.
.lii*^
(b) SPECTRUM OF SUNLIGHT AT TOTAL ECLIPSE OF AUGUST 31, 1932.
A separate image of the bright crescent Is given by the light of each
wave-length. The strongest images are clue to calcium, hydrogen,
helium and sodium, (The photograph is a negative.)
>(c) PORTION OF SPECTRA OF SUN (TOP AND BOTTOM) AMI> OF ALPHA
CENTAURI (CENTRE),
SPECTRUM OF ZETA URS.AS MAJORIS AT Two DIFFERENT DATES,
The lines in the lower spectrum are doubted.
PLATE XXII.
THE STARS OUR BLOOD RELATIONS 159
ture of the outer layers of the star and these tempera-
tures for the giant and dwarf red stars are the same.
The surface temperatures of the hottest stars are
about 60,000 C. and of the coolest stars about
2,000 degrees. The ratio of the temperatures is 30
to i, but this ratio corresponds to a ratio in the
energies given out by equal areas of surface of the
two stars of 810,000 to i. Whereas every square
inch of the Sun's surface sends out energy at the rate
of 62 horse-power, the hottest stars send out energy
at a rate of somewhere about 500,000 horse-power
from each square inch of surface and the coolest at
a rate of considerably less than i horse-power from
each square inch of surface.
The lines present in the spectra of the hottest and
bluest stars are mainly due to hydrogen, helium,
oxygen, nitrogen and silicon; lines due to the
metallic elements are absent. In the spectra of the
yellow stars, occupying the middle range of tem-
perature, the lines due to helium, oxygen, nitrogen
and silicon are weak or absent, and the lines of the
metallic elements, such as calcium, iron, titanium,
aluminium and manganese, are prominent. In the
coolest stars we find evidence in the spectra of the
presence of simple compounds, such as carbon
monoxide, cyanogen, zirconium oxide and titanium
oxide. It might be thought that these differences
in spectra were related to differences in chemical
constitution, and that stars like the Sun might be
composed largely of metals, whilst the hottest stars
might be composed largely of hydrogen and helium.
This is not the true explanation of the differences in
the spectra.
l6o WORLDS WITHOUT END
We have remarked that the spectrum of each
element is characteristic of that element, just as a
finger-print is characteristic of an individual. But
actually matters are rather more complicated. If
we place a little of an iron compound in the electric
arc passing between two pieces of carbon, the spec-
trum will show the presence of iron : we can also
obtain the spectrum of iron by passing a high-ten-
sion electric spark between two pieces of iron. But
the two spectra do not correspond. Some lines
appear in the spectrum of the spark that are not
present in the spectrum of the arc ; other lines are
stronger in the spark spectrum and others again,
that are present in the arc spectrum, are absent
or weak in the spark spectrum. Both spectra are
characteristic of iron and neither can ever be given
by any other substance ; depending upon the condi-
tions under which the spectrum is produced, we
may obtain the one .type of spectrum or the other or
even a spectrum which is intermediate between
them. It would be a considerable complication for
our finger-print experts if a criminal could have a
dual personality a Jekyll and Hyde existence and
give a different finger-print according to whether he
was Jekyll or Hyde at the moment, and at times a
finger-print which is a mixture of the two.
To account for this change of behaviour we must
remember that an atom is a complicated structure
forming a sort of miniature solar system; the nucleus
of the atom corresponds to the Sun and the electrons
moving in orbits around it correspond to the planets.
The electrons are minute particles carrying a charge
of negative electricity; the nucleus has a charge of
THE STARS OUR BLOOD RELATIONS l6l
positive electricity sufficient to balance the com-
bined negative charges of the electrons. The
heavier the a torn , the greater is the number of
satellite electrons that it contains, and the more
complicated is its structure.
Sometimes an atom loses one or more of its outer
electrons. It is then said to be ionized. An ionized
atom is positively charged and gives a spectrum
which diners from that of the neutral atom. Atoms
can be ionized by raising their temperature suffi-
ciently. We may compare ionization with the pro-
cess of breaking up a chemical compound into its
constituent atoms by the action of heat. Just as by
increase of temperature we can break up molecules
of hydrochloric acid into atoms of hydrogen and
atoms of chlorine, so also by a sufficient increase of
temperature we can split atoms of neutral iron into
atoms of ionized iron and electrons. The electric
spark is considerably hotter than the electric arc;
therefore in the spectrum of the spark, the lines
that are due to ionized atoms of iron are prominent,
and in the spectrum of the arc, the lines that are
due to the neutral atoms of iron are prominent.
If the temperature is still further increased, many
of the atoms will lose two electrons ; such atoms are
said to be doubly ionized, and they give rise to a
spectrum that is different from the spectrum of
either the neutral atom or of the singly ionized atom.
We are now in a position to understand how it is
that the differences between the spectra of different
stars are due mainly to differences of temperature.
In the spectrum of the Sun, for instance, we find
many lines that are produced by neutral atoms of
ii
l62 WORLDS WITHOUT END
iron and many other lines that are produced by
ionized atoms of iron; the lines produced by the
neutral atoms are relatively stronger because the
atmosphere of the Sun is not sufficiently hot to
ionize most of the atoms of iron. If we pass through
the sequence of stellar spectra from the Sun towards
the low temperature side, we shall find that the lines
due to ionized atoms of iron will completely dis-
appear; on the high temperature side we find first
that the lines due to neutral atoms of iron disappear
and that at still higher temperatures the lines due to
the ionized atoms disappear. The reason for the
disappearance of these lines at high temperatures is
that more and more atoms of iron lose two electrons
as the temperature increases, until all the atoms of
iron have become doubly ionized. But the doubly
ionized atoms of iron have no lines in the range of
wave-lengths normally accessible to observation.
All evidence of iron therefore disappears from the
spectrum of the star, although the atoms of iron in
the atmosphere of the star are just as numerous as
they were previously.
The very fact that it is possible to arrange the
spectra of most of the stars into a single continuous
series points strongly towards a uniformity in the
composition of the stars or at least in the composi-
tion of their outer layers. The elements which we
may conclude from the evidence provided by
stellar spectra are most abundant in the stars are
hydrogen, silicon, sodium, magnesium, aluminium,
carbon, calcium, iron, zinc, titanium, manganese,
chromium, potassium, vanadium, strontium and
barium. The elements which are the most abun-
THE STARS OUR BLOOD RELATIONS 163
dant in the crust of our Earth are oxygen, silicon,
hydrogen, aluminium, sodium, calcium, iron, mag-
nesium, potassium, titanium, carbon, chlorine,
phosphorus, sulphur, nitrogen, manganese, fluorine,
chromium, vanadium, lithium, barium, zirconium,
nickel and strontium. Eight of the elements of this
list, viz. oxygen, chlorine, phosphorus, sulphur,
nitrogen, fluorine, zirconium, nickel, have spectra
which are not well suited for estimating their
abundance in the stars. The spectra of the hotter
stars contain lines which are due to ionized oxygen,
sulphur and nitrogen, and it is reasonably certain
that these three elements are abundant in the stars.
There seems, therefore, to be a remarkable parallelism
between the composition of the Earth and the com-
position of the stars.
Not only are the stars in all parts of the Universe
built from the same sorts of bricks the atoms of the
known elements as those of which our Earth is
made, but also there is an unexpected similarity in
the relative proportions of the different varieties of
bricks: the elements which are most abundant on
the Earth are also on the whole the most abundant
in the stars. We may therefore assert with some
justification that the stars are our blood relations.
CHAPTER IX
TWIN STARS, PULSATING
STARS AND NEW STARS
IN this chapter we shall consider three particular
classes of stars which are of special interest twin
stars, pulsating stars and new stars.
Twin Stars
We have seen already that some stars, such as
Kriiger 60 (Plate XXII), are twin systems, for
the two stars are visible in the telescope, and by re-
peated observations extended over a sufficient length
of time the motion of the two stars about each other
can be detected. Two stars may appear in the tele-
scope as a twin system merely because they lie nearly
in the same line of sight, though they are at very
different distances from the Earth. Such optical
pairs, as we may call them, will show in time motion
relative to each other, because of the motion of each
star in space. But the relative motion of one star
with respect to the other will be in a straight line.
It will not be curved as it must be when the stars
form a true twin system. It is only when curved
paths, due to orbital motion, have been detected
that a pair of stars can be regarded as a twin system.
The time taken for each star to go once round its
orbit ranges for different systems from several years
to many thousands of years. The distance apart of
the two stars is the main controlling factor; when
they are very far apart, their mutual gravitational
164
TWIN, PULSATING AND NEW STARS 165
pull, which determines how fast they move, is weak.
If, for instance, the two stars are each as massive as
the Sun, they will take 10 years to go once round
their orbit if their average distance apart is 540
million miles; 100 years if it is 2,500 million miles;
and 1,000 years if it is 12,000 million miles.
A twin star which appears in the telescope as two
stars is called a visual binary star. We do not know
of any visual binary star in which the components
move completely round their orbits in a shorter
time than about 5 years. It does not follow that
there may not be twin systems which are circling
about each other in periods of less than 5 years.
In any such system, the two stars must be compara-
tively near together (less than four times the distance
of the Earth from the Sun) to move so rapidly.
Because of the great distances of the stars we are
unable to distinguish the slightly different directions
in which the lignt from two such close stars reaches
the eye. The star may be a twin, but even in a
powerful telescope it appears as a single point of
light.
It would therefore seem that, if there are twin
systems in which the two stars are close together and
in rapid motion about one another, we shall not be
able to learn anything about them. Fortunately,
however, there are several methods by which we can
recognise these close twin systems and learn some-
thing about them. There is, for instance, a star
called Algol, a bright star of the second magnitude
in the constellation of Perseus, which shines with a
practically steady light for about 60 hours; its
brightness then commences to decrease and at the
l66 WORLDS WITHOUT END
end of 5 hours it is only one-third of its normal
brightness. It then commences to brighten again
and 5 hours later has returned to its original
brightness, at which it stays for another 60 hours,
when the fading repeats itself. The changes in the
brightness of the star were known of old to the
Arabian astronomers, who gave it the name Algol,
meaning the demon star. John Goodricke in the
eighteenth century showed that the changes in
brightness occurred with perfect regularity.
The explanation of the inconstancy in the light
of Algol is that the star is a twin system., though it
appears in the telescope to be merely a single star.
The two stars move in an orbit which is nearly in
the line of sight and one star is much brighter than
the other. Once in each revolution the faint star
passes in front of the bright star and partially
eclipses it. As soon as the eclipse commences, we
see Algol begin to fade in brightness. As long as
more and more of the bright star is being hidden, the
fading continues ; then, as the faint star passes away
from the face of the brighter, the light increases.
Half a revolution later, the bright star will partially
eclipse the faint one. The one star is so much
brighter than the other that we cannot by eye detect
any fading when the faint star is eclipsed, though
careful measurements reveal that there is a very
slight fading.
Many stars are known whose light is variable be-
cause they are really twin systems and each star
alternately eclipses (or partially eclipses) the other.
These twin systems are of great interest because
from a careful study of the light variations we can
TWIN, PULSATING AND NEW STARS 167
learn a good deal about the sizes of the stars and
their relative brightness.
In Fig. 1 1 are shown drawings to scale of three
THE SUN
VPUPPIS(BIu)
Z HERCULIS (Whitish Yellow!
CASTOR C (Red)
Fio. ii. Relative sizes of typical twin stars.
The relative orbits, as they appear from the Earth,
are shown by the dotted lines. The areas of the
black squares indicate the relative weights of the
stars. The Sun is shown to scale ; the orbit of
the Earth around the Sun is nearly 12 times the size
of the relative orbit of V Puppis, and is too large
to be shown in this diagram.
typical twin systems ; the orbit of the smaller star
relative to the larger, as seen from the Earth, is re-
presented as a dotted line. The size of the Sun on
the same scale is shown for comparison. The
colours of the three systems are respectively blue,
l68 WORLDS WITHOUT END
whitish yellow and red. These typical systems
illustrate the general rule that the blue stars are
large and massive ; the yellow stars are comparable
with the Sun ; the red stars are of small size and
mass. In the case of each of these systems, the
orbits of the one star with respect to the other are
much smaller than the orbit of the Earth about the
Sun. It is because of the smallness of the orbits
that we are unable to distinguish the two stars even
with the aid of a powerful telescope.
If the orbits of the systems shown in Fig. 1 1 were
tilted so as to make a greater angle with the line of
sight, the orbit would appear more nearly circular
and, if the angle is sufficiently large, eclipses will no
longer occur, but the one star will pass above or
below the other as seen from the Earth. Variation
in the light will no longer be seen and we shall no
longer have this indication of the twin nature of the
system. There is, nevertheless, yet another way in
which it may be possible to detect the presence of
two stars.
In Plate XXII portions of the spectra of the Sun
and of Alpha Centauri are shown. Line by line
they correspond to each other, but every line in the
spectrum of Alpha Centauri is displaced slightly
from the corresponding line in the spectrum of the
Sun. Such displacements of the lines of the spec-
trum of a star are produced by a motion of the star
in the line of sight towards us or away from us. If,
for instance, the star is moving towards the Earth,
the waves of light sent out towards us are slightly
crowded together by the motion of the star. The
wave-length of the light is therefore effectively short-
TWIN, PULSATING AND NEW STARS 169
ened; in other words, the light appears slightly
bluer than it would have done had the star been at
rest. The lines in the spectrum of the star are there-
fore all slightly shifted towards the blue, or short
wave-length region. Similarly, if the star is moving
away from us, the waves are slightly spread out.
They are therefore reddened slightly and the lines
in the spectrum of the star are all displaced a little
towards the red or long wave-length region.
The spectrum of the star Zeta Ursae Majoris at
two different dates is shown in Plate XXII; the
star spectrum is the central band, and above and
below it a terrestrial spectrum is photographed for
comparison, for the purpose of identifying the lines.
The lines in the lower spectrum of the star corre-
spond exactly to lines in the upper spectrum. But in
the lower spectrum each line appears doubled. Ob-
servations of this star show that each line in the spec-
trum first opens and appears double, and then closes
again and appears single with perfect regularity.
We see here the effect of two stars moving in orbits
about their centre of gravity. When the two stars
are at the extreme positions in their orbits, one has
a relative motion towards us and the other a motion
away from us. Corresponding lines in the spectra
of the two stars are therefore slightly shifted in
opposite directions. We actually see a composite
spectrum formed by the spectra of the two stars
superposed. Each line in this spectrum therefore
appears double. But when one star passes above or
below the other, so that each star is moving at right
angles to the line of sight with the same speed to-
wards or away from us, corresponding lines in the
WORLDS WITHOUT END
two spectra fall in exactly the same position and the
composite spectrum appears to be the spectrum of
a single star.
The two stars in the system of Zeta Ursae Majoris
do not differ greatly in brightness. But in some
twin systems one star is much fainter than the other,
and then only the spectrum of the brighter star is
seen. The orbital motion of the star causes all the
lines in the spectrum to swing backwards and for-
wards ; the lines of the terrestrial comparison spec-
trum form the basis of comparison for detecting the
to-and-fro movement of the lines.
Thus there are various ways of detecting the exist-
ence of twin systems. Of the stars visible to the
naked eye, it appears that at least one in five is a
twin system; for the fainter stars, information is
necessarily incomplete, though there does not seem
to be any reason why twin systems should be less
common amongst these stars than amongst the
bright ones. We may therefore conclude that twin-
ning is a common phenomenon amongst the stars.
Some stars are more complex. The bright naked-
eye star, Castor, appears in the telescope as a twin
star. When the spectra of these two stars are
photographed, we find that each of them is a close
twin. One pair describes its orbit in about 3
days, the other in about 9 days. The two pairs
revolve about each other in somewhere about 300
years. In addition, the telescope reveals a faint
star which is also a member of the same system, and
which is itself a close twin. This faint star must
move round the other stars in a period of many
thousands of years. Thus, whereas with the naked
TWIN, PULSATING AND NEW STARS
eye we see only one star, there are actually six stars
involved in the system.
Pulsating Stars
We have recognised some stars as twins because
their brightness is not constant, but varies in a
regular periodic manner. But variation in the
brightness of a star is not necessarily an indication
that the star is a twin system. Some stars which are
single stars vary in brightness; in some cases in a
perfectly regular manner, in others in an irregular
manner. We know very little about the causes of
the irregular variations. We have remarked that
the recurrent ice-ages and warm periods on our
Earth, of which we have proof from geological evi-
dence, were probably caused by small fluctuations
in the output of energy from the Sun. The Sun
would therefore seem to be a star whose light varies
slowly and irregularly by a small amount.
Of particular interest to the astronomer is a class
of stars whose light varies in a perfectly regular man-
ner. These stars are called Cepheid variables be-
cause the first star of this type to be discovered was
the star Delta Cephei. The sequence of changes of
brightness of such stars is of one general type
and can be illustrated by the case of Delta Cephei.
The complete sequence of changes takes about 5 J
days for this star. Starting at the time of greatest
brightness, the star fades gradually during 4 days
until it is only about half as bright as it was initially.
It then commences to brighten again at a more
rapid rate than it faded, so that in i| days it has re-
gained its initial brightness. The fall in brightness
172 WORLDS WITHOUT END
always takes place more slowly than the rise in
brightness. The complete sequence of changes is
completed in a time which ranges for different stars
from several hours to about a couple of months.
The changes in brightness of a Cepheid variable
are attributable to a regular pulsation of the whole
star. The star swells up and then contracts again
with perfect regularity; it may be compared to an
inflated rubber balloon which is blown up, then
partially deflated, blown up again, and so on in a
perfectly regular manner. The difference between
the radius of the star when it is most distended and
the radius when it is smallest is generally somewhere
about one-tenth of the mean radius, but may be as
large as one-quarter.
We can obtain direct evidence that a Cepheid
star is pulsating by photographing its spectrum.
When the star is expanding, the surface of the star
that is facing us is moving towards us and the lines
in its spectrum are therefore all displaced and have
wave-lengths which are somewhat shorter than the
normal. When the star is contracting, the surface
is moving away from us and the spectrum lines are
shifted slightly in the opposite direction, the wave-
lengths then being somewhat longer than the nor-
mal. All the lines in the spectrum therefore swing
together slightly backwards and forwards in syn-
chronisation with the pulsation of the star. But it
was exactly this behaviour of the lines that we in-
terpreted above as evidence of a twin system in
which one star was much brighter than the other.
