R A R E E A R T H
92
In the distant past, some early eucaryans (perhaps already endowed
with a nucleus, but still very small and lacking other organelles) may have
routinely ingested other prokaryotic cells for food. This in itself was a major
advance over the prokaryotes, for it necessitated the evolution of an outer
cell wall that could devour, or phagocytize, other cell material—in other
words, it made predation possible. Some of these ingested prokaryotic cells,
however, were not rapidly digested and hence destroyed within the host cell.
Instead, they may have lived there for some time. (Alternatively, these organelles
may have invaded the host cell, rather than being captured by it;
perhaps they bored in and established parasitic colonies within the larger eukaryotic
cell environments.) Eventually, the host cell came to benefit from
this association in some way: Prokaryotes, being very efficient chemical factories,
may have performed services the host could not carry out for itself,
such as energy transformation or even energy acquisition, and metabolic
functions. The organelles known as mitochondria (which are involved in energy
formation and transformation), plastids (the sites of chlorophyll), and
perhaps even flagella (which are used for locomotion) may have evolved in
this fashion.
The prime evidence for this hypothesis comes from DNA. Mitochondria
and plastids contain their own strands of DNA, which are closer in structure
to prokaryotic than to eukaryotic DNA. Mitochondria may have been
free-living bacteria that were capable of oxidizing simple carbohydrates into
CO2 and water and liberating energy in the process. There are living bacteria
today, such as forms known as purple nonsulfur bacteria, that may be close
to the ancestral mitochondrial form. When incorporated into the host cell,
these “guests” eventually lose their cell walls and become part of the host.
With the addition of the cell organelles, our eukaryote approaches or attains
a level of organization that would be familiar to us.
We can now sketch the evolutionary steps necessary to arrive at the eukaryotic
grade of organization. We start with a cell membrane enclosing
DNA—a simple bag of protoplasm and DNA—and then evolve the ability to
phagocytize (or engulf material), evolve a cytoskeleton (which allows us,
among other things, to get larger), evolve aerobic respiration, and then bring
How to Build Animals
93
into our much larger bag various organelles: mitochondria, the nucleus, ribosomes,
and so forth.
This last is among the most interesting and controversial aspects of the
evolution of the eukaryotic cell type. Few scenarios have been proposed that
make evolutionary and adaptive sense, but one intriguing possibility has been
described by Dr. Joseph Kirschvink of Cal Tech, who has summarized the
problems faced by the evolving eukaryote as follows:
The problems for the eukaryotic host cell are that:
The host must be large enough to engulf other bacteria.
The host cell must be capable of phagocytosis, so that the
invaders are put into a membrane-bound vacuole (a small
space within the cell), leading to the characteristic double
membrane of the mitochondria and chloroplasts.
The cell should have at least a rudimentary cytoskeleton.
The host cell should offer a better, more controlled environment
for the symbionts, so that natural selection would
favor the association.
The only known bacterium that meets all of these constraints
is called Magnetobacter, discovered in Germany, which
dwarfs most other protists (in size). Each cell of this bacterium
makes several thousand organelles called magnetosomes, which
are tiny crystals of the mineral magnetite (Fe3O4) encased in a
membrane bubble—a bubble that forms by phagocytosis. These
magnetosomes are held in place in chain-like structures that keep
each crystal aligned properly; this can only be done if an intracellular
mechanical support structure such as the cytoskeleton exists.
Magnetobacter has the ability to keep itself in the optimal environment
by swimming along the magnetic field lines generated by the
Earth’s magnetic field. This ability makes it an attractive partner
for symbiosis, as many organisms spend a great deal of their metabolic
energy staying in the correct environment.
R A R E E A R T H
94
This scenario for the evolution of the eukaryotic cell has two
major implications for the timing of higher life on Earth. First, the
most probable path for the evolution of magnetotaxis (the ability
to align along magnetic fields) and magnetite biomineralization
(the formation of minerals by living organisms) is a result of natural
selection for iron storage. Anaerobic microbes do not need iron
storage mechanisms, as ferrous iron is freely available in solution.
But in oxygen-rich environments the iron rusts out the [ferrous]
form, which drops out of solution. Hence, magnetotaxis is unlikely
to evolve in an anaerobic world, which on Earth ended
about 2.5 to 2 billion years ago. The oldest magnetofossil—the
fossil remains of bacterial magnetosomes—date to about 2 billion
years ago. Second, magnetotaxis requires the presence of a strong
planetary magnetic field. On Earth, a strong early field probably
decayed after 3.5 billion years ago, only to reach its present level
after nucleation of the inner core about 2.8 billion years ago.
Kirschvink has thus postulated a novel scenario—and perhaps the most
plausible scenario—for the formation of the eukaryotic cell: a pathway necessitating
the presence of magnetite and a strong planetary magnetic field.
As we shall see in a later chapter, not all planets maintain magnetic fields. If
this pathway is the only way to large eukaryotic cells (a hypothesis that still
awaits verification), then we have another requirement we must impose on
planets that aspire to host animal life—a magnetic field.
E N V I R O N M E N T A L CO N D I T I O N S L E A D I N G
TO T H E E V O L U T I O N O F E U K A R Y O T E S
What environmental conditions led to the evolution of the forerunners of animal
life? New discoveries of the 1980s and 1990s have given us a much
clearer view of the early Earth during the great evolutionary transitions we
How to Build Animals
95
saw in the last chapter. The Earth’s earliest life seems to have formed during
or soon after cessation of the heavy comet bombardment. By about 3.8 billion
years ago that heavy cosmic bombardment ended, and by 3.5 billion
years ago we find the first fossilized evidence of life.
The region that has yielded Earth’s oldest fossils found to date is known
to Australians as the North Pole, because even in this isolated continent, it is
uniquely remote and inhospitable. The rocks in this region belong to a unit of
interbedded sedimentary and volcanic rocks known as the Warrawoona Series.
Geologists have deduced that the deposits were consolidated in a shallow
sea over 3.5 billion years ago. There is the evidence of storm layers and evidence
as well that on occasion, a hot sun evaporated small pools of seawater
into brine deposits. But it is not these structures that have created so much excitement
about the Warrawoona rocks. This ancient bit of Australia holds the
world’s oldest stromatolites, low mounds of lime and laminated sediment that
have been interpreted as the remains of microbial mats—in other words, life.
Stromatolites (the “stone mattresses” we mentioned earlier as an anomaly,
multicellular prokaryotes) are the most conspicuous fossils and the most
commonly preserved evidence of life for more than 3 billion years of Earth
history; they provide our best record of early life. They have been found on
every continent in rocks half a billion years old and older. Today, they are
found in only one type of environment on Earth, in quiet, briny tropical waters.
Such environments are refuges from algal grazers; stromatolites can no
longer exist on most of our planet’s surface because they would quickly be
eaten. The photosynthesizing bacteria termed cyanobacteria are modern
equivalents of these ancient deposits.
The presence of stromatolites is a sure clue that by 3.5 billion years ago,
life on this planet had left its earliest, probably hydrothermal or deep-earth environments
and diversified onto the surface of the planet. For a billion years the
prokaryotes were masters of the world, but life was still scattered. According to
the fossil record, it was not until about 2.5 billion years ago that the organisms
that produced stromatolites had released sufficient quantities of oxygen to form
sedimentary deposits known as banded-iron formations. Prior to the appearance
R A R E E A R T H
96
of common stromatolites, there was no dissolved oxygen in the sea, no gaseous
oxygen in the atmosphere, and hence no possibility of mineral oxidation. With
the appearance of oxygen, however, large volumes of iron that had been dissolved
in seawater precipitated out as it oxidized into iron oxides—rust, in
other words. Today, there still exist at least 600 trillion tons of such iron oxides
deposited before 2.5 billion years ago in these banded-iron formations.
The time interval commencing about 2.5 billion years ago is marked by
a change in the very tectonic nature of the planet Earth—its rate of mountain
building and continental drift. By this time, the heat production from radioactive
elements locked in Earth’s rocks had diminished, for some of the
radioactive elements decayed rapidly early in Earth’s history. This material
was like a finite amount of fuel within the interior of the planet, and as it was
used up, heat flow declined. It turns out that the processes of continental drift
and mountain building are by-products of heat rising from within Earth, and
as the amount of heat decreased over time, so did these two activities. There
is also some evidence that around this time, a major pulse of land formation
occurred, allowing larger continental land masses to form. As the new continents
formed, many shallow-water habitats were created, and these proved
favorable environments for the growth of photosynthesizing bacteria. We
can speculate that from about 4 billion to about 2.5 billion years ago, there
were few large continents, but numerous volcanic island chains dotted the
world. After 2.5 billion years ago, continental land masses began to form, and
volcanism on a global scale lessened.
With this increase of habitat, ever more stromatolites grew and flourished.
This in turn relentlessly pumped ever more oxygen into the sea. As
long as there was dissolved iron in the seawater, all of the liberated oxygen
was quickly locked up in the banded-iron formations. By about 1.8 billion
years ago, however, this reservoir of dissolved iron material was used up. We
know this because after that time, no more banded-iron formations were laid
down. This changeover left an indelible mark on the sedimentary record of
Earth, for as the sea became saturated with oxygen, the time of banded-iron
formations ended forever—or at least until some far-distant future when our
planet may again no longer have oxygen. With nowhere else to go, oxygen
How to Build Animals
97
began to emerge into our planet’s atmosphere and, in so doing, probably gave
life its first impetus toward animal life.
T H E OX Y G E N R E V O L U T I O N
It is probably impossible for us to conceive how entirely alien to ours this
world truly was. Yet the strange microbial world of 2 billion years ago may be
the norm in the Universe for those planets that harbor life. Traces of it exist
still, here on Earth, in the bacterial froths and pond scum that persist across
our planet, and perhaps nowhere more prolifically than in the rotting garbage
dumps and landfills created by our own species—places where huge, visible
colonies of rapidly growing bacteria exist still. But the rainbow slick of the
oozing swamp is the exception in a world where the eucaryans are so much
more in evidence than the prokaryotic forms. What would that 2-billionyear-
old world look like? The best description we know of was penned by
two scientists who have journeyed back to this world, in their imaginations,
many times. We owe the following image of the ancient Proterozoic era (the
formal name for the time interval of 2.5 to 0.5 billion years ago) to Lynn Margulis
and Dorion Sagan, in their 1986 book Microcosmos:
To a casual observer, the early Proterozoic world would have
looked largely flat and damp, an alien yet familiar landscape, with
volcanoes smoking in the background and shallow, brilliantly colored
pools abounding and mysterious greenish and brownish
patches of scum floating on the waters, stuck to the banks of rivers,
tainting the damp soils like fine molds. A ruddy sheen would coat
the stench-filled waters. Shrunk to microscopic perspective, a fantastic
landscape of bobbing purple, aquamarine, red, and yellow
spheres would come into view. Inside the violet spheres of Thiocapsa,
suspended yellow globules of sulfur would emit bubbles of
skunky gas. Colonies of ensheathed viscous organisms would
stretch to the horizon. One end stuck to rocks, the other ends of
R A R E E A R T H
98
some bacteria would insinuate themselves inside tiny cracks and
begin to penetrate the rock itself. Long skinny filaments would
leave the pack of their brethren, gliding by slowly, searching for a
better place in the sun. Squiggling bacterial whips shaped like
corkscrews or fusilli pasta would dart by. Multicellular filaments
and tacky, textilelike crowds of bacterial cells would wave with the
currents, coating pebbles with brilliant shades of red, pink, yellow
and green. Showers of spheres, blown by breezes, would splash
and crash against the vast frontier of low-lying mud and waters.
This prokaryotic world was creating what has been called the Oxygen
Revolution. The initiation of an oxygenated atmosphere was one of the most
significant of all biologically mediated events on Earth. Prokaryotic bacteria,
using only sunlight, water, and carbon dioxide, ultimately transformed the
planet by generating an ever-increasing volume of atmospheric oxygen. This
outpouring of oxygen created both biotic opportunity and biotic crises. Many
of Earth’s primitive organisms were metabolically incapable of dealing with
abundant oxygen. For most of the archaeans, the oxygen boom of about 2 billion
years ago was an environmental disaster, driving some species into airless
habitats, such as lake and stagnant ocean bottoms, sediments, and dead organisms.
Others were incapable of such migration and simply died out. For yet
other creatures, however, the profound change in atmospheric conditions created
new opportunities. Some prokaryotic cells began to exploit the enormous
power of oxygen metabolism to break down food sources into carbon dioxide
and water. This new metabolic pathway yielded far more energy than any of
the anaerobic pathways. Organisms that adopted it soon began to take over
the world. The most efficient of these were members of the domain Eucarya,
which, more than 2 billion years ago, evolved true eukaryotic cell machinery.
The oldest known fossils of an organism that appears to have attained
the eukaryotic grade of organization have been found in banded-iron deposits
located in Michigan. The fossils themselves are about 1 millimeter in
diameter and are found in chains as much as 90 millimeters long. The organism,
then, is far too large to be a single-celled prokaryote or even a singleHow
to Build Animals
99
celled eukaryote. This creature, which has been named Grypania, is preserved
as coiled films of carbon on smooth sedimentary rock bedding planes, the
places where sedimentary beds split apart. Its 1992 discovery indicates that
the evolution of the first eukaryotic cell occurred during the banded-iron formation
process, when there was still little free oxygen in the sea and probably
none in the atmosphere. These early eukaryotes may have been vanishingly
rare, for other eukaryotes do not occur in the fossil record for 500
million years after this first appearance, but with this form, a beachhead in
life’s advance had been established.
For the period between 2 and 1 billion years ago (see Figure 5.2), few notable
achievements of life are recorded as fossils in the rocks. The first common
appearance of eukaryotes begins about 1.6 billion years ago, when microscopic
fossils called acritarchs begin to appear in the geological record. These are
spherical fossils with relatively thick, organic cell walls. They are interpreted to
be the remains of planktonic algae, forms that used photosynthesis and lived in
the shallow waters of the world’s oceans. Other life forms evolved as well, but
as is also true of most living protists, such as the amoeba and the paramecium,
their lack of skeletons renders them invisible in the fossil record. With a proliferation
of plant-like forms, new varieties of predatory protists surely evolved.
Whole armadas of single-celled, floating pastures and the somewhat larger
and more mobile grazers on these fields of plankton lived and died in this
seemingly endless epoch of geological time. The open ocean would have had
little life, but the coastal regions richer in nutrients would have been awash
with life—microscopic life. It was the Age of Protists, the Age of the Small.
We have now reached 1 billion years ago, in our march through evolutionary
time. Finally, the tempo of evolutionary development increased, if we
are correctly interpreting the fossil record, for there is a burgeoning in the
number of eukaryotic species found in the rock record at this time. Some of
these new forms include the first red and green algae, forms still crucial and
varied in marine ecosystems. This diversification of eukaryotic species, including
protozoans and plants, set the stage for the evolution of larger, multicellular
forms and may have been triggered by the evolution of important
new morphologies within the eukaryotic cell.
R A R E E A R T H
100
E V O L U T I O N I N E U K A R Y O T I C
FORM AND F U N C T I O N
Four biological innovations may have been especially significant in paving
the way for the emergence of larger animals: (1) the development of sexual
cycles; (2) new methods of shuffling information coded along the chromosomes
(through the nascent ability to excise and relocate entire gene sequences);
(3) new methods of communicating between cells via substances
called protein kinases, and (4) the development of a new type of intracellular
skeleton, called a cytoskeleton, that allowed eukaryotic cells to increase
enormously in size. These innovations greatly enhanced the ability of cells to
Archaean Proterozoic Phanerozoic
Paleoproterozoic Mesoproterozoic Neoproterozoic
Grypania megafossils
Undoubted multicellular algae
Chuaria-Tawuia assemblage
Longfengshania
Worm-like megafossils
Ediacara-type fossils
Simple trace fossils
Cloudina
Skeletal fossils
Complex trace fossils
Trilobites
Chengjiang fauna
2.5 2.0 1.5 1.0 0.5 0 Ga
Varanger ice age
?
Figure 5.2 Early multicellular fossils. Broken bars indicate uncertain time ranges.
E A R L Y MU L T I C E L L U L A R F O S S I L S
How to Build Animals
101
evolve new morphologies in response to natural selection and their ability to
band together into multicellular creatures.
We can now better categorize what we term “advanced” life: eukaryotic
multicellular organisms. There are, of course, many types of multicellular organisms,
including a considerable number of prokaryotic forms. In most cases
these multicellular prokaryotes are composed of only two cell types. Cellular
slime molds are multicellular, as are some cyanobacteria. In a way, however,
these forms are evolutionary dead ends. They have existed on Earth for several
billion years and are highly conservative in an evolutionary sense. It is
multicellular creatures of the other category that became so important in the
history of life. We refer here to true metazoans.
The jump from single-celled organisms to organisms of multiple cells
requires numerous evolutionary steps. The jump from single-celled organisms
to metazoan animals, where a high degree of intercellular cooperation in organization
exists, involves even more. In their recent book Cells, Embryos and
Evolution, biologists John Gerhart and Marc Kirschner discuss this evolutionary
accomplishment. The first step, they argue, seems almost paradoxical: It
was not some new structure gained that allowed this transition, but an important
structure lost. Long ago in our planet’s past, some organism of the eukaryotic
lineage made a brave (or lucky) morphological change—it shed its
external cell wall. Why this occurred is still unclear, but the net effect was
far-reaching. A tough outer coating protects most unicellular creatures from
their surrounding environment. At the same time, however, it isolates these
cells from other members of their own kind. By divesting themselves of this
outer wall, individual cells could begin exchanging living material—and
information—with one another. The naked cells could adhere to each other,
crawl over each other, and communicate. These were the first steps in the formation
of a tissue, which is an aggregation of cells united for mutual benefit.
Larger animals require highly integrated systems of cells that can accomplish
the myriad functions necessary for all life. Respiration, feeding,
reproduction, the elimination of waste material, information reception,
locomotion—all require the integration of many cells acting in concert. Each
of these functions ultimately requires one or many types of tissues.
R A R E E A R T H
102
Among tissue types, the outer wall of any organism (the epithelium) is
of utmost importance. The epithelium must protect the organism from the
rigors of the external environment but at the same time allow adsorption of
critical gas and, sometimes, nutrients. The evolution of the epithelium was a
decisive first step in the evolution of metazoans.
Which group of unicellular creatures first achieved this breakthrough?
The most primitive and enigmatic of larger eukaryotic metazoans are
sponges. These curious creatures seem to bridge the gap between singlecelled
eukaryotes or even colonial protozoa and the highly integrated invertebrate
metazoan phyla. Sponges have several cell types that perform specialized
tasks, but there is a very low level of organism-wide organization.
There is no gut or body cavity specialized for processing food, nor is there
any nervous system. Yet the sponges may be an important clue to the identity
of our actual metazoan ancestors.
The stem, or ancestral, metazoan probably had a larger number of cell
types than sponges (perhaps 10 to 15 rather than the 3 to 5 individual cell
types found in sponges). There was probably a body cavity of some sort segregated
into two cell layers: an outer ectoderm and an inward-facing endoderm.
This two-tissue plan seems to have been an evolutionary dead end, and
it wasn’t until a third layer—the mesoderm—was added that animals with
real internal complexity formed. Eventually, a small worm-like shape with
three tissue layers evolved, a creature with a gut running through the long
axis of the body and a separate space known as a coelom to serve as an internal
hydrostatic skeleton. With this tiny organism (the first may have been
less than a millimeter long), the evolutionary stage was set for the emergence
of animals on planet Earth.
T H E TWO DI V E R S I F I C A T I O N S
O F A N I M A L P H Y L A
With the advent of this form—what evolutionary biologists call the “roundish
flatworm”—a body plan was in place that could be modified to shape all the
How to Build Animals
103
categories of metazoan life, the major body plans that we call phyla. The
phyla living today include the arthropods; the mollusks; the echinoderms;
our own group, the chordates; and about 25 more. These are the complex
metazoans that we hope to find—but that may be very rare—on other planets.
These are animals. They appeared relatively late in the history of life on
Earth. One of the great novel insights of the 1990s was our realizing that
their origin and their subsequent diversification and rise to abundance were
two separate events, not one, as had been believed since the time of Charles
Darwin.
Fossils of macroscopic animals (those visible to the unaided eye) first
appear in abundance less than 600 million years ago, during the “Cambrian
Explosion,” a diversification event resulting in the rapid formation of thousands
of new species; we will describe it in more detail in the next chapter.
Yet the appearance of abundant animal fossils at this time actually marks the
second of the two diversification events that led to the proliferation of larger
animals on the planet. As we will show, fossils of such complex animals as
trilobites and mollusks—common members of the Cambrian Explosion—are
advanced descendants of a much earlier, diversification event that took place
between 1 billion and 600 million years ago. Yet there is no fossil record of
this first diversification—paleontologists have been stymied by an almost
compete lack of fossils in strata older than 600 million years, when this initial
event must have taken place. Our understanding of the initial diversification
of animals comes not from paleontology but from an entirely different line of
investigation: genetics. Geneticists have arrived at answers about the “when”
of the first diversification event by examining the genetic code of living animals
via a technique called ribosomal RNA analysis.
Gene sequences are simply strings of base pairs lined up along the double
helix of a DNA molecule. As we saw earlier, if a DNA molecule is likened
to a twisted ladder, the base pairs can be considered the steps of the ladder,
and it is the sequence of the steps that is used in this type of analysis. Genes
are simply instructions for protein formation coded by the sequence of nucleotides
on the DNA ladder. There are only four types of nucleotides, but
they provide the genetic code that is the basis for all Earth life. All organisms
R A R E E A R T H
104
share more genes with their ancestors than with nonrelated species. By comparing
the genes from various organisms, it is possible to produce a model of
evolutionary history (an evolutionary tree, as it were) with the branches of
the tree showing which species gave rise to which other species. Yet according
to many geneticists, such an analysis not only tells us how the branching
occurred, it can also tell us when.
In 1996 G. Wray, J. Levinton, and L. Shapiro published a paper claiming,
on the basis of results obtained by using this genetic technique, that the
first event—the earliest divergence of animals—occurred 1.2 billion years
ago. This result drew a collective gasp from the paleontological fraternity: It
seemed much too ancient. The fundamental assumption of the Wray et al.
paper is that gene sequences evolve with sufficient regularity that a sort of
molecular “clock” can be used to date the divergence of various groups. The
reasoning behind the molecular clock technique is that changes in the genetic
code—evolution, in other words—occur at a rather constant rate. The
more distinct two DNA sequences are, the longer it has been since they diverged
from a common ancestor. Other scientists, however, dispute that
changes in gene frequency occur at a constant rate, and therefore they do not
believe in the molecular clock. It is these molecular clock data that led the
Wray group to their conclusion. This finding was a bombshell. If animals
evolved this early, why did they not appear in the fossil record until less than
600 million years ago? What were they doing for such a long time?
The Wray group’s findings were extremely controversial not only because
they contradicted long-held paleontological dogma but also because
they provoked criticism among other geneticists. There is fierce debate
among geneticists about the reliability of the molecular clock technique. The
Wray study itself, yielded both minimum and maximum figures for the earliest
divergence. One group of genes suggested that the fundamental splitting
of the phylum made up of annelids (worms) from the phylum of chordates
(our phylum) occurred only 773 million years ago, whereas a second group of
genes (in the same organisms) suggested 1621 million years ago—a very wide
spread indeed! These results give us minimum and maximum ages for the diHow
to Build Animals
105
vergence. Even with the minimum figure, however, there were (according to
the molecular data, anyway) recognizable chordates and annelids 700 million
years ago—yet there is no trace of their presence in the fossil record. Where
were they? Or were they not there at all? Could it be that no rocks of this age
survive or that no fossils from the interval of about 1 billion to less than 600
million years ago were preserved? This seems to be stretching things, as suggested
by British paleontologist Simon Conway Morris:
Appeals to gaps in the rock record and pervasive metamorphism
of the sediments are not going to work: if there were large metazoans
capable of either fossilization or leaving traces, they had an
uncanny knack of avoiding areas of high preservation potential.
Since the original, tantalizing analysis by the Wray group, other geneticists
have reconsidered the basic data. Most concede that the 1.2-billion-year
figure is too old. (However, a report published in Science magazine in late 1998
by a team headed by Adolf Seilacher of Yale University announced the discovery
of billion-year-old trace fossils (worm-tracks) possibly derived from small,
worm-like organisms. Critics of this finding suggest that the marks in question
could just as easily have been produced by inorganic actions, and even if these
trace fossils turn out to have been produced by organisms, the question remains:
Why are no further such fossils found for hundreds of millions of years?)
Let’s say, then, that divergence occurred less than a billion years ago. We must
still account for a significant period of time with animals but without fossils. Paleontologists
have long believed that only a single major diversification event
occurred—the event coincident with the appearance of fossils, the so-called
Cambrian Explosion that began about 550 million years ago. Now this evolutionary
event is seen as a follow-up to the much earlier first event.
The answer to this seeming conundrum is that the animals were indeed
present, but they were so small as to be essentially invisible in the fossil record.
A recent and spectacular discovery of microscopic fossil animal embryos
seems to confirm this view. Using newly developed techniques of searching
for tiny (but complex) animals in minerals called phosphates, paleontologist
R A R E E A R T H
106
Andy Knoll and his colleagues have uncovered a suite of tiny but beautifully
preserved fossils interpreted to be the embryos of 570-million-year-old
triploblasts—animals with three body layers, like most of those found today.
These fossils tell us that the ancestors of the modern phyla were indeed present
at least 50 million years before we find any conventional fossil record of
them. The combination of genetic information and new discoveries from the
fossil record now give us a robust view of the rise of animals: They did not
exist 1 billion years ago, and perhaps not 750 million years ago. Animals are
indeed very late arrivals on the stage of life on Earth.
Thanks to these new discoveries and interpretations, the question of
“when” has been answered to most people’s satisfaction: The emergence of animals
was a two-stage event. The initial stage seems to have occurred less (and
perhaps much less) than the billion years ago proposed by Wray and his colleagues.
But even recalibrated, the Wray group’s finding has given us yet another
tantalizing insight into the potential incidence of animal life in the Universe.
The Wray work confirms that there were indeed two “explosions.” The
first was the actual differentiation of the various body plans; the second was the
differentiation and evolution, in these various phyla, of species large and abundant
enough to enter the fossil record. The geneticists can show that genes of
annelid worms and genes of chordates were differentiating hundreds of millions
of years before the emergence of these creatures as large entities that
could appear in the fossil record. This leads us to ask a crucial question: Even if
they evolve, do animals necessarily, or inherently, go on to diversify, enlarge, and survive?
Does the second flowering of animal life—the Cambrian Explosion event so
long known to geologists—inevitably follow the first diversification, or is it yet
another threshold of possibility that may be (but is not necessarily) attained?
Perhaps on some worlds in the Universe, animals diversify but never attain
larger size and greater numbers in some Cambrian Explosion equivalent. This
particular insight was first expressed by paleontologist Simon Conway Morris:
We need to discuss to what extent metazoan history was implicit a
billion years ago, at least in outline, as opposed to what was inevitable
500 million years later at the onset of the Cambrian exHow
to Build Animals
107
plosion. Even if metazoans have a deep history, which paleontologically
remains cryptic, the actual organisms would have been of
millimeter size and perhaps without the potential for macroscopic
size and complex ecology. . . . Wray et al. may have been correct
in tracing the gunpowder back as far into the mists of the Neoproterozoic
(the late Precambrian time period of a billion years
ago), but the keg itself still looks as if it blew up in the Cambrian.
In other words, it seems that the development of animals was a two-step process,
with step two—the Cambrian Explosion—not necessarily being an outcome
predetermined by the initial differentiation of the animal phyla.
Over and over the same question arises: Why did it take so long for animals
to emerge on planet Earth? Was it due to external environmental factors,
such as the lack of oxygen for so long in the history of this planet, or to
biological factors, such as the absence of key morphological or physiological
innovations?
T H E E V O L U T I O N O F A N I M A L S :
B I O L O G I C A L B R E A K T H R O U G H O R
E N V I R O N M E N T A L S T I M U L U S ?
Complex animals surely cannot appear on any planet without following some
evolutionary pathway from simpler, single-celled organisms. The change
from single-celled microbes to multicellular creatures must be the common
route on any planet, and even if the molecules of life are different from world
to world, the pathway from simple to complex may be universal. Because of
this, the example of how animals evolved on our Earth may be of the utmost
importance in understanding the frequency with which animals occur on
other planets.
If we are to understand how animals evolved from single-celled ancestors,
we must first understand the environments where these monumental
R A R E E A R T H
108
Million
years Traditional view Wray et al. view Compromise view
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
Perm
Dev
Sil
Ord
Cam
Carb
Mollusca
Annelida
Echinodermata
Arthropoda
Agnatha
Gnathostomata
Mollusca
Annelida
Echinodermata
Arthropoda
Agnatha
Gnathostomata
Mollusca
Annelida
Echinodermata
Arthropoda
Agnatha
Gnathostomata
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
Proterozoic
Riphean Vendian
Phanerozoic
Million
years
Figure 5.3 Differing views of metazoan phylogeny. Most paleontologists follow the “traditional view”
(left), accepting the fossil record as a fairly reliable indicator of original events. Molecular clocks are interpreted
by Wray et al. (center) as indicating very deep origins for the principal metazoan phyla. The recognition
that some molecular clocks run much faster than others suggests a “compromise view” (right), which implies
that our search strategy for the first metazoans should be concentrated in the interval from about 750
million years onward. Perm, Permian; Carb, Carboniferous; Dev, Devonian; Sil, Silurian; Ord, Ordovician;
Cam, Cambrian.
evolutionary advances were made. We know well the “when” of this change—
it took place during a 500-million-year interval from 1 billion to 550 million
years ago. The second event, the Cambrian Explosion of between 550 and 500
million years ago, included the morphological diversification of the phyla into
subdivisions based on body plans, as well as the appearance, within the various
phyla, of species with skeletons and large size (see Figure 5.3).
During this interval of time, Earth went though major environmental
changes, among them ice ages of unprecedented severity, rapid continental
movements, and drastic changes in ocean chemistry. We are thus left with
perplexing questions: Did the environmental changes of this interval (which
are described in more detail below) somehow trigger the diversification of anHow
to Build Animals
109
imals? Or would the rise of animals have occurred even in the absence of
these profound environmental changes? These questions, which of course are
central to understanding the evolution of life on our planet, have great relevance
to understanding the frequency of animal life on other planets as well.
Does animal life always (or even commonly) evolve, once a suitable ancestor
appears? Or does there need to be an additional trigger of some sort, a sequence
of environmental steps? We might compare this whole process to
baking a cake. Say the ingredients for the batter were all assembled and
mixed by 1 billion years ago. Does the cake need to be cooked for a given
time at a highly restricted temperature in order to rise? Or will any amount of
cooking at any temperature accomplish the task just as well? Or will our cake
be completed without any cooking at all? (That is, does simply assembling
the ingredients into a batter ensure success?)
The beginning of this fecund period in Earth history is marked by the
appearance not of new types of animals, but of plants. Around 1 billion years
ago, many types of algae begin to appear in the fossil record, including the
green and red algae still so prominent on Earth today. These were not the ancestors
of animals, of course, but their appearance was the opening salvo of
an evolutionary assault that was the most significant up to that time. It was
followed, hundreds of millions of years later, first by the initial diversification
of animal phyla and then (after more hundreds of millions of years) by the
Cambrian Explosion of animal life.
What were the environmental events of this interval of time 1 billion to
600 million years ago? By this period, land masses approaching the size of
today’s continents had formed, and the total area of land on the planet may
not have been significantly different from what we see in the present day.
The land, however, was not a tranquil place. The period was one of significant
mountain building and continental drift. It was also marked by episodes
of continental glaciation unmatched in severity since that time. Did these
events have anything to do with the diversification of animals? One school of
thought says yes. Work by Martin Brasier and others suggests that rapid
changes in sea level, and especially the formation of broad, shallow seas
within the new continents, would have opened up many new habitats very
R A R E E A R T H
110
hospitable in terms of temperature and nutrients. This, in turn, may have
stimulated the diversification of animals and plants. There are dissenters, notably
James Valentine, who cautions, “the link between plate tectonics . . .
and the origins and radiation of animals remains to be demonstrated.” But, as
Harvard paleobiologist Andy Knoll points out, there is another way in which
the new and active tectonic events could have influenced the initial radiation
of animals that occurred during this time. In 1995 Knoll noted that “tectonic
processes could have influenced one or more of the great radiations (of animals)
. . . through their participation in the biogeochemical cycles that regulate
Earth’s surface environments.”
Examples of such effects include the role of hydrothermal influences on
ocean chemistry. The hydrothermal vents, as we saw in Chapter 1, are submarine
regions where great volumes of hot and chemically distinctive water
are mixed with seawater. The amount of this volcanically derived water entering
the oceans fluctuated during the interval of 1 billion to 550 million
years ago, and these fluctuations had marked effects on the chemistry of the
seawater, on the composition of the atmosphere, and on climate. The tectonics
events also affected the rate of burial and exhumation of organic carbon
in sediments. Oxygen and carbon dioxide values shifted, and as they did
so, major changes in the temperature and oxygenation of the planet ensued.
Yet another environmental stimulus may also have contributed to the
initial animal diversification. Changes in ocean chemistry caused by increased
tectonic activity beginning a billion years ago facilitated the evolution of
skeletons. This period is marked by the appearance of rocks called phosphorites.
Some authors credit these rocks with bringing about an increase in the
fertility of the oceans at this time, which may in turn have helped trigger the
sudden appearance of many diverse animals beginning about 600 million years
ago. Phosphorus is much more concentrated in living things than in the environment,
so it is a limiting nutrient. The sudden presence of abundant sources
of this element could have acted as a veritable fertilizer for growth.
Knoll has discussed all of these disparate factors and has proposed three
alternatives. First, it may be that the complex physical events and the equally
complex series of biological events that occurred from 1 billion to 550 milHow
to Build Animals
111
lion years ago are simply coincidental—they had nothing to do with one another.
If this is true, the great biological diversification must be attributed
solely to biological innovations (such as the ability of cells to bind together,
build an outer cell wall, and evolve internal cooperation between contained
cells) that were unrelated to concurrent environment changes.
The second alternative is that evolution was indeed facilitated by
changes in the physical environment. The most important of these changes
may have been in levels of oxygen. The first appearance of larger metazoans,
the ediacarans, about 600 million years ago occurred immediately after a sudden
increase in atmospheric oxygen (evidence for this comes from stable isotopes).
Thus it may be that the initial animal diversification of around 700
million years ago was itself a response to the oxygen level reaching some critical
threshold.
The third alternative is that the biological revolutions themselves somehow
triggered some of the physical events—just the opposite of alternative two!
In this scenario, the common use of calcium carbonate shells by newly evolved
animals changed the way calcium was distributed in the oceans. Similarly, organisms
may have favored the formation of phosphorus, not the other way
around: The presence of many organisms may have changed the physical chemistry
of the ocean environment, boosting the formation of this mineral type.
