-
Sigma Xi, The Scientific Research Society is collaborating with
JSTOR to digitize, preserve and extend access to
AmericanScientist.
http://www.jstor.org
Sigma Xi, The Scientific Research Society
THE ORIGIN AND EVOLUTION OF THE UNIVERSE Author(s): G. GAMOW
Source: American Scientist, Vol. 39, No. 3 (JULY 1951), pp.
392-406Published by: Sigma Xi, The Scientific Research
SocietyStable URL: http://www.jstor.org/stable/27826381Accessed:
01-04-2015 19:58 UTC
REFERENCESLinked references are available on JSTOR for this
article:
http://www.jstor.org/stable/27826381?seq=1&cid=pdf-reference#references_tab_contents
You may need to log in to JSTOR to access the linked
references.
Your use of the JSTOR archive indicates your acceptance of the
Terms & Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars,
researchers, and students discover, use, and build upon a wide
range of contentin a trusted digital archive. We use information
technology and tools to increase productivity and facilitate new
forms of scholarship.For more information about JSTOR, please
contact [email protected].
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
AGE PHYSICAL CONDITION EVOLUTIONARY PROGRESS
ZERO POINT TO
5 MINUTES LATER
Temperatures of
many billion
degrees. Uniform very high densities.
Primordial y/em, in
the dense field
of radiant energy.
o - e/ectron
> QC ?
0-neutron Q
5 MINUTES
TO
HALF-HOUR
Temperatures of one billion
degrees and below.
Uniform material densities below 0.1 percent of
atmospheric air.
Neutrons and protons begin to stick
together forming
composite nuclei.
-deuferon
o
ABOUT 30
MILLION YEARS
Temperature of 300? abs. (room
temp.) Mean
density about
10-24g-'m3
Lukewarm gas breaks
up into primordial
gaseous galaxies.
ABOUT 3
BILLION YEARS
Temperature of intergalactic space, few
degrees abs. Same density
within galaxies.
Material within galaxies condenses
into stars. Planetary systems
formed. Elementary life begins.
FEW
HUNDRED
MILLION YEARS
Temperature maintained at
few degrees abs. by stars. Same density within
galaxies.
The developed human intellect attacks the
problems of the origin of the universe.
Pictorial History of the Universe
(See pages 397-405.)
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
AMERICAN SCIENTIST
SUMMER ISSUE JULY 1951
THE ORIGIN AND EVOLUTION OF THE UNIVERSE
By G. GAMOW The George Washington University
The Age of the Universe
THE problem of the origin of the world has been occupying
human
mind ever since the dawn of history. All ancient religions,
which were, essentially, the first attempts of awakening intellect
to find its
place in the surrounding world, discussed the problem of
creation at considerable length. Some of them even went so far as
to give the exact date of the "creative act." Thus Archbishop
Ussher, in the seventeenth century, concluded from the narratives
of the Old Testament that the world was created in the year 4004
b.c. Much more elaborate calcula tions by the occult scientists of
ancient India lead to the date which would make the world
1,972,949,052 years old as of today. Modern estimates, based on
detailed studies of various evolutionary features of the universe,
do not claim the precision of the ancient thinkers, but they all
agree that the zero point of the history of the universe must be
placed at a few billion years ago.
There are two different geological methods for estimating the
date when the earth was formed: one of them leads to the age of the
oceans, the other to the age of the continents. We can get a fair
idea concerning the age of oceans by studying the salinity of ocean
water. This water contains about 3 per cent of dissolved salts,
which, if extracted and piled up on the land, would cover the area
of the United States by a layer almost two miles thick. How did all
this salt come into the ocean?
Strange as it may sound at first, salt is being brought into the
oceans by the rivers, which wash it away from the rocks forming the
crust of the earth. While water evaporates from the ocean surface,
and falls again on the continents to repeat its eternal cycle, the
dissolved salt stays in and gradually increases the salinity of
ocean water.
Based on the Sigma Xi-Resa National Lectureship, 1950-1951, and
on the author's
book, Creation of the Universe, soon to be published. All rights
reserved.
393
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
394 American Scientist
Geologists estimate that every year the rivers bring into the
ocean about 400,000,000 tons of salt. Since the present amount of
ocean salt is 40,000,000,000,000,000 tons, the process must have
lasted for at least 100,000,000 years. This figure must be
increased by a factor of a few tens, since it is known that at the
present epoch the erosion of con tinents is abnormally high, and
that during most of the geological time (when there were much fewer
mountain ranges than now) the erosion was
only a small fraction of its present value. Thus, the fact
itself that the oceans are not saturated with salt proves that they
could have existed only for a limited period of time, while the
date of their formation may be set at a few billion years ago.