This interpretation cannot be applied to the
Cepheid variables, however, for two reasons. In
TWIN, PULSATING AND NEW STARS 173
the first place we find that the size of the orbit comes
out to be smaller than the bright star, which is a
reductio ad absurdum\ and, in the second place, the
changes in brightness are accompanied also by
changes in the colour and in the temperature of the
star, indicating that the pulsation is related to the
physical properties of the star.
The Cepheid variables are all giant stars, large in
size, low in density, of very great candle-power and
generally much more massive than the Sun. In the
table below, data for several typical stars are sum-
marised. The following details are given : the name
of the star, the period of a single pulsation, the
luminosity or candle-power in terms of that of the
Sun as unit, the radius in terms of the radius of the
Sun, the total change in radius in millions of miles,
the mass in terms of the mass of the Sun as unit, and
the mean density in terms of the density of water
as unit.
Period of
pulsation
Days.
Lumin-
osity
(Sun - i).
Mean
Radius
(Sun-i).
Change
in Radius
( millions
of miles).
Mass
(Sun = i).
Density
(water = i).
RR Lyra
0-6
125
6
O'2I
5
031
SU Cassiopeiae
2'O
260
13
0-37
6
004
Polaris .
4-0
460
22
0-19
8
00 1 1
8 Cephei
5-4
700
26
1-6
ii
0008
i? Aquilae
7-2
IjOOO
35
2'2
13
OOO4
Geminorum
IO
1,700
43
2-3
18
0003
X Cygni
16
3,200
70
7-6
26
OOOII
Y Ophiuchi
i7
3.500
72
2'2
28
OOOI I
1 Carinac
36
9,600
116
10-7
50
00004
The stars in this table have been arranged in the
order of increasing period of pulsation. It will be
174 WORLDS WITHOUT END
noticed that this arrangement has placed them also
in the order of increasing candle-power, increasing
size, increasing mass and decreasing density. This
cannot be the result of chance but must be connected
in some way with the physical nature of this partic-
ular type of star.
The relationship between the period of pulsation
of a Cepheid variable star and its candle-power has
proved of the highest importance in astronomy.
We notice that the greater the luminosity of the star,
the longer is the time taken for a single pulsation.
So close is the correlation between the two quanti-
ties that we need only measure the time of the pul-
sation or, in other words, the time in which the
light of the star varies through one complete cycle
and we know at once the candle-power of the star.
Comparing the candle-power with the apparent
brightness of the star, we are able at once to deduce
its distance. In this way we are provided with a
powerful means of exploring space, for we see from
the third column of the above table that the Cepheid
stars are intensely luminous and can consequently
be seen at very great distances. For instance, the
last star in the above table, if it were at a dis-
tance of 50,000 light-years, would appear as a star
of the eleventh magnitude, easily visible in a six-
inch telescope. Suppose we observe the star, com-
paring its brightness with the brightness of several
neighbouring stars, over an interval of some months ;
we find that it is pulsating in a period of 36
days. We can conclude at once that its candle-
power is 9,600 times the candle-power of the Sun,
and therefore, since it appears as a star of the
TWIN, PULSATING AND NEW STARS 175
eleventh magnitude, that its distance must be 50,000
light-years. When we remember that direct meas-
urements of stellar distances greater that 500 light-
years are practically meaningless because errors of
observation cannot be reduced below a certain
small value, we shall realise what a powerful instru-
ment the Cepheid variable is for the exploration of
space. In a later chapter we shall see to how great
an extent our knowledge of the size of the Universe
depends on this property of the Cepheid variable
stars.
New Stars
The third class of stars of particular interest is the
small group of stars known as new stars or novae.
A nova is a star that undergoes a rapid and consid-
erable increase in brightness, which may be any-
thing from several thousand- fold to several million-
fold; after the peak of the flare-up has been reached,
the brightness diminishes again, usually rapidly at
first and then more and more slowly. The name
of " new star," suggesting the birth of a star that
did not exist before, is therefore misleading. But
the name was given at a time when the appearance
of a nova was viewed with wonder and when it was
believed that a star had come into existence where
no star previously existed. The surprise with which
the appearance of a brilliant new star in the year
1572 was regarded can be judged from the account
given by the great Danish astronomer, Tycho
Brahe. The following is a translation of a por-
tion of his account of this star, entitled De Nova
Stella :
176 WORLDS WITHOUT END
" Last year (1572), in the month of November,
on the eleventh day of that month, in the evening,
after sunset, when according to my habit, I was con-
templating the stars in a clear sky, I noticed that a
new and unusual star, surpassing the other stars in
brilliancy, was shining almost directly above my
head ; and since I had, almost from boyhood, known
all the stars of the heavens perfectly (there is no
great difficulty in attaining that knowledge) , it was
quite evident to me that there had never before been
any star in that place in the sky, even the smallest,
to say nothing of a star so conspicuously bright as
this. I was so astonished at this sight that I was not
ashamed to doubt the trustworthiness of my own
eyes. But when I observed that others, too, on hav-
ing the place pointed out to them, could see that
there was really a star there, I had no further doubts.
A miracle indeed, either the greatest of all that have
occurred in the whole range of nature since the be-
ginning of the world, or one certainly that is to be
classed with those attested by the Holy Oracles, the
staying of the Sun in its course in answer to the
prayers of Joshua, and the darkening of the Sun's
face at the time of the Crucifixion."
The star which Tycho Brahe observed was, when
at its greatest brightness, much brighter than any
other star in the sky, brighter also than Jupiter and
nearly as bright as Venus. It faded gradually and
ceased to be visible to the naked eye about 16
months after it had flared up. In the position in
the sky where it appeared only very faint stars are
now to be found, and it is not possible to identify
Tycho's star with certainty. It may be concluded
,/ Yl-MONIbRAHF. (.^TTOXIDIX Dx\>s' I
OE KKVU.TTRVP ET ARcii- VRANIENBV^G t*
DANICI HVEN'NA
N5V.-A M t
PLATE XXIII. PORTION OF THE MILKY WAY IN THE CONSTELLATION OF AQUILA,
SHOWING THE TWO BRANCHES SEPARATED BY A RlFT RELATIVELY DEVOID
OF STARS.
TWIN, PULSATING AND NEW STARS 177
that at its greatest brightness it was not less than
1 6 million times brighter than it now is.
Another brilliant nova appeared a few years later
in 1604 and was observed, amongst other astrono-
mers, by Kepler, whose patient investigation of
Tycho Brahe's observations of the planets led to the
discovery of the celebrated laws of motion of the
planets, which made it possible for Newton to dis-
cover the law of gravitation. The nova of 1604,
though not so bright as Tycho' s nova, became
brighter than Jupiter or than any other star. It
remained visible to the naked eye for about 16
months.
The next brightest nova of which we have record
appeared comparatively recently, in June 1918, and
was discovered independently by many observers.
The rise in brightness of this star was rapid. On
June 5 it was of normal brightness. At its discovery
on June 8, it had increased in brightness ten-thou-
sand-fold. The next evening it was the brightest
star in the whole sky, with the exception of Sirius
and Canopus.
The most recent nova was discovered on Decem-
ber 13, 1934, by Mr. J. P. M. Prentice, a young
British amateur astronomer. The star was of about
the third magnitude when it was discovered, but
subsequently brightened to the first magnitude. On
photographs taken before the outburst, it appeared
as a very faint star, the total increase in brightness
being about 4OO,ooo-fold. The star remained
visible to the naked eye for about 4 months, after
which there came a rapid fall in brightness.
The great increase in brightness which occurs
122
178 WORLDS WITHOUT END
when a nova flares up must be clue to one or both
of two factors to an increase in the surface tempera-
ture of the star, which is equivalent to an increase
in the candle-power per square foot of the surface of
the star, or to an increase in the total surface area
of the star. We can estimate the surface temperature
of the nova from its spectrum and there is no evi-
dence of any abnormally high temperature at maxi-
mum. So although in general we have little or no
information about the temperature before the flare-
up, we can conclude that the brightening must be
due in the main to increase in surface area. In
other words, the outburst is accompanied by a rapid
and extensive swelling up of the star. If, for in-
stance, the increase in brightness at the flare-up is
a million-fold, the star must swell up so that its
radius becomes 1,000 times its initial value if there
is no change in surface temperature, but if there is
a five-fold increase in temperature the radius will
only be 40 times its initial value.
At the peak of the outburst the star throws off
its outer shell of atmosphere. This gaseous shell
travels outwards from the star with a velocity usu-
ally of some hundreds of miles per second. In the
case of the nova in the constellation Aquila, which
flared up in June 1918, the expanding shell of gas
can still be photographed, growing uniformly as it
continues to recede from the star with the high velo-
city of about 1,100 miles a second.
After the peak is passed, the nova slowly sinks again.
This shrinkage is accompanied by a considerable in-
crease in temperature. There is some doubt as to the
final state of the star, because its faintness makes it
TWIN, PULSATING AND NEW STARS 179
difficult to learn much about it. It seems probable
that it finally becomes a small star of very high
density, possibly comparable to the small, dense
companion of Sirius.
The intrinsic luminosity or candle-power of a nova
when at its brightest is very high, being some ten or
twenty thousand times brighter than the Sun.
Novas have been detected in some of the distant
stellar systems, and they have provided valuable
corroborative evidence of the distances of those
systems inferred from the observation of Cepheid
variables.
It is only when a nova is within a distance of a
few thousand light-years that it becomes a conspicu-
ous naked-eye object. If a star at a distance greater
than 10,000 light-years becomes a nova, it is
not probable that it could be seen with the naked
eye. But many such outbursts of distant stars have
been detected on astronomical photographs. Bright
naked-eye novae have appeared of recent years in
1918, 1920, 1925 and 1934. The true frequency of
nova outbursts is not to be judged from these some-
what rare and spasmodic appearances. It seems
probable that in our Milky Way system of stars, the
number of novae which flare up averages about
30 a year. The total mass of this system is of
the order of 2,000 million times the mass of
the Sun. As there is not a wide range in stellar
masses, we may conclude that in the course of a few
thousand million years there will be as many nova
outbursts as there are stars. At the present time
astronomers are somewhat divided in opinion as to
the ages of the stars ; some favour a relatively short
l8o WORLDS WITHOUT END
life of a few thousand million years (comparable
with the age of the Earth) ; others favour a much
longer life. Whichever view we accept, wemustcon-
clude that on the average every star passes through
a nova outburst at least once in its life-history. It
may be that some stars pass through this stage many
times during their lifetime. If the stars have the
longer lifetime mentioned above, this would seem
to be inevitable. It is therefore not improbable that
the nova flare-up is a sort of distemper which few
stars are likely to escape.
The cause of these remarkably violent outbursts,
which are accompanied by an enormous expansion
of the star and a vastly increased output of energy,
is not known for certain. From one point of view
they are of special interest to us ; they provide our
only opportunity of seeing the evolution of the stars
actually in progress. For the most part this evolu-
tion takes place with such extreme slowness, judged
by ordinary human standards, that if we could see
the stars as they appeared to our first parents we
should be able to detect little or no change in the
vast majority of them.
The most plausible theory that we can suggest at
the present time supposes that the star, during the
slow progress of its evolution, arrives at a stage
when its equilibrium becomes unstable. When this
occurs, there will be a rapid transition to a new state
of stability. We can illustrate in a crude way by a
jug standing on a flat table. If the jug is tilted
slightly and then released, it will fall back to its first
position; it is said to be in stable equilibrium. But
if it is tilted more than a certain amount, it will no
TWIN, PULSATING AND NEW STARS l8l
longer return to its former position but topples over
and takes up a new position of stable equilibrium on
its side.
In collapsing from one stable position to another
stable position there is a sudden release of gravita-
tional energy within the star. The outrush of this
energy will temporarily distend the star and increase
its candle-power. When the energy has been re-
leased, the star settles down in a new stable state as
a small star of very high density.
Our own Sun has not passed through the nova
stage. Some astronomers believe that it is showing
incipient signs that it is approaching this stage.
Even if this were the case, the outburst might not
occur for millions of years yet a short interval in
the life-history of a star. But should the Sun be-
come a new star, everything on the Earth would very
quickly be burnt and in the course of a few hours
the Earth itself would become merely a cloud of hot
gases ; the Sun might even swell up to such an extent
that it would swallow the Earth. A sudden death
by heat is one possible end of our Earth.
CHAPTER X
OUR STELLAR UNIVERSE
IN the preceding four chapters we have considered
the stars in general ; how their distances, luminosi-
ties, masses and temperatures are determined and
how one star differs from another. We^have also
discussed in more detail some special classes of stars
of particular interest. But we have not yet con-
sidered how the stars are distributed in space. Are
they scattered at random, as though sprinkled
through space from a vast celestial pepper-pot, or is
there any indication that the Universe has a definite
structure ? The first serious attempt to answer this
question was made about 150 years ago by William
Herschel, who has been called " the father of
modern astronomy."
If we look carefully at the sky on a dark moonless
night, we shall notice that the few thousand stars
which can be seen by the naked eye are not by any
means distributed uniformly over the sky, but that
they are considerably more numerous in or near the
Milky Way than in the other portions of the sky.
The Milky Way or Galaxy is the name given to the
broad belt of faint luminous haze which encircles
the whole sky. In the northern sky it passes within
about 30 of the north pole, runs through the
constellations of Cassiopeia, Perseus and Auriga to
the horns of Taurus. Then it passes between
Orion and Gemini, through Monoceros to Argo, the
Southern Cross and the feet of the Centaur. Here it
182
PLATE XXIV. PORTION OF HIE MILKY WAY IN IHE CONST ELLAI ION OF
SAGITTARIUS.
PLATE XXV. PORTION OF THE MILKY WAY IN THE CONSTELLATIONS OF
SCORPIO AND OPHIUCHUS.
OUR STELLAR UNIVERSE 183
divides into two branches, the brighter of which
passes through Ara, Scorpio, the bow of Sagittarius
and Aquila to Cygnus, where it rejoins the other
branch. A portion of the Milky Way in the con-
stellation Aquila is reproduced in Plate XXIII; the
two branches, with the dark rift between, are
shown.
The Milky Way has had an important place in
the mythology of many primitive peoples, and has
often been regarded as the road traversed by the
souls of the departed. Longfellow relates how,
when the wrinkled old Nokomis nursed the little
Hiawatha :
" Many things Nokomis taught him
Showed the broad, white road to heaven.
Pathway of the ghosts, the shadows,
Running straight across the heavens"
The concentration of the bright stars towards the
Milky Way suggests that we may there find the clue
for unravelling the problem of the structure of the
Universe. When William Herschel first developed
an interest in astronomy he bought a copy of Fer-
guson's Astronomy, the best text-book of the day,
which passed through many editions. The almost
complete ignorance about the stars at that time is
reflected in Ferguson's book, in which twenty-one
chapters are devoted to the solar system and one
chapter only to the stars. The then current know-
ledge about the Milky Way is summed up in one
brief paragraph:
" There is a remarkable tract round the Heavens
184 WORLDS WITHOUT END
called the Milky Way from its peculiar whiteness,
which was formerly thought to be owing to a vast
number of very small stars therein; but the telescope
shows it to be quite otherwise; and therefore its
whiteness must be owing to some other cause."
This is not much more informative than the old
Greek view that the Milky Way arose from a few
drops of milk which the infant Hercules let fall from
the bosom of Juno.
When Herschel turned the telescopes of his own
making, which far surpassed in optical quality any
telescopes previously made, upon the Milky Way,
he found that Ferguson's description was wrong and
that Bacon's statement was correct that " the Milky
Way in the sky ... is a meeting or knot of a num-
ber of small stars, not seen asunder but giving light
together." Herschel describes how " the glorious
multitude of stars of all possible sizes that presented
themselves here to my view was truly astonishing."
Herschel found that the number of stars brighter
than any given limit of apparent magnitude seen in
the field of his telescope was greatest in the Milky
Way and that the number decreased progressively
with increase in distance from the Milky Way; the
disparity in numbers becomes greater the fainter the
stars. If we divide the sky into small squares
measuring one degree each way, then in the Milky
Way there is on the average one star visible to the
naked eye in every 8 squares; but at a distance
of 90 degrees from the Milky Way, there is only one
such star in every 27 squares. But when we con-
sider stars down to magnitude 20, we find from re-
cent counts of stars that there is an average number
OUR STELLAR UNIVERSE 185
of 40,000 in each square in the Milky Way, but only
i ? 200 in each square at a distance of 90 degrees from
the Milky Way.
Herschel's investigations led him to the conclusion
that the stars are grouped in space in the form of a
flattened disc, like a millstone, with the radial ex-
tension of the system much greater than its thick-
ness. He believed that the Sun was situated some-
where near the centre of this system. In the direc-
tion of the Milky Way we look through the system
to its most distant limits, but the system is so flat-
tened that in a slightly different direction we look
through a much smaller depth. The large number
of faint stars to be seen in the Milky Way is a con-
sequence of the great extension of the system in the
radial direction.
The Milky Way is by no means uniform in bright-
ness, nor is the distribution of the stars to be seen
within it at all uniform. There are numerous local
aggregations of stars, or star-clouds, in the Milky
Way. The brightest region, containing the densest
aggregation of stars, is in the constellation of Sagit-
tarius, in the southern sky. A portion of this region
is shown in Plate XXIV; the irregular distribution
of the stars is well illustrated by this photograph.
Cepheid variables have been found in some of the
star-clouds, and that has made it possible for their
distances to be estimated. The star-clouds are all
very remote from us ; some of the distances which
have been measured are as great as 30,000 light-
years. It would seem probable that the Galaxy
extends to distances much greater than this, for we
may suppose that there are more distant star-clouds,
l86 WORLDS WITHOUT END
hidden from view by the nearer ones. We shall see
that there are methods by which a reasonably good
estimate of the total dimensions can be obtained.
In many parts of the Milky Way we find not stars
only but hazy patches of faint greenish light, which
the most powerful telescope will not separate into
faint stars. These patches of delicate wispy light
are called nebulae, from the Latin word for a cloud.