Knoll leans toward the last alternative. He stresses that the first major
evolutionary radiation among protists and algae (about 1 billion years ago)
may have occurred because of the first evolution of sexual reproduction. The
invention of sex, rather than an environmental trigger, stoked the fires of diversification.
But Knoll also acknowledges the central role of oxygenation in
the evolution of larger animals. Without oxygen, larger animals could never
have evolved, and oxygenation during this interval was facilitated by tectonic
processes—specifically, the role of changes in sea level and erosion of continents
in complex geochemical cycles. For a variety of physiological reasons,
oxygen is a key to the appearance of larger animals; the metabolism of animals
requires oxygen.
Indeed, we may well ask whether oxygenation, and hence the rise of animals,
would ever have occurred on a world where there were no continents to
R A R E E A R T H
112
erode. Perhaps “water worlds” are ultimately inimical to animal life. But there
may have been even more sudden and catastrophic changes than those listed
by Knoll—most important, dramatic changes in planetary temperature. Evidence
uncovered in the late 1990s has led to a radical new concept: that Earth
almost completely froze over at least twice in its history—once 2.5 billion
years ago and a second time (perhaps repeatedly) during the interval from
about 800 to some 600 million years ago. These times of intense global cold,
when even the oceans were covered with ice, are known as Snowball Earth.
Their biological significance is explored in the next chapter.
Snowball
Earth
“Let it snow, let it snow, let it snow.”
—Christmas song
It is hard to hide our genes completely.
—Philip Kitchner, The Lives to Come, 1996
Spring is universally associated with birth, growth, and fertility. It is a
time of warmth and renewal after the frigid lifelessness of winter. And so
it would seem that the emergence of animals long ago on Earth should
have resulted from a protracted period of warm and fertile, spring-like conditions.
But new information uncovered by several insightful scientists suggests
that the birth of animal life on Earth was initiated not by a time of warmth
but, rather, by the most fearful winter ever to grip the planet. If this phenomenon,
known as Snowball Earth, turns out to be linked to the origins of animal
life, what will it mean for the possibility of animal life on other planets?
As we noted earlier, a majority of astrobiologists believe that the temperature
of early Earth from the time of the first life, about 3.8 billion years
ago, until the origin of eukaryotic cells, about 2.5 billion years ago, was
high—probably too hot for the existence of animal life. (Yet there are others
113
6
R A R E E A R T H
114
who suggest that Earth may have undergone a “cool start,” because the sun at
that time was giving off much less energy than now). Both camps agree that
the planet’s atmosphere was almost devoid of oxygen. Those who believe in
a “hot start” suggest that gradually, as greenhouse gases in the atmosphere
were reduced in volume, the temperatures declined. But Earth may have
cooled too much (or, if you are of the “cool start” persuasion, failed to warm
up enough), at least in the short term. There is evidence of as many as four
major episodes of glaciation on a scale far exceeding anything before or
since—times of cold and ice that make the last ice age, the Pleistocene epoch
of 2.5 million to 10,000 years ago, seem but a brief cold snap.
The first known Snowball Earth episode began about 2.45 billion years
ago, and a second protracted siege of several such events occurred between
800 and 600 million years ago. These two dates are of great interest, because
they are also the times of the two most signal events in biological history
since life’s first appearance here: Around 2.5 billion years ago the first eukaryotic
cells appeared, and the fossil record reveals that about 550 million
years ago, diverse and abundant animal life blossomed, in the event known as
the Cambrian Explosion, the subject of the next chapter. Perhaps it is just coincidental
that these two spectacular and far-reaching biological events occurred
immediately after the two most severe episodes of glaciation and ice
cover in Earth history. But according to a controversial new theory, both may
have been triggered by the Snowball Earth episodes.
I M P R I S O N E D I N I C E
Continental glaciations leave evidence of their former presence: a characteristic
topography on the landscape, grooves and scratches caused as the passing
glaciers ground over hard rock, and (perhaps most important) telltale sedimentary
deposits called tillites. The latter are deposits of angular rock
fragments, which were carried and then left by moving glaciers. The recently
concluded ice ages of 2.5 million to 12,000 years ago left many such deposits
Snowball Earth
115
in both the Northern and the Southern hemispheres. Such tillite deposits are
also found in much older rocks. Thick tillite deposits have been recovered
from two different intervals in Precambrian Earth history: around 2.4 billion
years ago and during the interval from about 800 to 650 million years ago.
The unusual aspect of these features is that they are recovered from virtually
all latitudinal regions of the globe, which shows that the glaciations extended
to near equatorial latitudes (in contrast to the more recent glaciations, which
extended from the poles to mid-latitudes). It may be that no region then on
Earth escaped the glaciation. So much of the planet was covered by ice in
these two Precambrian ice ages that in 1992, Dr. Joseph Kirschvink of Cal
Tech dubbed them “Snowball Earth” events. Far different from the later ice
ages, they were times when Earth teetered dangerously close to becoming
too cold for any life. The Snowball Earth theory received a boost in August
1998 with Harvard geologist Paul Hoffman’s publication, in Science, of new
evidence that ice extended to near equatorial latitudes in the late Precambrian,
about 700 million years ago.
The more recent glaciations, those that occurred since skeletons
evolved about 550 million years ago, affected only land regions; except for an
increase in icebergs, or at most ice cover near the continents, the oceans remained
open. Such may not have been the case in the Precambrian glaciations.
During these two “Snowball Earth” episodes, all of the oceans may have
been covered with ice to considerable depths. And although the deeper regions
of the seas remained liquid, thick icebergs, or pack ice to depths of 500
to 1500 meters, may have covered the ocean. Earth would have been cold indeed.
Average surface temperatures on the planet would have varied between
20°C and 50°C.
These extremely cold temperatures would have had an enormous influence
on the surface of our planet. For example, continental weathering would
have slowed or even stopped. In the interior of continents, the covering of ice
would eventually ablate (evaporate) away, just as it does in the dry valleys of
Antarctica today, leaving behind a sterile rock surface. Dust from these regions
would be blown out to sea, making the pack-ice cover of the oceans
R A R E E A R T H
116
brown from terrigenous material. From space, Earth would have looked white
and brown—the white being the ice covers on the oceans, the brown the denuded
land areas.
The presence of the pack ice covering the sea would act as a lid on a
pot. Normally, much free exchange occurs across the vast interface of ocean
and atmosphere. Water evaporates from the sea into the air and then rains
back into the sea. If the sea were covered with ice, however, the ocean and
the atmosphere would become “decoupled.” Chemical changes in the ocean
would be separated from the atmosphere by the kilometer-thick lid of ice on
the ocean surface. Very drastic chemical change could—and according to
Kirschvink and others, did—occur within the sea itself.
Even with the icy cover, volcanism would have continued both on the
land’s surface and along the mid-ocean volcanic ridges at the bottoms of the
world’s oceans. At such sites today (see Chapter 1), great volumes of metalrich
fluids gush forth from these submarine volcanoes. In a covered ocean,
this material would have become toxic, producing what are known as reducing
conditions. The oceans would have begun to accumulate with metal ions,
mainly iron and manganese. For as long as 30 million years, the glaciers and
ice never relaxed their frigid grip on the planet’s surface.
All of this global cold would surely have adversely affected life in the
shallow-water regions of all the world’s oceans. The biosphere became restricted
to a narrow belt around the equator and to deep-sea hot springs and
hydrothermal vent settings. Perhaps some life also survived in occasional
Yellowstone-like hydrothermal systems.
Astronomers once thought that a previously warm world’s descent into
such an “icehouse” or “snowball” would be irreversible. Their reasoning was
that as a planet gets more and more thickly coated by ice, the fraction of sunlight
reflected back into space increases and solar heating of the surface declines.
On Earth today, sunlight is adsorbed by the darker land and seas but
is reflected into space by cloud cover. A planet completely covered with ice
would reflect most sunlight into space, causing the planet to become ever
cooler. Yet it is clear that Earth was able to escape from the deep freeze—not
once but several times. The means of that escape was through the volcanic
Snowball Earth
117
emissions of greenhouse gases such as carbon dioxide into the atmosphere,
producing a “greenhouse effect.”
E S C A P E
As we saw in Chapter 2, a planet’s average temperature is greatly affected by
the volumes of greenhouse gases in its atmosphere. Much of this gas enters a
planet’s atmosphere from actively erupting volcanoes. Although there are
abundant volcanic eruptions in the sea as well, most of the carbon dioxide
from these events does not make its way into the atmosphere. Cold seawater
can hold large amounts of dissolved carbon dioxide, and below 700 meters,
CO2 will settle to the bottom of the ocean as it reaches saturation in the
water. At the time of Snowball Earth, enough CO2 would eventually reach
the atmosphere to melt back the sea ice and, in so doing, expose the metalrich
waters of the sea to the atmosphere. The time necessary for this “meltback”
has been estimated by Hoffman and his group to be between 4 and 30
million years. With the ice melted back from the sea, and temperatures again
warming, Earth would have undergone spectacular changes. Here is how
Kirschvink has described these events:
Escape from this “icehouse” condition was only accomplished by
the buildup of volcanic gases, particularly carbon dioxide, mostly
from undersea volcanic activity. Deglaciation during the end of
these glacial events must have been spectacular, with nearly 30
million years of carbon dioxide, ferrous iron, and long buried nutrients
suddenly being exposed to fresh air and sunlight. Hundreds
of meters of carbonate rock are preserved capping the glacial sediments,
at all latitudes, on all continents, as a direct result of wild
photosynthetic activity. For a brief time, the Earth’s oceans would
have been as green as Irish clover, and the sudden oxygen spikes
may have sparked early animal evolution.
R A R E E A R T H
118
The most important source of biological productivity in the oceans of
today derives from the growth of phytoplankton, the single-celled plants that
are the pastures of the sea. The growth of these plants, so important for producing
oxygen, is limited by the availability of nutrients and iron. If iron is
dropped into the oceans of today, a great bloom of phytoplankton results.
Such was probably the case soon after the end of the first Snowball Earth
event. As the ice-covered seas began to melt, the fine iron- and magnesiumrich
dust coating the surface of the sea ice would have acted as a fertilizer,
tremendously stimulating growth of the blue-green “algae” (really photosynthesizing
bacteria known as cyanobacteria). Enormous populations of
cyanobacteria would have clotted the surface regions of the liberated seas, releasing
huge volumes of oxygen as a consequence of their photosynthetic activity.
This sudden appearance of so much life, after the millions of years of
cold and dearth of life, would have been a great revolution, and it probably
stimulated new evolutionary changes.
These events would have had profound geological as well as biological
ramifications. The sudden rush of oxygen into the sea and air would
have caused the iron- and manganese-rich oceans to precipitate out iron
and manganese oxides. In a previous chapter we saw how banded-iron deposits
began to accumulate about 2.5 billion years ago. Kirschvink and his
group argue that the appearance of banded-iron deposition occurred soon
after the first Snowball Earth ended. Not only iron deposits but magnesiumrich
deposits as well were immediate results of the end of the first Snowball
Earth event. Evidence of this is seen in South Africa, where the world’s
largest land-based deposit of manganese minerals has been dated at 2.4 billion
years of age and sits just above sedimentary deposits that were laid
down during the 2.5-billion-year-old Snowball Earth episode. Like the
banded-iron formations, these manganese-rich deposits appear to be a direct
consequence of the oxygen bloom that occurred when the planetary
snowball melted.
The cessation of the 2.5-billion-year-old Snowball Earth thus appears
to have resulted in a rise in the amount of oxygen both dissolved in the sea
and free in the atmosphere. Probably for the first time in Earth’s history, the
Snowball Earth
119
sunlit portions of the sea became too oxygen-rich to allow iron to exist in solution
in seawater. Kirschvink and his colleagues argue that this dramatic
change in the chemistry of the sea would have exerted intense evolutionary
pressure on life on Earth, then no more advanced than prokaryotic bacteria.
Oxygen, indispensable to the survival of animals, was at that time a poison to
perhaps the majority of life forms. Having evolved in environments with little
or no oxygen, most life experienced the sudden appearance of the chemically
reactive element as a global disaster—but for the rest it was a powerful
evolutionary spur. There were but two choices facing life on Earth in that
long-ago time: Adapt through evolution, or die.
All organisms in the sea had to adapt in two major ways. First, they had
to evolve enzymes capable of mitigating the ravages of dissolved molecular
oxygen and chemicals called hydroxyl radicals. (We humans are still trying to
do this. Our ingestion of antioxidants such as vitamin E and vitamin C is an
attempt to reduce the ravaging effects that dissolved oxygen and “free radicals”
have on living cells.) Second, with the banded-iron formations’ precipitation
out from seawater, living cells no longer inhabited a solution rich in
iron. After having been surrounded by high-iron solution since the first formation
of life, proteins within cells had to be reengineered for life in an environment
low in iron.
Recent DNA sequencing has shown that several enzymes found in archaeans
and eukaryotes are left over from this event of 2.5 billion years ago.
No such enzymes occurred in the older bacteria. The implications of this are
profound: Kirschvink and his colleagues are proposing no less than complete
rejection of the Tree of Life models we examined at the end of Chapter 3,
which suggest that the three great domains (Archaea, Bacteria, and Eucarya)
all arose soon after life’s first evolution at least 3.8 billion years ago. The new
study has not only uprooted this tree; it has burned it. If the Kirschvink group
is correct, two of the three domains—Archaea and Eucarya—arose only after
the 2.5-billion-year-old Snowball Earth and are thus much younger than the
bacteria. Soon after this, in rocks about 2.1 billion years of age, we find a
record of the oldest organelle-bearing eucaryan—the creatures known as
Grypania, which we mentioned in Chapter 3.
R A R E E A R T H
120
This new version of the Tree of Life is a revolutionary scientific discovery,
and if true, it will utterly reshape our understanding of life’s evolutionary
path. The Snowball Earth events can be seen as biologically important in two
ways. First, the inception of the Snowball produced what may have been the
largest “mass extinction” (the subject of Chapter in our planet’s history.
The persistence of globally freezing temperatures, the isolation of the ocean
from sunlight, the change in the precipitation patterns on Earth, and the removal
of all water from the surfaces of continents would have removed the
majority of surface habitats then available for microorganisms. In only a few
places could microorganisms have survived: in the deep earth, around hot
springs, and in hydrothermal deposits. Second, Earth’s release from this icy
prison after 30 million years brought about a new catastrophe: from cold to
hot, from oxygen-free to oxygen-rich. Again, organisms had to adapt rapidly.
It is this legacy that we may be seeing in the DNA of all living organisms;
those that survived all bear witness in their DNA to this dual catastrophe—
first cold, then warmth and oxygen. Life on the early Earth went through an
icy bottleneck, and it came out the other side radically changed.
The Snowball Earth of 2.5 billion years ago may have given our planet
eucaryans and the eukaryotic cell necessary for animal life. The second series
of Snowballs (there were several in rapid succession) may have bequeathed
our planet an even more interesting biological legacy—animal life as we
know it.
T H E S E C O N D GL O B A L GL A C I A T I O N
As we saw in Chapter 5, by the next round of Snowball Earth events, those
spanning the time interval from 800 to 600 million years ago, animal life was
present on Earth, but it was newly formed. Either simultaneously with or
soon after the appearance of the new animal phyla, Earth was once more
locked into a global icehouse. Once again, there must have been a period of
mass extinction, as the warm planet froze and the heat-loving organisms of
Snowball Earth
121
Earth had to retreat to oases of heat, such as around volcanoes and hydrothermal
vents, or die. Yet the very severity of these events may have benefited
the newly arisen animals. The great stress inflicted by environmental
conditions imposed by the Snowball events would have stimulated inordinately
rapid evolution among the newly evolved animals. It would also have
caused the isolation of various populations, because the small populations of
life huddled around the undersea volcanoes would have been cut off from any
exchange of genes with other animal groups. This very isolation may have
been largely responsible for the diversity of phyla that emerged at the other
end of these crises, for when the final Snowball Earth event ended, about 600
million years (or less) ago, an entirely new group of creatures was ready to
take over the planet. This is the interval when animal life began to diversify
dramatically, in an event known as the Cambrian Explosion, the subject of
the next chapter.
Would this have happened if the glaciations had not occurred?
Kirschvink and Hoffman suggest that there is a causal link between the cessation
of these major glaciations and the emergence of animals. Hoffman has
noted, “Without these ice events, it is possible there wouldn’t be any animals
or higher plants.” He believes that the melting of the ice at the end of these
ice ages boosted biological productivity—and in the process stimulated evolutionary
activity. This idea has yet to be confirmed, but it remains a tantalizing
possibility.
Both of the two great episodes of Snowball Earth nearly ended life on
Earth, as we know it. But each, ultimately, may have been crucial in stimulating
the great biological breakthroughs necessary for animal life: the evolution
of the eukaryotic cell and then the diversification of animal phyla. This leads
us to ask whether Snowball Earth events are necessary to produce animal life as
diverse as that seen on Earth today.
The end of the last Snowball Earth event brought the time interval
known as the Precambrian to a close. Soon thereafter, abundant skeletons of
larger animals began to fill the sea, in the Cambrian Explosion. If the two
groups of scientists led by Joseph Kirschvink and Paul Hoffman are correct
R A R E E A R T H
122
about Snowball Earth, a good case can be made that life on Earth is to some
extent due to these events.

P L A N E T A R Y S U R F A C E T E M P E R A T U R E S
AND T H E E M E R G E N C E O F L I F E
The discovery of the Snowball Earth episodes suggests that temperatureinduced
events in planetary history may profoundly affect the course of biotic
evolution. This argument can perhaps be extended not only to specific
episodes of planetary temperature change but also to actual temperature values
over time. Could cooling planetary surface temperature reaching some
critical value have been the stimulus for other major breakthroughs in biological
evolution?
As we saw in Chapter 2, the habitable zone is most commonly defined
in terms of the presence of liquid water; this definition thus includes everything
from life forms capable of living in boiling water to those capable of life
in ice or snow. It may be that over much of its history, Earth was either too
hot or too cold to allow the emergence of animals. Environments with temperatures
near the freezing point or the boiling point of water are occupied
largely by microbes; animals tolerate a much narrower temperature range.
David Schwartzman and Steven Shore have pointed out that eukaryotic organisms
with mitochondria (the organelles that convert fuel into energy)
have an upper temperature limit for viable growth of 60°C. This limit is apparently
determined by the chemical structure of the mitochondrial wall. Because
eukaryotes evolved from prokaryotes, the habitable zone of a planet is
narrowed from the region that allows the presence of water (0–100°C) to the
narrower range of 0–60°C. Schwartzman and Shore note, “We assume that
the emergence of relatively simple life forms is almost inevitable on Earthlike
planets. Such organisms are remarkably robust. Complex life, however, requires
a more restrictive set of physical conditions—in particular, lower temperatures.”
Snowball Earth
123
Schwartzman and Shore provided the following list of the critical upper
temperatures for various organisms on Earth.
Approximate Upper Time of First Appearance
Group Temperature Limit (°C) on Earth (billions of years ago)
Multicellular plants 45–50 0.5
Animals 50 1–1.5
Eukaryotic microbes 60 2.1–2.8
Prokaryotic microbes
Cyanobacteria 70–73 3.5
Methanogens 100 3.8
Extreme thermophiles 100 3.8
Schwartzman and others have proposed that Earth’s surface temperatures
have been the critical constraint on microbial evolution, determining the timing
of major innovations. They believe that when Earth’s surface cooled
below 70°C more than 3.5 billion years ago, cyanobacteria were able to
evolve. These microbes colonized the surface of the land and, in so doing, increased
weathering rates and soil formation. The new soil in turn acted as a
sink for removing carbon dioxide from the atmosphere, thereby causing further
cooling of the planet. Each innovation among microbes resulted in biotic
enhancement of weathering. This process has been dubbed “biotically mediated
surface cooling.” With the evolution of higher plants with their often
elaborate root systems, this process increased greatly in efficiency. It is at the
heart of the “Gaia Hypothesis,” wherein Earth is conceived of as a selfregulating
“superorganism,” a perspective shared by many scientists, including
Lynn Margulis, Tyler Volk, and the originator of the term, James Lovelock.
We remain agnostic on this particular interpretation but see much merit
in the view that the emergence of animals may have been strongly influenced
by surface temperatures—and that life itself on this planet has had an enormous
impact on planetary temperatures.
R A R E E A R T H
124
Could there be any way in which (or any planet whereon) animals
could evolve faster than they did on Earth? The physical events affecting
Earth immediately before the emergence of large, skeletonized animals were
among the most complicated in all of Earth history. Was this just coincidence,
or did it make the acceleration of animal evolution possible? These
questions, and the curious and dramatic fashion in which the major body
plans of animals suddenly began to commonly appear in the fossil record on
our planet about 540 million years ago, are the topics of the next chapter.
The Enigma
of the
Cambrian
Explosion
Evolution on a large scale unfolds, like much of human
history, as a succession of dynasties.
—E.O. Wilson, The Diversity of Life
Our planet was without animal life for the first 3.5 billion years of its
existence and was without animals large enough to leave a visible
fossil record for nearly 4 billion years. But when, 550 million years
ago, sizable and diverse animal life finally burst into the oceans, it did so with
a figurative bang—in a relatively sudden event known as the Cambrian Explosion.
Over a relatively short interval of time, all of the animal phyla (the
categories of animal life characterized by unique body plans, such as arthropods,
mollusks, and chordates) either evolved or first appear in the fossil
record. Undoubted fossils of metazoan animals have never been found in
600-million-year-old sedimentary strata, no matter where on Earth we go.
125
7
R A R E E A R T H
126
Yet the fossils of such animals are both diverse and abundant in 500-millionyear-
old rocks, and they include representatives of most of the animal phyla
still found on Earth. It appears that in a time interval lasting at most 100 million
years (and in fact, as we will see, in an interval considerably shorter than
that), our planet went from a place without animals that could be seen with
the unaided eye to a planet teeming with invertebrate marine life rivaling in
size almost any invertebrate species on Earth today. This follow-up to the initial
animal diversification of more than 700 million years ago (described in
the last chapter) is the Cambrian Explosion.
The rate of evolutionary innovation and new species formation during the
Cambrian Explosion has never been equaled. The prior animal diversification
must have involved very few species, each growing to a very small size; the
Cambrian Explosion, on the other hand, produced huge numbers of new
species, many with completely novel body plans. The Cambrian Explosion
presents a great challenge to astrobiology, as we will show in this chapter. Questions
abound. For instance, can there be animal life on a habitable planet without
this type of event? Is the Cambrian Explosion an effect or a cause? That is, could
it be that the remarkable animal diversity on Earth today is a by-product of this
sudden diversification and would not have come about if the Cambrian event
had been a mild bang rather than an explosion? Was it inevitable once the late
Precambrian first event had occurred, or was another set of stimuli required?
What animals were involved? What were the event’s biological origins? What
caused it to occur? (Was there some sort of biological or environmental trigger?)
And, most relevant to astrobiology, was the Cambrian Explosion inevitable once
a certain level of biological organization had evolved? In other words, is there
any way that the Cambrian Explosion might not have occurred?
WH E N DI D T H E CA M B R I A N
E X P L O S I O N T A K E P L A C E ?
The Cambrian Explosion is marked by the sudden appearance of larger fossils,
which can be easily seen at many places around the globe. There is
The Enigma of the Cambrian Explosion
127
nothing subtle about this evidence, and it was known to even the earliest geologists.
In Washington State, for instance, telltale signs of the Cambrian
event are readily visible near the small town of Addy, where a slow country
road meandering through the Coast Range foothills cuts through low outcrops
of quartzite, the lithified remains of what, more than 550 million years
ago, was a white, sandy beach. If we could travel back to that time, the
beach itself would probably not elicit our wonder, for it would appear completely
unremarkable. Beaches are beaches regardless of time. But the appearance
of the nearby shore and inland vistas would be remarkable; there
would be no plants (or animals) to be seen. On Earth today there are a few
areas where plant life is not immediately visible—the harshest deserts, the
Arctic and Antarctic regions—but these are exceptions on a planet otherwise
carpeted with life. Yet such was not the case 550 million years ago. And
it was not only the land that was barren: If we were able to wade through the
shallow, warm sea, we would not encounter any shimmering fish or scuttling
crabs, no starfish and no sea urchins. There would be no clams burrowed in
the sand and almost none of the other animals we see so commonly along
the seashores of our world. There might be a few worms or jellyfish, but
nothing with a readily visible skeleton. We would conclude that this world
boasted little life, or at least little that we would recognize as animals or land
plants.
The quartzite in this region is exposed as sedimentary layers stacked
one on top of another, and if we were to count the individual layers (or beds),
they would number in the thousands. The lowermost beds are devoid of fossils.
Yet because these rocks are stratified, they are organized by time. If we
wander a short distance farther along the roadside outcrops, moving upward
through the succession of these stratified beds (and thus into younger intervals
of time preserved in the rocks), we see a wondrous thing. Suddenly, as
though by magic, an abundance of fossils appears. We find the remains of
shelled creatures called brachiopods, which look like small clams, and a few
other types of fossils, such as sponges and a tiny mollusk or two. But by far
the most common fossils to be found are also the most spectacular, for the
first fossil-bearing beds at Addy are packed with trilobites.
R A R E E A R T H
128
Along with ammonites and dinosaurs, trilobites are perhaps the most
iconic of all fossils. At first glance they seem to resemble large bugs or crabs
of some kind, but upon closer examination they look like nothing still alive;
their closest living relatives are horseshoe crabs and pill bugs, but these are
only distant cousins. Trilobite fossils range in size from the microscopic to
nearly 3 feet in length. They have numerous spines, great helmet-like heads,
and a variety of peculiar eyes, and their undersides housed an array of legs,
gills, and other assorted arthropod tools. All in all, they are complicated fossils
from complicated creatures—and, for that reason, are unlikely candidates
for the honor of being the world’s oldest animal fossils. If Darwin’s theory of
evolution is correct, the first fossils should be far simpler than a trilobite—as
indeed they are. Yet at Addy, as at so many other localities around the world
with sedimentary rocks of this age, the first obvious fossils are indeed trilobites
perched atop thick sequences of strata apparently devoid of fossils. This
observation suggests that animals of staggering complexity appeared on
Earth without evolutionary precursors. It is as though an orchestra began
playing without sounding a single tone to tune up.
This sudden appearance of larger animals in the fossil record is the most
dramatic aspect of the Cambrian Explosion. It drove Charles Darwin to distraction
and challenged the newly evolving field of geology, which had taken
as its guiding principle the idea that important events in Earth history unfolded
gradually, not abruptly. Yet even to the earliest geologists, the Cambrian
Explosion seemed anything but gradual.
In the early nineteenth century, geology was a newly born scientific discipline
established largely for economic motives, such as the search for fuel
and metals. It was clear that the discovery of these valuable commodities depended
on finding the relative age of rocks. By that time it was also recognized
that fossils were the remains of ancient life and that they appeared in a relative,
superpositional order and thus could provide a practical and reliable method
for determining the relative ages of rock bodies. With the aid of fossils, geologists
soon began to subdivide Earth’s sedimentary strata into time units.
In 1823, English geologist Adam Sedgewick named one such unit the
Cambrian. Sedgewick observed that a thick sequence of sedimentary rocks in
The Enigma of the Cambrian Explosion
129
Wales contained a characteristic assemblage of fossils, including numerous
trilobites. Overlying these strata were sedimentary rocks with a different
suite of fossils that represented a time unit eventually named the Ordovician.
Yet as he continued mapping and describing the mineral and fossil content in
his field area, Sedgewick encountered something novel: strata without fossils.
The Welsh sedimentary rocks studied by Sedgewick were composed of an
enormous thickness of unfossiliferous strata, overlain by an equally thick pile
of strata containing trilobites and brachiopods. Even more curious, the transition
between the unfossiliferous and fossiliferous strata was abrupt, not
gradual.
The strata beneath the fossiliferous, Cambrian rocks became known as
the Precambrian. The Cambrian period was defined as the block of time during
which the fossil-bearing strata recognized by Sedgewick in Wales were
deposited. Thanks to modern dating techniques, we now know that this unit
of time started about 540 million years ago and ended about 490 million years
ago. Although Sedgewick’s strata are found only in a part of Wales, we refer
to all rocks on Earth that formed between 540 and 490 million years ago as
belonging to the Cambrian system.
Sedgewick defined the base of the Cambrian as the stratal level where
the first trilobite fossils could be found, and that view prevailed for over a century.
Anywhere in the world where trilobite-bearing strata overlay unfossiliferous
strata was considered to mark the base of the Cambrian. Recently, however,
the way in which the base of the Cambrian is recognized has changed.
It is now marked at a level that Sedgewick would have considered below the
“base” of the Cambrian. Today geologists use the first occurrence of a particular
trace fossil (the fossilized record of animal behavior, rather than the preserved
hard parts of the animal itself) as the base of the Cambrian system.
Sedgewick’s discovery of the seemingly instantaneous appearance of
complex fossils convinced most scientists of his time that life was spontaneously
created—put on Earth through the action of some deity; this observation
is still cited by creationists as evidence against the theory of evolution.
This observation was perhaps the most difficult for Charles Darwin to reconcile
with his newly proposed theory of evolution, for the apparently sudden
R A R E E A R T H
130
appearance of large, complex animals in the fossil record ran utterly contrary
to his expectations. In On the Origin of Species, he speculated that the Precambrian
interval must have been of long duration and “swarmed with living creatures.”
Yet where were the fossils of these swarms? Surely, if Darwin was correct,
a long period of evolutionary change with simpler precursors would
have been necessary to produce the complex creatures collected by
Sedgewick and others in the lowest strata now known as the Cambrian. Darwin
was never able to refute this stringent criticism of his theory. Instead, he
railed against the “imperfections” of the fossil record, believing that there
must be a missing interval of strata just beneath the first trilobite-bearing beds
everywhere on Earth. He was convinced that there must be Precambrianaged
fossils. As it turns out, he was right, but he went to his grave unvindicated.
Paleontologists have since proved Darwin correct, for the supposedly
“barren” strata beneath those that bear what were thought to be the first fossil
skeletons do indeed contain the ancestors Darwin sought and theorized
about. They were long overlooked or missed, however, because of their rarity
or very small size. Most organisms from the youngest “Precambrian” time
both were tiny and lacked skeletons, so they rarely left obvious traces in the
fossil record. They are very hard to detect unless special processing techniques
are used to extract them from their entombing matrix; Darwin and his
contemporaries had not yet dreamed of such methods. The supposedly “sudden”
appearance of skeletonized life, more than 540 million years ago, is simply
the first appearance of creatures with large skeletons, which produce fossils
that are easily noticed. Because of this, the base of the Cambrian has now
been “lowered” into the supposedly barren strata beneath the first trilobitebearing
beds. Just as Darwin supposed, trilobites do appear only after a longer
period of evolution of simpler forms that rarely fossilize.
The twentieth century has witnessed a revolution in the science of geology.
No longer are fossils the sole means of dating rock. Sophisticated laboratory
analyses of volcanic and some sedimentary rock give accurate ages
in years, and the entire rock record (including the Cambrian) has been far
more accurately dated. In the 1960s the base of the Cambrian was deterThe
Enigma of the Cambrian Explosion
131
mined to be 570 million years old, and this date appears on age compilations
even into the late 1980s. Recently, however, there have been significant improvements
in radiometric dating techniques. The Precambrian/Cambrian
boundary is now dated at 543 million years old. The “Middle” Cambrian is
dated at about 510 million years ago, whereas the oldest trilobites are no
more than 522 million years old, which suggests that the bulk of Cambrian
time was “pre-trilobite.” Interestingly enough, although the “base” of the
Cambrian has gotten younger; its “top” has not changed in age. The Cambrian
Explosion remains a relatively sudden and signal outburst of animals—
an unleashing of abundant and voracious creatures upon the earlier bacterial
world, which continues, unabated, more than half a billion years later. With
the exception of life’s first formation, it remains the most profound biological
event to have occurred on this planet. And we propose that the Cambrian
Explosion has an even greater significance than Charles Darwin (or modern
scholars of the fossil and evolution record) realized: We believe that it yields
crucial evidence for estimating the frequency of animal life in the Universe.
WH A T A N I M A L S WE R E I N V O L V E D
I N T H E CA M B R I A N E X P L O S I O N ?
No one disputes that a huge diversity of large animals emerged with alacrity
between 600 and 500 million years ago. The event itself took place in the sea,
for the land areas of the time were largely barren except for lichens and perhaps
a few low plants; there were no trees, no shrubs, no stemmed plants at
all. Because of the lack of rooted vegetation, little soil would cling to the land
surfaces.
In the shallow seas and waterways, however, life was plentiful (though
clearly different from that in the seas of today, as we noted above) and was
rapidly changing in composition. Stromatolites, the layered bacterial forms
that had been the dominant type of life on Earth for most of the 4-billionyear
Precambrian era, were by 500 million years ago nearly absent from the
planet. There were literally being eaten out of existence, for a great biological
R A R E E A R T H
132
revolution was creating entire suites of organisms adapted for utilizing plants
as food. These newly evolved grazing animals (many looked like small
worms) used the stromatolites as food. After 700 million years ago, a steep
decline in stromatolite diversity took place, and newly evolved herbivores
were surely its cause, although these grazing creatures left no fossil record.
(They were too small and had no mineralized skeletons that could fossilize.)
In most instances we simply infer their existence.
Thus the stage was set for the great evolutionary drama we call the
Cambrian Explosion. It was grand theater, composed of four acts, each with
its own set of characters, although some of them hung around for successive
acts before exiting—by going extinct!
Act 1: The Ediacarans
The first act introduced a truly odd assemblage of creatures that looked like
bizarre jellyfish, mutated worms, and quilted air mattresses somehow brought
to life. This opening cast of characters is collectively known as the Ediacaran
fauna.
We now know that the Ediacarans opened Act 1 about 580 million
years ago and were largely gone by 550 million years ago (although a few appear
in much younger rocks). Most of the Ediacaran fauna somewhat resemble
members of the phyla Cnidaria and Ctenophorata—the jellyfish, sea
anemones, and soft corals of our world. Two of the most common types of
Ediacaran fossils resemble jellyfish and stalked, colonial sea anemone-like animals
known as sea pens (still quite common in our world), and they were first
interpreted as early versions of these modern forms. Other members of the
fauna were more worm-like in appearance, but these were minor players.
In some cases these are large organisms—some have left fossils nearly 3
feet long, making them veritable behemoths for their time. Yet they seemed
to have little organization of the sort we are so familiar with. For example,
they had neither an observable mouth nor an anus. Their organization suggests
a series of tube-like structures quilted together. In a 1988 essay, Stephen
Jay Gould proposes that these odd animals are indeed the flowering of the
The Enigma of the Cambrian Explosion
133
“diploblastic,” or two-cell-layer body plan, a type of body plan found today
only in the corals and jellyfish.
