Skip to main content

Full text of "The Pageant of the Stars"

See other formats










1:7". . 






j*.* aJUI IX^^mLa *«i. 

*&^Buiirr * 





Assistant Professor of Astronomy 
Harvard University 

STANLEY PAUL fcf CO. (1928) LTD. 

Photograph Harvard Observatory 


The Large Magellanic Cloud, an island universe in miniature, and ihe nearest neighbor of 
the Milky Way system. It is 100,000 light years distant, and contains millions of srars, 
many nebulre, star clusters, and variable stars. Although its southern position in the sky 
makes it invisible in our latitudes, it is a conspicuous object south of the equator. 



Whose penetrating insight first perceived the light that has guided 
the astronomers of to-day through the darkness of the universe. 

Sic ilur ad astra 

rnn' ■-■ 



Introduction xv 

Chapter I. The Origin of the Solar System 1 
The scientific interpretation of Genesis. The 
disastrous visit of a strange star. 

Chapter II. The Earth. 14 

Our planet, the earth. How large it is, and 
how old. Its motion in space. 

Chapter III. The Sun. 31 

The sun, as the giver and maintainer of life. 
Its size, its temperature, and what it is made 
of. Sun-spots, solar volcanoes, northern 
lights, and magnetic storms. The hidden 
source of energy of the sun; its past and its fu- 

Chapter IV. The Moon. 48 

Our satellite, the moon. Its distance, size, 
and motion around the earth. The moon as 
a timekeeper, and the origin of the calendar. 
The moon's landscape. 

Chapter V. Eclipses. 63 

Nature's greatest spectacle: a total eclipse of 
the sun. Why an eclipse takes place. Eclipses 
of the moon. Tests of the theory of rela- 
tivity through observations made during an 




Chapter VI. The Terrestrial and Minor 

Planets 77 

Fugitive Mercury, the Sun's closest neigh- 
bour. Venus the beautiful, our evening and 
morning star. Fiery Mars and its possibilities 
of life. The asteroids, celestial fragments, 
pygmies of the solar system. 

Chapter VII. The Major Planets. 99 

Jupiter, the giant of the planets. Saturn the 
unique. Pale Uranus and Neptune; the great- 
est triumph of scientific prediction. 

Chapter VIII. Comets and Meteors. 117 

Comets, wanderers of the wasteland, the semi- 
permanent members of the sun's family. 
Meteors,, our daily visitors from the great 
beyond and our only clue to the emptiness of 
space. The frontier of the sun's domain. 

Chapter IX. The Nature of the Stars. 138 

The brightness and the number of the stars. 
Application of photography and of the spec- 
troscope. The physical constitution of the 
stars, their temperature, and their size. 
Stellar distances. Measuring the universe 
with a new yardstick, the light year. 

Chapter X. Stellar Motions and the Near- 
est Stars. 164 

Apparent motion in the sky and motion in the 
line of sight, proper motion, and radial 
velocity. What we know about the brightest 
stars and about the nearest stars. Families of 
stars. Traffic laws of the universe. 




Chapter XI. Double Stars, Variable Stars, 

ANDNoViE. 188 

Distant suns revolving around each other 
under the influence of the law of universal 
attraction. _ Eclipses among the stars. Stars 
that vary in brightness. Explosions in the 
cosmos: new stars, bursting forth from in- 
significance to unrivalled splendour. 

Chapter XII. Clusters and Nebula. 212 

Gregariousness among the stars, clusters, and 
star-clouds. Nebula?, dust-clouds, and cos- 
mic fog. The enigma of nebulium solved. 

Chapter XIII. Stellar Evolution. 233 

The biography of a star, its birth, its radiant 
life, the decalogue for its behaviour, and its 
ultimate death. The interior of a star and 
the riddle of the universe: what becomes. of 
the light of the stars? 

Chapter XIV. The Milky Way System. 245 
The structure of our Galaxy, its laws and its 
changes. Traffic regulations, celestial col- 
lisions. The search for the centre of the 

Chapter XV. Island Universes. 257 

Other galaxies, the Magellanic Clouds, the 
Andromeda Nebula. The era of spiral nebu- 
las. The evolution of island universes and the 
finiteness of our material cosmos. 

Chapter XVI. From Chaos to Cosmos. 280 

A review of the material creation and of 
man's utter insignificance in it. Infinity and 

Name Index. 291 

Subject Index. 297 
















The Large Magellanic Cloud 




The Sun. A Sun-spot 

Northern Portion of the Moon 

A Great Solar Prominence. The 

Photographs of Planets 

Comet Morehouse. The Orbit of a 

Alpha Centauri and the Southern 

Proxima Centauri. Nova Pictoris 

Omega Centauri 

The Horsehead Nebula 

The Region of Rho Ophiuchi 

The North America Nebula 

The Milky Way in Scorpius and 

Three Island Universes 

The Spiral Nebula in the Big Dipper 269 

The Spiral Nebula in Triangulum 273 







figure . page 

1. Comparative Sizes of the Planets 12 

2. Precession 27 

3. The Real Motion of the Earth in Space 29 

4. Phases of the Moon 51 

5. Diagram of an Eclipse 67 

6. Phases of Venus 79 

7. Planetary Orbits inside Jupiter 93 

8. Sizes of Asteroids, and of the Satel- 

lites of Mars 95 

9. Planetary Orbits Outside Jupiter 105 

10. Sizes of the Moons of the Outer 

Planets 112 

11. Sizes of the Stars 150 

12. Sizes of the Stars 151 

13. Parallax of a Star 155 

14. Change in the Big Dipper 182 

15. Change in Alpha Centauri and the 

Southern Cross 183 

16. An Eclipsing Variable 197 



"'And who the deuce was Pythagoras?' 
" 'A sage who held that the earth is round and that it 
moves round the sun.' 
"'What an utter fool! Couldn't he use his eyes?'" 

— Shaw. 

Man cannot live by faith alone. Surrounded as 
he is by a world of facts, he seeks knowledge and 
understanding of these facts. On a knowledge of 
facts, however imperfect, man must build the super- 
structure of faith. His knowledge represents his 
determination to be fully conscious of the material 
universe; his faith represents his desire to be at 
peace with the spiritual universe. Civilization is 
man's effort to achieve such knowledge and to attain 
such faith. In the pursuit of these ends astronomy 
plays a unique and significant part, since it is the 
only science that deals with the material reality 
outside this earth. 

Astronomy was born out of wonder at the mys- 
tery of the dark and starlit night, wonder at the 
countless host of stars, so familiar and yet so 
remote; that wonder which Plato called the soul of 
science. Emerging from this primitive wonder, 
astronomy has matured down the centuries, widening 



its scope as man's mind turned from itself to press 
on in its bold and undeterred quest of the bounda- 
ries of his universe, boundaries which have now 
receded so far that his knowledge of fact and his 
exercise of faith unite to set his finiteness in infinity. 
Consequently, advance in astronomy is a phase 
of the advance of civilization — as man's outlook 
grew less parochial astronomy progressed from an 
anthropocentric to a geocentric point of view. At 
this stage it was sufficiently dominated by the author- 
ity of Aristotle, lingering throughout the Middle 
Ages, and by ecclesiastical interpretation of the 
Scripture, to postpone all further development until 
the general intellectual awakening of the Renaissance. 
It is no mere coincidence, therefore, that we find the 
formulation of the new truth in astronomy taking 
place simultaneously with the struggle for new ideas 
in religion. In 1512 Copernicus first published his 
views on the rotation of the earth and the central 
position of the sun in the planetary system — five 
years before Luther's dramatic gesture at Witten- 
berg. Copernicus's heliocentric system led to New- 
ton's discovery and demonstration of the principle 
of universal attraction, and with this first expression 
of a perfect law of nature it may be said that astron- 
omy came of age as a science. In the meantime the 
telescope had been invented, and its introduction 
into astronomy, coupled with Newton's law, entirely 
changed the aspect of our science. Naked-eye astron- 
omy ceased to exist, the universe became increasingly 



telescopic, and as a natural consequence astronomy 
developed into a pure science, thus severing its con- 
nection with the theological view of creation. The 
next century saw the development of celestial me- 
chanics and with it the desire to inquire into the 
motions of stars and planets; researches into the 
structure and the mechanism of the cosmos sup- 
planted the former simple description of the visible 
heavens. Astronomy to-day is almost exclusively 
telescopic, the naked-eye stars constitute considera- 
bly less than one millionth part of those that are 
now visible in our largest telescopes; the discovery 
of thousands of island universes and the introduction 
of the doctrine of relativity have entirely changed 
the concept of space. But in all this tremendous 
development we find unity: matter is the same every- 
where, chemical elements, atoms, and electrons are 
the same in the stars and nebula as on the earth,- and 
they obey the same laws everywhere. 

Through the introduction of giant telescopes and 
of photography, and through the application of 
modern physics and chemistry, new vistas have been 
opened far beyond the wildest dreams of our prede- 
cessors. At the same time, astronomy, though 
grown more diversified, has yet preserved the unity 
of its basic truths. To-day more than ever before 
we stand silent in admiration before the truths 
unveiled by astronomy, before the unity of fact 
throughout creation. 

Chapter I 


"Parturiunt montes, nascetur ridiculus mus." 

— Horace. 

Millions of millions of years ago our sun was 
travelling through space — alone. It was still a 
young sun, much more brilliant and much larger, 
and quite alone, having no attendant planets. Then, 
out of the millions of other stars, one seemed to 
single itself out. The others all appeared to move 
haphazardly, approaching the sun for a while, pass- 
ing at a considerable distance, then losing them- 
selves again in the depths of space; this one star, 
however, appeared to aim straight for the sun. For 
thousands of years it approached, and there seemed 
no doubt but that the sun and this star were destined 
to meet, both driven helplessly down their courses 
by the force of gravitation. In the eternal silence 
of empty space these two denizens of the universe 
were hastening toward their inevitable cataclysm, 
unwatched by witnesses, unconscious of their sur- 
roundings. As they continued to approach, their 
mutual attraction began to make itself felt, for not 



only did they both acquire greater and greater 
speed, while at the same time curving into one 
another, but, both being gaseous, they raised huge 
tides on each other's surface. Both assumed a 
pearlike shape, their speeds increased to a hundred 
miles a second or more, and still they approached. 
Then, when the fateful hour had struck, the stems 
of the two pear-shaped bodies melted into one 
another, the stars were thrown into convulsions, and 
a desperate combat ensued, with each star trying to 
battle its way to freedom and attempting to get 
away again with all its belongings. 

Here Nature stepped in and came to the rescue : 
for the very force of attraction that had made them 
melt into one another was also the source of their 
enormous speeds, and it was their terrific velocities 
that enabled them to separate again. The connect- 
ing link could not stand the strain and pull resulting 
from these great speeds, it broke into a thousand 
fragments, and for a while the scene was one of 
turmoil. When the smoke cleared there emerged 
from the chaos two very much mutilated suns, each 
tenaciously clinging to thousands of small bits of 
matter, each in furious rotation and each trying 
desperately to escape from the scene of recent con- 
flict as quickly as its great speed would carry it. As 
the two stars separated, each gradually established 
order in the nondescript mass of accompanying frag- 
ments, and out of this chaos of matter there rose, 
in the course of time, an orderly and well-regulated 


system of one large, central body surrounded by a 
number of smaller bodies revolving about the for- 
mer : the planetary system. 

Such is the picture which the present-day astron- 
omer and cosmogonist paints of the origin of the 
solar system. There were no eye witnesses of the 
event; the other star has become lost in the voids 
of infinite space during the millions or billions of 
years that have elapsed. Yet, the astronomer can 
reason back all this length of time, and from his 
knowledge of present-day facts of the solar system, 
and with the aid of his mathematics and physics 
reconstruct the major aspects of the cataclysm. 

We have sketched the occurrence essentially from 
the point of view of an inhabitant of a planet. To 
such a being the effect on the sun of the disastrous 
visitor was indeed gigantic in proportions; from 
the sun's point of view, however, it was little more 
than a passing disturbance, an incident in its life 
which had but small lasting influence. After the 
other star had wandered away, things quieted down 
again and only those fragments that had gone too 
far, or had acquired too much speed, were able to 
maintain a separate existence. The majority of the 
ejected material dropped back again into the sun, 
and when stock was taken of the actual losses it 
was found that the losses did not amount to more 
than one per cent. 

Before a theory of as great importance as this 
can be accepted it is always well to inquire whether 


it is reasonable. By reasonable we mean not 
improbable. In other words, what are the odds 
that two stars do come close enough together to 
give rise to such great disturbances as the birth of 
a planetary system, and how does the theory fit in 
with our ideas about the age of the earth? We 
know that head-on collisions between two stars 
occur not much more often than about once in every 
trillion years for the whole universe. A close 
approach will be much more frequent, and espe- 
cially in the denser parts of the universe it may 
well occur thousands or millions of times more 

Geologists tell us that the earth's crust is probably 
more than one thousand million years and less than 
eight thousand million years old. On the other 
hand, mathematicians can arrive at an estimate of 
the time required to change the original chaotic 
state of the solar system into its present more 
orderly state. Jeffreys, in England, has calculated 
that this would take one thousand million years at 
the very least, and might well take much longer. 
Other estimates have been as high as eight, or ten 
billion years. Considering these excessively long 
times, it appears as if the odds are no longer against 
the theory, and herein really lies the crux of the 
whole matter: given time enough, almost anything 
may happen, and of time astronomers have plenty. 

The theory as we have outlined it here is largely 
the work of four men: Chamberlin and Moulton 


of Chicago and Jeans and Jeffreys of Cambridge, 
England, although the germ of the idea is already 
to be found in the writings of Buffon. The theory 
is by no means complete, and we must not expect it 
to explain everything. There are still many un- 
solved problems in the solar system, still many 
unanswered questions, and even some obstacles 
which, for the sake of the theory, we prefer to ig- 
nore at present. Taken on the whole, however, the 
Chamberlin-Moulton theory explains the salient 
points in the problem of the formation of the solar 
system, and meets with no direct contradictions. 
And this, after all, is about all we can ask of a 
theory of the genesis of the planets, an event which 
must have happened billions of years ago. 

Previous to Chamberlin and Moulton, the "nebu- 
lar hypothesis" of Laplace held the field against all 
others. According to this theory, the whole solar 
system once was an enormous nebula which, by 
some means or other, got into a state of rapid 
rotation. As time went on, the nebula shrank and 
thus increased its rapidity of rotation, until finally 
it threw off rings of matter. These rings were then 
supposed to condense and form planets, which in 
turn rotated and threw off smaller rings which 
formed the satellites. Unfortunately, it has been 
shown by recent work of Jeans on the behaviour of 
gases that such a nebula could not throw off rings, 
and that, even if rings were formed, they could 
never condense into a planet. And thus, in spite 


of its otherwise attractive features, Laplace's nebu- 
lar hypothesis has had to be abandoned; one serious 
objection was sufficient to kill it. 

Returning once more to Chamberlin and Moul- 
ton's idea, or the hypothesis of "dynamic encoun- 
ter," as it is called, we note that the chances for 
such an event to happen, though not prohibitively 
insignificant, are still small enough to render improb- 
able the existence of large numbers of planetary 
systems. It is rather gratifying to our terrestrial 
vanity to be able to think that planetary systems 
are not common in the universe. We know of only 
one with any degree of certainty, viz., that one 
appertaining to the star that was so instrumental 
in the creation of our own. However, this other 
star, with its attendants, may now be in some remote 
corner of the universe, lost among its millions of 
neighbours, and, in spite of our curiosity, it seems 
unlikely that we shall ever be able to identify it. 

Having explained why we are here, it is time 
to consider the question why we continue to be here, 
and to study the why and how of the motions of the 
planets and their satellites around the sun. The 
first desire for an explanation of these motions 
originated with the ancient astrologers who sought 
to use the change in position of the heavenly bodies 
for making forecasts concerning the weal and woe 
of humanity. From their point of view the planets 
and stars were mere puppets moved by the hands 
of invisible gods to indicate their wishes to the 


human race, and it was therefore sufficient to find 
a descriptive explanation of the motions; the reason 
why never entered into the question. Considering 
that the world conception of the ancients was geo- 
centric, 1 it is not surprising that the first systems 
proposed all postulated the planets moving in circles 
with the earth as centre. As observations became 
more precise this simple idea proved untenable, and 
complicated additions were made to it: the planets 
were now supposed to revolve in circles, called epi- 
cycles, the centre of which revolved in another circle 
around the earth as centre. This system was 
devised by Hipparchus and Ptolemy, and published 
by the latter in his Almagest, the great astronomical 
encyclopaedia of antiquity. Having risen to its 
highest point in ancient times in this work, astron- 
omy was left there for many centuries, partly 
because the Church of the Middle Ages, in its inter- 
pretation of the Scriptures could not allow the earth 
to be dethroned as the centre of the universe. And 
so long as the earth remained fixed in the centre 
of the universe no better interpretation of the 
motions of the heavenly bodies was possible. 

Toward the end of the Fifteenth Century how- 
ever, the power of the mediaeval church began to 
wane. New thoughts and new ideas sprang up 
everywhere, not only in the arts, but also in science. 
In Germany these found their expression in the 
theories of Copernicus, who asserted that the earth 

1 Earth as centre. 


rotated on its axis, and that all the planets, includ- 
ing the earth, revolved about the sun. For various 
reasons Copernicus withheld his manuscripts from 
publication, and it was not until he lay on his death- 
bed, in 1543, that he saw the proofs of his work: 
De Revolutionibus Orbium Coeleslium. 2 His cause 
was espoused in Italy by Giordano Bruno and Gali- 
leo Galilei, who, for this reason, were opposed by 
theologians and scientists alike. Galileo's treatise 
on the Copernican system was interdicted and it 
is interesting to note that this ban was not lifted 
until 1835. In Denmark Tycho Brahe fought the 
new system, and undertook a long series of accurate 
observations to prove it wrong. Tycho died before 
he could complete his work. His observations fur- 
nished Kepler with the means to prove that the 
planets move in ellipses instead of in circles around 
the sun, and to formulate his laws upon the rapidity 
of this elliptic motion — laws which were in direct 
contradiction to what Tycho had expected to obtain 
from his observations. 

Thus far all explanations of the planetary system 
had been descriptive. It remained for Newton to 
formulate the simple law behind it all: the law of 
gravitation or universal attraction. With one stroke 
of genius Newton solved all difficulties of the planet- 
ary system. The enunciation of the simple but 
fundamental principle that all things material attract 
each other with a force proportional to their weight 

2 Concerning the Movements of the Heavenly Bodies. 


and inversely proportional to the square of the 
distance between them sufficed to explain all motion. 
Copernicus's book appeared in 1543, Newton's 
Principia in 1687 : in less than one hundred and fifty 
years astronomy had developed from a primitive, 
descriptive science to one which has found its funda- 
mental law. It is true that Einstein's theory of 
relativity with its accompanying change in our 
physical and philosophical concepts has given us a 
new insight into the nature of gravitation, but its 
resultant change in the mechanics of the solar system 
has been practically nil. Except for a very small 
correction in one instance, Newton's law still holds 
in its entirety, and it stands to-day, as it did two 
hundred and fifty years ago, as the basic law of our 
concept of planetary motion. 

Newton's law applied to the Copernican system 
gives a complete picture of the mechanism of the 
planetary system. It is shown that all planets must 
move in ellipses, with the sun slightly outside the 
centre. It might be well at this point to say that 
an ellipse is, roughly speaking, oval in shape, and 
it has two points in its interior called foci which 
have singular geometric properties; in the planetary 
ellipses the sun occupies one of these. These ellipses 
are all very nearly circular, their eccentricity, which 
determines the ratio of the greatest length to the 
greatest width, being very small, and for practical 
purposes we may well consider them as circles here. 
At present eight planets are known. In order of 






O 00 

O «-< 1"~ t^ T}< 
HW5O0 1ON 



»-< 0\ «-< »-i o 
<"0 »-< 



© © o 

00 t"» C5 fi vo 









CO t» 

f5 r- 

cn O *& O f» 



to ^^ 
C co 



S3 f- t: rt Q.5 rt g- 




their distance from the sun they are: Mercury, 
Venus, the Earth, Mars, Jupiter, Saturn, Uranus, 
and Neptune. In the table are given the most 
important things we know about them at present: 
their mean distance from the sun, in millions of 
miles and expressed in the distance sun-earth as 
unit; the times required to complete one revolution 
around the sun; their diameter in miles, and their 
mass in terms of that of the earth. In the last two 
lines of the table similar data for the sun and the 
moon are appended. 

It is seen that the first four planets are all much 
smaller in size than the next four; in addition, they 
are much closer to the sun, forming, so to speak, 
the inner half of the solar system. Sometimes they 
are called the terrestrial planets, to distinguish them 
from the other four, the major planets. Between 
the orbit of Mars, the outermost of the terrestrial 
planets, and that of Jupiter, the innermost of the 
major planets, a large number of very small objects 
are situated, the minor planets or asteroids. At 
present already more than one thousand of these 
are known; however, they are of interest collec- 
tively, rather than as individuals. 

A glance at the table reveals a regularity in the 
distances of the planets from the sun which can be 
expressed to a fair degree of approximation by a 
simple and curious series of numbers known as 
Titius's or Bode's law. If we write down the 
geometric series 3, 6, 12, 24, 48, 96, 192, add zero 



to the beginning, then add 4 to each member, we 
obtain the new series 4, 7, 10, 16, 28, 52, 100, 196. 
Comparing it with the numbers in the third column 
of the table of elements for the planets, we see that 
the approximation is indeed fair. The number 28 


may be taken to represent an average minor planet, 
somewhere between Mars and Jupiter. It is not 
known at present whether this agreement is a mere 
coincidence or whether there is some law of nature, 
as yet not understood, behind it. 

Looking through the table, we further notice 
that the combined mass of the eight planets is very 



small compared with that of the sun. Undoubtedly, 
the total mass represented by the thousands or more 
minor planets is insignificant compared even with 
that of the planets. Practically the whole mass of 
the system, therefore, is at present concentrated in 
the sun. Reflecting on the past history of our solar 
system, however, it is surprising to find that more 
than half of the mass of what was once the solar 
system has now disappeared. The sun, we know, 
is radiating light and energy into space, but as a 
result of that, so relativity assures us, the sun is 
losing weight at the prodigal rate of more than 
two million tons a second. We have seen that the 
planetary system is probably about five billion years 
old, and the sun itself not less than a few trillion 
years. Furthermore, it is likely that the sun has 
kept on shining during all this time at least as 
brightly as at present, and probably increasingly 
so the farther back we go in time. A simple calcu- 
lation then shows that the present mass of the sun 
is considerably less than that which it had at the 
birth of the planetary system, and probably less 
than half of that with which it was born. Thus, 
we may readily say that the larger part of the solar 
system has now ceased to exist: it has been radiated 
away into space. 



Chapter II 

". . . 'What makes me believe that nobody lives 
here,' said the Saturnian, 'is that it appears to me 
that common-sense people would not want to live 
here.' 'Well,' ventured Micromegas, 'perhaps the 
inhabitants lack common sense.' " 

— Voltaire. 

From the point of view of the universe our sun 
may be only a second-rate star, and our earth only 
a second-rate planet of the sun. To us, terrestrials, 
the earth always will be the most important and 
the most interesting of all objects in the universe. 
In describing it, the astronomer feels embarrassed 
by the wealth of material at his disposal; where 
to begin, and what to consider. There are so many 
points of interest that in most fields any discussion 
will soon lead into the borderlands of the various 
related sciences: geology, meteorology, geodesy. 
Most important of all to the living human being, 
the biological sciences have their say; even theology 
voices her claim to a place in the sun. 

We have seen how the modern scientist visualizes 
the formation of the earth as an independent body, 


as the outcome of a calamity which befell the sun 
between one and ten billion years ago. Thus the 
earth came into being, but it was then by no means 
fit to live on. For the first 15,000 years of its 
existence the earth was gaseous or liquid, and 
rotated on its axis in about four hours; the day 
and night at that epoch, therefore, were only two 
hours long. At the same time the earth had not 
yet quieted down, but was still contracting and 
expanding, vibrating so to speak, in a period of 
about two hours, while the sun was raising huge 
tides on it, with "high tide" twice a day, that is on 
the four-hour-day basis, every two hours. We all 
know what happens when we start a swing, and 
then give it a push every time it reaches its highest 
point: the length of the swing will become greater 
and greater. The same happened to the earth 
which, although liquid, was of a syrupy consistency. 
It was swinging in a period of two hours, and every 
two hours the sun gave it an additional impetus by 
means of the tides. Consequently high tide became 
higher and higher, until finally the top of the 
syrupy wave was racing around so fast, and creating 
such a strain, that a large piece broke off, left the 
earth, and henceforth pursued a separate path: the 
moon was born. After the earth had performed 
this duty it settled down to a more normal state, 
its temperature, which had originally been as high 
as 500 degrees, quickly dropped, enabling the crust 
to solidify, and water to condense and form the 



oceans. The time elapsed since this process was 
completed has been estimated in several ways, but 
the most reliable method appears to be that which 
uses the breaking up of radioactive substances. You 
have undoubtedly heard of radium, used so much 
in therapeutics. To the chemist radium is just one 
of those ninety elements, those simplest of sub- 
stances, from which everything is built up. These 
elements constitute the "bricks" from which the 
whole universe is made. When we list them in 
order of their weight, we find that radium, uranium, 
thorium, and several others, all of which occur at 
the heavy end of the list, are also the ones that 
decompose most easily. Being so heavy and com- 
plex, they seem to have taxed Nature's resources to 
the limit, they become unstable, and are slowly but 
constantly disintegrating. Give them time enough, 
and they will all turn into simpler elements, princi- 
pally lead. The rate at which this decay goes on 
is controlled entirely by forces in the innermost part 
of the atom, so near the nucleus, and so well 
protected against outside interference, that noth- 
ing we can do to an atom of such an element in 
the laboratory seems to have the slightest effect. 
If we may then assume that such has been the case 
since the formation of the solid crust of the earth, 
we have merely to pick up a rock that contains 
uranium and radium, determine the amount of lead 
that has been formed in it, and we have its age, 
and with it the age of the earth's crust. Estimates 



made in this way indicate that the solid crust of the 
earth is more than one billion but less than eight 
billion years old. 

When it comes to describing the successive stages 
of evolution of the earth's surface since the forma- 
tion of the solid crust, and through the early periods 
of life up to the present day, we should have to 
follow the geologist through his list of impressive 
names of the various periods of the development 
of the earth, names which, apart from their impres- 
siveness, can have but little meaning for the unini- 
tiated. Let us rather take the earth for granted, 
take it as it is to-day, with its solidified crust, its 
mountains, its oceans, and its atmosphere, and first 
consider its figure. 

Far back in antiquity, and by all primitive peoples 
alike, the earth was supposed to be a flat disk. 
Some of the Greeks, however, held different views. 
Eratosthenes, who lived about 250 B. C, noticed 
that on the longest day of the year the sun cast no 
shadow at Syene, in upper Egypt, and was therefore 
straight overhead, while at Alexandria the shadow 
cast by the sun corresponded to a deviation of 
7 degrees from the point overhead. Then, Eratos- 
thenes reasoned, if the earth is a sphere, the dis- 
tance from Syene to Alexandria, 500 miles, must 
be equal to an arc of 7 degrees on its surface, and 
the circumference of the earth must measure 25,000 
miles: remarkably near the true figure. More 
precise measurements made in modern times have 



shown that the earth is not a perfect sphere, but 
flattened at the poles, or an oblate spheroid as the 
mathematicians will have it. Its greatest diameter 
is at the equator, 7,927 miles in length, while the 
polar diameter is 27 miles shorter, totalling only 
7,900. This flattening of the earth is not really 
conspicuous; on a terrestrial globe of one foot 
diameter the indentation at both poles would not 
amount to more than one-fiftieth of an inch. The 
highest mountain and the greatest ocean depth, both 
of the order of five or six miles, would be less than 
one hundredth part of an inch, and might be repro- 
duced, on such a terrestrial globe 12 inches in size, 
by a change in the coat of varnish. Seen from a 
great distance, the earth would appear almost as 
smooth as a billiard ball. 

Very recent, and sensitive measures have indi- 
cated that the earth is not even a perfect spheroid 
but an ellipsoid, with three unequal axes. There 
seems to be a slight bulge amounting to at most 200 
yards along a certain meridian, with a corresponding 
deficiency along the meridian at right angles. Curi- 
ously enough, the meridian favoured by the bulge 
seems to coincide almost exactly with the meridian 
of Greenwich. 

Using a very delicate pair of scales, the Eotvos 
balance, physicists have been able to "weigh" the 
earth, by comparing its attraction to that of a large 
ball of lead or quartz, of known weight. Their 
answer is : six thousand million million million tons, 



a 6 with 21 ciphers (6,000,000,000,000,000,000,- 
000). From the diameter given above we calculate 
that the total volume of the earth is 259,000 million 
cubic miles, then dividing this into the weight we 
find a density of 5.5, that is to say, taken on the 
average, the earth is five and one-half times heavier 
than water, almost as heavy as pig iron. Since 
the superficial layers of the earth are composed 
of rock, and are considerably lighter, having a 
mean density of only 2.7, it follows that the inner 
core must be much denser. This can be explained 
by the enormous pressures exerted by the outer 
layers on the inside, which pressure already amounts 
to 300 tons per square inch at a depth of only 100 

The best way we have of determining what the 
interior of the earth is like, is by observing, by 
means of a seismograph, earthquakes at distant 
points. The different ways in which these tremors 
are transmitted along the surface and through the 
inner core of the earth make it possible for us to 
"see through" the earth; the earthquake vibrations 
constitute a sort of X-rays. The conclusion is 
reached that most of the inner part of the earth is 
more rigid than steel, although at the very centre 
it may be liquid in consistency. To this knowledge 
of the earth's interior, Jeffreys has added some data 
concerning the temperature, having found that 
where the increase in temperature is about 1 degree 



centigrade for every 100 feet downward, layers 
deeper than 400 miles below the surface are still 
at practically the same temperature as when the 
earth was born. 

A description of the earth would not be complete 
without mention of the blanket that protects us 
from the biting cold of empty space, the atmosphere. 
Near the surface of the earth this gaseous envelope 
is composed principally of two gases, nitrogen and 
oxygen, roughly in the proportion of 4 to 1, with 
slight additions of water vapour, carbon dioxide, 
and some of the lighter gases such as hydrogen and 
helium. As we ascend in this atmosphere, the heav- 
ier constituents, water vapour and carbon dioxide 
diminish rapidly, while the atmosphere as a whole 
is getting rarer and colder. At a height of IS miles 
above the surface we have passed through all but 
4 per cent, of the atmosphere, and yet we know 
that there is enough atmosphere left at heights of 
100 miles or more, to produce white heat by friction 
in meteors speeding through it. 

This is what we know about the earth itself, its 
origin, its constitution, and its surface conditions, 
and to us who live on the surface it is almost 
enough ; to us the earth is the symbol of immobility. 
This tranquillity is only apparent, however, for to 
an onlooker in the universe at large this earth would 
present a spectacle of extreme restlessness, of almost 
distressing complexity of motion. Suspended in 
empty space, the earth would appear to be agitated 



by a variety of forces, each of which made it move 
in a different path, with the result that the combined 
"true" motion would present so baffling a problem 
that its solution appears to contain insurmountable 
difficulties. First of all there is the diurnal motion, 
the rotation of the earth on its axis, to us the most 
important of all perhaps, since it gives us day and 
night and as such controls our mode of living, and 
almost all life on earth. From the point of view 
of explanation by mathematical formula? it is imma- 
terial whether we assume the earth as fixed, and 
make the sun, moon, stars, and planets revolve 
around us in twenty-four hours, or whether we 
ascribe the rotation to the earth. The former point 
of view was naturally taken by all primitive people, 
to whom the earth was the centre of the universe. 
When Copernicus dethroned the earth from this 
position, and postulated its rotation on an axis, the 
chief argument for his side was that it was far more 
probable that only the earth moved than that all 
the heavenly bodies together moved in unison. 
Later an experimental and absolutely convincing 
proof for the rotation of the earth was devised by 
Foucault and tried in 1851 in the Pantheon in Paris. 
His apparatus was quite simple and consisted of a 
heavy iron ball suspended from the ceiling of the 
dome of the Pantheon by a wire 200 feet long. The 
ball was set in motion by being pulled sideways and 
released, and allowed to swing freely. It was then 
noticed that the plane in which this pendulum was 



swinging slowly rotated from east to west. A math- 
ematician can easily prove that, once a pendulum is 
set in motion, and in the absence of external forces, 
the direction of its swing docs not change; there is 
then but one conclusion to be drawn from Foucault's 
experiment, namely that not the plane of the pendu- 
lum but the floor under it was turning, and turning 
from west to east. In the rotation on its axis the 
earth carries everything on its surface around with 
it, and thus all objects on the equator cover a dis- 
tance equal to the circumference of the earth, 25,000 
miles, every day, corresponding to a speed of 17 
miles per minute. This velocity diminishes as we 
go from the equator to the poles; in the latitude of 
New York it is only 11 miles a minute; at the poles 
themselves it is zero. 

One of the greatest benefits we derive from the 
rotation of the earth is that it provides us with a 
clock, in fact with almost the only clock we possess. 
As space around us is empty, there is no friction, 
and we may expect the speed of rotation to be abso- 
lutely constant, in other words, the earth's rotation 
on its axis should constitute a perfect clock. This 
is indeed so; it is far more perfect a clock than any 
mechanical or electrical apparatus human intelli- 
gence has yet been able to devise. To a scientist, 
however, perfection is impossible to attain, and 
through an extremely long series of observations, 
carried to the limit of accuracy, astronomers have 
been able to discover that this "earth-clock" docs 



have its small imperfections. By comparing it with 
clocks based on the motion of the moon, on that of 
Mercury, and of the moons of Jupiter, slight irregu- 
larities in the behaviour of mother earth have been 
detected. Although we have no way of knowing if 
any of these other "clocks" are any more perfect 
than ours, we have to trust their larger number 
rather than their individual accuracy, for it would 
be absurd to suppose that all these other clocks 
were slow by the same amount at exactly the same 
time, and fast at some other time. The reason for 
this inconsistency in our rotational clock lies in the 
fact that the earth has not yet "set" completely. 
It is still contracting and expanding at intervals, 
and each slight change will either accelerate or 
retard the speed of rotation. The amounts of these 
changes, it should be added, are so small that they 
are of interest only to the astronomer. At present 
our day is getting consistently longer by about one 
thousandth of a second per century, while the total 
cumulative error of our clock due to the erratic 
changes will never be greater than twenty seconds 
at a time. 

Another consequence of the oscillations of the 
crust of the earth is that the axis of rotation, 
although keeping the same direction in space, moves 
within the body of the earth. The poles therefore 
seem to wander over the surface; these wanderings 
can be very easily detected with our modern meth- 
ods of observation, although, actually, they are 



confined within a space smaller than the hall of 
Grand Central Terminal in New York City. 

The second of the conspicuous motions of the 
earth is its motion around the sun. We have seen 
before how Newton's law of gravitation demands 
that every planet describe an elliptic path around 
the sun, with the sun in one of the foci. In the case 
of the earth the orbit is very nearly a circle, with 
the sun practically in the centre, and the distance 
between us and the sun varies but little, reaching 
its smallest value of 91.5 million miles in January, 
and its largest, 94.5 million miles, in July. As in 
the case of the earth's rotation on its axis, there 
is a direct proof in the present instance that it is 
the earth which is revolving around the sun, and 
not the sun around the earth, but this proof is based 
on the principle of the aberration and its explana- 
tion falls outside the scope of this book. 

In our annual path around the sun, the axis about 
which the earth rotates remains parallel to the same 
direction in space. Since this axis is not perpendicu- 
lar to the plane of the orbit, that is the plane of the 
ecliptic, each pole of the earth is alternately inclined 
toward the sun, or turned away from it. On 
June 21st the north pole is tipped toward the sun 
by its maximum angle, and receives sunlight during 
the entire twenty-four hours of the day, while the 
south pole is entirely hidden from the sun's rays. 
On December 22d the situation is reversed; at the 
intermediate epochs of March 21st and Septem- 



ber 23d, the axis is inclined sideways, and both 
poles are equally illuminated, for both poles the 
sun is just visible on the horizon. As a result the 
day at the north pole lasts from March 21st until 
September 23d, the night from September 23d to 
March 21st. 

At the equator, where every point is carried 
through the plane of the ecliptic twice a day, night 
and day are always of equal length, and there are 
no seasons. The angle between the axis of the earth 
and the perpendicular to the plane of the ecliptic, 
which angle is equal to that between the ecliptic and 
the equator, is 23 1 /- degrees. If, then, we select a 
place on earth 23 l /z degrees away from the north 
pole, we see that on December 22d, when the north 
pole of the axis is tipped away from the sun by 
exactly that angle, the sun will no more than graze 
the horizon there; it will not really rise. Or> 
June 22d, on the other hand, the north pole is tipped 
toward the sun, and the sun, although it again 
grazes the horizon at our place of observation, is 
constantly above the horizon; it never sets. The 
circle drawn around the north pole at a distance of 
23 Y> degrees therefore marks the limit of the 
"midnight sun," the period during which the sun 
does not set, increasing steadily from one night at 
the arctic circle itself to half a year at the pole. 
The phenomenon of the midnight sun is so strange 
that it is difficult to conceive an impression of it, 
without having seen it. In the regions north of the 





arctic circle the sun may be seen crawling along the 
horizon, sinking into it almost imperceptibly, then 
rising again, and during the course of several hours 
around midnight the sun seems to follow along the 
horizon rather than move upward or downward. 

We have stated that in its annual motion around 
the sun the earth kept its axis of rotation always 
parallel in space, but this is not strictly true. The 
earth, not being a perfect sphere but somewhat 
flattened, is subject to an extra attraction from the 
sun and the moon which tends to bring the planes 
of the equator and of the ecliptic into coincidence. 
It would long ago have succeeded were it not for 
the fact that the earth is rotating on its axis; the 
effect now produced is similar to that of a spinning 
top, rotating rapidly while its axis is being attracted 
toward the ground. It is a curious feature of the 
spinning top that, where the force of attraction is 
directed downward, its effect is acting sideways. We 
all know what happens in the case of the top; it 
begins to gyrate, that is, while it continues to spin 
around on its axis, this axis revolves, describing a 
cone, always making the same angle with the 
ground. Precisely the same thing happens to the 
earth, its axis of rotation gyrates very slowly, mak- 
ing one complete turn in 26,000 years, thus pointing 
to different directions in space but always making 
the same angle with the plane of the ecliptic. The 
direction toward which this axis points will always 
be, to us, the north pole of the sky, and thus we see 


STAR \ pole 


IN 13.000 YEARS 

* - 







that, as a result of the gyration, or precession as 
astronomers call it, the north pole of the heavens 
moves among the stars. Our present pole star will 
not always be that — it will yield its place to others : 
in 6,000 years the star Alpha of the constellation 
Cepheus will guard the north pole, in 12,000 years 
Vega, the brightest star of the northern sky, will 
shine at the top of the vault of heaven, as it did 
14,000 years ago, while in both instances Canopus, 
second only to Sirius in glory, hovered near the 
south pole. At the same time the part of the sky 
visible from any one spot on earth will change, new 
constellations will rise, and familiar ones disappear. 
The Southern Cross was visible all over the United 
States 6,000 years ago and will be so again in 20,000 
years, while Sirius, the Dog Star, at present the 
glory of our winter nights, will disappear below the 
southern horizon 13,000 years hence. 

