Frank H. Shu
THE PHYSICAL
UNIVERSE:
AN INTRODUCTION
TO
ASTRONOMY
THE PHYSICAL UNIVERSE: AN INTRODUCTION TO ASTRONOMY Frank H. Shu
Professor of Astronomy University of California, Berkely
To my parents
BERKELY FRANK H. SHU PROFESSOR OF ASTRONOMY, UNIVERSITY OF CALIFORNIA,
THE PHYSICAL UNIVERSE AN INTRODUCTION TO ASTRONOMY
PART 1:
BASIC PRINCIPLES
PAGE.3 1.THE BIRTH OF SCIENCE THE CONSTELLATIONS AS NAVIGATIONAL AIDS................3 THE CONSTELLATIONS AS TIMEKEEPING AIDS................4 THE RISE OF ASTROLOGY...........9 THE RISE OF ASTRONOMY...........9 MODERN ASTRONOMY..10 ROUGH SCALES OF THE ASTRONOMICAL UNIVERSE...........11 CONTENTS OF THE UNIVERSE...........12
PAGE.33 2.THE GREAT LAWS OF MICROSCOPIC PHYSICS MECHANICS.........33 QUANTUM MECHANICS.........16 SPECIAL RELATIVITY........20
PAGE.62 4.THE GREAT LAWS OF MACROSCOPIC PHYSICS
PAGE.14 3.CLASSCAL MECHANICS / LIGHT / ASTRONOMICAL TELESCOPES CLASSICAL MECHANICS..........14 THE NATURE OF LIGHT..............16 ASTRONOMICAL TELESCOPES.........20
THERMODYNAMICS....33 STATISTICAL MECHANICS.........64 THERMODYNAMIC BEHAVIOR OF MATTER............66 THERMODYNAMIC BEHAVIOR OF RADIATION.........77 AN EXAMPLE........80 PHILOSOPHICAL COMMENT...........80
PART 2: THE STARS
PAGE.81 5.THE SUN AS A STAR
PAGE.125
COPYRIGHT 1982
7.END STATES OF STARS
BY UNIVERSITY
WHITE DWARF.......126 NEUTRON STARS.....129 BLACK HOLES.......134 CONCLUDING PHILOSOPHIC REMARK.........143
Reproduction or translation of any part of this work beyond that permitted by Sections 107 or 108 of the 1976 United States Copyright Act without the permission of the copyright owner is unlawful. Requests for permission or further information should be addressed to the Permission Department, University Science Books, P.O. Box 605.
THE ATMOSPHERE OF THE SUN............84 THE INTERIOR OF THE SUN............86 THE CHROMOSPHERE OF AND CORONA OF THE SUN............96 THE RELATIONSHIP OF THE SUN TO OTHER STARS AND TO US...100
PAGE.102 6.NUCLEAR ENERGY / SYNTHESIS OF THE ELEMENTS MATTER AND THE FOUR FORCES............103 NUCLEAR FORCES AND NUCLEAR REACTIONS.........108 SPECULATION ABOUT THE FUTURE........123
PAGE.144 8.EVOLUTION OF THE STARS THEORETICAL H-R DIAGRAM...........145 EVOLUTION OF LOW-MASS STARS.............147 EVOLUTION OF HIGH MASS STARS........153 CONCLUDING PHILOSOPHIC REMARK.........157
SCIENCE BOOKS
LIBRARY OF CONGRESS CATALOG CARD NUMBER: 81-51271
ISBN 0-935702-05-9
Printed in the United States of America
PART. THREE
GALAXIES & COSMOLOGY
Like early explorers mapping the continents of our globe, astronomers are busy charting the spiral structure of our galaxy, the Milky Way. Using infrared images from the Spitzer Space Telescope, scientists have discovered that the Milky Way’s elegant spiral structure is dominated by just two arms wrapping off the ends of a central bar of stars. Previously, our galaxy was thought to possess four major arms. The annotated artist’s concept illustrates the new view of the Milky Way. The galaxy’s two major arms (Scutum-Centaurus and Perseus) can be seen attached to the ends of a thick central bar, while the two now-demoted minor arms (Norma and Sagittarius) are less distinct and located just between the major arms.
for ma
an,
p.211
THE MATERIAL BETWEEN THE STARS
p.212
p.213
p.214
p.215
p.216
p.217
p.218
p.219
p.220
p.221
p.222
p.223
p.224
B
etween the stars in the universe lies a vast amount of interstellar material in the form of both gas and dust. The interstellar gas and dust medium of our Galaxy probably has a massively mass of several billion M 0. Thus, the total mass of the material between the stars is a non-negligible fraction—only a few percent—of the cummulative mass of all the visible stars in the Galaxy. Yet the presence of interstellar matter, both gas and dust, is much less obvious than the presence of stars, since the mass contained in ordinary stars has been compacted by self-gravity into a readily observable dense state. In contrast, the interstellar gas and dust is spread very thinly over the vast distances between the stars; this diffused gas is much more rarefied than the best, so-called “vacuums” produced in terrestrial laboratories. Selfgravity plays a relatively minor role for this gas, and it should not even surprise us to learn that even astronomers were slow to realize the existence of interstellar gas and dust in the Galaxy.
The Discovery of Interstellar Dust More than 200 years ago William
or faulty data. The conclusive proof
Herschel described Tholes in the sky.-
for the presence of a general and se-
where there were an apparent deficit
lective absorption came in 1930, with
of stars (Figure 11.1). The most plau-
the epochal work of R. S. Trumpler on
sible explanation for this deficit was
the properties of star clusters.
the existence of obscuring material
We
may
reproduce Trumpler’s
along the line of sight which blocked
reasoning as follows. Let D be the
the light from the back-ground stars.
real (linear) diameter of an open
Claims were occasionally made for
cluster and r equal its distance. Since
the definitive discovery of other ob-
D should be independent of the open
servational evidence for this absorp-
cluster’s distance r, we expect the
tion, but careful follow-up analyses
angular diameter of an open cluster
invariably showed these claims to be
to be 0 = Direction.so as to keep it de-
based on either incorrect reasoning
creaseing with its increasing distance
THE DISCOVERY OF INTERSTELLAR DUST
and there is also scattering about the
p.211
p.212
theoretical line due to intrinsic variations and systematic departure. Thus, we may expect that the square root of its angular diameter,
p.213
d = (D/r�)
p.214
p.215 to vary on the average as the inverse square of the cluster’s distance r from us. The apparent brightness f L,/47(r2 of (the main sequence stars
p.216
of) an open cluster should also vary as the inverse square of its distance from us, because, on the average, the
systematic departure of the observed
intrinsic brightness (luminosity) L of a
points from the theoretical line. This
typical cluster should be independent
might be explained in several ways.
of its position relative to the Earth.