The Ediacarans were not discovered until the 1940s, when an Australian
geologist named R.C. Sprigg noticed some odd-looking fossil remains on scattered
slabs of sandstone mines in the Ediacaran Hills of southern Australia, a
desolate and isolated locality in very arid country. The fossils were simply impressions
in the sandstone, rather than preserved skeletons of any sort. Some
were worm-like; others looked like giant leaves; a third group were circular in
shape. Sprigg collected a few of these, noting that many of the circular impressions
made in the sandstone looked like modern-day jellyfish, also known
as Cnidarians. But such soft creatures as jellyfish are preserved in rock only
under the most extraordinary circumstances, and because of this, many wondered
whether these were fossils at all. Nevertheless, Sprigg briefly announced
his find in a scientific journal, described them as “among the oldest direct
records of animals in the world,” and noted that “they all appear to lack hard
parts and to represent animals of very varied affinities.” The fossils ranged in
length from less than an inch to more than 40 inches. Other fossils from this
region began to turn up (see Figure 7.1), and they eventually became the passion
of Australian paleontologist Martin Glaessner. He created the first biological
reconstructions of the odd Ediacaran fossils and made astute observations
about the nature of the environment in which these organisms lived.
Intensive study of the taxonomic affinities of this varied fauna soon followed.
Glaessner ultimately placed the entire Ediacaran fauna, as he called
them, into known phyla, such as the Cnidaria, a phylum thought to be among
the most primitive of all animals. To him, the Ediacarans thus represented the
first flowering of the animals and belonged to taxonomic groups still present
on Earth today. Using the tree analogy, Glaessner viewed his Ediacarans as
“missing links” between the small, presumably simple ancestors of all animals,
and the jellyfish and anemones still alive today. The Ediacaran world seemed
to resemble a Cnidarian world, and this agreed nicely with most biologists’
view of how the metazoan radiation may have unfolded, beginning with the
most “primitive” of phyla, the sponges and Cnidarians, followed only later by
more complex fauna such as arthropods (and the trilobites, which are members
R A R E E A R T H
134
Ernietta Pteridinium Dickinsonia Phyllozoon Spriggina Charniodiscus
Charnia
Rangea
I cm
Fractal quilting Serial quilting
Figure 7.1 Lifestyles of bilateral Vendobionta. Ediacaran fauna. Drawings by A. Seilacher.
of the phylum Arthropoda). According to this view, our modern-day animals
are descendants of the Ediacarans. This opinion is still held today by many
specialists, including Simon Conway Morris of Cambridge University.
In the nearly four decades since Glaessner’s interpretations were first
published, the Ediacaran fauna has achieved new importance. First, the
strange fossils making up this assemblage have been found beyond Australia.
In the White Sea region of Russia, Arctic Siberia, Newfoundland in Canada,
and Namibia in southern Africa, other fossilized examples of these strange
creatures are preserved, showing that the Ediacaran fauna was essentially
worldwide in distribution at the end of the Precambrian. (Many of these localities
have now been dated via radiometric geochronometry; some of the
oldest known on Earth are those at Mistaken Point, in Newfoundland, dated
as 565 million years old.) Second, the Ediacarans seem to have a longer stratiThe
Enigma of the Cambrian Explosion
135
graphic range than previously supposed, and in a few places on Earth, they
may actually have coexisted briefly with the undoubted animal faunas. Finally,
some workers believe that the Ediacaran fauna do not represent animals
at all but are large plants, fungi, or even lichens. Others class them as animals,
but animals belonging to taxonomic groups now extinct. The Ediacarans
have thus gone from logical precursors of the Cambrian Explosion to much
more controversial players in the evolutionary drama.
This latter view, that the Ediacarans are not a main branch of the tree
leading to animals, but rather represent a side branch now extinct (and thus
have nothing to do with the ancestry of all current animal life), has been most
effectively championed by paleontologist Adolf Seilacher of Yale and Tübingen
University. He suggests that the resemblance between the Ediacarans and
living creatures such as jellyfish and sea pens is coincidental. In his view, the
Ediacarans represent an extinct assemblage of organisms—a separate biological
“experiment” involving creatures with tough outer walls and fluid-filled interiors.
Seilacher has speculated that during the time of the Ediacarans, a thick
mat of bacteria covered the ocean bottoms, and this could answer the very
perplexing question of how organisms without hard parts became fossilized.
This bacterial mat may explain why the soft-bodied Ediacaran fossils
were so commonly preserved as fossils. As sand settled over the Ediacarans,
they were pushed downward into the bacterial mat. Their tough outer walls
could not be crushed readily, and the impressions they made in the mats were
preserved in three dimensions by the overlying sand. Seilacher suggests that
the evolution of new, efficient grazing animals such as mollusks at the start of
the Cambrian rapidly brought an end to such mats and changed the way sediment
accumulated in the earliest Cambrian.
Perhaps the most intriguing aspect of the Ediacarans is that there is no
evidence of predation on them; we have no records of Ediacarans preserved
with bite marks or missing pieces. Did these creatures live in a time without
predators—in a “Garden of Ediacara,” as paleontologist Mark McMenamin
has dubbed the time?
What actually happened to the Ediacarans? Conway Morris asks
whether they were diluted out of existence. That is, did the great radiation of
R A R E E A R T H
136
animals at the base of the Cambrian simply overwhelm them in the fossil
record? (In other words, they existed but were too few to fossilize.) Was the
disappearance of the Ediacaran body form a result of the first mass extinction
on planet Earth? Were they ecologically replaced or even preyed on by the
new animals—driven to extinction by efficient new predators against which
they had little defense? The fossil record is still enigmatic. In some places on
Earth, the Ediacarans are gone before the first “Cambrian” animals appear,
which suggests that the new animals were simply filling the niches of the extinct
Ediacarans. But as we have said, in other places there is a clear overlap
between the two, suggesting that a competitive interaction ensued.
The Ediacarans do make good theater—enigmatic, mysterious, and the
first on the stage. They were a hard act to follow, but a great wave of diversification
was occurring at the end of their time in the limelight, a wave that
continues on planet Earth still. With the second act of the Cambrian Explosion,
undoubted animals appear on the scene.
Acts 2 and 3: Trace Fossils and Small Shellys
We can combine the next two acts, because the cast of characters is both incomplete
and poorly characterized. In Act 2, a new group of players, seemingly
wearing masks to disguise their true identity, replace most of our opening
troupe. We detect them only by the footprints they left on the stage
itself, for we have no true “body” fossils (usually the remains of skeletal hard
parts). The second assemblage of life making up the Cambrian Explosion has
left only squiggles and tracks in the ancient sediment. Such fossilized remains
are known as trace fossils; they are not the remains of animals, but evidence of
their behavior, and thus record the trackways or feeding patterns of ancient
organisms. Yet they are of enormous significance. Whereas the Ediacarans
simply sat in place their entire lives, these first trace fossils tell us that large
animals capable of locomotion had appeared on Earth. Perhaps they were
large worms or flatworms, or perhaps they belonged to phyla now extinct.
The first and most primitive trace fossils appear in rocks as old as the Ediacarans,
but they diversify and take center stage in younger rocks. Trace fosThe
Enigma of the Cambrian Explosion
137
sils are still being formed today and have been common in the rock record
since the Cambrian. But they are clearly formed by many different organisms,
and it is doubtful that the organisms that formed the first trace fossils survived
into much more recent times than the Cambrian period itself.
Our Act 3 introduces an assortment of tiny calcareous tubes, knobs, and
twisted spines, none larger than about 1⁄2 inch, all coming from animals that
it is still impossible to reconstruct completely. Some are the remains of larger
skeletons that have been fragmented into pieces, but most are single elements
of some sort of a multielement skeleton, like individual spines coming from a
porcupine. Collectively, they are known as small shelly fossils, or SSFs. The
small shellys are first found in rocks dated to around 545 million years ago.
These extremely significant fossils tell us that another great biological breakthrough
had been achieved: The SSFs are the first large animals with mineralized
skeletons.
Act 4: The Trilobite Faunas
Act 4 of our play is a grand finale featuring fossil icons much more familiar to
us than the previous actors. They include the first trilobites, brachiopods, and
a host of newly evolved mollusks and echinoderms. The characters are now
far larger—and greater in number—than in any of the three previous acts,
and ironically, these actors were long thought to mark the start of the Cambrian
Explosion, rather than its end. This last group didn’t appear until about
530 million years ago. Its diversification proceeded for another 30 million
years. By about 500 million years ago, the Cambrian Explosion was finished.
The trilobites are by far the most diverse and obvious part of this assemblage.
The oldest trilobites, of which the genus Olenellus is diagnostic,
were spiny, somewhat resembled annelid worms, and had large crescentshaped
eyes. They all had walking legs and gills, and all appear to have fed by
ingesting sediment or particulate material on the sea floor. They showed little
adaptation for defense against predation.
Another curious group to appear contemporaneously with the trilobites
were immobile, coral-like animals called archeocyathids. This group had
R A R E E A R T H
138
conical skeletons made of lime and lived gregariously. They appear to have
been the world’s first reef-forming organisms and seem to have lived in the
same environments favored by corals today. In addition to being the first of a
long line of reef formers, the archeocyathids have another somewhat dubious
claim to fame: They were perhaps the first animal phylum to go extinct. The
basic body plan of the archeocyathid skeleton is unlike anything alive today.
Taxonomists place these creatures in the same phylum as the living sponges,
but this is as much for convenience as anything else. They appear to have
constituted a separate phylum—and one of the few phyla we know of to have
suffered utter extinction.
The remarkable Burgess Shale fauna of British Columbia has yielded extraordinary
insights into animals living among the trilobites. Because of the
lack of oxygen in this ancient environment, even soft parts were preserved,
and these remains offer us an unparalleled window into the past. The Burgess
Shale reveals how diverse the marine ecosystems were by the time trilobites
evolved. Yet by the time of the Burgess fauna, some 505 million years ago,
the majority of animal phyla appear to have been present.
WA S T H E CA M B R I A N
E X P L O S I O N I N E V I T A B L E ?
Darwin’s theory of evolution describes two of the most important scientific
discoveries ever made: (1) that all life has descended from a single common
ancestor, and (2) that the various species descending from this ancestral creature
have descended with modification. The great advances of physics and
chemistry are milestones in human understanding, but they do not themselves
describe life. We are life, and we have appeared on this planet through
the processes of evolution; it is a central law affecting us. Yet for all its importance,
the theory of evolution remains one of the most misunderstood of
scientific views. One popular misconception equates evolution with increasing
complexity and assumes that evolutionary change (Darwin’s “descent
The Enigma of the Cambrian Explosion
139
with modification”) always means an unbroken series of ever more complex
organisms or structures within organisms. Although greater complexity often
does evolve, it is not an end result of the evolutionary process; modification
can occur without increases (or decreases) in complexity. We have only to
look at the domains Archaea and Bacteria to see that this is true. From what
we can tell from their fossil record, the archaeans and bacteria are no more
morphologically complex now than they were 3.5 billion years ago (although,
as we have noted, their biochemistry has diversified almost endlessly).
They have evolved, to be sure, but that evolution has not involved
dramatic increases in morphological complexity.
Of the three domains of life, only one, Eucarya, has undertaken wholesale
experimentation in new morphology and body plans. If the process of
life’s creation were to be repeated innumerable times, it is not at all certain
that eucaryan equivalents (lineages exploiting the morphological route of
adaptation, rather than the chemical route utilized by the archaeans and bacteria)
would appear each time—or even ever again. But on this planet, at least,
the eucaryans did arise, and it was from this group that the multicellular animals
now dominating planet Earth arose. The pattern and timing of their evolution
on Earth may provide major clues to understanding whether, and how
often, equivalents of our planet’s complex animals could have arisen on other
planets.
There are important astrobiological implications in this: Will animal life
(or some other type of complex life) inevitably develop on all worlds in a
planetary habitable zone? In our estimation, it has always been assumed that
forming the first life was the hardest aspect, but that once life originated, it
inevitably proceeded “up” gradients of complexity, culminating in very complex
animals. Yet the actual history of life on this planet tells a different story.
The first life appeared about 4 billion years ago. Eukarytotic organisms did
not appear for another 1.5 billion years, and multicellular animals did not appear
until more than 3 billion years after the first life. On the basis of this information
alone, we would have to conclude that forming animal life is a much
more difficult—or at least a more time-consuming—project than the initial
R A R E E A R T H
140
formation of nonanimal life. Perhaps the timing observed on Earth was just
chance; perhaps on any number of other Earth-like planets with newly
evolved prokaryotic equivalents, animals would appear not billions, but millions,
of years after life originated. Abundant evidence from our planet’s history
casts doubt on this possibility, however.
On Earth it is clear that the evolution of animals occurred not as a gradual
process but as a series of long periods of little change, punctuated by great
advances. This pattern of evolutionary “thresholds” was succinctly described
by paleontologists Douglas Erwin, James Valentine, and David Jablonski in a
1997 article in American Scientist: “The fossil record of the last 3.5 billion years
shows not a gradual accumulation of biological forms, but a relatively abrupt
transition from body plans of single cells to those of a rich diversity of animal
phyla.” Evolution thus did not gradually create complex metazoans. They
evolved quickly, probably in response to a set of environmental conditions
quite different from those that allowed the evolution of life in the first place.
There were several of these “great leaps forward.” One was the evolution
of the eukaryotic cell type with its enclosed nucleus; another was the initial radiation
of the animal phyla, described in the last chapter. The most profound,
however, was the Cambrian Explosion, that short burst of evolutionary innovation
that resulted in the appearance of the larger, complex animals we believe
to be so rare in the Universe. In this single, approximately 40-millionyear
interval, all major animal phyla (all of the basic body plans found on our
planet) appeared, each represented by some number of species.
This event has profound implications for the possibility of life on other
planets. Is the pattern on Earth—a single, short-lived diversification of larger
animals—unique or the standard for all planets? And why did the Cambrian
Explosion not take place until 3 billion years after life’s first appearance on our
planet? Does evolution always require 3 billion years to transform a bacterium
into a multicellular animal, or was evolution simply waiting for the environment
to become conducive to the proliferation of animal life? This may be
one of the most critical questions facing the emerging field of astrobiology.
The Cambrian Explosion signaled a major change in the tempo of evolution
then prevailing on Earth. Prior to this, our planet’s most complex life
The Enigma of the Cambrian Explosion
141
Recent
Cenozoic
Cretaceous
Jurassic
Triassic
Permian
Carboniferous
Devonian
Silurian
Ordovician
Cambrian
65
250
530
300 250 200 150 100 50 0 50 100 150 200 250 300
Number of Families
Millions of Years Ago
Figure 7.2 Fluctuations in the number of families, and hence in the level of diversity, of well-skeletonized invertebrates
living on the world’s continental shelves during the past 530 million years are plotted by geological
epoch in this graph. Time proceeds upward.
consisted of algae, slime molds, and single-celled animals characterized by
low rates of evolutionary change. There was little morphological change, and
few new species arose over vast stretches of time (see Figure 7.2). The first
evolution of metazoans changed all this. The staid tempo of evolutionary
change that had characterized the first 3.5 billion years of life’s history shifted
into a higher gear. New species appeared at a far more rapid rate. They—
we—have been diversifying at breakneck speed ever since.
A study of the various animal phyla is thus the study of a few dozen stable
and long-lived body plans. This study has resulted in three great surprises.
The first was the recognition that evolution has produced only a relatively
few body plans. The discovery that the perhaps tens of millions of animal
species on Earth today belong to between 28 and 35 phyla was a major surprise
to nineteenth- and twentieth-century paleontologists and zoologists.
Why this number and not a hundred? Or a thousand? Or five, for that matter?
Diversification of the huge number of different species on Earth today (it
is estimated to be between 6 and 30 million) has been through elaboration
R A R E E A R T H
142
or convolution of simple and conservative structural designs. Astrobiologists
seek to discover whether this is how all animals (or their equivalents) evolve.
Or is this simply Earth’s way, and might there be worlds in space where there
are nearly as many body plans as there are species?
A second surprise, and perhaps the most astounding, was that virtually
all of the phyla appear to have originated no later than the end of the Cambrian
and that none have appeared since. This cannot be proved, for there are
some minor phyla (such as the rotifers) that have left no fossil record, and
perhaps some phyla originated after the Cambrian. But there are none that
we know of. For all the great changes that have occurred in the last 500 million
years, with all the evolutionary events and mass extinctions of that long
history, it would seem that at least a few new body plans would have appeared.
Yet the fact that every phylum with a fossil record is represented in
Cambrian strata makes such a supposition problematic.
The third surprise was that there may have been far more phyla on
Earth in the Cambrian than there are today. Fewer than 40 extant animal
phyla are recognized today. Yet according to some paleontologists, in the
Cambrian that number may have been as high as 100! Although the number
of species on the Tree of Life has been increasing through time, the number
of higher taxa, such as phyla, has been decreasing. Thus the tree keeps adding
ever more twigs and leaves on a dwindling number of major branches. Perhaps
the Tree of Life on some other planet is quite different, with great new
branches appearing continually through time.
WH A T—I F A N Y T H I N G—TR I G G E R E D
T H E CA M B R I A N E X P L O S I O N ?
Near the end of the last chapter, we pondered whether the initial diversification
of the animal phyla was stimulated by evolutionary or environmental
causes—specifically, the Snowball Earth events of between 800 and 600 million
years ago. This same question can be posed about the subsequent Cambrian
Explosion: Did it occur as late as it did in Earth history because it took
The Enigma of the Cambrian Explosion
143
that long for the establishment of an environment conducive to animals of large
size, most with skeletons, or because it took that long for the necessary genes
to evolve—genes that allowed the diversification of these metazoans? Genes
or environment? Much new research into environmental conditions before
and during the Cambrian Explosion, and into the physiological, anatomical,
and genetic innovations leading to larger multicellular animals, has given us
new views of this crucial moment in Earth history.
Many hypotheses have been proposed to account for the Cambrian Explosion.
These can be categorized as attributing it to environmental causes or
to biological causes.
Environmental Causes
• Oxygen reached some critical threshold value.
This is perhaps the most often discussed and widely favored of all environmental
hypotheses. According to this hypothesis, the amount of available
oxygen reached some critical level, or threshold, that made possible the
great diversification of new organisms. Presumably, this level was higher
than that during the first animal diversification event of 700 million years
ago. Many scientists suggest that the biological response to a new, higher
oxygen level was a biochemical breakthrough allowing animal life to construct
hard skeletons for the first time. In the absence of abundant oxygen,
organisms have a great deal of difficulty precipitating minerals as skeletal
structures. As early as 1980, Heinz Lowenstam and Lynn Margulis postulated
that skeletons in the form of collagen—an elastic, proteinaceous material
similar to human fingernails—could have appeared as early as 2 billion
years ago, for collagen formation does not require as much oxygen. Calcareous
and siliceous skeletons and shells, however, were not possible at that
early date.
• Nutrients became available in large amounts.
Just as a lawn needs fertilizer, ecosystems—and especially marine
ecosystems—need a supply of organic and inorganic nutrients to remain at
R A R E E A R T H
144
high levels of productivity and diversity. Abundant evidence suggests that
the late Precambrian interval witnessed a relatively sudden and dramatic increase
in nutrients, which may have had a significant effect on the evolution
of organisms.
One of the mineral types most commonly found in rocks of this age is
phosphorite, a mineral rich in phosphorus, one of the most important of the
inorganic nutrients necessary for life. (The others are nitrate and iron.) There
appears to have been a long interval during the late Precambrian when phosphates
and nitrates were unavailable to organisms of the time, because they
were buried in deep-water bottom sediments. However, the latest Precambrian
was a time of changing oceanographic conditions; episodes of upwelling
became common, in which deep waters were brought up to the sea
surface, in the process liberating nutrients formerly locked in bottom sediment.
This upwelling appears to be related to changing continental configurations.
The latest Precambrian was a time of intense plate tectonic activity. In
particular, a giant “supercontinent” named Rodinia began to tear apart, and as
this occurred, it changed the global patterns of ocean current circulation,
thereby triggering the upwelling. According to this hypothesis, the release of
the phosphate nutrients accompanying that new tectonic activity sparked the
Cambrian Explosion.
• Temperatures moderated following the late Precambrian “Snowball Earth” events.
Just as the evolution of humanity is set against the backdrop of a global
ice age, so too is the Cambrian Explosion associated with glaciation, for it
occurred soon after the cessation of the “Snowball Earth” events profiled in
the last chapter. As the last of these glaciation events finally ended, it signaled
a protracted warming of the planet that followed 200 million years of glacial
advances and retreats. Was this the trigger that unleashed the Cambrian Explosion,
as suggested in the last chapter?
• The Inertial Interchange Event
There is a final environmental possibility that borders on the fantastic—
as well as the believable. For many years, paleomagnetists have known that
The Enigma of the Cambrian Explosion
145
most, if not all, of the continents underwent large amounts of continental
drift during Cambrian time. As we will see in more detail in Chapter 9, continental
positions have an extraordinary effect on global climate, often controlling
where warm and cold currents flow, the formation of ice caps, and
even the abundance of greenhouse gases in the atmosphere. During the past
few years, advances in the numerical calibration of the Cambrian time scale
and improvements in the paleomagnetic database have revealed something
astounding: Much of this continental drift happened during the Cambrian
evolutionary explosion, and the entire episode lasted no more than 10 to 15
million years. The continental shifts were quite dramatic. North America
moved from a position near the south pole to the equator, and at the same
time, the entire supercontinent of Gondwanaland spun around a point in
Antarctica, sending North Africa from the pole to the equator as well. It is as
though the continents suddenly became ice skaters, gliding about Earth’s surface
with unprecedented ease for a short period of time, before turning to
stone once more.
In 1997 the ubiquitous Joseph Kirschvink and two colleagues published,
in the prestigious journal Science, a controversial interpretation of the
cause of this tectonic movement—an explanation that is either the harbinger
of a revolution in our understanding of planets and their histories, or unmitigated
balderdash. As of this writing, the scientific community is about evenly
divided and is awaiting further developments with keen anticipation.
Kirschvink, David Evans, and Robert Ripperdan proposed that the Cambrian
Explosion might have been triggered by another unique event in Earth history:
a 90-degree change in the direction of Earth’s spin axis relative to the
continents. Regions that were previously at the north and south poles were
relocated to the equator, and two formerly equatorial positions on opposite
sides of the globe became the new north and south poles. This interesting hypothesis
will be confirmed or discredited only through the acquisition of
much new paleomagnetic data.
Kirschvink and his colleagues noted that all of the world’s continents
experienced a major increase in continental plate motions (the “drift” movement
of continental drift) during the same short interval of time when the
R A R E E A R T H
146
great evolutionary diversification took place—between 600 and 500 million
years ago. This rapid movement of Earth’s upper surface relative to its interior is
thought to have been brought about by an imbalance in the mass distribution of
the planet itself. During this redistribution, the theory goes, all the solid portions
of Earth move together. But because Earth also has liquid portions (its inner
core, for instance), the outer layer essentially flips over relative to the spin of the
planet. This phenomenon would not be limited to Earth; it may also have occurred
on Mars. Kirschvink and his colleagues point to Tharsis, a large plateau
located on the Martian equator. Tharsis sits atop the largest gravity anomaly (a
center of high mass that creates forces of gravity greater than those in the surrounding
rock) known for any planet in the solar system—a place of such high
density that it creates a measurable perturbation in the planet’s gravitational
field. It is unlikely that Tharsis formed on the equator, according to Kirschvink
and his colleagues. They believe the law of conservation of mass caused it to migrate
later to its current equatorial position. Its movement would have been
caused by an “inertial interchange event” similar to that posited for the Cambrian
Earth. Once the volcano was at the equator, Mars would rotate so that its
maximum moment of inertia was aligned with the spin axis.
The Earth’s own inertial interchange event (IIE) would have taken only
about 15 million years, and it is the very rapidity of this movement that
causes Kirschvink and his colleagues to speculate that the IIE might have
been associated with the Cambrian Explosion of life. During this period, existing
life forms would have had to cope with rapidly changing climatic conditions,
such as polar regions sliding to the hotter equatorial zones and
warmer, low-latitude sites moving upward into the high-latitude, cold regions
of the planet. These motions would have disrupted oceanic circulation
patterns and would have perturbed most ecosystems on Earth. It might be
only by chance that a unique tectonic event during Earth’s 4.5-billion-year
history coincided with a unique biological event. But how often does someone
win million-dollar lotteries on two successive days?
The inertial interchange event can also explain one of the most curious
aspects of Earth at this time. It is well known among geologists that the late
The Enigma of the Cambrian Explosion
147
Precambrian and earliest Cambrian Earth underwent some sort of event that
is preserved as large swings in carbon isotopes. (These are chemical signals
found in the oceans in response to varying amounts of life on the planet; such
signals were used to detect the first life on Earth in the Isua strata of Greenland,
as described in Chapter 3.) About a dozen of these swings occurred near
the end of the Precambrian time interval, and they have long puzzled geologists.
The swings in these isotopes suggest that large amounts of organic carbon,
long buried in ocean sediment, were suddenly exhumed and reintroduced
into Earth’s carbon budget. Repeated, major changes in the oceanic
circulation patterns could produce these effects, yet such global changes
would require massive tectonic changes in short periods of time. These
changes would have fragmented ecosystems and could have prompted evolutionary
diversification. The inertial interchange event would accomplish this.
If the Cambrian Explosion was necessary for animals to become so diverse
on this planet, and if the inertial interchange event occurred as postulated,
and if the Cambrian IIE event contributed to the Cambrian Explosion
or even somehow was required for the Cambrian Explosion to take place,
then Earth as a habitat for diverse animal life is rare indeed.
Biological Causes
In his pivotal book Oases in Space, paleontologist Preston Cloud suggested that
there were four biological prerequisites for the Cambrian diversification
event to take place: the prior presence of life itself, the attainment of oxidative
metabolism (the ability to live and grow in the presence of oxygen), the
evolution of sex in the domain Eucarya, and the presence of an appropriate
protozoan ancestor to give rise to more complex animals. In Cloud’s view, attaining
all of these milestones took nearly 4 billion years—85% of Earth’s history.
He thus seems to believe that biological actors were more important in
creating the Cambrian Explosion event than were the environmental aspects
we considered in the previous section. But other biological factors must have
played a pivotal role as well.

R A R E E A R T H
148
• The advent of precipitated skeletons
Skeletons are critical to large body size for many animals. Skeletons
usually perform several functions, such as protection (from predation, desiccation,
and ultraviolet rays), muscle attachment (thus allowing locomotion),
and maintenance of body form. Yet building such structures required many
evolutionary breakthroughs. Oxygen levels would have been critical for two
reasons. First, large skeletons such as a shell covering (found in the earliest
trilobites and mollusks) restrict the access of seawater to the soft body parts.
In most early animals, respiration took place by direct adsorption of oxygen
from seawater, across the body wall. Second, the presence of a shell means
that a larger area of the body is no longer available for this type of respiration.
In low-oxygen conditions, animals have a difficult time getting enough oxygen
as it is, and adding a body cover only makes this problem worse. Thus,
skeletons such as shells would not have evolved until relatively high oxygen
levels were available in seawater.
Professor Adolf Seilacher (whom we met in our discussion of the Ediacaran
fauna) is convinced that acquisition of skeletons played the dominant
role in causing the sudden appearance of multicellular animal phyla. He notes
that hard skeletons are not simply additions to preexisting body plans. Their
very evolution modifies body plans. Seilacher argues that the Cambrian Explosion
was triggered not by environmental conditions that allowed larger animals
to develop but by those factors that allowed skeletons to appear. This is
a subtle but important distinction. With the ability to produce hard parts,
new animal groups could use these hard parts for jaws, legs, or body support,
and this enabled them to exploit entirely new ways of life and new environments.
• Attainment of evolutionary thresholds made large size possible.
A second possibility is that evolutionary breakthroughs allowed, for the
first time, the advent of large body size. We know that the majority of living
organisms up until this time were less than a millimeter long; most were far
smaller. Did genetic innovations allow larger body sizes and thus trigger the
Cambrian event? Examples of such innovations include more efficient organ
The Enigma of the Cambrian Explosion
149
systems, such as improved circulatory, respiratory, and excretory systems.
Each had to evolve before larger body size was attainable.
• The predation hypothesis
In 1972 paleontologist Steven Stanley (and later Mark McMenamin)
proposed that the evolution of predators played a part in stimulating the
Cambrian Explosion. Survival was enhanced in those animals that evolved
the ability to defend themselves against predators by producing shells, burrowing
deeply, or swimming or otherwise rapidly moving away from danger.
And these creatures incidentally found themselves in a position to exploit food
resources that had been underutilized or had not been utilized at all during
the Precambrian. The evolution of shells made possible new forms of filterfeeding,
and deep burrowing gave these animals access to new food resources.
Cambrian predators thus forced animals to undertake new lifestyles,
which turned out to be successful.
I S T H E CA M B R I A N E X P L O S I O N S I M P L Y
A N A R T I F A C T O F T H E F O S S I L R E C O R D ?
In the simplest sense, the Cambrian Explosion was a relatively sudden proliferation
of animal types. The number of new species involved in this event is
unknown, but it was several thousand at most and perhaps far less than this.
The extraordinary aspect of this event was that the new species were spread
among many new body plans. As we have said, each body plan defines a
higher taxonomic category, such as a phylum or class. The Cambrian Explosion
thus involved a large number of higher taxa, each of which was composed
of just a few species. But are we simply seeing the advent of effective
fossilization, rather than a real diversification “event”?
We recognize the Cambrian Explosion as such for a simple reason: We
see a large number of fossils suddenly appear in the fossil record. But are we
seeing a genuine flowering of new forms, or do the fossils merely mark the
first appearance of skeletons in groups that had already been long established?
R A R E E A R T H
150
In other words, is the Cambrian Explosion merely an artifact of a very imperfect
fossil record? Skeletons make fossilization possible; it may be that the actual
diversification of body plans that appears to mark the Cambrian Explosion
actually took place long before but is invisible to us because it took place
among small animals without skeletons, which left no fossils.
This latter view can be considered the null hypothesis. What if there really
wasn’t a Cambrian Explosion at all? It may be that the various animal
phyla accumulated in gradual fashion over the last billion years of the Precambrian,
evolving one from another, but doing so without leaving any identifiable
fossil record. It was thus only the evolution of large size, and of skeletons
allowing the preservation of fossils, that accounts for the “Cambrian
Explosion.”
Whether the Cambrian event included the diversification of body plans
or consisted simply of the first evolution, by these various body plans, of
skeletons and large size is a moot point. Something stimulated the evolution of
many large animals with skeletons in a brief period of geological time. Furthermore,
M. McMenamin and R. McMenamin, in their 1990 book The Emergence
of Animals, emphasize that mineralized skeletons—and especially
shells—profoundly influenced the evolution of new body plans. A wide variety
of forms use their shells not only for protection but also as an integral part
of feeding. Brachiopods and bivalves (both invertebrates with two shells) use
the shell as an integral part of a filter-feeding process. It is hard to see how the
basic body plan of each of these groups could have formed before shells did.
CA M B R I A N E X P L O S I O N ,
CA M B R I A N CE S S A T I O N
For all of the animal phyla to appear in one single, short burst of diversification
is not an obviously predictable outcome of evolution. Although there
was certainly a long (200-million-year?) interval from the appearance of the
first metazoans until the 20- to 30-million-year Cambrian Explosion, most of
their morphological diversification during this time, including acquisition of
The Enigma of the Cambrian Explosion
151
the skeletal parts so characteristic and diagnostic of many invertebrate phyla,
took place in a relatively short period.
Yet as wondrous and unexpected as that finding is, a second aspect of the
evolution of the phyla is equally puzzling. The Cambrian Explosion marked
not only the start of the majority of phyla as recognized in the fossil record but
also the end of evolutionary innovation at the phyla level: Since the Cambrian,
not a single new phylum has evolved. The extraordinary fact is that the diversification
of new animal body plans started and ended during the Cambrian
Period. Is this evolutionary pattern a characteristic of animal life on all (or
any?) planets that succeed in producing animals, or is it unique to Earth?
The lack of new phyla and the paucity of new classes after the end of
the Cambrian Explosion may again be an artifact of the fossil record; perhaps
many new higher taxa did evolve and subsequently went extinct. This seems
unlikely. It is far more likely that the great surge of innovation that marked
the Cambrian came to an end as most ecological niches became occupied by
the legions of newly evolved marine invertebrates.
Yet a puzzling mystery remains: Subsequent to the Cambrian explosion,
Earth suffered several major mass extinction events—short periods
when a majority of the species then living on Earth went extinct. These
events, profiled in detail in the next chapter, drastically reduced diversity.
The most catastrophic of these, the Permo-Triassic mass extinction of 250
million years ago, eliminated an estimated 90% of marine invertebrate
species, and thus provides a natural experiment that we can examine to understand
better those factors that caused the Cambrian Explosion. And what
we observe is that even after this major reduction in diversity, no new phyla
appeared. Although the number of species plummeted to levels similar to the
very low species diversity found early in the Cambrian, the subsequent diversification
in the lower Mesozoic involved the formation of many new species,
but very few higher taxonomic categories. The evolutionary events during
the Cambrian and the Early Triassic are dramatically different. Both produced
myriad new species, but the Cambrian event resulted in the formation
of many new body plans, whereas the Triassic event resulted only in the formation
of new species exhibiting body plans already well established.
R A R E E A R T H
152
Two hypotheses have been proposed to explain this significant difference.
The first supposes that evolutionary novelty comes about when ecological
opportunities are truly large. During the Cambrian, for instance, there
were many habitats and resources that had not been occupied or exploited by
marine invertebrate animals, and the great evolutionary burst of new body
plans was a response to these opportunities. This situation was not duplicated
after the Permo-Triassic mass extinction. Even though most species were exterminated
in this catastrophic event, enough representatives of various body
forms survived to inhabit most of the available ecological niches (even if at
low diversity or abundance) and, in the process, to discourage evolutionary
novelty.
The second possibility is that new phyla did not appear after the
Permo-Triassic extinction because the genomes of the survivors had changed
enough since the early Cambrian to inhibit wholesale innovation. In this scenario
the evolutionary opportunities were available, but evolution was unable
to create radically new designs from the available DNA. This is a sobering
hypothesis and one not easily discredited, for we have nothing to which to
compare the DNA we find in living animals. It could be that genomes gradually
become encumbered with ever more information—they gather more and
more genes—and in the process become less susceptible to a critical mutation
that could open up the way to innovation.
DI V E R S I T Y AND DI S P A R I T Y
One of the central (and controversial) aspects of the Cambrian Explosion—
especially with reference to the wondrous assemblage of fossils found in the
Burgess Shale localities in Western Canada (where not only early animals
with hard parts but also forms without skeletons are preserved as smears on
the rocks)—concerns what are called diversity and disparity. Diversity (or in
this case biological diversity) is a term familiar to most of us. Overtly, it is usually
understood as a measure of the number of species present. Biologists use a
more sophisticated meaning, that encompasses not only the number of
The Enigma of the Cambrian Explosion
153
species present in a locality, but also the relative abundance of those species.