While this is the principal motion of our pole, it 
is not the only one. The moon, revolving around 
the earth in an orbit which itself rotates, produces 
small oscillations in the precession of the pole. In 
addition, the moon displaces the earth to and fro 
out of its orbit in a period of one month, the angle 
between the plane of the ecliptic and that of the 
equator changes slowly, the whole of the earth's 
orbit is slowly and irregularly changed by the action 
of the planets. In this way we could go on almost 
indefinitely; with every refinement in observation 
and calculation a further source of change in the 




position of the earth would manifest itself, until at 
length we should find ourselves with a combination 
of movements of such bewildering complexity as to 
defy understanding. However, it seems sufficient to 
indicate this type of involved mathematical calcula- 
tion as an illustration of the intricacies of the prob- 
lem, and of the hair's-breadth accuracy with which 
the modern astronomer analyzes the motions of the 
heavenly bodies. As such the problem is one rather 
for the advanced astronomer, and need not detain 
us here in our more general survey of the earth. 

Chapter HI 

"At whose sight, like the sun, 
All others with diminished lustre shone." 

— Cicero. 

From time immemorial the sun has been worshipped 
as the ruler of the sky, as the source of light and 
heat, as the originator and preserver of life, the 
symbol of ultimate and immaculate purity. Science, 
which has shattered so many idols of the past, has 
not only left the sun unmolested but has even exalted 
its significance. From a mere attendant of the 
earth, created for the benefit of the human race, 
the sun has been shown by astronomy to be the 
central and dominant body of the planetary system, 
dominating not only by virtue of its great mass, 
which forces all other objects in its vicinity to obey 
its will, but also because it is the only one that leads 
an independent existence, the only one shining by its 
own light. 

Naturally, one of the first questions that comes to 
the mind is: how far away is the sun? Difficulties 
present themselves immediately when the astron- 




omer tries to answer it. In space distances are 
measured by much the same methods which the 
surveyor uses in mapping out a piece of land, by 
selecting a base-line, the length of which is known 
with great accuracy, and by further measuring the 
directions of the various landmarks against certain 
fixed points on the horizon. Unfortunately, this 
method cannot give good results in the case of the 
sun, for when the sun is visible we can see no stars 
that could serve as fixed directions in the sky, and 
furthermore, the longest baseline available is at the 
most as long as the diameter of the earth, a paltry 
8,000 miles. The astronomer, however, is resource- 
ful, and takes to indirect means when direct meth- 
ods fail him. Instead of measuring the distance 
from the earth to the sun directly, he first draws 
a map of the whole solar system, to scale, in which 
all distances are relatively correct, although not 
one of them is known in miles. He then selects the 
one distance which, for the time being, is the one 
most easily measurable, and in this way determines 
the scale of his picture and, indirectly, the distance 
sun-earth. From a variety of measures it has been 
found that this distance, which because of its great 
importance, in astronomy, is called the astrotwmical 
unit, equals : 

92,870,000 miles. 
To give a better impression of the immensity of 
this distance, we might add that it would take an 
automobile, going at the rate of sixty miles per hour, 



175 years to reach the sun. Even a ray of light 
that can travel seven times around the earth in one 
second, requires 8 minutes and 19 seconds to bridge 
the gap between us and the sun. 

From the sun's distance and from the size it 
appears to have as observed from the earth, we can 
at once calculate its actual size. It appears a trifle 
larger than half a degree in diameter, or about as 
large as a silver dollar at a distance of 14 feet; at a 
distance of 93,000,000 miles, half a degree corre- 
sponds to 864,000 miles, and this then is the diame- 
ter of the sun. Since the earth has a diameter of 
7,900 miles, the sun is 109 times as large in diame- 
ter, 12,000 times larger in surface, and 1,280,000 
larger in volume! If we represent the sun by a 
sphere, the size of a football, the earth would be 
no bigger than a large grain of buckshot, one tenth 
of an inch in size, and it would revolve about this 
football at a distance of more than 100 feet. An- 
other way of illustrating the sun's enormous size is 
by stating that if the earth were exactly in the centre 
of the sun, the moon could still revolve around us 
at its distance of 240,000 miles and the entire opera- 
tion be confined to the inner core of the sun; the 
moon would be only a little more than half way out 
to the surface. 

As in the case of the earth, we cannot, obviously, 
weigh the sun directly, and we must calculate its 
mass from a comparison of the time it takes the 
earth to go round the sun, with the speed a stone 



acquires during the first second of its falling to the 
ground: the same law of gravitation controls both 
motions. We find that the sun weighs 328,000 times 
as much as the earth, or, when expressed in tons, 
a 2 with 27 ciphers. From the sun's volume we then 
deduce that on the average the sun is one and one- 
half times as dense as water, and that, on its surface, 
the force of attraction is 28 times that on earth. 
A stone would fall 450 feet in the first second, and 
a mass of 70 pounds would weigh a ton on the sun. 
When we turn a telescope on the sun, and study 
it in more detail, we notice that the surface, the 
emblem of purity for the ancients, is no longer 
immaculate, but that from time to time dark, 
irregularly shaped spots appear. These are known 
as sun-spots, and were first discovered by Scheiner, 
a Jesuit priest of Ingolstadt, Germany, and later 
confirmed by Fabricius in Holland, and Galileo in 
Italy. It seems, however, that the Chinese, although 
probably having no telescopes at their disposal, 
anticipated these discoveries, for the ancient chron- 
icles testify that they observed a number of spots 
in the years from 300 to 1200 A. D. Seen through 
a large telescope, a sun-spot appears as a jet-black, 
irregularly shaped object, the umbra, surrounded 
by a grayish ring of cloudy appearance, the pen- 
umbra. In size the central core of the spot may 
vary from 500 miles to as much as 50,000 miles, 
or more than six times the diameter of the earth, 



while the penumbra surrounding a group of spots 
may run up to 150,000 miles in diameter. 

When sun-spots are observed from day to day 
it is soon noticed that they move from east to west 
across the sun's disk; this is simply a result of the 
rotation of the sun on its axis. The speed with 
which sun-spots move then forms a good means of 
determining the speed of the sun's rotation, pro- 
vided we are sure that the spots remain fixed on 
the surface. Unfortunately, this is not so. By the 
time a spot has crossed from the eastern limb of 
the sun to the western, it may actually have moved 
on the surface itself, for we are, after all, not stand- 
ing on rock bottom when we are dealing with sun- 
spots, but on a turbulent vortex of exceedingly hot 
gases. One peculiar thing about the sun's rotation 
is quite apparent; the sun does not rotate as a solid 
body, such as the earth. Here everybody, from pole 
to equator, is carried around with the earth once 
in twenty-four hours. On the sun, on the other 
hand, a spot near the equator takes 25 days to go 
round, while one near the pole would take almost 
30. If we traced a straight line on the sun's surface 
from pole to pole, it would not remain straight, but 
look a corkscrew at the end of a year. 

On the whole, sun-spots are short-lived, many of 
them lasting but a single day, and often the sun is 
entirely devoid of them. There is a certain regu- 
larity in their appearance; some years the sun seems 
covered with them, at other times we may scan the 


sun's surface for days without seeing even one. 
They come and go in an eleven-year period; in 1917 
their number was very large, while throughout 
1923-1924 only very few were observed; 1929 will 
again see a maximum number of spots. The exact 
cause of this periodicity in the appearance of sun- 
spots is not yet known, although it has become 
increasingly evident, especially through the work of 
Hale at Mt. Wilson, that their origin must be deep- 
hidden, probably situated at considerable depth 
below the sun's surface. 

A sun-spot is in reality a cylinder of hot gases 
shot up from the sun's interior and arriving at the 
surface while revolving at great speed. Natu- 
rally, when coming into the outer layers of the sun's 
atmosphere where the pressure of the gases is much 
lower, this whirlpool will expand and cool off. By 
cooling off it decreases in brilliance, and thus appears 
black, simply by contrast with the dazzlingly bril- 
liant surface of the sun, although a sun-spot really 
is hotter than molten steel. 

To an observer on the earth this is not all that 
sun-spots mean for him. They have a more tangible 
interest. Their influence is strong enough to reach 
across the gap of 93,000,000 miles and produce 
severe magnetic storms on the earth, or a mag- 
nificent display of northern lights. The connection 
between sun-spots and magnetism on earth has been 
proved beyond question through a comparison, made 
over a large number of years, of the number of sun- 

Pko'.owapkj Mount Wilson Observatory 


Above: \ typical sun-S|iot. 

BtloU): 1'lie sun, photographed in hydrogen light, showing sun-spots, facular, and currents 
on the surface of the sun. 



spots and the daily irregularities of the needle of 
the compass. Every time there is an abundance of 
sun-spots, disturbances of the compass are also fre- 
quent, while at the times when sun-spots are few and 
far between, the magnetic needle is quiet and regu- 
lar. The reason seems to be that sun-spots them- 
selves are strong magnets, and consequently appear 
always in pairs, though they may not always be 
visible as such. There are such things as invisible 
sun-spots, that only betray their presence when 
their light is examined in the spectroscope. Natu- 
rally, if a sun-spot is a magnet, we may expect that 
occasionally a very large and powerful spot may 
make its influence felt to a considerable extent all 
over the earth. Witness the occurrence of northern 
lights (aurora borealis), which are nothing but 
manifestations from our upper atmosphere that it 
is receiving powerful electric blows from the sun. 
A large sun-spot almost invariably produces a strong 
display of northern lights about twenty-five hours 
after it has crossed the centre of the sun's disk, and 
if it is a particularly tenacious spot, it may come 
back again after the sun has completed one revolu- 
tion on its axis, and produce another aurora. This 
happened, e. g., in January, 1926, when the same 
sun-spot came back four times and at its last appear- 
ance was still powerful enough to produce an aurora 
of great splendour, and to put the telegraph and 
telephone cables out of commission for several 



Sun-spots are phenomena of the sun's surface, 
usually called the photosphere, as are also the 
facula, bright streaks and patches of light often 
found to surround spots. Above the photosphere 
lies the sun's atmosphere, in its turn divided into 
two parts, viz., th* reversing layer, composed of 
the vapours of many familiar substances, such as 
iron, calcium, sodium, and having a thickness of a 
few hundred miles, and the chromosphere. The 
reversing layer is so named because, being composed 
of gases at a much lower temperature than those 
in the photosphere, it absorbs a certain amount of 
light, according to the laws of optics, thus produc- 
ing black lines. These lines are discovered when 
the sun's light is analyzed by the spectroscope. 
Above this reversing layer lies the chromosphere, 
composed principally of the lighter elements, such 
as hydrogen, helium, and calcium. It is from the 
chromosphere that the prominences rise, those great 
crimson flames of hydrogen which sometimes reach 
heights of hundreds of thousands of miles. Beyond 
the chromosphere extends the corona, the outermost 
thin, gaseous layer of the sun's atmosphere, thus 
far observed only at times of an eclipse of the sun. 
Both the corona and the prominences are extensively 
studied at such times, and we shall have occasion 
to refer to them later. The corona, though it may 
extend millions of miles outward, does not represent 
the farthermost boundary. Beyond it comes the 
zodiacal light. This may be seen on any clear, 



moonless evening in spring as a faint glimmer of 
twilight extending upward along the ecliptic. In the 
tropics, or in Japan, when conditions are particu- 
larly favourable, the zodiacal light can be followed 
around the entire ecliptic, and there is but little 
doubt that it is caused by reflection of sunlight 
from a vast number of exceedingly small "dust" 
particles which seem to fill up all the space inside 
the earth's orbit, and even extend beyond it. From 
the total amount of illumination in this zodiacal 
light it can be calculated how many and how large 
these dust particles must be. We find that if the 
whole space between the sun and the earth is filled 
with particles one twenty-fifth of an inch in size, 
and twenty-five miles apart, the total light would 
be more than accounted for. 

When the spectroscope came into use in astron- 
omy, it became possible to study, not only the 
appearance of the sun's surface, but also its physical 
and chemical constitution, its temperature, and the 
behaviour of the gases in the surrounding atmos- 
phere. While the observations stop at the surface of 
the sun, our minds can push farther, and with the aid 
of modern physical and chemical theories, and with 
what we know of the behaviour of atoms, we can 
penetrate deeply into the interior of the sun, where 
all its strange secrets are kept — "pluck out the heart 
of its mystery." 

When sunlight is made to pass through a prism, 
a piece of glass with two edges not parallel to each 



other, the light rays are deflected from their path. 
Rays of different colours are deflected by different 
amounts: blue light more than green, green more 
than yellow, and red light least of all. The result 
is that where a beam of white sunlight enters the 
prism, a band of coloured light emerges. On this 
rests the principle of the spectroscope. All we have 
to do is to put a photographic plate in this coloured 
band, and record the sun's spectrum. One of the 
first uses the observations of the sun's spectrum are 
put to is the determination of the temperature of 
the surface where the light comes from. Physicists 
have calculated that bodies of different temperatures 
give us spectra with different proportions of coloured 
light, and conversely, if we can measure the amount 
of heat contained in the green rays coming from 
the sun, as compared to the heat in the red or violet 
rays, we can tell the temperature of the sun. This 
has been done by Wilsing in Potsdam, and by Abbott 
of the Smithsonian Institution. They agree in put- 
ting the sun's temperature between 5,500 and 6,000 
degrees centigrade, or about 10,000 Fahrenheit. 
Stating such a temperature as a mere number does 
not convey much meaning; perhaps a practical illus- 
tration is more to the point: 

An interesting experiment was once performed 
by Langley in the steel works at Pittsburgh. As the 
molten steel cascaded forth from the Bessemer con- 
verter, lighting up the great room with a million 



Roman candles, the intensity of its menacing glare 
was compared with that of the sun taken over an 
equal surface. It was found that the heat radiation 
of the sun was 87 times greater, while in light-giving 
power the sun surpassed the molten steel 5,000 

When the spectrum of the sun is analyzed more 
minutely, it is seen that it does not simply consist of 
a band of coloured light. On the contrary, this band 
is crossed by many dark lines, and these lines become 
more numerous as our apparatus for examination 
becomes more refined. We now know that each of 
these lines indicates the presence of a certain chem- 
ical element on the sun. The two strong lines far 
in the violet prove beyond question, for example, 
that there must be a considerable amount of calcium 
on the sun. A whole series of black lines running 
from the red to beyond the violet indicates the 
presence of hydrogen — lightest of all gases. 

In the chromosphere the situation is slightly dif- 
ferent; here the gases are observed, during a total 
eclipse of the sun, without any luminous background, 
and their characteristic rays of coloured light appear 
bright, instead of dark. One instance of identifi- 
cation of lines in the spectrum with a chemical ele- 
ment must be mentioned specifically: it concerns a 
bright yellow line seen in the spectrum of the 
chromosphere. There was no chemical substance 
known on earth that could produce a line in that 
exact place in the spectrum, therefore it was attrib- 



uted to a new element, helium (so named after the 
Greek for sun). Later on, helium was found by 
Ramsay in the atmosphere of our own earth, and 
proved itself to be one of the most useful substances, 
lighter than any other gas except hydrogen, and non- 
inflammable; an ideal filling material for airships. 
To realize fully the splendour of our sun we 
should obtain an idea of its prodigious output of 
heat, and to do this we must first go through some 
more cold analyses of scientific facts. Physicists have 
adopted as their unit of heat the calorie, the amount 
of heat which will raise the temperature of one gram 
of water by one degree centigrade. Accurate meas- 
ures of the sun's heat, carried out by the Smithson- 
ian Institution, have shown that every square inch of 
the earth's surface receives from the sun 12 calories 
every minute — and this at a distance of 93,000,000 
miles I Therefore, the total amount of heat given 
off by the sun in one minute is equal to 12 calories 
multiplied by the number of square inches contained 
in the surface of a sphere with a radius equal to 
that of the earth's orbit: 93,000,000 miles. In this 
way we find that the total energy output of the sun 
exceeds half a million million million million horse- 
power, enough to melt a cube of ice 2,500 miles 
long, 2,500 miles wide, and 2,500 miles high every 
minute, and equal in one year to the heat produced 
by four hundred sextillion tons of the best anthra- 
cite coal. Before such staggering figures even the 
brain of a scientist reels, and we can but look up in 



admiring wonder at this immense celestial furnace, 
to whose liberal expenditure of light, heat, and 
energy we owe our very existence. 

How is it possible, one may ask, that the sun can 
go on dissipating energy at this rate, without burn- 
ing out? The earth, we are told, has existed for 
well-nigh a billion years, and the sun must have been 
shining in full glory for a much longer time. How 
does it make up for the enormous loss of energy? 
Confronted with this problem the Mohammedan 
would exclaim: "Allah is great." The scientist will 
add: "True enough, but just how does he do it?" 
The first theory we had concerning this problem 
came from Helmholtz, who pointed out that, as the 
sun shrinks, it heats up because the gases in the 
outer atmosphere are falling in toward the centre. 
Unfortunately, however, later calculations showed 
that even if the sun had once been as large as the 
orbit of the planet Neptune, or 5,000,000,000 miles 
in diameter, it would have shrunk to its present size 
in less than 25,000,000 years. We have seen that 
the earth alone must have lived a great deal longer 
than that; consequently, Helmholtz's theory must 
be abandoned. 

Again atomic physics has provided us with a 
closer-fitting explanation. Let us suppose that the 
interior of the sun, which is now believed to be 
several million degrees in temperature, is hot enough 
to effect a transformation of hydrogen into helium. 
In this process a certain amount of matter would 



go to waste, since out of four pounds and one-half 
ounce of hydrogen we should get no more than an 
even four pounds of helium. Half an ounce of 
matter has disappeared, has "gone up in smoke." 
Not in "smoke," says the theory of relativity: this 
excess matter has simply been transformed into 
energy; heat and light, in other words. And 
"half an ounce of light" is a good deal, equal 
to about three billion kilowatts! Of course, 
we must remember that all this is a mere 
conjecture, we cannot place a thermometer in the 
inside of the sun and measure its temperature; we 
can use only our mathematical formulas for this pur- 
pose. Nevertheless, we should not forget that the 
present theory is the only key to the mystery that 
we can now see; it may not be the correct one, but 
at least it explains satisfactorily how the sun has 
been able to live as long as it has, and it gives us a 
feeling of relief that, if the sun is really capable of 
doing this trick, it has obtained a new lease of life, 
and can go on shining for several billions and 
possibly trillions of years. 

On the other hand, even though the sun has been 
granted a reprieve, the day of reckoning has only 
been postponed; the sun cannot escape its destiny. 
While our present-day theories may have succeeded 
in slowing up the grinding process of the mills of 
the gods, and consequently in lengthening the span 
of life of our sun, nemesis is bound to come. Ulti- 


mately the sun will cool off and die. Life, such as 
we now know it, will long before have ceased to 
exist, though the planets themselves, like wandering 
tombs, will still be revolving around the sun in the 
oppressive silence of eternal night: spectres of the 
past, but perhaps — cradles of the future. 

Chapter IF 

"Whilst she, in the vault of heaven, 
Moves with silent, peaceful motion." 
— Heine. 

The moon is as old as the earth. It has been plough- 
ing its way through space and has been a constant at- 
tendant of the earth for millions of years past. Its 
soft and silvery light had brightened our terrestrial 
nights long before there were human beings on earth 
to appreciate it. In all probability it was in the moon 
and its motion among the stars that astronomy and 
science found their first inspiration. From the moon 
the first calendars were made; on the moon and on 
its tide-raising effect the early seafarers depended. 
The sole reason for the high esteem in which the 
moon is held, is its nearness, not its inherent impor- 
tance, for in space the moon is an insignificant 
object. In size it is no more than 2,160 miles in 
diameter, less than one third of the earth ; in volume 
it is almost fifty times smaller, and in weight it is 
less than one eightieth that of our own planet. 
Of all the heavenly bodies the moon is the nearest 




to us, its mean distance from the earth being 240,000 
miles, or only thirty times the diameter of the earth. 
We have to say "mean distance" here, for in its 
motion around the earth in 27 days, 7 hours, 43 
minutes the moon does not always stay at the same 
distance. Like the earth in its path around the 
sun, the moon follows an ellipse, but a much more 
elongated one than that of the earth. While re- 
volving in this ellipse, the moon rotates on its axis 
in exactly the same time, thus always presenting the 
same face to the earth. This is one of the tragedies 
about the moon: we have never seen the other side 
of it. It is no whim of chance that this is so, but a 
direct result of the way in which the moon was born, 
as Darwin has shown. When, under the attraction 
of the sun, moon and earth separated, they were 
both still in a liquid state, or at least in a state re- 
sembling a syrup in consistency. The moon, as we 
all know, is the cause of the tides on earth, but, con- 
versely, the earth would, by reason of its greater 
attraction produce greater tides on the moon when 
the moon possessed oceans. At present the moon is 
all solid rock, but in the first few million years after 
its birth the moon was wholly liquid, and the earth 
raised huge tidal waves on it. Thus, while the moon 
was rotating on its own axis and revolving about the 
earth, the latter persisted in producing "high water," 
both in those places that were nearer and those fur- 
ther away. The moon, therefore, was never round, 
out always longer in the direction toward the earth' 



and when it cooled off enough for parts of the sur* 
face to solidify, the solid part still tried to rotate on 
an axis in its own time, while the tides were running 
over the surface in the time demanded by the earth. 
Naturally, this caused great friction; the more the 
moon solidified, the more this friction increased 
until, finally, it became so great that the moon gave 
up the struggle, and meekly followed the time of 
rotation imposed upon it by the earth: twenty-seven 
days. Thus, for ages past, the moon has bowed 
to the superior will of the earth and always rotated 
on its axis in exactly the same time as it takes to 
revolve around us. To us on the earth this gives 
the appearance that the moon does not rotate at all, 
but this conclusion is a fallacy. It is much the same 
as when a man walks around a city square, always 
keeping his eye on a statue in the centre. Any ob- 
server standing near the statue would always see 
the pedestrian's face, and to him the latter would 
not seem to be turning. Yet any third person, stand- 
ing outside the square, would immediately notice 
that in his walk around the square our pedestrian 
must face all directions of the compass, therefore 
he must be turning around — a rotation which con- 
sumes as much time as one complete turn around 
the square. 

In the case of the moon, we do see a little more 
than half; this is owing to the fact that the moon's 
orbit is not entirely circular, and to certain slight 
irregularities in its motion. We can thus see about 



59 per cent, of the moon's surface, the remainder 
being forever hidden from our view. 


The moon is a cold and dark body, it gives no 
light of its own, and we can see it only because it is 
illuminated by the sun. That side alone of the 



moon which is turned toward the sun is visible, there- 
fore, how much we see of it then depends on the 
moon's position relative to the sun and the earth. 
When it is between us and the sun, we see nothing; 
it is new moon. When it is on the side opposite the 
sun, we see the whole disk illuminated; it is full 
moon. At all intermediate positions of the moon 
we see only part of its disk illuminated : a crescent 
when it is a small part, shortly before, or shortly 
after the new moon ; or gibbous in appearance near 
the times of full moon; while at the times halfway 
between new moon and full moon exactly half the 
circle appears bright, the other half black. These 
are the first and last quarters. The moon is not a 
good mirror; it reflects only 7 per cent, of the inci- 
dent sunlight. At full moon its light is 450,000 
times fainter than sunlight; at first or last quarter 
it is more than 4,000,000 times fainter. 

In 27 days, 7 hours, 43 minutes the moon com- 
pletes one revolution about the earth ; that is to say, 
as seen from a distant star, it would appear in the 
same position after that length of time. In the 
meantime, however, the earth itself has not been 
idle, and has advanced in its orbit around the sun 
by a considerable extent, with the result that the 
moon still has some distance to travel before it is 
again in the same position relative to both the sun 
and the earth. In our illustration above, this would 
correspond to the statue in the middle being put on 
a turntable and rotated very slowly. Suppose the 



pedestrian started out facing north, and walked 
westward. By the time he was facing north again, 
he would have made one complete turn around the 
statue, as an outsider would see it. By that time, 
however, the statue would have turned enough to 
make the pedestrian go through till northwest be- 
fore he would be in the same position relative to the 
statue. Seen from the stars, the moon revolves about 
the earth in 27 days, 7 hours, 43 minutes, but it 
takes more than two days longer, 29 days, 12 hours, 
44 minutes, to cover the distance between one full 
moon and the next. Astronomers call the first time 
the sidereal period; the second, the synodic; it is this 
latter which is the basis of our month. 

One thing on which we should lay particular 
stress is that the moon, while describing an ellipse 
around the earth, which itself revolves about the 
sun, follows a path that is always concave toward 
the sun. If we represent the orbit of the earth 
around the sun by a circle of seven feet in diameter, 
the moon would never deviate from this circle by 
more than one tenth of an inch on either side. 

The moon's motion around the earth and the sun 
is only a first approximation; its exact motion is very 
complicated, as is that of the earth. In its orbit 
around the earth the moon describes an ellipse, and 
the longest dimension of this ellipse, the major axis, 
rotates in space, the shape of the ellipse also chang- 
ing slightly. Next, we know that the plane of this 
ellipse makes an angle with the plane of the ecliptic, 



and this angle varies somewhat, not only in amount, 
but in direction also, the moon being sometimes 
between the ecliptic and the equator, sometimes be- 
tween the ecliptic and the pole. In this way, we 
could continue ad infinitum, for the number of intri- 
cacies in the motion of the moon is legion, making 
the whole problem resemble a "blindfolded chess 
game in three dimensions," as Professor Brown so 
aptly puts it. A fortunate occurrence, since this very 
complexity has caused the mathematical theory of 
such celestial motions to be thoroughly analyzed — 
a stupendous task, and one with which many of the 
great names in celestial mechanics are forever asso- 
ciated: Newton, Delaunay, Hansen, Poincare, New- 
comb, Brown, to mention but a few. 

The weight of the moon may be found by various 
laborious processes, all of which involve a compari- 
son of the respective attractive forces of the moon 
and the earth. The answer given by all of them is 
that the moon wefghs less than 1/81 part of the 
earth. We know that the moon's volume is 49 
times less than that of the earth; it follows that, 
volume for volume, the moon must be lighter than 
the earth and composed of materials that weigh on 
the average only three times as much as water 
instead of more than five times as much, as is the 
case of the earth. Thus, the moon weighs exactly 
the same as if it were composed of rocks similar to 
those we find immediately below the surface of the 
earth, which fits in well with the theory that the 



moon came off the earth at the time everything was 
in a semi-liquid state. In that case the lighter ma- 
terials would come off, while the heavier inner core 
of iron would remain with the earth itself. 

The moon is so near that even the naked eye can 
see some detail on its surface, while an opera glass 
shows the darker areas in contrast to the whiter 
areas quite plainly. A small telescope, even as small 
as the first used by Galileo, reveals the mountains 
and the great plains and valleys; more powerful 
telescopes will show a wealth of detail. The naked 
eye sees the moon at a distance of 240,000 miles, 
or as large as a silver dollar at a distance of ten 
feet; the great telescopes on the Pacific coast bring 
it as close as 100 miles. With the 100-inch telescope 
at Mt. Wilson direct photographs of the moon can 
be taken which will show it with a diameter of 30 
inches, or on a scale of 1 : 20,000,000. Such a pho- 
tograph would stand considerable enlargement, and 
by means of such an enlargement we should be able 
to see things two miles in size. If, under favourable 
conditions, we looked through the telescope directly, 
we might see shadows cast by objects as large as the 
Great Pyramids or the Woolworth Building. 

The more prominent mountain ranges, such as 
the Apennines, and the conspicuous circular craters 
of the type of Copernicus, Plato, Archimedes, are 
undoubtedly familiar to all who have ever observed 
the moon through a telescope of any size; we need 
not here fatigue the reader with an enumeration 



of the mountains of the moon, which would simply 
be a catalogue of astronomers of the past. Lunar 
topography has offered unparallelled opportunities 
for immortality to astronomers who were of suffi- 
cient mediocrity to escape remembrance on earth; 
even after the great names of ancient and mediaeval 
astronomers had been exhausted on the main points 
of interest, there was always a crater vacant on the 
moon, in desperate need of an insignificant name. 

The mountains on the moon are relatively much 
higher than those on earth, for against a diameter 
of the moon, which is only slightly more than one 
fourth of that of the earth, the highest mountain 
on the moon is more than 22,000 feet high, not 
much inferior to Mt. Everest. The most curious 
feature of the lunar mountains is their shape: most 
of them consist of a high circular wall sometimes 
as much as 100 miles in diameter, with a small cone 
rising in the centre, not unlike some volcanic struc- 
tures on earth. The two chief theories in the field 
concerning the origin of these craters are that they 
are caused by volcanic eruptions or by the impact 
of a swarm of meteors. At present the available 
evidence rather appears to favour the second alter- 

When describing the large, dark areas on the 

moon the early astronomers chose names such as 

. Ocean of Tempests, Ocean of Tranquillity, Sea of 

Serenity, etc., simply supposing from analogy with 

the earth that these areas must contain water. We 


Tin- northern portion of die moon at last quarter, photo- 
graphed with the ioo-inch reflector, and shown as it is seen 
in an ordinary, inverting telescope, with north pointing 
down. The large, flat ring crater near the bottom is 
Plato, about 7,000 feet high, the large, flat crater near the 
centre, Archimedes, just one mile high, and 60 miles in 
diameter. Of the two craters to the left of Archimedes, the 
smaller and higher is Aricrillus, over 10,000 feet high. The 
mountain range toward the left, and above, where the sun 
is just setting, is called the Apennines. 


know now that there is no appreciable amount of 
water on the moon, neither is there any atmosphere. 
A body as small as the moon, with a weight less 
than one eightieth that of the earth, cannot retain 
any gases on its surface; the molecules of a gas are 
in too violent a motion. Flying haphazardly in all 
directions, these molecules soon escape to the voids 
of space, and all that the moon can retain are a few 
of the lazier particles of the heavier gases and a 
little water vapour. The direct consequence of the 
absence of an atmosphere on the moon is the intense 
contrast it lends to the lunar landscape, making it 
much sharper, more beautiful, and majestic, but at 
the same time giving it a grim and savage, almost 
cruel, aspect, entirely devoid of the softness that 
is so typical of our terrestrial surroundings. Let 
us try and imagine the view of an observer who, 
endowed with the same senses as ourselves, could 
yet live in the great ring crater Copernicus. 

The horizon would be much nearer, owing to the 
fact that the moon is so much smaller than the 
earth, and much more curved therefore. It might 
even be possible to be in the centre of one of the 
great ring craters, and not be able to see the walls 
at all. For us, who are observing from the earth, 
the ring craters form the most typical features of 
the lunar landscape. For inhabitants of the moon 
they might be very hard to recognize. 

To an observer in the centre of Copernicus the 
earth would appear very nearly overhead, and, 




except for a slight swaying to and fro, it would 
remain there. Thirteen times larger than the full 
moon appears to us, the earth would go through 
all its phases in 29 days, from "new earth," when 
it would shine with but a pale, bluish lustre to "full 
earth," forty times more brilliant than the full 
moon on earth. In addition, the earth would be 
seen to rotate on its axis, and the continents and 
oceans, Europe, Asia, and Africa, the Atlantic, the 
two Americas, and the Pacific, would be seen march- 
ing from left to right in a majestic and eternal 
pageant. Our Copernican could watch the sun rise 
and set every day upon the various parts of the 
earth. He could see giant storm clouds gather over 
the Mississippi Valley, he could discriminate be- 
tween the bright yellow Sahara and the tropical 
forests of the Dark Continent and of South Amer- 
ica. He would observe the straight line of the 
Andes, the small, compact block of the Alps, and 
the snow-covered Himalayas with the towering 
Mt. Everest. 

For two weeks he would see the sun; for another 
two weeks he would be plunged into semi-darkness, 
receiving light only from the earth. Sunrise and 
sunset would each consume a whole day, and in the 
absence of an atmosphere they would be devoid of 
all colour. On the moon there never is a rosy dawn, 
never a golden sunset, nor even an azure sky. Every- 
thing is perfectly black, but exceedingly transparent, 
giving a spectacle so magnificent that we on earth 


have nothing to compare it with. The sun, shining 
brilliantly in a jet-black sky, with the stars visible 
up to its very edge. The solar corona which astron- 
omers on earth have to travel thousands of miles 
to see, and which no human can ever see for more 
'than an hour during a lifetime, can be enjoyed on 
the moon for fifteen days at a time. 

Though in its absence of an atmosphere the moon 
can have no aurora borealis, it always has the zo- 
diacal light, a long oval of light surrounding the 
sun on all sides. Imagine the beauty of the Milky 
Way under those circumstances; its brightest patches 
might well look as brilliant as a white cloud in the 
moonlight in our sky. Thousands and probably tens 
of thousands more stars could be seen on the moon 
than on earth. 

Last, but not least, comes a total eclipse of the 
sun, by the earth. On the moon this is not a matter 
of a few precious minutes or even seconds; it lasts 
several hours, during which time all light and heat 
from the sun are completely cut off. Darkness and 
intense cold descend upon the moon as the sun is 
obscured by the great black disk of the earth, four- 
teen times larger than the sun itself. Round the 
earth is a thin ring of rosy light, the earth's at- 
mosphere, and beyond that a few streaks of light of 
the outer corona, and finally, on two opposite sides, 
the zodiacal light. 

The saddest thing about the moon is its complete 
lack of sound; as it has no atmosphere, no sound 



waves can be propagated and the moon is plunged 
into eternal silence. The only music one could 
experience on the moon would be the music of the 
spheres, the song sung by the grand march of the 
heavenly bodies as they rise and set. 

Thus we see the moon through the eyes of science : 
dead, dry, and desolate, unapproachable and for- 
bidding, silent and savage, but still beautiful. 

Chapter V, 

"... a darkness which may be felt." 

— Exodus: 8, 21. 

In the whole realm of Nature there is probably 
no spectacle of such awesome solemnity, nor one 
which leaves so unforgettable an impression, as a 
total eclipse of the sun. Those who have not seen 
a total eclipse can not truly say that they have run 
the whole gamut of human emotions. A total 
eclipse is one of those events that render one ob- 
livious to material surroundings and thrust one 
beyond temporal considerations — it is something 
before which even the most lethargic mind becomes 
fervently enthusiastic. 

In the full light of day the sun suddenly seems 
attacked by a deadly shadow, creeping on it from 
the west, advancing slowly but inexorably. With 
the gradual dimming of light, colours begin to fade, 
a menacing gloom takes possession of everything, 
and, as the darkness grows deeper and deeper, an 
oppressive silence sets in. The spectator becomes 
mentally chilled, he feels the vast and palpable pres- 




ence of something cataclysmic overwhelming the 
world: the darkness becomes almost tangible. The 
sun presents a melancholy sight; reduced to the 
thinnest of crescents imaginable, it seems weary and 
ready to give up the struggle against the relentless 
moon. A pale opalescent glow limns the horizon, 
vividly silhouetting the landscape. 

All eyes are intent now, watching for the great 
shadow of the moon approaching with terrifying 
speed, advancing almost as a wall, swift as thought, 
silent as doom. In an instant it is on us, and black- 
est night pervades all. It seems as if the end of 
all things is imminent. But out of this darkness 
there flashes the soft and silvery light of the incom- 
parable corona. Its pearly and unearthly radiance, 
mercifully soft, yet mercilessly intense, makes us 
forget all things terrestrial and think only of the 
universe. In its ethereal glow, its aloofness, this 
halo of immaculate purity seems beyond all human 
conceptions of light and fire; while we are watching 
it, spellbound, it seems as if Time is not. We can 
but gaze at it and admire — silently, reverently, as 
did the ancients, who, taking an eclipse to be a 
manifestation of divine wrath, saw in the corona 
the sign manual of mercy, the symbol of victory in 
defeat. Though as a spectacle it is but short in 
duration, the memory lingers, unforgettable, etched 
in the mind with an accuracy, a vivid exactness, 
which even time can never efface. Totality is over, 
the light returns, and the celestial drama, symphonic 



in its sweep, is ended. The earth, bathed in new 
splendour, appears to revive almost instantaneously; 
the moon's sinister spell is broken. 

In former ages men's hearts trembled with fear 
at the thought that "the sun shall be turned into 
darkness." To-day a total eclipse of the sun must 
make its reckoning with the "unconquerable mind of 
man" as that mind has been strengthened by the de- 
velopment of scientific inquiry. The mechanism of 
a total eclipse was difficult to grasp for primitive 
man, but it appears simple to us if we only consider 
the relations in space of the three bodies, sun, earth, 
and moon. 

All three are approximately spherical, the sun 
self-luminous, the earth and moon dark and opaque, 
casting long, funnel-shaped black shadows behind 
them where the sun's rays cannot penetrate. When, 
in its path around the earth, the moon becomes 
interposed exactly between us and the sun, the peak 
of its shadow falls on earth and hides the sun from 
our view: an eclipse of the sun takes place. When, 
on the other hand, the earth is between the sun and 
the moon, it intercepts the sunlight otherwise falling 
on the moon and causes the moon to be eclipsed. 
It is at once obvious that an eclipse of the sun can 
only happen at new moon, an eclipse of the moon 
only at full moon. That they do not occur at every 
new moon or full moon is due to the fact that the 
moon revolves about the earth in a path which de- 
viates from that followed by the earth in its course 



around the sun. For one half of the year the new 
moon occurs slightly south of the sun, as seen from 
the earth; during the other half the new moon is 
seen north of the sun, and only at two short inter- 
vals during the year does the new moon occur so 
nearly coincident in direction with the sun that an 
eclipse may result. 

One may wonder about the fact that a body as 
small as the moon 2,160 miles in diameter, can 
hide from view a sphere as large as the sun, 864,000 
miles across. The answer to this puzzle is: dis- 
tance. The sun is 93,000,000 miles away, the moon 
no more than 240,000, with the result that, as seen 
from the surface of the earth, the sun and the moon 
appear to be almost exactly the same size, a circum- 
stance which renders an eclipse just possible, though 
it remains a rare phenomenon. Precise calculation 
proves that the length of the moon's shadow varies 
from 228,000 to 236,000 miles, while the distance 
of the new moon from the surface of the earth may 
vary between 218,000 and 248,000 miles, thus giv- 
ing rise to a multiplicity of phenomena. When the 
shadow exceeds in length the moon's distance from 
the earth, the peak of the shadow cone reaches the 
earth, and the eclipse is total; when the reverse is 
true, the shadow falls short and there is no place 
on earth where the moon appears as large as the 
sun, so that at best a ring of light remains around 
the black disk of the moon; the eclipse is annular. 
Even in the most favourable case of a total eclipse 







the shadow has a diameter of only 167 miles, and 
the path of totality, the path swept out on earth by 
the moon's shadow, rarely exceeds 8,000 miles in 
length; thus, any one eclipse is total for but a small 
fraction of the surface of the earth. This explains 
why a total eclipse is such an extremely rare occur- 
rence for any given place although, on the average, 
there is an eclipse of the sun more than once a year 
somewhere on earth. An observer outside the cen- 
tral line of the eclipse will not be within the sweep 
of the actual shadow. He will, so to speak, see 
sideways around the moon and still see a part of the 
sun's disk; for him the eclipse is partial. 

Another consequence of the approximate equality 
in apparent size of the sun and moon is that an 
eclipse cannot last very long. In its motion in the 
sky the moon gains on the sun by an amount equal 
to its own size in about one hour; thus, on earth 
its shadow would sweep 2,100 miles per hour if the 
earth were standing still. Owing to the rotation on 
its axis, the earth imparts to an observer on the sur- 
face near the equator a speed of 1,040 miles per 
hour in the same direction as the moon's shadow is 
travelling; the observer, therefore, would be passed 
by the shadow with a speed of 1,060 miles per hour. 
At this rate, totality can never last longer than 7 
minutes, 40 seconds at the equator, while in higher 
latitudes, where the speed of rotation is diminished 
and the speed of the shadow resultingly increased, 
the maximum duration is reduced; in our latitudes 



totality can last but a little more than 6 minutes 
at maximum. 