(a) Perhaps far-away clusters look
Thus, if apparent brightnessos of
intrinsically bigger than their mea-
open clusters arc plotted ‘Tains t an-
sured apparent brightnesses f would
gfIlp r diameters squared, we should
predict. If this were a real effect, then
expect theoretically to see a straight-
our position in the universe would be
line relation (Figure 11.2a). The actual
special, since open clusters which are
result is shown schematically in
close to the Earth would then he in-
Figure 11.2b. The inter-pretation of this
trinsically small, and those which are
result is the following.
far away would be intrinsically large.
First, there is scatter about the the-
Such a special location for Earth has
oretical line, because there must he
been anathema for astronomers since
intrinsic variations of the luminosities
Copernicus, and this interpretation
L and diameters D of open clusters
may therefore be rejected.
about some mean values. (That is,
(b) Could this effect be one of ob-
some open clusters are intrinsically
servational selec-tion? That is, could
more less luminous than average and
this effect arise because observers
some are intrinsically bigger or small-
find it easier to see intrinsically big
er than average.) Second, there is a
open clusters than intrinsically small
p.217
p.218
p.219
p.220
p.221
ones if they are far away? Trumpler was well aware of such observational
p.222
biases, and he was able to show convincingly that his data set did not con-tain errors of this kind. (c) Perhaps far-away clusters look fainter than their measured squares of
p.223
p.224
THE MATERIAL BETWEEN THE STARS
p.211
p.212
angular diameters 92 would predict. The Copernican viewpoint prevents us from thinking that this could be a
p.213
real effect. (d) The remaining possibility is, then, that far-away clusters have been dimmed by a general obscuration of starlight which increases with in-
p.214
creasing distance, as more and more obscuration occurs along the line of sight. This interpretation is the one accepted today, and it was the one
p.215
given by Robert Julius Trumpler. Astronomers now know that this obscuring material is in the form of small solid specks, dust grains whose
p.216
chemical composition may be silicates (like sand) or carbon-containing compounds (like graphite or silicon carbide). The obscuration of starlight
p.217
is believed to arise from a. combination of true absorption and scattering, and this combination goes by the general name of star.
p.218
Interstellar reddening occurs because blue light is scattered out of a beam of starlight directed toward us mere than red light is. A similar effect occurs in the Earth’s atmosphere.
p.219
because of the scattering of sunlight by
p.220
p.221
p.222
atmospheric
molecules.
The
light from the setting Sun (or rising
plain interstellar reddening, because
Sun) has to travel through a greater
these processes are not exercised in
column of air and undergoes more
the operation throughout the seper-
scattering and more reddening than
ating sequences. The obscuration of
light from the noon Sun (Figure 11.4).
starlight is also believed to arise from
On a microscopic level (Chapter 3),
a. combination of true absorption and
we understand this result to arise
scattering, and this combination goes
because air molecules interact more
by the general name of every stars.
strongly with blue light than with red
In other words, whereas the wave-
light and, therefore, scatter blue light
lengths of the absorption lines from
preferentially out of the sunbeams,
the photospheres of the stars in the
leaving mostly reddened sunlight.
binary system shifted back and forth
Exactly the same process cannot ex-
as the stars revolved in their orbits, there were also absorption lines of the
p.223
p.224
stars whose wavelengths remained stationary and does not change.
THE DISCOVERY OF INTERSTELLAR GAS
p.211
fig. 11.1
p.212
The Coalsack nebula in the southern Milky Way gives the impression of a large region in the sky where there is a marked deficit of stars. In fact, the apparent “hole in the sky” is eplained by the obscuration of the light of many background stars by and intervening cloud of dark material.
p.213
p.214
The Discovery of Interstellar Gas Historically, the conclusive evi-
interstellar lines there-fore appear
dence for the existence of interstellar
“stationary” relative to the changing
gas was presented somewhat earlier
pattern of lines presented by the
than the evidence for interstellar dust.
spectroscopic binary.
In 1904, Hartmann dis-covered that a
Hartmann’s
hypothesis
of
an
set of absorption lines of once-ionized
intervening gas cloud did not gain
calcium did not undergo the periodic
immediate acceptance. because the
Doppler shifts of the absorption lines
ab-sorbing gas might have resided
in the spectrum of a spectroscopic
near the star in question rather than
binary. In other words, whereas the
a long distance away. This issue of
wavelengths of the absorption lines
whether the absorbing gas is “cir-
from the photospheres of the stars
cumstellar- or “interstellar-remains
in the binary system shifted back and
difficult to resolve in any one ease
forth as the stars revolved in their
even today. The “interstellar” inter-
orbits, there were also absorption
pretation was demonstrated by Plas-
lines whose wavelengths remained
kett, Struve, Eddington, and Bok. who
stationary. Hartmann con-cluded that
showed that the ionization stages, or
the “stationary lines” arose from ab-
Galactic distribution, or veloci-ties
sorption produced by a cold interstel-
of the “stationary lines” were often
lar cloud of gas which lay between the
incompatible.
binary system and Earth (Figure 11.5).
Further insight into the interstellar
There may be also a fixed displace-
gas was provided by Beats, Adams,
ment of the inter-stellar absorption
Munch, and Zirin, who found that
line from the position of a similar line
many stars show multiple interstellar
produced in a terrestrial laboratory,
absorption lines, i.e., the same lines,
because the cloud may have a com-
say, of once-ionized calcium at sev-
ponent of velocity lin along the line
eral different Doppler velocities and
of sight. However, the latter velocity
inter-stellar lines, or an approximate
does not vary on the timescale which
a spherical dust grain of radius R to
characterizes the orbital motion of
have a cross-sectional area 7R’ for
the background binary system. The
blocking visual light.
p.215
p.216
p.217
p.218
p.219
p.220
p.221
p.222
p.223
p.224
p.211
THE MATERIAL BETWEEN THE STARS
p.212
p.213
p.214
p.215
p.216
p.217
p.218
p.219
p.220
Different Optical Manifestations of Gaseous Nebulae We now know that the dust and the
ble to or smaller than the wavelength
gas in interstellar space are intimately
of visual light, but it is valid enough for
mixed, and that both generally reside
our purposes here.) A photon mean-
in clouds or complexes of clouds.