For instance, in this more technical sense, an assemblage of organisms comprising
some given number of species, each with the same number of individuals
in each species, is considered more diverse than a second assemblage
with the same number of species but a highly unequal distribution of the numbers
of individuals that compose each species group. Disparity is a measure of
the number of body plans, types, or design forms, rather than the number of
species. This distinction, first articulated by paleontologist Bruce Runnegar,
seems at first glance rather subtle. Surely each different species has a somewhat
different body plan from every other, and thus disparity and diversity
should always be equal. But this is not the case. There are millions of species
on Earth today. Yet the number of general body plans is far less than this.
Among animals, the major body plans are found within the major evolutionary
lineages, the phyla. As we have seen, these animal groups all arose
in the Cambrian Explosion. Yet the surprising finding of paleontology is that
there were very few species in the Cambrian. In his 1989 book Wonderful Life,
Stephen Jay Gould describes this finding as “a central paradox of early life:
How could so much disparity in body plans evolve in the apparent absence of
substantial diversity in number of species?”
The history of diversity and disparity during the Cambrian Explosion
(or, more properly, creating the Cambrian Explosion) is another puzzling aspect
of planet Earth’s diversification of animals: Is this the only way to create
animals, or just one way? Will every planet with animals create them as ours
did, by evolving the entire spectrum of body plans in one great evolutionary
rush among a low number of species? Or could this process be more gradual,
with a slowly increasing number of species over long periods of time incrementally
enlarging the number of body plans?
The Burgess Shale is clearly of major importance in understanding the
initial diversification of animal life. It is largely responsible for showing us
that most or all of the various animal phyla (or major body plans) originated
relatively quickly during the Cambrian. But the Burgess Shale may also be
telling us that not only were the body plans found on Earth today around in
the Cambrian, but so too were other body plan types now extinct. One of the
R A R E E A R T H
154
central messages of Gould’s Wonderful Life is that the Cambrian was a time not
only of great origination but also of great extinction, for Gould (and others as
well) assert that far more phyla were present in the Cambrian than exist
today. How many were there? Some paleontologists have speculated that
there may have been as many as 100 different phyla in the Cambrian, compared
to the 35 still living today. Gould clearly believes that there were more
Cambrian than present-day phyla: “[W]e may acknowledge a central and surprising
fact of life’s history—marked decrease in disparity followed by an outstanding
increase in diversity within the few surviving designs.”
This view—so forcefully and elegantly described in Gould’s Wonderful
Life—is vigorously disputed in British paleontologist Simon Conway Morris’s
1998 book Crucible of Creation, about the Burgess Shale and Cambrian Explosion.
Conway Morris is, ironically enough, a central and sympathetic figure
in Gould’s book. He is one of the architects of our new understanding of the
Cambrian Explosion. But Conway Morris denies that disparity has been decreasing
since the Cambrian, citing several cases that suggest just the opposite.
He also attacks Gould’s metaphor of “replaying the tape” by showing
how convergence in evolution (where distinct lineages evolve similar body
types in response to similar environmental conditions) can produce the same
types of body plans from quite unrelated evolutionary lineages. Conway
Morris argues that even if the ancestor of vertebrates went extinct during or
soon after the Cambrian, it is likely that some other lineage would have
evolved the body plan with a backbone, because this design is optimal for
swimming in water. This view is quite antithetical to that espoused by Gould.
We thus have several models for diversification (see Figure 7.3), and how it
actually occurred on Earth is still in doubt.
A F T E R T H E CA M B R I A N E X P L O S I O N :
T H E E V O L U T I O N O F DI V E R S I T Y
Another aspect of the Cambrian (and one that we admit to rather cavalierly
omitting) is that not only did diversity and complexity of species increase
The Enigma of the Cambrian Explosion
155
Time
(A) (B) (C) (D)
Dispar i t y
Figure 7.3 Various interpretations of the history of life and its disparity. (A) The traditional view, whereby
disparity steadily increases through geological time. (B) The view presented by S.J. Gould, whereby maximum
disparity occurred in the Cambrian. (C) The view that disparity increased very rapidly in the Cambrian
and thereafter stayed much the same. (D) The view that disparity increased rapidly in the Cambrian and
since then has generally increased, though at varying rates. (From Simon Conway Morris.)
through time, but the ecosystems in which they lived changed as well. The
evolution and emergence of eukaryotic creatures, culminating in the Cambrian
Explosion, was accompanied by a shift from bacterial ecosystems to assemblages
far more diverse and complex. The dramatic decline of stromatolites,
the layered bacterial structures that were so common until a billion
years ago, may be evidence of this transformation from a prokaryotedominated
to a eukaryotic world. With the rise of animals, efficient herbivores
appeared, and the passive bacterial mats that we call stromatolites when
they are fossilized, served as food for the emergent herbivores.
The Cambrian was the time when the most profound of these changes
occurred. Yet it was not the last period of major diversification. Paleontologist
Jack Sepkoski of the University of Chicago has spent more than two
decades analyzing the diversity of organisms through time. He identifies two
significant episodes of diversification after the Cambrian: one in the lower
Ordovician (the period immediately succeeding the Cambrian) and one at
the start of the Cenozoic era, the time interval some 65 million years ago that
immediately followed the great extinction that wiped out the dinosaurs and
R A R E E A R T H
156
so many other species. The major unanswered question is whether the great
rise in species that characterized the last 500 million years was inevitable
once animals arose or was itself due to chance.
R E L E V A N C E TO T H E F R E Q U E N C Y O F L I F E
ON OT H E R P L A N E T S
Is the long wait for animals—the story of animal life on Earth—the exception
or the rule for any other planets with emerging life? Could either oxygenation
or the evolutionary steps necessary for completion of the complex animal
body plan have occurred more rapidly on some other planet, and if so,
under what conditions? The lesson of Earth’s Cambrian Explosion is that two
parallel preparatory steps must be taken if complex metazoans—animals—
are to appear. First, an oxygen atmosphere must be constructed. This is surely
the most critical environmental step. Second, a very large number of evolutionary
adaptations must be concluded to allow the evolution of an ocean
liner—our animals—from the toy sailboat—the bacteria—that began it all.
Both of these parallel tracks require time. There do not appear to be any
shortcuts. On Earth, one or both required several billion years. And during
that time, Earth had to maintain a temperature that allowed the presence of
liquid water and avoid what we might call “planetary disasters” of sufficient
magnitude to sterilize the evolving root stocks of animals. In the next chapter,
we shall see why no such disaster put an end to animal evolution on
planet Earth.
Mass Extinctions
and the
Rare Earth
Hypothesis
Much of the work we do as scientists involves filling in the
details about matters that are basically understood already,
or applying standard techniques to new specific cases. But
occasionally there is a question that offers an opportunity
for a really major discovery.
—Walter Alvarez, T. Rex and the Crater of Doom
Imagine that we are in a spaceship orbiting Earth 65 million years ago—
about 500 million years after the Cambrian Explosion described in the last
chapter—on the day that an asteroid enters the atmosphere and streaks
down toward what is now the Yucatan region of Mexico. We are about to
witness the collision that will eliminate the dinosaurs (and 60% of all other
species) from the rolls of the living.
157
8
R A R E E A R T H
158
The asteroid (perhaps it is a comet) is between 6 and 10 miles in diameter,
and it enters the Earth atmosphere traveling at a rate of about 25,000
miles an hour. At such speed the body takes 10 seconds to pass through the
atmosphere and then smashes into Earth’s crust. Upon impact, its energy creates
a non-nuclear explosion at least 10,000 times as strong as the blast that
would result from humankind’s entire nuclear arsenal detonating simultaneously.
The asteroid hits the equatorial region in the shallow sea then covering
the Yucatan and creates a crater as large as the state of New Hampshire.
Thousands of tons of rock from the ground-zero impact area, as well as the
entire mass of the asteroid itself, are blasted upward. Some of the debris goes
into Earth orbit, while the heavier material reenters the atmosphere after a
suborbital flight and streaks back to Earth as a barrage of meteors. Soon the
skies over the entire Earth glow dull red from these flashing small meteors.
Millions of them fall back to Earth as blazing fireballs and, in the process, ignite
the verdant Late Cretaceous forests; over half of Earth’s vegetation burns
in the weeks following the impact. A giant fireball also expands upward and
laterally from the impact site, carrying with it additional rock material that
clots the atmosphere as fine dust is transported globally by stratospheric
winds. This enormous quantity of rock and dust begins sifting back to Earth
over a period of days to months. Great dust plumes and billowing smoke from
burning forests also rise into the atmosphere, cloaking Earth in a pall of darkness.
From space we begin to lose sight of the surface of the planet and can
see only darkening gauze obscuring Earth’s once green and blue surface. It is
a vision from Dante’s Inferno, a nightmare of red fires and black soot.
The impact creates great heat both on land and in the atmosphere. The
shock heating of the atmosphere is sufficient to cause atmospheric oxygen
and nitrogen to combine into gaseous nitrous oxide; this gas then changes to
nitric acid when combined with rain. A prodigious and concentrated acid rain
begins to fall on land and sea, and before it ends, the upper 300 feet of the
world’s oceans are acidic enough to dissolve calcareous shell material. The
impact also creates shock waves spreading outward through the rock from
the festering hole in the crust; Earth is rung like a bell, and earthquakes of unMass
Extinctions and the Rare Earth Hypothesis
159
precedented magnitude occur. Huge tidal waves spread outward from the
impact site, eventually smashing into the continental shorelines of North
America, and perhaps Europe and Africa as well, leaving, when they recede,
a trail of destruction and a monstrous deposit of beached and bloated dinosaur
carcasses skewered on uprooted trees. The surviving scavengers of the
world rejoice. The smell of decay is everywhere.
For several months after this fearsome day, no sunlight reaches Earth’s
surface; the atmosphere is darker than the oil-fueled miasma that blanketed
Kuwait following the Gulf War. After the initial rise in temperature from the
blast itself, the ensuing darkness causes temperatures to drop precipitously
over much of the planet, creating a profound winter in a previously tropical
world. The tropical trees and shrubs begin to die; the creatures that live in
them or feed on them begin to die; the carnivores that prey on the smaller
herbivores begin to die. The Mesozoic era, which began 250 million years
after the Cambrian Explosion featured in the last chapter, comes to the end
of its nearly 200-million-year reign.
Following months of darkness, Earth’s skies finally begin to clear, but the
extinction—the death of myriad species—is not yet finished. The impact of
winter comes to an end, and global temperatures begin to rise—and rise. The
impact has released enormous volumes of water vapor and carbon dioxide into
the atmosphere, which now create an intense episode of greenhouse warming.
Climate patterns change quickly, unpredictably, and radically around the
globe before Earth’s temperature regains some equilibrium. From tropical to
frigid, then back to even more tropical than before the impact, all in a matter
of a few years. The temperature swings produce more death, more extinction.
All of this havoc creates death: the death of individuals, the death of
species, the death of entire families of organisms. This event is a planetary catastrophe.
Had the impacting object been only twice the size it was, it might
have sterilized the surface of planet Earth. It was a narrow escape for complex
metazoans.
Just 65 million years ago, such an impact event did end the Mesozoic
era, and it ended the Age of Dinosaurs as well. It was but one of many impacts
R A R E E A R T H
160
and other assorted global catastrophes that have imperiled complex life on
Earth over the past 500 million years. Such events must happen on planets
elsewhere in the Universe, and they would surely be the greatest obstacle to
the continued existence of any complex metazoan that might exist there. Extinction
events are an important aspect of the Rare Earth Hypothesis. Although
the animals and plants of Earth have suffered grievously in the assorted
mass extinction events through time, the damage could have been
worse—and on many other planets where life may have evolved, it probably
has been, or will be. If hit at an inopportune time, a planet’s higher life might
be snuffed out—or it might never be allowed to evolve in the first place.
As we saw in the last chapter, the Earth of 500 million years ago was teeming
with complex animals and plants. Attaining such a world, for the first time
populated by animals, required a large number of evolutionary and environmental
changes and took 3 to 3.5 billion years. Maintaining these organisms required
other conditions. Complex metazoans tolerate a far narrower range of environmental
conditions than do microbes; there are no extremophile or anaerobic
complex metazoans, for example. Complex metazoans are also far more susceptible
to extinction caused by short-term environmental deterioration.
DE F I N I N G MA S S E X T I N C T I O N S
The frequency of animal life in the Universe must be some function of how
often it arises and of how long it survives after evolving. We believe that both
of these factors are significantly influenced by the frequency and intensity of
what are termed mass extinctions, brief intervals when significant proportions
of a planet’s biota are killed off. There is no mystery about what kills organisms:
too much heat or cold; not enough food (or other necessary nutrients);
too little (or too much) water, oxygen, or carbon dioxide; excess radiation;
incorrect acidity in the environment; environmental toxins; and other organisms.
Mass extinctions occur when one or some combination of these factors
kills a significant percentage of the planet’s biota. There has been no shortage
of them in the past.
Mass Extinctions and the Rare Earth Hypothesis
161
Mass extinctions have the potential to end life on any planet where it
has arisen. On Earth there have been about 15 such episodes during the last
500 million years, 5 of which eliminated more than half of all species then inhabiting
our planet. These events significantly affected the evolutionary history
of Earth’s biota. For example, if the dinosaurs had not suddenly been
killed off following a comet collision with Earth 65 million years ago, there
probably would not have been an Age of Mammals, because the wholesale
evolution of mammalian diversity took place only after the dinosaurs were
swept from the scene. While dinosaurs existed, mammals were held in evolutionary
check. Mass extinctions are thus both instigators of and impediments
to evolution and innovation. Yet much of the research into mass extinctions
suggests that their disruptive properties are far more important than their
beneficial ones. If planets with life are gardens, then the mass extinctions are
the pests and droughts as well as, perhaps, the fertilizer. Yet as any gardener
knows, plants are most susceptible when they are young, and disasters are
most pronounced early in the growing season. A late frost, a catastrophic
hailstorm, the emergence of early spring pests, a lack of sun—all make the
early growing season the most hazardous time. So, too, with animal life on
any planet. We believe that the early period in the evolutionary history of
complex metazoans is by far the most dangerous interval. In our view, planetary
disasters (resulting in mass extinctions) that occur before the evolution of
complex metazoans and those that occur after they are established through
the process of species diversification are far less likely to end in the extinction
of all life. The fossil record of life on Earth supports this prediction in that the
Cambrian period, when complex animal life had recently evolved, shows the
most significant losses of higher taxa.
Unlike animals, which are fragile and easily killed, microbes are less
susceptible to mass extinction events. Once established as a deep microbial
biosphere, the bacterial grade of life is probably very difficult to eradicate.
Short of planetary sterilization through destruction of the planet by a supernova
or collision with a very large asteroid, the deep microbial biosphere of
any planet must act as an effective reserve of life, because the regions several
kilometers beneath the surface are insulated from even prodigious disasters
R A R E E A R T H
162
that affect the surface regions. Surface life, on the other hand (even bacterial
surface life), is surely susceptible to major planetary catastrophes, such as the
impact of truly large comets or asteroids. It may be that life on Earth’s surface
was repeatedly sterilized during the period of heavy bombardment about 4
billion years ago, only to be reseeded by the deep-earth microbes or by the
return of rocks ejected by the collisions. But for animal life, quite the opposite
is true. Animals are not capable of the safer subterranean existence or of
hibernating in the vacuum of space. If they are wiped out by catastrophe,
they cannot be immediately restocked from some underground reserve. They
have to evolve again in a slow, step-by-step process that lasts hundreds of
millions or even billions of years.
On every planet, sooner or later, a planetary catastrophe can be expected
that either seriously threatens the existence of animal life or wipes it
out altogether. Earth is constantly threatened by planetary catastrophe—
mainly by impact from comets and asteroids crossing the Earth’s orbit, but
also from other hazards of space. Yet it is not only the hazards of outer space
that threaten the diversity of life on this planet and on any others where it exists.
There are Earth-borne causes of catastrophe as well as extraplanetary
causes. Both types have brought about mass extinction on this planet in the
past and would be likely to do so on other planets as well.
T Y P E S O F P L A N E T A R Y DI S A S T E R S
The immediate, or direct, cause of all mass extinctions appears to be changes
in the “global atmosphere inventory.” Changes to the atmospheric gases
(which may be changes in volume or in the relative constituents of the atmosphere)
can be caused by many things: asteroid or comet impact, degassing of
carbon dioxide or other gases into the oceans and atmosphere during flood
basalt extrusion (when great volumes of lava flow out onto Earth’s surface),
degassing caused by liberation of organic-rich ocean sediments during
changes in sea level, and changes in the patterns of ocean circulation. The
killing agents arise through changes in the makeup and behavior of the atMass
Extinctions and the Rare Earth Hypothesis
163
mosphere or through factors such as temperature and circulation patterns
that are dictated by properties of the atmosphere.
Planetary disasters can occur for a great number of reasons. We shall examine
a few, in no particular order of importance.
• Changing a planet’s spin rate
We take Earth’s 24-hour spin rate for granted, when in fact it appears
to be unusual if we compare it to other planets and satellites in our solar
system. Jupiter and Saturn, for instance, each far greater in mass and diameter
than Earth, spin much faster. Many other bodies, however, such as
Venus and Mercury (and even our own Moon), spin much more slowly. In
lower-mass stars, planets in the habitable zone become “tidally locked” by
the gravitational force of the larger star or planet they revolve around.
When one side always faces the star in question, that particular face becomes
very hot, whereas the other side is always facing cold space and becomes
frigid. Either environment would be lethal to surface life and prevent
its evolution.
Planets can change spin rates, and when they do, any life already
adapted to a particular spin regime would be likely to face planetary disaster
because of the major temperature changes it would encounter. Earth itself has
been gradually slowing, a phenomenon that has probably altered the distribution
of cloud cover over time.
• Moving out of the animal “habitable zone”
Animal life needs liquid water, so it requires a mean global temperature
that allows liquid water to exist. Any movement of a planet out of an
orbit that allows such temperatures will create a planetary disaster.
Though such changes of orbit are unlikely, they could be caused by another
planet in a stellar system. Such perturbations would be common in
open star clusters.
• Changing the energy output of the sun (star)
Complex animal life on any planet is dependent on stellar energy. If
stellar output either increases or decreases such that liquid water can no
R A R E E A R T H
164
longer exist, the result will be disastrous to animal life or to the prospects for
its evolution. Short-term and long-term changes in stellar energy output may
be one of the most common forms of planetary extinction—and even sterilization.
Some scientists are convinced that the end of life on Earth will be
caused by an increase in the sun’s energy output. This is nothing new. As we
have seen, the amount of energy being produced by the sun—and indeed, by
most stars—increases over time. On Earth, the maintenance of a relatively
constant temperature has been attained through a gradual reduction in greenhouse
gases as the amount of energy from the sun has increased, thus keeping
temperatures in check. We seem to be nearing the end of this type of
planetary temperature regulation, however. There are now very small volumes
of carbon dioxide in the atmosphere compared to earlier periods of geological
time, and the sun’s energy output continues to increase. Some scientists
have predicted that temperatures on Earth will become too high for
animal life within several hundred million years from now. That event, when
it comes to pass, will produce the last greatest mass extinction on Earth, its
sterilization.
• Impact of a comet or asteroid
Any planetary system is rife with cosmic debris: asteroids and comets,
the residue left over from planetary formation. Great quantities of this material
will eventually strike all members of a planetary system, and the energy
released can spell planetary disaster. Such disasters are now known to
have caused mass extinctions on Earth. In 1980, Luis and Walter Alvarez,
Frank Asaro, and Helen Michel from the University of California at Berkeley
proposed that one of the greatest of all mass extinctions, the 65-millionyear-
old event that killed off the dinosaurs and many other species living
near the end of the Mesozoic Era, was caused by the impact of a large meteor
or comet striking Earth, as described at the beginning of this chapter.
As evidence for this view mounted, most scientists realized that collision
with a meteor or comet could cause a biotic crisis on any planet and that it
has done so at least once (and probably other times as well) during Earth’s
past (see Figure 8.1).
Mass Extinctions and the Rare Earth Hypothesis
165
Many variables affect the degree of lethality resulting from a collision,
such as the meteor’s size, composition, angle of impact, and velocity and the
nature of the impact target area. In the case of the Cretaceous event (also
known as the K/T impact), for instance, the target rock was rich in sulfur,
which exacerbated the impact’s environmental effects. (The sulfur reacted
with air and water to produce a highly toxic acid rain that lasted many
months after the impact event itself.) Moreover, not only the geology of the
impact site, but also its geography, may play an important part. An impact in
Equivalent size of bomb
(tons of TNT)
Approximate initial
meteor diameter
Every hour
Every day
Every year
Every century
Every millennium
Every million years
Every hundred
million years
Once in Earth's
history
10
kilotons
1
megaton
100
megatons
10,000
megatons
1,000,000
megatons
1 m 10 m 100 m 1 km 10 km
(10 m) (100 m) 1 km 10 km 100 km 1000
Crater diameter on Earth or Moon
Wipes out
most
species
Wipes out global
agriculture, disrupts
civilization
Devastates
continentscale
region
Meteor breaks up in Earth's
atmosphere and may make no
crater during a terrestrial impact
Effects
Frequency of impact on entire Earth
Figure 8.1 The rate of meteor impacts at the top of Earth’s atmosphere as a function of meteor size. Bottom
scale gives crater size for typical impact velocity of about 15 kilometers per second. Top scale gives meteorite
size and energy in terms of tons of TNT. Dotted line shows the scale of the Siberian meteoritic explosion of
1908. Dashed line shows the scale of the impact 65 million years ago that wiped out dinosaurs and other
species. (After Hartmann and Impey, 1994; 1993 data from E. Shoemaker, C. Chapman, D. Morrison, G.
Neukum, and others)
R A R E E A R T H
166
a low-latitude site will have entirely different consequences from a similar
body hitting a high-latitude site at a similar angle and speed, because the distribution
of lethality across the globe may be produced by atmospheric circulation
patterns. Finally, the nature of both the biota and the atmosphere at
the time of impact are surely important. An impact in a highly diverse world
of ecological “specialists”—animals and plants with little tolerance for environmental
change—might produce more extinction than the same event in a
low-diversity world of “generalists.” And an impact in a greenhouse world
might have different effects from one where greenhouse gas inventory or
oxygen content was lower than that on Earth today.
In the early years after the Alvarez hypothesis was advanced, some investigators
thought that a general synthetic model linking most or all mass
extinctions to impact would emerge. This was the thinking behind the
“Nemesis” hypothesis of astronomer Rich Muller from Berkeley, and it underlies
the work of David Raup and Jack Sepkoski of the University of
Chicago, who hypothesized in 1984 that mass extinctions show a 26-millionyear
periodicity. Since then, elevated levels of iridium (the platinum-group
element used by the Alvarez team as a sign of impact) have been found from
11 different time intervals in the geological record. Yet most of these are at
such low concentration that they are not indicative of larger impacts. The evidence
to date suggests that only the major mass extinction at the end of the
Triassic and that at the end of the Cretaceous Period (the K/T) were brought
about by the effects of impact.
The presence of numerous impact craters on every stony planet or
moon of the solar system is stark evidence of the frequency of these events,
at least early in the history of our solar system. It is probable that impact is
a hazard in most, or perhaps all, other stellar systems as well. Impacts are
probably the most frequent and important of all planetary catastrophes.
They could completely reset the course of the biological history of a planet
by removing previously dominant groups of organisms, thus opening the
way for entirely new groups or for the rise to dominance of previously minor
groups.
Mass Extinctions and the Rare Earth Hypothesis
167
• Nearby supernova
Another mechanism that could produce a mass extinction is the occurrence
of a supernova in the sun’s galactic neighborhood. Two astronomers
from the University of Chicago calculated in 1995 that a star going supernova
within 10 parsecs (30 light-years) of our sun would release fluxes of energetic
electromagnetic and charged cosmic radiation sufficient to destroy Earth’s
ozone layer in 300 years or less. Much recent research on ozone depletion in
the present-day atmosphere suggests that removal of the ozone layer would
prove calamitous to the biosphere and to the species residing within. A depleted
ozone layer would expose both marine and terrestrial organisms to potentially
lethal solar ultraviolet radiation. Photosynthesizing organisms, including
phytoplankton and reef communities, would be particularly affected.
Judging by the number of stars within 10 parsecs of the sun in the last 530
million years, and by the rates of supernova explosions among stars, astronomers
have concluded that it is very plausible that one or more supernova
explosions have occurred within 10 parsecs of Earth during the last 500 million
years. They also believe such explosions are likely to occur every 200 to 300
million years. The probability of nearby supernovae would be much greater
closer to the galactic center, as suggested in Chapter 2.
• Sources of gamma rays
Astronomers have detected sudden bursts of intense gamma radiation
being emitted from various galaxies (gamma rays are the most dangerous radiation
emitted by atomic bombs). Although very little is yet known about
these short but extremely violent releases of energy, they would be lethal to
any life on nearby planetary systems.
• Cosmic ray jets and gamma ray explosions
A new entry into the mass-death rogues gallery is lethal bursts of radiation
produced by violent stellar collisions. Cosmic ray jets and gamma rays
might both result from the same source: merging neutron stars. Astronomers
Arnon Dar, Ari Laor, and Nir Shaviv have postulated that cosmic ray jets may
account for several of the major mass extinctions and might explain the
R A R E E A R T H
168
rapid evolutionary events that follow them. They propose that high-energy
fluxes of cosmic rays follow the merger or collapse of neutron stars, themselves
the residues of supernovae. These explosions are the most powerful in the
Universe, releasing in a few seconds as much energy as the entire output of a
supernova. When two of these objects coalesce, they create a broad beam of
high-energy particles that, if it hit Earth, would be capable of stripping away
the ozone layer and bombarding the planet with lethal doses of radiation.
The frequency of these events is the critical issue. New calculations suggest
that these events may be both more frequent and more dangerous to life
in any galaxy than previously supposed. Chicago physicist James Annis proposed
in 1999 that gamma ray explosions are so lethal that a single such event
could obliterate life over much or all of an entire galaxy. Annis has calculated
that the rate of such explosions is about one burst every few hundred million
years in each galaxy. For instance, Annis suggests that if the energy from such
an event hit Earth, it would kill all land life on our planet, even if the explosion
occurred at the center of our galaxy. If such violent and dangerous collisions
are rare, they are but one more low-probability event. Yet both Annis and Dar
argue that such collisions occur relatively often and were even more common
earlier in the history of the Universe. They calculate that such effects would
cause a major mass extinction on Earth every hundred million years.
• Catastrophic climate change: Icehouse and Runaway Greenhouse
Under certain circumstances, radical changes in climate can cause mass
extinction. Major glaciations and greenhouse heating are examples, and both
depend on the amount of carbon dioxide or other greenhouse gases in the atmosphere.
These are the actual killing mechanisms brought about by the reduction
or increase of stellar output or by a planet’s orbit becoming either
closer to or farther from its sun. Climate changes intense enough to threaten
the biosphere with major mass extinction would involve great swings in mean
planetary temperature, as well as relocation of oceanic current systems and
shifts in planetary rainfall patterns.
The two most catastrophic such conditions can be called Icehouse (the
Snowball Earth events are examples) and Runaway Greenhouse. In both
Mass Extinctions and the Rare Earth Hypothesis
169
cases, global temperatures move outside the 0–100°C range that allows the
presence of liquid water on the planet. We will see possible examples of each
when we examine the fates of Venus and Mars in the next chapter.
• The emergence of intelligent organisms
There is abundant evidence that the emergence of humanity as a globally
distributed species armed with technology has triggered a new episode of
mass extinction on Earth. It can be argued that the emergence of any intelligent
species co-opting a planet’s resources in the service of advanced technology
and agriculture will necessarily cause a planetary mass extinction.
T H E F R E Q U E N C Y O F MA S S E X T I N C T I O N S
How often do mass extinctions take place? Perhaps the best way to address this
question is to use the same methods that meteorologists and hydrologists use in
assessing the risk from weather and floods. Many natural phenomena—such as
floods, earthquakes, and droughts—are distributed through time in a similar
way. Small events are common, large ones rare. The best way to see how frequent
the really rare events are, is to assemble all the available data and arrange
them by return time or waiting time. For instance, we might ask how often in a
century, or in a thousand years, a flood of some given intensity occurs? We can
then define “10-year” floods (events of such magnitude that we can expect one
every 10 years on average) and compare them to much larger, 100-year events
and the even more catastrophic 1000-year events. This does not mean that we
cannot get two 100-year events in successive years, only that the probability of
two such events taking place in successive years is vanishingly small. Hydrologists
use a technique called extreme-value statistics to extrapolate waiting times
beyond the length of historical records. These estimates are, of course, imperfect.
But they allow scientists to make estimates about, say, 1000-year events
when only 100 years of historical records are available.
Paleontologist David Raup has adapted this same technique to investigating
questions of mass extinction. Raup’s questions are very similar to those
R A R E E A R T H
170
posed by meteorologists interested in estimating how often giant floods may
take place. Raup notes that we have a good record for the past 600 million
years of Earth history, so we can define the 10-million-year and 30-millionyear
events with confidence. Using these statistics, Raup calculated what he
called a kill curve.
The kill curve is a graph showing the expected waiting times for mass
extinctions of varying magnitude. It depicts the average “species kill”—the
percentage of all species on Earth at a given time suddenly going extinct as a
result of some mass extinction event—for a series of waiting times. Raup’s
curve is not entirely theoretical; he derived it by first amassing the extinction
records of more than 20,000 genera of organisms on the basis of their actual
geological longevities. Raup used the Zoological Record, a compendium of the
life now on Earth, found the first occurrence and last occurrence of all of
these genera of organisms, and then tabulated the results in a colossal database.
His data are thus derived from our best information about the actual
geological ranges of the organisms he surveyed.
The kill curve gives us a sense of how many species go extinct over a
given period of time. According to this curve, there is negligible extinction in
the case of natural phenomena that occur about every 100,000 years. The
million-year event is more consequential, with perhaps 5% to 10% of all
species on Earth going extinct. That figure rises to 30% of all species at the
10-million-year event and to nearly 70% of all species at the 100-million-year
event. These are frightening numbers. If nearly three-quarters of all species
go extinct in a short-term planetary catastrophe of some sort every 100 million
years, it suggests that we are living on quite an unsafe planet.
Raup discusses that last concern in the 1990 book Extinction: Bad Genes or
Bad Luck. How often might we expect the event that kills off the entire biosphere
of the planet—the complete sterilization of Earth of all its huge diversity
of living things? “I once tried extreme-value statistics on extinction
data to ask, ‘How often should we expect extinction of all species on Earth?.’
I don’t have much confidence in the results, but they are at least comforting:
Extinctions sufficient to exterminate all life should have an average spacing of
well over 2 billion years.”
Mass Extinctions and the Rare Earth Hypothesis
171
Yet this should not be a comforting figure. Indeed, it goes to the heart
of the Rare Earth Hypothesis. If we might expect a planetary catastrophe to
exterminate all life on this planet every 2 billion years, and if life has already
lasted 4 billion years, we are truly pressing our luck! And luck may be just
what animals need to evolve for a long time when the grim reaper of planetary
extermination is put off only through blind chance.
T H E E F F E C T O F MA S S E X T I N C T I O N S
Are mass extinctions that stop short of complete sterilization necessarily
detrimental to planetary diversity? Perhaps it can be argued that instead of
being deleterious to diversity, they are actually forces that increase diversity.
For example, it can be argued that the various Paleozoic extinctions caused
archaic reef communities to be reassembled with more modern types of
corals. Mass extinctions paved the way for a takeover of bottom communities
previously dominated by brachiopods (archaic shellfish) by the more modern
(and more diverse) mollusks. In another case, the extinction of the dinosaurs
paved the way for the evolution of many new types of mammals, and it appears
that there are more types of mammal species than there were dinosaur
species. If these mass extinctions had not occurred, would planetary diversity
(the number of extant species) be higher or lower than it is today, other variables
(the history of continental drift, for instance) remaining the same?
We can illustrate the enigma of mass extinctions and their effect on
global biodiversity as follows: The Cambrian Explosion results in a sudden
rise in diversity, followed by an approximately steady state during the Paleozoic.
The mass extinctions during the Ordovician and Devonian cause shortterm
drops in diversity, but these are soon compensated for by evolution of
new forms. The great mass extinction that ended the Permian creates a
longer-term deficit in diversity, but eventually, in the Mesozoic era, it also is
compensated for. In fact, after every mass extinction that occurred on Earth
over the past 500 million years, biodiversity has not just returned to its former
value but has exceeded that value. Today, in our world, biodiversity is
R A R E E A R T H
172
higher than it has been at any time in the past 500 million years. If there had
been twice the number of mass extinctions, would there be an even higher
level of diversity than there is on Earth now? Perhaps mass extinctions exert
a positive effect, creating new opportunities and fostering evolutionary innovation
by weeding out decadent or poorly adapted but entrenched and
resource-hogging species. On the other hand, perhaps just the opposite is
true: If the mass extinctions had not occurred, biodiversity would be higher
than it is today (see Figure 8.2). How do we choose?
Interesting as this question is, it has not yet been tested in any way. The
fossil record, however, does yield some clues that mass extinctions must be
entered on the deleterious rather than the positive side of the biodiversity
Increasing number of
mass extinction events
(A)
(B)
Diversity of species
Low
High
0
Figure 8.2 Two models of how mass extinction affects diversity. (A) Increasing number of mass extinction
events leads to ever lower diversity. (B) There is some critical number of mass extinction events
that causes lower diversity.
Mass Extinctions and the Rare Earth Hypothesis
173

EXTINCTION SURVIVAL RECOVERY
Diversity
Time
Figure 8.3 Phases of a mass extinction crisis. (After Kauffman, 1986; based on information in
Donovan, 1989.)
ledger. Perhaps the best clue comes from the comparative history of reef
ecosystems following mass extinction. Reefs are the most diverse of all marine
habitats; they are the rainforests of the ocean. Because they contain so
many organisms with hard skeletons (in contrast to a rainforest, which bears
very few creatures with any fossilization potential), we have an excellent
record of reefs through time. Reef environments have been severely and adversely
affected by all mass extinctions. They suffered a higher proportion of
extinction than any other marine ecosystem during each of the six major extinction
episodes of the last 500 million years. Reefs disappear from the
planet after each mass extinction, and it usually takes tens of millions of years
for them to be reestablished. For instance, there were no reefs whatsoever
after the Cambrian, Ordovician, Devonian, Permian, Triassic, and Cretaceous
mass extinctions. When they do come back, they do so very gradually.
It appears that complex ecosystems take a long time to build and to rebuild
(see Figure 8.3). When reef systems eventually reappear, they are composed
R A R E E A R T H
174
of entirely new suites of creatures. The implication is that mass extinctions, at
least for reefs, are highly deleterious and yield net deficits of biodiversity.