For an eclipse of the moon, the situation is en- 
tirely different; here we are dealing with the smaller 
moon being plunged into the shadow of the much 
larger earth. Because of the greater diameter of 
the earth, the shadow cone cast by the earth does 
not taper off as quickly as that of the moon; 
whereas the moon's shadow on the surface of the 
earth can never be larger than 167 miles in diame- 
ter, the shadow of the earth, where it reaches the 
moon, is never smaller than 5,500 miles in width, 
or more than two and one-half times the diameter 
of the moon. We have seen that the moon does 
not follow the same path in the sky as the sun, 
therefore an eclipse of the moon will not happen 
at every full moon, but only when the full moon 
occurs near the places where the two paths cross, 
or twice a year. When a lunar eclipse transpires, 
however, the moon may entirely disappear within 
the shadow of the earth; it may even remain ob- 
scured for one hour and forty minutes. An eclipse 
of the moon is a real eclipse ; the moon receives no 
light from the sun, and, since it is not itself a lumi- 
nous body, it disappears and is invisible not only 
from the earth, but throughout the universe. A 
solar eclipse, on the other hand, is merely an oc- 
cupation of the sun by the moon; no actual change 
in the sun occurs. An observer situated outside 
the earth would see simply a small black dot mov- 



ing rapidly over the surface of the earth. When 
we say that the moon disappears from view during 
a lunar eclipse we must not take this statement too 
literally; it holds true only in theory. It is true that 
no direct sunlight can fall on the moon at such a 
time, but the atmosphere of the earth bends the 
sun's rays into the shadow cone, with the result that 
the moon usually shines with a dark brown tinge 
and only rarely becomes entirely invisible. 

As we have repeatedly remarked, the earth, in 
its path around the sun, and the moon, in its revo- 
lution around the earth, pursue orbits that do not 
lie in the same plane but make an angle of about 5 
degrees. The two points, or rather, to be more 
exact, the directions on the celestial sphere of the 
two points where these two planes intersect are 
called the nodes; eclipses, as we have further seen, 
can take place only when full moon or new moon 
occur at, or near, one of these nodes. The "line 
of nodes," however, is not fixed in space; it rotates 
in such a way that every 18 years, 11 1-3 days the 
sun, the earth, and the moon are in almost exactly 
the same position relative to one another. Once 
an eclipse has occured, therefore, it will be repeated 
again 18 years, 11 1-3 days later, as was already 
known to the Chaldeans in prehistoric times; they 
named this period of repetition the saros. Since 
the saros is not equal to a whole number of days, 
the eclipse will not duplicate itself exactly, but be 
shifted westward on the surface of the earth. In 



this way the total eclipse of the sun that occurred 
in New York on January 24, 1925, will reappear 
on February 4, 1943, but in Siberia, Japan, and 
Alaska. At its next appearance totality will be 
visible in France, Italy, the Balkans, and Russia, on 
February IS, 1961, and the original eclipse will not 
return to New York until 3075 — after sixty-four 
saros periods have elapsed. 

A total eclipse of the sun is visible on but a small 
portion of the earth's surface, and usually most of 
the path of totality falls over the ocean, or passes 
through regions of comparative inaccessibility or 
regions where clear weather is scarce. As a result, 
the number of locations with favorable observing 
prospects is usually very small and may require long 
travels. In making these long journeys to observe 
eclipses astronomers are not prompted chiefly by 
the desire for the privilege not to see the sun for a 
few minutes, but by the opportunity offered for 
making special observations which can be made only 
at the time of a total eclipse — such as photographic 
and spectroscopic observations of the corona, the 
chromosphere, and the "Einstein effect" (the bend- 
ing of light rays near the sun's limb) which consti- 
tutes the most sensitive and most conclusive observa- 
tional test of the theory of relativity. The exist- 
ence of a deflection of light grazing the sun's limb 
in the direction and of the amount demanded by 
relativity was first indicated by the results of the 
British eclipse expedition of 1919 as discussed by 



Dyson and Eddington. The question has now been 
settled to the satisfaction of the majority of astron- 
omers, principally through the conclusive observa- 
tion of Campbell and Trumpler at the eclipse expe- 
dition to Australia in 1922. In the future, there- 
fore, efforts of eclipse observers will follow the path 
mapped out by Mitchell and others and concentrate 
largely on the chromosphere and the corona, con- 
cerning which our knowledge still leaves much to be 

When the light of the corona is examined in a 
spectroscope, there appears a ray of a particular 
yellow-green colour, which cannot be identified with 
light given out by any substance known on earth. 
Yet, the accepted ideas about the constitution of 
matter do not allow the existence of any more gases 
than we already know, and we must find the solu- 
tion of the difficulty by attributing this line to one 
of the known gases, but one which is in a very pe- 
culiar state of electrical excitation. The second 
question is what causes the light of the corona: is 
it all reflected sunlight? and if so, what reflects it? 
The theory that at present seems to hold the field 
is that the coronal light is reflected by free elec- 
trons, particles 25 million million times smaller than 
an inch in diameter and carrying the smallest pos- 
sible amount of electricity. 

A total or partial eclipse of the moon is of much 
less importance, and used principally to determine 
the conditions of the earth's atmosphere, since the 


■ lii'ite: The great solar prominence of May 29, 1919, phorographed by Crommclin, of rhe 
eclipse expedition sent our by the Royal Observatory, Greenwich, ro Sohral, Brazil. The size 
of the earth is indicated by rhe small white Spot at the upper left. 

Below: |"hc solar corona ar rhe rotal eclipse of June 29, 1927. Photograph taken at Jokk- 
. mokk, Lapland) by the eclipse expedition of the Hamburg Observatory, under the direc- 
tion of I'rolessor Schorr. 



colour and brightness of the totally eclipsed moon 
give a clue to the clearness or opaqueness of our 

Within the boundaries of the United States three 
total eclipses of the sun have occurred so far in the 
Twentieth Century — on June 8, 1918, September 
10, 1923, and January 24, 192S. The last-men- 
tioned eclipse passed through a densely populated 
region, extending in width from New York to Provi- 
dence, and was observed and enjoyed by millions of 
people. On April 28, 1930, the moon's shadow will 
barely touch the earth while it is passing through 
California, Nevada, Idaho, and Montana, but total- 
ity will not last more than a second and a half. A 
much more interesting eclipse will occur on August 
31, 1932, when the line of totality runs from north- 
ern Canada, over Montreal, through Vermont, 
New Hampshire, Maine, and Massachusetts, the 
maximum duration being about one minute and 
forty seconds. The eclipse of May 9, 1929, in 
Sumatra will be total for five minutes, that of June 
8, 1937, will have a duration of no less than seven 
minutes, almost equal to the theoretical maximum. 

The prediction and calculation of solar eclipses 
may be done with great precision; practically their 
only source of error is due to uncertainties in the 
position of the moon. Even this source of error, 
however, hardly amounts to more than a second or 
two, and the certainty and accuracy of eclipse pre- 
dictions thus stand as one of the greatest triumphs 



of human intelligence. Despite his material insig- 
nificance in the universe, man is able to foretell the 
event with such accuracy, and describe the phenome- 
non in such detail, that it almost seems as if, in some 
way, he controlled the mechanism of the universe. 
A solar eclipse can no longer launch a surprise at- 
tack as in the elder time; we are waiting for it and 
we censure with fine impatience a second's impunctu- 
ality, as though old chaos were come again. 

Before leaving the subject of eclipses we feel that 
we should again impress upon the reader that the 
sight of a total eclipse of the sun is something to 
look forward to. Its majesty defies prose por- 
trayal, its beauty escapes the photographic plate. 
All the superlatives available in the English lan- 
guage are inadequate to describe even approxi- 
mately the full glory of it, and though the photo- 
graphic plate may record things which the human 
eye overlooks, and preserve details which the 
human mind forgets, it cannot reveal the magnifi- 
cence of a total eclipse. Only a direct view of 
the eclipse itself makes us realize so vividly the ut- 
ter insignificance of man, the futility of all things 
material; only the eclipse experience itself can 
make us reach that crucial point in philosophical re- 
flection where we feel we can rise above things ter- 
restrial and become one with the conception of space 
and time. 

Chapter VI 


"And what are the planets? Drops of a mixture 
of mind, a little mire, and plenty of moisture." 

— France. 

Now that we have come to the description of the 
other members of the sun's family, it is natural 
that we should begin at the centre and travel from 
the sun outward. The first planet we meet, then, 
is Mercury, not only the nearest to the sun, but the 
smallest, lightest, and swiftest of all planets. The 
path of Mercury around the sun is also the most 
elongated of all, the planet's distance changing from 
28,000,000 miles in perihelion, the point nearest 
the sun, to 43,000,000 miles in aphelion, the point 
farthest away from the sun. It requires 88 days 
for the planet to complete this course, traveling at 
an average speed of 30 miles per second. Since the 
path of Mercury around the sun lies entirely inside 
that of the earth, we can never see the planet recede 
from the sun by more than a given angle, which, in 
the case of Mercury, varies from 18 to 28 degrees, 




depending upon the position of the planet in its 
ellipse. Consequently, Mercury is visible only a 
short while after sunset, or before sunrise, and is 
one of the most difficult objects to observe with the 
naked eye. Many astronomers indeed, among them 
the celebrated Copernicus, have never seen it. Like 
the moon, Mercury is a dark body, shining only by 
reflected sunlight; like the moon, therefore, it will 
show phases depending upon its position relative 
to the sun and the earth. With the phases, its dis- 
tance from the earth varies considerably, from 
136,000,000 miles when the planet is on the 
other side, presenting a fully illuminated disk, to 
50,000,000 miles when it is between us and the 
sun. The first phenomenon is called superior con- 
junction, the second, when it is "new" Mercury, 
inferior conjunction. Occasionally it may happen 
that Mercury passes so exactly between us and the 
sun, that we see it projected against the sun's disk, 
crossing it slowly as a small, black spot as large as 
a silver dollar half a mile distant. These transits 
of Mercury are rather rare occurrences, happening 
about twelve times a century. They are, however, 
of great astronomical importance, since they furnish 
an opportunity for determining Mercury's position 
with great precision. From observations made at 
such times the French astronomer Leverrier, and 
latevr Ncwcomb, calculated that the elliptic orbit of 
Mercury was turning around in space, and more- 
over, turning around faster than could be explained 


by the attraction of the other planets. The speed 
of this rotation was such that in three million years 
the orbit made one more turn than it should, an 
amount so great that, for years, it has presented 
the most serious discrepancy in celestial mechanics. 



The explanation was not given until Einstein came 
with his theory of relativity, and the fact that this 
"advance of the perihelion" of Mercury had al- 
ready been observed was a strong and significant 
factor in the argument for the acceptance of 

In concluding, a few more figures may be added: 
Mercury has a diameter of 3,100 miles, and weighs 



about one thirtieth part of the earth. The density 
is 0.7 that of the earth, or almost four times that 
of water. 

Between Mercury and the earth circles Venus, 
of all planets the one most comparable to the earth. 
Like Mercury, Venus can never recede further than 
a certain distance from the sun, but, since Venus is 
nearer the earth, and much larger in size than Mer- 
cury, it is a much more conspicuous object in the 
sky. When Venus is to the left of and following 
the sun, it may set as many as three or four hours 
later. It is then seen as our brilliant "evening 
star," far surpassing all other objects in the sky in 
brightness. At these times of greatest apparent 
distance from the sun, Venus may easily be seen in 
the daytime, if one knows where to look for it. 
When the planet is to the right of the sun, it pre- 
cedes and rises before the sun: it is then our "morn- 
ing star." The ancients did not notice for a long 
time that the morning star and the evening star were 
the same object. The Greeks even had a different 
name for each: Phosphorus for the morning, Hes- 
perus for the evening star. 

On the scale of the astronomer who expresses 
brightness on a logarithmic scale where smaller and 
negative numbers indicate greater brightness, Venus 
may become as bright as -4.3, more than one hun- 
dred times as bright as the average star of the first 
magnitude. On the same scale Mercury never 


reaches greater brightness than —1.2, not quite nine 
times a first magnitude star. 

In its path around the sun Venus travels in an 
almost perfect circle with a radius of 67,000,000 
miles; the time it takes for one complete revolution 
is 225 days, moving on the average with a speed of 
22 miles per second. Its diameter is 7,700 miles, 
or almost as large as that of the earth, but owing 
to the great variation in distance from the earth, 
ranging from 26,000,000 miles at inferior conjunc- 
tion to 160,000,000 miles at superior conjunction, 
the apparent size of the planet varies from 64 sec- 
onds of arc to only 10, or from 29 times smaller 
than the full moon to 180 times smaller. The 
occasions upon which Venus passes so nearly between 
us and the sun that we see it cross the sun's disk 
are very rare indeed, the last event of this kind 
having taken place in 1882, while the next one will 
not happen until 2004. 

The mass of Venus, determined by the effect it 
has upon other planets, is found to be 81 per cent, 
of that of the earth. Since its volume is 92 per cent, 
of the earth, it follows that the density is 88 per 
cent, of that of the earth, almost five times that of 
water. It has been shown by observations made 
near the time of inferior conjunction that Venus 
has an atmosphere which can be observed to be 
more than a mile high, but which contains practi- 
cally no oxygen. However, it is entirely probable 
that what we observe is really the outer part of the 



atmosphere, and that the whole planet is surrounded 
by a thick layer of clouds. 

Venus and Mercury are the only planets known 
inside the orbit of the earth; they are sometimes 
called the inferior planets. Neither has any satel- 
lites so far as we know. It has been suggested from 
time to time that there is still another planet nearer 
the sun than Mercury. The most careful observa- 
tions made during an eclipse of the sun have thus 
far failed to reveal a trace of it, and it appears 
that, if such a body exists at all, it can be no more 
than thirty miles in diameter. Beyond this we have 
no information concerning the existence of this 
intramerairial planet; faith, however, has outrun 
knowledge on this occasion, and the planetary infant 
has been formally christened Vulcan. 

Seen from Venus, the earth is the brightest object 
in the firmament; with the moon, it forms a beauti- 
ful double star. At the time of its closest approach 
to Venus the earth would outshine an average first 
magnitude star more than 300 times, while our 
moon, revolving around the earth at a distance of 
one degree (twice the diameter of the full moon as 
seen from the earth), is 40 times fainter, but still 
one of the brightest lights in the sky. 

From Venus we travel farther out, pass the orbit 
of the earth, and come to Mars, the ruddy planet 
symbolic of the war god of the ancients and even 
now the subject of many a quarrel between scien- 
tists. Of all the celestial bodies visible to man, 


Mars occupies the most outstanding position. Fiery 
red, almost menacing in its savage lustre, at times 
the brightest star in the sky, Mars was from time 
immemorial regarded as the symbol of destruction, 
the personification of the god of war. It was made 
the scapegoat of all human crimes and follies, and 
to its pernicious influence was ascribed all disasters 
that befell humankind. To-day Mars occupies the 
centre of the discussion of the habitability of other 
worlds. It is the subject on which writers with a 
fertile imagination have lavished too much atten- 
tion, and consequently it has become the chief sub- 
ject of disagreement between the layman and the 
professional astronomer, one of the permanent 
liabilities of the profession in the public eye. 

Before we venture into the precarious question of 
habitability, let us first examine the facts in the 
case: Mars revolves around the sun in an elliptic 
orbit, changing its distance to the sun from 128,- 
000,000 to 154,000,000 miles. As a result of this 
eccentric orbit of Mars, the planet changes its dis- 
tance of closest approach to the earth considerably. 
It may come as close to us as 35,000,000 miles, 
while at other times it does not come within 62,- 
000,000 miles. Its brightness may vary, as a con- 
sequence, from -2.8 or considerably brighter than 
the brightest star in the sky, to -1.1, not quite as 
bright as Sirius, dependent on whether the nearest 
approach, or opposition, to the earth is "favorable" 
or not. It takes the planet 687 days to complete 



one revolution around the sun, but 780 days to take 
up the same position again relative to both earth 
and sun. This is the time that elapses between 
two oppositions, the positions such that the planet 
is opposite the sun, as seen from the earth. 

Mars is a small planet, only 4,215 miles in diam- 
eter, a little more than half that of the earth, and 
only one seventh of the earth in bulk. Observa- 
tions of the satellites give for its mass the value 
one tenth of the earth, thus making its density 0.72 
and gravity at the surface 0.38; that is to say, a 
man weighing 160 pounds on earth would weigh no 
more than 60 pounds on Mars. From observations 
of markings on the surface of Mars it has been 
possible to determine with great precision the time 
it takes the planet to rotate on its axis; from data 
gathered during the past two hundred years this 
time has been calculated as 24 hours, 37 minutes, 
22.58 seconds, with an uncertainty of no more than 
a few hundredth parts of a second. The axis of 
rotation on Mars makes approximately the same 
angle with the plane of its orbit as on the earth. 
The seasons, then, must be very nearly the same 
as ours. An observer on Mars, however, would 
not see the heavens rotate around our pole star, 
but around the star Delta Cephei, a faint star situ- 
ated about midway between Cassiopeia and Deneb, 
the brightest star in the Swan. 

The planet possesses two satellites, discovered 
by Asaph Hall at the Naval Observatory, Wash- 


ington, in 1877; the outer one, which revolves at 
a distance of 15,000 miles in 30 hours, is called 
Deimos (Terror) ; the inner one, which is no more 
than 6,000 miles from the centre of Mars, or less 
than 4,000 miles from the surface, revolves in 7 
hours, 39 minutes and is named Phobos (Fear). 
They were so named after the legendary compan- 
ions of Mars, or Ares, the god of war. These two 
moons are among the smallest and most curious 
bodies known in the whole solar system. The inner 
one, Phobos, revolves around Mars from west to 
east in less than one third of the time the planet 
itself rotates on its axis from west to east. In our 
own case the earth rotates much faster from west 
to east than the moon revolves around us; conse- 
quently, the moon lags but little behind the stars 
and rises in the east. On Mars, however, Phobos 
overtakes the rotation of the planet, rises in the 
west, and sets in the east. Deimos, much slower 
in its revolution around Mars, behaves more nor- 
mally and does rise in the east, but it takes its time 
about it and remains above the horizon for almost 
three days. Compared to the planets and the other 
satellites of the solar system, these two moons are 
but small fragments of dust. Phobos is probably 
no more than ten miles in diameter, Deimos even 
less, probably about five; even so, they are both so 
close to the surface of the planet that, seen from 
there, Phobos would be only three times smaller 
in diameter than the full moon and twenty five times 



fainter, while Deimos would be of the same ap- 
pearance as Venus seen from the earth. 

Since Mars can come very close to us the topog- 
raphy of its surface is better known than that of 
any other object in the sky except the moon. A 
small telescope will show only the polar cap, a 
brilliantly white spot near one of the extremities of 
the axis of rotation, and without doubt due to sun- 
light reflected from the ice and snow that gathers 
there during the Martian winter. This polar cap 
is seen to diminish in size, due to melting of the 
ice and snow as Martian spring and summer ad- 
vance, as may easily be seen on the excellent photo- 
graph of Mars taken by E. C. Slipher at Lowell 
and by Wright at Lick. Recent work done by 
Coblentz and Lampland at the Lowell Observatory 
and by Pettit and Nicholson at the Mt. Wilson 
Observatory has shown that the temperature in 
these regions is about 100 degrees below zero, Fah- 
renheit, so long as the snow is present, but that 
after it has melted the temperature may rise to SO 
degrees above zero and become equal to the midday 
temperature at the Martian equator. The night 
temperature, even at the equator, must be very low, 
however, and might well be below zero the year 
round; at sunrise and sunset it may be just above 
zero. Spectroscopic investigations carried out at 
the Lick and Mt. Wilson Observatories have indi- 
cated the presence of a little water vapor and 
oxygen, in amount about 5 per cent, and 15 per cent. 

Phniiftaphi by A". ('.. Sliphrr, Iwrrll Hh»trvnii>r\ 


Tin- telescopic appearance of three of the brighter planets. In the Upper pari 
of the illustration a drawing and corresponding photograph of Mars are 
compared. In the photograph (on the tighO the polar caps are well shown, 
while the "canals" are less distinct than in the drawing. In the lower left, 
Saturn and its rings arc photographed in one of the positions most lavor- 
able for observation. In the lower right, on the disk of Jupiter, arc the 
equatorial heirs, and, near the right margin, the Bay of the (iteat Red Spot. 


respectively of that normally in the atmosphere of 
the earth. On the whole there is probably no more 
than 10 or 20 per cent, of the atmosphere we have 
on earth. 

When more powerful telescopes are used, not 
only the polar caps but a wealth of other details 
are revealed on the surface of Mars, among them 
the famous "canals." These were first discovered 
by Schiaparelli at Milan, in 1877, and named canali 
because they appeared as straight lines of a dark 
colour up to fifty miles wide, several hundreds 
of miles long, and traversing the Martian conti- 
nents in all directions. When later Lowell, at 
Flagstaff, Arizona, saw the whole Martian surface 
covered with a maze of these canals it became al- 
most inevitable that they should come to be inter- 
preted as artificial — the work of intelligent beings. 
Human beings? Possibly, though no one has as yet 
dared to go that far. At any rate, it has brought 
up the question of the possibility of life on Mars, 
and this has led to the problem of what "life" is, 
and has brought with it a disagreement between 
the astronomer and the public. When the astron- 
omer talks of life on Mars he would include all 
possible kinds; the layman, on the other hand, 
would mean principally human life. The latter is 
not concerned with fungi or bacteria, or even 
plants. Bacteria, so small that they run through 
the finest filters in our laboratories, may be of 
interest here if they cause cancer: they are of no 



interest whatever across a chasm of 50,000,000 

Percival Lowell did go so far as to suppose a 
race of intelligent beings on Mars, capable of such 
engineering feats as the construction of canals 
2,000 miles long. The underlying reason was ob- 
vious: since Mars is a dying planet, its inhabitants 
found it becoming so difficult to live there that they 
had to turn, as a last resort, to herculean methods 
for irrigating the arid equatorial deserts, the only 
regions warm enough to live in, and irrigate them 
by means of the snow and ice stored up at the poles. 
Lowell even went to the extent of calculating that 
a pumping system necessary for such planetary cir- 
culation of water required an expenditure of energy 
4,000 times the power of Niagara. 

Other astronomers have taken opposite views. 
The existence of canals in such numbers as indi- 
cated by Lowell is denied, and the fact that the 
canals are seen to cross the seas is taken as proof 
that they cannot contain water. W. H. Pickering, 
one of the most profound students of the subject, 
has come to the conclusion that it is not the canals 
themselves we see, but the much wider strips of 
vegetation on their banks. The whole question is 
far from settled as yet, and it will take many years 
before observations are in sufficient agreement to 
enable us to discuss the question of life on Mars 
properly. All that we can say now is that when 
reviewing the different prerequisites a planet must 


fulfill in order to allow life comparable to life on 
earth, we find that the conditions are just met on 
Mars, but no more than that. If, therefore, life 
does exist on Mars, it must be under severe con- 
ditions and can hardly be compared to the easy 
ways of our planet. 

With Mars we have finished the description of 
the terrestrial planets; outside these we find the 
four major planets, Jupiter, Saturn, Uranus, and 
Neptune. The gap between the orbits of Mars and 
Jupiter, however, is disproportionately large and 
destroys the symmetry of the system, as older 
astronomers, even those of Kepler's time, had re- 
marked long before Uranus and Neptune were 
known. When Bode's "law" of the increase in the 
mean distance of the planets from the sun was 
formulated, and when shortly afterward Uranus 
was discovered and found to conform to Bode's 
law, the lacuna between Mars and Jupiter became 
more and more evident. Astronomers were now 
firmly convinced that there must be still another, as 
yet unknown, planet in this region, and societies 
were formed whose aim it was to search for the 
missing link. It was the good fortune of Piazzi at 
Palermo, Sicily, to discover an object which later 
proved to be the long-sought-for planet; using his 
right of naming it, he chose the name Ceres after 
the ancient goddess of Sicily. After having first 
noticed it on the first day of the preceding century 
(January 1, 1801,) as a star-like object in the con- 



stellation Taurus, Piazzi followed its course for a 
time, but was taken ill shortly afterward; before 
he recovered the planet had disappeared in the 
evening twilight. The difficulty then arose to cal- 
culate the path of the new object from these obser- 
vations, in order that it might be rediscovered in 
future. It was in this way that Gauss was furnished 
with the inspiration which led to his invention of 
the method of deriving the orbit of a celestial body 
from three observations. 

A year later a second planet, Pallas, was discov- 
ered by Olbers; a third one, Juno, by Harding in 
1804, and a fourth, Vesta, again by Olbers, in 
1807. Then followed thirty-eight years of fruitless 
search, but from 1845 on hardly a year passed in 
which there were not one or more new objects of 
this type discovered, some years yielding as rich a 
crop as one hundred new discoveries. During the 
past few decades the search has been greatly inten- 
sified by the introduction of photography, and the 
total number of these bodies has grown so enor- 
mously that to date already more than one thousand 
have had reliable orbits determined, while almost 
as many others have been discovered, but observed 
insufficiently to permit a calculation of their path. 
Instead of merely filling a vacancy in the solar sys- 
tem, the small planets have become almost a nui- 
sance through their large numbers: once hailed as 
the "missing link," they are now maligned as the 
"vermin of the sky." 




The large number of small planets (or asteroids, 
as they are unfortunately and infelicitously called) 
now known more than fill the gap between the 
orbits of Mars and Jupiter; they fill it to overflow- 
ing. Their system of orbits is so complicated, how- 
ever, that, if each orbit were represented by a thin 
ring of steel wire, one could lift them all, by moving 
only one. In their paths around the sun the 
asteroids follow ellipses as do the planets, but their 
orbits are much more elongated, much more eccen- 
tric, than those of the planets. Contrary to the 
behaviour of the planets, the asteroids do not, as a 
rule, follow the plane of the ecliptic very closely, 
but may make large angles with it. In mean dis- 
tance from the sun they range from 1.46 astro- 
nomical units for Eros to 5.71 for Hidalgo; since 
the mean distances of Mars and Jupiter are 1.52, 
and 5.20 respectively, we see that the asteroids 
have indeed filled out the gap to overflowing; they 
have not, however, filled it uniformly. They seem 
to avoid certain values for the mean distance from 
the sun, as first noted by Kirkwood, and mathemati- 
cal analysis has shown that this is not due to chance, 
but a direct consequence of the perturbations exerted 
by Jupiter, the giant planet of the solar system. The 
influence of Jupiter on an asteroid at any one time 
is very small indeed, but allow Jupiter enough time 
and its disturbing attraction, recurring with deadly 
precision every twelve years, will drown out all 
asteroids that are moving in synchronism. Synchro- 


nous action is one of the most powerful destructive 
agents of nature; no swing can withstand the effect 
of periodic blows recurring at intervals equal to the 


time of swinging. We have already seen how the 
earth, through the recurrent tides which it raised 
on the moon, has subjected that body to its will and 
made it rotate on its axis in the same period in 



which it revolves around the earth. Indeed, though 
the mills of the gods grind slowly, yet they grind 
exceeding small ! 

The asteroids are all small bodies, the largest 
known one, Ceres, being only 480 miles in diameter, 
while many of the smaller ones are undoubtedly 
no more than a few miles in size, no larger than 
Manhattan Island. Their mass, compared to that 
of the earth, must be very small, one ten-thousandth 
at most. Ceres and Pallas, the two largest ones, 
are estimated to weigh about this much, but the 
combined mass of all asteroids, known and un> 
known, is probably not much over one thousandth 
part of that of the earth. 

When asteroids were first discovered, the theory 
was proposed that they perhaps formed the re- 
mains of one single planet which had exploded, 
leaving its fragments to pursue their own paths 
around the sun, each being subject to a different 
perturbative influence of Jupiter. Mathematical 
analysis has shown, however, that Jupiter never 
could have altered the orbits of such a collection 
of asteroids into the present heterogeneous mix- 
ture. Although, for this reason, the theory that 
all asteroids had a common origin has to be aban- 
doned, it may still be true for certain smaller aggre- 
gations. A striking tendency to gregariousness has 
been noticed among these small celestial bodies; 
Hirayama in Japan has identified several "asteroid 


families," some containing as many as forty mem- 

Of all asteroids known at present, Eros, number 
433 in order of discovery, is undoubtedly the most 
interesting. Eros describes a very elongated 
ellipse, which lies for the larger part inside the 
orbit of Mars. It may approach the earth within 
13,000,000 miles, closer than any other planet; it is 
then of great value for determining the exact di- 
mensions of the solar system by direct triangula- 
tion. As we have seen before, the solar system can 
be very easily mapped on a relative scale, the chief 
difficulty lying in the determination of the exact 
scale value. It is in this instance that Eros is use- 
ful, because of its short distance from the earth. 
Other remarkable asteroids are Albert (No. 719) 
and Ganymede (No. 1036), both of which may come 
inside the orbit of Mars, and Hidalgo (No. 944), 
whose orbit is an ellipse almost touching the orbit 
of Mars, on the one hand, and that of Saturn on the 
other. Lastly, there is the group of Trojan plan- 
ets, composed of Achilles (588), Patroclus (617), 
Hector (624), Nestor (659), Priamus (884), and 
Agamemnon (911), all of which revolve around 
the sun at the same distance as Jupiter. Thus, it 
can be proved mathematically that they must al- 
ways be near a point which forms an equilateral 
triangle with the sun and Jupiter. This is actually 
the case, Patroclus and Priamus following Jupiter, 
and the other four preceding. 



Perhaps the strangest asteroid of all is that dis- 
covered recently in Japan and provisionally named 
Tokio. A preliminary orbit determined for it indi- 
cates that it moves in an ellipse far more elongated 
than those of many comets and reaching as far as 
Uranus. The path of Ceres, as an example of a 
typical asteroid, and those of Eros, Ganymede, 
Achilles, Hidalgo, and Tokio, as exceptional aster- 
oids, are portrayed in figures 7 and 9. 

Chapter VII 

"If they are inhabited, what worlds of misery; 
if they are not, what a waste of space." 

— Carlylb. 

There is no greater contrast in the solar system 
than the transition from the asteroids, that host 
of small and innocuous planetary fragments, to 
Jupiter, the giant of the planets, larger and heavier 
than all the rest put together. Although never 
attaining as great splendor as Venus, and even 
occasionally surpassed in lustre by Mars, Jupiter 
is yet, through its much slower motion, the most 
conspicuous object in our night skies. Its steady, 
pale yellow light, coupled with its slow, majestic 
progression along the firmament, have earned for 
it its name, that of the god among gods. 

At its equator Jupiter has a diameter of 88,640 
miles, slightly more than eleven times that of the 
earth, but its polar diameter, owing to the very 
rapid rotation of the planet, and the consequent 
flattening, is considerably less, only 82,880 miles. 
This causes the disk to appear very flattened, so 




much so in fact that this flattening can be easily seen 
through a small telescope. The weight of Jupiter is 
the one most accurately computed in the whole solar 
system; it has been determined from the planet's 
influence on the motion of asteroids and from its 
own satellite, and has been found to be 313 times 
that of the earth, or 1,047 times less than that of 
the sun. Since the volume of Jupiter is 1,312 times 
larger than that of the earth, its mean density is less 
than one fourth of ours, or 1.34 times that of water. 

Jupiter travels around the sun in an ellipse which 
does not greatly deviate from a circle. The circle 
has a mean radius of 483,000,000 miles, and it 
takes the planet almost twelve years to complete 
one turn, while its average speed is 8 miles per 
second. While it revolves about the sun, Jupiter 
spins on its own axis in 9 hours, 55 minutes. A 
"year" on Jupiter, therefore, has no less than 
10,000 Jovian days. Since the axis of rotation is 
almost perpendicular to the plane of the orbit, 
there are no seasons on Jupiter, only perpetual 

Even a small telescope will show the principal 
features of Jupiter's surface, several wide, diffusely 
outlined belts, usually identified as clouds. Mete- 
orologically speaking, Jupiter is quite different from 
the four terrestrial planets, as may be observed 
even in an ordinary telescope. A magnifying power 
of fifty makes Jupiter's disk appear as large as 
the moon to the naked eye; under such magnifi- 


cation one can see very clearly that the surface is 
crossed by several wide, diffusely outlined belts, 
more or less parallel to the planet's equator. A 
larger telescope, and a higher magnification will 
reveal more detail, especially a beautiful contrast 
in colours. It will then also be noticed that this 
detail on the surface is continually changing in posi- 
tion, due to the planet's rotation, and in character, 
due, no doubt, to real changes in Jupiter's atmos- 
phere. It was formerly supposed that, since Jupi- 
ter has such a dense atmosphere, it must also be 
considerably warmer than the earth, at the surface 
at least, and perhaps hot enough to emit some light 
of its own. Recent accurate measures made by 
Lampland at the Lowell Observatory indicate that 
the light we receive from Jupiter is entirely reflected 
sunlight, and that the temperature at the surface 
cannot be far from 200 below zero, Fahrenheit. 
For this reason, it has been suggested that the 
planet has a central core of dense material, not 
unlike the rocks on our surface, but is surrounded 
by a thick layer of ice, and the whole enveloped 
in a dense atmosphere. 

As we might have anticipated from its great 
mass, Jupiter is not a solitary planet, but has a 
number of attendant satellites: nine are known at 
present. The four brightest were found by Galileo 
as soon as he turned his telescope on Jupiter on 
January 17, 1610, and they may be seen with any 
telescope, even with the help of a good field glass. 



Under exceptional conditions some of them may be 
seen with the naked eye. Although they have been 
given names, they are usually referred to by num- 
ber, in order of their distance from Jupiter. The 
masses of these satellites are determined from their 
mutual attractions and the disturbance of their 
orbits, and are found to be not very different from 
that of our moon. Since the first two are of the 
same size as our moon, they are probably composed 
of similar material, resembling the rocks on the sur- 
face of the earth; the third and fourth are almost 
four times the bulk of the moon and may, according 
to an interesting suggestion made by Jeffreys, con- 
sist largely of ice or solid carbon dioxide. 

All four satellites revolve around their primary 
in circles which lie very nearly in the plane of the 
ecliptic, and they are thus seen sideways from the 
earth. Consequently, we see them move only from 
right to left, and they always appear to be very 
nearly in a line which passes through Jupiter. As 
seen from Jupiter's surface, where sunlight is only 
one twenty-seventh as intense as on earth, the first 
satellite would appear five times fainter than the 
moon, and the other three still fainter. The inner 
three of Jupiter's moons are so close to the surface 
of the giant planet that once during every revolu- 
tion they pass behind it and into the great shadow 
cone; they are eclipsed. The fourth satellite is far 
enough away so it may escape eclipse occasionally. 
Conversely, during every revolution around Jupi- 



ter, the satellites pass in front of it and cast their 
shadows, visible from the earth as small black spots 
moving rapidly over the planet's surface. These 
two types of phenomena, eclipses of the satellites 
and transits across the disk, form a spectacle most 
interesting to watch from the earth. When Jupiter 
is in opposition to the sun — that is, when the earth 
is between it and the sun — the planet's distance 
from us is considerably less than when the planet 
is in conjunction, or on the other side of, behind, 
the sun, so to speak. Eclipses of satellites, which 
in reality occur at regular and stated intervals, will 
appear to be retarded in the latter case and ad- 
vanced in the former, simply because the rays of 
light which are carrying the message of the event 
have a longer way to travel in the second case. 
Arguing from this principle, Olaf Roemer, a 
Danish astronomer of the Seventeenth Century, 
first determined the velocity of light from the 
known distance of Jupiter to the earth and from 
the observed advance and delay of the eclipses. 

For almost three hundred years the four Gali- 
lean satellites remained the only ones known — until, 
in 1892, Barnard at the Lick Observatory discov- 
ered a fifth one very close to the surface of the 
planet and performing one whole revolution in less 
than twelve hours. This fifth satellite appears as a 
star of the thirteenth magnitude, one million times 
fainter than Jupiter, and it is without doubt one of 
the most difficult objects to observe. The sixth and 



seventh satellite were discovered photographically 
by Perrine at the Lick Observatory, the eighth by 
Melotte at Greenwich, and the ninth by Nicholson, 
again at Lick. They are all exceedingly faint ob- 
jects, observable only with the most powerful photo- 
graphic telescopes, although the sixth has occasion- 
ally been seen visually. 

A curious feature of the last two moons is that 
their motion around Jupiter is opposite to the usual 
direction of motion in the solar system. To an 
observer situated far north of the plane of the 
ecliptic, the earth and all other planets would be 
seen revolving around the sun in a direction con- 
trary to that of the hands of a clock; the same 
would be the case for the motion of the moon 
around the earth, the moons of Mars, and the 
innermost seven satellites of Jupiter; they would 
all appear to move counter-clockwise. The outer 
two, however, would move clockwise. As a result, 
it has often been the subject of speculation whether 
Jupiter has come by these two of its moons honestly, 
or whether, perhaps, they are not ancient asteroids 
captured by Jupiter. On the whole, mathematical 
astronomers are now inclined to believe that the 
nine known satellites are all permanent members of 
Jupiter's family, although the possibility that Jupi- 
ter has lost satellites in the past or will capture 
others in the future has not been entirely ruled out. 

With Saturn, the planet next to Jupiter, we ap- 
proach the ancient frontier of the solar system; the 




distance of 886,000,000 miles that separates Saturn 
from the sun is the remotest to which the unaided 
human eye had penetrated in the planetary system. 
On account of its greater distance from the sun, 
Saturn is the slowest of all planets readily visible to 
the naked eye; in addition, it is also the faintest, 
though still as bright as a star of the first magnitude. 
The peculiar, "livid" hue of its light together with its 
sluggishness of motion made this planet the symbol 
of immutable fate to the ancients. By the astrolo- 
gers of old, Saturn was regarded as the chief 
"malefic," the principal inciter of disaster that made 
all those unlucky ones born under its spell suffer 
forever after: through Saturn spoke the "voice of 

It takes Saturn about 29.5 years to complete its 
orbit around the sun, an orbit which is more ellip- 
tical than that of Jupiter, the distance from the sun 
varying between 840,000,000 and 940,000,000 
miles. The planet itself, too, is very far from a 
perfect sphere, and the flattening at the poles may 
be seen through any small telescope. The equa- 
torial diameter is 74,000 miles, the polar diameter 
only 66,000, which makes the bulk of the planet 
734 times larger than that of the earth. Its mass, 
which is determined with great accuracy from its 
satellites as well as from its influence on Jupiter, is 
not proportionately greater, but only 95 times that 
of the earth. It follows that the mean density is 
only one eighth that of our planet, or only seven 



tenths that of water: if we could only find an ocean 
large enough, Saturn would float. From the fact 
that Saturn turns on its axis in a little more than 
ten hours, mathematicians have deduced that most 
of the surface layers must be gaseous, and that 
much of its total weight is concentrated in a rather 
small core at the centre. These surface layers are 
at a temperature slightly lower than that of Jupiter, 
about 240 degrees below zero. 

In size, weight, and brightness Saturn must bow 
to the superiority of Jupiter. As a sight in the 
telescope, however, Saturn far surpasses all other 
objects in the solar system; it is unique in our expe- 
rience with nature in that it is surrounded by a ring. 
When Galileo turned his first telescope on Saturn, 
he noticed two "supplements," one on each side 
of the planet, and he thought the object to be 
triple. His telescope was not quite good enough 
to show the ring, and it was not until Huygens ob- 
served it in 1655 that the true form of the ring 
became known. Later, the French astronomer 
Cassini found that the ring really consists of two 
concentric rings, separated by a thin gap which 
appears as a black line in the telescope. Still later, 
Bond of Harvard and Dawes in England found a 
third ring, rather hazy in appearance, which forms 
the innermost boundary; it is called the "crepe 
ring." This extreme inner edge of the ring is only 
7,000 miles from the planet's equator and 11,000 
miles in width; the outer portions are 16,000 and 





10,000 miles wide respectively, while the black gap 
is no more than 3,000 miles wide. The ring forms 
an angle of about 28 degrees with the ecliptic, and 
during Saturn's revolution around the sun the ring 
remains parallel to itself. Consequently, it must 
happen twice during their course around the sun 
(that is to say twice during Saturn's year, or once 
every fifteen years) that the rings are pointed di- 
rectly at us and we see them "edge-on." This is 
really identical with saying that we do not see the 
rings at all for they are so thin that even with the 
largest telescopes one may scan in vain for them at 
such times. From this it has been estimated that 
they cannot be much more than 10 miles thick. 