The most beautiful is the Great Nebula in Orion
(Plate XXVI), which can be seen with the naked
eye as the faint hazy patch in the middle of the
dagger of Orion. This nebula was mentioned by
Peiresc in 1611, and was rediscovered by Huyghens
in 1656. Huyghens believed that it was a " hiatus
in the sky, affording a glimpse of more luminous
regions beyond."
The nebulae were studied in great detail by Wil-
liam Herschel, who had catalogued 2,500 of them by
the year 1802. Herschel was at first of the opinion
that they were aggregations of very distant stars.
Just as to the naked eye the Milky Way appears as a
hazy nebulosity which the telescope resolves into
faint stars, so, thought Herschel, the nebulae might
be much more distant aggregations of stars, that
were beyond the power of his telescope to resolve.
But when in 1790 he discovered a star surrounded
by a faint luminous atmosphere of a circular form,
he concluded that the nebulae were rarefied clouds
of gas. Direct proof of the gaseous nature of the
nebulae was not obtained until 1 864, when Sir Wil-
liam Huggins first observed their spectra. He found
that their light does not contain all the colours of the
rainbow but consists of a number of discrete radia-
OUR STELLAR UNIVERSE 187
tions. This is the type of spectrum given by a
glowing gas of low density, as we have already
seen.
The gaseous nebulae appear green because the
light which they emit in the visual region of the
spectrum consists mainly of two radiations in the
green. The spectral lines corresponding to these
radiations have never been found in the laboratory.
When they were discovered in the nebulae, they were
attributed to an unknown element which was called
nebulium, following upon the precedent of calling by
the name helium the element responsible for the
strong unknown line in the flash spectrum of the
Sun. But as physical investigation unravelled the
structure of the atoms of the various elements, it was
realised that there was no gap between the known
elements remaining to be filled by nebulium. The
only possible alternative explanation of the strange
lines was that they were emitted by a known ele-
ment, which was under conditions which the phy-
sicist had not reproduced in his laboratory. The
conditions in the nebula are, in fact, different from
those obtainable in the laboratory; the density of the
nebulous gas is not more than one-millionth that of
the residual gas left in the most perfect vacuum
that can be obtained with the aid of modern high-
vacuum technique, and the radiation from the
stars, which is absorbed and emitted again by the
nebulosity, is extremely weak. The atoms are there-
fore in a much more quiescent state than is possible
in the laboratory. The secret of nebulium has now
been unravelled, for theoretical calculations have
established that the green lines are produced by
l88 WORLDS WITHOUT END
doubly ionized atoms of oxygen, i.e. by atoms of
oxygen which have lost two electrons.
The nebulae are not self-luminous but shine by
the light of the stars which are embedded in them.
This is more clearly apparent in the case of some
nebulae than others, but it seems certain that all the
green nebulae shine in this way. The small con-
stellation of the Pleiades provides the clearest illus-
tration. A photograph with a moderate exposure
shows a bright patch of nebulosity around each one
of the brighter stars ; a photograph with a long ex-
posure shows that much fainter nebulosity extends
throughout a considerably larger area.
But though it is the light of the stars embedded in
the nebulae which causes them to shine, they do not
merely reflect the light, in the way that the Moon
shines by reflecting the light of the Sun, or scatter
it, in the way that the light from motor-car head-
lamps is scattered by a foggy atmosphere. The
light from the stars is actually absorbed by the atoms
in the nebula and is re-emitted in radiations of dif-
ferent wave-length. We can compare the action of
the nebulae to that of the luminous paint on the
hands of a watch, which absorbs light but emits a
different kind of light. It is only the blue stars of
high temperature that can stimulate the nebulae
to shine; the Sun and even the giant red stars such
as Antares or Betelgeuse cannot do so, because their
temperatures are too low.
Another feature of the Milky Way which at-
tracted HerschePs notice was dark patches, looking
very much like small clouds. In some of the densest
star-clouds in the Milky Way there are what appear
X
X
B
rH
s
OUR STELLAR UNIVERSE l8g
to be gaps, with few or no stars in them. The most
striking example is one called the " Coalsack," ad-
jacent to the Southern Cross, which is easily visible
to the naked eye and has the appearance of a small
cloud hiding the portion of the Milky Way behind
it. Herschel thought that these vacant regions were
lanes or channels through the star-clouds. A long
and careful study of these dark patches was made by
Barnard, who catalogued nearly two hundred of
them. Some examples of these dark regions are
shown in Plate XXVII; many others will be seen
in Plate XXV. That there should be so many
vacant lines through the great depth of the star-
clouds in the Milky Way pointing directly to the
Earth is so exceedingly improbable that another ex-
planation must be found. Barnard suggested that
there are opaque clouds between us and the Milky
Way. These clouds screen from us all the stars
which lie behind them and we see only those stars
which lie between us and the clouds. Some of the
clouds must be relatively near, for not a single star
can be seen within their boundaries ; others are more
distant and we see the foreground stars projected on
them. Some are small and others of enormous
extent.
We may have dark gaseous clouds which are simi-
lar in every respect to green nebulae, but are not
luminous because they do not contain any very hot
stars. But such clouds must be almost completely
transparent, because their density is extremely
small. Light passing through a gaseous nebula
extending over a distance of one thousand light-
years would be less absorbed than in passing through
IQO WORLDS WITHOUT END
the atmosphere of the Earth. The opaqueness of
the obscuring clouds must be due to the presence of
extremely fine dust. Dust particles comparable in
size with the wave-length of light have very great
obscuring power. If the average amount of dust in
the cloud is only one-fifty-thousandth of an ounce
in each square inch of cross-section, the cloud will
be completely opaque, whatever its thickness may
be. It is probable that the clouds consist of a mix-
ture of dust particles of various sizes, and they may
even contain small meteoric stones. We have men-
tioned that many small meteors enter the atmo-
sphere of the Earth from outer space.
The opaque clouds and the luminous nebulae fre-
quently occur in close association with one another.
A striking example is provided by the nebulous
region around Rho Ophiuchi (Plate XXVIII),
where there is a conspicuous dark lane and at its
head an extensive region of luminous nebulosity.
It is probable that the two classes of nebulae are
^essentially the same ; the presence of the fine dust
particles will make any nebula opaque unless there
are high-temperature stars suitably placed near its
surface to illuminate it. Some of the bright nebu-
lae, however, such as the diffuse nebula in Cygnus
(Plate XXVI), show no traces of absorbing dust and
may be almost, if not entirely, gaseous.
We have mentioned that over a great portion
of its extent the Milky Way is divided into two
branches. The appearance of two branches is
caused by the existence in the central regions of
the Milky Way of an obscuring cloud of very
wide extent. This widespread cloud hides the
OUR STELLAR UNIVERSE IQI
distant star-clouds in the middle of the Milky Way
zone (Plate XXIII).
The luminous nebulae and the obscuring clouds
afford direct evidence that both gaseous matter and
fine dust are very prevalent throughout the galactic
regions. May it not be possible for extremely dif-
fuse matter to be scattered throughout other parts
of the Milky Way, where we see neither the lumin-
ous nor the dark clouds ? It is possible to give a
definite answer to this question, and the answer is in
the affirmative. The evidence for this answer is
provided in several different ways. We shall men-
tion two only.
For the first of these we must anticipate somewhat
the next chapter. In that chapter we show that
our Galaxy or Milky Way system is but one of many
millions of separate island universes scattered
through space. These stellar systems are distri-
buted more or less uniformly in all parts of the sky
except those which are near to or within the region
of the Milky Way. As we approach the Milky Way,
the numbers begin to fall off, and within a zone
which forms a continuous irregular belt along the
Milky Way, of a normal width of from 10 to
20 degrees, not a single one can be found. The
explanation is not that no island universes exist
within such a zone, but that there is sufficient ab-
sorbing matter throughout the Milky Way region,
even in directions where there is no apparent
evidence of opaque clouds, completely to cut off
their light from our view.
The second piece of evidence is provided by the
stars themselves. When we analyse, by means of a
WORLDS WITHOUT END
spectroscope, the light reaching us from the Sun, we
see the band of prismatic colours crossed by numer-
ous fine dark lines produced by the atoms of calcium,
iron, titanium, etc., in the atmosphere of the Sun.
But in addition to these lines, we see also other dark
lines that are due to absorption by oxygen, nitro-
gen, water-vapour, etc., in our own atmosphere.
There is a simple way by which we can distinguish
between the absorptions that are present in the light
when it leaves the Sun and the absorptions that
occur during the passage of the light through the
atmosphere of the Earth. It is only necessary to
compare the light from the eastern edge of the Sun
with the light from the western edge. As a result
of the Sun's rotation, one edge is moving towards us
and the other edge away from us. The wave-lengths
of the lines in the spectrum of the light from the one
limb are therefore slightly decreased and those in
the spectrum of the light from the other limb are
slightly increased. But the lines which originate in
the Earth's atmosphere remain undisplaced and can
therefore be distinguished.
A method identical in principle can be used to
tell whether any lines in the spectrum of a star may
be due to absorption by diffuse clouds of^gas in
interstellar space. We have seen how in the spectra
of close twin stars the lines swing backwards and
forwards because of the orbital motion. In 1904
Hartmann was observing the spectrum of Delta
Orionis, one of the three stars in the belt of Orion.
This star is a close twin star, one of the stars being
much brighter than the other, so that we only
see the spectrum of the brighter star. Hartmann
PLATE XXVIII. NEBULOUS REGION IN OPHIUCHUS, SHOWING OPAQJJE CLOUDS
AND LUMINOUS NEBULOSITY.
(a) THE GLOBULAR CLUSTER, OMKCA CENTAURI.
SPIRAL NEBULA, SEEN EDOEWISE-ON, IN THE CONSTELLATION
OF BERENICE'S HAIR. NOTK THE ABSORBING BELT.
PLATE XXIX.
OUR STELLAR UNIVERSE 193
noticed that as the lines moved backwards and for-
wards whilst the star revolved in its orbit, there were
certain lines due to calcium vapour that remained
stationary. Such lines could not originate in the
Earth's atmosphere and must therefore presumably
originate in interstellar space.
Evidence has since accumulated that proves this
beyond any possibility of doubt. We find, for in-
stance, that the stationary lines are only seen in the
spectra of stars at distances greater than about
i ,000 light-years and that the more distant the
star the stronger they are. The principal lines in
the spectrum of a star which are produced by ab-
sorption in interstellar space are the lines of calcium
ana sodium vapour.
It is possible to make an estimate of the average
density of this interstellar matter. The density
proves to be so low that there are only about half
a dozen atoms in every cubic inch. To appreciate
how extremely small such a density is, we must
realise that in the most perfect vacuum that the
physicist can produce in the laboratory, with the
aid of the most elaborate modern high-vacuum
pumps, there still remain about 100,000 million
atoms in every cubic inch. Another illustration is
perhaps even more striking ; if we throw a cup of
water into the sea and let it mix thoroughly with the
water of the oceans, and if we then draw out a
cupful of sea water from any part of the sea, it will
contain several dozen molecules of the water that
was originally thrown into the sea.
The number of atoms in each cubic inch of inter-
stellar space is small, but atom after atom takes its
13
194 WORLDS WITHOUT END
toll from the light as it passes through, and at length,
after it has travelled a distance of about 1,000
light-years, we are just able to detect the effect.
The interstellar cloud of gas is not uniform in distri-
bution; it pervades the whole of the galactic regions,
but here and there it is strongly condensed, and where
these condensations occur we say that there are
nebulae. The nebulae are but the visible signs of
all-pervading gaseous matter which for the most part
we cannot see, though we can detect its effects.
The distances of some of the star-clouds in the
Milky Way are measured by tens of thousands of
light-years. We should like to obtain a more con-
crete idea of the actual dimensions of our galactic
system and of the position in it that the Earth oc-
cupies. This is made possible by a group of objects
of particular interest, called globular clusters. The
name expresses accurately their appearance. Each
cluster is a system, globular in shape, containing
many thousands of stars ; the stars are most densely
clustered at the centre, the density decreasing from
the centre outwards, at first rapidly and then more
gradually. The appearance of a globular cluster is
very much like the appearance of a target which
has been shot at for a large number of rounds by a
good marksman. The shots are thickly scattered
within the bull's eye; in the inner they are less
numerous; in and around the outer they are com-
paratively scarce. The brightest and nearest of the
globular clusters is called Omega Centauri (Plate
XXIX). To the naked eye it appears as a hazy
star of the fourth magnitude; it was noted by Halley
in 1677. Most of the clusters were observed by
OUR STELLAR UNIVERSE
Messier in the eighteenth century, though he knew
them only as nebulosities. It was William Herschel
and his son John who showed that they consisted of
myriads of stars.
The progressive increase in the optical power of
telescopes has revealed to the astronomer more stars,
more nebulae and more planets. But for the last
half-century it has made practically no addition to
the number of the star clusters. We may conclude
that the reason is that there remain few, if any, still
to be discovered. The distribution of the clusters
is rather remarkable, for they are nearly all to be
found in one hemisphere of the sky.
The distances of the globular clusters can be
determined, because Cepheid pulsating stars are
found to occur in them. They are all distant sys-
tems ; the nearest clusters are Omega Centauri and
47 Tucanae which are at a distance of about
18,000 light-years. The distances of other clusters
range up to about 140,000 light-years. Knowing
the distances, we can infer the sizes of the clusters;
we find that they are fairly large systems, having
diameters of the order of several hundred light-
years.
When we map out the positions of the clusters,
they are seen to form a flattened group, more or less
symmetrically distributed with regard to the Milky
Way. Our Sun is not centrally placed with respect
to the group, but lies well out towards one side of it.
This explains why most of the clusters appear to us
to be in one half of the sky. The centre of the whole
group is in the direction towards the constellation
of Sagittarius and at a distance of about 32,000
ig6 WORLDS WITHOUT END
light-years. It would seem not to be without signi-
ficance that the centre of the group lies in the direc-
tion of the densest and brightest portion of the
Milky Way, and we are tempted to infer that the
centre of our galactic system is also the centre of
the group of clusters. The greatest diameter of the
system outlined by the globular clusters is some-
where about 150,000 light-years; the thickness of the
system is from 25,000 to 40,000 light-years at its
central portion, but is not so great in the neighbour-
hood of the Sun.
The galactic system formed by the star-clouds and
gaseous nebulae is probably coextensive with the
system of the globular clusters, though it may be
that the clusters extend out beyond the star-clouds.
There is no doubt, however, that we must regard
the clusters as belonging to our galactic system, and
that the greatest dimension of this system is not
less than 150,000 light-years.
Such a distance is the greatest of which we have
so far had occasion to speak. If we reduce the scale
so as to get a better appreciation of relative dimen-
sions and represent 150,000 light-years by the dia-
meter of the Earth 8,000 miles then on the same
scale the distance of the Earth from the Sun is only
one-twentieth of an inch. The diameter of the Sun
becomes only about three times the wave-length of]
light, so that the Sun is comparable in size to a
microscopic dust particle.
We have another method, fortunately, by means
of which we can, get confirmatory evidence of the
size of the galactic system. The flattened shape of
this system suggests that it is in rotation ; if it were
OUR STELLAR UNIVERSE 197
not rotating we might expect it to be globular. It
is only within the last few years that the rotation has
been detected. The rotation of the system must be
controlled by the general gravitation of the whole
system. The outer parts will therefore revolve more
slowly than the inner parts, just as in the solar sys-
tem the outer planets revolve more slowly than the
inner planets, and as the outer parts of Saturn's rings
rotate more slowly than the inner parts. This is
different from the rotation of a rigid body such as
a wheel, in which the outer parts revolve more
rapidly than the inner parts.
If, then, we assume that the galactic system is in
rotation, we shall expect to find that the stars lying
between us and the centre of rotation are moving
more rapidly than the Sun, and the stars more re-
mote from the centre are moving more slowly.
Superposed on the motion of the stars resulting from
the rotation of the system there are, of course, the
random motions of the individual stars. A statis-
tical analysis of the mean motions of groups of stars
in different parts of the sky is necessary in order to
average out the random motions and to reveal the
differential motion produced by the rotation. The
greater the distance of the groups of stars from the
Sun, the larger the differential motion will be.
We do not make any assumption about how far
away the centre of the rotation is, nor about its
direction from the Sun.
The results of such a statistical analysis prove to
be exactly in accordance with our expectation. The
centre of the system indicated by this analysis is in
the direction towards the dense star-clouds in Sagit-
ig8 WORLDS WITHOUT END
tarius. It will be remembered that this was also the
direction towards the centre of the globular clusters.
But we can go further : we can get an estimate of the
distance of the Sun from the centre of rotation and,
since the rotation is controlled by the gravitation of
the system as a whole, the total mass of the system
can be deduced. The distance of the Sun from the
centre proves to be about 32,000 light-years; this is
in agreement with the estimated distance of the
centre of the globular clusters and confirms the
assumption which we previously made that the
group of clusters could be supposed distributed hap-
hazardly throughout the galactic system.
The controlling mass of the system proves to be
about iGOjOgo million times the mass of the Sun.
In this total mass is included the masses of all the
stars, including any stars which may have ceased to
shine and also the mass of the diffuse matter scat-
tered throughout the system. We might think that
the total mass of the interstellar gas, whose density
is some thousands of millions of times smaller than
the most perfect vacuum we can make, would not
amount to very much. But when we make the
computation we find that the total mass of this ex-
tremely rarefied gas is approximately equal to the
total mass of all the stars. The mass of the Sun is
about 2,000,000,000,000,000,000,000,000,000 tons;
to write down the total mass of the interstellar gas
in tons we start with the figure i or 2 and follow it
with 38 zeros. The comparison between the total
mass of this rarefied gas and the total mass of the
stars is a striking commentary on the one hand on
the great distances apart of the stars, and on the
OUR STELLAR UNIVERSE IQ9
other hand on the vast dimensions of our galactic
system.
The time of one complete rotation in the neigh-
bourhood of the Sun is about 225 million years.
This may appear a slow rate of rotation, but the
dimensions of the system are so great that the Sun
has a motion through space, arising from the rota-
tion, of 170 miles in a second. If we accept 3,000
million years as the probable age of the Earth, it
follows that more than a dozen complete rotations
have occurred during the lifetime of the Earth.