R I S K AND CO M P L E X I T Y
Can we discern a relationship between the complexity of life and its risk of
succumbing to mass extinction? Recent evidence suggests that as organisms
become more complex, their risk of undergoing extinction increases. As an organism’s
complexity increases, so too does its fragility; hence increasing complexity
in most cases narrows the environmental tolerances of a living organism.
Any bacterium can withstand the rigors of outer space (at least for a short
period of time), but no animal can. As we move from bacterial life forms to
protozoa and then to metazoans, the range of temperature, food supply, and
environmental chemistry in which life can persist becomes more restricted.
This generalization seems to apply not only to individuals of a species
but also to the species itself. One of the strongest messages communicated by
the fossil record is that extinction rate is a function of complexity. On average,
simple animals are far more successful at avoiding extinction and thus
persist (in terms of geological time) far longer than complex ones; the simpler
the species, the longer its reign on Earth. Many bacterial fossils found in
3-billion-year-old rocks are identical to living forms found commonly on Earth
today. Are they the same species? Unless we can compare the DNA content
of the ancient form with that of its living analog, we cannot tell. But our best
guess is that they may indeed be the same species; they certainly have the
same external morphology. Simple bacterial species, once evolved, seem to
last a long time, perhaps because of their very simplicity and ability to adapt
without resorting to new body forms. Complex metazoans, on the other
hand, show far shorter ranges, and even among metazoans the inverse correlation
between complexity and evolutionary longevity seems to hold. For example,
mammals (the most complex animals on the planet) have average
species longevity only slightly greater than a million years, whereas bivalve
mollusks, which are far simpler, last an order of magnitude longer.
Mass Extinctions and the Rare Earth Hypothesis
175
But how can complexity be measured? Perhaps a bacterium really is no
simpler than a complex metazoan, and the relationship purportedly observed
is due to chance or something other than complexity. It turns out that there
are ways of comparing complexity. Determining the length of the genome
(and the number of genes it includes) is one such way, and an even easier
method of characterizing complexity in a metazoan is by describing the number
of different cell types it includes, as initially suggested by paleontologist
James Valentine.
Zoologists and physiologists have carried out the differentiation of cell
types in animals for years. A bacterium or a paramecium has only a single cell,
of course, but with the advent of animal life exhibiting multicellularity, various
body cells became specialized. A sponge, among the simplest of multicellular
animals, has at least four cell types: one for catching food, one for secreting
spicules (primitive supporting structures), one for transporting
material around the body, and one that acts as a type of skin cell. Vertebrate
animals such as ourselves have many more than this; mammals have in excess
of 100 cell types.
Oddly enough, no one has yet tried to correlate complexity as measured
by number of cell types with evolutionary longevity. The latter, also
known as extinction rate, has been calculated for most groups of animals and
plants by paleontologist Jack Sepkoski. Here we have combined these two
sets of data and searched for correlation. The results seem to substantiate the
view that complexity comes at a price of lower evolutionary longevity. This
finding suggests that ever more complex animal or plant species show ever
shorter evolutionary ranges, whether on Earth or elsewhere. And it suggests
that risk of extinction increases through time (see Figure 8.4).
T H E HI S T O R Y O F MA S S E X T I N C T I O N S
ON E A R T H : T E N E V E N T S
Paleontologists have discovered many mass extinction events that occurred
since the Cambrian Explosion (that is, in the past 540 million years). Yet
R A R E E A R T H
176
other mass extinction events of earlier times are largely unknown to us, because
they occurred when organisms rarely made skeletal hard parts and thus
rarely became fossils. Perhaps the long period of Earth history prior to the advent
of skeletons was punctuated by enormous global catastrophes that decimated
the biota of our planet—mass extinctions without record. Yet very little
attention has been paid to earlier extinction events. For instance,
astrobiologist James Kasting believes that the greatest mass extinction of all
time may have been brought about by the “Snowball Earth” events of around
750 million years.
First life
Prokaryotes
First Eukaryotes
First animals
High
Low
4 3 2 1 0
Time before present (billions of years)
Susceptibility to mass extinction
Figure 8.4 Hypothesized curve of extinction “risk,” or susceptibility, through time. Extinction risk
is highest soon after a new evolutionary type appears and then lessens as diversification occurs. Diversification
is insurance against extinction.
Mass Extinctions and the Rare Earth Hypothesis
177
Of the various mass extinctions known from the last 500 million years
(15 events over that time interval are officially classed as “mass extinctions”),
six have been especially catastrophic as measured either by the number of
families, genera, or species going extinct or by the effect on subsequent biotic
evolution. To this list we propose adding three more that occurred before
500 million years ago, as well as the current biodiversity crisis caused by the
effects of a runaway human population. This latter mass extinction event is
presently under way, so its final extinction total cannot yet be tallied. Nevertheless,
it may be representative of what happens whenever an intelligent
species arises on a planet with complex metazoans.
Our understanding of the various events is in inverse proportion to their
age: The older they are, the more mystery still surrounds them. The modern
extinction that is still under way will also be dealt with only briefly. The most
recent of the ancient events (the K/T event) is by far the most studied and
best known. Accordingly, we will discuss it in greatest detail.
• The bombardment extinctions, 4.6 to 3.8 billion years ago
The period of heavy bombardment is thought to have sterilized Earth’s
surface at least several times. There is no other information known.
• The advent of oxygen—Snowball Earth, 2.5 to 2.2 billion years ago
The rise of oxygen certainly doomed to extinction most anaerobic bacterial
species then on Earth. There is little or no fossil record of this phenomenon,
which may have coincided with the first “Snowball Earth” event.
• Snowball Earth events of 750 to 600 million years ago
We have almost no information about these events, which may have included
three or four separate extinctions coinciding with repeated glaciations.
There did appear to be wholesale extinction among stromatolites and
planktonic organisms called acritarchs. The lack of fossilized animals from
this time period obscures these events.
• The Cambrian mass extinctions, 560 to 500 million years ago
The extinctions that took place immediately before and then during the
Cambrian period remain the most enigmatic of all extinction episodes. We
R A R E E A R T H
178
believe that they are also the most important, in terms of their effects on animal
life on this planet.
As we saw in an earlier chapter, the “Cambrian Explosion” was the single
most significant event in the history of animal life on Earth. In a relatively
short interval of time, all of the animal phyla still on Earth appeared. Since the
end of the Cambrian, no new phyla have evolved. Yet however celebrated it
is for its period of diversification, the Cambrian was also a time of significant
extinction. It seems that some of the phyla that appeared in the Cambrian did
not survive for long. The Burgess Shale fauna described by paleontologist
Stephen Jay Gould contains numerous organisms that seem to belong to no
living phylum, and many paleontologists believe that many phyla went extinct
in the interval between 540 and perhaps 500 million years ago.
Some scientists argue that the Cambrian Explosion was preceded by the
first of all mass extinctions, which caused the disappearance of the Ediacaran
fauna. This fauna consisted of the odd assemblage of jellyfish and sea
anemone-like animals found in strata immediately below the base of the
Cambrian from many parts of the world. They appear to be the first diverse
assemblage of animals—perhaps early forerunners of familiar animals such as
the Cnidaria and various worms, or perhaps an assemblage of phyla now all
extinct. In any case, they disappeared in sudden and dramatic fashion immediately
before the Cambrian period. The disappearance of the Ediacaran animals
remains a mystery. It may have been brought about through competition
with the newly evolving and more modern groups of animals that typify
the Cambrian fauna, or they may have been driven into extinction by sudden
environmental change.
A second wave of extinctions occurred about 20 million years after the
Ediacaran crisis. This second crisis unfolded over several millions of years,
and it gravely affected the first reef-forming organisms (called archeocyathids),
as well as many groups of trilobites and early mollusks. Again, there
is little direct evidence to suggest what may have brought about these extinctions,
other than that they seem to be linked to changes in worldwide sea
level and to the formation of anoxic bottom water.
Mass Extinctions and the Rare Earth Hypothesis
179
The Cambrian extinctions remain an enigma (see Chapter 7). Here we
have consequential events without overt cause. How consequential they were
is discussed later in this chapter.
• The Ordovician and Devonian mass extinctions, 440 and 370 million years ago
During the Paleozoic era there were two other major mass extinctions.
About 370 million years ago, in the Devonian period, and more than 430 million
years ago, in the Ordovician, major mass extinction events decimated
the marine faunas of the time. Because our record of land life is poor for both
of these intervals, we have much to learn about the severity of these events on
land. However, it is clear that the majority of species in the sea went extinct.
Both extinctions eliminated more than 20% of marine families.
The causes of both of these extinctions remain obscure. Impact has
been proposed as the cause of the Devonian mass extinction, but in spite of
intensive searching, little evidence for such a cause has emerged. No evidence
of impact exists for the Ordovician extinction either. Anoxia, temperature
change, and change in sea level are the favored causes, but it is hard to
account for such extensive extinction on this basis alone. These are two major
extinction events in search of a cause.
• The Permo-Triassic event, 250 million years ago
On the basis of various measures of extinction (percent of existing
species, genera, or families eliminated worldwide during the event), the Permo-
Triassic mass extinction of 250 million years ago appears to have been the
most catastrophic of all mass extinctions that have occurred on Earth. Specialists
in compiling extinction records through time, such as Jack Sepkoski
and David Raup of the University of Chicago, point out that this particular
event stands alone in severity compared to all other such events. More than
50% of marine families died out, and this figure is more than twice that for
any other extinction. Estimates of the percent of species (belonging to various
families) that went extinct in this event vary from nearly 80% to more than
90%. It is clear that the vast majority of animal and plant life on the Earth was
extinguished.
R A R E E A R T H
180
Intensive research into the cause of this extinction over the past decade
has yielded a clearer view of what has been a perplexing problem: Although
it has long been recognized as the most catastrophic mass extinction, its
cause has been unknown. It now seems that although several causes contributed
to the event, the most important derived from a short-term degassing
of carbon dioxide from sediments sequestered on the ocean floor and
from volcanic gas emitted during unusually severe volcanic eruptions occurring
about 250 million years ago. A sudden release of huge volumes of carbon
dioxide directly killed marine organisms by carbon dioxide poisoning and indirectly
decimated terrestrial life via sudden and intense global heating. The
excess carbon dioxide released into the atmosphere greatly enhanced the
greenhouse effect, thus increasing the amount of heat trapped in the atmosphere.
A heat spike of perhaps 5 to 10 degrees and from 10,000 to 100,000
years in duration probably caused the terrestrial extinctions.
• The end-Triassic mass extinction, 202 million years ago
The end of the Triassic period witnessed a significant mass extinction in
which about 50% of genera were eliminated. There is still a very poor record
for the fate of land life at this extinction boundary, but it is clear that marine
life at this time was extensively and catastrophically affected. This mass extinction,
like the K/T event described at the beginning of this chapter, was
thought to have been brought about by the impact of a large extraterrestrial
body, either comet or asteroid. The Manicouagan Crater in Quebec is 100
kilometers in diameter (compared to the approximately 200-kilometer diameter
of the Chicxulub Crater associated with the K/T event). This crater has
been dated at 214 million years in age, which is older than the age of the Triassic/
Jurassic boundary. Environmental changes other than impact have been
associated with this extinction event as well, most notably oceanographic
changes creating anoxia in many shallow-water environments at the end of
the Triassic. It is difficult, however, to see how such changes could have affected
land life, which also suffered significant extinction at this time. The
probable cause remains unknown.
Mass Extinctions and the Rare Earth Hypothesis
181
• The Cretaceous/Tertiary boundary event, 65 million years ago
The mass extinction of the dinosaurs, as well as 50% or more of the
other species then on Earth, has been recognized for more than a century and
a half as one of the most devastating periods of mass death in Earth’s long history.
Although numerous explanations for this event have been proposed, an
asteroid impact cause is now largely accepted.
In 1969, Dr. Digby McLaren of the Geological Survey of Canada proposed
that a largely unrecognized mass extinction had exterminated much
sea life about 400 million years ago. McLaren suggested that this extinction
was brought about by environmental consequences following the collision of
a large meteor with Earth. At that time, McLaren had no actual proof of the
impact, such as a suspicious-looking crater of the correct age. Nobel Prizewinning
chemist Harold Urey of the University of Chicago also proposed
that the great mass extinction that ended the Mesozoic era resulted from a
comet, the size of Halley’s, colliding with Earth. This proposal too was disregarded
for lack of evidence. Soon afterward, however, the Alvarez team at
Berkeley repeated the claim, this time providing a wealth of scientific evidence.
The Alvarez hypotheses of 1980 that (1) Earth was struck some 65 million
years ago by an Earth orbit–crossing object of some sort (asteroid or
comet), and (2) the environmental effects stemming from that impact
brought about a mass extinction are today widely (though not universally)
accepted. What is unequivocal, however, is that these groundbreaking hypotheses
changed the landscape of paleontological research and are of central
importance to astrobiology as well. The central paradigm shifts were that
asteroid and comet impacts can cause mass extinctions, and that mass extinction,
through whatever cause, can occur relatively rapidly.
Like most great science, the Alvarez theory is conceptually simple.
Judging by the concentrations of platinum-group elements found in three
Cretaceous/Tertiary boundary layers (or K/T boundary layers, as they came
to be called) from Italy, Denmark, and New Zealand, the Berkeley group proposed
that an asteroid at least 6 miles in diameter struck Earth 65 million
R A R E E A R T H
182
years ago and that the environmental aftereffects of this impact caused the
great mass extinction. The ultimate killer, according to the Berkeley group,
was a several-month period of darkness, or blackout, following the impact.
The blackout was due to the great quantities of meteoric and Earth material
thrown into the atmosphere after the blast, and it lasted long enough to kill
off much of the plant life then living on Earth, including the plankton. With
the death of the plants, disaster and starvation rippled upward through the
food chains.
By 1984, high iridium concentrations had been detected at over 50
Cretaceous/Tertiary boundary sites around the world. Shocked quartz
grains were also discovered from many Cretaceous/Tertiary boundary sites.
Fine particles of soot were found disseminated in the same K/T boundary
clays. The type of soot found comes only from burning vegetation, and the
quantity of soot ultimately found in boundary clays from many parts of the
globe suggested that at some time about 65 million years ago, much of
Earth’s surface was consumed by forest and brush fires. A truly horrifying vision
emerged: Soon after the impact, the majority of plants then on Earth
apparently burned. Some of the fires may have started from the fireball and
searing heat produced by the impact, but the majority were probably set
days later as rocky fragments, initially blown into orbit by the explosive
force of the titanic impact, streaked back to Earth as bright missiles of destruction.
There was but a single comet or asteroid strike. The impact created a
crater between 180 and 300 kilometers wide and now named Chicxulub. The
impact target geology (especially the presence of sulfur-rich sedimentary
rocks in the target area) may have maximized subsequent killing mechanisms.
Worldwide changes in the atmospheric gas inventory, temperature drop, acid
rain, and global wildfires are all proposed killing mechanisms.
The litany of disasters related above is enough to make us wonder how
any complex metazoans survived this event. Given an impact object of sufficient
size, it is clear that impacts such as that which caused the Cretaceous/
Tertiary mass extinction are capable of killing off all complex metaMass
Extinctions and the Rare Earth Hypothesis
183
zoans on a planet. If the object that caused the K/T extinction had been only
twice the size it was, that might have been Earth’s fate.
• The modern extinction
Suggesting that we have entered a mass extinction episode would have
been controversial at best during the 1980s, but most investigators now concede
that the number of extinctions that have occurred since the end of the
last glacial period, some 12,000 years ago, clearly defines the Holocene time
interval as one of pronounced and elevated extinction rates. There are many
estimates of how many species are currently going extinct each year, though
there are few hard data for many regions. What is clear is that the world’s
forests are being felled inexorably to make way for agriculture and that the
removal of forests leads to extinction. At the moment the seas are more insulated,
and there is little evidence of major extinction occurring there at present,
although this could change quickly as pressure on the world’s fish stocks
increases. As runoff and chemical pollution increase over the next several
centuries, the extinction rate in the sea may rise substantially. Estimates of
the faunal tally vary, but all carry the grave message that Earth is losing a
great number of species rather quickly. Perhaps the most sobering estimate
comes from Peter Raven of the National Academy of Sciences, who has suggested
that two-thirds of the world’s species may be lost by the year 2300.
The ultimate cause of this extinction is the runaway population of Homo
sapiens.
CO M P A R I N G T H E S E V E R I T Y
O F T H E MA S S E X T I N C T I O N S
The conventional means of comparing the severity of the various mass extinction
events has been to compute the percentage of taxonomic categories that
went extinct. This monumental work has been carried out largely by paleontologists
at the University of Chicago, through literature research initiated by
R A R E E A R T H
184
David Raup and Jack Sepkoski. The first tabulations were of families of marine
animals. Several years later Sepkoski tabulated the number of genera
going extinct, and he is now working on the number of species. It was
through the use of such statistics that the “Big Five” (the Ordovician, Devonian,
Permian, Triassic, and Cretaceous extinctions) were differentiated from
other Phanerozoic extinctions. If the number of genera going extinct is used
as a means of comparing the mass extinctions, then the Permo-Triassic event
was the most catastrophic, followed by the Ordovician, Devonian, Triassic,
and end Cretaceous. The Cambrian extinctions do not appear to be so
“major.” Sepkoski’s most recent compilation (which he kindly sent to us as a
personal communication in 1997) yields per-family extinction rates of 54%
for the Permian, 25% for the Ordovician, 23% for the Triassic, 19% for the
Devonian, and 17% for the famous K/T extinction. In two other analyses,
however, performed separately by paleontologists Helen Tappan and Norman
Newell, the Cambrian extinctions of marine families exceed those of the
Permian. The Cambrian extinctions are the most consequential of all; about
60% of marine families went extinct in the Cambrian compared to about 55%
in the Permian.
These results pose a dilemma. In the Permian period, the cause of the
great mass extinction is quite clear. The continents coalesced into one large
supercontinent, greatly affecting worldwide climate and temperature in the
process, and near the end of the period an additional, sudden, and catastrophic
event occurred as well: Enormous quantities of carbon dioxide were
released into the seas and atmosphere, causing a sudden and deadly rise in
global temperature. In the Cambrian, however, we have no overt cause. Our
best guess is that the garden analogy holds true here. The garden of animals
during Cambrian time had only just emerged, and although there were many
different types or body plans (more so than now, in fact), there were very few
species in each category. Even slight changes in environmental conditions
were sufficient to wipe out entire categories, entire phyla. The Cambrian was
the riskiest period of all time for animal life. Its extinctions, in our view, were
the most important in the history of life on Earth, so the Cambrian should be
judged as more important than any other such events, including the PermoMass
Extinctions and the Rare Earth Hypothesis
185
Triassic extinction. There is more to comparing extinctions than simply
adding up the number of species killed off.
E X T I N C T I O N R I S K THROUGH T I M E
Do the risks of extinction vary through time? This question involves two
variables: (1) Do the environmental conditions that affect life on a given
planet change through time? (2) Does the susceptibility of life to extinction
change as its evolution progresses? Just as we have had to modify the concept
of the “habitable zone,” this question cannot be answered without some qualification
related to degree of complexity, for the extinction rates of microbes
are far different from those of more complex forms. If, as we surmise, extinction
risk or rate varies with an organism’s complexity, we might expect rather
low extinctions during the long period prior to the evolution of animal life,
followed by increasing extinction rates (the percentage of the total biodiversity
or species richness going extinct during any interval of time) as animal
life evolves. Yet, as we have said, this question involves two variables, the second
being the possibility that the frequency or intensity of mass extinctions
may change as well. There is much evidence that the proximal causes of
extinction—the planetary catastrophes that cause the mass extinctions in the
first place—vary through time as well.
The history of major mass extinctions on Earth suggests that only two
causes have operated in Earth history: impact and global climate change.
There are other phenomena that may also have caused extinctions, such as
nearby supernovae, but we have no credible evidence that this latter mechanism
has indeed occurred. With regard to impact, there is good evidence that
the frequency of impact has changed through time, as we saw earlier in this
chapter. The most obvious of these changes was the cessation of major impacts
during the period of “heavy bombardment” that lasted from approximately
4.3 to 3.8 billion years ago. But even after this rain of major comets
ceased, there is evidence of a long, slow decline in impact rates, as documented
by Richard Grieve and others. This decrease would have reduced the
R A R E E A R T H
186
overall extinction rate during the same period in which the vulnerability of
emerging animal groups was increasing. One could argue that even in the last
500 million years, the time of complex animals, there should have been
enough comet or asteroid strikes to exterminate animal life on this planet.
That has obviously not happened.
T H E T R A D E - O F F O F CO M P L E X I T Y :
R I S K AND DI V E R S I T Y
With complexity can come diversity. Although organisms of greater complexity
are more susceptible to extinction, their defense seems to lie in numbers.
We are just beginning to study their forms in detail, but it does appear
that the number of bacterial species is far lower than the number of insect
species. Yet if complexity brings with it the price of greater vulnerability to
extinction, how have complex animals and plants survived as long as they
have on this planet through the various mass extinctions? There must be
some aspect of complex animals and plants that helps protect them from
extinction—not as species, but as higher taxa—categories above the species
level.
The history of animal phyla during and after the Cambrian Explosion
described in Chapter 7 is an example of this. Some paleontologists have suggested
that as many as 100 animal phyla may have evolved during the Cambrian
period (although the consensus seems to be far fewer than this). Some
of these phyla went extinct during the Cambrian or at its end. Since that time,
not a single phylum has gone extinct. It is probably not a simple case of weeding out
the bad from the good, where the survivors were those body plans best suited
for our world. Rather, it appears that the surviving phyla have endured subsequent
planetary disasters by having large numbers of species. As long as a
single species survives, the phylum survives and is in a position to rediversify.
In the Cambrian, on the other hand, all phyla contained just a few
species each; the Cambrian disasters eliminated whole phyla because there
was such low species-level diversity within the various phyla. As far as aniMass
Extinctions and the Rare Earth Hypothesis
187
mals are concerned, the Cambrian (or just before) was the most dangerous
period in the long history of Earth. Since that time, millions of species have
evolved in the various phyla, making them far more “extinction-proof.”
Diversity—the stocking of body plans through the evolution of numerous
species—may be the best protection against extinction.
How do higher animals and plants create and then maintain this diversity?
First, they have evolved rapid (compared to bacteria) speciation rates.
Because higher organisms use sexual reproduction as the dominant means of
reproducing, they create a great deal of variability within populations, upon
which natural selection can act. Speciation (the creation of new species) occurs
when small populations split off from larger parent populations and can
no longer exchange genes. Gradually, adaptation of the smaller population
makes it sufficiently distinct from the parent population to prevent successful
interbreeding if the two populations come in contact again.
This process is critical to maintaining diversity. Constant formation of
new species is needed if diversity is to persist, for there is a relatively high rate
of species extinction in all groups of animals and higher plants. Since the
Cambrian Explosion, the engine driving the creation of new species has been
causing the diversity of complex life on Earth to increase, though long-term
gains have been periodically—and temporarily—reversed by the various
mass extinctions. During the mass extinction events, it is diversity that saves
higher taxa. If a given phylum or major body plan is sufficiently diversified—
is represented in many forms living in many different environments—it has a
high likelihood of withstanding the extinction event.
A P L A N E T ON T H E B R I N K
The “close calls” of planetary catastrophe are repeatedly written large in the
rock record of Earth. We have teetered on the brink many times. In this chapter
we have recounted the obvious near misses, as manifested by the various
mass extinctions. Yet a more subtle if no less sobering record can be found in
the changes in atmospheric composition through time.
R A R E E A R T H
188
It is clear that the levels of carbon dioxide and oxygen have changed
dramatically during the Phanerozoic (the last 530 million years), and these
changes may themselves have been trivial compared to the longer but less
readily sampled Precambrian interval of time. The ecological effects of these
changes are very poorly understood. During the Paleozoic Era, CO2 values
20 times that of the present day are now confidently inferred to have been
present in the lower Paleozoic, followed by a rapid decline in the Permo-
Carboniferous time. The world then underwent a massive glaciation, as its
greenhouse conditions gave way to far cooler climates.
Changes in the amount of oxygen in the atmosphere have also been
profound, but they are far less well documented (or understood) than for carbon
dioxide. Estimating ancient oxygen levels is fraught with uncertainty:
One scientific group extracts oxygen from ancient amber, but others decry
this method as yielding totally erroneous results. Without direct readings, we
must estimate the ancient oxygen values in the atmosphere by studying the
rate of organic carbon burial in sediment through time or by studying rates of
rock weathering. Neither approach is very satisfactory. Yet if these methods
have any validity, they tell us that the amount of oxygen has varied over the
last 500 million years. For instance, there appear to have been far higher levels
of oxygen in the atmosphere around 400 to 300 million years ago; levels
as high as 35% by volume (compared to 21% today) are possible. This much
oxygen would have made forest fires far more common and more devastating.
There were also periods of depressed oxygen, and it is not too farfetched to
envision scenarios where lowered oxygen levels had large-scale effects, perhaps
even inhibiting the evolution or development of certain forms, such as
those with very high metabolisms requiring abundant oxygen.
Thus even on a planet such as ours, the most important of all systems for
life—the atmosphere—can be unstable over an interval 100 million years
long. Planetary atmospheres can change enough to cause mass extinctions,
and maintaining an atmosphere conductive to animal life for the staggering
periods of time necessary for animal life to evolve and diversify may be the
most difficult feat of all.
Mass Extinctions and the Rare Earth Hypothesis
189
A MO D E L O F P L A N E T A R Y E X T I N C T I O N
We can summarize the implications of Earth’s history of mass extinctions
with regard to the Rare Earth Hypothesis as follows. Mass extinctions probably
occurred rarely during the long period in Earth history when life was
only of a bacterial grade. With the evolution of more complex creatures, such
as eukaryotic cells, however, susceptibility to extinction increased. With the
advent of abundant complex animals in the Cambrian, vulnerability to mass
extinction may have reached a peak, because diversity was very low. As more
and more species evolved within the various body plans, susceptibility to extinction
decreased again.
On any planet, the number of mass extinctions may be one of the most
important determinants of where animal life arises and, if so, how long it lasts.
In planetary systems with large amounts of space debris—and thus a high impact
record—the chance that animal life will arise and persist will surely be
much lower than in systems where impacts are few. In similar fashion, inhabiting
a cosmic neighborhood where large amounts of celestial collisions, supernovae,
gamma ray bursts, or other cosmic catastrophes occur will also reduce
a planet’s likelihood of attaining and maintaining animal life.
It appears that the best “life insurance” is diversity. In the next chapter
we will document our view that plate tectonics, also known as continental
drift, has been the main process promoting high diversity among animals on
Earth.
The
Surprising
Importance
of Plate Tectonics
Imagine that we have a spacecraft capable of swiftly taking us to each
planet in the solar system. Our goal on this voyage is to try to determine
what features of Earth are essential to animal life. On our voyage, therefore,
we are looking for some clue to why animal life has been able to survive
on Earth for a time period approaching a billion years. What are those factors
that have fostered diversity on Earth?
We begin our celestial survey with Mercury, a cratered world of great
heat on the sunlit side and great cold on its dark side of the slowly spinning
planet. Yet we quickly find that Mercury is not only free of atmosphere, liquid
water, and life but is also volcanically dead. Its surface shows mainly numberless
craters of a meteor-ravaged world, scars left by the bombardment by
comets and asteroids. In contrast to Earth, in the 4 billion years since that
time of stony rain, little of geological importance has happened on this
planet. Mercury looks like our Moon.
191
9
R A R E E A R T H
192
Next we travel to cloud-covered Venus. Its surface looks curiously
young, like the face of a child, yet Venus is the same age as Earth. We find
that the crust of Venus appears to have been geologically “resurfaced” in
some sort of cataclysmic event that caused its surface to melt sometime in the
last billion years. Because of this, the numerous craters we saw on Mercury
are less common. But Venus has two other prominent geological features:
crustal plateaus and volcanic rises that look something like a string of volcanoes
whose cones have been lopped off. There is no animal or plant life and
no oceans or liquid water of any kind. The surface of Venus is simply too
hot—hot enough to melt solder.
Mars is the next leg of this long voyage, and as we orbit the red and
ochre planet, we see astounding volcanoes rising high above a cratered and
rock-strewn landscape. These volcanoes (the largest in the solar system) are
enormous by Earth standards. But they are relatively few in number—
solitary, lonely sentinels dispersed across the planet’s face. Curiously, there
are no other mountains, no equivalents of the Alps, or even the Appalachians.
And there are no seas, no lakes, no rivers, and no liquid water, although many
geomorphic features of the planet’s surface indicate that water was present
here long ago.
With Mars we have finished our survey of the so-called “terrestrial”
planets. We have already learned much: No other planet has linear mountain
chains. Now we travel toward the outer regions of the solar system, arriving
in the realm of the gas-giant planets. First we pass Jupiter, with its
writhing, multicolored atmosphere and the distinctive Great Red Spot racing
around this rapidly rotating colossus of a planet. There are no land features,
for there is no distinct planetary surface, no place where the atmosphere
ends and land begins. Jupiter is unsuitable for animal life (as we know
it, anyway) because it has no solid planetary surface. Perhaps bacteria-like
organisms live within its roiling atmosphere, perhaps not. Its satellites, however,
might be places where life has arisen and survived, so we swing by each
of the four large “Galilean” moons (so named because they were first seen by
the great Italian astronomer Galileo): Europa, Callisto, Ganymede, and Io.
The Surprising Importance of Plate Tectonics
193
Each is somewhat smaller than Earth, and all have frozen surfaces (although
Io has active volcanoes). There are no animals or even liquid oceans here, although
the frozen ocean of Europa seems a tantalizing possibility for life
simply because liquid water may lie deep below its ice-covered surface.
Ganymede and Callisto are also likely to harbor subterranean regions of liquid
water or brine.
From Jupiter we travel on to the other gas giants of the solar system:
Saturn, Uranus, then Neptune. Like Jupiter, each is a great gas ball without
any definable surface, but each has smaller, rocky satellites orbiting it, some
cratered, some ice-covered. None has animal life, although Saturn’s moon
Titan does provide an exotic environment with frigid hydrocarbon liquids at
its surface and liquid water at warmer depths.
We finally arrive at Pluto, a solid world, but a world without mountains
or volcanoes. Like Mercury (the innermost planet of the solar system), frigid,
distant Pluto is devoid of volcanic activity.
As we return to Earth from this trip, we ponder what is unique about
Earth that may offer us clues to why animal life exists here but not on other
planets and their moons in our solar system. A crucial difference, its seems, is
Earth’s unique possession of liquid water at its surface. Water, the universal
solvent, seems indispensable for animal life. Earth has other unique attributes,
too, including its oxygen-rich atmosphere and a temperature range that allows
liquid water to exist.
Another unique attribute of Earth at first glance seems extraneous to animal
life but may indeed be crucial to it: linear mountain ranges. There are, of
course, giant mountains elsewhere in the solar system, the tallest being the
great volcano Olympus Mons on Mars. Yet such mountains are always single
and never occur in chains, unlike most mountains on Earth. There is no
equivalent to the Rockies, the Andes, the Himalayas, or the score of other
linear mountain chains we are so familiar with. Even at this crude level of observation,
oceans, mountain chains, and life make Earth unique in this solar
system. Life has had little to do with creating oceans and mountain chains.
Yet these features of Earth may have been crucial to the origin of life. In this
R A R E E A R T H
194
chapter, we argue that all three of these precious attributes of Earth are connected
in a complex interrelationship. All three, furthermore, may be the result
of plate tectonics. This process, the movement of the planetary crust
across the surface of the planet, is found in our solar system only on Earth,
and it may be vanishingly rare in the Universe as a whole. It is not the mountains
as such that are so important to life on Earth but the process that creates
them: plate tectonics.
It may seem odd to think that plate tectonics could be not only the
cause of mountain chains and ocean basins but also, and most enigmatically,
a key to the evolution and preservation of complex metazoans on Earth. But
there are several reasons to consider this view. First, plate tectonics promotes
high levels of global biodiversity. In the last chapter, we suggested that the
major defense against mass extinctions is high biodiversity. Here we argue
that the factor on Earth that is most critical to maintaining diversity through
time is plate tectonics. Second, plate tectonics provides our planet’s global
thermostat by recycling chemicals crucial to keeping the volume of carbon
dioxide in our atmosphere relatively uniform, and thus it has been the single
most important mechanism enabling liquid water to remain on Earth’s surface
for more than 4 billion years. Third, plate tectonics is the dominant
force that causes changes in sea level, which, it turns out, are vital to the formation
of minerals that keep the level of global carbon dioxide (and hence
global temperature) in check. Fourth, plate tectonics created the continents
on planet Earth. Without plate tectonics, Earth might look much as it did
during the first billion and a half years of its existence: a watery world, with
only isolated volcanic islands dotting its surface. Or it might look even more
inimical to life; without continents, we might by now have lost the most important
ingredient for life, water, and in so doing come to resemble Venus.
Finally, plate tectonics makes possible one of Earth’s most potent defense
systems: its magnetic field. Without our magnetic field, Earth and its cargo
of life would be bombarded by a potentially lethal influx of cosmic radiation,
and solar wind “sputtering” (in which particles from the sun hit the upper atmosphere
with high energy) might slowly eat away at the atmosphere, as it
has on Mars.
The Surprising Importance of Plate Tectonics
195
WH A T I S P L A T E T E C T O N I C S ?
Geologists of the eighteenth and nineteenth centuries had little difficulty
understanding the origin of volcanoes: Hot magma from deep within the
planet rose to the surface regions and spewed forth lava, ash, and pumice to
form a cone. Understanding how nonvolcanic mountains and mountain ranges
could form, however, was more problematic. Countless hypotheses were
proposed. These included buckling of the crust as a result of sediment loading
(where the weight of slowly accumulating sediment finally causes the
crust to crack in linear fashion), shrinking of the planet (causing ridges to
form as on a dried prune), and an expanding Earth (where the expansion creates
mountain ranges). In 1910 American geologist Frank B. Taylor proposed
a radically new idea: The drifting of continents caused the great
mountain chains. This heresy was immediately decried by nearly all other
geologists and geophysicists, who could envision no mechanism by which
such “drift” could occur.
Taylor’s hypothesis, however, kindled a spark of interest that would not
die. Soon other scientists began toying with the idea and searching for supporting
evidence. The most dogged of the new converts was a German meteorologist
named Alfred Wegener, who from 1912 until his death in 1930 on
Arctic ice was obsessed with the idea. Drawing on evidence from geology
and geophysics, Wegener was the first to show how the fit of various coastlines
supported the idea that all the continents were once united in a single
“supercontinent.” He was also the first to use paleontological evidence to support
this claim: He argued that the presence of similar fossil species on land
masses now widely separated could have come about only if the various continents
had once been in contact. He convinced some other geologists that
continents did and do drift, although the majority remained skeptical until
the 1960s.