Observations with the spectroscope have proved 
that the rings do not constitute one solid body, but 
that they are composed of a great swarm of minute 
particles — meteors, so to speak — each revolving 
separately around Saturn. In a way then, the ring 
becomes analogous to the belt of asteroids in the 
solar system, and indeed there are many points of 
resemblance. We have seen before how there are 
certain gaps in the mean distances of the asteroids 
from the sun, owing to the impossibility of main- 
taining an asteroid in motion where it moves in 
synchronism with the powerful disturbing force of 
Jupiter. Here, in Saturn's ring, we have a minia- 
ture edition of the same story; the asteroids are 
replaced by the particles of the ring, and the role of 
the wicked uncle is played by the satellites nearest 

to Saturn. The black division discovered by Cas- 
sini is thus explained by the action of Mimas, the 
nearest satellite. 

This brings us to the satellites, for, in addition 
to its rings, the distant world of Saturn is accom- 
panied by no less than ten moons. Nine, named in 
order of their distance from Saturn: Mimas, Ence- 
ladus, Tethys, Dione, Rhea, Titan, Hyperion, 
Iapetus, and Phoebe, are known with certainty, 
while the tenth one, Themis, was seen but a few 
times and lost again; it cannot now be identified. 
Titan, the first to be discovered, is the largest of 
them all, slightly exceeding our moon in size, while 
the others range from ISO to 1,000 miles in diame- 
ter. Phoebe, the outermost of all, revolves back- 
ward around the planet, as do the eighth and ninth 
satellites of Jupiter; the other eight moons of 
Saturn behave normally. The masses of the satel- 
lites were determined by the influence of their mu- 
tual attraction and the resulting perturbations of 
their orbits; they are by far the smallest masses 
that have ever been determined in this way. Titan 
is almost twice as heavy as our moon, Mimas less 
than one thousandth part of our moon, while the 
others range in between. 

The solar system of the ancients, consisting of 
the sun, the earth, the moon, and the five planets 
visible to the naked eye, Mercury, Venus, Mars, 
Jupiter, and Saturn, had been enlarged and ex- 
tended by the discovery of several satellites, but 



the invention and application of the telescope had 
not added any new planet; rto new independent body 
had joined the ranks. When, on March 13, 1781, 
William Herschel discovered a new object which 
proved to be a planet revolving about the sun in 
an orbit lying entirely outside that of Saturn, the 
discovery created great excitement. The iron- 
bound limits of the solar system had been broken, 
new vistas had been opened, and free rein could be 
given to the imagination concerning the completion 
of the planetary system. 

^ Herschel named his star Georgium Sidus, after 
George III, then king of England, but fortunately 
for the impersonal character of science, the mytho- 
logical name Uranus, the father of Saturn, pre- 

After Herschel's discovery it was found that the 
same object had been observed many times before 
but that it had always been mistaken for a star. It 
has even been said that the inhabitants of some 
island in the South Seas knew of it, not as a star, 
but as a planet wandering through the heavens. 
This is just possible, for Uranus, under favourable 
conditions, is well visible to the naked eye if one 
knows where to look for it. 

The mean distance of Uranus from the sun is 
19 astronomical units or 1,782,000,000 miles; the 
actual distance, however, fluctuates between 1,700,- 
000,000 and 1,870,000,000, owing to the ellipticity 
of the orbit. Travelling at the rate of only 4y 4 



miles a second, Uranus needs 84 years to complete 
one turn about the sun. The other facts known 
about Uranus are that its diameter is about 32,000 
miles, its volume 64 times that of the earth, and 
its mass IS times greater than that of our planet; 
it rotates on an axis in about 1 1 hours, and has four 
satellites. In order of their distance from Uranus, 
these are named Ariel, Umbriel, Titania, and 
Oberon. They are all exceedingly faint objects and 
therefore very small, probably no more than a few 
hundred miles in size. 

When Uranus had been observed some fifty years 
it became evident that the planet did not follow the 
"true" course mapped out for it by mathematical 
calculation. During this interval the planet often 
strayed away from its computed position by as 
much as 20 seconds of arc; that is, by as much as 
the distance across a dime when seen from a dis- 
tance of 400 feet. In 1845 these deviations had 
accumulated to almost 2 minutes of arc, one fif- 
teenth of the moon's diameter, an "intolerable 
quantity." Such a state of affairs could not be en- 
dured, the mysterious source of these discrepancies 
must be discovered; and almost simultaneously two 
astronomers, Leverrier in France and Adams in 
England, undertook to do it. Independently, and 
almost simultaneously, they came to the same con- 
clusion: there was an unknown planet, revolving 
far beyond Uranus, but of sufficient mass to pull 
Uranus out of its path. Both computers made the 





same prediction concerning the position of the new 
planet, but Adams had the bad luck of finding no 
responsive audience, while Leverrier was astute 
enough to write to the Berlin Observatory, asking 
them to verify his discovery. But we are not doing 
Leverrier justice. He did not write and ask for 
verification; he laconically told Galle, an astrono- 
mer in the Berlin Observatory, that "if he looked 
in a certain position in the sky, he would find the 
new planet." So convinced was Leverrier of the 
correctness of his calculations that he never gave 
himself the trouble of looking for the new planet; 
in fact, he died in 1877, on the exact thirty-first 
anniversary of the discovery, without ever having 
seen it. He had reason indeed to trust his calcula- 
tions: when Galle, after receiving Le Vcrrier's 
letter, pointed his telescope to the sky, he imme- 
diately found the new planet, within one degree of 
the predicted position. The name Neptune was 
unanimously adopted for this remotest of planets. 
The discovery of Neptune, from mathematical 
calculation alone, is one of the most eloquent wit- 
nesses of the power of the human mind. We may 
well say that it represents the crowning achieve- 
ment of abstract reasoning in any science. The 
presence of a new planet had been felt, owing to 
some apparently trifling discrepancies in the motion 
of an object 1,780,000,000 miles distant; the new 
planet had been seen with the spiritual eye of the 
astronomer by means of his equations, his formula?, 


and his tables of logarithms. Verification then re- 
mained merely as one of the commonplace details 
of everyday life. 

Neptune revolves about the sun in an almost 
circular orbit, at a distance of 30 astronomical 
units, or 2,800,000,000 miles. It is the only one 
of the planets for which Bode's law of planetary 
distances breaks down, since this law would predict 
a distance of 39 astronomical units. Neptune's 
speed is only a little more than 3 miles per second, 
and it therefore needs 165 years to complete one 
revolution around the sun. The planet's diameter 
is 31,000 miles, its mass 17 times that of the earth; 
it appears to us as a star of the eighth magnitude, 
invisible to the naked eye, but, if one knows where 
to find it, visible in an opera glass. Neptune has 
one satellite, sometimes called Triton, an exceed- 
ingly faint object, which revolves around Neptune 
at a distance of 220,000 miles, and which is prob- 
ably very similar to our moon. 

With the modern increase in accuracy of obser- 
vations, slight irregularities have been noticed in 
the motions of Uranus and Neptune, which, again, 
cannot be explained by the action of the sun and the 
known planets. Again the conclusion has been 
reached that there may be another planet beyond 
Neptune, which may be responsible for these devia- 
tions. Unfortunately, the discrepancies in the pres- 
ent case are so small that no very definite conclu- 
sions can as yet be reached concerning such a 



trans-Neptunian planet, except that there cannot 
be a planet as large as Neptune within twice the 
distance of Uranus, or one as large as Jupiter within 
2J4 times the distance of Neptune. W. H. Picker- 
ing has announced as a not unreasonable prediction 
that the new planet, planet O, revolves about the 
sun in a very elliptic orbit at about the same mean 
distance as Neptune, that it cannot be much heavier 
than about half the weight of the earth, and that it 
should appear to us approximately as a star of the 
twelfth magnitude. 

With Neptune we have reached the boundary of 
the planetary system. Before we pass on to a de- 
scription of the more vagrant members of the sun's 
family, the comets, let us view in retrospect what 
we have passed through. If we should try to make 
a model of the solar system on the scale of one to 
one billion, we should have to represent the sun by 
a sphere four feet in diameter. At a distance of 
70 yards we should find Mercury, one fifth of an 
inch in diameter; at 120 and 170 yards respectively, 
Venus and the earth, both half an inch in diameter; 
at 250 yards a smaller globe, a quarter of an inch 
in size, representing Mars. Then would follow a 
host of sand-grains, as the asteroids, and at half a 
mile from the central sun we should find Jupiter, 
four inches in diameter. One mile away would 
reach Saturn, almost as large as Jupiter, and at 
distances of two and three miles respectively, 
Uranus and Neptune, each two inches in size. By 


the time we arrived at these outposts the inner 
parts would have disappeared from view: our 
human eyes, if transferred to Neptune, would be 
unable to see one single planet with ease. Jupiter 
and Saturn might be observed close to the sun, and 
Venus and the earth perhaps during a total eclipse, 
while the other three would remain invisible. The 
sun would appear 900 times smaller in area and 
900 times feebler in light than from the earth. The 
sun's heat would be inadequate to warm us; it could 
not raise the temperature above 360 degrees below 
zero. Here, at last, at the barrier of the sun's 
empire, the perpetual cold of empty space has van- 
quished the life-giving heat of the sun. 

Chapter Fill 

"Wan, dishevelled, slow, fatigued, glides before 
us the wandering comet in the night of its 

— Flam marion. 

Comets have long held the unenviable position of 
scapegoats in astronomy; they have for ages past 
been blamed for all the misfortunes befalling the 
human race, in man's simple desire to shift the 
responsibility for things terrestrial to forces and 
things out of his reach, in the sky. Comets, as 
weird apparitions, rare and sudden in their occur- 
rence, offer only too good a target for such accu- 
sations. The sight of a great comet blazing forth 
in the blackness of the nocturnal sky is an imposing 
sight, indeed, and not likely to be easily forgotten. 
Hence the numerous allusions to comets through- 
out the literature, sometimes as descriptions of 
painstaking minuteness, sometimes in language of 
grotesque extravagance. Especially the ancient 
Chinese records reveal some past masters in this 
noble art. 




Our word comet is derived from the Greek 
komeles, "the long-haired," a most obvious descrip- 
tion. Contrary to our present views, however, the 
ancients considered comets to be vapors in the 
atmosphere, or exhalations from the earth. Among 
the minor disasters that were supposed to accom- 
pany these missionaries of evil, were famine, pesti- 
lence, and wars. Now all these associations have 
disappeared, and only the name stands, though even 
this proves hardly deserved. Great comets, with 
long tails like those that terrified the ancients, are 
still seen as frequently as ever, but in addition the 
modern observing methods have revealed the exist- 
ence of large numbers of faint comets entirely de- 
void of tails. Strange to say, however, the great 
modern telescopes have not played a prominent 
part in the discovery of comets; that is largely due 
to amateurs, observing with comparatively small 
telescopes, and to the use of photographic cameras. 
The reason is that our giant telescopes have such a 
high magnifying power that only a very small por- 
tion of the sky can be examined at one time, and 
the chances for discovery then are much less favor- 
able than for amateurs who sweep the sky with 
small telescopes with large fields of vision. "Comet 
hunting" is quite a sport, but the rewards it yields 
are small: one comet a year is considered a good 
average for the assiduous observer, although some- 
times a lucky catch is made, as by the late Rev. Joel 


Metcalf, who once found three new comets within 
forty-eight hours. 

When a comet has been discovered the news must 
be communicated to the world at once, and in 
astronomy all roads lead, not to Rome, but to Har- 
vard and Copenhagen. A code telegram announc- 
ing the discovery is immediately dispatched to these 
two centres, whence the news is relayed to all other 
observatories. Now the large telescopes come into 
play, accurate observations of the new object are 
secured, and these, too, are at once cabled to Har- 
vard and to Copenhagen, and again relayed. As 
soon as three accurate observations have been ob- 
tained the computers set to work, and within 
twenty-four hours they produce the orbit of the 
comet, its path around the sun, which will enable 
them to predict its future course among the stars. 
Very often such an orbit is only a first approxima- 
tion, and further observations are required to clear 
up doubtful points. Some comets disappear again 
so quickly that there is no chance for further refine- 
ment, and the first preliminary orbit is the best we 
can do. In that case it is generally impossible to 
say whether the comet is a permanent member of 
the solar system, if it will return, and how soon. 
When we do get sufficient observations to calculate 
the path of a comet in more detail, we almost in- 
variably find that it describes an ellipse, a closed 
curve, and thus it will come back again after a 
number of years. It is only very rarely, and then 



always with suspicions and reservations, that the 
computations lead us to suppose that the comet 
travels along a parabola, or a hyperbola, an open 
curve, in which case it will recede again to infinity, 
never to return. On the whole it looks as if prac- 
tically all comets are describing closed paths around 
the sun, that they all come back to the sun at their 
appointed times. Sometimes these paths are very 
long ellipses, ovals, with the sun near one of the 
ends of the curve. In those cases it may take the 
comet thousands and perhaps millions of years to 
complete one turn; at other times the ellipse found 
is very short, and the comet may be expected back 
in a few years. And thus it seems as if really the 
vast majority of the comets are permanent members 
of the sun's family; we have no longer the right 
to consider them as interlopers from space, as mes- 
sengers from some far-distant part of the universe 
that have merely come to pay the sun homage, but 
have not come to stay. This does not mean, how- 
ever, that comets are as regular members of the 
solar system as the planets. Far from it; comets 
are very erratic in their behaviour. From the fact 
that they traverse the solar system in long ellipses, 
reaching beyond the orbit of Neptune in some cases, 
it follows that sometimes these wanderers of the 
wasteland may come very close to one of the larger 
planets, and be deflected from their course. Their 
entire orbit may be changed, as happened, e. g., to 
the comet Brooks. Delving into its history by 



means of calculations, it was found that this object 
used to travel around the sun once in thirty years, 
but that it imprudently came very close to Jupiter 
in 1886. The attraction of this giant among the 
planets was so powerful that it shortened the 
comet's path: thereafter it went around the sun in 
seven years. It was first discovered by Brooks in 
1889, but split into four pieces at that time, and 
when it came around again in 1896, in 1903, and 
in 1911 it had lost so much of its brilliance that it 
could only be seen in the largest telescopes. In 
1918 it was not found, and astronomers feared that 
its career might come to a sudden and tragic end, 
when computations showed that in 1921 it would 
again approach Jupiter very closely. Nothing 
much seems to have happened at that occasion, 
however, since the comet returned in 1925, on time 
but looking very much the worse for wear. 

In contrast to this one, some other comets behave 
almost as normally as the planets themselves. 
Encke's comet, first seen in 1786 and since then ob- 
served every three and one-half years, is as regular 
in its appearance and disappearance as may be de- 
sired. A remarkable comet was discovered by 
Schwassmann and Wachmann in 1927, for compu- 
tations of its path showed that it remains always 
outside the orbit of Jupiter, and always inside that 
of Saturn. If it were not such an exceedingly hazy 
and tenuous object we should probably call it an 
asteroid; now we call it a comet, though really there 



is not much to choose between these two names. 
In volume this comet must have been larger than 
Jupiter, yet it undoubtedly weighed many times less 
than the moon. 

The most famous comet of all times is unques- 
tionably Halley's, so named after Edmund Halley, 
the contemporary and close friend of Newton. 
When in 1682 a brilliant comet appeared in the 
sky, Halley and Newton set out to calculate its 
path, following the rules and principles just enunci- 
ated by Newton under his law of universal attrac- 
tion. Halley then applied himself to do the same 
for notable comets that had appeared in the past, 
and found to his amazement that the orbit of the 
new comet was practically identical with those of 
the comets of 1531, and 1607. Halley quickly 
grasped the significance of this: it must have been 
the same comet, which therefore must have a period 
of about 75 years, and he immediately ventured to 
predict its next return for 1758. 

Nowadays, when it is possible to observe a comet 
for three nights and to have its complete path 
around the sun less than twenty-four hours later, it 
is hard to realize what an epoch was marked in 
astronomy by Halley's prediction. It was the first 
real test of Newton's great law of universal attrac- 
tion, the first time a scientific forecast had been 
made for the remote future and with full confidence 
of its ultimate realization. And yet little attention 
was paid to it in the beginning. Halley died in 



1742, seventeen years before his comet could re- 
appear to make his fame. As the year 1758 drew 
nearer, astronomers revived Halley's work and pro- 
ceeded to look into the matter a little more closely. 
With the great progress celestial mechanics had 
made in the meantime, it was possible for the 
French astronomers Clairaut and Lalande to take 
into account the influence of Jupiter and Saturn. 
They came to the conclusion that the comet would 
not return until the beginning of 1759, and pre- 
dicted that it would reach its closest distance from 
the sun, its perihelion, toward the middle of April, 
1759, with an uncertainty of a month. And, in- 
deed, the comet was first picked up on Christmas 
Eve, 1758, by a farmer in Saxony, and it did pass 
through perihelion on March 12, 1759. It ap- 
peared next in 1835-36, and from observations 
made at this epoch the French astronomer Ponte- 
coulant predicted its subsequent return for May, 
1910. In 1908 Cowcll and Crommelin in England 
began their new computations, using the best pos- 
sible data for the solar system and taking into 
account the perturbations not only of Jupiter and 
Saturn but of Uranus and Neptune as well. For 
more than seventy years the comet had been invisi- 
ble even in the most powerful telescopes, but the 
eye of the calculations had nevertheless followed it 
closely; had, so to speak, seen it turn in 1873 at the 
extreme end of its orbit, more than three billion 
miles from the earth and beyond even the orbit of 


Neptune, the farthest outpost of the solar system. 
The calculations had followed the comet so closely 
through all its encounters with the various planets 
that when it was ultimately discovered by Wolf, on 
September 11, 1909, it was only seven minutes of 
arc from its predicted place. Less than one fourth 
of the moon's diameter, and less than a quarter of 
an inch on the original photographic plate 1 

By looking into old records and searching Chi- 
nese chronicles, Crommelin and Cowell were able 
to trace back, with only one exception, every return 
of Halley's comet since the year 240 B.C. Notable 
appearances were those of 11 B.C. (which by some 
is identified with the star of Bethlehem), when it 
"foretold the death of Agrippa in Rome," of 1066, 
when it was regarded as a portent for the Norman 
Conquest, and of 1456, when it was immortalized in 
France on the famous Bayeux tapestry. 

At its last appearance, in April and May, 1910, 
Halley's comet showed a tail 120 degrees long; in 
the tropics the tail would rise several hours before 
the head of the comet became visible. In linear 
measure the tail must have been at least 20,000,000 
miles long. On May 19, 1910, this enormous comet, 
tail and all, passed between us and the sun, at a dis- 
tance from us of only 1 5 million miles, and in such a 
way that the tail pointed practically straight at us. 
Seen from the earth, the comet would appear pro- 
jected on the sun's disk for more than an hour. For 
all this time astronomers searched and searched in 



vain; not a trace of the comet was visible, not the 
slightest obscuration of the sun's light did the comet 
effect. In spite of its great light-reflecting power, 
the comet and its enormous tail appeared perfectly 

Another comet, that of 1680, must have passed 
very close to the sun itself, and cannot have been 
more than 600,000 miles from the sun's surface. 
The great comet of 1843 did better than that even: 
it approached the sun to within 80,000 miles of its 
surface, at which point it received more than two 
million times as much light and heat from the sun 
as we do at noon on a summer day. Yet it got 
away again by virtue of its great speed, several 
hundred miles per second, which enabled it to run 
more than halfway around the sun in less than two 
hours. Even so, when we know that solar explo- 
sions may reach a height of 200,000 miles above 
the sun's surface, and that the corona, the sun's 
outer atmosphere, extends well over a million miles 
beyond the sun's surface, it seems incredible that 
the comet got away unscathed. More than that: 
it got away in style, too, supporting a tail of more 
than 300,000,000 miles long. If this tail had been 
a rigid fixture of the comet, the other extremity of 
the tail would have swept over a distance of no less 
than 800,000,000 miles during the two hours that 
the comet itself turned halfway round the sun. This 
would correspond to a speed of more than 100,000 
miles per second, and even if such great speeds were 



possible, particles of the comet's tail travelling at 
this rate would soon become dissociated from the 
comet; they would pursue their own course in space, 
and pay very little attention, even to the sun. We 
are thus forced to the conclusion that the tail does 
not really belong to the comet; it is already lost, 
and will be left behind in space. 

The cornet of 1843 stands by no means alone in 
this peculiar behaviour; the comets of 1668, 1680, 
1702, 1882 I (discovered during a total eclipse of 
the sun), of 1S82 II, the greatest spectacle of the 
Nineteenth Century, and that of 1887 all came ex- 
ceedingly close to the sun, and what is more they 
all described similar paths round the sun. Although 
it is perfectly certain that these comets are not 
identical, it seems probable that they are all rem- 
nants of one and the same really magnificent comet, 
which once came a little too close to the sun and 
broke in pieces, leaving each piece to travel along a 
path of its own liking. 

Other comets, though not able to produce such 
a prodigiously long tail as that of 1843, are still 
remarkable for the large number of tails they had. 
The great comet of 1744, according to a fairly 
authentic account, had six enormous tails, Dunlop's 
comet of 1825 had five; but the record is held by 
Borelly's comet of 1903, a fairly small object, but 
still able to grow nine tails, as is shown on photo- 
graphs taken at Greenwich. Morehouse's comet 
of 1908 was even more freakish. On September 


/ v 




/ / / 


Above: Path of a comet around the sun, showing the lengthening of the rail as it approaches 
the sun. 

Relow: Two photographs of Comet Morehouse, taken at the Ycrkes Observatory, three and 
one-hall hours apart, showing the motion of the tail in this interval. (Since, as is evident 
from close inspection of the two photographs, the comet moves among the stars, and the 
telescope had ro he pointed directly at the comet throughout each exposure, the stars have 
formed short "trails" on the picture.) 



29th of that year it appeared perfectly normal, on 
September 30th the tail began to explode and dis- 
rupt, little by little, and on the following day it had 
entirely disappeared. A few days later, however, 
the comet, not unlike a lizard, had grown a new 
tail and it underwent several more complete meta- 
morphoses in less than a day. 

Biela's comet of 1845 neatly divided into two 
parts which, for a while, kept on travelling together, 
and at the next return of the comet, in 1852, were 
only a million miles apart. The two fragments 
disappeared from view in September, 1852, and 
they have never been seen again. They were ex- 
pected back in 1859, 1865, and 1872, and especially 
at the last-mentioned epoch the comet should have 
been very favorably located for observation. Noth- 
ing was seen, except a brilliant display of meteors 
in 1872, and again in 1885 and 1892. There seems 
no other conclusion left but that Beila's comet has 
disintegrated, worn out, and vanished, and we have 
had to write it off our books. Other comets have 
exhibited similar testimony of decay and it is gen- 
erally believed that the majority of comets are only 
temporary phenomena which, like moths at a 
candle, burn their wings when they get too close 
to the sun and ultimately disintegrate and vanish. 
Some comets seem to have hardy constitutions, on 
the other hand. There is the case of Encke's, 
which has occasionally returned with new vigour; it 
has even been accused of having been to a restau- 



rant. As yet, we have been unable to detect either 
the location or the bill of fare of this all-night 
restaurant that apparently operates in darkness. 

All this gradually brings us to the question: 
What are comets? Why do they, and they alone 
of all the bodies in the solar system, have tails? 
We have already seen that in all probability they 
are permanent members of the solar system, debris, 
so to speak, left behind when the other star, which 
caused the catastrophic birth of the planets, left the 
solar neighbourhood. Being so very light and tenu- 
ous in substance, they may have strayed away very 
far before they ultimately bowed to the stronger 
will of the sun's attraction. This may explain why 
we find comets revolving about the sun in all pos- 
sible directions, whereas the planets are practically 
confined to the same plane. That comets are very 
light we cannot doubt; even the gigantic comet of 
1843, which, tail and all, contained more than 
1,000,000,000,000,000,000,000 cubic miles of lumi- 
nous material, failed utterly and completely to make 
any impression on the planets. Its total mass was 
so small that not the slightest deflection could be 
observed in the motion of the planets. When we 
say small, we mean astronomically small, of course; 
small compared with the earth or the moon — 
expressed in tons it may still be a redoubtable num- 
ber. Donati's comet of 1858 has been estimated 
to weigh certainly less than one third of one million 



million million tons, or less than one twenty- 
thousandth part of the earth. 

When it comes to explaining what the tail of a 
comet is, we must keep in mind that a tail is really 
only a temporary fixture: all the material particles 
in the tail, which we see because they reflect sun- 
light, are already lost to the comet. Through the 
work of the Russian astronomer Bredichin we know 
that the tail consists of large numbers of exceed- 
ingly small particles which, under the action of the 
strong sunlight, or the light pressure as it is called, 
are streaming away from the comet. For ordinary 
matter this pressure of light is insignificant when 
compared with the attraction exerted by the sun. 
When things get smaller and smaller, the pressure 
of light becomes relatively larger, and for particles 
only one twenty thousandth of an inch in diameter, 
light pressure and gravitation balance; for still 
smaller particles light pressure wins. And indeed, 
as observations have shown, the small particles of 
matter driven out of the comet's head by the action 
of the sun's light and constantly accelerated by this 
same force are acquiring more and more speed as 
they get farther out; ultimately they may reach 
speeds of sixty miles a second. 

As for the physical constitution of comets, we 
believe at present that the central part of the head 
may contain large quantities of iron, but the spec' 
troscope has proved that the gases immediately 
surrounding the head are largely made up of carbon 



monoxide and cyanogen, with some sodium vapor 

In recent years our discoveries of new comets 
(i. e., not counting predicted returns of already 
known periodic comets) have averaged about five 
a year. Now it is perfectly obvious from the hap- 
hazard way in which these discoveries are made 
that we are not getting more than a fraction (and 
a small fraction at that) of the total number of 
comets each year, not even of those that are within 
reach of our telescopes. So that, conservatively 
speaking, we may put the number of new comets 
appearing each year at twenty-five. We do not 
know how long we could go on finding this many 
new comets each year before exhausting the whole 
stock, but we do know that comets, on the whole, 
wear out pretty rapidly, astronomically speaking; 
furthermore, we know that many of them have ex- 
cessively long periods of revolution, counted in 
thousands and possibly in millions of years. Since 
we have no reason to suppose that the present time 
is peculiarly plentiful in comets, and since we know 
further that the solar system must have existed for 
a very long time, probably millions of millions of 
years, it follows that the total number of comets 
in the solar system must be enormous. There must 
be billions, possibly trillions of them, and no matter 
how small and light they may be individually, their 
total mass may be considerable. It may very well 



be comparable to or even exceed the mass of the 

If we consider comets as the rather numerous 
but vagrant and independable members of our solar 
system, we can go a step farther, and take a look 
at meteors or "shooting stars," which obviously sur- 
pass comets in all these characteristics. Meteors 
constitute our daily visitors from the great beyond; 
they form our only direct contact with the material 
world outside the earth, and as such are of the 
greatest importance. Meteors are the only things 
that can give us reliable information concerning the 
emptiness of space, and yet we know very little 
about them. We know that they are generally 
small particles of matter that come through our 
atmosphere with terrific speeds, heat up through 
the friction they encounter, and when they are hot 
enough to emit light appear to us as "shooting 
stars." Usually they are very small, sometimes no 
larger than a grain of sand, and their glory lasts 
but a short time; in a few seconds they are burned 
out, but during these few seconds they are raised 
to a temperature of 6,000 degrees centigrade, and 
give off energy at the rate of 4,000 horsepower. 
Occasionally, meteors are so huge that even their 
racing through the whole of our dense atmosphere 
does not suffice to burn them up, and they strike 
the ground with a terrific roar, often accompanied 
by a hail of smaller particles. The largest meteor 
thus far found was discovered by Peary in Green- 


land, and is now on exhibition in the American 
Museum of Natural History, in New York; it 
weighs no less than thirty-six and one-half tons. 
There are hopes that we may find a bigger one yet 
in the near future, for scientists are now pretty well 
convinced that "Coon Butte," a great crater in 
Arizona near Canon Diablo, was formed by the fall 
of a huge meteorite. It is a crater in the middle 
of a desert 4,000 feet in diameter, with walls rising 
ISO feet above the surrounding plain and descend- 
ing precipitously to the floor, 600 feet below. 
There are no signs of volcanic action in the neigh- 
borhood, and all around the crater, even imbedded 
in the walls of it, there have been picked up frag- 
ments of meteorites, or stones containing a mixture 
of iron and nickel. We have learned through chem- 
ical analysis that many meteors consist of a mixture 
of very pure iron with some nickel or cobalt, and 
such a combination of metals is rarely found else- 

Since meteors form our only means of direct 
communication with the stars, their chemical com- 
position is of special interest; it furnishes us with 
the only tangible information concerning the chem- 
ical make-up of the stars. It is fascinating to think 
that when we take up a piece of a meteor, we may 
be holding in our hand a piece of matter belonging 
to a star which has exploded millions of millions 
of years ago. How can we be so certain that 
meteors come from the stars? you may ask. Why 



cannot they too, like comets, all belong to our solar 
system? For this belief we have conclusive evi- 
dence in the velocities of the meteors. If they 
really were permanent attendants of the sun, none 
of them could have speeds exceeding 26 miles per 
second. Even if a meteor fell into the sun from 
an infinite distance, it could not attain a speed 
higher than 26 miles a second by the time it crossed 
the earth's path around the sun. Direct observa- 
tions of meteors have revealed that at least some 
of them have speeds as high as 45 miles per second, 
putting them well outside the class of solar meteors. 
Among the slower meteors, those that travel 
with speeds less than 26 miles per second, and are 
therefore permanent members of the solar system, 
we find many great aggregations apparently travel- 
ing the same path around the sun, and some of 
these can be identified with the orbits of comets. 
We have already seen that Biela's comet has dis- 
appeared and given way to a swarm of meteors; 
likewise, we know of meteors traveling the same 
path as Halley's comet. During the summer of 
1927, when Pons-Winnecke's comet was unusually 
close to the earth, Yamamoto at Mukden, Man- 
churia, saw and photographed an exceedingly bright 
meteor, many times brighter than the full moon, 
which was shown by the calculated path to have 
been connected with that comet. A swarm of 
meteors travelling round the sun in such a manner 
that they meet the earth annually in August, thus 



causing a brilliant display, a "shower" of meteoric 
fireworks, are probably the remains of some pre- 
historic comet. 

While the chemical analysis of meteors furnishes 
a clue to the composition of the stars, the abundance 
of meteors in space gives an indication of the empti- 
ness of space. That is to say, if we knew how many 
meteors of each different size strike the earth every 
day, we should be able to tell how much matter 
there really is in empty space, a thing all astron- 
omers are very anxious to find out. Unfortunately, 
our knowledge in this instance is deplorably incom- 
plete, and all we can say is that probably not less 
than ten million meteors strike the earth every day. 
As an interesting by-product of this, it follows that 
the earth is getting heavier all the time, and must 
therefore gradually slow down in its rotation. Since 
the added mass of the meteors would also increase 
the sun's attraction, it would tend to make the 
earth's orbit smaller, and thus shorten the year. 
The day would be getting longer and the year 
shorter, but, as Young has calculated, this change 
would not amount to more than one one-thousandth 
of a second in a million years. 

We have now reached the end of the sun's do- 
main. We have passed beyond the orbit of Nep- 
tune, the furthest outpost among the regular mem- 
bers of the sun's family, we have passed beyond 
even the most far-flung of the comets and streams 
of meteors, we have surveyed all that shines by the 



reflected glory of the sun. All this will fade into 
insignificance as we continue farther. As we go 
beyond, into the vast stretches of interstellar space, 
and look back, we see only the sun. Only the sun 
remains, and even the sun is now reduced to the 
level of merely one among countless millions of 
other stars. 

Chapter IX 


" 'The time has come,' the walrus said, 
'To talk of many things: 
Of shoes and ships and sealing wax, 
Of cabbages and kings.' " 

— Carroll. 

In the solitude and silence of the nocturnal sky the 
stars give the only manifestation of a something 
existing throughout the unfathomable depths of 
space; were it not for them, the whole realm of 
space would be in the grip of perpetual darkness. 
Small wonder indeed that the ancients saw the heav- 
ens as the celestial stage upon which the gods per- 
formed and wielded their magic, using stars and 
planets as pawns to dictate their caprices to human- 
ity. Out of the mists of antiquity there come to 
us whispers from the book of Job, from Homer, 
concerning the symbolism by which the sky of pre- 
historic times was identified — a symbolism laden 
with portents boding good or ill for human kind. 
It was thus that the constellations originated, as 
also the star names, and with them the first descrip- 
tion of the heavens. Astronomy of to-day has, of 




necessity, out-grown this nomenclature of yester- 
day; no longer do we refer to the stars by fanciful 
names such as "the star on the left claw of the 
Scorpion" or "the red star on the eyelid of the 
Bull," but rather designate them by letters and 
numbers, an identification much less ambiguous and 
less subject to the individual imagination. Thus 
Sirius, the brightest star in the sky, is astronomically 
called Alpha Canis Majoris, indicating that it is the 
brightest star in the constellation of the Greater 
Dog; the Pole Star, for similar reasons, is called 
Alpha Ursa; Minoris, or in this special case, Polaris. 
For fainter stars the designation of Argelander is 
employed, in whose system the sky is divided into 
180 narrow zones one degree wide running parallel 
to the equator, and each star given a number in the 
zone to which it belongs. 

Within the past forty years a number of observa- 
tories have cooperated to map and catalogue, by 
means of photographic plates, the precise positions 
of three to four million stars, distributed over the 
whole sky. At present about three quarters of this 
work has been completed. The number of very 
faint stars is so vast that it is of little use to devise 
elaborate systems of identification; usually only a 
rough position in the sky is given, and, for a few 
stars that merit special attention, small charts of the 
region immediately surrounding them are prepared 
and filed away. In recent years photography has 
completely supplanted the older method of mapping 


the sky by plotting the positions of all stars ob- 
served on a chart, not only because the new method 
is much less laborious, but principally because a 
direct photograph constitutes a lasting record of 
the sky, often invaluable for later investigation. 
Through the great foresight of Pickering, the Har- 
vard Observatory was one of the earliest in the field 
of astronomical photography. Since 1887 it has 
continuously and systematically photographed the 
entire sky from its two stations in Cambridge and 
Arequipa, Peru. 1 The photographic library thus 
accumulated numbers to date almost 400,000 plates, 
and contains such a wealth of information that 
we may well say that its value will not be exhausted 
as long as the photographic film remains intact. 

A first arrangement of the stars immediately sug- 
gests itself when we observe them differing consider- 
ably in brightness. The Greeks already distin- 
guished several classes among them ; they named the 
brightest stars in the sky, stars of the first magni- 
tude, and stars just visible to the naked eye, stars 
of the sixth magnitude, the others being graded in 
between. When telescopes revealed increasingly 
large numbers of still fainter stars, these had to be 
fitted into the system by assigning them to the 
seventh magnitude, etc., greater numbers indicating 
smaller brightness. Lack of uniformity between 
estimates made by different observers naturally led 

'This southern branch of the Harvard Observatory was moved 
to Bloemfontein, South Africa, in 1927. 



to the establishment of a rigorous mathematical sys- 
tem of grading stellar brightness. In this system 
the faintest stars visible to the naked eye are still 
called stars of the sixth magnitude, but stars of the 
fifth magnitude are defined as those 2.512 times 
brighter than the former, stars of the fourth magni- 
tude as 2.512 times brighter again. This number 
2.512 has been chosen for the reason that now five 
magnitudes difference in brightness corresponds ex- 
actly to a ratio of 100 times in actual light, thus 
making a first-magnitude star exactly 100 times 
brighter than one of the sixth. Insistence on pre- 
cision subsequently resulted in the use of decimals, 
so that we now speak of a star of magnitude 5.52; 
in addition, it has been found that there are a few 
stars in the sky brighter than the first magnitude, 
such as Vega, Arcturus, and others. Not to upset 
the mathematical basis of magnitude scales, these 
had to be labelled stars of zero magnitude, and the 
two stars still brighter, Canopus and Sirius, were 
designated as of the magnitudes minus one and 
minus two respectively. Exact measurement has 
shown that the light we receive from Sirius is equal 
to that of a candle at a distance of 200 yards, while 
an average first-magnitude star corresponds to a 
candle at 600 yards. Continuing in the direction of 
greater brightness, it has been found that the light 
of the full moon corresponds to a stellar magnitude 
of —12, that of the sun to -27. Principally owing 
to the efforts of Pickering at Harvard, we are now 


in possession of accurately measured magnitudes or 
all stars brighter than the sixth magnitude, and of 
tens of thousands of stars fainter than that limit. 

Telescopes with a light-gathering power much 
greater than that of the human eye will naturally 
show stars much fainter than the sixth magnitude. 
An opera glass of 1-inch aperture shows stars of 
the ninth magnitude, or 15 times fainter than those 
just visible to the naked eye; the 36-inch telescope 
of the Lick Observatory shows stars down to the 
seventeenth magnitude, the 100-inch at Mount 
Wilson stars of the nineteenth magnitude. Long- 
exposure photographs made on sensitive plates will 
reveal still fainter stars, and it seems safe to con- 
clude that with the 100-inch telescope at Mount 
Wilson we could photograph, if we wished, all stars 
brighter than the twenty-first magnitude; that is, all 
stars not more than 1,000,000 times fainter than 
the limit of visibility of the naked eye. 

Stars of this extreme faintness, however, exist in 
such multitudes that it seems at present not worth 
the effort to record them all. The chief thing we 
want to know about them, namely, their approxi- 
mate total number, may well be derived by other, 
and simpler, though indirect means. Statistics here 
come to the aid of the astronomer, and the problem 
is solved in much the same way as the medical man 
solves his public health problems. When the latter 
wants to determine the influence of a certain dis- 
ease, he is usually unable to gather reliable informa- 



tion for the whole country and has to be content 
with obtaining them for certain communities where 
he knows he is getting the whole truth and nothing 
but the truth. Similarly with the astronomer in 
his attempts to count the stars in the sky. He is 
unable to count them all individually, and therefore 
takes photographs of a large number of restricted 
areas distributed all over the sky and counts the 
number of stars per square inch on these plates. 
Knowing how large an area of the sky is repre- 
sented by a square inch on his plates, he then takes 
the average of his counts, multiplies by the whole 
area of the sky, and the result will give him the de- 
sired number. For the brighter stars, of course, 
this procedure need not be used, since they can be 
counted directly; only the fainter stars require 
statistical methods. 

The whole sky contains twenty stars which are of 
the first magnitude or brighter, some sixty stars 
brighter than the second magnitude, five thousand 
stars brighter than the sixth, and more than one 
million stars brighter than the twelfth magnitude, 
the total number of stars observable with the ioo- 
inch telescope, stars brighter than the twenty-first 
magnitude, is probably not less than one-thousand 
million. In spite of the rapid increase in numbers, 
the fainter stars do not contribute much to the total 
light of the stars in the sky, and this total amount 
of light is known with far greater precision than is 
the number of stars. Newcomb first made an esti- 


mate of the total brightness of the sky; more recent 
measures by Van Rhijn indicate that total starlight 
is equal to about 1200 stars of the first magnitude, 
or 120 stars as bright as Sirius, and less than 1 per 
cent, of the light of the full moon. 