The age of the continents can be estimated by measuring the age
of various rocks from which they are formed. It is known that many
rocks contain small deposits of radioactive elements, uranium and
thorium, which are slowly decaying into lead. Once the rock is
solidified from the originally melted state, this radiogenic lead
stays together with the original radioactive elements. Therefore,
by measuring the uranium/ lead and thorium/lead ratios, we can get
a rather exact figure for the age of a given rock, in the same way
as one can find how long a furnace was
burning by comparing the amounts of remaining coal and
accumulated ashes. Using this method, one finds different ages for
the rocks of dif ferent geologic formations, but in not a single
case does this age exceed the value of two billion years. We can
consider this figure as the lower limit, and, possibly, as a good
actual value for the age of the earth.
It may be noted that a few years ago a British geologist, A.
Holmes, proposed a more intricate method based on the study of
uranium-lead transformation prior to the solidification of the
crust. By comparing the relative amounts of radiogenic leads in the
rocks of different geological ages, and making certain assumptions
concerning the processes of ore deposition, he arrived at the
result that the formation of radiogenic lead must have started
3,350,000,000 years ago. This figure is supposed to represent, not
so much the age of the earth itself, as the date at which the
"freshly formed" radioactive atoms must have started to decay into
lead.
Astronomers have essentially three different methods for judging
the age of the stellar universe. The first is based on the study of
stellar motion within our system of the Milky Way, and refers to
the statistical distri bution of stellar velocities which is
expected to approach a certain "limiting distribution" (the
so-called equipartition of energy between all stars) when the
stellar system has existed for a sufficiently long time. The
observed velocity distribution is still some way off from that
"limiting distribution," which, according to mathematical
calculations, indicates that the system must have existed for only
a few billion years. The second astronomical method is based on the
study of stellar
energy sources. We know in fact that stars, and in particular
our sun, derive their energy from the slow nuclear transformation
of hydrogen into helium taking place in their hot central regions.
Thus, the natural life span of a star is determined by the rate of
its burning (that is, by its
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
The Origin and Evolution of the Universe 395
absolute brightness) and by the original amount of hydrogen it
contains. Since the brightness of stars is known to increase as the
cube of their mass, and the amount of nuclear fuel is simply
proportional to the mass (hydrogen forms about half of the total
mass in a normal star), the brighter stars will burn out faster
than fainter ones, their life span being inversely proportional to
the square of their mass. Our sun is a com paratively faint star;
its total life span can be calculated to be about 50 billion years.
If the sun is only a few billion years old, it may be com pared
with a baby who is just learning to walk.
The stars, which are five times heavier than the sun, burn 25
times as fast and have a life span of only about two or three
billion years. Ob servational astronomy reveals that the stars of
just about that mass, seem to be on the verge of hydrogen
exhaustion. The dwindling of their fuel supply is manifested in all
kinds of "unquiet behavior" ranging from regular pulsations of
their giant bodies (Cepheid variables) to terrific explosions
(novae and supernovae) which tear these stars apart. It therefore
seems reasonable to conclude that most of the stars forming the
system of the Milky Way were originally formed two or three billion
years ago, and that in the case of pulsating and exploding stars we
ob serve the death agony of those members of the stellar community
who are coming to the end of their natural life during the present
epoch of the history of the universe.
The third astronomical method of estimating the age of the
universe is based on the phenomenon of universal expansion
discovered by the
Mount Wilson astronomer, E. Hubble, about a quarter of a century
ago. We know that the stellar system of the Milky Way, containing
our sun
along with several billions of other stars, is not a lonely
island in the infinite expanses of the universe. Large telescopes
reveal that the space outside our Galaxy is populated by myriads of
similar stellar systems scattered more or less uniformly all the
way to the limit of telescopic vision.
There are nearly one billion such galaxies within the range of
the 200-inch telescope of the Palomar Mountain Observatory, and the
author was told by Professor Harlow Shapley of Harvard that when
ever he has a new graduate student he sends him up to the telescope
with orders "to discover a new galaxy and to name it." The striking
thing about these distant galaxies of stars is that the light
emitted by them, while similar to the light coming from nearby
galaxies, shows however the peculiar phenomenon of a shift of all
spectral lines towards the red end of the spectrum. A simple
physical explanation of this "red shift" lies in the assumption
that the galaxies are receding from us at rather
high speeds. This so-called Doppler effect, consisting in the
change in
frequency of waves emitted by an approaching or receding source,
is a familiar phenomenon in the field of acoustics. Everyone has
noticed how the aggressive high-pitched honk of an approaching,
fast-driven car goes into a much lower departing tune as the car
passes by.