The greatest obstacle to the idea (and the rallying point of all “antidrifters”)
was the seeming absence of any sort of reasonable, underlying
mechanism. How could the massive continents “float” over the surface of the
planet’s stony surface? The answer to this question, it was eventually discovered,
R A R E E A R T H
196
lies in the different phase states of Earth’s uppermost layers, known as the
crust and upper mantle, and the presence of thermal convection in these regions.
Scottish geologist Arthur Holmes first proposed that the upper mantle
acts like boiling water, producing large moving “cells” of material. Deep
below the surface, the fluid, hot material composing the upper mantle is
heated and begins to rise; as it rises it cools, and eventually it begins to flow
parallel to the planet’s surface. When it cools sufficiently, it sinks again.
Holmes proposed that where it rises, the convection cells might rupture the
rigid, solid crust and then carry it along, piggyback fashion, in those regions
where the mantle moves parallel to the surface.
The outlandish ideas behind the early theory of continental drift were
eventually shown to be correct. Evidence came from many sources, including
paleontological data and even the fit of the continental coastlines, as first
proposed by Wegener. Yet the two most powerful lines of evidence for plate
tectonics (another term for continental drift) came from fields unknown to
Wegener: From the study of paleomagnetics, which allowed the reconstruction
of ancient continental positions, and from oceanographic studies of the
ocean floor, which revealed the presence of enormous underwater volcanic
centers, areas where the sea floor literally pulls away from itself.
We know now that all continents are masses of relatively low-density
rock embedded in a ground mass of more dense material. The low-density
rocks have the average composition of granites, whereas the higherdensity
rocks that make up the ocean crust are basaltic in composition. Because
granite is less dense than basalt, the granite-rich continents essentially
“float” on a thin (relative to Earth’s diameter) bed of basalt. Earth scientists
like to use the analogy of an onion; the thin, dry, and brittle onion skin is the
crust, sitting atop a concentric globe of higher-density, wetter material. Continents
are like thin smudges of slightly different material embedded in the
onion skin. Unlike an onion, however, Earth has a radioactive interior and
constantly generates great quantities of heat as the radioactive elements, entombed
deep within in the planet, break down into their various isotopic byproducts.
As this heat rises toward the surface, it creates gigantic convection
The Surprising Importance of Plate Tectonics
197
cells of hot, liquid rock in the mantle, just as Arthur Holmes envisioned. Like
boiling water, the viscous upper mantle rises, moves parallel to the surface
for great distances (all the while losing heat), and then, much cooled, settles
back down into the depths. These gigantic convection cells carry the thin,
brittle outer layer—known as plates—along with them. Sometimes this outermost
layer of crust is composed only of ocean bed; sometimes, however,
one or more continents or smaller land masses are trapped in the moving
outer skin.
Under the pressures and temperatures encountered at depths many
kilometers beneath Earth’s surface, the familiar rocks of our crust act in ways
very different from what we are used to. Victor Kress of the University of
Washington pointed out that all but a tiny fraction of the upper mantle is entirely
solid. Yet it acts like a liquid in certain ways, most significantly in its
“convection”: the process whereby a liquid, when heated, flows upward and
then across the top of its container. The mantle convects in the manner of a
liquid only because the movement is so slow, and the temperatures so high,
that individual crystals have time to deform in response to stress. The upper
mantle is a hot, highly compressed mass of crystal that acts like a very viscous
liquid.The “plates” of plate tectonics are composed of all of the crust and a thin
section of mantle that underlies it, which together act as a relatively rigid
composite layer. Plates are of varying thickness, and their “bottoms” are
thought by many scientists to coincide with the 1400°C isotherm (a region
where the rock is heated to that very high temperature at which mantle rock
material melts into a plastic-like medium). Another way of visualizing the
plate foundation is to recognize that this region is characterized by much decreased
viscosity. The difference in viscosity between the overlying plate and
the underlying region of lowered viscosity is highly important in plate tectonics.
It allows the relatively rigid crust to slip as a unit over the zone of high
viscosity. Plates composed of oceanic crust and mantle are about 50–60 kilometers
thick, whereas the plates with continental crust average about 100
kilometers in thickness.
R A R E E A R T H
198
Let’s begin our examination of the plate tectonic process with ocean
basins. The crust we find lining the bottom of the world’s oceans is largely
made up of basalt, the same type of volcanic rock that makes up the Hawaiian
Islands. This material originates within the deeper mantle region of Earth; it
ascends along the rising zones of the convection cells. As this hot, dense
mantle material rises toward the surface, it moves into regions of ever lessening
pressure, because the weight of overlying material decreases. A lowerdensity
liquid separates from the higher-density mantle material, rising to the
surface as the “lava” we are familiar with from so many movies of erupting volcanoes.
The magma enters a huge crack in the surface of the planet formed by
the pulling apart of two plates and solidifies into basaltic ocean crust. It too
begins to move away from the “spreading center” where it first lithified, and
more new magma wells up to take its place—an endless conveyor belt.
The basalt produced in the spreading centers has a much different composition
from its “parent,” the mantle material rising along the limbs of the
convection cells. Because it contains a much higher percentage of silica
atoms, it is much lower in density than the mantle material. The basalt has differentiated
from the parent material (which, when occasionally found on the
surface, has the name peridiotite). This differentiation from a peridiotite
composition to a basaltic composition is the final step of oceanic crust formation.
Continents, however, have an even lower density than the oceanic
crust. The recipe for their creation requires a further step in this arcane lithic
cooking: the formation of the rock types granite and andesite. The characteristic
speckled appearance of both of these rocks, compared to the more
somber, chocolate to black color of basalt, comes from their containing even
more of the white (and low-density) silica. The major step in forming continental
crust is thus the differentiation of granite from material of a basaltic
composition. This process takes place in several steps, but the key ingredient
is water, and the key mechanism is called subduction.
Over many millions of years, oceanic crust moves away from its birthplace,
the spreading centers, all the while being carried piggyback on the
convecting mantle beneath it. Like all journeys, however, this long ride must
eventually end; the oceanic crust cannot expand forever. The basalt has
The Surprising Importance of Plate Tectonics
199
Mantle
Outer core
Inner
core
Asthenosphere
Backarc
basin
Subduction zone
Transform fault
Ridge
Triple
junctions
Pull-apart
basin
Volcanic
arc
Hot spot
tracks
Lithosphere
Crust
Figure 9.1 Schematic diagram displaying the principal features of the lithosphere, the rigid plates
50 –150 km thick that incessantly move about on Earth’s surface, created along the 56,000 km of
spreading-ridge systems and consumed along the 36,000 km of subduction zones.
cooled through time, and even more significantly, it has gained some heavy
freeloaders—piles of dense igneous rock known as gabbro that attach to the
base of the basalt. The basalt now just barely floats, and as it cools, it gets
heavier. Given any good excuse, it simply sinks, descending as deep as 650
kilometers. Eventually, then, the convection cell begins its downward journey
back into the deep mantle, and when it does, it carries its veneer of
oceanic crust back down with it, at regions called subduction zones.
Subduction zones (see Figure 9.1) are long, linear regions where
oceanic crustal material is driven deep into Earth, not so much by being
pushed down as sinking down through gravity. It is near and parallel to these
R A R E E A R T H
200
subduction zones that linear mountain ranges are constructed. The mountains
form partly as a by-product of the collision of two plates, which causes
buckling and crumpling of the leading edges, and partly by the upward
movement of hot magma, which eventually solidifies into granites and other
magmatic rocks parallel to the subduction zones. The Cascade Mountains of
Washington State are an example; the still-active peaks, such as Mt. Baker,
Mt. Rainier, and Mt. St. Helens, are direct evidence of the power and importance
of subduction in creating mountain ranges. Most of the world’s volcanoes
and mountain chains are found near these subduction zones (or where
ancient subduction zones used to operate), further testimony to the fundamental
link between subduction and mountain building. That mountain
chains are not found on other planets or moons of our solar system is clear evidence
that only Earth now has plate tectonics.
Volcanoes occur along subduction zones because by the time (which
may be millions of years after its formation) that oceanic crust reaches a subduction
zone and begins to descend, it is of slightly different composition
from when it was created in the spreading centers. As the basalt created in
spreading centers moves away from its birthplace, water is gradually added to
the crystal structures of key minerals—in other words, the basalt becomes
hydrated. Over long millennia, seawater works its way down through many
cracks and crevices of the oceanic crust and reacts chemically through the addition
of water molecules to the crustal lattices of minerals making up the
basalt. Water-poor minerals actually incorporate significant amounts of water
in their structure. The newly hydrated minerals have a lower melting point
than nonhydrated minerals, so as the oceanic basalt descends in the subducting
slab, the hydrated, silicate-rich minerals making up the basalt melt, and
the liquid that is produced rises back toward the surface. This water leads to
a decrease in the melting temperature of the overlying mantle rock that now
surrounds it, creating liquid magma where one would otherwise expect to
find only solid rock. This magma, when eventually cooled, becomes the rocks
we call andesite and granite, and its rise back to the surface is a key force in
producing new mountains and the line of volcanoes we find along subduction
The Surprising Importance of Plate Tectonics
201
zones. But the crucial aspect of these volcanoes is that they are made up of
magma of lower density than the basalt that parented them, and in this way,
a new, lower-density rock type is created. This rock starts out as andesite
(named after the Andes Mountains) and becomes part of the continental
crust. Because andesite and granite (which is created in similar fashion) are so
rich in silicate mineral, they are less dense than basalt. They become the
backbone of the continents—and their flotation devices! With andesite- and
granite-rich cores, continents can float on a sea of basalt. They can never be
sunk in subduction zones. Continents cannot be destroyed (though they can
be eroded). They can be split and fragmented, to drift from place to place,
but their basic volume cannot be reduced. Through time, in fact, the number
of continents on Earth has seemingly increased.
One of the most important findings about Earth history is that since the
formation of our planet, the total area of oceanic plates has gradually diminished
as the area of continental plates has grown (see Figure 9.2). This seems
counterintuitive, because the oceans are continuously enlarging as a result of
sea floor spreading. Yet as we have just seen, ocean crust can sink (and be
remelted back to magma in the process), whereas the lighter continental crust
remains afloat like a cork on this sea of basalt. Furthermore, the continents
enlarge through the process of mountain building, for the volcanoes lining
subduction zones and many continental edges receive vast quantities of
granitic and andesitic magma. Geologist David Howell, in his book Principles
of Terrane Analysis, estimates that the volume of continents increases by between
650 and 1300 cubic kilometers of rock each year. This estimate is for
the modern day, and some geologists believe continental volume increased
more rapidly in the past, especially early in Earth history, when plate tectonic
processes may have occurred much faster than they do now because more
heat emanated from the early Earth.
Plates thus intersect with each other in three ways: at the spreading centers
(where new magma reaches the surface along enormous linear cracks,
such as the mid-Atlantic ridge); areas where plates grind by each other side
by side (such as the San Andreas Fault of California); and regions where
R A R E E A R T H
202
plates collide—the subduction zones—which are associated with linear
chains of active volcanoes (such as the Cascades and the Aleutian Islands).
WH Y I S P L A T E T E C T O N I C S
I M P O R T A N T TO L I F E ?
The rate of continental growth is of major importance to life and its ecosystems.
The majority of Earth’s biodiversity is today found on continents, and
there is no reason to believe that this relationship has changed over the last
300 million years. As continents have grown through time, they have affected
global climate, including the planet’s overall albedo (its reflectivity to
sunlight), the occurrence of glaciation events, oceanic circulation patterns,
and the amount of nutrients reaching the sea. All of these factors have biological
consequences and affect global biodiversity.
Oldest rocks
Volume of continental land mass
relative to present
100%
50%
1 2 3 4
Time since Earth's formation
(billions of years)
First evidence of
life on land
Figure 9.2 Estimate of the growth of continental land mass with time (adapted from Taylor, 1992). For
nearly a third of its history, Earth was a “water world” nearly devoid of land.
The Surprising Importance of Plate Tectonics
203
In the last chapter we proposed that diversity (roughly the number and
relative abundance of species on the planet at any given time) is a major
hedge, or defense, against planetary extinction or sterilization of life: High
levels of diversity can counter the loss of body plans during mass extinctions.
Plate tectonics can augment diversity by increasing the number and degree of
separation of habitats (which promotes speciation). For example, as continents
break apart, the seaways forming between them create barriers to dispersal.
This in turn reduces gene flow and enhances the formation of new
species through geographic isolation. Plate tectonics also increases the nutrients
available to the biosphere, which may (or may not) also promote increased
biotic diversity.
Plate tectonics promotes environmental complexity—and thus increased
biotic diversity—on a global scale. A world with mountainous continents,
oceans, and myriad islands such as those produced by plate tectonic forces is
far more complex, and offers more evolutionary challenges, than would either
totally land- or ocean-dominated planets without plate tectonics. James
Valentine and Eldredge Moores first pointed out this relationship in a series of
classic papers in the 1970s. They showed that changes in the position and
configuration of the continents and oceans would have far-reaching effects on
organisms, causing both increased diversification and extinction. Changes in
continental position would affect ocean currents, temperature, seasonal rainfall
patterns and fluctuations, the distribution of nutrients, and patterns of biological
productivity. Such varying conditions would cause organisms to migrate
out of the new environments—and would thus promote speciation. The
deep sea would be least affected by such changes, but the deep sea is the area
on Earth today with the fewest species. Over two-thirds of all animal species
live on land, and the majority of marine species live in the shallow-water regions
that would be most affected by plate tectonic movements.
The most diverse marine faunas on Earth today are found in the tropics,
where communities are packed with vast numbers of highly specialized
species. In higher latitudes the number of species lessens, and in Arctic regions
there may be only a tenth as many species as in equivalent water depths

R A R E E A R T H
204
or habitats found in the tropics. Not only are there fewer species in the higher
latitudes, but the composition of species is also different there. Physiological
adaptation constrains most species to fairly narrow temperature limits: Animal
species adapted to warm, tropical conditions cannot survive in the cold,
nor can the cold-adapted species tolerate the warmth of the tropics. Given
that temperature conditions change rapidly with latitude, it’s not surprising
that north-south coastlines of continents show a continuously changing mix
of species. North-south coastlines promote diversity because of the latitudinal
temperature gradients. East-west coastlines, on the other hand, often
show similar species.
As continental positions change through time, the relative abundance
of north-south and east-west coastlines can change. Also, the larger the continents,
the lower the environmental heterogeneity. If many or all of the continents
were welded together into “supercontinents,” biodiversity would be
expected to be lower than if there were many smaller and separated continental
masses. On one large continent, groups of land animals have fewer
barriers to dispersal—and thus less opportunity to form new species. Clearly,
continental size and position should affect biodiversity, and this appears to
have been the case in Earth history.
WH A T WO U L D HA P P E N I F P L A T E
T E C T O N I C S CE A S E D ?
The fossil record suggests that there are more species of animals and plants
alive on Earth today than at any time in the past; estimates vary between
about 3 and 30 million species. This great diversity has come about through
many physical and evolutionary factors. We contend that the effects of plate
tectonics are among the most important. But once created, does high biodiversity
require the continued presence of plate tectonics? We can examine this
question with a thought experiment.
The Surprising Importance of Plate Tectonics
205
An End to Volcanism
Imagine that all volcanism on Earth’s surface suddenly ceases. This will stop the
many dozens of volcanic eruptions that occur on the continents each year (usually
causing great media fanfare and little damage). But the cessation of volcanism
will have a far more profound effect. If all volcanism stops, so does sea floor
spreading—and thus plate tectonics as well. And if plate tectonics stops, Earth
eventually (through erosion) loses most or all of the continents where most terrestrial
life exists. In addition, CO2 is removed from the atmosphere via weathering,
causing our planet to freeze. Of all of the attributes that make Earth rare,
plate tectonics may be one of the most profound and—in terms of the evolution
and maintenance of animal life—one of the most important.
Only cessation of the flow of heat up from Earth’s interior or thickening
of the crust would stop volcanism. It is this heat that causes convective motion
of the interior, the subterranean motor of plate tectonics. To stop plate
tectonics would require eliminating these great lithic boiling pots, and that
cannot be done unless all heat emanating from the Earth’s interior is stopped
(which would require that all the radioactive minerals locked away there
decay to stable daughter products) or the composition of Earth’s crust or
upper mantle changes such that movement can no longer occur. This could
happen if the crust became too thick or the mantle too viscous to allow
movement. None of these conditions is likely to occur on Earth in the foreseeable
future, but there is speculation that just such events occurred on both
Venus and Mars in the past.
If Earth’s tectonic plates did suddenly stop moving, subduction would
no longer occur at the contacts between colliding plates. Mountains—and
mountain chains—would cease to rise. Erosion would begin to eat away at
their height. Eventually, the world’s mountains would be reduced to sea level.
How long would it take? The problem is a bit more complicated than simply
measuring average erosion rates and calculating the number of years required
for the mountains to disappear. This is because of the principle of isostacy.
Mountains (and continents) are a bit like icebergs: If you cut off the top, the
bottom rises up relative to sea level, causing the entire iceberg (or mountain)
R A R E E A R T H
206
to rise. Eventually, however, even this isostatic rebound effect would be
overcome by the extent of the erosion.
What would sea level be in a world without further plate tectonics? All
of the sediment produced by the simultaneous erosion of the world’s mountains
would have to go somewhere—and that somewhere is the ocean. The
eroding continental mass carried into the oceans by river and wind transport
would displace seawater and cause the level of the sea to rise. Calculations by
David Montgomery, a geomorphologist at the University of Washington,
suggest that the entire Earth might become covered by a global ocean—
much shallower than the oceans of today, of course, but global in extent nevertheless.
Our planet would have returned to its state of 4 billion years ago: a
globe covered completely (or nearly so) by ocean. And with the continents
awash, Earth would witness a mass extinction more catastrophic than any in
the past. All land life would die off under the lapping waves. Paradoxically,
the increase of ocean area would probably also be accompanied by extinctions
in the sea. Ocean life depends on nutrients, and most nutrients come
from the land as runoff from rivers and streams. With the disappearance of
land, the total amount of nutrients (though initially higher as so much new
sediment entered the ocean system) would eventually lessen, and with fewer
resources, there would be fewer marine animals and plants.
How long before such a water world would be achieved? Tens of millions
of years would be required for the mountains and continents to erode to
sea level. Yet mass extinction would ensue long before that. Planetary
calamity for complex life would occur shortly after the cessation of plate
movement, for plate tectonics is not only the reason we have mountains; it
turns out to control our planet’s climatic thermostat as well.
Loss of Planetary Temperature Control
The temperature of Earth must remain in a range suitable for the existence of
liquid water if animal life is to be maintained. The range of temperature that
Earth experiences is the result of many factors. One is the existence of the atThe
Surprising Importance of Plate Tectonics
207
mosphere. The average temperature of the Moon is 18°C, for example,
well below the freezing point of water, simply because it has no appreciable
atmosphere. If Earth did not have its cloaking atmosphere, including such insulating
gases as water vapor and carbon dioxide (producing the muchdiscussed
Greenhouse Effect), its temperature would be about the same as
that of the Moon. Yet the Earth, thanks to the greenhouse gases, has an average
global temperature of 15°C (33°C warmer than the Moon). Greenhouse
gases are keys to the presence of fresh water on this planet and thus are
keys to the presence of animal life—and many scientist now believe that the
balance of greenhouse gases in Earth’s atmosphere is directly related to existence
of plate tectonics.
Greenhouse gases are those with three or more atoms, such as water
vapor (H2O; three atoms), ozone (O3), carbon dioxide (CO2; three atoms),
and methane (CH4; five atoms). All can capture outgoing infrared energy from
Earth’s surface and, in so doing, warm the planet. Their role in keeping Earth’s
temperature within the critical levels necessary not only for allowing the
presence of liquid water (0°C to 100°C) but also for maintaining animals
(about 2°C to about 45°C) has been beautifully summarized by Columbia
University geologist Wally Broecker in How to Build a Habitable Planet. Broecker
describes the following scenario. Imagine that the sun’s energy was diminished
for a period of time brief by geological standards but long enough for
the oceans to freeze. If the sun then resumed its normal output of today,
Earth would remain frozen. Once frozen, water reflects much of the light that
hits it, and even the current volume of greenhouse gas would be insufficient
to reheat the planet to a temperature at which the water would thaw. This
condition is called a Global Icehouse, and it is one way a planet can lose its
animal life. They freeze to death.
Now, say we reversed this situation and allowed the sun’s energy to increase
for a geologically short period of time, but long enough so that all of
Earth’s oceans boiled away, filling the atmosphere with steam. If we then reduced
the sun’s energy to its present-day levels, the oceans might not recondense,
and the planet would stay hot. Once in the atmosphere, the steam
R A R E E A R T H
208
would keep the planet hot through its properties as a greenhouse gas, even
when solar radiation hitting the planet had decreased. This situation is called
a Runaway Greenhouse.
The Earth’s greenhouse gases are rare compounds of our planet’s atmosphere.
It turns out that the major constituents of our atmosphere, nitrogen
and oxygen, play little role in the greenhouse warming, because they do not
absorb infrared radiation. Carbon dioxide and water vapor, on the other
hand, do, even though they make up only a tiny fraction of the gas volume of
the atmosphere (carbon dioxide constitutes only 0.035% of the atmosphere).
Plate tectonics plays an important part—perhaps the most important part—
in maintaining levels of greenhouse gases, and these in turn maintain the temperatures
necessary for animal life.
P L A T E T E C T O N I C S
A S GL O B A L T H E R M O S T A T
Over and over again we come back to to a common theme: the importance of
liquid water. For animal life based on DNA to exist and evolve, water must be
present and abundant on a planet’s surface. Even on the water-rich Earth
today, slight differences in water content obviously affect life. In desert regions
there is little life; in rainforests at the same latitude, life teems in abundance.
For complex life to be attained (and then maintained), a planet’s water
supply (1) must be large enough to sustain a sizable ocean on the planet’s surface,
(2) must have migrated to the surface from the planet’s interior, (3) must
not be lost to space, and (4) must exist largely in liquid form. Plate tectonics
plays a role in all four of these criteria.
Earth is about one-half of 1% water by weight. Much of this water arrived
among the planetesimals that took part in Earth’s formation and accretion.
Other volumes of it were dumped here by incoming comets after Earth accreted.
The relative importance of these two processes is largely unknown at this time.
Once liquid water is established on the surface of a planet, its maintenance
becomes the primary requirement for attaining (and then supportThe
Surprising Importance of Plate Tectonics
209
ing) animal life. The maintenance of liquid water is controlled largely by
global temperatures, which are a by-product of the greenhouse gas volumes
of a planet’s atmosphere. The temperature of Earth’s (and of any planet’s
surface) is a function of several factors. The first is related to the energy
coming from its sun. The second is a function of how much of that energy
is absorbed by the planet (some might be reflected into space, and this relationship
is dictated by a planet’s reflectivity, or albedo). The third is related
to the volume of “greenhouse gases” maintained in a planet’s atmosphere.
Greenhouse gases have a residence time in any atmosphere and are
eventually broken down or undergo a change in phase. If their supplies are
not constantly replenished, the planet in question (such as Earth) will grow
colder gradually until the freezing temperature of water is reached, at
which point it will grow colder rapidly (as we have noted, when a planet
starts accumulating ice, its albedo increases, boosting its rate of cooling).
Greenhouse gases are thus vastly important in maintaining a planet’s thermostatic
reading. Both plate tectonic and non–plate-tectonic planets regularly
produce greenhouse gases, because the most important source of these
planetary insulators is volcanic eruption, which occurs on most or all planets.
On Earth, volcanoes daily exhale vast volumes of carbon dioxide from
deep within. Even so-called “dormant” volcanoes are venting carbon dioxide
into the atmosphere. On any planet with volcanism there is usually an
abundance of greenhouse gases—too much in some cases, and this is where
plate tectonics becomes crucial.
Greenhouse gas compositions, and thus planetary temperatures, are byproducts
of complex interactions among a planet’s interior, surface, and atmospheric
chemistry. One of the most important by-products of plate tectonics is
the recycling of mineral and chemical compounds locked up in any planet’s
sedimentary rock cover. On non–plate-tectonic worlds, vast quantities of
sedimentary material are produced by erosion. These materials and minerals
become sequestered and eventually buried and lithified through sedimentation
and the formation of sedimentary rocks, and in most cases, they are reexhumed
only through some process leading to mountain building. Yet, as we
have seen, mountain building on non–plate-tectonic worlds is largely confined
R A R E E A R T H
210
to the formation of large volcanoes over hot spots. With plate tectonics,
however, the motion (and collision) of plates, the formation of mountain
chains, and the process of subduction all lead to a recycling of many materials.
This recycling plays a large role in maintaining Earth’s global temperature
values in a range that allows the existence of liquid water. One of the
most important of the recycling processes is putting CO2 back into the atmosphere.
As limestone is subducted deep into the mantle, it metamorphoses
and, in the process, returns CO2 into the atmosphere. This is clearly an important
aspect of global warming.
The most important element in reducing atmospheric carbon dioxide
(which leads to global cooling) is the weathering of minerals known as silicates,
such as feldspar and mica (granite has many such minerals within it).
The presence or absence of plate tectonics on a given planet greatly affects
the rates and efficiency of this “global thermostat.” The basic chemical reaction
is CaSiO3  CO2  CaCO3  SiO2. When the first two chemicals in
this equation combine, limestone is produced and carbon dioxide removed
from the system. The feedback mechanism at work here was first pointed out
in a landmark 1981 paper by J. Walker, P. Hays, and J. Kasting. (James Kasting
has told us that he first had this insight in the middle of his Ph.D. exam!)
The mechanism is related to the rates of weathering—that is, the physical or
chemical breakdown of rocks and minerals. Although weathered entails the
reduction in size of rocks (big boulders weather into sand and clay over time),
a very important chemical aspect is also involved (see Figure 9.3). Weathering
can cause the actual mineral constituents of the rocks being weathered to
change. Weathering of rocks that contain silicate minerals (such as granite)
plays a crucial part in regulating the planetary thermostat. Walker and his
colleagues pointed out that as a planet warms, the rate of chemical weathering
on its surface increases. As the rate of weathering increases, more silicate
material is made available for reaction with the atmosphere, and more carbon
dioxide is removed, thus causing cooling. Yet as the planet cools, the rate of
weathering decreases, and the CO2 content of the atmosphere begins to rise,
causing warming to occur. In this fashion the Earth’s temperature oscillates between
warmer and cooler as a result of the carbonate–silicate weathering and
The Surprising Importance of Plate Tectonics
211
precipitation cycles. Without plate tectonics, this system does not work efficiently.
It also works less efficiently on planets without land surfaces—and
much less efficiently on planets without vascular plants such as the higher
plants common on Earth today.
Calcium is an important ingredient in this process, and it has two main
sources on a planet’s surface. It is found in igneous rocks and (more important)
in the sedimentary rocks called limestone. Calcium reacts with carbon dioxide
to form limestone, the material that marine animals use to build their shells
(and that we humans use to build our cement and concrete). Calcium thus
draws CO2 out of the atmosphere. When CO2 begins to increase in the atmosphere,
more limestone formation occurs, but only if there is a steady
Ocean
Oceanic plate
Carbonate deposits
form on the sea floor
Weathering carries
CO2 into the sea
Subduction
Release of
volcanic CO2
Decomposing
carbonates
release CO2
Figure 9.3 The CO2–rock weathering cycle. This remarkable cycle has controlled the amount of atmospheric
carbon dioxide, a greenhouse gas, to regulate Earth’s surface temperature for billions of years. Because
this process requires both surface water and plate tectonics, it is not known to occur elsewhere.
R A R E E A R T H
212
source of new calcium available. The calcium content is steadily made available
by plate tectonics, for the formation of new mountains brings new
sources of calcium back into the system by exhuming (in magmas) ancient
limestone, eroding it, and thus releasing its calcium to react with more CO2.
The planetary thermostat requires a balance between the amount of
CO2 being pumped into the atmosphere through volcanic action and the
amount being taken out through the formation of limestone. On non–platetectonic
worlds, buried limestone stays buried, thus removing calcium from
the system and producing increases in carbon dioxide. On Earth, at least,
plate tectonics plays an integral part in maintaining a stable global temperature
by recycling limestone into the system.
Although most accounts of habitability of planets refer to the range between
0°C and 100°C, required temperature range is really much narrower if
animals are to survive. As we have seen, life such as bacteria can withstand a
range of temperatures that may approach 200°C in high-pressure environments.
But animals are much more fragile. Animal life on Earth—and perhaps
anywhere in the Universe—depends on the narrowest of temperature
ranges within the wider range that permits liquid water to exist. Extended
periods of anything above 40°C or much below 5°C will stymie animal life.
The planetary thermostat must be set to a narrow range of temperatures indeed,
and it may be that only the plate tectonic thermostat makes this finetuning
possible.
P L A T E T E C T O N I C S AND
T H E MA G N E T I C F I E L D
Outer space is not a particularly friendly place. One of its hazards is cosmic
rays, which are elementary particles—electrons, protons, helium nuclei, and
heavier nuclei—traveling at velocities approaching the speed of light. They
come from many sources, including the sun and cosmic rays from distant supernovae,
the explosions of stars. These catastrophic events send great numbers
of particles hurtling through space.
The Surprising Importance of Plate Tectonics
213
In The Search for Life in the Universe, D. Goldsmith and T. Owen speculate
that without some sort of protection, life on Earth’s surface would be extinguished
within several generations by cosmic rays hitting our planet’s surface.
However, the vast majority of cosmic rays are deflected by Earth’s magnetic
field. The innermost layer of our planet, its core, is made up mainly of iron,
which in the outermost region of the core is in a liquid state. As Earth spins, it
creates convective movement in this liquid that produces a giant magnetic field
surrounding the entire planet. What produces the convection cells in the core
is loss of heat. Heat must be exported out of the core, and this liberation of heat
appears to be greatly influenced by Earth’s plate tectonic regime. Joseph
Kirschvink of Cal Tech has suggested that without plate tectonics, there would
not be enough temperature difference across the core region to produce the
convective cells necessary to generate Earth’s magnetic field; no plate tectonics,
no magnetic field. The magnetic field also reduces “sputtering” of the atmosphere,
a process whereby the atmosphere is gradually lost into space. No magnetic
field, perhaps no animal life. Plate tectonics to the rescue again.
WH Y DO E S E A R T H ( B U T NOT MA R S
O R VE N U S ) H A V E P L A T E T E C T O N I C S ?
Why is there plate tectonics on Earth? The recipe for plate tectonics seems
simple enough at first glance. You need a planet differentiated into a thin,
solid crust sitting atop an underlying region that is hot, fluid, and mobile. You
need this underlying region to be undergoing convection, and for that you
need heat emanating from even deeper in the planet. And you are likely to need
water—oceans of water: Much new research suggests that without water you
cannot have plate tectonics (though perhaps it is simply that without water
you cannot get continents).
As in so much else in planetary geology, there is still a great deal we
don’t know about why our planet (and, more important, any planet) develops
and then maintains plate tectonics. Because ours is still the only planet we
know that has plate tectonics, we have nothing with which to compare it.
R A R E E A R T H
214
Much of the data pertaining to plate tectonics lies so deep that we are unlikely
ever to sample it directly.
As an illustration of the degree of uncertainty about what we might call
planetary plate tectonics, which we can define as the theoretical (as opposed
to the Earth-based actual) study of plate tectonics, we cannot be certain
whether plate tectonics would operate if Earth were 20% larger or smaller, or
if it had a crust with more iron and nickel than it does, or if its surface had
only 10% of the present-day volume of water. The best current work on these
types of questions is being done by planetary geologists V. Solomatov and
L. Moresi, who are using computational models to study how convection (the
driving force of plate tectonics) works. Yet the abstract of their 1997 paper
on the subject concluded, “The nature of the mobility of lithosphere plates
on Earth has yet to be explained.” We know the plates move, and we know
convection moves them. The physics behind the convection is well understood,
but its application to subduction is still an enigma.
When we asked about the physical condition necessary to produce plate
tectonics on a planet, Solomatov responded, “It is a very interesting problem
and we’ve just started exploring the physical conditions required for plate tectonics
to occur on a planet. So far, we have been moving to the conclusion
that water might be the factor which is crucial for plate tectonics: no water,
no plate tectonics.” Without water, the lithosphere (which is the plate of plate
tectonics, the rigid surface region composed of the crust and uppermost part
of the mantle) is strong and cannot break and descend back into the mantle—
the process known as subduction that occurs along the linear subduction
zones described earlier in this chapter. According to Solomatov, subduction
is a major requirement for plate tectonics. Apparently, subduction zones operate
only when the crust is “weak,” or able to bend and break, which allows
it to descend into the regions where the mantle convection cells sink downward.
All of this work is being done with mathematical modeling. Solomatov
and his colleagues are using computers to arrive at these generalizations—not
trips to the center of the Earth with Jules Verne’s heroes.
Even in the absence of water, plumes of hot magma may rise to a
planet’s surface. But this new material must ultimately go somewhere, and if
The Surprising Importance of Plate Tectonics
215
subduction is not operating, the plates will not move, for the new crustal material
must ultimately duck down into the mantle, along the linear subduction
zones. Without subduction zones, there is no plate tectonics, even if mantle
convection cells are operating inside a planet.
Venus and Mars both lack subduction zones and thus lack plate tectonics.
Although both might have the internal mantle convection necessary to
move surface plates, the surface itself is composed of “strong” rock (Solomatov’s
term) that cannot move. Because of its thickness and strength, the crust
on these planets is now immobile. The lack of water on both of these plates
may be the reason why this is so. Because both of these planets may in the
past have had liquid water and crustal composition similar enough to that of
Earth, we may find that Venus and Mars once did have plate tectonics—and
perhaps lost it when they lost their liquid water. Venus and Mars may be experiencing
what Solomatov and Moresi describe as a “stagnant lid regime”:
The viscosity difference between the convecting mantle and the solid surface
is so great that little or no movement of the crust can occur. Yet heat continues
to flow upward, and in the case of Venus, this heat caused the entire surface
of the planet to melt about a billion years ago (the planetary “resurfacing”
we alluded to at the start of this chapter). On Earth this great viscosity
difference does not occur. Earth has a “small viscosity contrast regime,” according
to the technical scientific papers describing all of this, and the result
is the very actively moving crust so important for mountain formation, nutrient
cycling, and life.
Yet perhaps we have this story reversed. Perhaps Mars and Venus had
water but lost it because they had no plate tectonics—and thus no planetary
thermostat.
HOW (AND WH E N ) D I D P L A T E
T E C T O N I C S S T A R T ON E A R T H ?
The time of onset of plate tectonics is controversial. Many sources believe it
began 1 to 2 billion years after Earth’s formation, whereas others view plate
R A R E E A R T H
216
tectonics as being far more ancient, its inception dating back over 4 billion
years. Much of the controversy involves the rate of heat flow from the early
Earth and how this would have affected the composition and rigidity of the
planet’s surface.