We have seen that the Sun occupies a very eccen-
tric position in the galactic system. When the dis-
tribution of stars in the immediate neighbourhood
of the Sun and out to a distance of about 1,000
light-years is investigated, a steady decrease in
density of stars in all directions outwards from the
Sun is found. It appears therefore that the Sun is
situated in a localised cluster, which is probably a
star-cloud similar to the star-clouds that are dis-
tributed throughout the Milky Way. This local
cluster or star-cloud, like the larger galactic system
of which it forms a part, is much flattened ; its median
plane does not lie exactly in the Milky Way but is
tilted at a slight angle to the Milky Way. The diam-
eter of this cluster is about 2,000 light-years; the
Sun is about 300 light-years from its centre, which
lies in the direction of the southern constellation
of Carina.
The galactic system appears therefore to be a vast
flattened rotating system extending over a distance
of about 150,000 light-years, and containing many
thousands of millions of stars sparsely scattered
2OO WORLDS WITHOUT END
throughout it. The distribution of the stars is far
from uniform; there are numerous local aggrega-
tions or clouds of stars. Permeating the space be-
tween the stars is an extremely rarefied gas that
in places reaches a higher density and is revealed to
us either as luminous nebulous clouds or as opaque
clouds. The total weight of this gas is comparable
with the total weight of the stars. The Sun is
situated in a somewhat eccentric position in a star-
cloud that lies far out from the centre of the sys-
tem. The stars are most closely aggregated around
the central nucleus of the system, which lies in the
dense star-clouds in Sagittarius.
A ray of light takes 8 minutes to come to us
from the Sun. To cross from one end of the solar
system to the other it takes 1 1 hours. To travel
to the nearest star it takes 4 years. The journey
across our local star-cloud takes 2,000 years ; that
from the Sun to the centre of the galactic system
takes some 32,000 years, whilst to travel from one
end to the other of the whole galactic system some-
thing like 150,000 years is needed.
CHAPTER XI
CELESTIAL CATHERINE-
WHEELS
THE bright green-coloured gaseous clouds and the
dark absorbing clouds, considered in the last chap-
ter, belong to our galactic system and occur ex-
clusively in the Milky Way regions. There is an-
other type of cloud or nebula, which does not show
the characteristic green colour but appears white to
ithe eye. The members of this group we may there-
fore denote for the present by the designation
" white nebulae." Sir William Huggins found in
1867 that the spectrum of a white nebula is entirely
different from that of a green nebula. As we have
seen, the spectrum of the latter consists of a number
of isolated bright lines, which is the type of spectrum
characteristic of glowing gas : the spectrum of a white
nebula contains radiations of all colours, giving the
prismatic band, crossed by dark lines, and bears
more resemblance, therefore, to the spectra of the
stars than to the spectra of the green nebulae.
The green nebulae are usually irregular in shape
and outline, and many of them extend over large
areas of the sky. The white nebulae are for the most
part fairly compact and regular in shape; they
usually appear as faint patches of light, with a
brighter but somewhat ill-defined nucleus at the
centre. With a few exceptions, they are incon-
spicuous objects in the telescope and can best be
studied by means of long-exposure photographs
obtained with large telescopes. The largest and
2O2 WORLDS WITHOUT END
brightest of the white nebulae is known as the Great
Nebula in Andromeda ; it can be seen with the naked
eye as a faint hazy patch of light (Plate XXXI).
The white nebulae are far more numerous than
the green and are found in all parts of the sky except
in or near the Milky Way. Thus the white nebulas
and the green nebulae are mutually exclusive. In
and near the Milky Way, the green nebulae only are
to be found; outside these limits, the white nebulae
only are to be found.
These differences suggest that the two classes of
nebulae are entirely different in nature. In 1845,
the discovery was made by Lord Rosse with his
large 6-foot reflector that one of the white nebulae,
No. 51 in the catalogue by Messier, and hence
known technically as Messier 51, had a definite
spiral structure. This nebula, in the constellation
of the Hunting Dogs, is now commonly known as the
Whirlpool. Messier had described it as a double
nebula, without stars; Sir John Herschel de-
scribed it as a bright round nebula, surrounded by
a halo or glory at a distance from it and accom-
panied by a companion. It is of interest to compare
the drawings of this object by Herschel and Lord
Rosse with one another, and with a modern photo-
graph with a large telescope (Plate XXX) . Though
HerschePs drawing does not show any spiral struc-
ture, there are marked similarities between it and
the photograph; the spiral structure is more pro-
nounced in Lord Rosse's drawing than it really is,
according to the photograph. It was also noted by
Lord Rosse that on the finest nights the convolutions
could be seen breaking up into stars. This appear-
*!
ei
I
(a) THE GREAT SPIRAL NEBULA IN ANDROMEDA.
(b) ENLARGEMENT OF SOUTHERN PORTION
OF ANDROMEDA NEBULA.
PLATE XXXI.
[203,
CELESTIAL CATHERINE-WHEELS 203
ance is well shown in the photograph. The dis-
covery of the spiral structure in the Whirlpool was
followed by further discoveries of a similar type of
structure in other white nebulae. Such a definite
structure is something quite different from the
diffuse and formless green nebulae.
When the spiral structure has once been realised,
it is easy to recognise a similar structure in the Great
Nebula in Andromeda (Plate XXXI). If the
Whirlpool nebula is a flat system, that happens
to be seen practically broadside-on, and we imagine
it tilted through a considerable angle towards the
edge-on position, we should expect it to appear
somewhat like the Andromeda nebula. But though
this nebula is easier to observe than the Whirlpool
nebula, the fact that it is not seen broadside-on
made the spiral structure more difficult to recognise.
We can find other white nebulae, similar in their
main features, inclined at all angles to the line of
sight, from " broadside-on " to " edge-on. 55 We
may conclude from this continuous transition from
one form to the other that a nebula, such as the one
in the constellation of Berenice's Hair reproduced in
Plate XXIX, is also a spiral system. The appear-
ance of this nebula, seen edgewise-on, confirms our
assumption that a nebula such as the Whirlpool is
much flattened, having a far greater extension in
the plane of the spiral arms than in the perpen-
dicular direction.
The spiral structure is therefore not uncommon
amongst the white nebulae. They have the appear-
ance of celestial Catherine-wheels, suggesting irre-
sistibly that they are spinning round in space.
204 WORLDS WITHOUT END
We naturally turn to the most favourably placed
of these nebulae, the Great Nebula in Andromeda,
to endeavour to learn something more of the nature
of these objects. It will be noticed that it has a
bright diffuse nucleus and that the brightness of the
spiral arms which extend outwards from the nucleus
is far from uniform. The outer portions of the
nebula, when photographed with powerful tele-
scopes, are resolved into what appears to be a multi-
tude of extremely faint stars. The doubts that were
at first felt about the interpretation of these faint
spots of light as images of stars have been completely
removed by the careful investigations of Dr. Hubble
at the Mount Wilson Observatory. By comparing
large numbers of photographs with one another he
has found that many of these bright points vary in
brightness in a regular manner, and that the changes
in brightness are of that particular type which we
have learnt to be characteristic of pulsating stars :
the fading from greatest to least brightness is slower
than the subsequent brightening. These minute
points of light must therefore be images of pulsating
Cepheid stars. We found in Chapter IX that these
pulsating stars are all giant stars of extremely high
candle-power, many hundreds or thousands of times
brighter than the Sun. In the Andromeda nebula,
therefore, stars of high candle-power appear as faint
spots of light which we can detect only with very
powerful telescopes. A star in this nebula, of the
moderate candle-power of the Sun, would not be
detected as a star but would contribute its quota to
the general faint unresolved luminosity. The ob-
vious conclusion to be drawn is that the nebula is
CELESTIAL CATHERINE-WHEELS 205
at a very great distance from us. By determining
the periods of pulsation of the Cepheid stars, we can
find their in trinsicjuminosities and therefore deduce
the distance of tfie nebula. This proves to be about
870,000 light-years. The distance is much greater
than tEe dimensions of our galactic system, and we are
compelled to conclude that the Andromeda nebula
is a system that lies far outside our galactic system.
This conclusion is somewhat startling and many
astronomers were not at first prepared to accept it.
Independent confirmation of the distance was much
to be desired. This confirmation was provided by
Dr. Hubble. He found that a particular star-image
might appear on a few photographs only and be
missing both on earlier and on later photographs.
Many instances of these temporary images were de-
tected. They suggested the flare-up of a new star
in the nebula. By taking series of plates at short
intervals, it was proved that the typical features of a
nova outburst were present the very rapid rise in
brightness, followed by a slower and rather irregular
decline. Now, though any one new star differs in
many details from any other, we have seen that at
greatest brightness their candle-power is always
very high. If we assume that the average candle-
power of the novae which have been detected in the
Andromeda nebula is the same as the average for
the novae in our galactic system, we have another
means of estimating the distance of the nebula.
This method is less precise than that based on the
pulsating stars, but it is sufficient to provide a check.
It is satisfactory to find that the distances derived by
the two methods are in reasonably close agreement.
206 WORLDS WITHOUT END
Knowing the distance of the nebula we can deter-
mine its actual size from measures of its angular
dimensions. The only difficulty is to decide just
where the limits of the nebula are. The brightness
falls off rapidly towards its extremities and it ap-
pears that even photographs with long exposures do
not show its full extension. By delicate observations
of the brightness with a photo-electric cell. Professor
Stebbins has recently shown that the brightness con-
tinues to fall off outside the limits recognisable on a
photograph, until it merges into the background of
the sky. The full extension of the nebula appears
to be not less than 60,000 or 65,000 light-years.
The conclusion is that the Andromeda nebula is a
system which is of the same order of size as our own
galactic system. The evidence indicates that it is
a smaller system than our own. This conclusion is
not necessarily final. The effect of the absorption
of light by the diffuse gas throughout the Milky Way
is to lead to a tendency to overestimate the dimen-
sions of the system. During recent years, the re-
vision of the dimensions determined for the galactic
system has resulted in reducing our estimate of its
size; the estimated size of the Andromeda nebula
has been increased. We cannot regard the dimen-
sions at present accepted for either system as neces-
sarily final, and in the course of a few years their re-
vision may still further reduce the present disparity.
It is indisputable, however, that the Andromeda
nebula is to be regarded as a separate universe of
stars, a universe which bears many strong resem-
blances to our own Universe. In Plate XXXI we
reproduce an enlarged photograph of the southern
CELESTIAL CATHERINE-WHEELS 207
portion of the nebula. This bears remarkable simi-
larity to many portions of the Milky Way : we see
the localised aggregations of the stars into star-clouds
and the presence both of luminous nebulosity and
of dark clouds. The presence of luminous clouds of
gas is proved by the appearance of bright lines in the
spectra of some portions of the nebula.
In a spiral seen edgewise-on, such as is illustrated
in Plate XXIX, the presence of a belt of absorbing
matter lying in and near the central plane of the
spiral is clearly shown. This is exactly analogous
to the widespread belt of absorbing matter in and
near the plane of the Milky Way, which causes the
apparent division of the Milky Way into two
branches through a great portion of its extent.
It is therefore with considerable justification that
we can argue that our own galactic system is a
spiral nebula. Situated as the Sun is, almost in the
central plane of the system, we can only infer the
spiral structure by analogy with many other systems
which possess this structure. We should never have
been able to guess it if we had seen no other spiral
systems in the sky. But when we consider the simi-
larity in size, when we note the presence of star-
clouds, of luminous clouds of gas and of opaque
clouds in the spiral nebulae, and when we remember
that our galactic system is flattened like the spiral
nebulae, we cannot deny that there is a strong resem-
blance. The spirals seen edgewise-on have exactly
the flat grindstone structure that Herschel imag-
ined for our own Universe.
Some of the spirals seem to have proceeded far-
ther than others in the aggregation of the matter of
208 WORLDS WITHOUT END
which they are composed into discrete stars. Some,
such as the spiral Messier 81 in the Great Bear,
shown in Plate XXXII, have stars in the outer por-
tions of the arms only; in others, such as Messier 101,
also in the Great Bear, star-clouds are found close
to the nucleus.
In these photographs of spiral nebulae it will be
noticed that the stars are not distributed evenly
along the arms, but that they tend to occur in local-
ised clusterings. We have seen that the Sun is a
member of a local cluster in our own Universe. To
complete the analogy between the galactic system
and the distant stellar universes, they also are found
to be in slow rotation, and from the rate of rotation
it can be inferred that the masses involved are many
thousands of millions of times the mass of the Sun.
It is only in the nearer universes that we are able
to detect individual stars. For the more distant
systems, where this is no longer possible, we have no
direct method of estimating the distances. But by
making certain assumptions, we can get some idea
of the distances.
The study of the nearer systems gives some justi-
fication for supposing that these stellar universes
are generally similar to one another both in size and
mass. We can make a rough estimate of the dis-
tances of the more remote systems if we assume
either that they are all equal in size, or that they are
all equal in total candle-power. It cannot be
claimed that the distance of any particular universe
derived on either assumption will be very accurate ;
we may expect, however, that the distances will be
of the correct order of magnitude.
CELESTIAL CATHERINE-WHEELS 2Og
This conclusion is confirmed in a way that we
shall now describe, which in turn leads to a method
of determining the distances of even the faintest and
most remote systems. We have explained how we
can measure the velocity in miles per second with
which a star is moving either towards or away
from us, by means of the shift of the lines in its
spectrum. The same method has been used for
the spiral nebulae and with surprising results. It is
very rare in our galactic system to find a star
moving faster than about 100 miles per second; the
velocities of the spiral systems are much larger,
usually amounting to many hundreds or thousands
of miles per second. Not only have they these
exceedingly large velocities, but they are all with
one or two trivial exceptions moving away from us,
and the farther the system is from us the faster it
is receding. These results may be illustrated by
the following figures for some of the distant
universes. The first column gives the name of the
constellation in which the system is to be found;
the second column gives the distance estimated in
the way described above ; the third column gives
Distance
Velocity
Constellation..
(millions of
(miles per
Velocity/106.
light-years).
second).
Virgo
6
560
5
Pegasus
24
2,400
Cancer
29
3,000
28
Perseus
36
3>3
3i
Coma
45
4,700
45
Ursa Major
72
7,400
70
Leo .
104
12,200
"5
2IO WORLDS WITHOUT END
the directly measured velocity. In the last
column of the above table, we give the values of
the velocities (in miles per second) divided by 106.
If the figures in this column are compared with
the figures in the second column, a close corre-
spondence is seen. In comparing the figures, it
must be remembered that the distances are based
on the assumption that the various island universes
are all equal in size. That the correlation between
observed velocities and estimated distances is so
close gives us considerable confidence in the validity
of this assumption. We see now that we can accept
the proportionality between velocity and distance
and use it to infer the distance of any universe
whose velocity has been measured. This provides
the best method of estimating the distances of the
very remote universes.
The greatest velocity yet measured is 24,300 miles
a second for a remote nebula in the constellation
of Bootes. We can infer that this nebula is at a
distance of about 230 million light-years. This is the
most remote system whose distance has been esti-
mated. It is at such a great distance that the light
by which it is photographed has been travelling
through space for 230 million years. During this
long journey the dinosaurs and flying reptiles have
appeared on the Earth and with v the slow march of
evolution have disappeared again. Many of our
mountain ranges have appeared, such as the Appa
lachians, the Rocky Mountains, the Pyrenees and the
Himalayas. And at length, when the journey was
all but finished, man appeared on the Earth. Per-
haps we can get a better picture of the length of this
CELESTIAL CATHERINE-WHEELS 211
journey if we change the time-scale and represent it
by the three score years and ten of human life. On this
time-scale, man has existed on the Earth for about 4
months ; the Christian era has lasted for 5 hours only
and the average span of human life is not more than
about 10 minutes. It is only within the last minute
or two that a telescope has been built sufficiently
powerful to photograph this remote universe.
These stellar universes are to be seen in large
numbers in all parts of the sky except where they are
hidden from our sight by the opaque clouds in the
Milky Way. It has been estimated that the 100-
inch telescope of the Mount Wilson Observatory, by
photographs with long exposure under good con-
ditions, could show us somewhere about 75 million
of these universes, if it could reach the whole of the
sky and if the opaque clouds were not blocking out
many of them from our sight. Up to the greatest dis-
tances to which this telescope can reach there is no
evidence at all of any thinning out in the frequency
with which these vast systems are scattered through-
out space. Their distribution is approximately uni-
form in space, each system being on the average at
a distance of somewhere about 2 million light-
years from its nearest neighbour.
Astronomers are not content with the distances to
which they have succeeded in exploring space and
are anxious to reach out into the unknown beyond.
So the great project of constructing a giant tele-
scope, with a mirror 200 inches in diameter (twice
the diameter of the largest hitherto made), has been
conceived. The telescope is now under construction,
but several years will be required for its completion.
212 WORLDS WITHOUT END
Meanwhile we may pause to look at the position
of the Earth in the Cosmos, as the astronomer now
pictures it. The Earth is one of the smaller satel-
lite bodies attendant upon the Sun, which is a dwarf
star in a system containing many thousands of mil-
lions of stars. In this Universe the Sun has no
pride of place, but lies far out from its centre, in one
of the many clusterings of stars which pervade it.
This Universe, of which we form so insignificant a
part, is but one of many millions of universes that
are generally similar to it. To what distances these
universes extend we know not. All we can at present
say is that out to the greatest distances, exceed-
ing 200 million light-years, to which we have been
able to explore, we have been unable to find any
sign of an approach to the limits of this system.
Each vast separate universe, held together by the
force of gravitation, is spinning round in space like
a gigantic Catherine-wheel, and it is to this rotation
that the flat shape is due.
We have mentioned that the other universes or
galaxies are all moving away from us, and the far-
ther away they are the more rapidly they are reced-
ing. At first sight it might appear that the Uni-
verse in which we happen to find ourselves is the
centre of the larger system, comprising all the uni-
verses, which we may term the Cosmos. But a mo-
ment's reflection will convince us that this is not so.
The rate at which the distance from us of any system
is changing is proportional to that distance ; in other
words, every distance is increasing at the same per-
centage rate. When one distance has doubled, all
the other distances have doubled. But this indi-
CELESTIAL CATHERINE-WHEELS 213
cates merely a uniform expansion of the whole
system. The distance between any two galaxies
must increase at the same percentage rate. It fol-
lows that if we were to be placed in another galaxy
we should observe the same phenomenon that all
the other galaxies were moving away from us. It
is not a case of an aversion of all the other galaxies
(to our own particular galaxy, but an aversion of
|every galaxy to every other galaxy. This is what
'has been termed the expansion of the Cosmos.