In optics the same effect will make the light of an approaching
source look bluer, and that of a receding source redder, than it
actually is.
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
396 American Scientist
There is a story of how this Doppler effect in optics almost
saved the famous American physicist, R. W. Wood, from paying the
usual fine for crossing an intersection on a red light. The story
goes that, being sum moned to the traffic court with the violation
ticket, Professor Wood gave a brilliant (as usual) lecture to the
judge on the subject of the Doppler effect, explaining how and why
one can see a red light as green if one drives towards it. But,
while the judge was highly impressed by that presentation, and was
ready to waive the fine, one of Wood's students (recently flunked
by him on an optics examination) happened to be in the courtroom
and proposed that the judge ask the professor to estimate the
velocity with which he must have been driving in order to see the
red light as green. As a result, the fine was changed from that for
crossing on a red light to that for exceeding the speed limit of
the city of Balti
more.
Hubble's measurements of the red shift in distant galaxies
indicate that they all move away from us with speeds proportional
to their distances. It does not mean, however, that we actually are
in the center of the universe with all its parts running away from
us, and can, in fact, be interpreted simply as an optical illusion
common to any observer located anywhere within a uniformly
expanding system. If we imagine an inflated rubber balloon with
black dots painted all over its surface in a polka-dot fashion (the
galaxies scattered through the space of the universe), an observer
sitting on any one of these dots will see all other dots receding
from him when the balloon is gradually swelling to a larger and
larger size. And the observed recession-velocity of more distant
dots will be larger in proportion to their distances.
The observationally established expansion of the universe gives
us a valuable clue to the history of the universe, indicating that
all present features of the universe must have originated as the
result of successive differentiation of a rapidly expanding
primordial matter. The date of the "beginning/' that is, the epoch
when the material forming the universe was in the original highly
compressed homogeneous state, can be obtained by a simple division
of the average distance between the neighboring galaxies by the
measured velocity of their mutual recession. The result, 1.8
billion years, is of the same order of magnitude as all other
approximate estimates of the age of the universe.
However, Hubble's exact figure of 1.8 billion years stands in
sharp contradiction with Holmes' figure of 3.35 billion years (the
discrepancy being far outside the claimed accuracy of both
methods), and at the present stage it is difficult to say which of
the two figures should be con sidered to be correct. It may be that
the geological estimate is to be changed, since Holmes'
calculations are based on certain specific assump tions concerning
the process of ore formation. On the other hand, there are several
ways in which the observed recession velocities of distant galaxies
could be fitted into a longer time scale. One of these
possibilities involves the introduction of the so-called
cosmological term, which corresponds physically to the assumption
of a repulsive force acting between individual galaxies and
increasing with their distance. The
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
The Origin and Evolution of the Universe 397
presence of such a repulsion would turn the expansion into an
accelerated process, and thus move the zero point farther back in
time. Another possibility was considered recently by Bondi, Gold,
and Hoyle who suggest that, while the space of the universe is
being gradually thinned out by the expansion, new matter is
continuously being created between the receding galaxies at a rate
which compensates for the effect of the recession. Thus, while
older galaxies get farther and farther from each other, new
galaxies are being formed in between to take their place, and the
show is going on without any beginning or ending. Although such a
hypothesis may be quite attractive from the philosophical point of
view, it encounters serious observational as well as theoretical
diffi culties, and should be taken at present with a good-sized
grain of salt.
At the present writing it seems that the discrepancy between
geologi cal and astronomical age-estimates can be removed by using
better data for intergalactic distances. In fact, a recently
published work by A. Behr indicates that, introducing various
corrections into the older estimates of intergalactic distances,
one may almost double the original Hubble figure for the duration
of the expansion process.
The Formation of Atomic Species When we inquire about the early
stages in the history of the universe*
we find that the most valuable archaeological document is
presented by the relative abundance with which different atomic
species are found in nature. In fact, there is every reason to
believe that chemical elements were "cooked" very early in history
when the density and temperature of the matter in the universe were
both exceedingly high. If we imagine history running back in time,
we inevitably come to that epoch of "big squeeze" with all the
galaxies, stars, atoms, and atomic nuclei squeezed, so to speak, to
a pulp.