By the time the crust had solidified, more than half of the heat that
could result from planetary accretion, core formation, and decay of radioactive
isotopes (such as uranium-235) had already been lost from Earth. During
the 2 billion years of the Archaean era, heat flow slowed. Some workers believe
that the early crust was still too hot and thin to act as the rigid plate necessary
for plate tectonics; according to this hypothesis, plate tectonics may
not have commenced until 2.5 billion years ago. There is evidence in much
older rocks, however, of fault lines and movements consistent with plate tectonics.
The rate at which plate tectonics built continental surfaces on Earth was
not constant. If we plot the size of the continents through time relative to the
present area, we do see not a linear increase but a logistic curve—a curve that
started slowly, picked up speed in its middle, and then slowed near the end.
We have spoken in another context of the “Cambrian Explosion.” Here, Earth
underwent a “continental explosion” that resulted in a rapid formation of land
area. Many lines of evidence suggest that by far the greatest growth took
place rather rapidly, during a period between about 2 and 3 billion years ago.
This rapid growth completely changed Earth from a planet dominated by
oceans to one dominated (at least in terms of its global temperatures and
chemistry) by continents.
CO U L D P L A T E T E C T O N I C S A C T U A L L Y
HA V E I N H I B I T E D T H E F O R M A T I O N
O F A N I M A L L I F E ON E A R T H ?
In this chapter we have contended that plate tectonics facilitated the rise and
then the maintenance of animal life on Earth. But might not the opposite acThe
Surprising Importance of Plate Tectonics
217
tually be true? Could it be that plate tectonics actually retarded the rise of animals?
This is the contention of two NASA scientists, H. Hartman and
C. McKay, who hypothesized that plate tectonics slowed the rate of oxygenation
of the Earth atmosphere. In a 1995 article, Hartman and McKay
proposed that plate tectonics slowed the rise of oxygenation on Earth and, by
inference, on any planet.
As we have detailed in an earlier chapter, animal life did not arise on
Earth until less than a billion years ago, whereas life on this planet antedates
the first animals by about 3 billion years. One of the most puzzling aspects of
life’s history on Earth is this singular gap between the first life and the first animal
life. Many factors were surely involved, but there is irrefutable evidence
that oxygen is a necessary ingredient for animal life (at least on Earth), and
there is much evidence that sufficient concentrations of oxygen were not
present in the oceans and atmosphere until less than 2 billion years ago.
Many scientists suspect that the long time it took for Earth to acquire an oxygen
atmosphere accounts for some, or even all, of the delay between the origin
of the first life and the origin of animal life on Earth. Hartman and McKay
make the novel suggestion that this delay was partly due to the existence of
plate tectonics on Earth.
It is universally agreed that the rise of oxygen on Earth was due to the
release of free oxygen as a by-product of photosynthesis. The earliest photosynthetic
organisms used an enzymatic pathway called Photosystem 1; however,
this system does not release free oxygen. The later-evolved Photosystem
2 does. This latter system may not have evolved until 2.7 to 2.5 billion
years ago. Eventually, photosynthesizing organisms such as photosynthetic
bacteria and single-celled plants floating in the early seas would have released
vast volumes of oxygen. There was probably some source of inorganically
produced free oxygen on the early Earth as well. It may be, for example, that
ultraviolet rays hitting water vapor in the upper atmosphere created free oxygen,
at least in small volumes. However, a net accumulation could not take
place until various reducing compounds (which bind the newly released oxygen
and keep it from accumulating as a dissolved gas in the oceans or as a gas
in the atmosphere) were used up. For example, the amount of iron in the crust
R A R E E A R T H
218
of a planet has a major effect, for all of it on the surface in contact with the
atmosphere must be oxidized before free oxygen can accumulate. Such reducing
compounds emanate from volcanoes, and it can be argued that planets
with a higher rate of volcanicity have more reducing compounds in their
oceans and atmospheres. Another important source of reducing compounds
is organic compounds, produced either through the death and rotting of organisms
or through the inorganic formation of organic compounds, such as
amino acids. Great volumes of such material are found in the oceans on Earth,
but it is usually buried in sediments. In the absence of plate tectonics, argue
Hartman and McKay, such sediments become buried in sedimentary basins
and are never brought back into contact with the oceans and atmosphere;
thus they are removed from active participation in oceanic and atmospheric
chemistry. Because they are taken out of the system, oxygen can accumulate
faster than in the case where reducing compounds are constantly being reintroduced
into the atmosphere—a case where the dead don’t stay buried.
Hartman and McKay make the intriguing point that Mars may have
seen the evolution of complex life within 100 million years of the formation
of that planet (assuming, of course, that life originated there at all). Their argument
is as follows: The rapid removal of reductants on Mars through burial
in deep and undisturbed sediment would have allowed oxygenation to
occur much more quickly than on Earth (see Figure 9.4), where plate tectonics
constantly recycles sediments via subduction, plate collision, and mountain
building. All of these processes can cause previously buried sediments to
be brought back up to the surface, where their reductants would once more
bind whatever atmospheric oxygen was available. Hartman and McKay also
point out that volcanicity on a planet like Mars that does not exhibit plate
tectonics is much lower than on Earth. Thus the amount of reducing compounds
(such as hydrogen sulfide) entering the atmosphere–ocean systems
on Mars from volcanic sources would also have been much lower.
Could it be, then, that Earth hosted the evolution and then the maintenance
of animal life in spite of plate tectonics? And that plate tectonics actually
discourages the attainment of animal life on a given planet because its presence
slows the accumulation of the necessary oxygen-rich atmosphere?
The Surprising Importance of Plate Tectonics
219
We cannot fault the arguments of Hartman and McKay concerning the
role of reductants in retarding oxygenation. However, we can point out that
plate tectonics would surely increase the rate of biologically produced oxygen
on any world, because it enhances biological productivity by recycling
nutrients such as nitrates and phosphates. The net productivity on plate tectonic
worlds should thus be expected to be far higher than on non–plate-tectonic
worlds, so the rate of oxygenation through photosynthesis should also
Complex animal life
First animal life
Eukaryotic life
Prokaryotic life
4 3 2 1
Mars life extinct
Mars
Earth
Figure 9.4 Comparative evolutionary history of a planet without plate tectonics (Mars) and a
world with plate tectonics (Earth) expressed in billions of years before the present.
R A R E E A R T H
220
be much higher on a plate tectonic world, perhaps offsetting the retardant effect
of the recycling of sediment-sequestered reductants.
MO S T CR U C I A L E L E M E N T O F T H E R A R E
E A R T H HY P O T H E S I S ?
Plate tectonics plays at least three crucial roles in maintaining animal life: It
promotes biological productivity; it promotes diversity (the hedge against
mass extinction); and it helps maintain equable temperatures, a necessary requirement
for animal life. It may be that plate tectonics is the central requirement
for life on a planet and that it is necessary for keeping a world supplied
with water. How rare is plate tectonics? We know that of all the planets and
moons in our solar system, plate tectonics is found only on Earth. But might
it not be even rarer than that? One possibility is that Earth has plate tectonics
because of another uncommon attribute of our planet: the presence of a
large companion moon, the subject of the next chapter.
The Moon,
Jupiter,
and Life on Earth
The surface is fine and powdery. I can kick it up loosely
with my toe. It does adhere in fine layers, like powdered
charcoal, to the sole and sides of my boots. I only go in a
small fraction of an inch, maybe an eighth of an inch, but I
can see the footprints of my boots and the treads in the
fine, sandy particles.
—Neil Armstrong’s first words from the surface
of the Moon (1969)
Perhaps an astronomer’s greatest fear is that sooner or later, someone will
mistake him or her for an astrologer. The ancient belief system known
as astrology posits that the stars and planets exert a major influence on
our daily lives—a set of beliefs frequently and fervently disputed by astronomers.
Recent research, however, has in an odd way proved the astrologers
slightly correct. Two heavenly bodies, the Moon and Jupiter, do, in
fact, play pivotal roles in our very existence as a species. Without the Moon,
and without Jupiter, there is a strong likelihood that animal life would not
221
10

R A R E E A R T H
222
exist on Earth today. Both are thus key elements in the Rare Earth Hypothesis,
but for different reasons.
LUNA
Without the Moon there would be no moonbeams, no month, no lunacy, no
Apollo program, less poetry, and a world where every night was dark and
gloomy. Without the Moon it is also likely that no birds, redwoods, whales,
trilobites, or other advanced life would ever have graced Earth.
Although there are dozens of moons in the solar system, the familiar
ghostly white moon that illuminates our night sky is highly unusual, and its
presence appears to have played a surprisingly important role in the evolution
of life. The Moon is just a spherical rock 2000 miles in diameter and 250,000
miles away, but its presence has enabled Earth to become a long-term habitat
for life. The Moon is a fascinating factor in the Rare Earth concept because
the likelihood that an Earth-like planet should have such a large moon is
small. The conditions suitable for moon formation were common for the
outer planets but rare for the inner ones. Of the many moons in the solar system,
nearly all orbit the giant planets of the outer solar system. The warm,
Earth-like planets that are close to the sun and that fall within the habitable
zone, are nearly devoid of moons. The only moons of the terrestrial planets
are ours and Phobos and Diemos, the two tiny (10 kilometers in diameter)
moons of Mars. Some of the solar system’s moons are huge. Jupiter’s
Ganymede is nearly as large as Mars, and Saturn’s Titan is nearly that large
and has an atmosphere denser than our own, though much colder. Our Moon
is somewhat of a freak because of its large size in comparison to its parent
planet. The Moon is nearly a third the size of Earth, and in some ways it is
more of a twin than a subordinate. The only other case in the solar system
where a moon is comparable in size to its planet is Pluto and its moon,
Charon.
The Moon, Jupiter, and Life on Earth
223
Tilt!
The Moon plays three pivotal roles that affect the evolution and survival of
life on Earth. It causes lunar tides, it stabilizes the tilt of Earth’s spin axis, and
it slows the Earth’s rate of rotation. Of these, the most important is its effect
on the angle of tilt of Earth’s spin axis relative to the plane of its orbit, which
is called “obliquity.” Obliquity is the cause of seasonal changes. For most of
Earth’s recent history, its obliquity has not varied by more than a degree or
two from its present value of 23 degrees. Although the direction of the tilt
varies over periods of tens of thousands of years as the planet wobbles, much
like the precession of a spinning top, the angle of the tilt relative to the orbit
plane remains almost fixed. This angle is nearly constant for hundreds of
millions of years because of gravitational effects of the Moon. Without the
Moon, the tilt angle would wander in response to the gravitational pulls of
the sun and Jupiter. The monthly motion of our large Moon damps any tendencies
for the tilt axis to change. If the Moon were smaller or more distant,
or if Jupiter were larger or closer, or if Earth were closer to or farther from
the sun, the Moon’s stabilizing influence would be less effective. Without a
large moon, Earth’s spin axis might vary by as much as 90 degrees. Mars, a
planet with the same spin rate and axis tilt, but no large moon, is believed to
have exhibited changes to its tilt axis of 45 degrees or more.
Because tilt of a planet’s spin axis determines the relative amounts of
sunlight that land on the polar and on the equatorial regions during the seasons,
it strongly affects a planet’s climate. On planets with moderate tilts, the
majority of solar energy is absorbed in the equatorial regions, where the noon
sun is always high in the sky. Each pole is in total darkness for half a year and
has constant illumination for half a year. The highest altitude that the sun
reaches in the sky at the pole is exactly equal to the number of degrees of the
tilt of the spin axis. For moderate tilt angles, the sun is never high in the polar
sky, and ground heating by sunlight is low even in the middle of the summer.
The planet Mercury provides a spectacular example of what can happen on a
planet whose spin axis is nearly perfectly perpendicular to the plane of its
R A R E E A R T H
224
orbit. Mercury is the closest planet to the sun and most of its surface is hellishly
hot, but radar imaging from Earth has shown that the poles of the planet
are covered with ice. The planet is very close to the sun, but as viewed from
the poles, the sun is always on the horizon. In contrast to Mercury’s lack of
tilt, the planet Uranus has a 90-degree tilt; and one pole is exposed to sunlight
for half a year while the other experiences cryogenic darkness.
Although our viewpoint is certainly biased, our planet’s tilt axis seems to
be “just right.” Constancy of the tilt angle is a factor that provides long-term
stability of Earth’s surface temperature. If the polar tilt axis had undergone
wide deviations from its present value, Earth’s climate would have been much
less hospitable for the evolution of higher life forms. One of the worst possibilities
is that excessive axis tilt could have led to the total freezing over of
the oceans, a situation that might be very difficult to recover from. Extensive
ice cover increases the reflectivity of the planet, and with less absorption of
sunlight, the planet continues to cool. Astronomer Jacques Laskar, who made
many of the calculations that led to the surprising discovery of the Moon’s
importance in maintaining Earth’s stable obliquity, summarized the situation
as follows:
These results show that the situation of the Earth is very peculiar.
The common status for all the terrestrial planets is to have experienced
very large scale chaotic behavior for their obliquity, which,
in the case of the Earth and in the absence of the Moon, may have
prevented the appearance of evoluted forms of life. . . . [W]e owe
our present climate stability to an exceptional event: the presence
of the Moon.
High obliquity has remarkable and seemingly counterintuitive effects
on planets (see Figure 10.1). Consider a planet that is tipped 90 degrees. Averaged
over the year, the poles would receive exactly as much solar energy as
the equator would with no tilt angle. The north pole would become the Sahara!
For the 90-degree tilt, however, the equatorial regions would receive
much less energy averaged over the year and would become colder. If a
planet is tilted more than 54 degrees, its polar regions actually receive more
The Moon, Jupiter, and Life on Earth
225
Tilt angle
0˚
22.5˚
45˚
60˚
90˚
Energy at pole
Energy at equator
0
0.4
0.9
1.1
1.6
Figure 10.1 The ratio of the annual amount of solar energy falling on a planet’s pole to that falling on
its equator varies with the angle of a planet’s spin axis. With a tilt angle of 22.5°, the Earth has very cold
polar regions, but if the tilt exceeded 54°, the polar regions would actually receive more sunlight than the
tropics. The lines parallel to the equator are the polar circles, where the Sun never sets in the midsummer and
never rises in the midwinter.
R A R E E A R T H
226
energy input from sunlight than the equatorial regions. If the Earth were
tilted more than this amount, the equatorial oceans might freeze and the
polar regions would be warmer: a topsy-turvy world. Recently uncovered evidence
has revealed that equatorial ice sheets did exist about 800 to 600 million
years ago, and ice-rafted sediments of this age have been found in formerly
equatorial regions. This has led to the “Snowball Earth” hypothesis that
Earth may have actually have frozen over, as we saw in Chapter 6. It has been
suggested that this may have been due to high tilt angle during a period of
time when the Moon did not have full control. We do not know for sure how
long the Moon has been successful in stabilizing Earth’s obliquity.
In the distant future, the Moon will lose its ability to stabilize Earth’s
spin axis. The Moon is slowly moving outward from Earth (at a rate of about
4 centimeters a year), and within 2 billion years it will be too far away to have
enough influence to stabilize Earth’s obliquity. Earth’s tilt angle will begin to
change as a result, and the planet’s climate will follow suit. Further complicating
the future is the slow but unrelenting increase in the brightness of the
sun. At the time when our planet’s spin axis begins to wander, the sun will be
hotter, and both effects will decrease the habitability of Earth.
There is currently much speculation about how rapid such changes of
planetary obliquity might be in the absence of the Moon. Estimates for the time
it would take Earth to “roll” on its side range from tens of millions of years to far
shorter periods. Astronomer Tom Quinn of the University of Washington has
suggested to us that the time of obliquity change could occur on scales as short
as hundreds of thousands, rather than millions, of years. Such large-scale fluctuations
would probably lead to very rapid and violent climate change. If the
tropical regions became locked in a permanent ice cover in 100,000 years or
less, there would certainly be a mass extinction of great severity.
Is the lack of a large moon sufficient to prevent microbial life from
evolving into animal life? We have no information, but because deep-sea regions
are insulated from climate change, it seems doubtful that rapid obliquity
changes would deprive a planet of animal life. What it could do, however,
is deprive a planet of complex life on land.
The Moon, Jupiter, and Life on Earth
227
Lunar Tides
A second benefit of Earth’s large moon is tides, which are due to gravitational
effects of both the sun and the Moon. The pull of these two bodies
produces bulges in the ocean pointing both toward and away from the
Moon and the sun. The complexity of Earth’s present tidal effects is well illustrated
by the tidal charts cherished by clam diggers, anglers, and sailors.
The daily variations seen in the charts are caused by the interplay of both
lunar and solar tides. Both the Moon and the sun cause ocean bulges on their
respective near and far sides of the planet. As Earth spins under the bulges,
the sea rises and falls at any particular location. When the Moon lines up
with the sun every 2 weeks, the tidal ranges are at a maximum, and when
they are 90 degrees apart in the sky (the quarter moon is overhead at sunrise
or sunset), the range of tidal change is minimized. With a smaller or more
distant moon, the lunar tides would be lesser and would have a different annual
variation.
Soon after the Moon formed, it was perhaps 15,000 miles from Earth.
Instead of being a few meters high, as they are today, it is possible that lunar
tides rose hundreds of meters or higher. The extreme effects of such a close
moon could have strongly heated Earth’s surface. The ocean tides (and land
tides) from a nearby Moon would have been enormous, and the flexing of
Earth’s crust, along with frictional heating, may have actually melted the
rocky surface. However severe their effects, the enormous tidal variations
would have been short-lived because the forces responsible for tides also
cause the Moon to move outward, thus diminishing the effect. Early land
tides may have been a kilometer high, but they dropped to moderate levels in
less than a million years.
The retreat of the Moon is a natural consequence of gravitational pull
between the Moon and the tidal bulges. The Earth’s lunar tidal bulges don’t
actually correspond with a line from Earth to the Moon but, rather, lead
ahead as the Moon orbits the planet (see Figure 10.2). This offset produces a
torque that causes Earth’s spin rate to decrease slowly and the distance beR
A R E E A R T H
228
Force from the
leading bulge
Earth
Force from the
trailing bulge
Moon
Figure 10.2 Earth’s leading tidal bulge produces a constant forward force on the Moon that is not
totally balanced by the backward pull of the more distant trailing bulge. The net forward force on the
Moon causes it to spiral outward from its place of origin close to Earth. If the Moon had formed orbiting
in the opposite direction, this effect would have caused it to spiral inward to a catastrophic collision.
Such a dramatic fate awaits Triton, Neptune’s large and backward-orbiting moon.
tween Earth and Moon to increase. Besides measurement by laser, the recession
of the Moon can also be detected in the fossil record. Daily and annual
layers in Devonian horn corals show that about 400 million years ago there
were 400 days in each year, the Moon was closer, and Earth was spinning
faster. The coupling of these two effects is due to conservation of angular momentum,
the same physical law that allows ice skaters to spin faster by pulling
their arms against their bodies. The outward movement of the Moon would
be reversed if the Moon happened to be orbiting in the opposite direction.
Instead of retreating, it would approach Earth and would eventually collide
with it. Although we have nothing to fear from Earth’s moon in this respect,
The Moon, Jupiter, and Life on Earth
229
Triton, the large moon of Neptune, is in a retrograde orbit and will collide
with Neptune within a few hundred million years.
A New Account of the Moon’s Origin
A quite remarkable aspect of the Moon is that its formation appears to have
been highly unlikely, a rare chance happening. The origin of the Moon has
inspired endless speculation ever since people first gazed into the sky, but interest
in this subject peaked in 1969, when Apollo 11 landed on the Moon
and returned lunar rocks to earthbound laboratories. A major goal of the
feverish research activity on these samples was to determine how the Moon,
the “Rosetta Stone of the solar system,” had formed.
Before the Apollo rocks were brought back, the most popular notion was
that the Moon had formed cold and consequently would retain records of the
earliest history of the solar system. When the lunar samples were returned,
there was breathless anticipation that the samples would finally show how the
Moon formed. However, no one satisfactorily solved the mystery of the
Moon’s origin in 1969 or even in the next decade. Ironically, the assembly of a
well-agreed-upon theory for the lunar origin was slowed by the fabulous wealth
of highly detailed data from the samples. Extensive work did show that the
Moon had a violent, high-temperature history and was thus not a benevolent
body for preserving records of its earliest past, as had been hoped. The rocks
did, however, record exquisite details of the Moon’s history between 3 and 4
billion years ago, a time interval when Earth’s history is very poorly known.
During the Apollo program, everyone talked about the origin of the
Moon; in the following years, though, most lunar scientists worked on details,
and the “big picture” was little discussed. As happens at times in science,
this situation radically changed when a sort of cosmic convergence occurred
at a meeting on the origin of the Moon held at Kona, Hawaii, in 1984. At this
meeting, many details of the analysis of lunar samples and advances in theory
came to light, and many scientists left Kona convinced that the Moon had a
quite peculiar and improbable origin. Theories on the lunar origin usually fit
into three categories: it formed in place, it formed elsewhere and then was
R A R E E A R T H
230
captured, or it was somehow ejected from Earth. The new idea was, in a
sense, a combination of all three (see Figure 10.3). The model was that during
its formation, Earth was hit by a Mars-sized (half the diameter of Earth)
projectile. Debris from this collision was ejected into space, and some of it remained
in orbit, where self-collisions would cause it to form a thin, orbiting
ring of rocks analogous to the rings of Saturn. The moon would form from
this ring by collision and sticking, the processes of accretion that built most
of the bodies in the solar system.
Several important aspects of this process were consistent with the properties
of the Moon deduced from lunar sample studies. One was that the great
violence of the event would deplete the Moon of the so-called volatile elements.
Compared to meteorites, the moon is depleted in elements such as
zinc, cadmium, and tin. These relatively volatile elements would have been
vaporized in the impact, and the resulting vapor would have had difficulty
completely recondensing from hot gas. Stuck in the gas phase, the volatile elements
would have been swept into space and lost to the Earth–Moon system.
Included in the lost elements and compounds were nitrogen, carbon,
and water. One of the surprising initial findings of Apollo was that the lunar
samples were exceedingly dry. Unlike terrestrial rocks, they contained no detectable
water. Another remarkable property of lunar samples is that they are
highly depleted in the siderophile, or iron-loving, elements that tend to concentrate
in the metallic iron cores of planets. When planetary cores form by
the molten iron sinking to the planet’s center, the siderophile elements (such
as platinum, gold, and iridium) are incorporated into the falling iron and are
highly depleted from the crustal and mantle materials left above. That the
lunar rocks were depleted in siderophile elements was unexpected, because
the Moon cannot have a substantial iron core. The mean density of the Moon
is 3.4 times the density of water, very similar to that of rocks on the lunar surface
and much lower than that of Earth (5.5 times the mean density of water).
If it had a substantial core of dense metallic iron, the mean density of the
Moon would be higher than is observed. Seismic and magnetic data also
show no evidence of a significant core.
The Moon, Jupiter, and Life on Earth
231
Impact
Figure 10.3 Impact origin of the Moon as modeled by Cameron and Canup (1998). A body several
times more massive than Mars impacts the edge of the half-grown Earth with spectacular effects.
After a glancing blow, the two distorted bodies separate and then recombine. The metallic cores (light
gray) of both bodies coalesce to form Earth’s core, while portions of the mantles (black) of both bodies
are ejected into orbit and accumulate to form the Moon. After its formation the Moon spiraled outward,
a process that continues to the present time. To produce such a massive moon, the impacting body had
to be the right size, it had to impact the right point on Earth, and the impact had to have occurred at
just the right time in the Earth’s growth process.

R A R E E A R T H
232
The collision model solves the siderophile mystery by suggesting that
metallic cores formed in both Earth and the projectile before the collision. In
the collision, both cores ended up in the center of Earth, and the debris
ejected into orbit was mainly from the mantles of both bodies. This sequestering
of siderophiles explains why gold and platinum are so rare on the
Moon and in the crustal rocks of Earth. The impact ejection of mantle materials
from both the giant impactor and the target Earth is consistent with
some of the remarkable similarities between the trace element content of the
Moon and that of rocks from Earth’s mantle. It is also consistent with Earth
and the Moon being identical in isotopic composition.
A collisional origin is very attractive, but did it actually happen? In earlier
times, it was often imagined that the accretion of planets occurred in a
regular fashion by collision with small bodies. This was why the Moon was
thought to have formed cold. A body that grows by accretion of small objects
does not bury heat in its interior. Even though the bodies may collide at high
velocity, if they are small they produce small craters, and the energy of impact
can be largely radiated back into space. For an impact to eject enough
material to form the Moon, the colliding body has to be huge, a Mars-size
body. Theoretical modeling by George Wetherill, a Medal of Science–winning
planetary scientist at the Carnegie Institute of Washington Department of
Terrestrial Magnetism, showed that a natural consequence of the accretion
process is that several large bodies do form in each planet’s accretion zone.
The growth process includes the impact of several bodies each of which carries
more than 10% of the mass of the final planet.
In the case of Earth, these big bodies were the size of Mars and larger.
Their impact not only ejected material into space to form the Moon but also
injected considerable amounts of heat into Earth’s mantle. This heat input
and great violence led to the forging of Earth’s core during the accretion
phase, before the planet was fully formed. This is in contrast with a planet
that formed cold, by slow accretion of small bodies. Core formation requires
high internal temperatures so that blobs of molten iron descend through the
mantle to reach the core. Such a planet could form a core only after longThe
Moon, Jupiter, and Life on Earth
233
term buildup of radioactive heat from the decay of uranium, potassium, and
thorium. In Earth’s case, the early heat from accretion of large bodies led to
core formation as accretion occurred. The Moon’s formation occurred after
its core formed. Both bodies had differentiated and already had metal cores at
the time of collision.
Recent computer simulations by A. Cameron and colleagues provide
the best fit with the actual properties of Earth and the Moon when the Moonforming
collision occurs when Earth has only grown to about half of its final
mass and the impactor mass is about a fourth of the final mass of Earth. When
the collision occurred, the effects on both bodies were incredible. They
briefly fused, but then inertial effects on the resulting plastic mass literally
ripped the assembly into two major pieces. The fragments separated for several
hours but then fell back in response to the forces of gravity. After a few
violent oscillations, the planet finally settled down. Like droplets ejected
from a stone tossed into a pond, a small amount of material was ejected and
formed a debris ring around Earth. Materials in the ring were derived from
the silicate mantles of both bodies. The Moon formed from that disk by accretion
of solid particles over a period of a few tens of thousands of years.
When it finally formed, the Moon was only 15,000 miles from the
planet’s surface. With the Moon so close, Earth would have been spinning at
such a rate that the day would be only 5 hours long. The height of the tides
would have been fantastic, and as noted earlier, the resulting heat probably
melted the surface of the planet. The importance of this heat may be purely
academic, because Earth would still have been exceedingly hot just from the
great impact. The impact event was so energetic that it actually vaporized
rocks, forming a “silicate atmosphere” that survived for a short time before it
cooled and condensed as a silicate rain. All of these effects would have been
detrimental to any attempts by life to establish an early presence on the
planet.
Although this is somewhat of a conjecture, it is possible that Earth’s violent
early history seeded the eventual development of plate tectonics. The
large-scale heating would have led to formation of a magma ocean covering
R A R E E A R T H
234
the surface of the planet. Differentiation of this “ocean” may have spawned
the first rocks that could form long-term continents. This violent early history
must also have had severe effects on the ocean and atmosphere.
If the Earth’s formation could be replayed 100 times, how many times
would it have such a large moon? If the great impactor had resulted in a retrograde
orbit, it would have decayed. It has been suggested that this may
have happened for Venus and may explain that planet’s slow rotation and lack
of any moon. If the great impact had occurred at a later stage in Earth’s formation,
the higher mass and gravity of the planet would not have allowed
enough mass to be ejected to form a large moon. If the impact had occurred
earlier, much of the debris would have been lost to space, and the resulting
moon would have been too small to stabilize the obliquity of Earth’s spin axis.
If the giant impact had not occurred at all, the Earth might have retained a
much higher inventory of water, carbon, and nitrogen, perhaps leading to a
Runaway Greenhouse atmosphere.
Of the many elements of the Rare Earth Hypothesis, the presence of
our huge Moon seems to be one of the most important and yet most perplexing.
Without the large Moon, Earth would have had a very unstable atmosphere,
and it seems most unlikely that life could have progressed as successfully
as it has. Even with Earth’s relatively stable long-term climate, it still
took over 90% of the planet’s lifetime to date for land animals to develop.
Unfortunately, there is no evidence on how common large moons are for
warm terrestrial planets close to their parent stars. We just don’t know, and
we probably won’t for some time. Detection of terrestrial planets will be possible
in the coming decades, but detecting their satellites will be much more
difficult.
The Moon, our closest neighbor in space, has thus figured prominently
in the origin and evolution of life on Earth. Other solar system bodies,
though far more distant, have also had effects remarkable enough to suggest
that the conditions that promoted development of life on Earth are rare if not
unique. A fascinating case in point is a body over 500 million miles from us,
the planet Jupiter.
The Moon, Jupiter, and Life on Earth
235
J U P I T E R
Even in a small backyard telescope, Jupiter is quite a sight. In the eyepiece, it
is a disk noticeably flattened because of its high spin rate of two revolutions
per day. With parallel equatorial bands and whitish color, its appearance is
totally distinct from what Carl Sagan called our “pale blue dot” Earth, a blue
planet shrouded with wispy clouds. The most remarkable property of Jupiter
visible with a backyard telescope is the presence of four moons, pinpricks of
light that can be seen to move over a few hours’ time. The rhythmic mathematical
dance of the four largest moons, first observed by Galileo in 1612,
was a stunning scientific finding, because it resembled a miniature Copernican
solar system where orbital motion could be directly observed.
What cannot be seen by telescope, of course, is the exotic interior of
this planet. Jupiter is a giant gas ball that gets hotter and denser with depth,
but like the other giant planets in the solar system, it has no surface. Jupiter is
mostly hydrogen and helium, and deep in its interior, the pressure is so high
that electrons are not bound to individual hydrogen atoms but instead move
freely from atom to atom, as in a metal. At pressures of a million atmospheres
in the interior of Jupiter, hydrogen is actually in a metallic state.
It is enchanting in a small telescope, it has fascinating properties and a
rich history, but observed with the unaided eye, this giant planet is just a spot
of light—another “star” in the sky. From a distance of half a billion miles, it is
difficult to imagine (at least for an astrology agnostic) that this distant planet
with its frigid upper atmosphere could have any effect on Earthlings. Remarkably,
however, Jupiter’s existence and its time and place of formation
have profoundly influenced our Earth’s ability to provide and maintain a stable
environment for life.
A Giant’s Influence When the Planets Formed
Jupiter is ten times larger than Earth (and over 300 times its mass), and it is by
far the most massive planet orbiting the sun. The origins of Jupiter and its
R A R E E A R T H
236
neighbor Saturn (the “Jovian” planets) differ from those of other solar system
bodies in that these planets grew largely by direct accretion of gas from the
solar nebula in addition to accretion of solids. They are a more democratic
sampling of the nebula, and accordingly, they have elemental compositions
that are fairly similar to that of the sun, mostly hydrogen and helium. Jupiter
formed very quickly, and its rapid growth had major effects on the interior
planets, especially those that were trying to accumulate just inside its orbit.
For example, a terrestrial planet was in the process of forming about halfway
between Jupiter and the sun, but because Jupiter formed first, its development
was aborted. This failed planet is now the asteroid belt, a region where some
of the original planetesimals and their fragments still survive but were never
assembled into a real planet. The largest of these planetesimals is the asteroid
Ceres, a rounded object 1000 kilometers in diameter.
The asteroids are the source of meteorites, and detailed examination of
these ancient rocks provides deep insight into the nature of planet formation.
Most meteorites are ancient “rubble pile” mixtures of materials as old as the
solar system. They are the oldest radiometrically dated rocks. The meteorites
indicate that initially there was a period of growth when colliding materials
led to accretion but that later, during most of the solar system’s history, collisions
have occurred with such high energy that they have led to erosion and
disruption, not growth.
The effects that aborted the formation of a planet in the asteroid belt
also severely affected the formation of Mars. Mars is often described as the
most Earth-like planet, but in fact it is only half the size and one-tenth the
mass of Earth. Presumably, both Mars and the asteroid planet would have
grown to the size of Earth if the rapid growth of their giant neighbor had not
occurred. If this had been the case, the solar system might have ended up
with three truly Earth-like planets, each with oceans and advanced life forms
living on or near their surfaces. If Mars had been as large as Earth, it probably
would have retained a denser atmosphere, and, with the increased radioactive
heat that comes with additional mass, it is likely that it would have been more
volcanically active, perhaps driving plate tectonics. (Because of its smaller
The Moon, Jupiter, and Life on Earth
237
size, the volcanic activity on Mars is only a few percent of what occurs on
Earth.) A larger Mars would also have had a larger core, presumably producing
a larger magnetic field. One of the most critical shortcomings of the
planet Mars, from the point of view of hospitality to life, is that it almost entirely
lacks a global magnetic field. Thus electrically charged particles (the
solar wind) flowing outward from the sun played a major role in sputtering
the Martian atmosphere off into space. A substantial, Earth-like magnetic
field deflects the solar wind and protects the atmosphere from erosion.
If Earth had been a little closer to Jupiter, or if Jupiter had had a somewhat
larger mass, then the “Jupiter effect” that aborted the formation of the
asteroid planet and nearly ruined the formation of Mars could also have affected
Earth, rendering it a smaller planet. And if Earth had been smaller, its
atmosphere, hydrosphere, and long-term suitability for life would surely have
been less than ideal.
Following the discovery that Martian meteorites arrive at Earth at a
rate of half a dozen each year, some investigators have suggested that Mars
played a role in seeding Earth with life. The reasoning is that Mars is tougher
than Earth to sterilize globally. Ironically, this aspect of habitability is
caused by the lack of a Martian ocean. During the first half-billion years of
the history of the solar system, during what is called “the period of heavy
bombardment,” the terrestrial planets were hit by projectiles larger than 100
kilometers in diameter. On Earth, impacts of such magnitude vaporized part
of the ocean, and heat from the impact and the resulting greenhouse effect
could warm the entire surface of the planet to sterilization temperature. On
Mars, with no ocean, such an impact could cause great regional damage but
would not sterilize the whole planet. With its thin atmosphere, surface heating
would also be more rapidly radiated into space. The low total abundance
of water on Mars may thus be the result of these giant impacts, coupled with
the planet’s lower mass and surface gravity. If the early Mars did have
oceans, they may have been effectively ejected into space by impacts. Even
if there were more water on the young Mars, most of the early impact history
of Mars occurred on a planet that was dry compared to Earth. If life

R A R E E A R T H
238
evolved on both planets, it may have been destroyed on Earth, once or even
several times, while it survived on Mars.