The expansion is taking place at such a rate that
every distance becomes doubled in about 1,300
million years. Long though this period is when
judged by ordinary terrestrial standards, the ex-
pansion, when considered from the astronomical
point of view, must be regarded as rapid. It implies
that when the Earth was born, the separate uni-
verses were much closer together than they are at
the present time; their mutual distances must all
have been less than one-quarter of their present
value. We are assuming that the expansion has
been proceeding at a uniform rate throughout the
life-history of the Earth, and this is an assumption
which may not be justifiable. If we had been liv-
ing on the Earth 3,000 million years ago we should
certainly have had a much better view of the distant
universes than we have at present; observations
which now require a telescope with an aperture of i oo
inches could then have been made with equal facility
with a telescope with an aperture of only 24 inches.
It is not possible in a book of this nature to enter
into a discussion of the theoretical implications of
the observed recession of the universes. One view
214 WORLDS WITHOUT END
which has been advanced supposes that originally
the separate universes formed one system which at
some instant exploded, the fragments scattering in
all directions. If such an explosion occurred, some
portions would have been shot off with higher
speeds than others ; but all the portions would con-
tinue to move outwards with their initial speeds,
except in so far as these were modified by the mutual
gravitational attraction. At any subsequent time
we should find that each portion would have
covered a distance exactly proportional to the speed
with which it is moving. Wherever we might find
ourselves in this exploded system, we should observe
that every portion was apparently moving away
from us, and we should have no means of identifying
where the centre of the system was. We do not
know, of course, what determined the " zero hour "
at which the explosion took place, nor why every
portion of the system was of about the same size.
Another suggestion which has been made is that
it is only by chance that we happen to see the
Cosmos expanding; it is possible that, just as some
stars pulsate, so the whole Cosmos may pulsate, alter-
nately expanding and contracting, going on endlessly
* through the same cycle. If we lived at some other
time, either in the past or in the future, we might ob-
serve an apparent approach of all the other galaxies,
which would have been even more difficult to explain
than a recession. We can picture an explosion caus-
ing all the fragments to fly apart, but the supposition
of scattered fragments all moving inwards at such a
rate that they would come together at one point at
the same instant appears much more artificial.
CHAPTER XII
THE AGE AND EVOLUTION
OF THE STARS
IN the preceding chapters we have seen the great
variety there is amongst the stars in size, in average
density, in surface temperature, in candle-power
and to a much lesser degree in weight. We are
tempted to enquire whether this variety is connected
iri any way with the ages of the stars and therefore
with their evolution. If the stars were all of the
same age should we find as great a variety as we do?
We are handicapped because we cannot see stel-
lar evolution in progress, except in the one instance
of the outburst of a new star, and we are not yet con-
fident that every star must necessarily pass through
this particular phase. The span of human life is
but an instant in the lifetime even of our Earth.
We can endeavour to obtain guidance from theo-
retical considerations. If, for instance, a certain
amount of matter is aggregated together to form a
star, can we predict what its size and candle-power
will be ? We are at once face to face with the diffi-
culty that we know nothing about the composition
of the interior of a star. Observation tells us a great
deal about the composition of the outer layer, which
is the only portion that we can actually see. But
can we necessarily assume that the composition of
the matter deep in the interior of the stars is in any
way similar to the composition of the outermost
layer ? Furthermore, we have no direct knowledge
215
2l6 WORLDS WITHOUT END
of the state of the matter in the interior. Are we
justified in assuming that a star is gaseous through-
out ? The pressure at a great depth in the star is
enormous; it is possible that the central portions of
the stars may be liquefied by the high pressure to
which they are subjected. All that we can do is
to make assumptions, which may or may not repre-
sent an approximation to the actual conditions;
then to work out the consequences of these assump-
tions and see whether they are in agreement or in
conflict with the results of observation.
Suppose we make the assumptions that the star is
gaseous throughout and that the material every-
where has the properties, which we can study in the
laboratory, of what is termed a " perfect gas." It
is evident that the density, pressure and tempera-
ture will all increase continually from the surface of
the star inwards to the centre. Calculation shows
that, under these assumptions, the temperature
within a star of the same size and of the same weight
as the Sun must exceed one million degrees through-
out the greater part of the interior and that at the
centre it must be many millions of degrees.
This high internal temperature has two important
consequences. In the first place it implies that we
do not need to concern ourselves much about the
composition of the material in the interior of the
star. We have seen that by raising the temperature
sufficiently we can split off one or more of the outer
'^satellite electrons from the atoms. This process is
icalled ionization. In the range of temperatures
<a few thousand degrees only available in the
^laboratory, we can merely knock off a few of the
AGE AND EVOLUTION OF THE STARS 217
outer electrons from some of the atoms. But when
temperatures of millions of degrees are involved the
ionization becomes enormously intensified. Prac-
tically all of the electrons are stripped from the
atoms and we have a mixture of atomic nuclei and
electrons. What we are really interested in, so far
as the composition of the material is concerned, is
the average weight of each particle. Suppose a star
was composed entirely of oxygen; the atomic weight
of oxygen is 1 6 and each atom is broken up into the
nucleus and 8 electrons, making 9 particles in all,
whose average weight is 16 divided by 9, i.e. 1-78.
If it was composed entirely of iron, whose atomic
weight is 56, each atom is broken up into the nucleus
and 26 electrons and the average weight per particle
is 2 "07; if composed of gold, atomic weight 197, the
average weight per particle becomes 2-46. Thus
whatever the material of which the star is composed,
with a single exception, we can assume an average
weight per particle of 2, with the confidence that
we shall not be much in error. The exception is
hydrogen. The atomic weight of hydrogen is i,
and the hydrogen atom contains only one proton
and one electron, so that for a star composed of
hydrogen the average weight per particle is only J.
We can therefore anticipate that the properties of a
star will depend appreciably on the proportion of
hydrogen which the star contains. It is sufficiently
accurate to consider any star as composed of hydro-
gen, with weight per particle of |, and not-hydro-
gen, with weight per particle of approximately 2.
This is a great simplification for the theoretical in-
vestigation of the interior of a star.
2l8 WORLDS WITHOUT END
The second consequence of the high internal tem-
perature is that it is not possible to neglect the pres-
sure exerted by the radiation generated within the
star as it endeavours to escape to the surface. It
has long been known that any form of radiation,
;uch as light, exerts a pressure upon any surface on
which it falls. We have already had direct evidence
Df this pressure in the tail of a comet. The pressure
exerted by the light from any terrestrial source is
extremely small, though it has been possible to de-
tect and to measure the pressure in the laboratory
by delicate experiments. The pressure increases
rapidly as the temperature of the source of radiation
is increased ; if the temperature is doubled the pres-
sure increases sixteen-fold. At the extremely high
temperatures within a star the pressure of radiation
becomes very large. Thus, whereas if the tempera-
ture were 5,000 C. the pressure of the radiation
would amount only to about one-twentieth of an
ounce per square foot, at the temperature of 20
million degrees it is about 3 million tons per square
inch.
We can now form a general idea of the state of
affairs prevailing inside the star. The star is held
together by its gravitation; were it not for the force
of gravitation, the matter of which the star is com-
posed would rapidly diffuse outwards and be scat-
tered through space. Within the star the nuclei and
electrons into which the atoms are split up are flying
about in all directions, with velocities which are ex-
tremely great because of the high temperature.
The electrons are speeding around with a velocity
of about 10,000 miles a second; the heavier nuclei
AGE AND EVOLUTION OF THE STARS
of the atoms move more slowly, but with speeds up
to several hundreds of miles a second. Atoms and
electrons are continually colliding with one another.
These numerous collisions are collectively equiva-
lent to a pressure which is tending to disperse the
material of the star outwards in all directions.
Through this medley of hurrying atomic nuclei and
electrons the radiation is attempting to escape. At
the centre of the star, the radiation is of extremely
short wave-length, comparable with the wave-
length of X-rays. The free passage of the radiation
is impeded by the atoms. The radiation is con-
tinually being captured by the atoms that it en-
counters in its path; the atoms retain the captured
radiation for a short time and then re-emit it, gener-
ally in an entirely different direction. The radia-
tion thus gradually buffets its way towards the sur-
face of the star. The net effect of the absorption of
radiation by the atoms and of its re-emission is
equivalent to the pressure of radiation to which we
have referred above, tending to drive the atoms
outwards. We may compare the escaping radiation
to a wind blowing outwards in all directions. The
continual absorption and emission of radiation by
the atoms gradually lengthens the wave-length of
the radiation until, when it has reached the surface,
it consists of the mixture of ultra-violet, visual and
ijifra-red radiation with which we are familiar.
The mathematical investigation of these processes
is one of great difficulty and has not been com-
pletely solved. There is the additional complica-
tion that we know little about the source whence the
energy that the star continues to radiate is de-
22O WORLDS WITHOUT END
rived: whether, for instance, the generation of
energy occurs entirely or mainly in the hottest cen-
tral region of the star, or whether it occurs more or
less uniformly throughout the star. The method
that has been used by Eddington, Milne and
others is to assume that the star is built according to
some particular model and to work out the con-
sequences. Some of the results do not depend to a
great extent upon what particular model is adopted,
and we can therefore regard them as reasonably
accurate.
It can be inferred that for a star of the size and
mass of the Sun the temperature at the centre is
about 20 million degrees. For the Sun and for stars
of smaller mass, the effect of the pressure of radiation
is relatively small compared with the gas pressure,
i.e. the pressure arising from the motion of the
atoms. But for stars of larger mass, the effect of the
pressure of radiation becomes increasingly impor-
tant and for such stars the central temperature is not
so high. For the giant stars of large mass and large
in size, the temperature at the centre is only one
or two million degrees.
If the size of the star and the amount of matter in
it are known, can we calculate its candle-power ?
We find that this is not possible without a knowledge
of the average composition of the star. But we have
seen that all the elements except hydrogen can be
lumped together, being practically equivalent to
one another when fully ionized. What we really
require to know before we can calculate the candle-
power of the star is therefore the percentage of
hydrogen in the star. The candle-power is very
AGE AND EVOLUTION OF THE STARS 221
sensitive to this percentage; by varying the per-
centage of hydrogen in a star of the size and mass
of the Sun, the candle-power can be varied in a
range of 600 to i . For a star of given weight and
size, the candle-power is lowest when the star con-
tains 85 per cent, by weight of hydrogen. If the
percentage is greater, the candle-power is higher
because the material of the star is less effective in
damming back the escaping radiation; if the per-
centage is lower, there is more obstruction to the
escape of the radiation, but the central temperature
becomes so much higher that there is a greater net
out-flow of radiation and consequently a higher
candle-power.
We have no means of directly determining the
percentage of hydrogen in the star. But we can
proceed in the opposite direction and infer how much
hydrogen the star contains by comparing the ob-
served candle-power with the candle-powers com-
puted by assuming several different percentages of
hydrogen.
The observed and calculated candle-powers of
the Sun can be brought into agreement for two
values of the percentage content of hydrogen, one
of which is lower than 85 per cent, and the other is
higher. The two percentage contents which give
agreement are 33 and 99 -5. It is highly improbable
that the Sun is nearly all hydrogen, as the latter
figure would suggest, and we may therefore con-
clude that the Sun is about one-third part by weight
hydrogen and two-thirds other substances.
It would seem that the majority of other stars
must also contain about one-third hydrogen. With
222 WORLDS WITHOUT END
this assumption we can calculate the candle-powers
of stars of different weights. The candle-power for
a given percentage content of hydrogen depends
mainly upon the amount of matter which the star
contains and only to a small extent upon its size.
The computed relationship between candle-power
and weight is closely satisfied by the actual data of
observation. There must therefore be some cause
which makes the hydrogen content practically the
same for all stars, with few exceptions. What this
cause may be is not known.
The massive stars have high candle-power; the
stars of small mass have low candle-power.
The theoretical investigations are based upon the
assumption that the material throughout the star is
compressible like a perfect gas. This assumption is
certainly reasonable for giant stars of low average
density, such as Antares or Betelgeuse. But is it a
reasonable assumption to make for the Sun, whose
mean density is nearly i| times the density of water ?
We know that water is practically incompressible.
The molecules of a liquid are so close together that,
however great the pressure we apply, we cannot
squeeze them much closer together. With air or
any other gas it is otherwise. The average distance
apart of the molecules is much larger than the size
of the molecules. Air is mainly empty space and
by applying pressure we can force the molecules
closer together.
When Sir Arthur Eddington first carried out these
calculations, he anticipated that the observed and
computed candle-powers would agree for the diffuse
giant stars, but that there would be an increasing
AGE AND EVOLUTION OF THE STARS 223
divergence for stars of larger and larger mean dens-
ity. His expectations were not realised; the agree-
ment in the case of the giant stars was satisfactory,
but, surprisingly enough, it was equally satisfactory
for the dense stars. The only possible explanation
is that these stars are also compressible like a perfect
gas. How can matter much denser than water be
as compressible as a gas ?
The explanation, given by Sir Arthur Eddington,
was simple and when given was obvious. Just
as air in a room is mainly empty space, so also each
molecule of the air is also mainly empty space.
Sir Oliver Lodge compared an atom to Westminster
Abbey, with a few gnats flying about in it. The
Abbey represents the size of the atom, the gnats the
electrons. Now conceive the walls of the Abbey
removed, the gnats still flying about in the space
which they had enclosed. We then have a fair pic-
ture of the emptiness of the atom, or of the molecule.
Normally we can only compress a substance until
the imaginary walls of the molecules are in contact
with one another. Though we still have mainly
emptiness, further compression is not possible.
Liquids and solids are in this condition.
But in the star the imaginary walls which sur-
rounded the atom have been demolished. The
electrons (gnats) are no longer imprisoned within
them but are free to mingle with the electrons of
other atoms. There is now plenty of scope for fur-
ther compression before the gnats begin to get
jammed together. This will only occur when the
density is of the order of 100,000 times that of water.
It is therefore apparent that the matter in the Sun
224 WORLDS WITHOUT END
and other similar stars is very far from the stage of
maximum compression, and it is to be expected that
the material will behave like a perfect gas, as indeed
it does.
There are a few stars known whose candle-power
is much lower than we should expect from the
amount of matter which they contain. These are
the stars which the astronomer has called the white
dwarfs; the name was given because in general the
white stars are giants. The small companion of
Sirius is the typical white dwarf. We have seen
that these stars have extremely high densities, and
we can infer that the reason why their luminosities
are low is that they have arrived at the stage at which
the electrons and nuclei are jammed so closely to-
gether that the stellar material can no longer be
considered as having the compressibility of a perfect
gas.
" Dense matter," as matter at these extremely
high densities has been called, has properties which
are very different from the properties of normal
matter. These properties have been investigated
theoretically by Professor R. H. Fowler, with the
aid of wave-mechanics. It is found from these
theoretical considerations that compression cannot
be continued until all the atoms are jammed closely
together; there is a limit to which the compression
can be carried. There comes a stage at which not
more than a certain number of slowly moving par-
ticles can be crowded into a given volume. When
this stage is reached, the only particles which can be
squeezed in are the more rapidly moving particles.
As the pressure increases the average energy in a
AGE AND EVOLUTION OF THE STARS 225
given volume therefore increases. But this energy
is not available to be radiated away. It is only the
excess of the total energy of motion of all the par-
ticles above the amount which is, as it were, tied up
by this crowding together which is available for
radiation. The rest is, as Sir Arthur Eddington has
expressed it, kept on deposit. The final limit is
reached when the whole of the energy is tied up on
deposit by this crowding together and none at all
can be spent. Thus, as the density increases, the
energy available for radiation rapidly decreases and
the candle-power of the star becomes progressively
smaller. In the final stage the total energy is still
great, but radiation of energy ceases and the tem-
perature may therefore be said to be zero. When
this stage is reached the star ceases to be visible and
may then be termed a " black dwarf." The white
dwarf stars are in a position intermediate between
normal stars and black dwarf stars. We know only
a small number of white dwarf stars ; their luminosity
is so low that it is only those which are in the neigh-
bourhood of the Sun that we can discover. But it
is probable that they are very abundant in space.
Black dwarfs may also be abundant, but because
they are invisible we have no hope of finding out
anything about them.
In these investigations dealing with the interior
of a star no assumptions have been made as to
the origin of the radiation which the stars emit.
We have mentioned (p. 129) that Lord Kelvin
supposed that the Sun maintained its radiation by
a gradual shrinkage and that the maximum life-
time possible on this hypothesis is about 25 million
15
226 WORLDS WITHOUT END
years. This estimate was quite inadequate for
geologists. The study of the geological changes
which the Earth has undergone forced them to the
conclusion that the Earth must have existed for a
much longer time than 25 million years. The
phenomenon of radioactivity was not known in
Lord Kelvin's day and the accurate method of de-
termining the ages of certain rocks which radio-
activity provides, as explained on p. 18, was not
available. We have seen that the Earth must have
existed for more than one thousand million years,
and the age of the Sun cannot be less than that of the
Earth. None of the sources of energy with which
we are familiar in everyday life can account for an
age of such length. We are forced to the conclusion
that the Sun and the other stars can tap the enor-
mous store of energy which is locked up in the atom.
The energy which is thus locked up is surprisingly
large. If we could release it, we should from one
ounce of coal obtain sufficient energy to run engines
of a total horse-power of 100,000 for one year. In
other words, the fuel requirements of a large generat-
ing station could be met by one ounce of coal per
year. The Queen Mary could be driven across the
Atlantic with the energy from a fragment of coal
no larger than a pea.
According to modern physical conceptions, mass
and energy are synonymous terms. The Sun is
radiating energy into space at the rate of 62 horse-
power from each square inch of its surface. The
total energy which it emits is equivalent to a loss of
weight of 4 million tons every second. Large as
this loss is, the mass of the Sun is so great
AGE AND EVOLUTION OF THE STARS 227
i
(25000500050003000,000500050005000,000 tons) that,
if the radiation were maintained at its present rate,
the Sun would last for about 1 6 million million years.