During that early stage of evolution, matter must have been dis
sociated into its elementary compounds: protons, neutrons, and elec
trons. We call this primordial mixture ylem since in Webster's
Dictionary this word is explained as: "O.F. Hem, fr. L. hylem, acc.
of hyle. See Hyle. The first substance from which the elements were
supposed to be formed. Cf. Hyle, 1. O?s." While the temperature of
ylem was still very high (above one billion degrees centigrade)
thermal motion of the particles
was too violent to permit their sticking together. This high
temperature also prevented neutrons from decaying into protons and
electrons, or, to state it more correctly, the production of fresh
neutrons in the processes of proton-electron collisions at that
time was compensating for their loss due to the decay process.
However, as soon as the density and temperature of matter
dropped as the result of the progressing expansion, two processes
must have started. The first process was the predominating neutron
decay which was cutting sharply into the number of neutrons
available for the nuclear reaction. The second was the aggregation
of neutrons and protons into complex groups : the prototypes of the
atomic nuclei of today. The result of the competition between these
two processes must have determined
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
398 American Scientist
the relative numbers of various composite nuclei which exist in
nature at the present time. If the expansion had been too fast or
the original density of matter too low, very few nuclear collisions
could have taken
place before all neutrons were destroyed (turned into protons)
by the natural decay process. In this case, practically no complex
nuclei would have been built, and the matter of the universe today
would consist
predominantly of hydrogen. If, on the contrary, the original
density had been too high, neutrons and protons would have had
ample chance to unite into complex units, and most of the material
of the universe would be present now in the form of heavier
elements.
Apparently, the actual situation was somewhere between these two
extremes, and we should be able to get rather exact information
con
cerning the physical conditions which prevailed during the early
stages of the expansion of the universe, by analyzing in detail the
processes of nuclear formation which took place during that epoch.
It must be re membered that, even though these processes occurred
billions of years ago, we can discuss them on the basis of
perfectly reliable nuclear in formation. In fact, the temperature
of a billion degrees corresponds to thermal energies of the order
of one million volts, and these are exactly the energies at which
nuclear reactions are being studied in our labora tories using
electrically accelerated nuclear beams.
The first attempt to calculate what must have happened to ylem
dur ing the early stages of the expanding universe was made by the
author and a former student, R. Alpher, several years ago.1 The
problem pre sented by the building-up process of atomic nuclei is
very similar to the classical problem of heat flowing along a bar
heated at one end. In the latter problem the increase of
temperature in any section of the bar is given by the difference
between the heat inflow from the heated side and the heat outflow
in the opposite direction. Similarly, the increase in the number of
representatives of any given nuclear species, say the nuclei with
atomic weight 100, is given by the difference between the rate of
their production through neutron capture in 99-weight and their
elimina tion as the result of moving into 101-weight through a
subsequent neutron capture.
One can write simple differential equations containing the known
neutron-capture cross sections, the solution of which will give us
the ex
pected distribution of the original material between different
atomic weights, for any given original density and any given time
of cooking. Since, as was stated above, the expansion process must
have started
spontaneous decay of neutrons, the entire "cooking period" could
not
1 When the preliminary communication concerning these
calculations was written, and signed by two names, Alpher and
Gamow, we felt that something was missing. Thus, in accordance with
the Greek alphabet, we have added the name of Bethe (in absentia),
which resulted in the theory's being often referred to as the a?y
theory. Dr. Bethe, who reviewed a copy of the article sent to
Physical Review, did not object and, in fact, was quite helpful in
further discussions. There was, however, a rumor that at the later
date when the theory went temporarily on the rocks, Bethe was
con
sidering changing his name to Zacharias. It may also be noticed,
with chagrin, that Dr. R. Herman, who joined the team later, still
refuses to change his name to Delter, which would be most
appropriate.
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
The Origin and Evolution of the Universe 399
have lasted much longer than the mean lifetime of a free neutron
(the order of magnitude of half an hour). It may look silly to talk
about the consequences of a process which took place a few billion
years ago and lasted for only half an hour, but the ratio of half
an hour to a few billion
Neutron capture cross section (in barns)
O 50 IOO I50 200 250
ATOMIC WEIGHT Fig. 1. Observed and calculated abundance
curves.
years is about the same as the ratio of a few microseconds to
several years, which represent respectively the reaction time
within an explod ing atomic bomb and the period of time after which
the radioactivity of fission products can still be noticed at the
explosion site!