Apart from their difference in brightness, the 
stars may be seen with the naked eye to differ in 
colour. Some stars appear white, others yellow, 
while a few even merit the adjective red. Vega and 
Sirius arc examples of the white class, Capella and 
Arcturus of the yellow, Betelgeuse and Antares of 
the red group. Since the stars are self-luminous 
bodies, a difference in colour immediately suggests a 
difference in temperature, and brings to the mind 
what happens when we heat up a bar of iron. At 
a comparatively low degree of heat the iron will 
appear red hot, at a higher temperature it seems 
pale yellow, while in the Bessemer converter or in 
the electric furnace it appears intensely luminous and 
white. The same holds for the stars, as indeed the 
spectroscope has shown that the white stars are hot- 
ter than the yellow or red stars. When the light 
of a star, concentrated by means of a telescope, is 
passed through a prism and spread out into a 
colored band of light, a spectrum, it is found that 
such a star-spectrum, like that of the sun, is not a 
continuous band of colours but contains fine black 

The first observations of star spectra were made 
by Fraunhofer as early as 1824, and later more 



extensively by Secchi and Huggins. It was soon 
found that the stars must be incandescent bodies 
surrounded by a cooler layer of gases which produce 
the black lines, and thus must be very similar in 
structure to the sun. When the black lines were 
studied in detail they were found to be identical 
with lines ascribed to chemical elements on earth; 
furthermore, it appeared from Secchi's work that, 
with very few exceptions, the spectra of all stars 
could be arranged in four groups. The first of 
these groups is typified by the presence of only a 
few black lines, principally those of hydrogen, while 
the second group is marked by numerous lines which 
may be proved to be of metallic origin and due to 
calcium, iron, magnesium, and other well-known 
metals. The white stars all belong to the first 
group, the yellow stars to the second. The red 
stars are divided over the other two groups and 
show shaded, banded regions due to titanium oxide, 
or to carbon compounds, in their spectrum. 

As in many other fields, so in the field of stellar 
spectroscopy visual observations were soon aban- 
doned in favor of photographic spectroscopy which 
operates not only more efficiently but also pro- 
duces a record which may be examined at ease in 
the laboratory rather than in an uncomfortable 
position behind the telescope. As in the case of 
direct photographs of the sky, Pickering at Har- 
vard became a pioneer in this work, and photo- 
graphic observation of star spectra on a wholesale 



basis has led to a classification of much greater de- 
tail than Secchi's. In the system developed by Miss 
Maury, Mrs. Fleming, and Miss Cannon ten prin- 
cipal types of stellar spectra are recognized. Classes 
O, B, and A comprise the former first type, class O 
stars showing very few lines, class B, lines due 
mainly to hydogen and helium, while in class A the 
hydrogen lines alone predominate the whole spec- 
trum. The second type of Secchi is divided into 
classes F, G, and K, where, in the order mentioned, 
the hydrogen lines become less and less conspicuous 
and the metallic lines, principally those of calcium 
and iron, grow more intense. Class M takes the 
lion's share of former type III, class N that of 
type IV, while classes R and S contain but a few 
stars. Miss Cannon and the late Professor Pick- 
ering at Harvard have already classified more than 
250,000 stars on this system, published in a cata- 
logue named in honor of Henry Draper, another 
pioneer of stellar photographic spectroscopy in 

Originally it was believed that the difference in 
stellar spectra was due to difference in chemical 
composition among the stars. Those showing strong 
hydrogen lines were assumed to consist largely of 
hydrogen, at least insofar as their upper layers were 
concerned. Researches by Lockyer and subse- 
quent application of modern physics by Megh Nahd 
Saha of Calcutta led to the conclusion that the 
chemical composition of the stars is of negligible 



influence on the spectrum, but that the difference in 
temperature alone is responsible for the fact that in 
some stars the lighter gases, hydrogen and helium, 
seem to come to the fore, while elsewhere the heav- 
ier metals preponderate. 

This explains at the same time why the vast ma- 
jority of stellar spectra arrange themselves so 
neatly into one sequence: classifying the spectrum 
of a star is reduced to a determination of its tem- 
perature, and, except for a few cases of strange be- 
haviour, the temperature of the outside layers of a 
star completely dominate the larger features of the 
spectrum. From facts known in the laboratory 
and from experiments conducted on earth with 
atoms of the different chemical elements it has be- 
come possible to conclude that the surface tempera- 
ture of the coolest stars visible, those belonging to 
Secchi's type IV, and Miss Cannon's class N, is 
about 3,200 degrees Fahrenheit. On earth we 
should still call it "white heat"; in the sky we call 
stars of this type "very red," since they appear dull 
red in comparison with other, much hotter stars. 
Going up in the scale of temperatures, we next find 
stars of class M, such as Betelgeuse, whose temper- 
ature is about 4,800 degrees Fahrenheit. That of 
the sun, which shows a spectrum of class G, is about 
10,000 Fahrenheit, while the white stars, Sirius, 
Vega, and others, are well over 20,000 degrees. 
The still hotter stars, belonging to class B and O 
and having a temperature of 35,000 degrees or 


more, even appear blue in comparison with sunlight. 

A very interesting application of the theory of 
stellar temperatures lies in the calculation, by theo- 
retical means, of the diameters of the stars. It has 
long been known, from the researches of Stefan, 
Boltzmann, and Wien, that the total amount of 
light radiated per square inch by a hot body depends 
almost entirely upon the temperature; for a "com- 
plete radiator" this amount of light can be ac- 
curately predicted. Once we have measured the 
temperature of a star from its spectrum, we can 
thus calculate the amount of light given off per 
square inch by that star; a comparison of this with 
the light we receive from the star on earth will 
yield a resultant value for its apparent size. 

It was in this way that Hertzsprung first pre- 
dicted the sizes of the stars, as early as 1905, when 
there seemed to be no hope whatsoever of experi- 
mental verification: the largest diameters predicted 
were no more than one twentieth of a second of 
arc, or as large as a dime seen from a distance of 
forty miles. Thanks to an instrument invented by 
Fizeau, greatly improved and successfully applied 
to astronomical measurements by Michelson, the 
interferometer, it has become possible in recent 
years to make actual measures of the diameters of 
the stars. Using the 100-inch reflector at Mount 
Wilson, Pease has already measured the apparent 
angular diameters of seven of the brightest red 
stars. These are especially suitable for such meas- 



urements since, on account of their low temperature, 
their surface brightness — that is, the total light 
emitted per square inch of the surface — is exceed- 
ingly low. Therefore, if they appear bright to us, 
it can only be because they have a very large sur- 
face; thus, they will appear to us with a large angu- 
lar diameter, "large" meaning one twentieth of a 
second of arc. For yellow stars the apparent di- 
ameter is already much smaller: Arcturus, for ex- 
ample, is no more than one fiftieth of a second of 
arc in size, and Capella which, square inch for 
square inch, gives as much light as the sun, but 
which has an actual diameter ten times greater than 
that of the sun, appears to be no more than seven 
one thousandths of a second of arc in size. For 
the white and blue stars the predicted diameters 
are so much smaller that there does not even now 
seem any possibility of determining them by meas- 

Actually, these apparently small diameters of the 
stars correspond to enormous sizes. Even if Betel- 
geuse appears to us only as large as a dime at a 
distance of forty miles, we must remember that 
Betelgeuse is not forty miles but 25 million million 
times forty miles distant, and is therefore 25 mil- 
lion million times larger than a dime in diameter, 
about 250 million miles, 300 times larger than the 
sun. If the sun, with its attendant planets, were 
placed in the centre of this gigantic star, the whole 
orbit of our earth would lie well inside it, and even 


that of Mars would not be very much outside. The 
total bulk of Betelgeuse must be about twenty-five 
million times larger than that of the sun; on any 


one thousand times more rarefied than the gases in 
our atmosphere; a vacuum we should call it on 
earth! Pease's measures were a complete vindica- 



reasonable assumption of its weight, the mean den- 
sity must be about one thousandth of our atmos- 
phere. Taken on the whole, this immense red star 
must be composed of tenuous material more than 




tion of Hertzsprung's early predictions, and 
astronomers feel confident now that they are not 
only on the right track in their theories concerning 


the temperatures and the total light output of the 
stars, but also that they can predict with reason- 
able certainty the diameter of any star for which the 
temperature is known, and whose light has been 
measured. The calculated linear diameters ex- 
pressed in miles are still somewhat uncertain, and 
a large part of this uncertainty is due to lack of 
precision in our knowledge of the distances of the 

With the problem of stellar distances we are 
really entering upon one of the most vital points in 
our study of the universe. On our knowledge of 
distances rests our conception of the structure of the 
universe, and it may well be said that the determi- 
nation of the distances of the stars constitutes the 
most urgently necessary task and at the same time 
the most difficult one in all practical astronomy. 
We have previously mentioned the difficulties en- 
countered in the determination of the distance of 
the sun, difficulties arising simply from the fact that 
our baseline for triangulation was no more than 
8,000 miles long, a length totally insufficient to 
measure with accuracy a distance of 93,000,000 
miles. Now that we turn to the stars, however, 
even that distance, a "mere 93,000,000 miles," 
fades almost into insignificance in view of the fact 
that the nearest star is 275,000 times 93,000,000 
miles away. To try and measure such distances of 
trillions of miles from a baseline no more than 8,000 
miles long is palpably ridiculous. We must look 



for other means and other baselines, and fortu- 
nately, we do not have to look far. 

In the course of its annual motion around the 
sun, the earth describes a circle with a radius of 
93,000,000 miles; consequently, the position in 
space from where we terrestrials observe the stars 
does not remain the same. If we look at the sky 
now, and again in six months' time, we shall then 
be viewing it from a point, 186,000,000 miles dis- 
tant from the place of first observation. We shall 
see that certain very minute changes have taken 
place. We all know that when we are riding in a 
train and looking out of the window we see the 
landscape undergo slow but progressive changes. 
If we fix our eyes on a church spire, for instance, 
and on a distant mountain range behind it, we no- 
tice that the mountains seem to travel along with 
us but the church spire appears to lag behind. In 
its path around the sun the earth does not travel in 
a straight line, but follows an almost circular orbit. 
To make our analogy more exact, therefore, we 
should have to make our train go in a circular track 
around a centre, the sun, and view our church spire, 
as one of the nearest stars, against the mountain 
peaks, acting as the background of very distant 
stars. Observing from the train, we should see the 
church spire move to and fro as outlined against 
the mountains. If, for instance, we look at the 
church spire while the train is going eastward, the 
mountains would "follow" us, and the church spire 


would appear to be moving westward with respect 
to the mountains, while at the opposite side of the 
circular track, where the train would be running 
westward, the church spire would again appear to 
move in the opposite direction with respect to the 
mountains, and appear to go eastward. 

Exactly the same motion may be observed in the 
nearer stars when compared with those more dis- 
tant: stars situated in a position analogous to that 
of the church spire, that is to say, stars lying nearly 
in the plane of the earth's orbit, would be seen 
swaying to and fro in the period of one year. Stars 
seen in a direction perpendicular to the plane of the 
earth's orbit appear to describe a small circle, when 
compared with the more stationary background, 
stars in intermediate positions describing ellipses. 
As is easily verified from the diagram, the diameter 
of such a circle and the largest dimension of such 
an ellipse, or the full length of the line along which 
the star appears to sway, are equal in every instance 
to the angle subtended by the diameter of the 
earth's orbit as seen from the star in question. Half 
this angle, or the angular size of the radius of the 
earth's orbit, of one astronomical unit, as viewed 
from the star, is called the parallax of the star. 
We notice from the diagram that the parallax of a 
star decreases as its distance from us increases. 

This apparent motion of the stars in the sky is 
a necessary consequence of the revolution of the 
earth around the sun, and the fact that such a mo- 






Hon had not been observed was immediately used 
by the opponents of Copernicus as an argument 
against his system. In the system of Ptolemy the 
earth occupied the fixed centre of the universe, and 
no motions of this kind could be observable among 
the stars; in that of Copernicus, only the sun re- 
mained fixed in space while the earth described a 
circle around it, and thus, as a result, or rather as a 
reflection, of this motion of the earth, all the stars 
should annually describe circles or ellipses in the sky. 
Copernicus, however, had himself foreseen this 
difficulty and had met it beforehand: the distances 
of the stars are so vast that their annual motion in 
the sky is insensible. Copernicus was right indeed, 
but we must admit that it was an act of faith on the 
part of his followers to accept it. For nearly three 
hundred years the adherents of this theory had to 
exercise this faith, as it was not until the middle of 
the Nineteenth Century that their faith was sub- 
stantiated by facts. Almost simultaneously Struve 
in Russia, Bessel in Germany, and Henderson in 
South Africa succeeded in measuring the parallax 
of a star. It then became obvious that with the 
older and less precise methods of observation no 
parallaxes could possibly have been measured. If, 
returning to our analogy of observing a church spire 
from a train window, we should have to replace the 
train running on a circular track by a toy train run- 
ning in a circle of one-foot radius, and put the 



church spire at a distance of fifty miles, even then 
the church spire would represent only the very near- 
est of all stars. 

When measuring the distances of the stars, the 
astronomer looks around for a new yardstick with 
which to express his distances. In the laboratory 
we measure in inches and feet, while the surveyor 
uses yards and miles. When we leave the earth be- 
hind us, a mile is of little use any more. Even the 
moon, the nearest of all celestial objects, is 240,000 
miles distant, the sun no less than 93,000,000. In 
the solar system we circumvent the problem of using 
large numbers by taking this distance of the sun, 
93,000,000 miles, as a new unit, calling it one astro- 
nomical unit. It does service throughout the plan- 
etary system, but no sooner do we pass beyond the 
frontier of the sun's domain than it sinks into in- 
significance. The nearest of all stars is 275,000 
astronomical units away. We are obliged to devise 
a new unit for measuring stellar distances, and we 
choose that distance from where the radius of the 
earth's orbit would appear to be one second of arc. 
The parallax of a star at that distance would be one 
second: and we call this new unit a parsec, a name 
first suggested by Turner. Seen from this distance, 
the orbit of Neptune, which constitutes the outer- 
most boundary of the solar system, would appear 
no larger than a silver dollar seen from a distance 
of 200 yards. 



Another way out of the difficulty lies in making 
use of the speed of light. Traveling at the rate of 
186,000 miles per second, a ray of light could go 
round the earth seven times in one second. It takes 
only eight minutes, nineteen seconds for a light ray 
to reach us from the sun, four hours to penetrate 
from the sun to Neptune. A ray of light could 
traverse the whole domain of the solar system in 
eight to nine hours, but it would require four and 
one-half years to reach the nearest star. Here, 
then, we have another chance of establishing a new 
and convenient yardstick to guage the depths of 
space : the light year, the distance covered by a ray 
of light in one year, and equal to a trifle more than 
one third of one parscc. Expressed in miles, a 
light year equals a number with twelve ciphers, 
nearly 6,000,000,000,000 miles while a parsec 
equals almost 18,000,000,000,000 miles. 

A star one hundred light years distant has a par- 
allax of one thirtieth of a second of arc; this means 
that, on photographic plates taken with the best 
telescopes available, the total displacement of the 
star due to parallax is no more than one five-thou- 
sandth part of an inch. If we wish to have any feel- 
ing of security in the determination of such a dis- 
tance, the least we may ask is that the uncertainties 
due to measuring shall not exceed one quarter of 
the quantity measured, that is to say, shall not ex- 
ceed one twenty-thousandth part of an inch, less 
than half of one hundredth part of the thickness of 


an ordinary playing card. With our best instru- 
ments, and taking all the precautions possible, we 
can barely attain such accuracy, but it is plain that 
we cannot place great faith in a single measure of 
the distance of a star with a still smaller parallax, 
that is of a star more than 100 light years away. 
Since the pioneer work of Schlesinger at the Yerkes 
Observatory, and through the concerted effort, dur- 
ing the past twenty years, of several observatories 
in America, notably the McCormick, Allegheny, 
and Mount Wilson Observatories, and in England, 
the Royal Observatory at Greenwich, we now have 
at our disposal reliable parallaxes of about 2,000 
stars. However, it is only too evident that the end 
is in sight. In fact, no sooner had determinations 
of stellar distances, perhaps the most fundamental 
observations in astronomy, begun, than we became 
aware of the limitations set by the problem itself. 
We may be able, ultimately, to determine the exact 
distances of all stars nearer than 300 light years, 
and perhaps, through accumulation of large num- 
bers of observations made at different observa- 
tories, be able to push this limit as far as 1,000 light 
years, but we shall not, as we now see it, ever be 
able to extend our researches beyond. That part 
of the universe beyond 1,000 light years will remain 
a closed book to us as far as direct determinations 
of distance are concerned. 

However, when direct methods fail, we search 
for indirect ones, methods based upon the physical 



properties of the stars themselves; in the first place, 
the intrinsic brightness. This is the "real" lumi- 
nosity of the star, not its apparent brightness, as 
we see it in the sky, since the apparent brightness is 
dependent upon the star's distance. As soon as 
the distance is known, the intrinsic luminosity of a 
star may be found, since its apparent brightness is 
always easily measurable. In order to have a stand- 
ard of comparison, astronomers have agreed to 
define the intrinsic brightness, or the absolute mag- 
nitude, as it is usually called, as the brightness 
under which the star would appear if it were seen 
from a distance of ten parsecs, or thirty-three light 
years. Our sun, removed to this distance, would 
shine as a star of the fifth magnitude, four million 
million times fainter than it appears to us now. 
Betelgeuse, on the other hand, would appear much 
brighter than it is now, and of the magnitude —3, 
and Rigel, the other bright star in Orion would 
again outshine Betelgeuse, being of magnitude —6. 
Thus we see that actually these two stars surpass 
the sun more than 1,000 and 15,000 times in light, 
respectively, although, owing to our proximity to 
the sun, our sun appears to outshine them both by 
more than sixty thousand million to one. 

As soon as the intrinsic luminosities of the stars 
became known, even though the early measure- 
ments of stellar distances were crude compared to 
the modern ones, a curious relation between lumi- 
nosity and colour, or temperature, was noted. When 



arranging his stars in order of their colour, or spec- 
tral class, Hertzsprung found that the whiter and 
hotter stars were, on the whole, more or less of 
the same luminosity, while the yellow stars appeared 
to divide themselves into two groups, a high-lumi- 
nosity group, and a low-luminosity group, the dis- 
parity in intrinsic brightness becoming greater as 
the stars increased in colour. Thus, the stars of the 
same spectral class as the sun are seen to cluster 
around two values of the absolute magnitude, one, 
which is equal to that of the sun, the other rating 
one hundred times brighter. Stars as yellow as 
Arcturus and Aldebaran form two groups that 
differ a thousandfold in brightness, while the very 
red stars such as Betelgeuse and Antares appear to 
have the choice between two groups differing in 
brightness as much as from ten thousand to one. A 
further remarkable fact is that the stars constitu- 
ting the brighter of those groups are of about the 
same luminosity for all different spectral classes — 
that is, approximately one hundred times as bright 
as the sun. Russell has very happily suggested the 
name "giants" for them. The other stars, the 
"dwarfs" of Russell's system, decrease very rapidly 
in intrinsic brightness as their colour becomes more 

We now begin to see an indication of the way in 
which we can use these facts to derive stellar dis- 
tances. If, knowing the apparent brightness of a 
star, we can calculate the intrinsic brightness, once 



the distance is known, then, conversely, we can esti- 
mate the distance when we know the intrinsic lumi- 
nosity. If, furthermore, we observe the color of 
the star, or what is really equivalent, its spectrum, 
it is then necessary only to know whether the star 
in question is a "giant" or a "dwarf," and the prob- 
lem is solved, since, as we have seen, the intrinsic 
luminosity of the star is then approximately known. 
The actual solution again finds its origin in a remark 
made by Hertzsprung in 1907, when he found that 
in the spectra of the yellow stars a certain black 
line, due to the chemical element strontium, was 
more conspicuous for the giants than for the dwarfs. 
Working in 1913 at the Mount Wilson Observatory, 
Kohlschtitter developed a method by which it be- 
came possible to determine the absolute magnitude 
of the stars from the relative strength of certain 
lines in their spectrum. Kohlschiitter's method was 
greatly elaborated and extensively applied by Adams 
and his collaborators at Mount Wilson, and by 
Young and Harper at Victoria. Their labours have 
resulted in the determination of several thousand 
"spectroscopic absolute magnitudes," and hence have 
greatly augmented our knowledge of stellar dis- 
tances. The great advantage of the spectroscopic 
method for deriving distances is that, if once the 
intensities of the lines in stellar spectra have been 
calibrated with known intrinsic luminosities, the 
distance of any star that can be observed spectro- 
scopically may be determined. Instead of standing 



helpless with our triangulation methods before an 
inadequate baseline of only 93,000,000 miles, we 
now approach the problem well equipped with a 
powerful spectroscope. Extending our gauges of 
the depth of space had become tantamount to in- 
creasing the light-gathering power of our telescopes. 
Hence the restriction of 1,000 light years as the 
greatest distance measurable is removed at once; 
with the 100-inch telescope a distance of 50,000 
light years could be measured with ease. 

Chapter X 


"Ships that pass in the night." 
— Longfellow. 

The "crystal sphere of fixed stars" is no more. To 
the peoples of ancient times it may have repre- 
sented the acumen of immobility and repose in its 
arrangement of perpetual order; for us the indi- 
vidual stars have come to life. To us they repre- 
sent glowing suns, great globes of matter in con- 
stant turmoil inside and, subject as they are to the 
law of gravitation, in chaotic motion in space, a 
bewildering picture of perpetual unrest. Indeed, 
when accurate observations of the positions of the 
stars are made and compared with others made in 
the past, it is found that no stars really remain 
fixed; they all move. Some stars move so rapidly 
that their displacement among slower, more dis- 
tant stars may easily be detected; others seem so 
inert that many years are required to reveal their 
motion. Sirius, for example, moves more than a 
second of arc per year, a motion that would carry 




it across the moon's disk in 1,400 years, while Alpha 
Centauri is even more rapid and traverses a dis- 
tance equal to the moon's diameter in 500 years. 
Astronomers call this a very large motion; there 
are, in fact, but few stars in the sky that move with 
such speed. To the naked eye, on the other hand, 
even this large motion would be utterly indiscern- 
ible, as the following illustration may show: 

Imagine that, one night, we put a dime on the 
sidewalk of lower Broadway, in New York City, 
and place a docile firefly in the center. Suppose 
further that we ascend to the top of the Woolworth 
tower, and, taking for granted that the dime as well 
as the firefly are still in place, observe them from 
this distance of 800 feet. If we are able to distin- 
guish the firefly and see it crawl from the middle 
of the dime to the edge in one year, we have ob- 
served a motion just as fast as that of Alpha Cen- 
tauri in the sky. To the unaided eye it may not 
seem impressive, but the astronomer, armed with a 
good telescope, could easily determine it in ten 

Actually, too, this motion is very rapid, for, al- 
though it does not appear to us to be more than 
halfway across a dime per year, that is less than 
one thousandth of an inch per day, when we view 
it from a distance of 800 feet we should not forget 
that Alpha Centauri is not 800 feet but twenty-five 
trillion miles distant from the earth. In order to 
cover the same angular distance in the sky, Alpha 



Centauri must fly with a proportionately greater 
speed, equal to 14 miles per second, as exact calcu- 
lation shows. 

These angular displacements in the sky are called 
proper motions, to indicate that they are inherent 
in the stars themselves; they are accurately known 
for all the stars visible to the naked eye, and for 
many thousands of others. The largest known 
proper motion is that belonging to a faint star in 
the constellation Ophiuchus, discovered by Barnard, 
and reaches the value of more than 10 seconds of 
arc per year. Perhaps it should be stressed that 
large proper motions may be due to two causes: 
either the star has actually a large motion, ex- 
pressed in miles per second, or the star is very near 
to us. Barnard's star belongs to both classes, as 
it is the second nearest of all stars in space and 
also has a linear motion of 56 miles per second. 
Take, on the other hand, the two stars Sirius and 
Antarcs, both of which are actuated by a speed of 
10 miles per second. Sirius is only 9 light years 
away, but Antares 360, with the result that Sirius 
appears to move 40 times faster than Antares. 

It is, of course, unreasonable to suppose that all 
the stars are moving sideways as seen from the 
earth, for in general any moving object, when 
viewed from a fixed place, will appear to be ap- 
proaching or receding as well as moving sideways. 
This other motion, the velocity of approach or of 
recession, cannot be determined from observations 

Photograph Harvard Ofatrvatory 

P I. A T K Y [ I 

Hie southern Milky Way, showing the Southern Cross, the Coal Sack, Alpha and Beta 
Centauri. The Southern Cross, in the centre of the plate, has ceased to be conspicuous as 
a cross, lor the sole reason that one of its stars is very red and does not register well on the 
photographic plate. The lowest and brighten star of the cross may be seen immediately 
to the right of the Coal Sack, the black spot in the centre of the plate. The next brightest 
star is to the left and above, while the red star is straight above the brightest star, and the 
fourth above and to the right. The bright star near the left-hand edge of the plate is Alpha 
Centauri, while Proxima, the nearest known of all stars, and a companion in space of 
Alpha, is one of the many faint stars indicated by the arrow. The distance in the sky 
between Proxiina and Alpha is almost three degrees, corresponding to 11,000 times the 
distance sun-earth. 



of the positions of the stars: if a star were coming 
straight at us it might appear to be growing 
steadily brighter, but it would never shift its posi- 
tion in the sky. Fortunately, we have in the spec- 
troscope a means of observing this radial veloc- 
ity, as it is called. The principle underlying such 
observations is that light is composed of waves 
similar to those of sound but traveling at a much 
greater speed, 186,000 miles per second, instead 
of 1 mile in 5 seconds. If you have ever atten- 
tively listened to an automobile blowing its horn 
while it was rapidly approaching and passing you, 
you will have noticed that while the car was still 
approaching, the pitch of the horn was compara- 
tively high, and that it dropped noticeably the mo- 
ment the car had passed and began speeding away. 
In the first case the sound waves, traveling toward 
us with a velocity of 1,000 feet per second, were 
helped along by the speed of the automobile, which 
might have been 60 miles per hour, corresponding 
to about 80 feet per second, while in the second 
case the sound waves had to overcome the unfavor- 
able speed of the automobile. In the former case, 
more vibrations reached us per second and the pitch 
was higher; in the latter case, fewer vibrations 
reached us and the pitch dropped. If we put a 
standard tuning fork on the automobile, giving out 
a C, for example, and compared it with another 
standard C tuning fork remaining stationary, we 


could easily determine the speed of the automobile 
from the difference in pitch, in each case separately. 

The same holds for the light of the stars as Dop- 
pler has shown; it too consists of waves, and these 
waves will be approaching us faster or slower, de- 
pending on whether the star itself is approaching us 
or receding from us. When the pitch of a light wave 
is changed, that is to say when the number of vibra- 
tions per second is changed, the colour of the light 
is changed: toward blue when the pitch is increased, 
toward red when the pitch is lowered. Since the 
speed of light is so excessively high, the correspond- 
ing shift in colour is minute and can only be ascer- 
tained through very delicate measurements. If there 
is an atmosphere of hydrogen gas around the star, 
producing a black line in the green part of the 
spectrum, the pitch of this black line, too, will be 
altered if the star has a velocity in the line of 
sight; if the star is approaching, the line will be 
shifted toward the blue-green, if the star is receding 
the line will be displaced toward the yellow-green. 
If now, when the star is observed, an observation 
is made simultaneously of the same black line but 
produced by terrestrial hydrogen gas, a small dif- 
ference in "pitch" of the line may be accurately 

The principle of measurement of the velocities 
in the line of sight is thus quite simple, but the 
actual execution of such measures in practice is an 
entirely different matter. Visual observations be- 



gun in England by Huggins were subject to such 
large uncertainties that they were soon abandoned 
when Vogel in Germany showed that a much 
greater accuracy may be secured by photography, 
but real success in this field was not attained until 
Campbell at the Lick Observatory applied himself 
to the problem. Using a powerful spectroscope 
attached to the 36-inch telescope, Campbell was 
able to measure velocities with an accuracy of about 
half a mile per second, corresponding to a measur- 
ing accuracy on the photographic plate of less than 
one ten-thousandth part of an inch. Through the 
cooperation of many observatories, notably the 
Lick, Mount Wilson, Victoria, and Cape of Good 
Hope Observatories, the radial velocities of more 
than 5,000 stars are now known. 

When the radial velocities of stars in different 
parts of the sky are compared, the fact is brought 
to light that in some regions the velocities are pre- 
ponderantly negative — that is, the stars are ap- 
proaching us — while in the opposite regions the 
stars are, on the whole, receding. Analysis of the 
proper motions reveals the same tendency : the stars 
seem to be approaching the region of the sky near 
Sirius, and avoiding the opposite part. Calculating 
the linear speeds in miles per second from the 
proper motions and the distances, when known, and 
combining them with the radial velocities, we obtain 
the total velocities of the stars. At first these total 
velocities seem to be haphazard, almost chaotic in 



-■ c 



5 u 


i +7 i + i + i ++7+++++ i + i + 

-* — <N — o 




> *o © c 


I CM © "J C 

OCMO©Ot^r*: — r-i-*»/}i/;*-*u">*'-*+ l Oi , --t--f*'>0 
CM t-# th CS CM »>h CM fO — C> *0 <-»-N«^ 

— c 

S a 




> O O "? O Q Q C* C-J O O — Ol^ 

CO — — -co 

— ^* CO NM Ol CO *0 ft 


I I 

e s ca a b cq a 

• O. 

III §s 

CJ « U-- £ (J 



g-.s a 

.S o " o 
"13 °> - 8 

J ? a •£ § 

c ** . - 5 

3 -— : en rt 

>» c cJi ._- 
£ 5 o-o u 


*« il 

»" c ° o 



w S ji 

.s § $•§ a 

8 S £ s - 
.".5 x-£.~ 

I S ||1 

■3 E " E -s 


•= -5 o « 3 
a S.c c « c 
c u . . c— ex 












c e « s 



8- c 



bUS - = s 

if « I g >1 

cj**- c*n >• h 
■fl o^-sS, 


,2 to oi a <-> o?, 
1-9 i 


u C 


* 2 


J >. 


Q 5 




i i i 

OOf^ OO r»»-< cn •»*■ *f 

+11 +1 ++ + II 

** rt> •O^'OO O-t'MQOON OO Q «* O COO WQ 
*-«-* lOfO'O*-* — CM — CO — — * t/} l- CM ro '»'- fO^ 

;Sg38s = 5 

lo oc 00 < 
■o •coo. 

co COO 





- <o o — o © 

; 2 gW5 : 


OJ*0 hhW)C\ «*)^0 0*««^ WO* CM** *0 CN ^^rfiC 

^*tj* *do6odcd c>c>c>©©© ©o' ^j-« »j jj ej f^^2 "g 

^-1 »-H *H — — VH ■-! ««« CM 


•= Jill" 

a * 

CO "2 0! 

C-1G0 CO C 

c c = 


X= Fi=.3 <M u 0~ a. V U XU ,13 

a. 1 ^ «?jc« 1,5 I* si. .3 Jto 




■M V 

■c to 
.5" 2 



c| e 

a o 3 



size as well as in direction, but a closer inspection 
shows that there is "method in their madness." If 
we select a large number of stars at random and 
examine their motions, we again find that, on the 
average, they are always moving approximately in 
the direction of Sirius. The fault, however, is not 
in the stars, but in our sun. It is the motion of the 
sun in space which makes the stars appear to move 
in the opposite direction. Imagine, for example, 
a large number of stars whose motions are really 
and truly chaotic, without any preference for a par- 
ticular direction. Place the sun amidst them and 
assume the sun to be going north. Even though 
the sun might be going so slow that it would be 
overtaken by a number of stars, we should still find 
that, on the average, the surrounding stars would 
appear to go south. The average motion of the 
stars surrounding us is a pure reflection of the sun's 
motion through space, and if the stars appear to be 
moving toward Sirius it means no more than that 
the sun is really moving in the opposite direction. 
Careful analysis of all the velocities known has led 
to the conclusion that the exact direction of the 
sun's motion is toward a point not far from the 
bright star Vega, and that the speed of this solar 
motion is approximately 13 miles per second. 

With the description of the movements of the 
stars we have finished our survey of their more 
important characteristics, and we may now see how 
our sun and the well-known stars measure up in this 




respect. In the first of the accompanying tables 
are given the principal data for the twenty stars 
which are of the first magnitude or brighter. The 
proper name of each star is first given, as well as 
the usual astronomical designation by constellation; 
then follow the apparent brightness in the sky, 
expressed in stellar magnitudes, the distance from 
the sun in light years, and the intrinsic luminosity, 
derived from the last two and expressed in that of 
the sun as unit. The temperature is given, expressed 
in degrees Fahrenheit, while the next two columns 
show the diameter and mass, insofar as these have 
been determined or may be estimated with some 
degree of probability, both again referred to the 
diameter and mass of the sun as unit. The last two 
columns give the proper motion and the radial 
velocity, both expressed in miles per second. A 
negative sign in the last column means that the star 
is approaching us, a positive sign denotes recession. 
At the bottom of the table similar data for the sun 
are appended for purposes of comparison. 

Such a comparison, especially of the data in the 
fifth, seventh, and eighth columns, is not very flat- 
tering for the sun. 

There is but one star in the table that is equal 
to the sun in light and in size; the others far surpass 
it. To draw the conclusion, however, that the sun 
must be a very insignificant star in the universe 
would be erroneous; a comparison with the first- 
magnitude stars is not really fair to the sun. These 



stars are not at all representative of the stars in 
space; they appear bright to us, not because they 
happen to be near, but principally because they are 
of singularly great intrinsic luminosity: they are the 
aristocrats of space. A much fairer sample of what 
to expect from the average star in space we may 
obtain by taking, not the twenty brightest, but the 
twenty nearest stars. Here we have to add, now 
known, as it is more than probable that a consider- 
able number of stars in the vicinity of the sun have 
as yet remained undiscovered, simply because the 
stars are so faint that their presence has not been 
detected. What we know and what we may con- 
jecture about these twenty nearest stars is gathered 
in the second table, arranged in the same way as 
that for the brightest stars, except that the diam- 
eters are now given in miles, instead of in millions 
of miles. 

The two sets of data present a striking contrast. 
To begin with, they have but three stars in common, 
Sirius, Alpha Centauri, and Procyon, which three 
are counted among the lesser lights in the first table 
but appear as the rulers in the second. The sun, 
so hopelessly outclassed by the first magnitude stars, 
ranks high among its neighbours in space. This 
becomes even more evident when we look at the 
names of these twenty nearest stars : only seven are 
bright enough to be seen with the naked eye and 
to have a constellation name; the remainder have 
to be designated by the number they bear in certain 



star catalogues, or by the name of their discoverer. 
Several of the stars in this second list are "double," 
some are even "triple"; for all of these, what is 
known about the individual components has been 
entered separately. Since the fainter components 
of the double stars among the stars in our first list 
would obviously not properly belong there as they 
would not be of the first magnitude, all the stars in 
the first table have been treated as single. We shall 
have occasion to return to the subject of double 
stars later. 

Two items may be emphasized especially; the 
first of these is the fact that two stars in the second 
list, the faint companion to Sirius and "Van 
Maanen's Star," seem to be of much smaller size 
than the rest. These two stars, with three others 
not in the present list, comprise the only five white 
dwarfs now known. They all are very curious 
objects, their mere existence seeming to be in fla- 
grant contradiction to the laws of physics; yet, 
when examined more carefully, it is seen, as Edding- 
ton has shown, that they not only conform to our 
present ideas of matter but even constitute one of 
the best proofs for the truth of these theories. 

The companion to Sirius, the best known of all, 
gives 400 times less light than the sun, and has been 
shown, from observations made with the spectro- 
scope, to be much hotter on the surface than the 
sun. Thus it gives more light per square inch than 
the sun, and from its low total luminosity we calcu- 





late that it must be a very small body, not any 
larger in size than the planet Uranus. On the other 
hand, we have ascertained that it weighs almost as 
much as the sun. From the combination of its size 
and its weight we derive its density and find the 
amazingly high figure of 27,000 times that of water. 
This remarkable star is made of such exceedingly 
dense material that one cubic inch of it weighs about 
half a ton; it is 1,500 times as heavy as gold! 
Strangely enough, however, when we examine its 
light in the spectroscope we find that in the outside 
layers of the star hydrogen abounds, the lightest 
of all gases, ten thousand times lighter than water. 
The explanation? Quite simple, if we only stop to 
consider that all matter, as the latest physical theo- 
ries maintain, is really full of holes. Hydrogen, 
the simplest form of matter, is made up of a very 
small, and very heavy, central particle, the proton, 
around which revolves another, much lighter parti- 
cle, the electron, the combination forming a minia- 
ture planetary system. Other chemical elements, 
though more complicated, are essentially built up 
of combinations of protons and revolving electrons. 
Under ordinary circumstances these multitudes of 
electrons whirl around the central core so fast that 
they set up a barrier through which nothing can 
penetrate, with the result that the size of a particle 
of matter is determined by the size of the largest 
orbit of the electrons. Under normal conditions the 
size of the smallest particle of hydrogen, of an atom 

of hydrogen, is about four one-billionths of an inch; 
consequently, we could never compress hydrogen 
gas more tightly than by lining up four billion atoms 
per inch. Compared with the size of such an atom, 
an electron is again totally insignificant; it is at 
least 10,000 times smaller, and the whole structure 
resembles, as Sir Oliver Lodge so felicitously 
expressed it, "a fly in a cathedral." The best we 
can do under ordinary circumstances is to compress 
the atoms until the cathedrals touch; beyond that 
point we cannot go. Deep in the interior of a star, 
however, things are different, and conditions are 
not "normal." Here we meet temperatures of 
several million degrees, temperatures which have 
the effect of exciting the electrons into violent agita- 
tion, so violent indeed, that instantaneously all the 
electrons are rudely separated from their protons. 
The two no longer belong together and each pursues 
its own destiny, entirely independent of the other: 
in other words, the walls of the "cathedral" col- 
lapse, and all that remain are the "flies." We can 
begin compressing again, and, instead of finding our 
limit when the cathedral walls touch, we can now 
go on until the flies touch; obviously, we can thus 
attain far greater densities than ever before. Espe- 
cially, since the walls of the cathedrals were not 
real walls; they were merely barriers set up by the 
fast whirling motion of the electrons. In other 
words, the cathedrals were built of the thinnest 
tissue paper, the flies made of compressed platinum. 



Matter under normal, terrestrial conditions is com- 
parable to a city, built of tissue-paper cathedrals 
touching each other; under the severe strain of the 
high temperature inside a star the city has collapsed 
and only the small flies buzzing inside are left. 
This is, in all probability, what has happened to 
the faint companion star of Sirius. In its interior 
the temperature is so high that the electrons and 
protons have severed their relationship, and the 
result is a density of 27,000 times that of water. 

According to relativity, a star of this extreme 
density must affect the rays of light emanating from 
it, since the strong gravitational attraction at its 
surface makes the light waves indolent; they vibrate 
slower, and appear shifted toward the red in the 
star's spectrum. Observations made by Adams at 
Mount Wilson have indeed borne out this predic- 
tion, thus proving not only the correctness of the 
calculated high density but also of the relativistic 

The other item of interest in the second table 
concerns the first two entries, Proxima and Alpha 
Centauri. The latter, the third brightest star in 
the sky, has long been regarded as the sun's nearest 
neighbour in space. In 1911, however, Innes, at 
Johannesburg, discovered a faint star in the vicinity 
of Alpha Centauri which showed a large proper 
motion. Accurate measures proved this motion to 
be practically identical with that of Alpha itself, 



leaving but little doubt that the two stars are con- 
nected and travel through space together. When 
the distance of the new object was measured it was 
found to be slightly closer to us than Alpha, and 
thus this faint star, a companion of Alpha Centauri, 
becomes the sun's nearest neighbour; hence its name, 
Proxima. The way in which the motions of the 
stars may be discovered is shown in Plate VIII, 
where three photographs of the region of Proxima 
are compared. The left-hand one, taken in 1901, 
shows Proxima to the left of another star of equal 
brightness; in the centre picture, taken in 1910, 
Proxima forms a close double with its apparent 
neighbour, although this neighboring star may be 
100 times farther away. On the right-hand photo- 
graph, taken in 1925, Proxima has passed its neigh- 
bour by a considerable distance, while the relative 
configuration of all the other stars in the photograph 
has remained unchanged. 