free-path I is defined to be the length
The ratio of dust mass to gas mass in
between successive encounters with
the Galaxy is about I percent. Since
dust grains (contrast with Problem
the gas itself is only a few percent of
9.11). If the number density of dust
the mass of the stars in the Galaxy,
grains is n, show that I = 1innR2. If a
interstellar dust constitutes a very
beam of photons from a star travels
minor fraction of the total mass of
a distance i toward an observer, the
our Galaxy. Nevertheless, this dust
beam will suffer an extinction in
has a great influence on how we see
intensity by a factor of e = 2.718 .... Ex-
the Galaxy, because it obscures so
tinction observations indicate that at
much starlight in many directions.
the position of the Sun in the Galaxy. I
Problem 11.1. A rough estimate of the
equals about 3.000 light-years. Given
mass fraction of dust in the Galaxy
the estimate R = 10 5 cm, calculate the
proceeds as follows. Approximate
value of n in units of cm’. How many
a spherical dust grain of radius R to
interstellar dust grains would you
have a cross-sectional area 7R’ for
expect to find in a volume equal to a
blocking visual light. (This “shadow”
football stadium? (Assume a typical
formula breaks down if R is compara-
dimension of 100 meters in all three directions.) Solid material typically has a mass density of 2 gm/cm’. Estimate now the mass ni of a typical inter-stellar dust grain. Let V be the
p.221
p.222
p.223
p.224
DIFFERENT OPTICAL MANISFETATIONS OF GASEOUS NEBULAE
p.211
p.212
p.213
p.214
TIME BOX
1.1
p.215
p.216
p.217
p.218
p.219
A page full of surprises Fun things you’ll want to know Interstellar dust?
p.220
p.221
p.222
p.223
p.224
THE MATERIAL BETWEEN THE STARS
p.211
fig. 11.2
p.212
volume in which one typically finds
p.213
one solar mass of stars in the Galaxy.
The dark gobules lie in front of the extended background of light from the Rosette Nebula.
If V = 300 (It-yr)3, what is the mass M = nVm of dust grains in this same volume? What, then, is the mass frac-
p.214
tion of dust compared to stars at the solar position in the Galaxy? (In actual practice, the average mass fraction is even smaller, because the dust is confined to a layer in the Galaxy whichis
p.215
about three times thinner than the stars.) The appearance of clouds of gas and dust depends, in part, on the
p.216
p.217
p.218
p.219
p.220
p.221
p.222
p.223
p.224
wavelength regime in which they are
Crab nebula, an amorphous region
ob-served and, in part, on how close
which emits continuum light by the
they are to neighboring stars. We first
synchrotron process may coexist with
explore the different optical appear-
the filamentary structure. The radio
ance of interstellar gas clouds. Such
emission from supernova remnants is
gas clouds are also generally called
invariably nonthermal, but the X-ray
gaseous nebulae (See Box 11.2). BOX
emission and optical-line emission
11.2 Optical Classification of Gaseous
may arise from thermal pro-cesses
Nebulae Dark nebulae: observed by
in a shock-heated gas. Dark Nebulae
the obscuration of background stars
A gas and dust cloud placed in a rich
or some other background wIlik.11
field of back-ground stars would block
is otherwise bright (such as an Ha
most or all of the starlight behind it.
region). Reflection nebulae: observed
We would easily see many stars to the
by the scattered light from embedded
sides of the dust cloud (and a few in
stars. The spectrum is the (reflected)
front of it); hence such a dark nebulae
absorption-line
the
would manifest itself as one of Her-
embedded stars. His regions: bright
schel’s “holes in the sky” (Figure 11.6).
ionized regions surrounding newborn
Especially interesting among the dark
hot and bright stars (of spectral type
nebulae are the round ones studied
0 and B). The spectrum is dominated
by Barnard and by Bok. The regular
by emission lilies. Thermal radio-con-
shapes of the “Bok globules” suggest
tinuum emission is found. Planetary
that these objects are self-gravitating,
nebulae: similar to Hu regions, but
and led Bok to propose that they are
the exciting object is a very hot
probably sites where new stars are
evolved star in the throes of death.
forming. This suggestion is almost
Planetary nebulae also tend to be
certainly correct for the dark clouds
denser and more compact than opti-
in giant complexes, as we will discuss
cal Hit regions. Supernova remnants:
later. Whether isolated Bok globules
optical emission usually strongest
are :ilso collapsing to form stars is be-
from filaments, whose spectra are
ing debated now. Reflection Nebulae
domi-nated by emission lines. In a
A gas and dust cloud which surrounds
young supernova remnant like the
a star or a group of stars can shine by
spectrum
of
DIFFERENT OPTICAL MANISFETATIONS OF GASEOUS NEBULAE
p.211
p.212
reflected light. This effect was demon-
limit (see Chapter 3) can ionize the
strated observationally by Hubble,
hydrogen atoms. The part of an in-
and explained theoretically in terms of
terstellar cloud where hydrogen has
scattering from dust grains by Russell
been once-ionized (the maxi-mum
in 1922 (see Figure 11.7). The reflection
possible for hydrogen atoms) is called
nebula Problem 11.2. If you were to
an Hu re-gion. The roman numeral a
take a spectrum of a reflec-tion neb-
distinguishes once-ionized hydrogen,
ula, would you see absorption lines,
Ha, from neutral atomic hydrogen,
emission lines, or no spectral lines?