The equivalence of matter and energy can per-
haps be better realised if we think of matter as com-
posed of protons, carrying a positive electric charge,
and electrons, carrying a negative charge. It is
conceivable that if we could bring a proton and an
electron together so that their charges coalesced and
neutralised one another, both particles would dis-
appear and a pulse of energy would remain. We
have no evidence that this process does happen, nor
even that it can happen in Nature. But if it can
happen, the longest life possible for any star will be
obtained on the assumption that the whole of the
material of the star is ultimately transformed into
energy.
The radiation does not take place at a uniform
rate. The greater the mass of the star, the greater
its candle-power or, in other words, the greater the
rate at which it is dissipating its energy. The stars
of large mass are very profligate of their resources ;
the star S Doradus, for instance, is losing more than
one million million tons in weight every second.
But as their capital gets smaller, they become less
profligate and gradually reduce their rate of spend-
ing. Suppose a star to be formed with a very large
initial mass ; at first the mass will decrease at a rapid
rate, but the rate will progressively slow down so
that by the time the mass has become small the de-
crease has become very slow. The reason why
there are few stars with masses exceeding twenty
times the mass of the Sun is probably that the mas-
228 WORLDS WITHOUT END
sive stars lose weight so rapidly that they soon cease
to be massive stars.
However large the mass of the Sun may have been
initially, it will have decreased to its present value
in not more than 5 million million years, if it can be
assumed that the Sun derives its energy from the
actual annihilation of matter. But we have no
proof of the validity of this assumption and it is
doubtful whether it is correct. Theoretical con-
siderations suggest that the annihilation of matter
cannot occur at temperatures lower than some thou-
sands of millions of degrees. The highest tempera-
ture in the interior of a star is about 20 million
degrees. Such temperatures seem hopelessly in-
adequate for annihilation to take place.
There is an alternative process by which not all
but a small fraction of the energy contained in mat-
ter can be liberated. This process is the building
up of heavier atoms out of atoms of hydrogen. A
helium atom, for instance, contains four protons and
four electrons ; the four protons and two of the elec-
trons are bound together to form the nucleus, the
two remaining electrons describing orbits around it.
Since a hydrogen atom contains one proton and one
electron, we can imagine a helium atom to be built
up from four hydrogen atoms. But the weight of
the helium atom is less than the weight of the four
hydrogen atoms by about one part in 140. The
weight which disappears when one helium atom is
formed from the four hydrogen atoms is accounted
for by the energy which is released in the process.
We can suppose also that the elements heavier
than helium are built up in a similar way from
AGE AND EVOLUTION OF THE STARS 22Q
hydrogen atoms. The energy set free is then rather
greater, though not appreciably greater, than in the
building up of helium. The process of building up
heavy atoms from lighter atoms such as helium may
also be taking place; but the energy which can be
obtained in this way is comparatively small. The
mass of the oxygen atom, for instance, is only
slightly greater than the combined mass of four
helium atoms. We may say that approximately
i per cent, of the mass of the star can disappear in
the form of radiated energy, if heavier atoms are
built up from hydrogen atoms, and if the star con-
sisted initially entirely of hydrogen.
There are thus two conceivable alternatives: the
complete annihilation of mass or its partial annihila-
tion by the building up of heavy atoms. On either
assumption, we may say that a star is able to continue
to radiate light and heat by a process of self-cannibal-
ism, slowly eating itself up. But the first assumption
supposes that it can eat itself up completely; the
second assumption requires that when only one-
hundredth of the meal has been completed the star
can go no further ; it can no longer radiate, it is dead.
If we think of protons and electrons as the funda-
mental units of which matter is composed, we must
suppose that the process of the building up of heavy
atoms does occur. For these atoms exist and they
must have been formed in some way. The interior
of a star, where so far as we can tell temperatures
higher than anywhere else in the Universe are to be
found, would seem to be as favourable a place as any
other that we can think of for this process to occur.
Recently, the transmutation of elements has been
23O WORLDS WITHOUT END
achieved in the laboratory. When atoms are bom-
barded by fast-moving particles protons, electrons,
or helium nuclei (called Alpha particles by the
physicist) the nuclei of some of the atoms are hit
by the bombarding particles. The proton or Alpha
particle may be captured by the nucleus and a new
element is formed. Similar processes may occur in-
side a star, where the particles have very high
velocities. The laboratory experiments make it
probable that, at the temperatures which prevail
in the interior of a star, the transmutation will pro-
ceed sufficiently fast to supply the energy required
to maintain the star's output of radiation.
The possible length of life of a star is many hun-
dreds of times longer, if annihilation of matter takes
place, than it is if it does not. In the latter case the
supply of fuel available for consumption is only
about the one-hundredth part of the supply avail-
able in the former case. The star then has prac-
tically the same weight throughout its life; when
i per cent, of the initial mass has been radiated
away, it dies. But if annihilation of matter occurs,
it will continue to radiate away its mass at a pro-
gressively slower and slower rate. The following
table gives the duration of time required for suc-
cessive changes of mass.
Mass (Sun = i).
very large to 35
35
10
37
0-92
0-31
10
37
0-92
0-31
0-18
Duration
(in millions of years).
38,000
65,000
214,000
930,000
5,210,000
36,300,000
28l,000,OOO
2,I90,OOO,OOO
AGE AND EVOLUTION OF THE STARS 23!
Thus, for instance, if the initial mass of a star was
0-92 of the mass of the Sun, about 36 million mil-
lion years would be required for the mass to de-
crease to 0-53 of the mass of the Sun.
We have seen that the stars of high intrinsic
brightness are the very massive stars and that there
is a progressive decrease in intrinsic brightness with
mass. If annihilation of matter occurs, we can in-
terpret this as evidence of evolution; as the star
grows older it loses weight and at the same time
becomes gradually less luminous. But if annihila-
tion does not occur, there can be no evolutionary
significance in this relationship. A massive star is
always a massive star; a star of small mass can never
have had a large mass. It therefore makes a big
difference to our ideas of the evolution of the stars
whether or not we can suppose that annihilation of
matter occurs inside the stars.
If annihilation does not occur, the possible age of
the stars is limited to a few thousand million years.
This time is comparable with the age of the Earth
and would imply that the Earth is about as old as
the stars. But if annihilation of matter does take
place, the age of the stars can be extended to several
million million years. We naturally look for evid-
ence in favour of one or other of these ages. We
find that the evidence is inconclusive ; some evidence
seems to point to the shorter age, other evidence to
the longer age. One of the great unsolved problems
of astronomy to-day is the decision between these
two widely differing ages.
We shall consider briefly a few pieces of evidence.
In our galactic system there are many thousands of
232 WORLDS WITHOUT END
millions of stars. Some are heavy-weight stars,
others are light-weight stars ; some are moving fast,
others slowly. Every star is attracting every other
star in the system according to the law of gravita-
tion and thereby tending to alter the speed with
which it is moving. But the average distance of one
star from another is so great that the attraction be-
tween two stars is in general extremely small. It
might be thought that the velocity of any star would
not appreciably be changed by the gravitation pulls
of the other stars. On the other hand, it must be re-
membered that this interaction is taking place con-
tinuously, and the cumulative result over many
millions of years may be far from negligible.
It is not possible to trace out the effect on any
particular star, for the positions of all the stars are
continually changing. But the general effect of the
interactions of the stars one with another can be
studied statistically by mathematics. It appears
that the result of the mutual gravitational pulls is
on the average to slow down the stars which have
more than the average energy and to speed up those
which have less than the average energy. If the
interaction continues for a sufficient length of time,
the average energy of the heavy-weight stars will
become equal to the average energy of the medium-
height stars and to the average energy of the light-
height stars. We cannot and do not need to in-
vestigate what happens to any individual star. We
group the stars, therefore, according to their weight,
determine the average speeds of the stars in each
group and compute the average energy for each
weight-group. When we do this, we find that not
AGE AND EVOLUTION OF THE STARS 233
only are the light-weight stars moving considerably
faster on the average than the heavy-weight stars,
but also that the velocities differ by just the amount
necessary to give approximately the same mean
energy for each group.
Given the actual weights, speeds and average dis-
tance apart of the stars, it is merely a matter of
mathematics to find how long the mutual interac-
tions must have continued in order that this equal
sharing out of energy will have resulted. The cal-
culation gives a time of about 10 million million
years. It would seem, therefore, that the stars must
have existed for at least this length of time.
A second argument which points to the same con-
clusion is based on clusters of stars which have a
common motion. The stars of the Great Bear, for
instance, are moving together with the same speed
through space; many other stars scattered over the
sky, including Sirius, the brightest star, share this
motion. Several other groups of stars that are
moving together are known. As these groups of
stars sweep onwards through the general system of
the stars, the effect of the gravitational pull of all
the surrounding stars is gradually to change both the
speed and direction of motion of every star in the
cluster. The effect will be most pronounced for
the light-weight stars. After a sufficient time, there-
fore, the motion of the light-weight members of the
cluster will have been changed to such an extent
that we shall no longer be able to recognise them as
members of the cluster. After a still longer time,
the cluster will have lost the medium-weight stars
and finally, after a further interval of time, the
234 WORLDS WITHOUT END
motions of the heavy-weight stars will have been so
distorted that the group of stars can no longer be
identified as a cluster of stars with a common mo-
tion. We actually find that these clusters have lost
their light-weight stars and most of their medium-
weight stars, and the mathematical discussion of the
problem indicates that several millions of millions of
years must have elapsed for this to have come about.
A third argument of a more speculative nature
supports the two preceding arguments. We have
seen that the island universes scattered throughout
space are of about the same size and the same weight.
We can suppose that these have all gradually con-
densed from a primitive gas, scattered more or less
uniformly throughout space. The time required for
this to happen can be estimated and again indicates
the long time-scale for the lifetime of the stars.
The principal argument against so long a life is
based upon the observed recession of the distant
Universes. We have seen that all distances become
doubled in about 1,300 million years. If we work
backwards in time, assuming the expansion to have
been taking place uniformly in the past, we find that
several thousand million years ago the Universes
were collected more or less together. We can sup-
pose that something in the nature of an explosion
then took place and the fragments began to scatter
outwards in all directions. But what preceded this
explosive outburst we do not know, nor whether the
stars and the galaxies existed as such. The Universe
may have existed for an indefinitely long time in
this sort of embryo state; but, if so, did the stars and
galaxies exist ? If they did, were the galaxies more
AGE AND EVOLUTION OF THE STARS 235
compressed than they are now ? May not the ex-
pansion of the Cosmos have been accompanied also
by an expansion of individual galaxies ?
These are questions that we cannot answer, to
which perhaps no answer will ever be forthcoming.
That it is necessary to ask them makes us feel some-
what insecure about our previous arguments in
favour of an age for the stars much greater than the
few thousand million years since the galaxies were col-
lected together. The stars were then possibly much
closer together than they are now. Their mutual
gravitation pulls would in that case have been much
larger and the equal sharing out of energy amongst
the stars would have been more rapid; it may have
been practically complete by the time the dispersal
of the system took place. The same considerations
apply to the moving clusters. We do not know how
or when these came into being. But we must suppose
that they existed when the separate Universes of
to-day were collected together; they may have been
practically in their present state and have lost their
light-weight stars before the initial explosion. The
argument based upon the condensation of galaxies
from a primitive gas scattered through space would
seem also to fall to the ground.
A possible means of escape from these difficulties
is perhaps to be found in the suggestion which has
been made that the recession of the galaxies is not
permanent but that the Cosmos may be in a state of
pulsation, alternately expanding and contracting.
It would then be just a matter of chance that we
happen to exist at a time when expansion is taking
place.
236 WORLDS WITHOUT END
The pulsating Universe is mathematically possible
according to the theory of relativity. There are
great difficulties in making it work in detail, but
these are perhaps not insuperable. It would appear
to provide a loophole by which we can fit in the
long time-scale, which the stars seem to demand,
with the present observed rapid expansion of the
Universe. On the other hand, our arguments in
favour of the long time-scale have implicitly assumed
that the interactions of the stars have taken place
throughout their past history at the same rate as at
the present time. We cannot now feel at all cer-
tain that this assumption is justifiable.
Thus, whereas a few years ago astronomers had
a theory of stellar evolution that seemed to be self-
consistent, with an age for the stars of several mil-
lions of millions of years, everything is now in the
melting-pot. It frequently happens in scientific in-
vestigation that a new discovery cannot be fitted in
with accepted theories; the theories have then to be
recast or, if necessary, abandoned and a new theory
formulated that will embrace the new discovery.
That is how progress in science is made. It is the
discovery of the recession of the galaxies and the
expansion of space that has called for some re-
vision of the ideas of stellar evolution recently cur-
rent. But at present we are not certain what we
must give up and what can be retained.
We can outline two tentative theories of stellar
evolution according to whether we do or do not
assume that annihilation of matter is occurring
within the stars. If we assume this, we suppose the
star to start as a tenuous condensation in a mass of
AGE AND EVOLUTION OF THE STARS 237
nebulosity. The star will be large in size and of
low mean density, somewhat like the giant red stars
such as Antares. At first the diffuse gaseous mass
will contract rapidly under the action of gravita-
tion. The contraction will release gravitational
energy, causing an increase in the central tempera-
ture ; at the same time energy may also be released
by the synthesis of more complex elements from
hydrogen. The output of radiation from the star
will remain practically steady, but the surface tem-
perature will rise ; the increased radiation from each
square foot of the surface, due to the rising tempera-
ture, will approximately counterbalance the de-
crease in surface area due to the contraction. The
star will pass rapidly from a red giant star to a blue
giant star. During this stage there is very little de-
crease in the mass of the star. When the central
temperature has risen to about 20 million degrees
we must suppose that the annihilation of matter
commences. There is no further contraction and
the whole of the energy which the star now radiates
is at the expense of its mass. The star now slowly
radiates away its mass; the decrease in mass is ac-
companied by decrease in luminosity. The central
temperature remains constant, but the surface tem-
perature slowly decreases, and the star passes
through the successive stages from a giant blue star
to a dwarf yellow star and then a dwarf red star.
The complete time for the transition from a giant
red star to a giant blue star and then to a dwarf red
star may occupy several hundred millions of mil-
lions of years.
If we do not accept annihilation of matter as pos-
238 WORLDS WITHOUT END
sible, we must draw a different picture. The syn-
thesis of heavier elements from hydrogen will now
provide the main source of energy within the star.
We suppose, as before, that in the early stages of
the evolution there is rapid contraction of the large
diffuse gaseous mass. The central temperature
rises rapidly until it becomes sufficiently high for the
synthesis of heavier elements from hydrogen to be-
come possible. This process then provides the
main source of energy. The star continues to draw
on this supply of energy, the mass and also the
luminosity remaining approximately steady. When
this source of energy begins to fail, contraction must
again set in and the mean density will rapidly in-
crease. At length the material at the centre of the
star will become so dense that the gas laws are no
longer obeyed. The star will gradually pass into
the white dwarf stage with rapid decrease both in
size and in output of radiation. In this stage of the
evolution instability may set in; the star may sud-
denly collapse, with the rapid release of a large
amount of gravitational energy. The star has be-
come a nova. The nova stage is succeeded by the
white dwarf stage, which is succeeded by the black
dwarf stage. The complete time for this sequence
of changes to occur will be a few thousand million
years.
At the present time, the balance of evidence would
seem to be in favour of the second picture, but a
definite decision between the two will probably
have to await some further knowledge as to the pro-
cesses by which energy is generated in the interiors
of the stars.
CHAPTER XIII
WHAT WAS WHAT IS TO BE
IN the preceding chapters we have been concerned
mainly with the results of observations of various
sorts and with the conclusions which can be drawn
from them. We have been building on fairly sure
foundations. Some of the conclusions are ad-
mittedly tentative because our information is incom-
plete. We are still in ignorance, for instance, of
the source of energy within the stars and whether the
energy is generated in a limited region near the
centre of a star or more or less uniformly throughout
the star as a whole. In this chapter our foundations
are much less secure; we shall attempt briefly to en-
visage what was the past of the Universe and what
will be its destiny. We can expect little help in
such matters from observation; we must rely instead
mainly on inference. It must therefore be em-
phasised that we are entering the uncertain regions
of speculation.
We have found that in the Universe as a whole,
matter is largely aggregated into vast discoidal
systems or galaxies. Each of these systems has a
weight many millions of times the weight of the
Sun; each is in slow rotation. In any single system
we find both diffuse clouds of gas and stars ; but
whereas some systems appear to be almost entirely
clouds of gas, others appear to consist largely of
stars. There is great variety amongst the stars ex-
cept in one respect their weight. With few excep-
239
24O WORLDS WITHOUT END
tions, the weight of any star is not greatly different
from the weight of the Sun. It would seem that
these uniformities must have been due to the opera-
tion of similar processes throughout the Universe.
If, at the present time, the whole of the matter in
the Universe was spread out uniformly, space would
appear extremely empty. One cubic inch of ordinary
air would occupy a vast volume of about 6 million
million cubic miles. As a result of the recession
of the galaxies, the average density of matter
in space is becoming steadily less. A few thousand
million years ago the average density may have
been about one thousand times greater than at the
present time; even so, the average density was
extremely low, judged by any terrestrial standards.
We may suppose that there was a time when the
matter in the Universe was distributed uniformly
throughout space as a diffuse gas of extremely low
density. We do not know and we cannot prove
that this was so; but the supposition provides a
model and we can examine the consequences.
If the density had been absolutely uniform the
system might have existed in this condition in-
definitely. But if the density was slightly greater
in some parts than in others, there would be a
tendency for condensations of matter to grow in the
regions where the density was above the average.
Any slight disturbance which upset the supposed
initial uniform density would give the force of gravi-
tation the opportunity to get to work ; provided the
disturbance was sufficiently large, gravitation would
overcome the tendency of the atoms to diffuse from
the regions of greater density to the regions of lesser
WHAT WAS WHAT IS TO BE 24!
density and thus to restore the conditions of uniform
density. Any condensation which was of sufficient
size to begin with would gradually grow; the small
ones would be smoothed out again. Corresponding
to any given initial density, there is a definite mini-
mum weight necessary in order that a condensation
may grow. The lower the initial density, the
greater the weight that the condensation must have
in order that gravitation can overcome the natural
tendency of the condensation to disperse.