The results obtained by the integration of the building-up
equations
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
400 American Scientist
for different original densities of y lem are shown graphically
in Figure 1, along with the empirical abundance curve (shaded band)
of different atomic species. We see that the best fit is obtained
on the assumption that in the beginning of the building-up process
(five minutes after the stage of maximum compression) the density
of matter in the universe was in the vicinity of one microgram per
cubic centimeter, that is, about one-tenth of 1 per cent of the
density of atmospheric air. This is quite a
high density considering that today the mean density of the
universe is about one atom per cubic meter or 10~30 g./cm3. Without
going into the details of the theory, we may indicate that the
characteristic shape of the empirical abundance curve, which shows
a steep slope for the ele ments of the first half of the periodic
system, and a horizontal run for all heavier elements, is a direct
consequence of corresponding behavior of
Helium Tritium ^Tralphium
v Deuterium
Hydrogen
Neutrons
1 -1-1-1-1-1?
5 io 15 2 0 25 30 35 minutes from zero point
Fig. 2. Building of lightest elements. CAfter Fermi and
Turkevich.)
neutron-capture cross sections which are shown in the inset
diagram in the upper-right corner of Figure 1.
One should expect, in fact, that the abundance curve must level
off when the capture cross sections become rather large, in the
same way that the temperature distribution along a heated bar
levels off at its far end if that part of the bar is made from a
material possessing much higher heat conductivity than the part
immediately adjacent to the source of heat.
The above-described calculations, involving the building-up
process of all atomic species from the beginning to the end of the
periodic system, are, by necessity, rather approximate. In fact, in
order to be able to carry through the calculations in a finite
period of time, we had to assume the "smoothed out" curve for
capture cross sections, and to make several other simplifying
assumptions. Another possible way of
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
The Origin and Evolution of the Universe 401
obtaining the density of the matter in the universe during the
"nuclear cooking" process would be to study exclusively the
reactions between the few simplest elements, but to make this study
in great detail using all the available data of nuclear physics.
Calculations of this type have been worked out by the author for
the process of deuterium formation in the reaction n+p-*d+y, and
later by Fermi and Turkevich for all reactions involving neutrons,
protons, deuterone, tritons, tralpha (He3), and alpha-(He4)
particles. The results of the later authors, who used the same
physical assumptions concerning the densities and temperatures as
used in the previous calculations, are shown graphically in Figure
2. We notice that, whereas the decay of neutrons into protons
begins almost immediately after the start of expansion, the
aggregation process leading to the formation of complex nuclei
starts only five minutes later when the temperature of ylem has
dropped below one billion degrees. By the end of half an hour most
of the original neutrons have decayed, whereas hydrogen and helium
are forming about half and half of the entire mixture.
Deuterium is present in the amount of a few per cent, and will
be completely exhausted in the process of building heavier
elements. (As stated above, these further building-up processes
were disregarded in the calculations.) Thus we see that, using this
detailed method, we also arrive at a reasonably good result, since
it is known that at present the matter in the universe consists of
somewhat more than 50 per cent of
hydrogen, somewhat less than 50 per cent of helium, and about 1
per cent of all heavier elements. It is still too early to state
that the above described theory of the origin of chemical elements
provides a complete explanation of all observed facts. Indeed,
there still remain some serious difficulties involving notably the
problem of carrying the building-up process across the atomic
weight five (because of the absence of any stable nuclei of that
weight), and the explanation of the so-called "shielded isotopes"
(that is, the isotopes which cannot be obtained directly through
the beta-decay of the nuclei with the excess of neu
trons). It seems, however, that the theory gives a reasonably
consistent picture of what must have happened during the very early
stages of the evolutionary history of our universe.
The First Thirty Million Years, and the Beginning of the
Differentiation Process
After atomic species were formed in the first half-hour or so of
the
history of the universe, the expansion of newly formed matter
continued in a rather monotonous way for quite a long time. The
most characteristic feature of this entire period was the
prevailing role of radiant energy as
compared with ordinary matter. It is well known that, according
to Einstein's law, radiant energy possesses ponderable mass, the
value of which can be obtained numerically by dividing the amount
of energy, expressed in ergs, by the square of the velocity of
light. Using this rule, we can easily find, for example, that the
light which is filling a lecture room weighs only a few billionths
of a microgram?about the weight
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
402 American Scientist
of one bacterium! This is, of course, a negligibly small figure
as com
pared with the weight of the air in the same lecture room.
If, however, we make a similar calculation for radiant energy
and ordinary matter during the early stages of the expansion of the
universe, we shall arrive at a rather different result. Assuming
that at the end of five minutes the temperature of the universe was
about one billion de grees (as it follows from the previous
discussion), we find that the mass
density of radiation (aTA/c2) was about 10 g./cm.3, thus being
com
parable with the density of iron! Since, at the same time, the
density of
ordinary (atomic) matter was only about one microgram per cubic
centimeter, we conclude that at that epoch the situation was ruled
ex
clusively by light (with very short wave length, of course) and
that the material particles were helplessly thrown around like
little chips of wood in the stormy ocean of radiation.