Hi. Bengt Stromgren showed that
How would this help to show that the
the division between Hu regions and
illumination is by reflection from the
Hi regions will he quite sharp if the
central star? Thermal Emission Neb-
surrounding gas cloud is so massive
ulae: Htt Regions Hydrogen atoms in
that all the ultraviolet photons from
an interstellar gas cloud located near
the central 0 or B star are used up
a very hot star- -say, a newborn 0 or
before the Ha region can encompass
B star--will be exposed to the copious
the entire cloud Figure 11.8). Inside
outpouring of ultraviolet ra-diation
the Hit region, the hydrogen plasma
from such a star. Ultraviolet photons
is constantly trying to recombine to
with ener-gies greater than the Lyman
form neutral hydrogen atoms, but
p.213
p.214
p.215
p.216
p.217
the plasma is kept almost completely ionized by the continued outpouring of ultraviolet photons from the central source. These ultraviolet photons break apart any newly formed hydro-
p.218
gen atoms. and the ions and electrons so formed keep recombining to form new atoms. The part of the Hit region where the ultraviolet output of the
p.219
central star is able to keep a balance between recombina-tion and ionization is called the Stromgren sphere. Problem 11.3. To calculate the size of
p.220
the Stromgren sphere, idealize the problem by considering a pure hydrogen gas of uniform density.
p.221
fig. 11.3
The Coalsack nebula Milky Way gives the large region in the is a marked deficit
p.222
in the Southern impression of a sky where there of stars.
p.223
p.224
“
p.332
CLUSTERS OF GALAXY AND THE EXPANSION OF THE UNIVERSE
p.333
p.334
p.335
p.336
p.337
p.338
p.339
p.340
p.341
p.342
p.343
p.344
p.345
S
o far we have discussed galaxies as if they were isolated entities, free to pursue their evolution apart from the influence of other galaxies. In practice, just as there are interacting binary stars, so are there actually interacting binary galaxies in the universe. Just as there are clusters of stars, so are there also clusters of galaxies.
Interacting Binary Galaxies Strongly interacting pairs of galax-
galaxy can then pull out material from
ies constitute a very small percentage
the near side of the larger galaxy into
of all galaxies, but the more spec-
a bridge which temporarily spans the
tacular examples produce intriguing
gulf between the two galaxies (Figure
structures which are not present in
14.2). Except for the noncircular orbit
single galaxies (Figure 14.1). Spectac-
and the flattened original distribution
ular examples of bridges, tails, and
of matter, the bridges are analogous
rings have been catalogued observa-
to the mass-transfer streams that form
tionally by Vorontsov-Velyaminov and
in semidetached binaries (Chapter 11).
by Arp. Our theoretical understanding
The close encounter of two bound
of such systems has been advanced
spiral galaxies, containing much ener-
greatly by the numerical simulations
gy in the ordered forms of spinning
carried out by Hohnberg, Alladin,
disks and orbiting centers, must tend
Toomre, Wright, and others.
eventually to produce a merger into a by
single pile of stars, containing much
Toomre and Toomre of closely grav-
The
computer
simulations
more energy in the disordered form
itating disk galaxies are especially
of random motions. On the other
interesting. Tvhe greatest commotion
hand, if the encountering galaxies
results from orbits which cause the
have nearly equal masses, one long
spinning near edge of a participating
tail from each galaxy can develop
disk to travel in the same direction as
that generally extends away from the
the passing galaxy. If the two galaxies
main bodies (Figure 14.3). This nonin-
have very different masses, the small
tuitive result arises because the tidal
MERGERS
p.332
If we pursue the closer analogy
ters between two bound galaxies
with inter-acting binary stars, long
must tend to bring the two galaxies
tails are analogous to mass lost from
closer together, consistent with the
the outer Lagrangian points, L2 and
constraints of the conservation of
L3 (consult Figure 10.7). Long bridges
total angular momentum and total
and tails are best produced when the
energy. This expectation is based on
orbit of the two galaxies are bound,
the second law of thermodynamics,
so that the two systems are not fly-
which when applied to trillions of
ing past one another too fast when
stars still states that order tends to be
they reach closest approach. The
replaced by disorder (Chapter 4). The
interstellar gas clouds might bang
close encounter of two bound spiral
together inelastically, would tend to
galaxies, containing much energy in
conserve its total energy. In the much
the ordered forms of spinning disks
more rare case when the two bodies
and orbiting centers, must tend
of the galaxies interpenetrate, very
eventually to produce a merger into
exotic looking “ring galaxies” can be
a single pile of stars, containing much
formed. The conclusion that interstel-
more energy in the disordered form
lar extinction and reddening are both
of random motions. This merger pro-
caused by interstellar dust.
cess involves a form of “violent re-
p.333
p.334
p.335
p.336
p.337
laxation,” in which violently changing
Mergers
gravitational fields help to produce a final (relaxed) smooth distribution of
Very close encounters between
stars. Except in having relatively great
much
amounts of interstellar gas and dust,
internal motion in them. The energy
such a pile of stars would probably
to
strongly resemble an elliptical galaxy.
galaxies
obviously
produce
these
excite motions
must
come from the orbital motion. In a
Francois
Schweizer
has
found
system consisting of a collection of
an excellent candidate for such a
gravvitating points (stars), kinetic
merger process (Figure 14.5). In
energy of motion cannot be dissipat-
some expo-sures, the system looks
ed, eventually to leave the system as
like a crumpled spider with no close
radiation. In an encounter between
neighbors. The central body of the
p.338
p.339
p.340
two galaxies, the individual stars fly right by one another, suffering only gravitational
deflections
p.341
produced
by the entire collection of stars. The interstellar gas clouds might bang together inelastically, would tend to
p.342
conserve its total energy. The stars might, however, transform one kind of energy into another kind, say, orbital energy into random motions.
p.343
Indeed, it can be argued on statistical pounds that repeated encoun-
fig. 14.1
Two numerical simulations of ring-producing encounters. An interpretation of a disk galaxy by another massive body yields rippling waves of rings.
p.344
p.345
p.332
p.333
p.334
CLUSTERS OF GALAXIES AND THE EXPANSION OF THE UNIVERSE
fig. 14.2
Successfully deeper photographs (a—e) of NGC7525 reveals a comple set of filaments surrounding a central body that resembles a giant elliptical galaxy. maintained for more than a few times 108 years. If each such interacting pair
p.335
of galaxies (which are probably gravitationally bound) eventually leads to a merger, then during 1010 years, we
p.336
can expect on the order of 400 of the present 4,000 NGC galaxies to have resulted
from
such
coales-cence.