The large angular motion of Alpha Centauri will 
gradually carry it into a different part of the sky. 
In about 4,000 years it will pass very close to its 
rival, Beta Centauri, thus forming the most spec- 
tacular double star in the heavens, while 14,000 
years hence it will stand at the head of Southern 
Cross. In 28,000 years from now it will come 
closest to the sun, and will then shine as a star of 
the magnitude minus one. Counting back 2,000 
years, we find that at the beginning of our era Alpha 
Centauri belonged to what is now the constellation 



Circinus; had the constellations then been mapped, 
the star might have been known as Alpha Circini. 
The case of Alpha Centauri is only one out of many 
and illustrates the effect of the proper motions of 
the stars upon the configuration in the sky. In the 
course of time all relative positions in the sky will 
be altered; the constellations will be "dislocated." 
Fifty thousand years ago there was no "Big 
Dipper"; fifty thousand years hence, the Big Dipper 
will again belong to the past. 


FIG wfc E lnmn HE CHANGING SKY r - the big dipper, as it 
Il.prf.'JSnn'ff ARS AG0 ' AS IT ,S N0W ' AND AS IT W,LL 


The companionship of Alpha Centauri and 
Proxima is not unique in the annals of the sky; we 
know of many cases where a number of stars appear 
to be travelling together in space. The most cele- 
brated of them all is undoubtedly that of the group 
of Big Dipper stars, or the Ursa Major cluster, as 
it is technically called. Proctor had already called 
attention to the fact that the five central stars of 
the seven bright stars that form the constellation 
of the Big Dipper have motions in the sky that are 
almost parallel. When later observations showed 



that the velocities in the line of sight also were 
nearly identical, and when, in addition, Hertz- 
sprung noticed that several other bright stars dis- 
tributed over the whole sky had motions in the same 
direction and of the same amount as those of the 



Dipper stars, it became evident that we were dealing 
with a physically associated group of stars, a moving 
cluster. Among the stars belonging to this family 
are Sirius, Beta Auriga?, Alpha Corona?, and several 
others. When a model is made of this cluster of 
stars, it is found that they form a flat, almost disk- 
shaped group, at present so situated in space that 
the sun is near the plane of the disk. The sun, how- 



ever, does not belong to this family, and, as seen 
from the sun, the whole disk appears to be moving 
with a speed of 1 1 miles per second, in a direction 
not far from the bright star Alpha Indi. The stars 
constituting this cluster are doubtless millions of 
millions of years old; the fact then that the diameter 
of the disk is at present no more than ISO light 
years shows that the motions of the individual stars 
must be very accurately parallel. Without any 
question we may assume that these stars had a 
common origin; equally without doubt we can admit 
that they are driven by the same hidden force to 
the same unknown destiny. But to say where and 
how they originated, or what will be their ultimate 
fate, is beyond our ken. From the point of view 
of the universe, they and the sun are but chance 
acquaintances. Compared with our own ephemeral 
existence, they may seem to be permanent fixtures 
of the sky, while in reality they linger but a short 
while. One million years from now the disk-like 
formation will have receded enough that we may 
see it all at once instead of having to find the con- 
stituent stars all over the sky. In one billion years 
the stars will have become lost in the deep abyss 
of space : ships that pass in the night. 

When the motion of the sun is calculated from 
the systematic behavior of the stars surrounding 
us, the true velocities of those stars may be 
derived by making an allowance for the solar 
motion. The resulting velocities of the stars again 



look chaotic at first sight. But again, when they 
are subjected to statistical analysis, regularities are 
soon discovered; the "traffic laws" of the universe 
are then revealed. Kapteyn, in Holland, was the 
first one to notice that there are two favored direc- 
tions of motion, that there appears to be a main 
highway in the direction of which many more stars 
are moving than at right angles. In a way, and 
very crudely, the situation may be compared to the 
traffic on Fifth Avenue in New York. There is a 
constant stream of uptown and downtown traffic, 
with relatively little motion crosstown. Actually, 
however, the situation among the stars is much 
more complicated. To begin with, the thorough- 
fare does not exist as such. It is only the directions 
of motion that show a strong preference for paral- 
lelism to a certain axis; all other directions of 
motion are still possible, though not as frequent. 
The direction of this celestial highway seems to 
point not far from the bright star Procyon. 

If the regulations on directions of traffic in the 
universe are lenient, the restrictions concerning 
speed limits are much more severe. The sun, we 
have seen, flies through space at the rate of 12 miles 
per second, a speed slower than the average; 20 
miles per second seems to be a normal velocity, 
30 is distinctly fast, and velocities more than 40 
miles per second are rather infrequent. Only a few 
stars still persist beyond 40 miles, some as fast as 
100 miles per second, and we know of only one 


moving with a speed of more than 600 miles per 
second. The faster a star goes, the more it begins 
to avoid certain directions in space; we might almost 
say that certain directions are forbidden to high 
velocities, that the speeders among the stars travel 
on "one-way streets." 

With stars existing that are 400 times larger in 
diameter than our sun, and others which have speeds 
up to several hundred miles per second, it appears 
reasonable to ask the question if celestial collisions 
ever occur, even taking into account the immensity 
of space at the disposal of the stars. The average 
star, as we now think of it, is probably a little 
smaller than the sun, a little lighter, and moves a 
little faster. If we make a model of the universe 
to the scale of one to thirty thousand million, that 
is to say, if we represent the average star by a 
billiard ball and place this billiard ball on top of 
the Woolworth tower, we must then put the next 
nearest star, also represented by a billiard ball, in 
San Francisco. To complete the picture, we must 
endow these average stars with their average 
motion, 20 miles per second; but, on this small scale, 
20 miles per second corresponds to only 2 miles per 
century. The traffic laws for speeds of 20 miles 
per second are not very strict in the universe; stars 
with such speeds may move in practically every 
direction they please. If then, we let a number of 
billiard balls as far apart as New York and San 
Francisco strike out in almost any direction they 



choose, but with a speed reduced to 2 miles a cen- 
tury, we have very nearly a true model of the 
universe. With the situation being such, it does 
not require much calculation to estimate the fre- 
quency with which collisions happen in the universe; 
it is plain that they must be exceedingly rare. In 
the whole universe a collision probably does not 
occur more often than once in many billions of 

Chapter XI 


"Age cannot wither her, nor custom stale 
Her infinite variety." 

— Anthony and Cleopatra. 

It was known even to the Arabs that Mizar, the 
middle star of the handle of the Big Dipper, is 
accompanied by a fainter star, slightly toward the 
north and visible to a trained eye, although the 
casual observer might well overlook it. Great was 
the surprise, however, of the Italian astronomer 
Riccioli when, in 1650, he turned his telescope on 
these stars and found that the brighter one again 
separated into two. After Riccioli's discovery, a 
few more double stars were casually noted during 
the Seventeenth and Eighteenth Centuries, but sys- 
tematic observation of them did not begin until 
Sir William Herschel entered the field. During a 
few years of observation he catalogued as many as 
700 doubles, while Wilhelm Struve, observing at 
Dorpat, and Pulkova, in Russia, subsequently pub- 
lished a list of more than 3,000 such stars. Since 




then, astronomers have paid a great deal of atten- 
tion to these interesting objects, and at present more 
than 20,000 double stars are known. 

With the great telescopes of our times stars may 
be found to be double when the separation between 
the two components is no more than one or two 
tenths of a second of arc — about the same as the 
separation of automobile headlights if these could 
be seen from a distance of 800 miles. Especially 
Burnham at Yerkes, and Aitken at Lick have found 
many such close pairs. Where the telescope fails, 
the interferometer, the instrument used for measur- 
ing stellar diameters, can still be made use of, since 
with it angular separations as small as one hun- 
dredth of a second of arc may be detected under 
favorable circumstances. So small a separation 
would correspond to our automobile headlights 
removed to a distance of 12,000 miles from us, or 
to two beacons, sixty feet apart, placed on the moon. 
Even an instrument as delicate as the interferometer 
has its limitations, however, and double stars closer 
than this limit of one hundredth of a second of arc 
would remain undiscovered, were it not for the 
spectroscope. This instrument, though unable to 
separate the two components, can yet indicate 
duplicity. When a seemingly single star is in reality 
a close double, its two components will, under the 
influence of their mutual attraction, revolve around 
each other; and thus, if we consider their centre of 
gravity as fixed, the two stars will each, in the course 



of their revolution around this centre of gravity, 
alternately approach the earth, or recede from it. 
Consequently, the radial velocities will vary periodi- 
cally, and the black lines in the spectrum of the star 
will periodically shift their positions. From a care- 
ful analysis of these changes in the positions of the 
black lines it may then be ascertained how long it 
takes the stars to revolve about each other, and the 
whole orbit of such a spectroscopic binary may be 
reconstructed in detail. Though the two stars may 
be so close together that we could not distinguish 
them separately even with our best telescopes, we 
can yet prove that there, at the immense distances 
in the stellar universe, two suns are whirling around 
each other, obeying the law of universal attraction. 
Not only the components of a spectroscopic 
binary revolve around each other, but such motion 
applies equally to a double star visually observed. 
William Herschel was perhaps the first to notice 
that in the course of time small changes occur in 
the relative positions of the components of a double 
star. In some cases it is quite evident that one star 
is revolving around the other, and when mathemati- 
cal analysis is applied to the observations it is found 
that the path described by the one star is always 
an ellipse which has the other star in one of the foci. 
Thus the attractive force that operates between the 
components of a double star and compels them to 
revolve is entirely similar to that which makes the 
planets describe ellipses around the sun. In fact, 



observations of the motions of double stars consti- 
tute a beautiful proof for the generality of New- 
ton's law of gravitation, since it can be proved that 
Newton's law is the only one that can make one 
star describe an ellipse with the other star in the 

Having come to the conclusion that Newton's 
law of gravitation is universal, we can in turn use 
it to determine the weight of the stars, from the 
behaviour of double stars. If we take the case of 
Alpha Centauri, for example, we find that we are 
dealing with two stars revolving around each other 
in 79 years, in an ellipse with a greatest diameter 
of 4,000 million miles. This is quite comparable 
to the case of the sun and Uranus, the latter revolv- 
ing around the sun in 84 years, in an ellipse with a 
diameter of 3,600 million miles. We know that 
the attraction between two stars diminishes rapidly 
as the distance between them increases; smaller dis- 
tances will thus result in quicker rotation and shorter 
periods of revolution. We know likewise that the 
force varies directly as the weight. Taking all this 
into account, it is easy to see how, from such a 
comparison as we have made between Alpha Cen- 
tauri and the system of the sun and Uranus, it is 
possible to calculate the weight of Alpha Centauri. 
Similarly, the weight of all binary stars, for which 
the distance between the components and the time 
of revolution is known may be computed. The rela- 
tive motion of double stars provides us with the 



only means of determining the masses of the stars 
and, thus, with a means of "weighing" the universe. 
From the data accumulated up to the present, it 
has been brought to light that the stars are, on the 
whole, very similar in mass. We have not yet 
found a star weighing less than one tenth as much 
as the sun; likewise, stars more than twenty times 
as heavy as the sun are scarce, though a sporadic 
few have been found with masses ranging up to 
almost one hundred. Comparing this with the tre- 
mendous range in luminosity among the stars, from 
less than one ten-thousandth part of the sun's light 
to more than ten-thousand times as much, both 
extremes being reasonably common, we see that 
indeed the stars have much less disparity in mass 
than in light, that they are much more alike in 
substance than in appearance. 

The mass of a double-star system can only be 
found if the system can be properly compared with 
that of the sun and its planets, that is when the 
linear distance between the components of the 
double star is known. The angular separation is 
always measurable, and the linear distance between 
the stars thus becomes dependent on the measure- 
ment of the distance of the double star from us. 
Without this distance the mass cannot be computed. 
From the fact, however, that there is such a small 
disparity in mass among the stars, Hertzsprung and 
Russell have reversed the problem, and, by suppos- 
ing the mass of a double star to be equal to the mean 



value for all stars, calculated the distance from this 
mass. Distances derived in this way are very 
reliable and often more accurate than direct deter- 

Not all double stars revolve around each other; 
some seem to move in a perfectly straight line with 
respect to each other. This does not mean, of 
course, that Newton's law does not hold in some 
cases, but simply that the two stars seen as a double 
star from the earth are not connected at all. They 
are merely two individuals of the manifold popula- 
tion of the cosmos that happen to be seen in nearly 
the same direction; actually, one may be very near 
and the other far behind it. Such double stars are 
called optical doubles, in contrast to the real, phys- 
ical doubles. Vega, the brightest star in Lyra, is an 
example of such an optical double. The proof that 
the stars are not connected rests, in this instance, 
upon the fact that the faint companion does not 
share in Vega's large proper motion in the sky. 

Of all the double stars in the sky, Mizar, or Zeta 
Ursa: Majoris, as the astronomers call it, has made 
most history. Its name is written on three mile- 
stones along the astronomical road of progress. It 
was the first double star to be discovered with a 
visual telescope; it was the first double star to be 
recorded photographically, by Bond at Harvard, 
in 1857, and it was the first spectroscopic binary to 
be detected, by Pickering in 1889, also at Harvard. 
Later, observations by Frost at Yerkes have shown 


that both components of the bright star are spec- 
troscopic binaries, and that in addition Alcor, the 
fainter companion just visible to the naked eye, is 
also spectroscopically double. Castor, the second 
brightest one of the Twins, was found to be a double 
star as early as 1719. As yet the precise period of 
revolution has not been determined but is estimated 
to be between 300 and 350 years. Both components 
of this double are again spectroscopic binaries, with 
periods of 3 and 9 days respectively, while at a 
considerable distance from these four stars is situ- 
ated another faint, red, dwarf star, undoubtedly 
belonging to the system, and one which is itself a 
double star. 

The brightest star in the sky, Sirius, is unique in 
that it is the first star for which the existence of a 
faint companion was computed before it had been 
observed. We have learned before that Sirius has 
a large motion in the sky; so large that, as seen by 
an observer standing on top of the Woolworth 
Building, Sirius, when watched for a whole year, 
would be found to move by the thickness of a 
half dollar placed on Broadway. In this motion, 
however, Sirius was not regular; instead of mov- 
ing in a straight line, it swayed to and fro per- 
ceptibly. This could not be tolerated; no other 
star in the universe behaved this way, and the 
cause of it must be found. Again, as in the case 
of Uranus and Neptune, astronomers set to work 
and found that these idiosyncracies of Sirius could 



only be explained if it had a faint, but rather heavy, 
companion which moved around it in 50 years and 
persisted in pulling Sirius out of its place. When a 
telescope was turned on Sirius, the companion, 
10,000 times fainter than the primary, was found 
in its predicted place. Subsequent developments 
have helped to make this faint object one of the 
prize exhibits in astronomy, since it has a density 
of 27,000 times that of water. 

Concerning the numbers of double stars, it has 
been found that one in nine among the stars visible 
to the unaided eye is a visual double, while at least 
one in eighteen of the stars brighter than the ninth 
magnitude can be seen to be double. If spectro- 
scopic binaries are included these numbers arc still 
more increased: duplicity must be rather common 
in the universe. Naturally enough, these double 
stars revolve around each other in all kinds of 
ellipses, with the planes of their orbit oriented at 
random, in all different ways. Some of those ellipses 
we see from above, so to speak; others we view 
obliquely, with the result that they appear fore- 
shortened; still others we see "on edge." Suppose 
now that such a double star, or rather a spectro- 
scopic binary, because there the two stars are closer 
together, consists of a small, very luminous star 
revolving around another, much larger, and com- 
paratively dark star. If we really see the orbit 
from the side, the luminous star will, once during 
every complete revolution around the larger star, 



pass behind it and suffer eclipse. Consequently, the 
total light of the star (which is all we can observe, 
since the components of such a star are too close 
together to permit us to distinguish them sepa- 
rately) will be appreciably diminished. When, on 
the other hand, the smaller and more luminous star 
passes in front of the larger star, it will in turn hide 
and eclipse a small fraction of the light of this 
larger and fainter component. Since a large portion 
of the total light is concentrated in the small star, 
the diminution of light in the latter case is only 

Such an eclipsing system we find in Algol, the 
second brightest star in Perseus. Here, a star 100 
times brighter than the sun periodically passes 
behind and is eclipsed by a slightly larger star only 
about ten times as bright as the sun. When the 
two are side by side, and the full light of both of 
them reaches us, Algol shines as a star of the second 
magnitude; the partial disappearance of the brighter 
star will make the magnitude drop to between the 
third and fourth, or to about one third of the normal 
amount of light, while the secondary eclipse, when 
the bright star is in front, diminishes the combined 
light by only 4 per cent. In the accompanying figure 
we have tried to give a representation of what 
occurs in the case of the star RW Tauri, another 
"eclipsing binary," which is invisible to the naked 
eye but shows a greater change in brightness than 
Algol. It is clear from the drawing that during 






the greater part of the time the two stars are visible 
side by side and the combined light must remain 
constant. When actual measures of this total light 
are made with a photometer and these measures 
are plotted in a graph, against the time at which 
they were made, a series of points are obtained. 
When these points are joined by a smooth curve, the 
light curve, our predictions are indeed borne out; 
for the light curve will be flat over the largest part 
of the period, indicating that no change occurs in 
the total light, and will further show two minima, 
a deep one for the primary eclipse, and a shallow 
one for the secondary. 

At present about 200 stars of this eclipsing type 
are known; among them all different degrees of 
light variation are found, owing to the necessarily 
different positions under which their orbits in space 
are viewed from the earth, and owing to the great 
diversity in the brightness and size of the com- 
ponents. In the case of Beta Lyrs the two stars 
are almost, in that of W Ursa: Majoris they proba- 
bly are, in actual contact; in the latter case the stars 
are even elliptical in shape. 

The light variations of Algol and of many other 
eclipsing binaries are so conspicuous that they can 
easily be seen with the naked eye. No wonder, 
therefore, that Algol was discovered as a variable 
star as early as 1667, by Montanari, in Italy. The 
real nature of the variability, however, was not 
brought to light until 1782, when Goodricke and 



Pigott took up the problem. In recent years mathe- 
matical analysis has been applied to the light curves 
of these variables, and especially through the work 
of Russell and Shapley at Princeton a great deal 
has become known about the properties of the 
individual stars. 

Variable stars of the eclipsing type are, in a sense, 
not really variable; they only appear so, dependent 
upon the orientation of their orbit in space. Neither 
of the two components really varies in light output, 
and this feature distinguishes them from other real 
variable stars. The explanation of an eclipse does 
not satisfy the conditions of change of light in the 
other known types of variable stars ; here we have 
to assume that the luminosity of the star is subject 
to inherent changes, although we can, at present, 
only conjecture as to the exact nature of these 

Among these other real variable stars, Mira, one 
of the brightest, was also the first one to be discov- 
ered. In 1596 Fabricius, in Holland, noted a star 
of the third magnitude in the constellation Cetus, 
which faded rapidly and disappeared in a few 
weeks. It was again seen in 1603, and for the third 
time in 1638. So great was the astonishment at the 
recurrence of such a phenomenon that the star was 
given the name Mira, the wonderful, a name which 
it has kept until to-day. For a long time Mira and 
the subsequently discovered Algol remained the only 



ones known; at the close of the Eighteenth Century 
hardly a dozen were known. One hundred years 
later, photography had entered the field of astron- 
omy, and more than 500 variable stars were known. 
At present the total number has grown to more than 
6,000, more than 4,000 of which have been found 
at Harvard. Again mass production has reduced 
discovery to a commonplace: finding new variables 
is no longer an event, and anyone with a large col- 
lection of photographic plates at his disposal could 
easily guarantee to find one new variable a day. 

When first seen, Mira Ceti was of the third mag- 
nitude; later it became invisible to the naked eye. 
Modern observations have shown it to descend to 
the ninth magnitude, while it has been known to 
become brighter than the second. Its total range 
in brightness, therefore, is more than seven magni- 
tudes, corresponding to a light ratio of one to one 
thousand. In addition, these light changes have 
been found to be periodic; not quite as regularly so 
as the variation of eclipsing stars, but more in the 
manner of the periodic recurrence of sun spots. 
Roughly speaking, Mira is at its brightest every 
330 days, then spends 240 days in descending to 
minimum brightness, and consumes only 90 days in 
its ascent to the next maximum. In every individual 
cycle of its variation the star does not follow this 
schedule very closely; it may be as many as twenty 
days ahead or behind. At present about 2,000 stars 
behaving in a manner similar to that of Mira are 



known, and periods have already been derived for 
more than 800. When these periods are examined 
it is shown immediately that the vast majority of 
them lie between 100 and 400 days, with a concen- 
tration around 250 to 330 days. Since these periods 
are much longer than those of other types of varia- 
bles, the Mira Ceti stars are usually called long- 
period variables. Observations with the spectro- 
scope reveal them all as red stars, mostly of spectral 
class M while determination of their distances and 
the actual measurement of the diameter of Mira 
proves that, at least during their time of maximum 
light, they are among the largest of stars, as large 
or larger in diameter than Betelgeuse or Antares. 
They are true giants, and at their maximum lumi- 
nosity surpass the sun several hundred times in 
light, while in minimum their light must be com- 
parable to that of the sun. Their speeds are large, 
more than 40 miles per second on the average, but 
some go as fast as 250 miles per second. 

The next largest group of variable stars is formed 
by those stars whose light fluctuations are less than 
one magnitude on the average, and whose periods 
are usually short, from three to seven days. The 
brightest representative of this class is the Pole 
Star; the earliest one for which the light variations 
became accurately known is a fourth-magnitude 
star in Cepheus, Delta Cephei, after which the 
entire group are named Cepheids. Unlike the Mira 
variables, the Cepheids are rigorously periodic in 



their light variations; they repeat themselves with 
such extreme accuracy that the light curve is subject 
to very precise determination. Although the cause 
of their variation has not yet been definitely settled, 
it seems highly probable that the Cepheids are pul- 
sating stars, according to a theory originally pro- 
posed by Shapley, and subsequently worked out 
mathematically by Eddington. 

A third group of variable stars, known as the 
cluster variables, because of their occurrence in 
large numbers in stellar clusters, have periods 
closely concentrated around 13 hours as a mean. 
They show light fluctuations very similar to those 
of the Cepheids, but are less rigorously periodic, 
subject rather to sudden and erratic jumps. These 
cluster stars may in some way be physically related 
to the Cepheids, though many points on which they 
differ may be cited: The regular Cepheids all have 
small velocities, usually not more than 8 or 10 miles 
per second, while the cluster variables are among 
the fastest stars known, having speeds ranging up 
to 800 miles per second. The Cepheids are all 
crowded near the Milky Way; the cluster variables 
may be found all over the sky. 

In 1912, when studying the light variations of 
some Cepheids among the 2,000 variable stars in 
the Magellanic Clouds, Miss Leavitt at Harvard 
noted the remarkable fact that the periods of these 
Cepheids stood in a distinct relation to their appar- 
ent brightness. Since the variable stars all presum- 



ably belonged to the Cloud, and were thus at the 
same distance from the earth, this meant that the 
periods were related to the intrinsic luminosities. 
This curious relationship lay unknown for some 
time, and it was not until Hcrtzsprung grasped its 
full significance that it took its rightful place as one 
of the most powerful means of penetrating the 
depths of space. Shortly afterward it was extended 
by Shapley to the cluster-type variables and applied 
with success in measuring the universe. 

If indeed the period of variation is correlated to 
the intrinsic brightness of a Cepheid, then it is suffi- 
cient to determine the luminosity of but one Cepheid, 
and the intrinsic luminosities — and with them the 
distances of all Cepheids whose periods of variation 
have been observed — may be calculated. Hertz- 
sprung had previously shown that the Cepheids may 
be counted among the brightest stars in the universe, 
being about 1,000 times more luminous than the 
sun. This is another incidental advantage of the 
method, since it means that Cepheids can be seen 
at far greater distances than ordinary stars, and 
thus they provide a very efficient and deeply pene- 
trating astronomical yardstick. Perhaps, as the best 
description of the method of using Cepheids, Jeans's 
graphic illustration may here be cited : "The method 
is simply that of a mariner who estimates his dis- 
tance from land by identifying a lighthouse, looking 
up its candle power in a book of reference, and 



noticing its apparent brightness at the spot where 
he happens to be." 

The long-period variables, the Cepheids, and 
cluster-type stars, and the eclipsing binaries form 
by far the largest portion of all known variable 
stars; the remaining portion consists of stars vary- 
ing in an irregular fashion. Among these is another 
group of stars that are not really variable, but only 
appear so because they are imbedded in an envelope 
of obscuring material. If this envelope, which is 
hardly ever uniform in light-obstructing power, 
moves, it will change the thickness of the occulting 
curtain in front of the star, and thus give the star 
the appearance of variability. In the Orion nebula 
a number of variables are known, which, in all 
probability, change in brightness for this reason. 
Another star of this type, discovered by Knox Shaw 
at Helwan, is situated in the blackest part of a 
dark cloud in the constellation Corona Australis. 
It was later found again at Harvard, and varies 
irregularly between the thirteenth and eighteenth 

The vast majority of variable stars behave so 
regularly that their actions may be predicted 
with some degree of certainty, and even the most 
erratic ones will always stay within certain bounds. 
Compared with such commonplace and monotonous 
behaviour, we can, however, contrast something that 
is really new and unexpected: the appearance of a 
"new" star. Where before there was but an invisi- 



ble and unknown denizen of the cosmos, there now 
appears a star so bright that it may outshine all 
others in the sky, a nova, as astronomers call it. 
Here we have a real display of the forces of Nature, 
an explosion which, in a few days, or even hours, 
may transform an object of comparative insig- 
nificance into one of absolute though ephemeral 
supremacy in the universe. 

What really happens, and how? Thus far we 
can only surmise; we still grope in the dark about 
the true origin of a nova. How is the stage set 
for this, Nature's greatest drama? We do not 
know. Probably it happens in the interior of the 
star, and in the silence of empty space. While the 
celestial dynamite is being stored up, slowly but 
surely, no outward sign betrays what is going on. 
When the hour strikes, the trigger is pulled mys- 
teriously and the explosion is immediate, but silent, 
for the voids of space do not admit of any sound. 
The most spectacular explosion in creation, it is no 
more than a blinding flash, without the roll of 
thunder 1 In one great leap, the hot gases come 
rushing out, the star begins to expand at the rate 
of 1,000 miles a second. Within a day it may have 
increased its light a millionfold, its volume a billion- 
fold. If it was a dwarf before, not unlike our sun, 
it now has become a giant among giants; for the 
moment of its glory it has no rival in the realm 
of transparent space. Meteoric as was its rise, its 
downfall must soon follow; a few hours, a few days 


at most, its brief reign may last. But what are a 
few days? A mere cosmical instant. The star's 
brilliance begins to decline, and though the star is 
unwilling to sink into oblivion again, and may make 
a few more futile attempts to regain the summit of 
its former splendor, it is doomed. It may be a 
matter of months, sometimes of years; the end is 
inevitable, the grave always claims its prey. 

As yet we do not know what causes the outburst 
of a nova. It cannot be due to a collision between 
two stars, since such collisions will not happen, on 
the average, more than once in a million years, and 
we observe novas at the rate of at least one a year. 
Neither can it be due to the passage of a star 
through a nebula, as the heat generated by friction 
accumulates too slowly to produce such an instan- 
taneous and immense increase in luminosity. And 
thus, cornered, with no explanation to offer, we are 
driven into the star's interior, and must attribute 
the explosion to the action of mysterious forces in 
the innermost layers of the star. Although we thus 
still grope in the dark when it comes to explaining 
the cause of the outburst, our knowledge of other 
characteristics of the phenomenon of new stars has 
advanced rapidly in recent years, especially through 
the researches of Wright and Lundmark. 

The first new star to be scientifically and sys- 
tematically observed was the one that appeared in 
1572 in the constellation Cassiopeia. It was seen 
by Tycho Brahe, the famous Danish astronomer, 



and still bears his name. At the height of its lustre 
this nova surpassed even Jupiter in splendor and 
was easily visible in broad daylight. When we now 
search for it in the spot where Tycho observed it, 
we cannot find a star brighter than the twelfth 
magnitude, and it is doubtful whether the remains 
of the nova are at all visible. At the very least 
this represents a drop in brilliance of more than 
seventeen magnitudes, a light ratio of six million 
to one! Another well-known nova was that of 
Kepler, appearing in Ophiuchus in 1604, which 
reached a brightness almost equal to that of Jupiter, 
and remained visible to the unaided eye for eighteen 
months. In 1918 a star in Aquila, which thereto- 
fore had been of the tenth magnitude, suddenly 
increased to magnitude minus one, 300,000 timer 
brighter than the sun, actually. The outer layers 
of this star were expanding at the rate of 1,000 
miles per second. 

A curious feature developed around the nova of 
1901, a diffuse, extended patch of light which 
appeared to be expanding with a speed of more 
than 100,000 miles per second. Such a speed was 
unheard of; it simply could not exist. Kapteyn 
immediately suggested the explanation: the star 
was surrounded by a cloud of dust, and as the light 
from the nova travelled outward, it illuminated these 
dust clouds in order of their distance from the 
centre. For the first time in history we saw the 
actual speed of light I 



The most recent of brilliant nova; was Nova 
Pictoris, born in May, 1925. Previously, it had 
been a star slightly fainter than the sun, intrinsically, 
and also a little smaller in size, somewhat cooler, 
and appearing as a star of the twelfth magnitude. 
On May 25th the nova had suddenly increased its 
diameter to 40,000,000 miles, 50 times larger than 
the sun. On June 9th it was of the first magnitude, 
and almost 80,000,000 miles in diameter. At 
present (May, 1928) the star has already decreased 
to the seventh magnitude and is no longer visible 
to the unaided eye. 

Perhaps the most significant thing about new 
stars is their great frequency of occurrence. Even 
with our present incomplete surveys of the heavens, 
we catch a nova rather more often than once a year, 
and since we can only find the brighter ones, the 
total number actually appearing must be greater. 
We know that our earth alone has existed for at 
least five hundred million years and the sun for 
many billions of years. It does not seem reasonable 
to suppose that we are living at a time particularly 
favoring nova, and it seems that we have every 
right to assume that outbursts of nova; have been 
as frequent during the past fifty billion years as 
they are now. During that time then, more than 
fifty billion nova should have appeared, a number 
at least equal to what we now estimate is the total 
number of stars in the universe. This leaves us 
with two alternatives: either all stars have been 

Photograph Harvard Observatory 


Abate: Three photographs illustrating the proper motion of Provima Centauri. (See page 

Btlnu:- Nova Pictoris. (See page 208) 

Since the scale of the last photo B raph il more than twice that of the other m it nives the appearance a. if Nova 
Pictoris were ImitlHest on tab last picture. In reality it represents a imichtncis ten times fainter than that of the 
mitUHc photograph. 



nova;, or will in time become nova;, or "once a nova, 
always a nova." Either the nova stage is something 
which every star has to go through at least once, or 
it is a habit, a disease, which, once contracted, 
returns time and time again. Although at present 
there seems to be little or no evidence either way, 
astronomical opinion in general appears to favor 
the second notion. 

In Plate VIII are gathered three photographs of 
this nova, the left-hand one showing it as it ap- 
peared before its explosive rise in brightness as a 
star of the twelfth magnitude (February 29, 1902). 
In the middle Nova Pictoris is shown as it appeared 
on September 19, 1925, four months after its 
greatest splendour, when it had already faded to less 
than one tenth of its maximum brilliance. On the 
right is given a much more enlarged photograph, 
taken on December 19, 1927, showing the nebulous 
envelope which subsequently developed around the 
star, then more than one trillion miles in diameter. 



Chapter XII 

"If shape it might be called, that shape had none." 

— Milton. 

While cruising through the vast emptiness of 
space the stars are not all subjected to solitary con- 
finement. Instead, they show a distinct tendency to 
gregariousness. Some, such as Alpha Centauri, are 
double, or have a faint, distant companion star; 
others, after the manner of Sirius, are double, but 
at the same time belong to a larger family of stars, 
a cluster. This family of Sirius, the Ursa Major 
cluster, has so far as now known, some twenty mem- 
bers, mostly bright stars. They appear bright to 
us because they are near, and because they are so 
near that the sun lies almost among them, they 
also appear scattered over a large part of the sky. 
When in the course of time the cluster has moved 
away from us sufficiently that we may view it in 
retrospect, we shall see a number of stars, not quite 
so bright as the cluster stars appear to us now, but 
much closer together; their clustering will then be 
very evident. Another group of stars, which already 


is in this position, is the Hyades, situated at a dis- 
tance from us of ISO light years, and appearing as 
a rather closely packed group of stars of the fourth 
and fifth magnitudes, surrounding Aldebaran in 
Taurus. The Pleiades, also in Taurus, are still 
farther away, some 400 light years distant, and as 
a result they appear more closely packed even than 
the Hyades. After the Pleiades comes the double 
cluster in Perseus, a twin conglomeration of stars of 
the eighth magnitude and beyond; finally we come 
to the large number of small groups of stars that 
are known under the name of open clusters. Most 
of these clusters are rather faint, and can only be 
seen through large telescopes; they contain from a 
few hundred to at most a few thousand stars, usually 
with the brightest stars near the centre. Most of 
the stars in such clusters are intrinsically much more 
luminous than the sun. About two hundred of such 
objects have been discovered up to the present. 

Another type of cluster, built on a vastly more 
magnificent scale than these open clusters, is formed 
by the globular clusters. Here thousands and tens 
of thousands of stars unite into one great conglom- 
eration, a whole firmament in one. These clusters 
do not, however, reveal their beauty to the naked 
eye, nor even when seen through a small telescope. 
Only one, Omega Centauri is visible to the unaided 
eye as a rather hazy star of the fourth magnitude, 
while a small telescope still shows the majority of 
clusters as hazy objects, without much detail, and 



with but a few individual stars in their midst. Seen 
through a large telescope, such a cluster may finally 
appear to be resolved in all its individual stars, but 
the photographic plate proves that this resolution 
is, indeed, only apparent. Long-exposure photo- 
graphs taken with the largest telescopes show the 
number of stars to be so great in the centre that 
the images overlap and create the impression of 
one tremendous globe of luminosity breaking into 
stars at the edge. 

The very multitude of their stars gives these 
globular clusters an impregnable appearance, as if 
to defy all analysis. No wonder that our knowl- 
edge of these objects advanced but slowly, almost 
imperceptibly at first. No sooner were the great 
telescopes of Mount Wilson turned on them, how- 
ever, than the barrier was broken down, and through 
the epoch-making work of Shapley many of the 
secrets of globular clusters were quickly unveiled. 
Bailey, at Harvard, had found large numbers of 
variable stars in these clusters, studied their be- 
haviour, and derived periods for the light-variation 
of many stars. Combining Bailey's discovery with 
Hertzsprung's work on the intrinsic brightness of 
such variable stars, Shapley was able to determine 
the luminosities of several stars in globular clusters, 
and thus, also, the distances of the clusters them- 
selves. The globular clusters proved to be among 
the remotest objects yet found, the nearest and 
brightest of them all, Omega Centauri, being 21,000 

UairarJ Obstrvatory 


Omen Centauri, the neatest of all globulai dusters. 21,000 light years distant. It 
contains Thousands and tens of thousands of stars of the thirteenth magnitude and 
fainter. The combined light of all these stars makes the cluster appear to the naked 
eye as of the fourth magnitude, as bright as the faintest of the seven stars in the liifi 



light years distant. When removed to such an enor- 
mous distance, the sun would appear as a star of the 
nineteenth magnitude, 100,000 times fainter than 
the faintest star visible to the unaided eye. And 
this distance concerns the nearest of all globular 
clusters. On the average they are four to five times 
farther distant, while the remotest of them all, so 
far measured, is no less than 200,000 light years 
away. No wonder that the introduction of such 
distances immediately placed the globular clusters 
on the outskirts of creation and revolutionized our 
concept of the extension of the universe. 

The majority of the variable stars in globular 
clusters complete their cycle of light variation in 
thirteen hours, and in many cases the light changes 
appear to take place with almost clock-like precision 
for thousands of oscillations. Indeed, as Professor 
Bailey so aptly put it, "these cluster variables might 
be called Nature's celestial time-keepers"; they 
would serve, if needed, as admirable watches. If a 
photograph of a cluster containing a number of such 
variable stars had been buried with King Tut-Ankh- 
amen, it might be made to reveal the exact epoch 
when the photograph was made. 

The brightest stars in a globular cluster are prob- 
ably of the same type as the "red giants" in the 
neighborhood of the sun, similar to Antares and 
surpassing the sun in luminosity by more than a 
thousand to one. The individual stars in such a 
cluster must be very closely packed; a space which 



contains one star in the neighborhood of the sun 
would have to accommodate one hundred in the 
globular clusters. Even so, the stars are far enough 
apart to avoid collisions, nearest neighbors being 
probably no closer than one thousand times the dis- 
tance between the sun and Neptune. 

Owing to their great distance, it has been impos- 
sible thus far to determine the proper motions of 
the globular clusters; at the same time, their faint- 
ness has made radial velocity observations extremely 
difficult. Slipher, at the Lowell Observatory, has 
determined radial velocities for about twenty of 
the more than seventy globular clusters now known. 
These velocities are all large and lead us to believe 
that the speed of the average globular cluster may 
be as high as 150 miles per second. 

Surpassing in splendour even the globular clusters 
are the nebula, those strange, seemingly shapeless 
clouds of matter, impressive through their evident 
immensity. Sometimes they appear luminous, creat- 
ing the impression that they are the cosmic labora- 
tories where stars are being generated; sometimes 
they are black, contrasting strangely with the 
myriads of luminous stars surrounding them, and 
through their omnipresence tragically suggestive of 
a celestial morgue. For centuries their existence, 
their constitution, and their role in the universe have 
been an enigma, and although the complete solution 
will undoubtedly take many centuries more, we feel 



confident that now we understand many of its major 

William Herschel, the father of modern astron- 
omy, noted early the fact that there are nebula: and 
nebula?. Not all objects that appear hazy and 
diffuse in a telescope are real nebula;. Some are 
simply conglomerations of stars, star clusters, ulti- 
mately resolvable in a large telescope, while others 
appear obviously gaseous, large masses of brilliance 
without definite form, "nebulous fluid," as Herschel 
used to call them. The spectroscope has shown in- 
disputably that this view is correct, that there are 
such clouds of luminous gas, while photography 
later added the fact that there must also be great 
clouds of non-luminous material. Especially Barnard 
at Yerkes and Wolf at Heidelberg have greatly 
advanced our knowledge of these objects. 

The dark nebulae are unquestionably great clouds 
of "dust," and are visible only because they obscure 
the light of the stars behind them, thus causing the 
appearance of "black holes" or "dark lanes." Note- 
worthy examples of such occulting clouds are the 
"coal sack" in the southern Milky Way, shown on 
Plate VII to the left of the Southern Cross and 
the "Horse-head" Nebula in Orion, portrayed in 
Plate X. The whole region of the star Rho 
Ophiuchi seems to be teeming with many such dark 
lanes, all apparently emanating from the same 
centre, where, in the immediate vicinity of the bright 
star, the black nebula: become luminous themselves, 



as is plainly shown in Plate XI. The same holds 
here as well as in the Horse-head Nebula, Plate X; 
the obscuring cloud changes into a bright wisp of 
nebulosity when it approaches a bright star. No 
doubt the dust cloud is then shining by reflected light, 
or is, perhaps, incited by the powerful radiations 
from the very hot star. 