We can therefore make an estimate of the mini-
mum weight of any condensation that could have
formed out of the uniformly distributed gas which
we have conjectured. The estimate cannot be very
precise, because we do not know what the composi-
tion of the gas was ; the average speed of the atoms
depends upon this composition and the minimum
weight of a condensation depends in turn upon this
average speed. We are able, nevertheless, to con-
clude that the minimum weight must have been
many millions of times the weight of the Sun. The
condensations could not have been stars, but must
have been systems comparable, in the amount of
matter which they contain, with the spiral nebulae. ,
Each condensation as it formed would be likely
to have a certain amount of spin, for it would be
difficult to initiate any disturbance in our hypo-
thetical primaeval gas without causing differences of
motion between different parts that would intro-
duce rotations. As each nebulous condensation
gradually contracted, its rate of spin would increase,
in accordance with a well-known dynamical law,
so as to keep the angular momentum constant. As
16
242 WORLDS WITHOUT END
the rate of spin gradually increased, the system
would first become slightly flattened, somewhat like
Jupiter; a bulge would next form round the equator
and become more pronounced until the system
assumed a lenticular shape ; after this, with still in-
creasing spin there would be no further flattening
but matter would be ejected through the sharp
equatorial edge into the equatorial plane. If the
nebula was absolutely symmetrical, the ejection of
matter would occur all round the equatorial edge.
But any slight disturbance, such as the gravitational
attraction of an adjacent system, would be sufficient
to introduce a bias and to cause the matter to be
ejected from two diametrically opposite points.
Although we cannot prove that the spiral nebulae
have been formed in the way we have conjectured,
we do in fact find systems of all the shapes which
according to theory a rotating contracting mass of
gas would pass through in succession spherical,
oblate, lenticular and systems in which matter
appears to be streaming from opposite ends of a
diameter.
The nebula formed in this way is initially of very
low density. As it contracts the mean density in-
creases considerably. The average density of the
material in the spiral nebulae as we now see them
is such that one cubic inch of ordinary air would
occupy a volume of several thousand cubic miles.
* This density is many times greater than the density
of the primaeval gas ; it follows that the condensations
which can form out of it are much smaller in mass.
The greater density gives gravitation a much better
chance. We find that in such a nebula gravitation
WHAT WAS WHAT IS TO BE 243
can hold together a condensation comparable in
weight with the Sun. It seems reasonable to sup-
pose that the stars have in fact been born in this
way by condensing out of the nebulous clouds in
the separate universes. This may have occurred
in two stages, for the photographs of nebulae such
as that in the Great Bear (Plate XXXII) show a
tendency for the stars to be aggregated into large
clusters. We have seen that the Sun is a member
of a localised cluster in our galactic system. It is
probable that as a nebula contracts it breaks up
first into a number of relatively large condensations
and that each of these in turn breaks up into groups
of stars. This does not exclude the possibility of
smaller condensations forming directly in the
nebula and giving birth to individual stars.
In this way, starting with an initial distribution
of gaseous matter of extremely low density, we have
three successive stages first the formation of large
systems weighing many millions of times more than
the Sun, which we identify with the great extra-
galactic nebulae; then the formation of localised
clusterings within each nebula and finally the breaks
up of these clusterings into individual stars, com*
parable in weight to the Sun.
Many of the stars are twin systems, as we have
seen in Chapter IX. How have these been formed ?
We can suppose the same process to have continued
after the individual stars were born. The rotation
that has been generated cannot disappear : it must
persist. We can suppose that, as a general rule,
each star when it was born was rotating. We know
that our own Sun is rotating and we have direct
244 WORLDS WITHOUT END
evidence of the rotation of many other stars. Some
stars are rotating so rapidly that the rotational
speed at the equator is 200 or 300 miles a second.
As the star contracts after its birth, its rate of
rotation will increase; the star may assume a flat-
tened lenticular shape and at length eject matter
from the equatorial edge. But it can be shown
that this ejected matter would not condense into
nuclei such as smaller stars or planets. It would
either be gradually dissipated away into space or
it would form a nebulous atmosphere about the
stars. Many stars with nebulous atmospheres are
known and they may have been formed in this way.
But the evolution of the star may not necessarily
-take place as we have supposed ; it will only do so if
the star is much condensed towards the centre. If
this is not the case, evolution follows a different
course. The initial spherical mass of gas will be-
come flattened as it contracts and its spin increases ;
then, as the spin still further increases, it will become
considerably elongated and a neck will begin to
form in the middle, the mass concentrating towards
the two ends. The neck will become deeper and
deeper until at length the body will divide into two
separate portions, rotating about each other prac-
tically in contact. Many of the eclipsing twin
systems seem to be in this stage,
The evolution will not end here. After the
cleavage into two bodies has occurred, they will
both continue to contract and to spin at a faster rate.
Each star will raise large tidal protuberances on the
other. The period of rotation of each star will be
shorter than the period of mutual revolution of the
WHAT WAS WHAT IS TO BE 245
stars about each other. The tidal protuberance on
either star, as it moves around the star under the
gravitational attraction of the other star, will exert
a braking action, which will tend gradually to
equalise the periods of rotation of the two stars and
the period of mutual revolution. During this pro-
cess the stars will gradually separate and the period
of revolution will increase.
There are limits, however, to the extent to which
this evolution can proceed. It can be shown mathe-
matically that the separation of the two stars and the
period of revolution will only increase to a limited ex-
tent. The twin-stars of wide separation and of long
period certainly cannot have been formed in the way
we have described. We must attribute the existence
of such systems to an entirely different process. We
can conceive of two processes by which these systems
of two widely separated stars with a long period of
revolution may have been formed. In the first
place, it is possible that two adjacent nuclei in the
original nebula condensed out sufficiently close to
one another to be held prisoners, the one to the
other, by their mutual gravitational pull ; they would
have fallen together but for their relative motion,
and the result was that each revolved about their
mutual centre of gravity. We should expect that
the distance between two separate condensations
would be much greater than the distance between
the two components in a system formed by the cleav-
age of a single star.
In the second place, it may have happened that
two stars which had condensed from separate nuclei
:hanced, as they sped along, to pass each other so
246 WORLDS WITHOUT END
closely that they became bound to one another and
henceforth had to follow a common existence as a
twin system. The probability of this happening
would depend upon the average distance apart of
nuclei at the time they condensed out of the original
nebula.
We can thus form reasonable hypotheses to account
both for the systems which consist of two stars nearly
in contact revolving rapidly about each other, and
for the systems which consist of two stars widely
separated from each other and with long periods of
revolution.
The examination of the sequence of events which
may be expected on purely theoretical grounds to
follow from the evolution of a primordial diffuse
nebula has led to a plausible explanation of many
of the formations which we see in the heavens. But
it has not suggested as a possible by-product a system
in any way comparable to the solar system, consist-
ing of a Sun surrounded by a family of planets. It
is not possible in the limits of this book to attempt
to give an account of the various theories which
have been suggested to account for the existence of
the solar system. There are fatal objections to
every theory which supposes the Sun to have been
alone in space : for a plausible theory we must sup-
pose another body to have passed near to the Sun.
We shall consider therefore the sequence of events
which may be expected to have happened on the
hypothesis that another star passed near the Sun,
but not so closely that the two bodies became bound
to each other by their mutual gravitational pull.
As the star approached, its gravitational pull raised
WHAT WAS WHAT IS TO BE 247
a large tide upon the Sun, causing a protuberance
or hump to form, pointing in the direction towards
the oncoming star. As the star drew still nearer
this hump grew to such an extent that a long tongue
of matter was drawn out from the Sun. When the
star was at its nearest, the rate of ejection of matter
from the Sun was greatest; then, as the star passed
by and its distance from the Sun increased, the ejec-
tion gradually diminished. The tongue of matter
at length broke off from the tidal hump, which
slowly subsided as the disturbance passed away. A
cigar-shaped tongue of matter remained as evidence
of the passage of the stranger star. The ejected
matter rapidly cooled, and after a comparatively
short time liquefaction set in near the two ends.
A break-up into detached masses followed. The
smallest masses would form out of the densest matter,
and it is in accordance with expectation that we find
the largest planets, Jupiter and Saturn, in the middle
of the series and smaller planets on either side. The
smaller planets were probably liquid or solid from
birth; the two largest planets may have been
initially gaseous.
The tidal forces exerted on the planets by the Sun
the central mass gave rise in a somewhat similar
way to the further ejection of matter from the
planets and the formation of satellites.
In the way which we have briefly outlined we
can give a logical explanation of the formation of
the solar system and of its principal features. The
main argument against the theory is that the prob-
ability of two stars approaching sufficiently close
for a tongue of matter to be ejected is extremely
248 WORLDS WITHOUT END
small. We can calculate from the present average
distance apart of the stars in the neighbourhood of
the Sun how often this is likely to happen. It is
found that it is unlikely for it to occur more than
once in several million years. This would seem to
suggest that the solar system is something excep-
tional in the stellar universe and that in our galactic
system there are not likely to be more than a few
stars surrounded by systems of planets. Whereas
the Earth was once believed to be the centre of the
Universe, we now are apt to regard with suspicion
any theory which makes it appear as something ex-
ceptional or as occupying some privileged position
in the Universe. But we are forced to admit that
no more plausible theory has been suggested, and
improbability is not by itself a sufficient ground for
rejecting it. A possible avenue of escape from this
one objection may be provided by the expansion of
the Cosmos. We put the age of the Earth at a few
thousand million years. If we go back that distance
in time, we find the separate universes much closer
together than they are now The average distance
apart of the stars may also have been appreciably
less than at the present time. Though we cannot
be sure of this, it suggests that planetary systems may
be much more common than was formerly thought.
We have sketched out what seems to be the prob-
able course of evolution of the Cosmos, if we can
assume that it started from a uniform distribution
of matter in the form of an all-pervading gas of ex-
tremely low density. The questions may be asked:
What preceded this uniform distribution of matter ?
How did it come into existence ? Was a definite
WHAT WAS WHAT IS TO BE 249
act of creation involved ? I do not pretend to be
able to give any answer to these questions. To
assert that the Universe may have existed in this
initial state for countless ages is only shelving the
question or to adopt the picture suggested by Sir James
Jeans " of the finger of God agitating the ether " is
merely a confession of ignorance. Astronomy can-
not take us any farther back in time. I am writing
as an astronomer, not as a metaphysician or as a
theologian, and I prefer therefore to leave these
questions unanswered.
We have looked backwards in time as far as we
are able. Looking forwards, we ask the questions:
Whither is the Universe bound ? What will be its
end ? Before attempting to answer these questions
about the Universe, we shall start at home and con-
sider what fate there may be in store for the Earth
and for Mankind. We can picture two possible
ends, depending upon what we consider to be the
normal course of stellar evolution. These two pos-
Ssibilities are a slow death by cold or a quick death
by heat. The Sun is consuming its substance at the
rate of four million tons every second; as its reserves
of fuel get less and less the Sun will gradually get
cooler, its output of light and heat will slowly de-
crease. The conditions for life on the Earth are
directly dependent upon the radiation which the
Earth receives from the Sun. The Earth has no
store of heat of its own; as the Sun slowly cools down
the Earth will gradually get colder and colder.
How long it may be before the average temperature
has dropped so much that human life is no longer
possible depends upon how great is the store of fuel
25O WORLDS WITHOUT END
upon which the Sun has to draw: in other words,
upon whether its energy is derived from the annihi-
lation of matter or from the building up of complex
elements from hydrogen. If annihilation of matter
provides the main source of the radiation, then in
one million million years' time the Sun will only have
lost about 6 per cent, of its weight. The Earth
will be a little colder than it is now and life a little
less bearable. The face of the Earth will have
changed considerably; erosion will have levelled
the mountains with which we are now familiar,
many thousands of millions of tons of material will
have been deposited on the ocean floors and this ex-
tensive redistribution of the weight of the surface
materials may have upset the equilibrium of the
Earth's crust and brought about the uplift of new
mountain masses. The oceans will have largely
turned to ice, and a permanent and universal ice age
will be slowly but inevitably in course of develop-
ment. The changes will be so slow that human life
will doubtless adapt itself to the gradually changing
Conditions, and we can conceive of its possible exist-
ence for many millions of millions of years. But at
length the inevitable would happen; life must suc-
cumb to the gradually increasing grip of the cold as
the Sun gradually passes into its long-drawn-out
old age, with but a shadow of its former glory and
but a fraction of its present weight.
Our previous discussion has suggested that it is
more probable that annihilation of matter does not
take place and that the energy radiated by the Sun
is provided by the building up of complex elements
from hydrogen. As the Sun now contains about
WHAT WAS WHAT IS TO BE 25!
one-third part by weight of hydrogen, the maximum
possible loss of weight is limited to about one-third
of i per cent. For somewhere about 40 to 50
thousand million years the Sun could continue to
radiate at a rate not differing very greatly from its
present rate ; during this period the temperature of
the Earth would not greatly change. But after this
period most of the hydrogen will have been con-
verted into heavier elements; the supply of fuel
necessary to provide the energy for radiation will
be rapidly failing and the Sun will then begin to
shrink and its output of heat to decrease. There
will be a relatively rapid fall of temperature on the
Earth and life would probably very soon become
extinct. This would not happen for several thou-
sands of millions of years; but as the Earth is already
a few thousand million years old, it follows that, on
this hypothesis, the Earth has lived more than an
insignificant fraction of its life. If we think of the
whole lifetime of the Earth as changed in time-scale
so as to equal the normal span of human life, on our
previous assumption the Earth is still a new-born
infant an hour or so old, but on the second assump-
tion it is a child of a few years.
There remains another possibility that the Sun
may pass into the nova stage. We do not know
what is the cause of a nova outburst; but we have
seen that there are reasons for supposing that it is a
stage in stellar evolution which every star may have
to pass through. We know, moreover, that our Sun
has not yet passed through it. So far as we can tell,
when the Sun is ripe for this catastrophe we shall
have little, if any, preliminary warning. In the
252 WORLDS WITHOUT END
course of a few days or even of a few hours its output
of heat would increase to such an extent that all life
would become extinct ; the oceans would be turned
into vapour; trees, forests, cities and everything
combustible would be burnt; the Sun would rapidly
swell and might even consume and swallow up the
Earth. It is therefore possible that the Earth will
not survive to its old age but that it may be cut off
while still in the prime of life.
We have sketched out a scheme of evolution
starting with a uniformly distributed primaeval
nebula. The individual universes have been
formed by condensation from this nebula, and the
stars in their turn have been formed by condensa-
tion out of these universes. In the attempt to trace
out the course of evolution of individual stars we
have encountered the difficulty that we are uncer-
tain as to the source of the supply of energy which
is needed to enable a star to continue to radiate.
But whatever the source of energy may be, it is
inevitable that sooner or later the supply will be
exhausted and the star will cease to send out radia-
tion. The radiation is at the expense of the weight
of the star; each individual star continually loses
weight and the emitted radiation is accumulating in
space. The radiation travels on and on through
space unchanged, except when it encounters inter-
stellar matter, in the form of atoms, electrons or
small dust particles. The effect of these encounters
is gradually to increase the wave-length of the radia-
tion, so that ultimately it will all be transformed
into waves of very long wave-length, essentially
similar to extremely long radio waves.
WHAT WAS WHAT IS TO BE 253
It has been suggested that there may be a con-
verse process taking place in interstellar space and
that the radiation may there be re-formed into elec-
trons and protons ; in other words, that matter may
be re-created out of energy and an all-pervading
nebulosity thus produced anew, which could go
through the same process of evolution. But it is
not easy to understand how birth could be given to
matter in the way supposed. It runs counter to all
our experience, summed up in the famous " Second
Law of Thermodynamics."
The science of thermodynamics is concerned with
the various forms of energy and the changes which
they can undergo. It is based on two laws. The
first law states that the total energy in the Universe
is constant ; energy can neither be created nor de-
stroyed. This is the principle of conservation of
energy. When this law was formulated the equi-
valence of mass and energy was not realised and it
was also believed that matter could neither be
created nor destroyed expressed in the principle
of conservation of matter. The original form of the
law must be widened to include matter, so that we
now have the principle of the conservation of matter
and energy jointly.
The second law of thermodynamics is concerned
with the availability of energy and can be crudely
expressed in the form that energy continually be-
comes less and less available, or, in other words, that
there is a progressive degradation of energy. For
example, in any heat engine we must supply more
heat energy than we can recover from the engine in
the form of work; a proportion of the energy is lost
254 WORLDS WITHOUT END
for practical purposes in the condenser of the
engine. It is dissipated in the form of increased
random motion of the molecules, which we recog-
nise as heat. When heat flows from a hot body
to a cooler body there has been a net loss in the
availability of energy; so also when light-energy is
changed into heat-energy or radiation of short wave-
length into radiation of long wave-length. All
these processes change the form of the energy, in
that they degrade it or make it less available for co-
ordinated or organised purposes. Another manner
of expressing the law is that there is a progressive
decrease in the amount of organisation in the Uni-
verse. Energy becomes more and more disorgan-
ised.
The second law of thermodynamics is a law based
upon experience; we accept it because we have
never found it to be violated. It teaches us that
processes in nature tend to be uni-directional. We
cannot reverse them, except by the expenditure of
organised energy. Water will run downhill; the
energy which it gains by running downhill can be
utilised, as in hydro-electric power schemes. We
could use the energy so obtained to lift the water
uphill again but we should find that we should not
be able to lift as much water as had run down.
Some of the energy has been converted into forms
that are not available. Every process of any sort
that takes place increases the total amount of dis-
organisation.
Thus it seems that the end of the Universe must
come when all the energy has been degraded to such
an extent that it has reached the lowest possible
WHAT WAS WHAT IS TO BE 255
state of availability. Disorganisation is then at its
maximum. The total amount of energy in the
Universe is the same as it was initially, but all cap-
acity for change has been lost. There will then be
thermodynamic equilibrium throughout the Uni-
verse; this state has been described as the " heat
death " of the Universe.
There seems to be only one possible means of
escape from these conclusions. The second law of
thermodynamics is a law based upon experience in
a Universe which is expanding. From the scienti-
fic point of view the background conditions, relative
to which we must measure the degree of disorgani-
sation, are changing. It has been suggested that
this expansion may not be permanent; that the Uni-
verse may really be pulsating and that it is only by
chance that we happen to exist at a time when it is
expanding; and that in a contracting universe
organisation may increase and energy become
increasingly more available. A scheme of relativ-
istic thermodynamics has been formulated by Pro-
fessor Tolman which requires that this should be
the normal condition in a contracting universe ; the
re-formation of matter out of radiation would then
be possible. We can thus conceive the Universe to
pass through a series of successive stages, first of run-
ning down and then of being wound up again ; we
need not think of any beginning or of any ending
to such a universe.