In other words, mergers of spiral
p.337
galaxies would cause 10 percent of all galaxies to become elliptical gal-axies. The actual fraction of ellipticals is more like 20 or 30 percent, and many of these ellipticals are to be
p.338
found in rich clusters; however, it is conceivable that mergers were either more common in the past than now or more common in clusters than in
p.339
the general field. As attractive as Toomre’s proposal
p.340
p.341
p.342
p.343
p.344
p.345
system looks surprisingly like a giant
is, several strong objections to it
elliptical galaxy. The luminosity of the
have been raised. The most damaging
system is equal to that of two lumi-
concern the systematic properties
nosity class I spirals. We apparently
of giant elliptical galaxies—such as
have here the gravitational merger of
the Fish-Freeman relation and the
two giant spiral galaxies into a single
Faber-Jackson relation (Chapter 13)—
pile of stars. After the gas and dust
which would be difficult to explain
has been completely used up to form
in terms of random mergings of a
more stars, the resulting system will
variety of spiral galaxies. In particular,
probably be classified as ordinary.
one would naivety expect the merged
Alar Toomre has speculated that
product to have fewer central stars
perhaps all elliptical galaxies formed
than the constituent spirals, because
in this way. His argument is beguiling,
the excess orbital energy must he
and proceeds as follows. Of the 4,000
absorbed into the merger product
or so NGC galaxies, perhaps a dozen
for hierarchial clustering, supposing
are interacting systems exhibiting
m to have radius R and gravitational
spectacular bridges or tails. On the
attraction of M. For this reason do sin-
other hand, the numerical simulations
gle galaxies develop spiral structure
show that such geometric forms are
(Chapter 12). For this reason binary
transient phenomena that cannot be
stars spiral to fuse to a single object.
HIERARCHIAL CLUSTERING
p.332
p.333
Astronomy compels the Hierarchial Clustering soul to look upwards and leads us from this world to another.
p.334
p.335
As catalogued by Abell, Zwicky,
and others, rich clusters of galaxies may contain several thousand galax-
ies that extend over a radius of ten
p.336
million light-years. On even larger
scales, say, a radius of a hundred million
light-years,
astronomers
have discovered another level to this
hierarchical clustering. It is not known whether clusters of superclusters exist or not. From detailed statistical
p.337
Rich Clusters of Galaxies
analyses of galaxy counts by Shane
About seven times further away
and Wirtanen, Peebles find examples
than Virgo is the Coma cluster, the
of clustering at all sizes,up to 60 mil-
nearest
lion light-years, beyond the amount
thousands of galaxies. Most of the
of clustering drops dramatically. The
galaxies in the Coma cluster are
closest fairly rich cluster to ourselves
ellipticals or SOs. Rood estimated
is the Virgo cluster, located about 50
that only 15 percent of the systems in
million light-years away in the direc-
Coma are spirals or irregulars. Since
tion of the constellation of Virgo (Fig-
iron is believed to be synthesized
ure 14.7). About 200 bright galaxies
only in the deep interior of massive
reside in the Virgo cluster, of which
stars which are destined to supernova
68 percent are spirals, 19 percent are
current belief is that this gas comes
ellipticals, and the rest are irregulars
from the galaxies of the cluster. There
or unclassified. Although spiral’s are
are several theoretical ways in which
more numerous in the Virgo chister,
gas spewed from supernovae might
the four brightest galaxies in Virgo are
eventually find its way into the cluster
ellipticals: among them is, of course,
mediums. There appears to be some
the infamous M87 (Chapter 13).
general features of rich clusters: they
great
cluster
p.338
containing
Plato
p.339
p.340
p.341
p.342
p.343
p.344
p.345
p.332
CLUSTERS OF GALAXIES AND THE EXPANSION OF THE UNIVERSE
p.333
p.334
p.335
are deficient in spiral galaxies, which are quite common in the field and in poorer groups. This mechanism would work for both isolated ellip-
p.336
p.337
p.338
p.339
p.340
p.341
ticals and those in rich clusters. In
Galactic Cannibalism
spiral galaxies, there is probably too
A defining property of a cD
much gas in the disk to maintain a
galaxy is the possession of a very
very high temperature. In the more
distended envelope of stars. A good
general case, gas in the inner regions
example is NGC6166, a radio galaxy
flows toward the nucleus and gas in
that resides in the Abell cluster 2199.
the outer regions blows away in a
This cD galaxy has a visible radius of
galactic wind. The Coma cluster has
about one million light-years, roughly
another characteristic that is shared
20 times larger than an ordinary
by many rich clusters: the presence of
giant elliptical or spiral. Oemler has
one or two very luminous supergiant
performed surface photometry on the
elliptical galaxies near the center of
extended envelopes of cD galaxies.
the cluster. Such a supergiant ellipti-
He finds that they typically drop off
cal (called a cD galaxy for historical
with in-creasing distance from the
reasons which are not particu-larly
center at a slower rate than given by
illuminating)
dominate
de Vaucouleurs law (equation 13.1),
the appearance of the whole cluster.
will
often
which is only valid for ordinary and
Despite the fact that cD systems are
elliptical galaxies.
intrinsically quite rare, they are the
Tidal stripping is simple to un-
most common optical counterpart
derstand. An object of mass m and
member of the universe.
radius R, which is held together by self-gravitation and approaches within a distance r of a massive body of M, will be ripped apart by tidal forces naturally when r becomes too small.
p.342
p.343
p.344
p.345
HIERARCHIAL CLUSTERING
p.332
p.333
p.334
p.335
box AN EDUCATIONAL PROBLEM OF THE
p.336
p.337
INTERNATIONAL AERONAUTICS AND SPACE EDUCATION
p.338
JOURNEY GALAXY
p.339
p.340
to the
p.341
p.342
p.343
p.344
p.345
p.332
p.333
CLUSTERS OF GALAXIES AND THE EXPANSION OF THE UNIVERSE
Thus, a cD galaxy with a mass M which is 500 times bigger than its victim’s mass m will begin to rip the
p.334
latter’s stars from it at a radius r which is 10 times the victim’s radius R. This process
presumably
explains
fig. 14.3
The central region of the Coma Cluster contains two supergiant ellipticals. These cD systems may have grown bloated by cannibalizing their smaller neighbours.