As an illustration of the former supposition, 
Hertzsprung calculated the brightness such a nebula 
should have if it were shining by reflected light, 
and found that in the case of the Pleiades the 
nebulosity in which the brighter stars are imbedded 
is only 5 per cent, as brilliant per square inch as a 
sheet of white paper would be. Slipher then deter- 
mined the spectrum and found it to be, as expected, 
of the type associated with reflected light. More 
recently, Hubble has thoroughly analyzed all the 
known "diffuse" nebuls. He has shown that, with 
a few exceptions, the stars are physically associated 
with the nebula;, and are their source of luminosity. 
Thus the "North America" Nebula (See Plate XII) 
may depend for its luminosity on Deneb, the very 
bright nearby star. The great nebula in Orion 
appears bright to us because of the abundance of 
bright stars imbedded in a vast mass of "black." 
material. To the naked eye this nebula appears as 
a hazy spot of light just under the three stars that 
form the "belt" of Orion ; to the eye of the photo- 
graphic camera, whose sensitivity increases with the 

Pholot'Ofh Mount ll'ilion Obirrralory 



The Horse-head Nebula ii Orion, near the star Zeta Ononis. It consists of a great mass of 
obscuring material illuminated at the edge by a bright star behind it. 1 he multitude oi 
faint stars in the upper half of the picture arc situated at Krearer distances than the cloud; 
while the few stars seen in the lower part arc all nearer than the cloud, and appear projected 
None of the stars shown is visible to the naked eye. 

against it. 



time of exposure, the nebula appears to extend over 
almost the entire constellation. 

There can be no reasonable doubt, then, that the 
dark nebula?, as well as many of those of the Orion 
t ype_often called diffuse and gaseous nebula?, are 
all cosmic dust clouds, shining only when illuminated 
by a neighboring star. When we call them clouds 
of dust, however, we must not confuse them with 
such dust clouds as we know on earth. From the 
probable value of the mass of the Orion nebula, 
it has been calculated that the mean density must 
be less than one millionth of a billionth of that of 
air under ordinary circumstances; that is to say, at 
most equal to the density resulting from expanding 
one cubic inch of atmospheric air over a volume of 
a cubic mile. On earth we should call this an ex- 
tremely good vacuum. In the stretches of space this 
density is so great that a cloud of gas of such tenuity 
acts as a very effective screen, cutting off completely 
the light of the stars behind it. 

According to the newest researches of Eddington, 
these nebulae are only condensations in the "gas" 
that fills all space nearly uniformly. The density 
of this inter-stellar gas must be considerably lower 
still, probably not more one ten-thousandth part of 
that of such a nebula, or about equal to one ounce 
of matter distributed evenly over a space of twenty- 
five thousand million cubic miles. This does not 
seem excessive; nevertheless, if it is added up 
throughout the millions of light years at the dis- 



posal of the astronomer, it amounts to a great deal. 
One cubic light year contains — or rather, one cubic 
light year of "empty" space "weighs" — about one 
septillion tons: a number with 24 ciphers. A thou- 
sand cubic light years weigh as much as the sun on 
this basis. This interstellar gas would be ionized 
by the ultra-violet light from all the stars. Thus, 
if it contained calcium, we would expect to find, in 
stellar spectra, absorption lines of calcium, which 
did not share the motion of the stars. Thus, an 
explanation is found for the strange phenomenon of 
the "stationary calcium lines," which have been the 
subject of much investigation by Hartmann at Got- 
tingen, Plaskett at Victoria, and Struve at Yerkes. 
In fact, Struve has used the increasing strength of 
these lines as a measure of increasing absorption, 
and thus determined the distances of a number of 

Although nebulas usually present too vague an 
appearance to allow of precise measurement of posi- 
tion, it has been found possible in two special cases 
to measure motion. One nebula in Taurus, called 
the "Crab" Nebula, is in the process of expanding 
rapidly, while another, the "Network" Nebula in 
Cygnus, consisting of two feathered wisps of nebu- 
losity in the shape of an oval, is likewise increasing 
its dimensions. In spite of the uncertainty involved, 
it is tempting to reverse this process in our imagina- 
tion and sec if it is not possible that, some time, long 
ago, those nebulae were caused by the bursting of a 

Pkototraph Ytrkei Ot^erraiory 


Tin- region !UtTOunding the star Rho Ophiuchi. showing extended irregular masses of 
obscuring clouds, an interstellar "fog" that hides from our view all the stars behind it. but 
which itself becomes luminous when excited by the radiation of a hot star near it. 



new star. In the case of the crab nebula we actually 
have a record, though admittedly uncertain, of a 
new star that blazed forth at about the required 
time, which makes the above speculation seductively 

To the measurement of radial velocities the ill- 
defined appearance of the nebula presents no ob- 
stacles; the lines in the spectrum are just as sharp 
as in a star, and many velocities in the line of sight 
have been measured at the Lick and Lowell Ob- 
servatories. It has even been possible to measure 
the internal turbulent motion in the great nebula in 


There is still a third type of nebula;, the planetary 
nebula?, so named because they are much sharper 
in outline than the others and present a disk some- 
what similar in appearance to that of the planet 
Uranus. Practically all contain a faint star in or 
near the centre and often show shells or rings of 
light of different intensities. These central stars 
are all very hot, some fifty thousand degrees Fahren- 
heit at the surface, but there appears to be some evi- 
dence that they are rather small in size. The plane- 
tary nebula themselves are probably not very dif- 
ferent in dimension, about 10,000 times larger in 
diameter than the orbit of the earth. Their masses 
are probably comparable to those of the heaviest 
stars, some 20 to 100 times that of the sun; but, 
even so, their density must be exceedingly small, in 



fact not much larger than that which we found for 
the Orion Nebula. 

The most puzzling feature of these planetary 
nebula; and of some diffuse nebula; concerns the char- 
acter of their light. Analysis with the spectroscope 
showed a number of lines which could not be ascribed 
to chemical elements of terrestrial origin. Espe- 
cially two strong and typical lines in the green, often 
identified with a hypothetical element "nebulium," 
provided a mystery difficult to solve. Very recently, 
however, even this secret was wrested from the 
nebula;, representing the culmination of a long and 
arduous labor in which the foremost astronomers 
of the country took part. Under the persistent attack 
of Wright at Lick, the problem slowly neared solu- 
tion, until only the exact identification of the green 
lines remained as the last stronghold to be conquered. 
And now, even this has capitulated; after a master- 
ful summary of the problem by Russell, Bowen at- 
tacked it and solved it with the aid of modern 
physics. The green nebular light so long ascribed 
to an unknown substance, the mysterious nebulium, 
now takes its place among all other celestial radi- 
ances as the easily explained, and necessary conse- 
quence of the modern theories of light and of the 
constitution of the atom. By the time Bowen began 
his work it had become increasingly probable that 
the green light was due to ordinary terrestrial gases 
emitted under conditions peculiar enough to excite 
a gas, which is millions of times more rarefied than 

Phototraph Yerkts Objtnaiory 


The North America Nebula in Cygnus, a large, diffuse nehula, near Deneb, the brightest 
star of the constellation. The light of the nehula may be a reflection of that of Deneb; its 
distance from us is probably around 60C-700 light years. 



our atmosphere and of itself intensely cold, into 
giving light. Physics knows of only two ways in 
which this can be done: by bombardment with high- 
speed electrical particles, or ultra-violet light, and in 
either case we are immediately referred back to hot 
stars. We must look for the solution of the mystery 
in the radiation sent out by the central star and in 
the peculiar conditions of the surrounding gas. To 
make a long story short, the solution lies in the 
extreme rarefication of the nebula. Even in the best 
vacuum we can produce in the laboratory there are 
enough molecules of gas left to cause each of them 
to collide several thousand times per second. If, 
now, the strong radiation caused by a very hot 
source of light has brought a gas particle into such 
a state that it could send out this green light a col- 
lision with another particle would destroy that pe- 
culiar condition. In the planetary nebula, on the 
other hand, matter is so tenuous that a particle may 
travel for weeks and over a distance equal to an 
astronomical unit, instead of a thousandth of an 
inch, before it collides with another. The peculiar 
state due to the radiation from the hot star is 
retained, the green light, so characteristic of the 
nebula, results, and the mystery is explained. The 
principal green lines, it is found, are emitted by 
oxygen, other lines in the spectrum of the nebula; 
formerly ascribed to nebulium also, now appear 
caused by nitrogen, and we may thus say with 


Russell that "nebulium has literally vanished into 
thin air." 

Bowen's coup is the crowning achievement of a 
long series of investigations which have resulted in 
the shattering of one more romance of astronomy: 
the nebula have ceased to exist as such. They have 
been reduced to cosmic dust, have lost their inde- 
pendence, so to speak, and are shown up as shining 
merely under stimulation of stars. When the star is 
very hot, the nebula takes in the light and trans- 
forms it into radiance of its own liking, principally 
green light; when the star is comparatively cold, 
however, the nebula can do no more than reflect 
the light. When there is no bright star near, the 
nebula remains "without form and void," and acts 
as a curtain, obscuring everything behind it: "Dark- 
ness is upon the face of the deep." Once more the 
stars have risen to the position of being the only 
independent denizens of the cosmos, the only ones 
that shine with a glory of their own. 

Chapter XIII 

"Thousands of human generations, all as noisy 
as our own, have been swallowed up of Time, and 
there remains no wreck of them any more; and 
Arcturus, and Orion, and Sirius are still shining 
in their courses clear and young as when the 
Shepherds first noted them." 

— Carlyle. 

How is a star born? How long does it live? 
How and why does it die? These and other ques- 
tions have been perplexing astronomers for the past 
few decades, have been the driving power of astro- 
physics. It is only during the past few decades, 
since the advent of photography and of the spectro- 
scope, that we have known enough to ask such ques- 
tions intelligently. It is only since then that we have 
really begun to consider the stars otherwise than as 
mere specks of light, that we have become inter- 
ested in their physical constitution. 

In a way it should be simple for the astronomer 
to write the biography of a star — the stars really 
write their own life histories. That is to say, the 
data are already there; all the astronomer needs 



to do is to heed the writing in the sky, traced in 
words of flame and fire, and decipher the code of 
these celestial autobiographies. Parts of them are 
written in infinitesimal characters, the language of 
the atoms and electrons. These are the easiest to 
interpret, for the recent atomic theories have sup- 
plied us with an excellent Rosetta stone, and we 
have been able to translate no small part of them. 
Especially the epoch-making work of Eddington 
has enabled us to penetrate deeply into the interior 
of the stars, into the innermost of the cosmic cruci- 
bles. Thus far, however, the Ten Commandments 
which govern the life-course of a star are not yet 
fully understood, though we have reason to believe 
that the tablet on which for trillions of years past 
the stars themselves have been chiseling this dec- 
alogue has been brought to light, largely through 
the efforts of Lockyer, Hertzsprung, Russell, Jeans, 
and Eddington. The alphabet employed in these 
stellar cuneiforms has not yet been established, and 
only one of the commandments seems to have been 
fully translated: "Thou shalt not be heavy." Ed- 
dington has shown that stars much heavier than the 
sun are extremely unstable and will explode at the 
slightest provocation. 

That astronomy with all the means at its disposal 
has not yet progressed beyond this point must not 
seem surprising. Indeed, the task of interpreting 
correctly the evidence of stellar evolution is much 
more formidable than it appears. Here on earth, 



where the average lifetime of a single individual 
falls considerably short of a century, and where the 
average interval between generations is about thirty 
years, we arc in possession of records of the history 
of man for a period of thousands of years. But 
for the history of the stellar universe we have rec- 
ords of not more than forty years, while the average 
life of a star must be at least a quadrillion years, a 
number with IS ciphers. On the earth, even the 
suggestion that evolution takes place seems to meet 
with at least a few objections; with the stars, 
although astronomers may hold different opinions 
as to the how and why of stellar evolution, they all 
admit its necessity. 

When we say that our records are only a few 
decades old, we mean those records that can be 
used as evidence in the search for traces of stellar 
evolution. Although the stars have been observed, 
wondered at, and thought about from time immemo- 
rial, man's knowledge never reached far beyond the 
mere fact that they existed. Their physical con- 
stitution remained behind a closed and sealed door 
until, under the combined attack of photography 
and spectroscopy, the seals were broken. Once 
the door was unlocked a glimpse was caught of a 
treasure so vast that its full importance is not yet 

We may say that we have, at most, forty years 
of spectroscopic research behind us, and what is 
forty years compared with the time scale of the 



astronomer, to whom a million years is "as but a 
day"? I believe it was Simon Newcomb who illus- 
trated the difficulty of the problem by the following 
comparison : imagine a person endowed with the 
keenest powers of reasoning and in possession of 
the finest and most accurate instruments we have as 
yet produced, a person, furthermore, who has 
neither seen nor heard of a pendulum. Imagine 
this person placed in a completely darkened room, 
where numerous pendulums are swinging noiselessly. 
Illuminate this laboratory for one single instant by 
a flash of lightning. Then demand from the ob- 
server that he explain the law that governs the 
motion of a pendulum. And yet, his task is child's 
play compared with the stupendous problem of the 
astronomer who has to derive a theory of stellar 
evolution, for the duration of the electric flash 
against that of the swing of a pendulum is, com- 
paratively speaking, millions of times longer than 
forty years in the life of a star. 

With such an immense task before us, it is well 
to assemble what we really know, before we venture 
forth on the perilous undertaking of theorizing. In 
gathering our evidence we must also be careful to 
select with the least possible prejudice, paying more 
attention to facts than to appearances. For things 
are not what they seem. The stars, as they appear 
to us in the sky, are not at all representative of the 
stars as a whole, and any theory derived solely from 
the stars we can see without the aid of a telescope 



might be very materially wrong when applied to 
the stars as we find them in space. We have seen 
before how much the stars may differ in luminosity, 
how the giants of yellow color may outshine the 
dwarfs by more than a thousand to one, and how 
the giants of all types exceed the sun more than a 
hundred times in brilliance. Of all the stars visible 
to the naked eye, some 6,000 in number, at least 
4,000 may be classed as giants; of the remainder, 
probably less than 200 are inferior to the sun in 
luminosity. If, on the other hand, we take a sample 
of 6,000 thoroughly representative stars in space, 
we should find that more than 75 per cent, are 
fainter, intrinsically, than the sun, while the giants 
contribute no more than a miserable 2 or 3 per cent. 
The stars visible to the naked eye, therefore, prove 
to be exceptions rather than typical specimens, which 
circumstance adds another difficulty to our already 
involved problem: the naked eye stars lend them- 
selves especially to accurate and detailed observa- 
tions by their greater brightness, and now we find 
that the larger part of our best observations has 
been made on exceptional cases. Persistent attacks 
and statistical evaluation of the material, avoiding 
as much as possible the selection introduced by the 
observations, have ultimately resulted in the estab- 
lishing of at least a few facts. 

The principal one among these is Hertzsprung's 
discovery that the vast majority of the stars arrange 
themselves very neatly in a sequence, running from 


the very luminous, and very hot, blue and white 
stars down to the insignificant and obscure red 
dwarfs. At the upper end of this series stand the 
redoubtable Orion stars, 35,000 degrees Fahren- 
heit in temperature, bluish-white in color, and sur- 
passing the sun between 100 and 10,000 times in 
light-giving power. Halfway down the line is our 
sun, not more than 10,000 degrees in temperature, 
while at the lower end we find millions of faint 
red dwarfs, 5,000 degrees or less in temperature, 
and between 100 and 10,000 times fainter than the 
sun. Hertzsprung immediately suggested that this 
sequence was the main branch of stellar evolution; 
the yellow and red giant stars which did not conform 
to it were, for the time being, treated as exceptions. 
It was Russell who, a few years later, first pro- 
posed a very attractive theory of evolution, taking 
into account the above-mentioned facts which he 
had come to independently of Hertzsprung. In 
Russell's theory the red giant stars of the type of 
Bctelgeusc and Antares head the list and represent 
the very first stages of stellar evolution. As such 
a star grows older, it contracts and becomes hotter, 
thus going through various metamorphoses, finally 
arriving at the hottest and most luminous stage, 
that of a blue-white giant. It has now reached the 
pinnacle of its fame, and shines without rival among 
the stars; from now on it is doomed to grow cooler 
and fade, although still continuing to contract. It 
passes through the stage now occupied by the sun 



and ultimately reaches that of the red dwarfs, which 
we thus see as the sad remains of the celestial fire, 
nothing but a lingering memory of the giant stars 
of the Orion type, once so proud and powerful that 
their light messages penetrated to the utmost depths 
of space, but now scarcely able to flicker to their 
immediate surroundings their last Morituri te 

For a while Russell's theory, though not entirely 
unchallenged, remained the only workable one. As 
time went on, however, new data accumulated, and 
data seemingly incompatible with the theory. Fore- 
most among these were the "white dwarfs," stars 
of enormous densities, whose existence had been 
indicated for some time but was definitely proved 
by Eddington. In his researches on the internal 
constitution of the stars Eddington first came to 
the remarkable conclusion that, paradoxically 
enough, as stars grow heavier, things on their sur- 
face begin to "weigh" less and less. A new force, 
the pressure of radiation directed outward and re- 
sulting from the enormous flow of energy emanating 
from the interior of a star, arrives on the scene. 
Since "light," according to relativity, has "weight," 
a stream of light rays exerts a pressure upon things 
material, and in very heavy and very luminous stars 
this pressure of light may even exceed the force 
of gravitation, and consequently such very massive 
stars may explode without warning. Conclusive 
proof of the theory lies in the fact that we do not 



know of one single star more than 100 times heavier 
than the sun. 

During the course of his subsequent investigations 
Eddington has indicated that the interior of a star 
will practically always behave as a "perfect gas." 
The physicist understands by a perfect gas a gas in 
a state of such extreme tenuity that the simplest gas 
laws are accurately obeyed. Under ordinary cir- 
cumstances there are always slight deviations from 
these simple laws, deviations which grow larger 
and larger as the density of the gas increases. For 
a density ten times greater than that of atmospheric 
air such deviations are already quite marked, and 
we had always, rather naturally, supposed that in 
stars of the density of the sun, more than one thou- 
sand times denser than our air, these deviations 
would be enormous. On the contrary, the deviations 
are far less, almost negligible, and for no other rea- 
son than that the temperature in the interior of a 
star is so excessively high, several millions of de- 
grees, that it has changed the gases to such an extent 
that they will again behave as a perfect gas. The 
effect of temperature on a gas is to set it in violent 
motion, not only the particles of the gas itself, but 
even the electrons inside the individual particles. 
When the temperature becomes sufficiently high, the 
electrons are shot off, the atoms are "stripped," 
and the size of the smallest particles in the gas are 
reduced so considerably that, although they may 
be condensed to weighing one thousand times more 



per cubic inch than before, there is, relatively speak- 
ing, so much more room for the smaller particles to 
move that they behave more nearly like a perfect 
gas than ever before. Continuing his argument in 
the same direction, Eddington finally explained the 
existence of white dwarfs, stars with a density 
25,000, 100,000, perhaps as much as 1,000,000 
times greater than water. 

As an alternative to Eddington's perfect gases, 
Jeans has recently advanced a singularly attractive 
theory based upon the idea that the stars really 
behave as liquids. Finding that, under the condi- 
tions possible in the inside of a star, it must behave 
either as a perfect gas or as a liquid, Jeans then 
indicates that when a star is in the "perfect gas" 
stage it is very unstable, and will soon try to become 
a stabler configuration, a liquid. Again, Jeans 
obtains his arguments from the modern theories 
of the atom: it is owing to the wishes of the indi- 
vidual atoms and electrons that a star behaves as it 
does. Apart from this majority rule in its interior, 
however, a star has not many ways to go. The few 
possibilities are a direct outcome of a race between 
the diameter of the star and the size of the atoms, 
the stripping of the atoms really being due to a 
change in temperature, which accompanies any 
change in diameter. In Jeans's happy figure of 
expression, a "case of the tortoise and the hare." 
"If the stellar diameter is the tortoise, the atomic 
diameter is the hare; its progress is by spurts and 





rests alternately. The spurts of the hare do not save 
it from ultimate defeat, but they result in its being 
alternately in front of and behind the tortoise." 
As in Eddington's theory, the ultimate stage is that 
of the white dwarfs, where the atoms are com- 
pletely stripped of all their electrons. The hare 
has jumped for the last time, and when next the 
tortoise catches up with it they become both caught 
in a trap from which no escape is possible. On 
previous occasions, whenever the atoms became so 
closely packed that they had no elbow room, it was 
always possible to strip the atom some more, to 
reduce its size again, and make room. In the white- 
dwarf stage the atom is stripped to the core, it is 
the final stage, and the star will remain in it until 
the end of its days. 

At the same time that they developed their 
theories on the interior of a star, both Jeans and 
Eddington approached the question of what keeps 
the stars going, how they replenish the incessant 
stream of light and energy they are pouring out 
into empty space. This, naturally, is intimately con- 
nected with the problem of stellar evolution, since 
a complete answer as to where the energy comes 
from will provide us with a time scale for evolution. 
Since the whole process goes on in the interior of a 
star, we cannot be certain but only conjecture what 
takes place. According to Jeans and Eddington, 
mailer is being annihilated on a large scale. That 
this is possible is a new idea, introduced by rela- 

tivity, which claims that matter and energy are, 
after all, merely different manifestations of one and 
the same thing, and may well be transformed one 
into the other. Also according to relativity, a small 
amount of matter will go a long way. Take, for 
example, the changing of hydrogen into helium. 
If we start out with one pound of hydrogen, we do 
not get our full pound of helium. We find that 
we are short one eighth of an ounce, and it is this 
eighth of an ounce of the life blood of matter that 
has been transformed into a prodigious amount of 
energy, light, heat, or radiation. It would suffice 
to heat one million tons of water from freezing to 
boiling. For this reason, hydrogen has often been 
called matter unborn; it is not matter in the ordi- 
nary sense of the word, since it carries this excess 
weight of eight tenths of one per cent, which is not 
matter. And, according to the laws of the uni- 
verse, this small percentage of explosive contents 
is cosmically intoxicating. Once, however, the catas- 
trophe has taken place, that is, from helium onward, 
matter behaves perfectly normally: after it has shot 
the rapids of this Niagara of matter, it flows on 
quietly in regular channels and creates no further 

According to Eddington and Jeans, the actual 
process that goes on in the inside of a star need 
not be that of converting hydrogen into helium; 
some similar action would do equally well. It does 
appear reasonably certain, however, that the source 


of stellar energy must be sought in the annihilation 
of matter on a large scale, since this is the only 
process known at present that will liberate an 
amount of energy sufficient for the maintenance of 
stellar radiation for trillions and quadrillions of 

From all the facts at present at our disposal, we 
may confidently surmise that the age of the stars 
must at least be of the order of magnitude of a 
trillion years. Our sun, not yet an old star, although 
undoubtedly its tumultuous youth is past, is probably 
not less than several trillion years old ; a star of the 
red-dwarf type may well be a quadrillion years old. 

No matter which way the course of stellar evolu- 
tion runs, no matter which way the star chooses to 
go, red dwarf of low temperature, or white dwarf 
of high temperature and exceedingly great density, 
the end is inevitable. The star's career must come 
to an end, and the curtain is lowered at the close 
of the last act of the celestial drama. The ghost- 
like body of the extinguished star may continue to 
roam through space, it may wander on for trillions 
of years to come as a celestial vagrant — as a star 
it is no more, its tale is told, its glory dissolved. 
In due time it may be gathered in and thrown on the 
scrap heap of the universe, but who knows that some 
day it may not be thrown into some cosmic crucible, 
together with a score of other dead bodies, be 
rejuvenated, and arise as a Phenix from its ashes? 

Chapter XIV 

"When the astronomer says that the light from 
a star takes one hundred years to reach us, the lie 
is too great to be artistic." 

— Shaw. 

Although following the general scientific tendency 
toward diversity and complexity astronomy never 
loses sight of the unity of creation. In our detailed 
description of the different kinds of objects in the 
universe, their dimensions and their velocities, this 
unity may have become obscured. Yet the stars are 
not distributed at random, nor do they wander aim- 
lessly through the shelterless deserts of astronomical 
space; the variable stars, the clusters, and nebula? 
all play their unobtrusive parts in making up that 
great organization known as the Milky Way system. 
In the heavens the Milky Way appears to us as 
that faint, hazy, but broad band of light encircling 
the whole sky, a band of light so faint, indeed, that 
moonlight causes it to disappear almost completely. 
The presence of a small amount of artificial light 
is enough to subdue it perceptibly and the city 



dweller rarely gets a good view of the Milky Way. 
If, however, we were to go into the solitude of the 
desert, or high into the mountains, preferably in 
southern latitudes, far from the disturbing illumina- 
tion of civilization and above the haze and dust of 
the lower regions of sea level, we should behold an 
entirely different spectacle. A great arch of won- 
drous beauty stretches across the vault of heaven, 
its vague windings sometimes reaching as high as 
the zenith, and at times of such resplendence as to 
cast a shadow. If we could but remove the earth, 
or place ourselves in such a position in space that 
the whole sky would be visible, the complete Milky 
Way would be seen as a great circle making the 
entire circuit of the sky. 

When we examine the Milky Way in more detail, 
we notice that its boundaries are not well defined, 
and that the whole broad and faint band of light 
varies in width, as well as in brilliance. In the 
constellations Auriga, Monoceros, Orion, and Argo 
it is very broad and faint, while it is very bright 
in the region of the Southern Cross, and reaches its 
greatest splendor in Scorpio and Sagittarius. In 
the constellation Cygnus the Milky Way divides in 
two, the main stream continuing down through 
Aquila into Sagittarius, while the secondary branch 
extends southwestward into Ophiuchus and gradu- 
ally dies out. In many places, especially in Sagit- 
tarius, the light seems to condense into star clouds 
and, as a contrast against these, there appear black 


A portion of ihc Milky Way, extending from Aquila to Centaurns, and containing the 
great star clouds in Sagittarius. The centre of our Milky Way system is probably situ- 
ated in the direction represented by the middle of the plate, but a large number of the 
stars belonging to this central condensation of our system seem to be obscured by dark 
clouds. The two bright stars on the right, near the edge of the plate, arc Alpha (left) and 
Beta Ccntauri. In the centre is Lambda Scorpii, while the consrellation Scorpius con- 
tinues from there upward, including the group of bright stars near the rop. The bright star 
near the bottom of the plate is Alpha 1'avonis. (The plate is a composite photograph ob- 
tained by fitting together prinrs made from thirteen different negatives. The lines 
produced at the junction of the separate prints show faintly. The original negatives 
were all taken at the Arequipa station of the Harvard Observatory, with a lens of i}-inch 



spots, seemingly void of stars and doubtless due 
to huge masses of obscuring material. The main 
features of such star clouds are well shown on the 
plate of the Milky Way in Sagittarius and Scorpio; 
the "coal sack," shown in the illustration of the 
Southern Cross, is a good example of the obscuring 

If a telescope is pointed toward any portion of 
the Milky Way, a multitude of faint stars imme- 
diately spring into view; if we sweep the whole 
domain of the Milky Way with even a small tele- 
scope, this army of stars grows into tens and hun- 
dreds of thousands. On long-exposure photographs 
of Bailey, Barnard, and others, hundreds of thou- 
sands of stars are sometimes shown on one plate; 
we possess one plate at Harvard on which no less 
than half a million stars may be counted. From the 
latest researches of Van Rhijn and Seares it is esti- 
mated that the total number of stars in the Milky 
Way that could be photographed with the 100-inch 
reflector at Mount Wilson is probably of the order 
of two or three billion. 

When contemplating this immense structure 
around us, the question arises as to what interpreta- 
tion we shall put on the data. It seems rather 
absurd to imagine that the sun is in the centre of 
an enormous ring of star clouds, and much more 
reasonable to suppose that the whole structure of 
the Milky Way system is shaped like a watch filled 
with stars throughout, and with the sun situated 





not too far from the central plane, although not 
exactly in the centre. William Herschel, the 
pioneer in this branch of astronomy, already held 
this latter view, and although, with the data avail- 
able in his day, it was impossible to arrive at any- 
thing but the crudest approximation, he foresaw 
the time when this problem could be solved in its 
principal aspects. In his words, the ultimate goal 
of astronomy consisted in the determination of the 
distance of every star in the universe; once this was 
done, the vast structure of the Milky Way system 
would reveal itself. We have intimated before how 
impossible such an ideal solution of Herschel is, 
even in our days: our best methods of triangulation 
do not reach beyond 1,000 light years, while the dis- 
tances of the remotest portions of the Milky Way 
may number 30,000 to 50,000 light years. Again 
we have to resort to statistical means, and in doing 
so, we incidentally enlarge Herschel's definition of 
the 'goal of astronomy,' to take in, not only the dis- 
tances of the stars, but also their luminosities, their 
sizes, their masses, and their motions; in short, we 
want a picture of the living and moving universe. 
The difficulties attached to a problem of such scope 
seem well-nigh insuperable. 

To mention but one instance: looking upon the 
photograph of the Southern Cross and its surround- 
ings, we note the fact that the nearest known of all 
stars, Proxima Centauri, is so faint that it is indis- 
cernible among the confused mass of other faint 

objects, and can be found only with difficulty on the 
original negative. On the other hand, the brightest 
star near the centre of the plate, Alpha Crucis, is 
so far away that its distance has not yet been meas- 
ured with any degree of certainty. On the plate it 
appears more than 10,000 times as bright as 
Proxima; in reality, its luminosity must be more 
than 4,000,000 times greater than that of our near- 
est neighbor in space. Apparent brightness, there- 
fore, is but a poor criterion of nearness, and for the 
host of faint stars in the Milky Way, where we 
lack all information concerning color or motion, we 
must resort to wholesale statistical devices in order 
to arrive at even an approximate first solution. 

The best and most comprehensive determination 
that has thus far been made is that of Kapteyn who, 
practically single-handed, inaugurated the modern 
era of statistical research and had the great hap- 
piness of seeing his life-long labors come to fulfill- 
ment in this, his picture of the Milky Way system. 
According to his results, this "Kapteyn Universe," 
as it is commonly called, is an immense conglomera- 
tion of stars, flattened in shape and somewhat re- 
sembling a bun, or a watch. The "dial" of the 
watch is five times larger in diameter than the thick- 
ness, although there really are no sharp boundaries, 
but merely a falling off in star density toward the 
edges. As a first approximation, the sun had to be 
assumed near the centre of this system, and it is 
found that the star density — that is, the number of 


stars per cubic million light years — decreases from 
the centre outward. Near the sun a cube 100 X 
100 X 100 light years, or equal to 1,000,000 cubic 
light years, contains on the average about 1,500 
stars, but at a distance of 10,000 light years in any 
direction in the plane of the Milky Way this num- 
ber has already been reduced to 150, while at a 
distance of 30,000 light years we find no more than 
15 stars per cubic million light years, only 1 per 
cent, of the star density near the sun. Since the 
"watch" of the Milky Way system is rather thin, 
the falling off in numbers of stars in the direction 
of the smallest thickness, that is in the direction of 
the poles of the Milky Way, is much more rapid 
than in the plane. Instead of 10,000 light years, 
we now do not have to go more than 2,000 to find 
the number of stars per cubic million light years 
reduced to 150, or 6,000 light years to find that 
number as low as 15. This state of affairs explains 
why we see the Milky Way in the sky as a luminous 
band, containing far more stars than the regions 
near the poles of this great circle. It is simply that 
in the direction of the Milky Way plane we see 
through five times as much distance, and a great 
many more stars combine their light to make up 
the total glow, than in the direction of the poles. 

Kapteyn also determined the relative numbers of 
stars of different intrinsic brightness as an incidental 
by-product of his researches on the structure of the 
Milky Way system. He found that for every 



1,000,000 stars of the luminosity of the sun, space 
contains 200,000 stars 10 times brighter, 16,000 
stars 100 times brighter, 450 stars 1,000 times 
brighter, and only 5 stars 10,000 times more lu- 
minous than the sun. Stars such as Rigel, Canopus, 
and Deneb are exceedingly scarce, therefore, and 
the only reason that we see a considerable number 
of them even with the naked eye is that, through 
their great luminosity, they are visible at far greater 
distances and over a far greater portion of space 
than their less luminous brothers. For stars in- 
trinsically fainter than the sun the numbers augment 
rapidly, stars 10 times fainter than the sun being 
represented by 1,800,000 in the above series, stars 
100 times fainter by at least 2,000,000. For stars 
still fainter our present knowledge is so sketchy that 
it is difficult to say how large their number is, 
although it seems indicated by the work of Hertz- 
sprung and others that they are far more numerous 
than was previously believed. 

Another outcome of Kapteyn's great investiga- 
tion is concerned with the velocities of the stars. 
If, again speaking in first approximation, they are 
assumed to be at random, in size as well as in direc- 
tion, the whole Milky Way system may well be 
likened to a gas in which the individual molecules 
representing the stars are in constant agitation. We 
have seen, however, that in the vicinity of the sun 
the star density is such that an analogous picture 
may be formed by taking a number of billiard balls 


and placing them 3,000 miles apart on the average. 
In the outer portions of the Milky Way the star 
density is so small that we should have to place 
these billiard balls 10,000 to 20,000 miles apart, 
and if such a galaxy is compared with a gas, it can- 
not be with a gas under ordinary conditions, but 
with a vacuum. In fact, similar conditions would 
not exist in a gas until we had rarefied it to such an 
extent that it corresponded to a cubic inch of air 
spread out over one thousand cubic miles or more. 
Collisions which form the basis of all theories of 
gases must be exceedingly rare occurrences under 
such conditions: precisely what we have found to 
hold for the stars. If, reasoning further on the 
path of this gas theory, we calculate the velocities 
acquired by the stars under the gravitational attrac- 
tion of the whole Milky Way system, or the Galac- 
tic system as it is often called, we find that their 
velocities should all be less than sixty miles per 
second. This leads to the conclusion that all stars 
moving faster than this limit will ultimately escape 
from the Galactic system, and in all probability 
they must have come from elsewhere originally: 
they are interlopers. On the other hand, it appears 
that the motions of these interlopers are, on the 
whole, remarkably parallel to the plane of the Milky 
Way, and seem to become more so as the speed 
increases. The plane of the Galaxy thus seems to 
have some essential connection with these stars, 
which could not be the case if they really came from 



the outside and were not members of the Milky 
Way system. The problem has not yet been settled 
and will probably require a great deal more data 
before a satisfactory solution can be reached. 

Kapteyn's ideas concerning the structure of the 
Galactic system were admittedly only rough and 
approximative: the few years that have elapsed 
since the completion of his work have indicated that 
probably the real structure is different and designed 
on a vastly larger scale. As a matter of fact, Shapley 
has shown that the Kapteyn Universe is little more 
than the local cluster, the group of stars surround- 
ing the sun; this group itself is considerably off 
centre in the much more extended Galactic system. 
This centre he has estimated to lie at a distance of 
some 25,000 light years in the direction of the great 
star clouds in Sagittarius, and coincides with the 
centre found for the system of globular clusters and 
for that of the planetary nebula. In addition to 
this, the region of Sagittarius is very rich in "new" 
stars, and contains enormous numbers of variable 
stars. Evidence is rapidly accumulating tending to 
show that these great star clouds in Sagittarius are 
really so immense in size that they alone may well 
contain several billion stars. The greater Galactic 
system, of which these clouds form the centre, is 
probably as much as 300,000 light years in diameter, 
and still shaped like a watch, although considerably 
flatter than Kapteyn's system. 

The great mass concentrated in this region will 



then act similarly to the sun in our planetary system, 
and will force the other stars to revolve around it. 
Linblad and Oort have investigated this matter and 
found that there is some evidence to show that the 
sun and all the stars in its vicinity are rotating 
around this centre with a speed of 180 miles per 
second, which, if they kept up this speed, would 
make them complete one turn in about 200 million 
years. The local cluster may then be likened to a 
swarm of bees: the sun, belonging to the swarm is 
actuated by a velocity of 1 2 miles per second with 
respect to the average member of the swarm, while 
the swarm as a whole is revolving around the hypo- 
thetical centre of the Galaxy with a speed of 180 
miles per second. 

Vast as the framework is which we have designed 
and upon which we have built our comprehension 
of the Galactic system, it does not represent the sum 
total of our knowledge. Our telescopes have now 
revealed many objects that cannot be placed within 
the confines of that Milky Way system, while our 
minds had already pondered on their existence. 
From the Galaxy, of which our own sun is but an 
insignificant constituent, we pass on again into empty 
space toward a still greater conception of the ma- 
terial universe. 

Chapter XV 

"The distance is nothing, it is only the first step 
that counts." 

— Mme. du Deffand. 

Inasmuch as universe means all, it is perhaps 
difficult to understand its use in the plural. By a 
change of meaning, which is as intelligible as it is 
illogical, universe to the astronomer is now limited 
to meaning all that was once thought to exist, 
instead of what really exists. Astronomical advance 
during the past few decades has been so rapid that 
astronomers find themselves hampered by the inade- 
quacy of their language. New concepts quickly 
supersede older ones, new words have to be coined 
or older meanings changed. The Galactic system, 
once thought to be the whole universe, has been 
found vast beyond the wildest dreams, but of no 
avail; even its size of 300,000 light years is now 
shown as insufficient to include any but the smallest 
fraction of the real cosmos. Our conception of the 
material cosmos to-day is much broader; we see the 
abysmal chasms of space strewn with countless 



"island universes," vast conglomerations of matter 
far beyond the confines of the Milky Way, scattered 
at distances of millions and perhaps billions of light 

Long before telescopic observations warranted 
it, the Swedish philosopher Swedcnborg and the 
English scholar Thomas Wright speculated upon 
the existence of such external stellar systems. In 
Germany Kant pursued this, but it was not until 
William Herschel's researches that the idea of 
island universes began to take on definite shape. 
The era of real observational evidence was inaugu- 
rated in 1845 when Lord Rosse discovered that a 
nebula in the constellation Canes Venatici appeared 
to be of a spiral shape. This most unusual and 
unprecedented discovery was doubted at first, but, 
as telescopes became more powerful, and especially 
as photography was introduced in astronomy, 
Rosse's discovery was established beyond doubt. 
The great nebula in Andromeda, one of the very 
few visible to the unaided eye, proved to be a spiral, 
likewise two nebula: in the Big Dipper and a large 
nebulous structure in Triangulum. With the aid of 
the photographic plate spiral after spiral was dis- 
covered, and now they form a well-established and 
very numerous class of celestial objects. In some 
parts of the sky, notably in the constellations Virgo 
and Coma Berenices, many hundreds of such objects 
may be seen on a single photographic plate; in some 
very localized areas the spiral nebula? may even be 



more numerous than the stars. The total number 
of objects of this type that can be photographed 
with the 100-inch reflector is estimated at several 

For some time after their discovery, astronomers 
speculated about the nature of these nebulous 
objects. Were they gaseous, like the real nebula; 
such as the Orion nebula, or were they great aggre- 
gations of stars, only appearing nebulous because 
of their tremendous distance from us? The spec- 
troscope soon put an end to this controversy when 
it showed that the light of these nebula does not 
resemble that of the real gaseous nebulas. The 
work of Slipher, especially, has established beyond 
doubt the fact that the spiral nebulas are aggrega- 
tions of stars, since their spectra are roughly similar 
to that of the sun : a band of light, crossed by numer- 
ous dark lines indicating the presence of many 
familiar chemical elements, such as iron and calcium. 
Such spectra could only be produced by a large 
number of stars composed of the same materials 
as those known in our universe. When reflecting 
upon this observation we may well marvel at the 
power of spectroscopic methods: long before we 
knew anything about the distance of these spirals, 
long before we could distinguish the individual stars 
in them, the spectroscope had proved to us that the 
behavior of atoms and electrons in these far-distant 
structures was the same as in the stars and on earth. 
We might almost be tempted to say that at these 



infinite distances we can study only the infinitely 

Another outcome of the spectroscopic observa- 
tions is the indication it affords concerning the 
velocities of the spiral nebula?. If we can take 
measured differences between the light emitted by 
these spirals and that produced in our laboratory 
as attributable to their velocity, we find that they 
all have high speeds. The great spiral in Andro- 
meda is probably approaching us with a speed of 
200 miles per second, while the other spirals, with 
only few exceptions, are receding from us, some as 
fast as 1,000 miles per second. 