From the strictly scientific view-point we can there-
fore no longer dogmatically assert that the heat-
death of the Universe at some finite time in the
future is necessarily required by the laws of thermo-
256 WORLDS WITHOUT END
dynamics. In the present condition of knowledge,
we are free to consider it equally possible either that
the Universe is slowly but inexorably pursuing its
course towards old age and inevitable death, or that
it is destined to undergo periodic rejuvenation and to
live its life over and over again. Sir Arthur Ed-
dington regards the second alternative as " wholly
retrograde " ; he has expressed himself to be an evo-
lutionist not a multiplicationist. Sir James Jeans
takes the same attitude: " It is hard to see what ad-
vantage could accrue from an eternal reiteration of
the same theme." But there are many others who
find the doctrine of the heat-death of the Universe
equally repugnant. To such it may be a consola-
tion to reflect that in the present state of knowledge
it is not possible to claim that this doctrine has been
definitely established. As a practical astronomer, I
must emphasise that these are at present realms of
speculation. Observation is the touchstone of every
theory or hypothesis in science; the two alternative
but divergent theories as to the future of the Uni-
verse cannot yet be tested by astronomical observa-
tions. Until this is possible, we are free to select
whichever we prefer.
INDEX
Absorbing clouds in Milky Way,
1 88
Achcrnar, 142
Adams, J. G., 76
Age of Earth, 15,94 ; of terrestrial
rocks, 19 ; of stars, 231
Algol, 165
Alpha Centauri, 139, 141, 144,
158, 1 68
Altair, 142
Ammonia on Jupiter, 64 ; on
Saturn, 70 ; on Uranus and
Neptune, 74
Annihilation of matter, 227, 228,
229
Antares, 150
Anti-trade winds, 4
Aristotle, 97
Asteroids, 59 ; size of, 60 ; total
weight of, 60 ; origin of, 61
Astraea, 59
Atmosphere of Earth, 10 ; of
Venus, 46, 86 ; of Mars, 54, 89 ;
of Jupiter, 64, 84 ; of Saturn,
69, 84 ; of Uranus and Nep-
tune, 74, 85
Atom, structure of, 160, 223 ;
ionized, 161
Atomic energy, 226
Availability of energy, 253
Bacon, F., 184
Barnard, E. E., 91, 189
Base-line for stellar instances, 137
Bayer, J., 131
Bayeux Tapestry, 100
Betelgeuse, 143
Biela's comet, 107
Binary stars, visual, 165 ; eclips-
ing, 1 66
Birth of matter, 252
Black dwarf stars, 225
Bode 5 J.E.,58 )7 5
Bode's Law, 58, 77, 80
Bolometer, 41
Bradley, J., 10
Branching of Milky Way, 183, 190
Brightest stars, the twelve, 141
Brightness of stars, 132
Brooks's Comet, 101, 107
Building-up of atoms, 228
Calcium clouds on sun, 122
Callisto, 67, 85
Canals of Mars, 51, 89
Candle-power of stars, 134, 145 ;
of Cepheid variables, 174 ; of
novae, 1 79 ; relation to mass,
222
Canopus, 141, 143
Carbon dioxide on Venus, 47, 86
Castor, 170
Cepheid variable stars, 171 ; size
and mass of, 1 73 ; candle-power
of, 1 74 ; in Andromeda nebula,
204 ; in globular clusters, 195
Ceres, 59 ; size of, 60
Challis,J., 76
Clouds in Earth's atmosphere, 1 2
Clusters, globular, 194 ; dis-
tances of, 195 ; distribution of,
195
Coal-sack, 189
Coma of comet, 102 ; size of, 105
Cornets, 97 ; orbits of, 99 ; perio-
dic, 101 ; coma or head of, 102;
nucleus of, 102 ; tail of, 102;
weight of, 105 ; disruption of,
107
Comets and meteor showers, 109
Companion of Procyon, 144 ; of
Sirius, 144
Conservation of energy and
matter, 253
Constellations, 130
Copernicus, N., i, 137
Corona, solar, 125 ; brightness of,
125 ; shape of, 125
Cyanogen in comet's tail, 106
Cyclonic disturbance, 4
Day, variation in length of, 3
Defoe, D., 98
Degradation of energy, 253
258 INDEX
Deimos, 57
Delta Cephei, 171
Delta Orionis, 192
Dense matter, 224
Density of Earth, 7 ; of Moon, 22 ;
of Jupiter, 6 1 ; of Saturn, 69 ; of
comet's head and tail, 105 ; of
Sun, 113 ; of giant and dwarf
stars, 150 ; of gaseous nebulae,
187 ; of interstellar gas, 193
Depressions, 4
Diffuse matter in Milky Way, 191 ;
density of, 193
Discovery of Uranus, 58 ; of
Neptune, 76 ; of Pluto, 78
Distance of Moon, 21, 136 ; of
stars, 138 ; of Milky Way star-
clouds, 185, 194 ; of globular
clusters, 195 ; of centre of
Galaxy, 195, 198 ; of spiral
nebulae, 209
Doughty, C.,44
Douglass, A. E., 124
Dunham, T., 65
Dwarf stars, 146, 224 ; densities
of, 150
Earth, rotation of, 2 ; speed at
equator, 4 ; size and shape of,
5 ; mass of, 5 ; density of, 7 ;
interior of, 7 ; atmosphere of,
10 ; age of, 15, 94 ; possible
end of, 249
Earth-light, 27
Earthquake waves, 7
Eclipse of Moon, 27 ; of Sun, 27,
119
Eclipsing binary stars, 166
Ecliptic, 37
Eddington, A. S., 222, 223, 225,
256
Electrons, 161, 216, 223, 227, 229
Encke's comet, 101
Equivalence of energy and mass,
226
Europa, 67
Evolution, stellar, tentative theories
of, 236
Expansion of Cosmos, 94, 212
Ferguson, J., 183
Flamsteed, J., 76, 131
Foucault, L., 2
Fowler, R. H., 224
Fraunhofer, J., 153
Fraunhofer's lines, 153
Full Moon, 26
Galaxy, 182 ; discoidal structure
of, 185 ; dimensions of, 196 ;
rotation of, 197 ; mass of, 198 ;
position of Sun in, 199, 212 ; as
spiral nebula, 207
Galileo, G., 67, 71, 114
Galle,J.G.,77
Ganymede, 67, 85
George the Third, 75
Georgium Sidus, 75
Giant stars, 146 ; densities of, 150 ;
Cepheid variables as, 1 73
Goodricke, J., 166
Gravitation, law of, 6
Gulliver's Travels, 57
Halley, E., 15, 99, 194
Halley's comet, 99, 102, 106
Hartrnann, J., 192
Heat-death of Universe, 255
Heliacal rising of stars, 10
Helium, u, 18, 156
Hcrschel,J., 195, 202
Hcrschel, W., 58, 75, 78, 182, 183,
184, 185, 188, 189, 195, 207
Hiawatha, 183
Hipparchus, 9, 132
Holbrook meteor fall, 112
Hubble, E. P., 204, 205
Huggins, W., 1 86, 201
Huyghens, G., 72, 186
Hydrogen content of stars, 221
Ice-ages, 171
Infra-red photographs, 45 ; of
Venus, 45 ; of Mars, 52 ; of
Jupiter, 63
Interior of Earth, 7 ; of stars, 218
Intrinsic brightness of stars, 134
10,67
lonization, 161, 216
Ionized atoms, 161
Ionosphere, 123
Jeans, J.H., 249, 256
Jeffreys, H., 65
Josephus, 100
Juno, 59 ; size of, 60
Jupiter, 6 1 ; distance of, 61 ;
velocity of, 61 ; size and weight
of, 6 1 ; rotation of, 62 ; atmo-
sphere of, 62, 64 ; red spot on,
62 ; temperature of, 64 ; rocky
core of, 65 ; ice coating of, 66 ;
satellites of, 67
Keeler,J. .,72
Kelvin, Lord, 17, 129, 225
Kepler. J., 39, 177
Kepler's laws, 39, 99
Kriiger 60, 147, 164
Kulik, no
Lalande, J. de, 131
Lassell, W., 78
Law of gravitation, 6
Length of day, variable, 3 ; on
Mercury, 43 ; on Venus, 48 ;
on Mars, 56 ; on Jupiter, 62
Leonid meteor shower, 108
Leverrier, V. J., 76
Life in other worlds, 82 ; condi-
tions for development of, 95
Lifetime of star, 230
Light time, 138
Light, total, of stars, 134
Light-year, 139
Lodge, O., 223
Longfellow, H. W., 183
Lowell, P., 51, 79, 89
Lunar craters, 30 ; origin of, 31
Magnetic storms, 117
Magnitude, stellar, 132
Mahomed II, 100
Mars, 49 ; distance of, 49 ; size
of, 49 ; weight of, 49 ; atmo-
sphere of, 49, 53, 89 ; polar caps
of, 49 ; surface of, 50 ; canals
on, 51 ; temperature of, 55 ;
INDEX 259
rotation of, 56 ; satellites of, 57 ;
possible life on, 90, 93
Marsh-gas on Jupiter, 64 ; on
Saturn, 70 ; on Uranus, 74
Mass of Earth, 5 ; of Moon, 22 ;
of Mercury, 42 ; of Venus, 44 ;
of Mars, 49 ; of Jupiter, 61 ; of
Saturn, 69 ; of Uranus and
Neptune, 74 ; of asteroids, 60 ;
of comets, 105 ; of Sun, 198,
227 ; of stars, 148, 173
Maunder, E. W., 124
Maxwell, J. G., 72, 104
Mercury, 41 ; distance of, 42 ;
size of, 42 ; weight of, 42 ;
surface of, 43 ; devoid of at-
mosphere, 42, 84 ; rotation of,
43 ; reflecting power of, 42 ;
temperature of, 43, 84
Messier, G., 202
Meteor crater in Arizona, 107
Meteor showers, 107, 108, 109
Methane, see Marsh-gas
Milky Way, 182 ; early knowledge
of, 183 ; star-clouds in, 185 ;
nebulae in, 186 ; absorbing
clouds in, 188 ; diffuse matter
in, 191. See also Galaxy
Minor planets, 59
Moon, distance of, 21 ; size of,
22 ; weight of, 22 ; devoid of
atmosphere, 23 ; speed of
motion of, 25 ; rotation of, 25 ;
reflecting power of, 26 ; con-
ditions on, 28 ; temperature of,
29 ; mountains on, 30 ; craters
on, 30 ; topography of, 32
Moving clusters of stars, 233
Nearest stars, the twelve, 143
Nebula, Great, in Andromeda,
202, 204 ; Cepheid variables in,
204 ; distance of, 205 ; novae
in, 205 ; size of, 206 ; similarity
to Galactic system, 206
Nebulae, gaseous, 1 86, 187; density
of, 187 ; origin of light of, 188
Nebulae, white, 201 ; spiral nature
of, 202
260
INDEX
Ncbulium, 187
Neptune, 74 ; size of, 74 ; mass of,
74 ; atmosphere of, 74 ; satellite
of, 78 ; discovery of, 76
New Moon, 26
New stars. See Novae
Newton, Isaac, 6, 40, 135
Not-hydrogen, 217
Novae, 1 75 ; swelling up of, 1 78 ;
candle-power of, 179 ; fre-
quency of, 1 79 ; cause of, 1 80 ;
in Andromeda nebula, 205
Number of stars visible to naked
eye, 131
Numbers of stars to various limits
of magnitude, 134
Nutation, 10
Oases on Mars, 89
Obscuring clouds, 190
Occultation of star by Moon, 24
Omega Centauri cluster, 194
Origin of planetary systems, 94 ;
of galactic systems, 240 ; of
stars, 242 ; of twin-stars, 243 ;
of solar system, 246
Oxygen, on Venus, 47, 87 ; on
Mars, 54, 89 ; origin of terres-
trial, 87
Ozone, in Earth's atmosphere, 1 2 ;
in atmosphere of Mars, 55
Pallas, 59 ; , size of, 60
Peiresc, 186
Penumbra of sun-spot, 116
Periodic comets, 101
Periodicity of sun-spots, 116 ; of
solar prominences, 122
Phobos, 57
Piazzi, G., 59
Planets, 36
Pluto, 78, 85 ; discovery of, 78 ;
size of, 80 ; temperature of, 80 ;
distance of, 80 ; orbital period
of, 80 ; velocity of, 80
Polar caps of Mars, 49, 50, 53
Pole-star, 9
Pons-Winnecke's comet, 106
Precession, 9
Prentice, J. P. M., 177
Pressure of radiation, 103, 218
Procyon, 142, 144 ; companion
of, 144
Prominences, solar, 119, appear-
ance of, 1 20 ; motions of, 120 ;
periodicity of, 122
Protons, 217, 223, 227, 229
Pueblo Bonito ruins, 125
Pulsating stars, 171 ; see also
Cepheid variables
Pulsating Universe, 235
Queen Matilda, 100
Radiation, pressure of, 103, 218
Radioactivity, 17, 129
Radium, 17
Rainbow, 151
Rameses II, 1 6
Ramsay, W., n, 157
Rare gases in Earth's atmosphere,
ii
Rayleigh, Lord, 11
Recession of spiral nebulae, 209,
234
Red spot on Jupiter, 62
Reflecting power of Moon, 26 ;
of Mercury, 42 ; of Venus, 45
Rigel, 142
Rosse, Lord, 92, 202
Rotation of Earth, 2 ; of Sun, 115;
of Galactic system, 197 ; of
stars, 243
Rowland, H. A., 156
Salinity of oceans, 15
Satellites, 37 ; of Mars, 57 ; of
Jupiter, 67 ; of Saturn, 73 ; of
Uranus, 78 ; of Neptune, 78
Saturn, 69 ; size of, 69, 70 ;
weight of, 69 ; atmosphere of,
69 ; temperature of, 70 ; con-
stitution of, 70 ; rings of, 71 ;
satellites of, 73
Saturn's rings, 71 ; nature of, 72 ;
origin of, 72
Schiaparelli, G. V., 51, 89
INDEX
26l
Schwabc, S. H., 116
Second law of thermodynamics,
253
Sedimentation, 16
Seismographs, 8
Shooting stars, 12, 109 ; size and
speed of, 109
Siege of Jerusalem, roo
Sirius, 135, 141, 144; companion
of, 144
Sky, blue colour of, 14
Slipher, V. M., 51
Solar system, scale of, 37 ; origin
of, 246
Source of stellar energy, 2 1 9
Space, emptiness of, 143, 240
Spectroheliograph, 1 2 1
Spectroscope, 151
Spectrum of gas, 153 ; of incan-
descent solid, 154 ; of sun, 156 ;
of stars, 159 ; arc and spark,
1 60
Spiral nebulae, 92, 202 ; Dotation
of, 208 ; distances of, 209 ;
velocities of, 209 ; number of,
211
Star-clouds in Milky Way, 185 ;
distance of, 194
Stars, number of, 134 ; total light
of, 134; candle-power of, 134,
145 ; sizes of, 140 ; tempera-
tures of, 140 ; the brightest,
141 ; the nearest, 143 ; giant
and dwarf, 146 ; weight of, 146,
148 ; elements present in, 163 ;
interior of, 2 1 8 ; hydrogen con-
tent of, 221 ; age of, 231 ;
gravitational interaction of, 232
Sun, 113; size of, 113; mass of,
198, 227 ; density of, 1 13 ; rotat-
ing, 115 ; output of heat and
light from, 126, 128 ; tempera-
ture of, 127 ; maintenance of
heat of, 128 ; spectrum of, 1 56 ;
elements in, 156 ; central
temperature of, 220
Sun-spots, 114 ; motion of, 115 ;
periodicity of, 1 1 6 ; nature of,
117; magnetic field of, 118;
connection with aurorae and
magnetic storms, 117, 122 ; and
weather, 123
Swift, Dean, 57
Tail of comet, 102 ; length of,
103, 105 ; direction of, 102 ;
nature of, 103
Temperature of Moon, 29 ; of
planets, 41 ; of Mercury, 43 ;
of Venus, 48, 88 ; of Mars, 55,
93 > f Jupiter, 64 ; of Sun,
127 ; of stars, 140 ; of stellar
interiors, 216, 220
Thermodynamics, 253
Thermopile, 41
Thorium, 17
Tides, 6, 33 ; spring, 34 ; neap,
34
Titan, 73, 85
Tolman, R. G., 255
Tombaugh, C. W., 78
Trade-winds, 4
Transmutation of elements, 229
Tree-rings, 124
Twin stars, 146, 164 ; origin of,
243
Tycho Brahe, 175
Ultra-violet photographs of Mars,
52 ; of Jupiter, 63
Umbra of sun-spot, 116
Uranium, 17
Uranus, 74 ; discovery of, 58, 75 ;
size of, 74 ; mass of, 74 ; atmo-
sphere of, 74 ; early observa-
tions of, 75 ; satellites of, 78
Vanovara meteor fall, 1 10
Vega, 142
Velocity of escape from Moon, 22 ;
from Mercury, 42 ; from Venus,
45 ; from Mars, 49 ; from
Jupiter, 6 1 ; from Saturn, 69
Venus, 44 ; size of, 44 ; weight
of, 44 ; phases of, 44 ; atmo-
sphere of, 46 ; temperature of,
48, 88 ; conditions on, 88 ;
possible life on, 88
262
INDEX
Verne, Jules, 21
Vesta, 59 ; size of, 60
Visual binary stars, 165
Water-vapour, 12 ; on Venus,
48, 88 ; on Mars, 54, 89
Wave-length of light, 1 52
Weighing the Earth, 5
Weight of Earth, 5 ; of Moon, 22 ;
of Mercury, 42 ; of Venus, 44 ;
of Mars, 49 ; of Jupiter, 61 ; of
Saturn, 69 ; of Stars, 1 46, 1 48,
173
Whirlpool nebula, 202
Whitaker*s Almanack, 69
White dwarf stars, 224
Year, length of, 10
Zeta Ursae Majoris, 169