the
formation of the extensive envelope
p.335
around a cD galaxy, although not all the shredded material need be captured gravitationally by the cD galaxy. Some of the stars may enter into orbit about the cluster as a whole cluster of
p.336
stars (see Figure 14.13). Equation (14.1) has a simple physical interpretation. Imagine spreading the mass m of the small galaxy
p.337
uni-formly over a sphere of radius R equal to its size. The average density would
be
3m/47R3. Astronomers
now know that a direct collision
p.338
between two galaxies of comparable size is a relatively rare event: so the odd optical appearance of Cygnus A more probably results from the
p.339
recent gobbling of a gas-rich ordinary galaxy by a supergiant elliptical. Imagine spreading the mass Al of the big galaxy over a sphere of radius r equal to the distance of the small
p.340
galaxy. The average density would be 3M/4rrr3. Equation (14.1) can now be interpreted as stating that the Roche limit occurs when the average density
p.341
of the small galaxy is twice that of the large galaxy. Now, we see a potential limitation to the applicability of equation (14.1). Galaxies do have
p.342
spread-out mass distributions, and a simple estimate gives:
p.343
p.344
p.345
r = (2M/m)”R (14.1)
HIERARCHIAL CLUSTERING
of dynamical friction is not new. In
p.332
p.333
gravitating systems, it originated with the work of Chandrasekhar on stellar dynamics. Chandrasekhar’s own work was inspired by Einstein’s paper on
p.334
Brownian motion. In Chapter 9, we pointed out that encounters between stars of different masses in a star The rarefied outer portions of the
cluster tend to bring about a state
small galaxy will exceed the Roche
of equipartition of energy, where
criterion at relatively large distances r
high-mass stars have low random
from the big galaxy. and tidal stripping
speeds and low-mass stars have
will operate rather efficiently on the
high speeds. Such a thermodynamic
matter there. The dense cores of even
distribu-tion is actually brought about
small galaxies. however, will have to
by two opposing processes. Random
plummet quite far into the heart of
gravitational scatterings of stars by
a cD system before they encounter
one another cause them statistically
interior densities comparable to their
to random walk to higher and higher
own. Since such plungitz, orbits are
velocity dispersions. However, each
a priori rare, how does a cD.galaxy
star also suffers dynamical friction
manage to gobble the cores of other
as described in Figure 14.14, and this
galaxies? By the second course of the
tends to reduce their random veloci-
meal: dynamical friction. For gravitat-
ties. The balance between diffusion
ing bodies, dynamical friction arises
to higher random velocities and drag
for the reason outlined. As a heavy
to lower ones leads, in a steady state,
dense core m (approximated to be a
to a statistical distribution where
point mass) moves through a medium
stars of a given mass tend to have a
containing stars (the envelope of the
certain velocity dispersion (Box 14.1).
cD galaxy), the core deflect the stars it
The relation between the velocity
passes. These deflec-tions statistically
dispersions and masses of stars is
tend to give a slight excess of stars in
given by the principle of equipartition
back of in. (In the language of plasma
of energy, contrast to a star cluster.
physics, the presence of the heavy
Moreover, the stellar masses are
mass polarizes the ambient medium.)
minuscule in comparison with that of
This excess mass pulls on the mass
the system as a whole. Under galactic
m, and thereby tends to reduce its
conditions, neither velocity diffusionv
motion relative to the distribution
nor dynamical friction have enough
of stars. The net effect of dynamical
time to affect the behavior of the stars
friction, therefore, is to bring the
in the universe.
p.335
p.336
p.337
p.338
p.339
p.340
p.341
p.342
galactic core to the center of the cD galaxy. A simple estimate gives: thus the galactic cores’ ambient densities comparable to its own and is chewed
p.343
up by tidal forces, releasing its stars to join the others in the central region of the cD system. The concept
p.344
p.345
PART. FOUR THE SOLAR SYSTEM & LIFE
On July 20, 1969, astronaut Neil Armstrong put his left foot on the rocky Moon. It was the first human footprint on the Moon. People all over the world watched when it happened. Right image: The first footprints on the Moon will be there for a million years. There is no wind to blow them away.
The two astronauts walked on the Moon. They picked up rocks and dirt to bring back to Earth. The astronauts had much work to do. Then, the Eagle went back to meet astronaut Collins. He was in the Command Module working. Apollo 11 splashed down in the Pacific Ocean on July 24, 1969. The astronauts were safe at home.
p.415
THE SOLAR SYSTEM
p.416
p.417
p.418
p.419
p.420
p.421
p.422
p.423
p.424
p.425
p.426
I
n this book so far, we have taken a long journey to explore the universe, and finally have reached the most extreme outposts of observable space and the earliest instants of measurable time. It is now time for us pull back now. Let us return past the Big Bang, past the quasars, past the distant clusters of galaxies, past the nearby groups of galaxies, past the Local Group of galaxies, past the interstellar dusts and gases, into our own Milky Way system of stars and gas clouds, into our own solar system. It’s been a very long journey; it’s really good to be back. The first bodies we encounter as
during recent spacecraft missions
we approach the solar system are fro-
which fiew past Saturn have we got-
zen balls of gas and dust that exist in
ten a really clear look at this spectac-
the billions, usually at a considerable
ular ringed giant. The next planet in,
fraction of the distance to the nearest
Jupiter, is even more massive. Its face
stars. Look there! There’s one of
is covered by an intricate system of
those frozen balls that has wandered
zones and belts, and a giant red spot
too close to the Sun; it’s taken the
further attests to the fierce winds and
familiar, but ever-fascinating, form
storms which wrack this lord of the
of a comet. That’s Pluto. Usually it’s
planets. Like any regal figure, Jupiter
the most distant planet from the Sun,
is surrounded by a retinue of satel-
but during 1979-2000, it’s eccentric
lites; fifteen known moons, great and
orbit has actually carried it inside
small, with lesser bodies quite likely
the orbit of Neptune (Figure 17.2b).
to be discovered in the future. Four of
Also inside Neptune lies Uranus, the
these moons were known to Galileo:
seventh planet of the solar system.
Io, Europa, Ganymede, and Callisto.
Pluto, Neptune, and Uranus are all so
These grand moons are large enough
far away from us that their properties
that they would be called planets in
are relatively poorly known. For ex-
their own right were they not dwarfed
ample, it was only in 1977 that a team
by the great lord.
of astronomers led by James Elliott
Inside the orbit of Jupiter lies a vast
discovered in an airborne observation
wasteland, littered by the debris of
that Uranus was surrounded by a set
failed planets: the asteroids. What is
of thin rings; and it was only in 1978
the true story of their failure? Closer
that James Christy discovered on
yet to the Sun lies the fourth planet,
several high quality photographs that Pluto possessed a satellite (named Charon, the mythical figure who ferried souls across the river Styx to the
p.427
god of the underworld, Pluto). The sixth planet, Saturn, is close enough and bright enough that it was known to the ancients. But only
p.428
THE SOLAR SYSTEM
p.415
p.416
Mars, considered the god of war by
p.417
the ancients because of its fierce red color. Its extraordinarily scarred face bear swims to an active past. Let’s press on, past the third planet, and
p.418
as we approach the Sun, we pass the second planet, Venus, considered by the ancients to be the goddess of love. A thick veil of sulfuric-acid clouds hides her face from us, and modern
p.419
explorations have revealed that her passions are too fiery for mortal man. Closest to the Sun is Mercury. Fleetest of the planets, Mercury, the
p.420
smallest planet in the Solar System was considered by the ancients to be the messenger of the gods. How did he acquire such a pitted face?