As the expansion of the universe proceeded, the situation was
gradually changing in favor of matter. Indeed, whereas the total
number of atoms was left unchanged, radiant energy was being spent
doing the work of expansion. It can be calculated that the density
of matter and radiation became equalized approximately at the age
of about 30,000,000 years, when the temperature of the universe
dropped to about 300? Kelvin (roughly room temperature), and its
material density to the value of about 10~24 g./cm.3 (one H-atom
per cm.3, or the present density within the galaxy).
At that point in history, matter took over the leading role in
further developments, and its first deed was to break up the
homogeneity of the hitherto continuous expansion process. The chief
agent in this break ing-up process, which ultimately led to the
present highly differentiated state of the material universe, was
Newtonian gravitation between ma terial particles. In fact, as it
was once shown by the British astronomer, Sir James Jeans, a
gravitating gas filling uniformly an unlimited space is
intrinsically unstable, and is bound to break up into separate "gas
balls"
with completely empty space in between. The size of the
condensations resulting from this so-called gravitational
instability is determined by the condition that each "gas ball"
should be sufficiently large so that the escape velocity from its
surface is larger than the thermal velocity of gas particles. Using
the initial temperature and density of the pri mordial gas, and
remembering that the temperature in the expanding universe falls
off in inverse proportion to the square root of its age, whereas
the material density falls inversely to the three-half power of it,
one can estimate the size and mass of "gas balls" which must have
been formed during various epochs of the expansion of the universe.
In doing this, one finds that the mass of the condensation always
comes out the same no matter when they are formed, since the
time-dependence cancels out in the mass formula.
Numerically, one obtains for the mass of the condensation the
value of about 108 sun masses, which, though being on the low side,
corresponds in order of magnitude to the average mass of galaxies.
This is a very gratifying result indeed, since that mass is
obtained exclusively by using
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
The Origin and Evolution of the Universe 403
nuclear constants such as neutron-capture cross sections, etc.
As was
mentioned above, no such condensations of gas could have taken
place prior to the age of 30,000,000 years, that is, so long as
radiant energy ruled the situation. As soon as matter took over,
however, the break-up process must have taken place, so that the
mean density of individual "gas balls" must equal the mean density
of the universe at that time.
This conclusion also agrees with observational evidence, since
the present mean density of matter inside the individual galaxies
is the same (10-24 g./cm.2) as the above-quoted mean density of the
universe at the moment of galactic separation. When, at the date
mark of 30,000,000 years, the originally homo
geneous gas broke up into separate clouds (the progenies of
today's galaxies), the space of the universe was quite dark since
the original brilliance of the first days of creation was already
dimmed out by ex
pansion, and the stars, which illuminate the universe today,
were not yet formed. There was nothing at that time but giant
clouds of luke warm gas which were being pulled away from each
other by the pro gressing expansion of space. It goes without
saying that the break-up of the expanding gas into separate clouds,
or fragments, must have re sulted in a rather rapid rotation of
these fragments around their axes distributed at random in all
directions. We observe the same type of rotation in the fragments
of an artillery shell exploded in mid-air. Here probably lies the
explanation of the fact that most gnlaxies are found now in the
state of rotation, manifested in their flattened elliptical shapes
and in their spiral arms winding around their centrally con densed
bodies.
Stars, Planets, Satellites
The next step in the evolution of the universe apparently was
the formation of stars, which must have originated as the result of
the secondary condensation, that is, the break-up of the original
"galactic gas balls" into billions of smaller "stellar gas balls"
by the same old process of gravitational instability. These smaller
gas condensations contracted quite rapidly, and, as the result of
compression, the material in their central regions was heated to
the temperature of some 20,000,000 degrees, representing the
threshold for nuclear reactions. The liberation of nuclear energy
had started, and the universe became illuminated by billions and
billions of stars.
Space does not permit us to consider here the detailed analysis
of stellar evolution, and, in particular, the problem of the origin
of plane tary systems. We shall mention only that, according to
recent theories of the German physicist, C. von Weizs?cker, and the
American as
tronomer, G. P. Kuiper, the formation of the planets took place
in a way very similar to that proposed centuries ago by Kant and
Laplace (the collision theory of Jeans, and Chamberlin and Moulton,
being abandoned
by modern cosmogony). Since, as was mentioned above, the
material
forming the original galactic gas balls was in a state of rapid
turbulent
rotation, stellar condensations (or pre-stars) were rotating
too. Thus,
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
404 American Sdentisi
whereas most of their material must have fallen towards the
center, forming the main body of the star, some of it must have
been left outside in the form of a strongly flattened or rotating
disk. About 99 per cent of this disk was gaseous hydrogen and
helium, whereas the remaining 1 per cent was formed by small dust
particles of silicates, iron oxides, ice crystals, etc.