The next question to be decided was: how far 
away are they? Since their apparent dimensions 
can be observed, this would involve also asking: 
how large are they? Placing them at moderate 
distances would make them over-sized star clusters, 
so to speak, while distances of a million light years 
or more would make them comparable to our 
Galactic system in size. For a while astronomical 
opinion was divided, as no definite evidence in favor 
of either supposition was available. The first 
fairly certain evidence came from the behavior of 
the new stars that had been observed to flash up in 
spirals from time to time. In the Galactic system 
such outbursts usually result in a "new" star which, 
at the time of its greatest brilliance, surpasses the 
sun more than ten thousand times in luminosity. If 
the reasonable assumption may be made that the 



new stars in spiral nebula? are about as bright, 
intrinsically, as those in the Galactic system, we 
merely have to determine the apparent brightness 
of the new stars in spiral nebula? and we can calcu- 
late their distance. In this way, Lundmark and 
Curtis reached the conclusion that even the nearest 
of spirals must be about one million light years 

About the same time, however, the opponents 
of such tremendous distances were greatly fortified 
in their opinions by the measures made by Van 
Maanen at Mt. Wilson, on the internal motions in 
the spirals. Van Maanen's conclusion was that 
these nebula? were in rapid rotation, some of them 
making one complete turn in as short a time as 
50,000 years. For an object so vast as a spiral 
nebula this is indeed incredibly short, altogether 
incompatible with the great distances derived by 
Lundmark. Suppose, e. g., that Landmark's dis- 
tance of 1,000,000 light years for the Andromeda 
nebula were correct, then its linear size must be 
about 50,000 light years. Suppose also that such 
a structure were rotating in as short a time as 
50,000 years; then a particle near its outer bound- 
ary would, in 50,000 years, describe a circle with a 
diameter of 50,000 light years, or rather, the 
particle would complete a path more than 150,000 
light years long in that time. Thus, it would travel 
more than 3 light years per year. In other words : 
;'/ would go faster than light. Here we have 



reached an absurdity, an impossibility, according to 
our present way of thinking. The doctrine of rela- 
tivity postulates that the speed of light is the great- 
est possible, and that no material particle can travel 
with such a speed, let alone surpass it. Whatever 
relativity decrees is the constitution of the universe, 
it allows of no violations of its laws, of no excep- 
tions to it. Either we have to abandon our theory 
of great distances or disbelieve the rapid rotation 
of spirals. Since the large distances were based 
upon indirect evidence of somewhat uncertain value, 
astronomical opinion on the whole seemed to favor 
Van Maanen's views, though Curtis and Lundmark 
dissented. The pendulum of progress again swung 
from island universes to star clusters, and the Galac- 
tic system once more rose supreme. 

The island-universe theory suffered no more than 
a temporary relapse, however, as again the pendu- 
lum swung forward — definitely, this time, as we 
now think. Hubble, at Mount Wilson, discovered a 
number of variable stars in the Andromeda nebula, 
variable stars of the Cepheid type. As we have 
seen before, such variables constitute a beacon 
whose candlepower is accurately known, and thus 
the distance of the Andromeda nebula may be 
found. Again, the distance proved to be nearly 
one million light-years, a complete vindication of 
Lundmark's earlier work. Hubble's evidence is of 
such fundamental importance in the problem of 

Photograph: Mount It'il'on Ob:rtvat0ry 


About: The Sombrero Nebula, an island universe in a very early stage of development. It is 
situated in the constellation Virgo, and probably not less' than 10,000,000 light years 

Below: Left: A close group of spiral nebular in I'egasus. containing three large spirals, all of 
about the same "age," and one, much "younger," elliptical nebula which, in all probabil- 
ity, is much farther away. Right: A twin universe in the constellation Virgo, just at the 
point of developing into two spiral nebula;. 




stellar distances, and is at present believed to be 
of such great weight, that his results are now uni- 
versally accepted. Island universes it must be, and 
the rotation, previously given credence, must go. 

The best known of all island universes is undoubt- 
edly the Andromeda nebula, the only one visible 
to the naked eye in our latitudes. It is a great, 
spiral-shaped structure about fifty thousand light 
years in diameter, but rather flattened in one direc- 
tion, not unlike the Galactic system. It is almost 
one million light years distant, and contains millions 
and possibly billions of stars, the vast majority 
being too faint to be seen individually, although 
actually they may greatly surpass the sun in lumi- 
nosity. In fact, our sun, if removed to such an 
immense distance, would be of the twenty-seventh 
magnitude, more than one hundred times fainter 
than the faintest star within reach of our most 
powerful telescopes of to-day. In 188S a new star 
blazed forth near the centre of the Andromeda 
nebula, and reached the seventh apparent magni- 
tude. It is generally supposed that the star was 
really connected with the nebula, and thus also situ- 
ated at a distance of 1,000,000 light years: in that 
case, during the time of its maximum brightness, it 
must have been 100,000,000 times as luminous as 
our sun, one tenth as bright as the whole Andromeda 
universe. During the few days of its maximum 
splendor, it reigned supreme in the cosmos; it was 
the brightest star we have ever observed. It was 



also the most wasteful, for at that time it was so 
prodigal in radiating away its light and energy that, 
according to the theory of relativity, it was losing 
mass at the rate of more than one hundred million 
million tons per second. In addition to this phe- 
nomenally brilliant nova, about eighty-five others 
have been observed to date. These do, indeed, 
justify Lundmark's original supposition that they 
are comparable to those in the Galactic system, since 
these distant novas, too, reach a maximum brightness 
which is, on the average, about 10,000 times that 
of the sun. In addition to the nova:, scores of 
variable stars are known in the Andromeda nebula, 
discovered by Hubble at Mount Wilson. Recently 
there were found at Harvard two bright variable 
stars which are, in all probability, connected with 
the spiral. In that case their light varies between 
the limits of 25,000 and 100,000 times brighter 
than the sun. 

Although the Andromeda nebula is the nearest 
of all spirals, it is not the nearest island universe. 
Three others are known at present: a faint, nebulous 
looking object, known under the technical name of 
N. G. C. 6822, and the two Magellanic Clouds, 
which we have had occasion to mention already. 
They are both well visible to the naked eye in South- 
ern latitudes, the Large Cloud being about seven 
degrees in diameter, the Small Cloud about three 
and one half. To the naked eye they look almost 
as if they were patches torn off the Milky Way. 



It was from them that our first clue came concerning 
the luminosity of Cepheid variable stars, and they 
were the first objects whose distance was determined 
from the Cepheids, by Hertzsprung, who found a 
value of 30,000 light years. Using much more 
up-to-date material, Shapley has recently arrived at 
a value of 100,000 light years for their distance. 
Since these clouds are seen at a considerable angular 
distance from the great circle of the Milky Way 
in the sky, such a distance places them well outside 
the Galactic system. They are island universes in 
miniature, being no more than 7,000 and 14,000 
light years in size. In them we find all the different 
types of objects with which we are so familiar in 
our own Milky Way system: stars, variables, gase- 
ous nebulae, globular clusters, and we might well 
consider them as dwarf galaxies. Shapley has also 
found that the largest of the nebula; in the Magel- 
lanic Clouds is of such enormous dimensions that, 
if it were placed at the same distance as the stars 
in Orion, it would fill the entire constellation and 
appear as bright as Venus. It contains in its centre 
a variable star that alone is more than 100,000 
times brighter than the sun. Thus far only one 
new star has been discovered in the Clouds; it 
reached a maximum brightness at least 10,000 times 
greater than that of the sun, in full accord with the 
behavior of nova in other universes. 

At the southern station of the Lick Observatory 
the velocities in the line of sight of several gaseous 


nebula in the Clouds have been measured, and it 
has been found that the Large Cloud is receding 
from us at a speed of 170 miles per second, the 
Small Cloud with a speed of 100 miles per second. 
From their proximity in the sky, their similarity in 
structure, and from the remarkable progression in 
the velocities of the nebula situated in different 
parts of the Clouds, Hertzsprung has tentatively 
reached the conclusion that they may form a twin 
system, which, as a whole, is animated by a velocity 
of 400 miles per second. This speed is probably 
great enough to overcome the attraction of the 
Milky Way system, and allow the Magellanic 
Clouds to escape from our vicinity. If, however, 
the total mass of the Galactic system is much greater 
than we now have reason to suppose, it becomes 
barely possible that the Galactic system will prove 
the stronger of the two and force the Magellanic 
Clouds to circle around us, and form a kind of 
satellite universe to ours. 

During the past few decades a large amount of 
material on spiral nebula and island universes in 
general has been painstakingly accumulated by 
Curtis at the Lick Observatory, Wolf at Heidel- 
berg, Reynolds at Helwan, Egypt, and by Hubble 
at Mount Wilson, the last of whom, having the 
100-inch reflector at his disposal, could penetrate 
much deeper into space than his predecessors. As 
a result, a great variety of island universes have 
been found and when an attempt is made to arrange 


Mount miton Ob, 

I he great spiral nebula in the Big Dipper, an island universe of medium ace still in rhe 
process ol unwinding us spiral arms. The distance from us is probablv not far from 
5,000.000 light years. None of the stars shown in the photograph is visible to the naked 



these in different groups, according to their appear- 
ance, as has been done especially by Wolf, Hubble, 
and Lundmark, the strange fact appears that such 
an arrangement is nearly identical with that pro- 
posed by Jeans in England from purely theoretical 
considerations. In such a case we may feel reason- 
ably confident that Jeans's attractive conception of 
the origin and development of spiral nebula must 
have more than a germ of truth in it. 

According to Jeans, the primordial island uni- 
verse begins as an immense chaotic mass of glowing 
gas, very nearly in the form of a sphere. As time 
goes on, and probably as a result of rotation, this 
sphere begins to flatten out, the nebula taking on 
the shape of a bun or a lens. After having spent 
some time in this flattening process, and having con- 
sumed some trillions of years in it, such a system 
will begin to show signs of internal disturbance. 
Disruptions and eruptions take place, and the nebula 
begins to throw up solid matter from the interior, 
spiral arms may develop, and even stars may appear 
on the outside. In a few more trillion years it may 
have gone through a complete transformation; from 
an amorphous-looking mass of gas it has become a 
real spiral. After the nebula has thus found its 
destination, the process of development appears to 
become more orderly. The spiral slowly unwinds 
its arms, opening and producing more stars. It 
goes through stages similar to those of the Andro- 
meda nebula, and of the spirals in the Big Dipper 



and in Triangulum, shown in Plate XV and XVI. 
Once started, it cannot stop the process of disin- 
tegration; it is doomed, and ultimately even the 
spiral arms disappear. Nothing is left now but a 
swarm of stars, a great star cloud, such as the 
Magellanic Clouds, and, in all probability, the 
Galactic system. From now on, it becomes possible 
that some of the more audacious stars begin wan- 
derings of their own, leaving the mother universe 
and venturing forth in the depths of space. The 
majority of the stars will probably remain faithful 
to the main body, but it is now only a question of 
time as to how long they will manage to keep alive. 
The smaller stars die first, after having sent their 
light and heat into the insatiable cold of empty 
space. The larger stars follow, and finally the 
whole galaxy is reduced to a conglomeration of 
dying embers treading their danse macabre through 
the voids of creation and waiting to be gathered 
into the scrap heap of the cosmos, ultimately, 
perhaps, to be re-kindled. 

Such is the drama of the evolution of an island 
universe as we now see it. With the aid of Jeans's 
theory and Hubble's observations we can scent the 
course of nebular evolution along an unmistakable 
path, looking into the past for hundreds of millions 
of years, and into space for sextillions of miles. 
For we must not forget that, in dealing with island 
universes, we are dealing, in Hubble's words, with 
a "history of receding horizons." If the light of 

Photograph Mount H'iUon Obiervatory 

PLATE xvi 

The spiral nebula in Triangulum, at a distance of 1,000,000 light years and about 20,000 
light years in diameter. If removed to this distance the sun would appear about 1,000 
times fainter than the faintest star on the plate. This island universe is on the verge of 
breaking up into individual stars and of losing its spiral character completely; it is prob- 
ably not much younger than our Milky Way system. 



the Andromeda nebula takes a million years to reach 
us, then we see this nebula, not as it is now, but as 
it was one million years ago. Even the nearest of 
all island universes, the Magellanic Clouds are seen 
only as they were 100,000 years ago, long before 
the last ice age on earth. The majority of island 
universes we observe in times far, far earlier than 
the appearance of man on earth. 

From the scheme of evolution of a universe we 
may perhaps infer that island universes occupying 
the same stage of evolutionary progress are com- 
parable in size. Basing our researches on this 
assumption, we can then make deductions concerning 
the relative frequency of each type of universe, and 
about the total number of galaxies in a given volume 
of space. In short, we can begin to study the popu- 
lation of that greater cosmos, the cosmos of island 
universes. One significant thing strikes us imme- 
diately; namely, that among all the different island 
universes now known there is not a single one that 
surpasses the Galactic system in size. They may 
all be more or less comparable to each other. The 
Andromeda nebula represents thus far the maximum 
dimensions, and even it, with its 50,000 light years 
diameter, is much inferior to our Galactic system, 
with its diameter of 300,000 light years. These 
extraneous stellar systems may all be island uni- 
verses; our Milky Way system is still the only 
continent among them. 



When the population of spiral nebula? is studied 
in detail, it is soon noticed that "twin universes" 
are not uncommon ; even whole clusters of universes 
have been observed, as, for example, the enormous 
cluster of spirals in Virgo and Coma Berenices, 
studied by Wolf, Lundmark, and Shapley. There 
appears to be little doubt, however, that this cluster 
is not a "super-universe," but rather an archipelago 
of sidereal coral reefs, perhaps mostly smaller than 
the Magellanic Clouds in size, and at a distance of 
some 10,000,000 light years. From the most recent 
data on island universes Hubble has calculated that 
they are on the average about 1,500,000 light years 
apart. Since the average size of these galaxies is 
probably not much more than 10,000 light years, it 
is obvious that island universes are not crowding 
each other, and that collisions between universes 
must be rather rare occurrences. 

Another significant result of the island universe 
development is that it permits us to make an esti- 
mate of the size of the cosmos. According to 
relativity, the cosmos is not infinite, but finite in 
all directions. So far as man can see it and observe 
it as a universe of three dimensions, the cosmos 
must be limited in size. The new data on island 
universes indicate that the maximum size of this 
great cosmos is perhaps in the neighborhood of 
100,000,000,000 light years. This does not mean 
that there is nothing beyond, but only that we ter- 
restrials, living and thinking in three dimensions, 



could never hope to observe it: "The fault is not 
in the stars, but in ourselves." Paradoxically, how- 
ever, Einstein's theory does not state that we could 
see the end of the universe. Far from it. To us 
the universe must always appear unbounded, though 
it is finite. It is much the same way with us as with 
a flat ant crawling on the curved surface of a sphere. 
If the ant stays on the surface, it will never come 
to an end; it can crawl through eternity, and will 
never find that the surface of the sphere is bounded 
anywhere, yet it cannot get more than a certain 
distance away from its starting point. Half the 
circumference of the sphere would be this greatest 
distance, and if the ant were only able to perform 
measurements, it would find that the surface of the 
sphere is finite and equal to a perfectly definite 
amount of square inches. So it is with us; we live 
not on the curved surface of a sphere, but in a 
"curved space" in a universe which is finite yet 
unbounded. To the philosopher it may seem an 
extraordinary idea to try to limit the universe; the 
astronomer is not much disturbed by it, especially 
since he cannot, even with the most powerful tele- 
scopes at his disposal, penetrate into space deeper 
than one hundredth of the "greatest" distance. 

What is the outcome of all this new information ? 
Our giant telescopes have temporarily brushed aside 
the eternal veil of empty space, and in less than 
ten years we have advanced our milestones from 
100,000 light years to 100,000,000, only to find 



l.a S 

5. S 

09 a 

-; <»8 H 

§•1? a 

«. 3 g 

fc-Sw g 


g M 

.5 o 












a 1 ! 

(flu. Q 




.-as; »< 

- ~2 ■< 

I 5! 




o^ §0 w-3 §"3«-. 

«J «! 5 «„ e-o S g 

EE2E^«SE n 
.2.2.2 »ai5 g = 5 


15 t> 

o ©"©" 0006 ©' 

- 8* 






c u 

€ - 





o •• 

<« a* 
s =» 







that still more are awaiting us. The "eternal 
silence of infinite space" which so frightened Pascal 
has ceased to exist for us. Silence reigns no more 
in the realm of space, but has given way to a Babel 
of light messages from countless island universes, 
past, present, and future. Not content with a mere 
description of the material universe, we have gath- 
ered on our photographic plates a true record of 
the evolutionary drama of this universe, past and 
present. In a perhaps not entirely futile attempt 
to comprehend this time scale of the cosmos, we 
have fatigued our imagination to the breaking 
point, only to realize that in our apparent advance 
against infinity and eternity we never really gain 
ground. Our mind stands perplexed, our intellect 
overwhelmed, before the joint concept of infinity 
existing throughout eternity. 

Chapter XVI 

"We end, not with a period, but with a question 

— Poincar6. 

In our analysis of the astronomer's findings of the 
material universe we have perhaps not sufficiently 
stressed our gratitude to the eternal darkness of 
space. If the sun were always visible we might even 
now be completely ignorant of everything else in 
the universe ; in the light of day it is as though 
seeing, we could not see, hearing, we could not 
understand. It is only the black and silent night, 
the perpetual gloom that pervades the chasms of 
infinity, that to us becomes the true torch and the 
true speech of science. Its very stillness is eloquent 
of that, it subdues a hostile universe, and makes 
audible the "voices of the past which speak to us 
from the depths of the vanished ages." 

Our interpretation of the facts has been, I fear, 
a trifle too terrestrial. Let us therefore, in a final 
summary, again review the pageant of the stars, 
freed, if such be possible, from our human preju- 




dices. To begin with, what is a star? A sphere 
of glowing gas, varying in size from a globe as 
small as the earth to one a thousand times larger 
than the sun in diameter, large enough almost to 
take in the entire orbit of Jupiter. In density such 
a mass of "gas" may vary from ten thousand times 
rarer than our atmosphere to several hundred thou- 
sand times denser than water. After a star comes 
a galaxy, which, when described in terms under- 
standable on earth, would have to be called a 
vacuum. In taking the census of the Milky Way 
system, the best known of all galaxies, we find that 
the number of stars, when compared to the unfath- 
omable stretches of space at their disposal, is so 
pitifully small that it is only equaled by the mole- 
cules of air when one cubic inch of normal air is 
allowed to expand over a cube of 150-miles side. 
Advancing still another step, we find that such a 
galaxy, though seemingly a vacuum when measured 
against terrestrial standards, represents a great 
condensation of matter in space: galaxies of sizes 
generally not more than 50,000 light years are 
parceled out at average distances of no less than 
1,500,000 light years. 

If for a moment we again have recourse to an 
illustration drawn from our inadequate terrestrial 
units, and liken the Milky Way system to some- 
thing as small, pro rata, as the state of Massachu- 
setts, we are effecting a reduction in the ratio of 
ten thousand million million to one. The Andro- 





meda nebula, on this scale some thirty miles in size, 
would be slightly farther away than Washington, 
D. C.J the great spiral nebula in the Big Dipper 
perhaps as far as San Francisco. The Hawaiian 
Islands might appropriately represent the extended 
cluster of spiral nebula in Virgo, in distance as well 
as in size, although perhaps being somewhat defi- 
cient in number. The remotest galaxy within reach 
of the 100-inch telescope would perhaps be 60,000 
miles distant, one fourth the distance from the 
moon, and the entire "Einstein Universe" would 
appear condensed within a space one eighth as large 
as the distance from the earth to the sun. Having 
reduced the cosmos to this state of helplessness in 
our imagination, even a ray of light must appear 
paralyzed: instead of 186,000 miles per second, it 
now moves at the sluggish rate of 3 feet per year. 
But our interest lies mainly in the Milky Way sys- 
tem, inside the state of Massachusetts in the above 
comparison. A vast assemblage of stars such as 
the globular cluster Omega Centauri has dwindled 
down to something comparable to the Harvard 
Stadium. The distance between the sun and the 
nearest star, four and one-half light years, has been 
reduced to thirteen feet, the sun itself to a particle 
only six one-millionth parts of an inch in diameter. 
If light appeared indolent to us on this small scale, 
we cannot be surprised to find the stars practically 
standing still; their average speed of 20 miles per 

second has been annihilated, and they retain a veloc- 
ity of only 4 inches in a thousand years. 

As terrestrials we are not satisfied until we have 
put our earth in this picture. We may do so, but 
we shall never find it again: encircling the pin point 
that represents the sun, in a circle of just one thou- 
sandth of an inch in diameter, we find our earth, 
as a small, completely dark speck of matter one 
twentieth of one millionth of an inch in size. Our 
best microscopes could not possibly reveal it, and 
it would remain what it is to-day in the cosmos, an 
obscure, and infinitesimal nonentity, inhabited by 
phantom pygmies. 

We have not yet finished. Our position of utter 
insignificance is made even more ignominious by the 
fact that what we, on earth, consider the usual 
aspect and the usual behavior of the universe are, 
in reality, features and conditions of extreme im- 
probability. What to us are the essentials of exist- 
ence are in the universe merely results of a fortui- 
tous coincidence, and we may perhaps, defiantly, 
find comfort in the thought that we are freaks of 

We have already seen that "daylight" such as 
we enjoy for half the time, is an exceptional state 
of affairs. Throughout space darkness reigns, 
almost Stygian darkness, broken only by the feeble 
radiation of the distant stars. Within the confines 
of a galaxy such starlight may attain a total inten- 
sity of slightly less than 1 per cent, of the full 



moon, about equal to that of a candle at a distance 
of fifty feet. In space at large, in the voids between 
galaxies, even this little is dimmed by distance, or 
by "fog" perhaps, and the total illumination might 
well be one thousand times less still. 

We have likewise seen that the conditions that 
cause the birth of a planetary system must be of 
extremely rare occurrence, and that, therefore, our 
earth and the solar system, though in all probability 
not unique, must yet be regarded as reasonably un- 
common in the realm of the cosmos. 

On the surface of the earth the greater part of 
matter is in the liquid or solid state. In the stars 
matter is not only gaseous, but the high stellar 
temperatures have broken it up in simpler constitu- 
ents, atoms, and electrons. Iron, on earth, is solid; 
in the universe at large it exists solely as vapor, 
and in this vapor in the form of mutilated atoms 
which have lost some of their electrons. Solid 
matter is an exceedingly rare occurrence, and it may 
well be that a large portion of all matter exists as 
protons and electrons. 

The universe, as we now see it, is but a transient 
structure; it is dissipating its energy and mass with- 
out, so far as we can now ascertain, receiving any- 
thing in return. The sun is losing weight at the rate 
of one trillion tons per second; the whole Andro- 
meda nebula is dissipating at least one billion times 
as much, and in two weeks loses as much mass as 
the whole earth weighs. Our own Milky Way 



system is in all probability even more prodigal, and 
if we combine the radiations of all the galaxies in 
the cosmos, we find that, at the very least, a mass 
equal to that of our sun is being destroyed daily. 

Here we have approached the Sphinx of modern 
science. The modern OZdipus who will solve her 
riddle has not yet risen, but, to draw the analogy 
closer, the ominous portents are already becoming 
apparent that, if he succeeds, it will only result in 
dire tragedy for his parents, the sciences of physics 
and astronomy. We must not forget, however, 
that no formulation of any law of the universe 
hitherto reached is certainly final; those laws found 
thus far by human science have to be modified 
continually. There is no special sanctity attached 
to the doctrine of relativity, or to the quantum 
theory. They do explain the great majority of 
physical facts now within our ken, but there can be 
no doubt that, at best, they are but imperfect por- 
trayals of the real laws of nature, framed with 
human fallibility. In our description of the cosmos 
the truth of to-day is no more than an ephemeral 
phantom ; it will turn into a falsehood when viewed 
in the light of the truth of to-morrow. In the 
last analysis the astronomer, and especially the 
cosmogonist, must abandon all positive assertions, 
and there is no need, therefore, to feel more than 
temporarily distressed by the logical and lamentably 
unavoidable conclusion of our present-day theories, 
so admirably voiced by Jeans in his sayings: "At 



present the universe seems to be running down like 
a clock which no one winds up. . . . The widely 
desired cyclic universe, in which just as much matter 
is created as destroyed, would seem to be a universe 
already dead. . . . With universes, as with hu- 
manity, the only possible life is progress to the 

With this we enter the domain of the un- 
known, and approach that of metaphysics, and 
it is with true scientific reticence that the astron- 
omer pauses on the threshold before presenting 
what might be termed his credo; for, unlike the 
metaphysician, the scientist does not try to divine 
a purpose behind it all — the scientist is satisfied 
with a descriptive explanation. After the portrait 
we have painted of the importance of the earth in 
the universe, and in view of the delicate balance 
needed between a multitude of diverse factors to 
produce and maintain human life, it would be 
superfluous to stress man's manifest insignificance 
in the material universe. Man's universe tran- 
scends the material; the reality of the immaterial 
takes its place beside the accepted existence of the 
material. In our studies of the material universe 
we may sometimes feel that we have lifted the veil 
that shrouds the mysteries of the empyrean, and 
reflect that the universe, so invincibly immense in 
size, may, after all, not be so invincible in concep- 
tion. Having reached the portal of infinity, we, 
like Byron's Prometheus, "in portions may foresee 



our own funereal destiny." In our contemplation 
of the immaterial we may then, freed from the 
bondage of our concepts of boundlessness and im- 
mortality to those of infinite extension and everlast- 
ing existence, feel that we have safely arrived at a 
point in our meditation from which we can ponder 
the enigma of infinity during the leisure of eternity. 




Abbott, 42. 

Adams, J. C, 111, 112. 
Adams, W. S., 162, 180. 
Aitken, 189. 
Argelander, 139. 
Aristotle, xvi. 

Bailey, 214, 217, 249. 

Barnard, 103, 166, 219, 249. 

Bessel, 156. 

Bode, 12, 91, 114. 

Boltzmann, 148. 

Bond, G. P., 107. 

Bond, W. C, 193. 

Bowen, 228, 229, 232. 

Brahe, Tycho, 8, 206. 

Bredichin, 131. 

Brown, 54. 

Bruno, Giordano, 8. 

Burton, 5. 

Burnham, 189. 

Campbell, 72, 171. 

Cannon, 146, 147. 

Cassini, 107. 

Chamberlin, 5, 6. 

Clairaut, 123. 

Coblentz, 86. 

Copernicus, xvi, 8, 9, 21, 78, 

Cowell, 123, 124. 
Croramelin, 123, 124. 
Curtis, 261, 262, 268. 

Darwin, 49. 
Dawes, 107. 

Delaunay, 54. 
Doppler, 170. 
Draper, 146. 
Dyson, 72. 

Eddington, 72, 234, 239, 240, 

241, 242, 243. 
Einstein, 9, 79, 277. 
Encke, 129. 
Eratosthenes, 17. 

Fabricius, 34, 199. 
Fizeau, 148. 
Fleming, 146. 
Foucault, 21. 
Fraunhofer, 144. 
Frost, 193. 

Galilei, Galileo, 8, 34, 55, 101, 

Galle, 112. 
Gauss, 92. 
Goodricke, 198. 

Hale, 36. 
Hall, 84. 
Halley, 122. 
Hansen, 54. 
Harding, 92. 
Harper, 162. 
Hartmann, 224. 
Helmholtz, 45. 
Henderson, 156. 

Herschef, 110, 188, 190, 219, 




Hertzsprung, 148, 151, 161, 162, 
183, 192, 203, 214, 220, 234, 
237, 238, 253, 267, 268. 

Hipparchus, 7. 

ilirayama, 97. 

Hubble, 220, 262, 265, 268, 271, 
272, 276. 

Muggins, 145, 171. 

Huyghens, 107. 

Junes, 180. 

Jeans, 5, 6, 203, 234, 241, 243, 

271, 272, 285. 
Jeffreys, 4, 5, 19, 102. 

Kant, 258. 

Kapieyn, 185, 207, 251, 252, 

253, 255. 
Kepler, 8, 91, 207. 
Kirkwood, 94. 
Kohlschiitter, 162. 

Lalande, 123. 
Lampland, 86, 101. 
Langley, 42. 
Laplace, 5, 6. 
Leavitt, 202. 
Leverrier, 78, 111, 112. 
Lindblad, 256. 
Lockyer, 146, 234. 
Lodge, 179. 
Lowell, 89, 90. 

Lundmark, 206, 261, 262, 271, 

Van Maanen, 177, 261, 262. 

Maury, 146. 

Melotte, 104. 

Metcalf, 119. 

Michelson, 148. 

Mitchell, 72. 

Montanari, 198. 

Moulton, 5, 6. 

Newcomb, 54, 78, 143, 236. 
Newton, xvi, 8, 9, 24, 54/ 122, 

191, 193. 
Nicholson, 86, 104. 

Olbers, 92. 
Oort, 256. 

Pascal, 279. 

Peary, 133. 

Pease, 148, 151. 

Perrine, 104. 

Pcttit, 86. 

Piazzi, 91, 92. 

Pickering, E. C, 140, 141, 145, 

146, 193. 
Pickering, W. H., 90, 115. 
Pigott, 199. 
Plaskett, 224. 
Plato, xv. 
Poincare, 54. 
Pontecoulant, 123. 
Proctor, 1S2. 

Ptolemy, 7, 156. 

Ramsay, 44. 
Reynolds, 268. 
Van Rhijn, 144, 249. 
Riccioli, 188. 
Roemer, 103. 
Rosse, 258. 

Russell, 161, 192, 198, 228, 232. 
234, 238, 239. 

Saha, 146. 
Scheiner, 34. 
Schiaparclli, S9. 
Schlesinger, 159. 
Schwassman, 121. 
Scares, 249. 
Secchi, 145, 146, 147. 
Shaw, Knox, 204. 


Shapley, 198, 203, 214, 255, 267, 

Slipher, E. C, 86. 
Slipher, V. M., 218, 220, 259. 
Stefan, 148. 
Struve, O., 224. 
Struve, W., 156, 188. 
Swedenborg, 258. 

Titius, 12. 
Trumpler, 72. 
Turner, 157. 


Vogel, 171. 


Wachmann, 121. 

Wien, 148. 

Wilsing, 42. 

Wolf, 124, 219, 268, 269, 276. 

Wright, Thomas, 258. 

Wright, W. H., 86, 206, 228. 

Yamamota, 135. 
Young, C. A., 136. 
Young, R. K., 162. 



Absolute magnitude, 160. 

Age, of earth, 4, 16, 17; of so- 
lar system, 4, 13; of stars, 
184, 235, 244; of sun, 13, 244. 

Algol, 196, 198, 199. 

Almagest, 7. 

Andromeda nebula, 258, 261, 
262, 265£f, 271, 275, 281, 284. 

Aphelion, 77. 

Asteroids, 11, 94ff, 108. 

Astronomical unit, 32, 157. 

Atmosphere of earth, 20, 61; of 
moon, 59; of Venus, 81; of 
Mars, 86; of Jupiter, 101. 

Atom, 16, 41, 178, 240, 284. 

Aurora, 36, 39, 61. 

Betelgeuse, 147, 149, 160, 161. 

Big Dipper, 182, 258. 

Binaries — sec double stars. 

Bode's law, 12, 91, 114. 

Brightest stars, 175ff. 

Brightness of stars, 160. 

Calcium, 40, 43, 14S. 

Calendar, 48. 

Calorie, 44. 

Candle-power, 141. 

Canals of Mars, 89. 

Cassini's division, 107. 

Celestial mechanics, 190. 

Cepheid variables, 201rr, 262. 

Ceres, 91. 

Chemical elements, 145. 

Chromosphere, 40, 71. 

Clusters, moving, 182ff; globu- 
lar, 213; open, 213; nebulae, 
276, 282. 

Cluster variables, 202. 

Coal sack, 219, 249. 

Collisions, 4, 186. 

Colour, of stars, 144; of sun, 42. 

Comets, 115, 117ff; Halley's, 
122ff, 135; Encke's, 129. 

Conjunction, 78, 103; inferior, 
78; superior, 78. 

Constellations, 138. 

Copernican system, 8, 9. 

Corona, 40, 61, 64, 71, 72. 

Cosmogony, 271ff. 

Crepe ring, 107. 

Deflection of light rays, 71. 
Density, of earth, 19; of sun, 

34; of Jupiter, 100; of stars, 

150, 178; of space, 223. 
Diameter of stars, 148ff. 
Dissipation of energy, 284. 
Distance, of sun, 3 Iff; of stars, 

152S, 193; of spiral nebulr, 

Double stars, 188ff; visual, 

18817; spectroscopic, 190ff; 

optical, 193. 
Dust clouds, 223. 
Dynamic encounter, 6. 

Earth, 14ff, 283. 
Eccentricity, 9, 94, 






Eclipse, of sun, 40, 61, 63 If; of 
moon, 65, 69, 72; of Jupi- 
ter's satellites, 102. 

Eclipsing variables, 196ff. 

Ecliptic, 24, 25, 41, 94, 102, 104, 

Einstein effect, 71. 

Electrons, 178, 180, 240, 284. 

Ellipse, 9, 49, 53, 94, 97, 100, 
119, 120, 190, 191. 

Equator, 22, 25, 35. 

F.ros, 94, 97. 

Evening star, 80. 

Evolution, of solar system, Iff; 
of stars, 233ff, 242; of nebu- 
la;, 27 Iff. 

Fa cula;, 40. 

Families of asteroids, 97. 
Focus of ellipse, 9, 190. 
Foucault's pendulum experi- 
ment, 22. 

Galactic system, Galaxy— see 

Milky Way. 
Gravitation, 8, 9, 190, 191. 

Heliocentric system, xvi. 
Helium, 20, 40, 44, 45, 147, 243. 
llyades, 213. 
Hydrogen, 20, 40, 43, 45, 145, 

146, 147, 170, 178, 243. 
Hyperbola, 120. 

Interferometer, 148, 189. 
Interior, of earth, 19; of sun, 

45; of star, 179ff, 206. 
Interstellar gas, 223. 
Infra-Mercurial planet, 82. 
Iron, 131, 134, 145, 284. 
Island universes, 257ff. 

Jupiter, 23, 91, 94, 99ff; satelli- 
tes discovered on, lOlff. 

Kapteyn Universe, 251ff. 
Kepler's laws, 8. 

Life on Mars, 89. 

Light, speed, 103, 207; pres- 
sure, 131, 239; year, 158; 
curve, 198, 202. 

Long period variables, 201. 

Luminosity of stars, 160, 192, 
203, 260. 

Magellanic Clouds, 202, 266ff, 

272, 275. 
Magnetic storms, 36, 39; fields 

on sun, 39. 
Magnitude, of star, 140ff; ab- 
solute, of star, 160; of moon, 

141; of sun, 141. 
Main sequence, 237ff. 
Mars, 82ff, 91, 94, 97. 
Mass, of earth, 18; of sun, 34; 

of moon, 48; of Jupiter, 100; 

of stars, 191ff, 234, 239. 
Matter, annihilation of, 46, 

242ff; structure of, 178ff. 
Mercury, 23, 77ff. 
Meteor crater, 134. 
Meteors, 20, 56, 108, 133ff. 
Midnight sun, 25. 
Milky Way, 61, 245ff, 262, 281, 

Mira, 199, 200. 
Mizar, 188, 193ff. 
Molecules, 59. 
Month, 53. 

Moon, 11, 15, 48ff, 141. 
Morning star, 80. 

Nearest stars, 176. 

Nebula;, diffuse, 218ff; dark, 
219; gaseous, 219; planetary, 
227ff, 231; spiral, 258ff. 

Nebulium, 228ff. 

Neptune, 45, 91, 113. 
New stars — nova:, 204ff; in 
spirals, 260ff, 265. 

Nova Pictoris, 208. 

Nucleus of atom, 16. 

Radial velocities, 169ff, 227. 

Radioactivity, 16. 

Relativity, 71, 79, 242, 243, 262. 

266, 276, 285. 

SuXlr^ff; of double Reverse layer, 40. 
t .„ isq 195- variable Rigel, 160, 353. 

:;:;:; mSUS i*K «>• ^ ° f saturD - io7 

Oppositions, 84, 103. 

Orbits, 83, 94, 106, 110, 114, 

115, 119. 
Origin of solar system, Iff. 
Orion nebula, 220, 227, 228, 259. 

Parallax, 154. 
Parsec, 157. 

Pendulum, Foucault, 21; New- 
comb, 236. 
Perihelion, 72. 
Period luminosity curve, 201, 

202, 267. 
Perturbations, of Uranus, 111; 

of Neptune, 114. 
Phases, of Mercury, 78; of 

moon, 52. 
Photography, 139, 235. 
Photometer, 40. 
Photosphere, 40. 
Planets, 11, 77ff, 99ff. 
Pleiades, 213, 220. 
Polar caps of Mars, 86. 
Polaris— Pole Star, 28, 139, 201. 
Poles, 23, 35. 
Precession, 28. 
Prominences, 40. 
Proper motion, 166. 
Protons, 178, 180. 
Ptolemaic system, 7. 
Pulsating stars, 202. 

Quantum theory, 285. 

Saros, 70. 

Satellites, 82, HI; of Jupiter, 
101 ff; of Mars, 84ff; of Sat- 
urn, 109. 
Saturn, 91, 97, 104ff. 
Scorpius, 246, 249. 
Seasons, on earth, 24; on Mars, 

84; on Jupiter, 100. 
Shadow, umbra of, 67; penum- 
bra of, 67. 
Shooting stars— see Meteors. 
Sirius, 28, 139. 141, 147, 164, 
166, 171, 177, ISO, 183, 194, 
Sodium, 132. 

Solar energy, source of, 44, 45, 
284; system of, 1, 10; mo- 
tion of, 174. 
Southern Cross, 181, 250. 
Spectra, of sun, 42ff; of stars, 
144ff, 161ff; of nebula;, 219, 
228, 259. 
Spectral class, 145, 147, 161, 

Spectroscope, 40, 41ff, 108, 144, 

163, 189, 235, 259. 

Spectroscopic binaries — sec 
double stars. 

Stars, dwarf, 161, 237; giant, 
161, 201, 217, 237; white 
dwarf, 177ff, 239, 241, 242, 
244; variable, 199ff; new, 



Stellar, motions, I64ff, 254; ev- 
olution, 233ff, 242; energy, 

Sun, 3 Iff, 141; spots, 34ff. 

Taurus cluster, 213. 
Temperature, of sun, 41, 42, 
147; of Mars, 86; of Jupiter, 
101; of stars, 144, 146, 147. 
Tides, 2, 48, 49, 95. 
Time scale of universe, 272. 
Transit, of Mercury, 78; of 
Venus, 81; of Jupiter's satel- 
lites, 103. 
Trans-Neptunian planet, 114. 
Trojan asteroids, 97. 

Universe, island, 257ff; limit* 

of, 276. 
Uranus, 91, 98, HOff. 
Ursa Major— see Big Dipper. 

Variable stars, 199ff. 

Variation of latitude, 23. 

Velocity, of light, 103; of clus- 
ters, 218; of radial, I69ff, 
190; of nebula;, 227, 260. 

Venus, 80ff. 

Vulcan, 82. 

Water, on moon, 56, 59; on 

Mars, 86. 
Worlds, plurality of, 284. 

Zodiacal light, 40, 61. 

it * 





%' ■ 



: ---. _)r ' 

r <- \