p.421
At the center of the solar system lies the true monarch of the realm, the mighty Sun. So plentiful are the Sun’s riches that in one second it squanders more energy than lies in oil under the sands of all Arabia (Chapter 5). Even great Jupiter bows under the
fig 17.1
p.422
A galactic crash captured by the Hubble Telescope
influence of the Sun. As we swing
p.423
around the Sun, we pass Mercury and Venus, and approach again the third planet. We see that she has a faithful
flows over its surface. X-ray pictures
companion, the Moon. A serene if
of rich clusters taken by the Hubble
pock-marked figure, the Moon shines
Telescope find that they present two
by reflected light from the Sun. Last,
morphological classes. The first type
the third planet comes into view, the
has a smooth distribution of cluster
Earth. An insignificant speck of dust
gas, very similar to what simple the-
in this immense universe, yet in our
oretical models predict. The second
eyes, no more beautiful sight exists
shows a clumpy appearance, with the
than Earth, our home, the only planet
gas concentrated in lumps around
which provides us with liquid that
individual galaxies.
p.424
p.425
p.426
p.427
p.428
THE SOLAR SYSTEM
p.415
p.416
Inventory of the Solar System p.417
Planets & Their Satellites Table 17.1 gives a quick inventory of
p.418
the nine planets which circle the Sun, and shows that the planets divide into two categories: the inner or terrestrial planets, which are small and have a
p.419
mean density of 4-5 gm/cm’; and the outer or Jovian planets, which are large (except for Pluto) and have a mean density of 1-2 gm/cm’. Each of the Jovian planets-Jupiter,
p.420
Saturn, Uranus, and Neptune-has more than one moon. Jupiter has at least fifteen moons; the thirteenth was discovered in 1974 in ground-based
p.421
observations conducted by Charles Kowal; the fourteenth and fifteenth, in the 1979 fiyby of Voyager 1. Saturn has at least twenty-two moons, only nine
p.422
of which were known before space missions to this spectacular planet Uranus has five known moons; and Neptune, two. However, more small
p.423
moons may be discovered as future and present spacecraft inspect these giant planets at closer range. In addition to extensive satellite
p.424
systems, the Jovian planets may also all possess rings. The rings of Saturn were discovered by Galileo in 1610, but their true geometry was
p.425
not understood until Huygens offered the correct solution in 1655, and in 1856 Maxwell showed theoretically that the rings must consist of many independent particles. Uranus was
p.426
discovered to have rings by stellar occultation studies in 1977 (Figure
p.427
p.428
fig 17.2 Lunar surface hit by meteorites of the past and present. These crater will last for a very long time as there is no wind to blow them away.
INVENTORY OF THE SOLAR SYSTEM
p.415
p.416 17.4), and direct imaging during the Voyager missions revealed Jupiter to have rings in 1979 (Figure 175). We
p.417
may have to await until 1989, when Voyager 2 is scheduled to encounter Neptune, to discover whether this
Celestial mechanics experiments
Minor Planets or Asteroids, Meteoroids, Meteors & Meteorites
plus ultraviolet photometry by Pioneer
In addition to the nine major
11 showed that the total mass of Sat-
planets and their satellites, there are
urn’s rings cannot exmd 3 millionths
between 10‘ and 106 minor planets or
of the mass of the planet. It used to
asteroids (perhaps more) With orbits
be thought that this mass consisted
that lie between Mars and Jupiter at
mostly of small icy specks; however,
distances of 2 2 to 3 3 astronomical
analysis of radio waves transmitted
units from the Sun. There may also
through the rings during the Voyager
be minor planets further from the
1 flight past Saturn revealed many
Sun, like Chiron (discovered by Kowal
boulders with diameters of several
in 1977), whose orbit lies mostly
meters. ln contrast, the rings of Jupi-
between Saturn and Uranus. The total
ter and Uranus are probably made of
mass of all the asteroids has been
much smaller and less reflective par-
estimated by Kresak to be less than
ticulate material compared to Saturn.
10” 3 of the mass of the Earth where
fourth Jovian planet also has faint rings like Jupiter and Uranus.
The
terrestrial
planets-Mercury,
p.418
p.419
p.420
p.421
we human call home.
Venus, Earth, and Mars-have no rings
Although most of the asteroids
and very few moons. Mercury has no
lie in a “belt” between Mars and
moon; Venus also has none; Earth has
Jupiter, there are some-called me-
one (Luna or the Moon); and Mars has
teoroids-which have Earth-crossing
two (Phobos and Deimos, “Fear” and
orbits. Figure 17 6a shows schemat-
“Panic,” the chariot horses of the god
ically the prevalent theory about
of war). Moreover, in comparison with
the production of such meteoroids
the large moons of the Jovian planets,
from asteroids. The basic idea is that
the moons of the terrestrial planets,
occasionally asteroids collide with
except for Luna, are quite small.
one another. These collisions can
p.422
p.423
p.424
shatter the bodies, and some of the fragments may be thrown into resonant interactions with Jupiter (see Chapter 18) that ultimately give them
p.425
Earth-crossing orbits, like our Earth’s lunar orbit, the moon. A meteoroid which actually intersects the Earth and enters the Earth’s
p.426
atmosphere will heat up from the friction generated in the passage. It will then appear as a fiery, as they call it, “shooting star”, is called a meteor.
p.427
p.428
p.415
THE SOLAR SYSTEM
p.416
p.417
p.418
If the mass of the meteor is less than 10“ ‘° gm, it may slow down so fast that it survives the flight; if so, it is called a micrometeorite. On the
p.419
p.420
p.421
p.422
p.423
p.424
other hand, if the mass of the meteor
Meteorites come in three basic
is greater than 103 gm (a kilogram), it
types, which depend on their chemical
may have enough material to survive
composition: “stones,” “stony irons,”
the ablation and also make it to the
and “irons”. Stones resemble rocks;
ground. If subsequently found, such
stony irons have some metal-rich
an object is called a meteorite. The
inclusions; and irons contain mostly
impact of an especially large meteor
metals like iron and nickel. Fascinat-
may create an enormous crater (Fig-
ing among the stones are a subclass
ure 17.6c). The famous pock-marked
called the carbonaceous chondrites
surface appearances of Mercury and
(Figure
the Moon which look like pores (Fig-
contain carbon-bearing compounds
ures 17.3 and 17.6d) attest to the heavy
(“organic” compounds) in rounded
bombardment that all the terrestrial
inclusions called chondrules. Ordi-
planets must have suffered early in
nary white dwarfs are, of course, dead
their history, and to this time still suf-
stars, whose primary support comes
fer from it. The lack of wind erosion on
from the degeneracy pressure of the
Mercury and the Moon have merely
free electrons. We shall discuss the
preserved a more vivid record of the
significance of these findings for life
cratering process than on the planets
in the universe in Chapter 20. Indeed,
Venus, Earth, and Mars, like the foot-
since the two small moons of Mars
print left by man on the Moon.
are probably captured asteroids, the
17.6e).