The dust particles of that swarm must have been always colliding
with one another (forming the units of larger and larger mass
"plane tesimals" of the Chamberlin and Moulton theory). Those
chunks of the material which happened to grow larger than the
others swept the space around them, capturing the smaller stones
and dust particles, until they found themselves moving in
practically empty space.
Mathematical analysis of this rather complicated process not
only gives us a reasonably complete picture of planetary formation,
but also leads to the understanding of various observed
regularities such as the
Bode law which governs the distance of various planets from the
sun. It goes without saying that this "rejuvenated Kant-Laplace
theory" predicts the existence of planetary systems around
practically any star within our own or any other galaxy. Recently
this theoretical conclusion found confirmation in the actual
discovery of planetary systems near two close stars. According to
the theory, the entire process of the condensa tion of stars and
the formation of planetary systems must have taken a few hundred
million years.
There is one more important point to be mentioned in connection
with the theory of planetary origin. As we have seen above, planets
were formed by the accumulation of solid dust particles which
were
floating in a hydrogen-helium, gaseous mixture. Such a process
would produce rocky bodies similar to our earth, to Mars, and to
two internal planets : Venus and Mercury. However, if the mass of a
planet exceeded certain limits (a few earth masses), it would
possess a sufficiently strong gravitational field to capture and
hold quite large amounts of interstellar hydrogen and helium gases.
Neither our earth nor the three other inner planets ever exceeded
that limit, and so they have remained rocky bodies as we know them
now. On the other hand, the original rocky bodies of outer planets,
such as Jupiter and Saturn, managed to grow above that limit
(because the original dust disk was thicker at these distances),
and thus have acquired a lot of interstellar gaseous
material.
It was recently shown by H. Brown of Chicago that only about 2
per cent of the giant bodies of Jupiter and Saturn is made from the
same
material as our earth. This material, forming the rocky cores of
these planets, is covered by layers of frozen water, methane, and
ammonia, which account for another 8 per cent. The rest of Jupiter
and Saturn is nothing but highly compressed mixtures of gaseous
hydrogen and helium. Thus, if Flash Gordon of the Sunday comic
strips, or a serious rocket explorer of the future, were to land on
the seemingly solid surface of these planets, he would sink deeper
and deeper into the compressed gas, and would finally be crushed by
the tremendously high pressures
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
The Origin and Evolution of the Universe 405
lying deep in the body of these planets at the surfaces of their
inner solid cores.
The formation of satellites took place in a way completely
similar to planet formation, by the condensation of dust particles
from the flat tened rotating envelopes which surround the
proto-planets. The only possible exception is presented by our own
moon, which stands out of the company of other satellites because
of its exceptionally large rela tive mass. It is believed that the
earth was originally born without any satellites (as were Venus and
Mercury), and that it was later broken into two pieces (larger one,
the earth; smaller one, the moon) by the tidal forces of the sun.
In fact, a British astronomer, George Darwin, was able to show that
in the distant past, the moon was much closer to the earth, and
that several billion years ago, the earth and the moon
must have comprised one single body. We may note here again that
the calculated date of the moon's birth fits well with other values
quoted for the age of the universe.
The history of the universe as described above is shown
schematically in the Frontispiece (page 392).
A Glance into the Future
Having learned that the universe, as we know it today, must have
originated a few billion years ago from hot homogeneous ylem which
was successively differentiated in the process of expansion of the
uni verse, we may naturally ask: What lies ahead of us? Will the
universe continue its present expansion beyond any limit, or will
it stop and start collapsing back on us (or rather on our
descendants) ? This question can be answered in a simple way by
comparing the kinetic energy of galaxies flying away, with the
potential energy of the Newtonian attraction be tween them. Using
the available data, one can easily find that the kinetic energy of
galactic recession is almost a hundred times as large as their
mutual gravitational energy. Thus, the situation is similar to the
case of a rocket fired from the surface of the earth with ten times
the escape velocity (one hundred times the escape energy). The
galaxies will fly apart forever without ever turning back. Within
each galaxy the process of stellar evolution will be continuing,
and the stars, which draw their death ticket by using up all their
hydrogen fuel, will be ex
ploding and going into oblivion. Some 47 billion years from now,
that fate will reach our own sun, and still later the other fainter
stars. But all this is still so far away that it is hardly cause
for anxiety.