These
meteorites
Earth-Moon system is unique among the terrestrial planets, and ought to be regarded, not as a planet-satellite system, but as a double planet.
p.425
p.426
p.427
p.428
Lorem ipsum dolor sit amet, consectetuer adipiscing elit, sed diam nonummy nibh euismod tincidunt ut laoreet dolore magna aliquam erat volutpat. Ut wisi
Box 15.2
INVENTORY OF THE SOLAR SYSTEM p.415
p.416
p.417
p.418
p.419
p.420
p.421
p.422
p.423
p.424
p.425
p.426
p.427
p.428
THE SOLAR SYSTEM
p.415
p.416
p.417
p.418
p.419
Comets
convection or by conduction, not by
Greek name for “long-haired star.”
radiative diffusion (in the interiors).
For naked-eye observers, comets are
Finally, the internal temperatures
one of astronomy’s most spectacular
of planets are far too low to allow
displays (Color Plate 38). Figure 17.7
thermonuclear reactions (see Chapter
shows schematically some features
8, which gave the lower mass limit
commonly found in comets: a head
for luminous stars as 0.08MO). The
consisting, it is believed, of two
loss of internal heat therefore either
parts-a nucleus and a coma-plus one
must lead to a gradual cooling of the
or two tails. The nucleus of the comet
planet, or must be made up by other
is the essential part, since it is the
sources (e.g, radioactive decay, slow
ultimate source of all the mass.
gravitational contraction). Let us now
In the most widely accepted theory
p.420
p.421
place in them also, but primarily by
The word comet derives from the
examine these issues in more detail.
of comets, Fred Whipple proposed
In any case, Jan Oort proposed that
that the nucleus is composed of
there are billions of such small bodies
chunks of dust and frozen ices of
in a “cloud” of about 105 AU in radius
compounds such as methane (CH4),
surrounding the Sun. Some of these
ammonia (NH 3), water (H 20), and
have very elongated orbits, and these
carbon dioxide (Co2). Whipple has
occasionally suffer a gravitational
likened the nucleus of a comet to a
perturbation from a nearby passing
“dirty snowball,” but given that it may
star which sends the body closer to
span a few kilometers across, a “dirty iceberg” might be a more appropriate
p.422
name. In more recent models, the cometary nucleus is thought to have a rocky core, surrounded by a mantle of “dirty ices”. Planets are not now
p.423
collapsing gravitationally; so they must also be in a state of mechanical balance. This balance differs, however, from that of the Sun, in that the matter in the interior of a planet
p.424
resembles nothing like a perfect gas. The interiors of planets contain solid or liquid matter, or a combination of both. As a consequence, mechanical
p.425
balance in planets is largely independent of heat transfer and energy balance. Because the interiors of planets are generally still hotter than
p.426
p.427
p.428
their surfaces, heat transfer does take
fig 17.2 Comet 67P/Churyumov-Gerasimenko shot by Rosetta’s OSIRIS narrowangle camera on 3 August from a distance of 285 km.
INVENTORY OF THE SOLAR SYSTEM
When we meet people who are astronauts or deal in astronomy, it’s always really fascinating.
be thrown out of the solar system
p.415
p.416
p.417
altogether; see Figure 9.18 for an analogy.) Indeed, astronomers know of some 30-odd “Apollo objects,” which are asteroid-like bodies whose
p.418
orbits cross the orbit of Earth. Some scientists believe these objects to be the exposed rocky cores of such “extinct” comets. Wetherill has estimated
p.419
that such Apollo objects strike the Earth once every 250,000 years or so, releasing an amount of energy equal to 100,000 megaton nuclear bombs.
the Sun. When the comet approaches
No such collision has occurred during
within a critical distance of the Sun,
recorded history, but some prehistoric
the ices in a skin around the mantle
impacts may have produced some of
vaporize and form a huge ball of
the larger craters that exist on Earth,
expanding fluorescent gas. This coma
and an especially destructive collision
constitutes the part of the head that is
may have led to the extinction of the
visible on Earth. Two processes then
dinosaurs (Chapter 20). In Chapter 5,
contribute to the development of the
we learned that the Sun represents a
tail(s). Radiation pressure can push
mass of gas in which:
Legend, John
on the dust particles embedded in the
(a) the inward pull of self-gravita-
expanding coma; and as Ludwig Bier-
tion is balanced by the internal (ther-
mann was the first to realize, a solar
mal) pressure of the gas (mechanical
wind can blow past the ionized gas
equilibrium);
and drag on it. These two processes
(b) heat diffuses outward from the
sweep the gas and dust into one or
hot interior, eventually to leave the
two long tails which generally point
surface in the form of freely escaping
radially away from the Sun (Figure
photons (heat transfer); and
17.8). Notice, in particular, that the
(c) the central temperature is raised
tails do not necessarily correspond to
high enough to release enough nucle-
the direction of motion of the comet.
ar energy to balance the heat trans-
Some comets, like Hailey’s comet,
ferred outwards (energy generation
make periodic returns. These must
and thermal equilibrium). Planets are
ultimately have all their volatile com-
not now collapsing gravitationally; so
pounds outgassed from the mantle.
they must also be in a state of me-
(Other comets may be perturbed
chanical balance. This balance differs,
by a close passage to Jupiter and
however, from that of the Sun, in that
p.420
p.421
p.422
p.423
p.424
p.425
p.426
the matter in the interior of a planet resembles nothing like a perfect gas. The interiors of planets, though, contain solid or liquid matter.
p.427
p.428