Another question we could ask pertains to the forces which
caused the initial expansion of the universe, and to the state of
affairs which
must have existed prior to the maximum stage of contraction
which was the starting point of all our discussion. Mathematically
we may say that the observed expansion of the universe is nothing
but the bouncing back which resulted from a collapse prior to the
zero point of time a few billion years ago. Physically, however,
there is no sense in speaking about that "prehistoric state" of the
universe, since indeed during the
stage of maximum compression everything was squeezed into the
pulp,
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
-
406 American Scientist
or rather into ylem, and no information could have been left
from the earlier time if there ever was one.
This conclusion is in complete agreement with the statement made
centuries ago by St. Augustine of Hippo who, in one of his
writings, was trying to answer the question of what God was doing
before He made heaven and earth. "He was making the hell," wrote
St. Augustine, "for the persons who ask that kind of
questions."
REFERENCES The A ge of the Universe
1. Holmes, Arthur. The age of the earth. Endeavour (London),
July 1947. 2. Hubble, Edwin. The problem of the expanding universe.
In: Science in
progress, third series. Yale University Press, 1942. 3.
LeMa?tre, George. The primeval atom. D. Van Nostrand Co., 1950. 4.
Hoyle, Fred. The nature of the universe. Harper & Brothers,
1951. 5. Behr, Alfred. Zur Entiernungsskola der Extragalactischen
Nebel. Astr.
Nachrichten., 279, 97, 1951.
The Formation of Atomic Species
6. Alpher, Bethe, and Gamow. The origin of chemical elements.
Phys. Rev., 78, 803,1948.
7. Alpher, Ralph, and Herman, Robeft. Theory of the origin and
relative distribution of the elements. Rev. Mod. Phys., 22, 153,
1950. (This paper con tains complete literature on the
subject.)
The First Thirty Million Years
8. Jeans, Sir James. Astronomy and cosmogony. Cambridge
University Press, 1928.
9. Gamow, George. The evolution of the universe. Nature, 162,
680, 1948.
Stars, Planets, Satellites
10. Weizs?cker, Carl von. ?ber die Entstehung der
Planetensystems. Zeit, f?r Astroph., 22, 319,1944.
11. Kuiper, Gerard. On the origin of the solar system. In:
Astrophysics: A topical symposium. McGraw-Hill Book Co., 1951.
12. Brown, Harrison. The composition of our universe. Physics
Today, April 1950.
This content downloaded from 163.178.101.228 on Wed, 01 Apr 2015
19:58:41 UTCAll use subject to JSTOR Terms and Conditions
Article Contentsp. [392]p. 393p. 394p. 395p. 396p. 397p. 398p.
399p. 400p. 401p. 402p. 403p. 404p. 405p. 406
Issue Table of ContentsAmerican Scientist, Vol. 39, No. 3 (JULY
1951) pp. 361-520Front MatterEDITORIAL MISCELLANY [pp. 366, 368,
370, 372, 374, 376, 378, 380, 382, 384, 386, 388, 390, 513-516,
518-519]THE ORIGIN AND EVOLUTION OF THE UNIVERSE [pp. 392-406]THE
RELATIONS BETWEEN ASTRONOMY AND GEOPHYSICS [pp. 407-411]THE
MEANINGS OF TIME AND SPACE IN PHILOSOPHIES OF SCIENCE [pp.
412-421]THE MEANING OF "ELEMENTARY PARTICLE" [pp. 422-431]ON THE
PSYCHOLOGICAL ASPECTS OF AUTHORITARIAN AND DEMOCRATIC POLITICAL
SYSTEMS [pp. 432-440, 451]ON THE BIOLOGICAL BASIS OF ADAPTEDNESS
[pp. 441-451]FISHER AND FORD ON "THE SEWALL WRIGHT EFFECT" [pp.
452-458, 479]COMMUNICATIONSTHE CITIZEN AND THE HISTORY OF SCIENCE
[pp. 459-461]SAMPLING THE UNIVERSE [pp. 462-465]LETTERS FROM A
MEXICAN EXPEDITION [pp. 466-471]WILLIAM PROCTER, 18721951 [pp.
471-472]
MARGINALIA [pp. 473-479]THE SCIENTIST'S BOOKSHELF [pp. 480-484,
486, 488, 490, 492, 494, 496, 498, 500, 502, 504, 506, 508,
510-513]Back Matter