There are sound reasons to believe that life may have formed during a
limited window of opportunity. This window of time may have closed before
the end of heavy bombardment on Earth. It is thus possible that the present
life on Earth is of Martian origin, transported to Earth by meteorites ejected
by major impacts. If Mars had been Earth-like, with oceans, then it too would
have been sterilized by impact. If Mars had been larger and had had a denser
atmosphere, it would also have been much more difficult for impacts to eject
meteorites into space.
A Distant Sentinel
Jupiter also played a crucial role in purging the inner solar system of bodies
left over from planet formation. Jupiter is 318 times more massive than Earth,
and it exerts enormous gravitational influence. Its gravitational interactions
very efficiently scatter bodies that approach it, and it has largely cleaned out
stray bodies from a large volume of the solar system. In the early solar system,
there were tremendous numbers of small bodies that had escaped incorporation
into planets, but over half a billion years, most of the larger ones inside
the orbit of Saturn disappeared. They were accreted by planets, ejected out
of the solar system, or incorporated into the Oort cloud of comets. Jupiter
was the major cause of this purging of the middle region of the solar system.
The objects that still impact Earth today are planetesimals that managed
to survive in three special ecological niches: the Oort comet cloud beyond
Pluto, the Kuiper belt of comets just beyond the outer planets, and
the asteroid belt, that special refuge located between Mars and Jupiter. The
current impact rate averages one 10-kilometer body every 100 million
years. The impact of just such a body occurred 65 million years ago, the
time of the K/T extinction that ended the age of the dinosaurs. George
Wetherill of the Carnegie Institute of Washington has estimated that the
flux of these 10-kilometer bodies hitting Earth might be 10,000 times higher
if Jupiter had not come into being and purged many of the leftover bodies of
The Moon, Jupiter, and Life on Earth
239
the middle region of the solar system. If Earth had been subject to collisions
with extinction-causing projectiles every 10,000 years instead of every 100
million years, and fairly frequently with even larger bodies, it seems unlikely
that animal life would have survived.
Do most planetary systems harbor planets like Jupiter? Ours has two
(Jupiter and Saturn), and the detection of Jupiter-mass planets around other
stars suggests that Jupiters exist in other planetary systems, but their frequency
is still unknown. It is likely that many planetary systems do not have
Jovian planets. The standard formation model of Jupiter has it accreting a
large, solid core before it can begin accreting gas, its main constituent. The
necessary conditions (available mass and rapid accretion before gas is lost
from the planetary system) may not be common, and so Jupiter-mass planets
may be rare in other planetary systems. When planetary systems lack a Jovian
planet to guard the outer boundary of the terrestrial planet region, the inner
planets may not be capable of supporting more than microbial life.
The Origin—and Fortuitous Stability—of Jupiter
Why and how did Jupiter form where it did? It is generally believed that
Jupiter’s formation began with the accretion of a solid core. This core grew by
collision and sticking of dust, ice, rocks, and larger bodies—a process similar
to the accretion of Earth. Jupiter, however, formed outside the “snow line,” a
special place in the solar system where water vapor condensed to form ice
grains and the presence of “snow” in this region would enhance the density of
solid matter and accelerate the accretion process. The mystery is why the
proto-Jupiter grew so rapidly. Apparently, Jupiter grew to a mass of 15 Earths
before Mars grew to 10% of an Earth mass. David Stevenson at Cal Tech has
suggested that outward migration of water vapor and condensation at the
“snow line” may have provided larger concentrations of condensed matter at
this location, thus speeding up the formation of the embryonic Jupiter.
Jupiter’s growth to a giant planet began when the rock–ice core mass
reached 15 Earth masses. At this mass, the gravity of the core can pull in and
hold hydrogen and helium, the light gases that account for 99% of the mass
R A R E E A R T H
240
of the nebula. When this gas accretion process begins, it is very dramatic because
the rate of accretion of gas is proportional to the square of the mass already
accreted. In other words, the bigger it gets, the faster it grows. If gas
could be continually fed to it, it would gobble up the Universe in a relatively
short time! What actually happens is that Jovian planet formation depletes its
feeding zone of matter, which in turn truncates planet formation. And although
the general properties of this process might be modeled, it just seems
to have been by chance that our Jupiter formed as it did.
Because it cleans our solar system of dangerous Earth orbit–crossing asteroids
and comets, Jupiter has had a beneficial influence on life on Earth.
However, it appears that we have been quite lucky that the Jupiter in our
solar system has maintained a stable orbit around the sun. A Jupiter and a
giant neighbor like Saturn are a potentially deadly couple that can lead to disastrous
situations where a planetary system can literally be torn apart. Recently,
it has become possible to use powerful computers to determine the
stability of the orbits of Jupiter and Saturn over the lifetime of the planetary
system. There are minor chaotic changes but no major changes, and the solar
system, at least to a first approximation, is stable over its lifetime. However,
this would not be the case if either Jupiter or Saturn were more massive or if
the two were closer together. It would also be dangerous to have a third
Jupiter-sized planet in a planetary system. In an unstable system the results
can be catastrophic. Gravitational perturbations among the planets can radically
change orbits, make them noncircular, and actually lead to the loss of
planets ejected into interstellar space. It is possible for chaotic disruption to
occur even after a system has been stable for billions of years, and in the
worst cases, planets can be spun out of the planetary system, escaping the
gravitational hold of the star. A life-bearing planet ejected into galactic space
would have no external heat source to warm its surface and no sunlight to
provide energy for photosynthesis. Although instability might start with just
two planets, the effects would spread to them all. In less severe cases, the orbits
of the planets would become highly elliptical, and the changing distance
between planets and the central star would prevent the persistence of conditions
required for stable atmospheres, oceans, and complex life.
The Moon, Jupiter, and Life on Earth
241
Numerical calculations first indicated that some planetary systems
might become unstable, and recent observations provide evidence that this
actually does occur. At the present time, planets are being discovered around
other stars by detecting small velocity changes of the central star. Of the
planets that have been detected, many are Jupiter-mass planets far from the
star, with highly noncircular orbits. This is quite different from the solar system,
where all giant planets have quite circular orbits. It is generally agreed
that the best explanation for the elliptical orbits is that these are planets
whose orbits have been altered by other planets, possibly by ejection of another
planet into interstellar space.
It is during the formation of planetary systems that Jupiter-like planets
pose the most draconian threat to terrestrial planets. With radial order
similar to that in the solar system, the terrestrial planets in the habitable
zone of typical planetary systems would form close to the star, the Jovian
planets farther out. There is reason to believe that this is the “natural way,”
because the formation of Jovian planets probably occurs only in the colder,
more distant regions outside the “snow line,” mentioned previously. It is
also to be expected that giant gaseous planets like Jupiter could not form
close to a star because tidal forces (the differential force of gravity) would
disrupt a planet in its diffuse, early stages. When a proto-Jupiter was very
diffuse and too close to the star, the differential force of gravity between
the sides near the star and far from it would pull the forming planet apart.
It was quite surprising, then, that several of the first extrasolar planets discovered
were found to be Jupiter-mass planets very close to their central
stars—closer than Mercury to the sun. All of these “hot Jupiters” have
highly circular orbits, and it is difficult to imagine that they actually formed
in these locations.
A popular explanation of this phenomenon is that the giant planets in
these systems actually did form at Jupiter-like distances but that their orbits
decayed and the planets spiraled inward. This cannot happen in an
evolved planetary system, but it can in the early, solar nebula phases when
extensive gas and dust still exist in the regions between the planets. Doug
Lin at the University of California at Santa Cruz has calculated that spiral
R A R E E A R T H
242
waves generated in the solar nebula phase can extract energy from a
young Jupiter and cause its orbit to spiral inward. In many cases the planet
actually hits the star; in others the inward drift stops before collision occurs.
The observed giant planets that are very close to stars may be examples
of this inward drift. Events such as this can be calamitous for terrestrial
planets. When a Jupiter spirals inward, the inner planets precede it
and are pushed into the star. If our Jupiter had done this, Earth would
have been vaporized long before life-tolerant conditions were ever established
on its surface. Lin has suggested that our solar system may have had
several Jupiters that actually did spiral into the sun, only to be replaced
with a newly formed planet. Perhaps Jupiter is at its “right” distance from
the sun only because it was the last one to form and it formed at a time
when the solar nebula had weakened to the point where orbital decay
ceased to be important.
The programs to detect extrasolar planets have revealed that nearly all
of the planets found either are “hot Jupiters” in circular orbits close to the star
or describe elliptical orbits farther from the star. All of these are “bad” Jupiters
whose actions and effects should preclude the possibility of these systems
having animal life on Earth-like planets in the habitable zones of the parent
stars. These life-unfriendly planetary systems have been found around 5% of
the nearby stars. The search techniques are most effective for detecting
Jupiters close to stars, however, and at present they cannot detect Jupitermass
planets at our Jupiter’s distance from the sun, nor can they detect planets
less massive than Jupiter. They could not presently detect our solar system
from the distance of nearby stars, and so it is possible that up to 95% of the
nearby solar-like stars have “regular” planetary systems similar to our own,
with terrestrial planets close to the star and Jovian planets in circular orbits
farther out. On the other hand, it appears that most other “Jupiters” so far detected
would have prevented the development of animal life anywhere in
their respective solar systems.
The Moon and Jupiter are two factors causing us to believe that complex
life requires disparate influences. We shall see, in the next chapter, how
we might put this hypothesis to the test.
Testing
the Rare Earth
Hypothesis
The Rare Earth Hypothesis is the unproven supposition that although
microscopic, sludge-like organisms might be relatively common in
planetary systems, the evolution and long-term survival of larger, more
complex, and even intelligent organisms are very rare. The observations on
which this hypothesis is based are as follows: (1) Microbial life existed as
soon as Earth’s environment made it possible, and this nearly invincible form
of life flourished over most of Earth history, populating a broad range of hostile
terrestrial environments. (2) The existence of larger and more complex
life occurred only late in Earth history, it occurred only in restricted environments,
and the evolution and survival of this more fragile variant of terrestrial
life seem to require a highly fortuitous set of circumstances that could not be
expected to exist commonly on other planets. This hypothesis can be tested.
Throughout human history, people have wondered what lies beyond
the limits of the known world. This instinctive obsession has driven humans
(and perhaps other species) to expand their own territories. This haunting
question permeates mythology and religion and has provoked some of the
deepest of human thoughts. In earlier times, the phrase beyond the known world
may have referred to regions only hundreds to thousands of miles distant. In
243
11
R A R E E A R T H
244
modern times, these musings extend to actual worlds—to other planets. Over
the past century and a half, the great advances in science and in our understanding
of nature and physical processes have refined our ability to imagine
other worlds realistically and evaluate the possibilities of life beyond Earth.
We now actually have the knowledge and technological tools to begin the
serious search for alien life, and for the first time in history, we have the capability
to test the Rare Earth Hypothesis. The tests that can be done are of
two types. One consists of efforts to detect the presence of microbial life in
other bodies in the solar system. The discovery of living microorganisms or
fossil evidence of microorganisms would support the contention that microbial
life originates readily, that it forms frequently, and that we might expect
it to occur in numerous bodies that have warm, wet environments somewhere
in their interiors. The search for microbial life can be done in the solar system
by sending specialized probes to seek life directly with in situ analysis techniques.
The second test of the Rare Earth Hypothesis is the search for evidence
of advanced life forms, which might range from simple, multicellular
organisms to large animals. We see no evidence of advanced life in the solar
system, except on Earth, so the main search for advanced life will focus on
planetary systems around nearby stars. These searches will be conducted via
large, space-borne telescopes. Both the in situ detection of microbial life and
the telescopic detection of advanced life are in the planning stages, and both
have high priority with funding agencies in the United States and Europe.
This is a very exciting time. It is our first opportunity to actually study the
processes that lead to the origin, evolution, and survival of life in the Universe.
A D V A N C E D L I F E
An alien astronomer, viewing Earth from a great distance, could detect the
presence of life on the planet with comparative ease. This could not be done
Testing the Rare Earth Hypothesis
245
directly, by imaging, but rather indirectly, by spectral analysis of the composition
of the atmosphere. Even with the most grandiose alien telescopes, it is
doubtful that extraterrestrial astronomers could directly detect organisms,
groups of organisms, or even the most immense structures life has formed:
coral reefs, forests, forest fires, red tides, city lights, the Great Wall of China,
freeways, and dams. At best, images of Earth would only slightly resolve what
Carl Sagan referred to as the “pale blue dot.” The principal clue alerting distant
astronomers to the presence of life would be a spectacular and unmistakable
signature. Spectral analysis of infrared light would reveal that life plays
such a major role on the planet that it controls the composition of the atmosphere.
Spectrum of a Life-Bearing Planet
Earth’s atmosphere would actually be quite “unnatural” for a nonbiotic planet.
It is clearly different from the nearly pure carbon dioxide atmospheres of its
neighbors, Mars and Venus. The mix of nitrogen, oxygen, and water vapor is
chemically unstable and would never arise on a dead planet. Without life, nitrogen
and oxygen in the presence of water would combine to form nitric
acid and become a dilute acidic component of the ocean. Earth’s peculiar atmosphere
is not in chemical equilibrium, and it succeeds in disobeying natural
chemical laws only because of the presence of life. The most peculiar aspect
of the atmosphere is the abundance of free oxygen. Oxygen is the most
abundant element in the whole Earth (45% by weight and 85% by volume!),
but in the atmosphere, it is a highly reactive gas that would exist only at trace
levels in the atmosphere of a terrestrial planet devoid of life. Oxygen is a poisonous
gas that oxidizes organic and inorganic materials on a planetary surface;
it is quite lethal to organisms that have not evolved protection against
it. The source of atmospheric oxygen is photosynthesis, the miraculous biological
process that utilizes the energy of sunlight to convert carbon dioxide
to pure oxygen and organic material. Ironically, it was the long-term photosynthetic
production of this poisonous gas, and life’s adaptation to it, that
R A R E E A R T H
246
made complex and energetic life possible on Earth. Except for the noble gas
argon, all of the major atmospheric constituents are also processed and recycled
on short time scales via biological processes.
Our distant alien astronomers would realize that life exists on Earth as
soon as they detected, in its infrared spectra, absorption bands due to the
presence of carbon dioxide, ozone, and water vapor (see Figure 11.1). Nitrogen
and normal oxygen (O2) are the major atmospheric gases, but they do not
produce detectable absorption effects. The telltale bands arise because of the
way Earth interacts with sunlight. Earth’s surface is warmed by visible sunlight,
and it reradiates in the infrared. The incoming energy is in the visible region
of the spectrum (near 0.5 micrometer in wavelength) where the sun, with its
surface temperature of 5400°C, emits most of its energy. The atmosphere is
largely transparent to visible wavelengths, and the light that is not reflected
into space is largely absorbed by Earth’s surface. This energy heats the surface
to “room temperature”—only about 5% of the absolute temperature of the surface
of the sun. Earth’s surface cools itself to balance exactly (averaged over
time) the absorption of sunlight by radiating infrared radiation back into
space. Because of Earth’s relatively cooler temperature, the bulk of this energy
is in the “thermal infrared” spectral region near 10 micrometers in wavelength.
Atmospheric transmission of parts of this spectral region are blocked because
of absorption by certain gases. Water vapor, ozone, and carbon dioxide absorb
part of the outgoing infrared radiation and block its escape from Earth.
This process and these same gaseous species are the root cause of the atmospheric
greenhouse effect that prevents Earth’s oceans from freezing. All of
these “greenhouse” molecules are minor constituents, but they cause warming
effects of some 40°C above the temperature of an atmosphere that is totally
transparent in the infrared. They also provide a very strong spectral signature
to be seen by alien astronomers. Water makes up a few percent of the atmosphere,
CO2 is currently only 375 parts per million, and ozone occurs only in
the parts-per-billion range. Although rare, they absorb significant chunks of
the infrared radiation streaming into space. The outgoing infrared has significant
absorption dips at wavelengths of 7, 10, and 15 micrometers, respectively,
for water, ozone, and carbon dioxide.
Testing the Rare Earth Hypothesis
247
Seeking Life’s Spectral Signature
Observation of ozone, CO2, and H2O in the atmosphere of an unknown
planet strongly suggests habitable conditions and the presence of life. Water
vapor at moderate levels can indicate that a planet is essentially in the habitable
zone of a star. Water vapor abundance depends on the temperature of an
atmosphere as well as on the availability of surface water. CO2 abundance
also provides a clue to the “habitability” of a terrestrial planet. At least on
Earth, the ability to keep this dangerous greenhouse gas locked in sediments
requires moderate surface temperatures and an active land–ocean weathering
cycle, whereby “excess” CO2 can be sequestered in carbonates and thus kept
out of the atmosphere. The strengths of the CO2 and H2O absorptions provide
Intensity
Water vapor
Ozone Carbon dioxide
Wavelength (micrometers)
7 8 9 10 11 12 13 14 15 20
Figure 11.1 The hypothetical infrared spectrum of an Earth-like planet with life. The abundances of water
vapor, carbon dioxide, and ozone would be clues indicating that the planet was in the habitable zone of its star
and that life was producing oxygen.
R A R E E A R T H
248
strong clues for identifying planets that are truly Earth-like, with land,
oceans, and moderate surface temperatures. The detection of ozone signals
the presence of active, photosynthetic life. Ozone (O3) is a molecule composed
of three oxygen atoms. It is a highly chemically active molecule, so it
is not very stable. It is produced in the atmosphere by the interaction of ultraviolet
light from the sun and normal oxygen (O2). Light splits O2 into individual
atoms, and they in turn react with O2 to form ozone. Only a minuscule
fraction of the atmosphere’s oxygen is in the form of ozone, but it
provides strong infrared absorption. Ozone implies the presence of oxygen,
which, in sufficient concentration, implies the existence of life. Maintaining
moderate oxygen levels requires continuous production to balance the many
processes that lead to its removal from the atmosphere.
Several space-borne projects have been proposed to detect infrared
spectral signatures that would suggest Earth-like planets, oceans, moderate
surface temperatures, and the presence of biological activity. A project in the
planning stage at NASA is the Terrestrial Planet Finder, and a similar project
being studied by the European Space Agency (ESA) is appropriately called
Darwin. Both of these missions would use very large telescopes to image terrestrial
planets and take infrared spectra. The basic challenge is to measure
the spectral signature of terrestrial planets close enough to their parent stars to
be in the habitable zones. This is a formidable task, in part because of the faintness
of planets but also because of the proximity of these planets to the star.
Viewed from a great distance, Earth itself would look very close to the sun and
comparatively very faint. From the distance of the nearest stars, the angular
separation of Earth and the sun is comparable to the diameter of a quarter as
viewed from a distance of 4 miles. This is near the limit of what can be resolved
with conventional ground-based telescopes, but it is easily within the reach of
space telescopes and of ground-based telescopes with adaptive optics to reduce
the blurring effects of the atmosphere.
The major problem with the study of extrasolar planets, even for nearby
stars, is that planets are much fainter than stars. At visual wavelengths, Earth
is seen only by the sunlight it reflects. The sun is a billion times brighter than
Testing the Rare Earth Hypothesis

249
the reflected “Earth-light,” and in any telescope system, the glare from such
an intensely bright object overwhelms the image of a faint nearby planet. In
the infrared, at wavelengths of 10 micrometers, the situation is better. The
sun is much fainter, and this is the peak of the region where the warm Earth
radiates energy back to space. At this wavelength the sun is only ten thousand
times brighter than Earth, and separation of the two images is feasible
via special techniques involving interferometry.
The grand plan for detecting life in extrasolar planets involves the construction
of very large telescopes. It is not efficient—or perhaps even
possible—to build a single-mirror telescope large enough to detect individual
extrasolar planets, and this limitation suggests using groups of smaller telescopes
combined to work in concert. The ability to resolve ultrafine angular
detail depends on the diameter of the telescope. Telescopes together in an
array have resolving power equivalent to a single mirror as large as the entire
array. Present plans for the Terrestrial Planet Finder call for a cluster of four
telescopes, each with an individual mirror 4 to 8 meters in diameter. These
could be mounted on a truss or could be free flyers with a total separation of
about 100 meters. In either case, the separation of the individual telescopes
would have to be controlled to incredible precision: small fractions of the
wavelength of light. The telescopes would combine light beams and would
function as an interferometer with a very special property. Their sensitivity
would be a minimum in the center of the imaged field and a maximum at a
slight offset equal to the expected angular spacing of a planet and the star.
When the telescopes were then pointed directly at the star, the image of the
star would effectively be attenuated by a factor of a million, whereas the light
from a nearby planet would not be attenuated.
This special design produces a “null” on the star, minimizing the great
difference in brightness between planet and star. The technique uses interference,
the same process the produces the iridescence of soap bubbles and
the brilliant color of some butterfly wings. A laser pointer projected onto a
distant wall shows an analogous effect in reverse. It has a bright spot in the
middle that is surrounded by faint dark and bright rings. The planet finders
R A R E E A R T H
250
will use interference to produce the opposite effect: a null in the center and
sensitivity in rings around it. Doing precision interferometry with such huge
telescopes in space will require extraordinary effort and billions of dollars.
The technique is just at the edge of available technology but appears practicable.
The system could search for Earth-like planets around several hundred
of the nearest stars, and in it lies our best hope for detecting life outside the
solar system any time soon.
Seeking Intelligent Life
Another approach to the search for extrasolar life is the detection of radio
signals sent by other civilizations. The modest attempts to pick up any such
signals have generated much interest, speculation, and debate. It is true that
the most powerful radio telescopes on Earth could receive signals from similar
telescopes, aimed directly at Earth, from any other spot in the galaxy.
Considerable thought has also been devoted to what wavelengths would be
used for communication and what types of information might be sent. The
laws of physics and radio propagation in the galaxy suggest that the best
wavelengths are in what is known as the “water hole” near 20 centimeters.
This activity is commonly referred to as SETI, the Search for ExtraTerrestrial
Intelligence. In 1990, NASA funded a modest SETI effort, but its budget was
cut after only a few years, before the program got seriously under way. Senator
Proxmire gave SETI one of his famed Golden Fleece awards and fumed
“not a penny for SETI.” Others in influential positions were concerned about
ridicule of a program that might run for decades or even millennia seeking
faint radio signals from other civilizations, and public funding for SETI
searches has been very limited. (Just as it is a problem on Earth, funding
would probably be a critical factor on other worlds too. On Earth, in our
most economically prosperous times, we cannot even listen. Sending the signals
would be much more complex and costly.)
Unfortunately, it is very difficult to know if SETI is an effective use of
resources. If the Rare Earth Hypothesis is correct, then it clearly is a futile efTesting
the Rare Earth Hypothesis
251
fort. If life is common and it commonly leads to the evolution of intelligent
creatures that have long, prosperous planetary tenures, then it is possible that
enlightened aliens might be beaming signals off into space. A key factor in
deciding whether SETI makes sense involves the lifetime of civilizations with
radio technology. Does such a civilization last only centuries before nuclear
war, starvation, or some other calamity causes its decline? Or does it last forever?
In the most optimistic minds, “Star Trek” societies might populate the
stars. But even if they do, it is a real question whether any of them would or
could beam enormous amounts of radio power into space to potential audiences
that are prevented by the vast interstellar distances from ever returning
the message in a timely manner. There probably are other civilizations in the
galaxy that have radio telescopes, but the vast numbers of stars and the vast
distances involved are barriers that may always keep SETI more an experiment
of the imagination than a large-scale scientific endeavor. An exception
might be made for the limited number of nearby stars that have planetary systems.
If some of these are found to have Earth-like planets with atmospheric
compositions indicative of life, then the public might support either sending
signals or listening. And, of course, even though we do not intentionally
beam radio messages to nearby stars, Earth is a potent transmitter of radio
power emanating from radar, television stations, and other sources.
MI C R O B I A L L I F E I N T H E S O L A R S Y S T E M
The search for microbial life in the solar system began in earnest with Apollo
11. Although it was clear that the Moon was not a teeming abode of life, it
was thought that the Moon might provide clues to early life or at least to prebiotic
chemical conditions. The astronauts and the samples they collected
underwent elaborate quarantine, lest lunar microbes attack Earthlings like the
disastrous diseases carried across the Atlantic Ocean just 450 years earlier.
Before Apollo, some thought that the Moon was similar in composition to
primitive meteorites—that it might contain abundant carbon and water in
R A R E E A R T H
252
the form of hydrated minerals. A popular theory for the Moon’s origin was
that it formed elsewhere and was captured during a close encounter with
Earth. This theory had it that the Moon initially had a much smaller orbit and
then retreated outward in response to tidal effects. Harold Urey, the Nobel
Prize–winning chemist (and one of the leading pioneers in the field of planetary
science), imagined that the Moon would have passed so close to Earth
that the immense tidal interaction would have resulted in parts of the ocean
sloshing into space and landing on the lunar surface. Although few believed
that any living organisms could flourish in the harsh environment of the airless
Moon, Urey thought the Moon might retain critical records of prebiotic
chemistry and desiccated remains of the earliest forms of life on Earth. Urey
called the Moon the Rosetta Stone of the solar system.
When the Apollo 11 samples were returned, the first tests were toxicological,
to see whether the samples had any dire effects on terrestrial life.
Some of the priceless cargo of lunar soil was fed to rats and placed in the root
systems of growing plants. No negative effects were observed, and detailed
analysis of the rocks and soils revealed no organic material of biological origin.
There was carbon, but it all appeared to have been derived from impacting
meteoritic bodies and implantation by the solar wind. As mentioned
above, the lunar samples were extraordinarily “dry”; they contained no bound
water. The Moon was found to be a lifeless body that did not even contain
the building blocks of life or a life-supporting environment.
The Viking program was the only space mission that directly included
life detection among its goals. This extraordinary program involved four
spacecraft: two that landed on the surface of Mars for detailed in situ studies
and two that went into orbit for global-scale mapping and to relay lander information
to Earth. With the possible exception of the Hubble Space Telescope,
Viking was the most expensive NASA mission launched purely for scientific
exploration. (Apollo had a large scientific component, but the mission
was largely motivated by national priorities—getting to the moon first.) The
Viking missions cost about 4 billion 1999 dollars and required robotic spacecraft
to land on another planet to conduct chemical searches for the presence
Testing the Rare Earth Hypothesis
253
of life. The first mission landed on Mars in 1976, the American Bicentennial,
and many of the scientists involved referred to each other as the Vikings of
’76. The large, cowboy-style brass belt buckles that many of them wore, with
the mission logo engraved thereon, are still seen at various meetings of planetary
scientists and engineers.
Viking was an enormously difficult and successful mission. And yet, in a
sense, the Viking mission was a failure in that it did not detect life. Not only
did it not detect life, but the results revealed the Martian surface to be a
highly inhospitable environment for life. There was less carbon in the soil of
Mars than there was on the Moon, and worse, the presence of highly oxidizing
conditions indicated that organic material could not survive in the soil. If
a dead mouse were buried in shallow Martian soil, its carbon would be converted
to carbon dioxide, which would flow into the atmosphere. The results
from Viking drove many nails into the coffin of the belief that Mars was an
Earth analog that might harbor life.
The Viking missions carried three major life detection experiments.
Each was a miniature, highly specialized chemical laboratory designed to detect
chemical changes characteristic of biological activity. Each lander had a
retractable arm with a scoop at the end. One of the joys of the mid-1970s was
watching these scoops actually dig trenches and collect samples on Mars, the
famous red planet of so many science fiction stories. The scoops would dig
soil samples and drop them through a screen into the analysis instruments.
The major life detection experiments were the gas exchange (GEX), labeled
release (LR), and pyrolitic release (PR) techniques. The first data, 8 days after
the Viking I landing, came from the GEX experiment, and the results were
positive—or appeared to be so. A gram of soil was placed in the experimental
chamber, and a small amount of water and nutrient was added. Two days
later a large amount of newly generated oxygen gas was detected, an expected
signal of biological activity. The LR experiment also yielded positive
results only a day later. In the labeled release experiment, water and nutrient
labeled with radioactive carbon-14 was added to a soil sample, and the equipment
recorded whether C14-labeled carbon dioxide or methane was released.
R A R E E A R T H
254
The signal was again positive, and—startlingly more positive, in fact, than in
many soils on Earth! In the pyrolitic experiment no nutrient or water was
added, but the soil was exposed to C14-labeled carbon dioxide and to carbon
monoxide gas and light. After an exposure, the soil was heated (pyrolized) to
see whether C14-labeled material would be released from any newly formed
organic compounds. There was a weak but positive signal.
In spite of high hopes and expectations, however, the Viking scientists
had built in some backup tests. The instruments were designed so that multiple
experiments and samples could be run repeatedly, as in a “real” Earthbased
laboratory. Repeated tests showed that the “positive” results could be
attributed to unusual chemical properties of Martian soil. With no ozone
layer to block it, the harsh ultraviolet light from the sun lands directly on the
soil and produces highly oxidizing and reactive compounds, such as peroxides,
that can produce the reactions observed. After severe heating that
would have killed any terrestrial organism, the soils still yielded “positive signals.”
The Viking team’s interpretation of their data indicate that instead of
the actions of living Martian organisms, the observed results were caused by
surface chemical reactions of nonbiological origin.
The Viking landers did not convincingly detect Martian life, but they
did show how difficult it is to identify microbial life, with unknown properties,
on a planet that is quite different from Earth. Viking was capable of detecting
organisms in most of Earth’s surface materials, but our planet teems
with life, and a gram of typical soil contains over a billion individual organisms.
Viking found that the surface soils on Mars did not and could not support
any of the forms of life found on Earth. If life does exist on Mars, we
must look for it in subterranean regions beneath the frozen “cryosphere” at
depths where liquid water can persist. Future missions cannot simply look for
life in spoonfuls of surface soil; they must search the warm, wet regions beneath
the inhospitable surface. To directly reach wet rock, future searches for
living organisms will have to drill. And drilling must not be done just anywhere,
because the frozen cryosphere normally extends to depths of several
kilometers. Instead of attempting to drill so deeply, future missions will
Testing the Rare Earth Hypothesis

255
search out rare geothermal hot spots where liquid water might reach close to
the surface. These missions will search the “Yellowstones” of Mars. It is also
possible that samples of life-bearing rocks might be found on the surface as
debris excavated by impact craters. Studies of terrestrial and lunar craters
have shown that large rocks—some the size of houses—can be lifted from
considerable depth and deposited on the crater rim. Organisms could not live
in these cold dry rocks, but they might survive in a dormant state for thousands
or even millions of years.
Although finding living creatures that could be observed to reproduce
would be the most convincing discovery, the next searches for life on Mars
will have less ambitious goals. They will search for fossils or for chemical,
isotopic, or mineralogical indicators of past life. Even if Mars is currently a
totally sterile planet, it may well have harbored life in its distant past. Channels
and other surface features suggest that Mars was much more Earth-like
three or four billion years ago. Mars occasionally had liquid water on its surface,
and it probably had lakes or larger bodies of water persisting for moderate
periods of time beneath thick ice crusts. If the Rare Earth Hypothesis
is correct and life forms readily, then we would expect life to have evolved
on Mars during the early period of its evolution when it had more Earth-like
surface conditions. The search for life on Mars is a key test of this hypothesis.
Searching for microscopic fossils or other indicators of life is enormously
complex and is difficult to do in a convincing way with spacecraft instruments.
Space instruments must be designed to perform very specific tasks.
Constraints on power, mass, cost, reliability, and remote operation in a hostile
environment mean that the kitchen sink and most of the other items that
scientists would like to include on the spacecraft are inevitably left behind.
Space instruments are usually marvels of their time, but their capabilities
rarely are competitive with those of their heavy, power-guzzling and inelegant
siblings that are used for day-to-day work in Earth-bound laboratories.
The most important limitation of spacecraft instruments is their inflexibility
to adapt to new findings. They usually do what they are designed to do, but
R A R E E A R T H
256
not more. This differs considerably from normal laboratory studies, where
initial results provide new insights and lead to investigations that were not
previously anticipated. For these reasons, the most detailed searches for life
and fossils on Mars will require sample return. Searches for evidence of life in
Martian meteorites have produced intriguing results and have shown what
types of investigations could be done on returned samples. The first such mission
is now planned for launch in 2005. Although they pose an enormous
technical challenge, the Martian sample return missions offer the best current
hope for finding evidence of life on Mars. Once returned to Earth, the samples
will be examined with the most sensitive instruments we have for clues
of the past existence of microbial organisms. Even if life exists on Mars, of
course, its discovery may require a series of exploration and sample return
missions.
In addition to Mars, there are many other bodies in our solar system that
might have microbial life. These include the three outermost large moons of
Jupiter (Europa, Ganymede, and Callisto), the large moon of Saturn (Titan),
and possibly other moons as well. Europa presently is the most attractive
prospect apart from Mars. Images of its surface show a complex landscape of
shifting ice and mysterious ridges. Heated by tidal energy, liquid water lurks
beneath the surface. Lesser amounts of water or brine are also thought to exist
inside Callisto and Ganymede. Although twice as far away, Titan is also an
exploration target of great interest. Its dense nitrogen atmosphere and
hydrocarbon-rich surface is tantalizing—in spite of cryogenic temperatures
at its surface. In 2004 the Cassini mission will parachute an instrumented
package onto the surface of Titan. It is not designed to detect life, but this
probe will measure environmental parameters that are important to life.
Most of our discussion up to this point has been couched in relatively
qualitative terms. It’s time now to look at some efforts that have been made to
quantify the probability that life will evolve and persist, and to suggest a few
numbers of our own. This is the subject of the next chapter.
Assessing
the Odds
Indeed, the only truly serious questions are ones that even a
child can formulate. Only the most naïve of questions are
truly serious.
—Milan Kundera, The Unbearable Lightness of Being
“Do you feel lucky? Well do ya?”
—Clint Eastwood, Dirty Harry
How rare is Earth? We have arrived at the end of a long grocery list
of ingredients seemingly necessary to make a planet teeming with
complex life. It involves material, time, and chance events. In this
chapter we will try to assess these various factors and their relative importance;
all can be thought of as probabilities. In some cases we understand
these probabilities, but in others almost no research has been done, and our
questions, like those referred to in the quotation above, are the simple questions
of children—questions as yet with no answers. Some of these questions
can thus be tackled only with our imagination. Others will be answered by the
257
12
R A R E E A R T H
258
space voyages and instrumented investigations we discussed in the previous
chapter.
Let us begin by imagining we have the power to observe 100 solar nebulae
coalesce into stars and the planets that will encircle them. How many of
these events will yield an Earth-like planet with animal life?
As we have seen, the first step in preparing the way for a habitable environment
is the formation of a suitable star: one that will burn long enough
to let evolution work its wonders, one that does not pulse or rapidly change
its energy output, one without too much ultraviolet radiation, and most important,
perhaps, one that is large enough. Of the 100 applicants, perhaps
only two to five will yield a star as large as our sun. The vast majority of stars
in the Universe are smaller than our sun, and although smaller stars could
have planets with life, most would be so dim that Earth-like planets would
have to orbit very close to their star to receive energy sufficient to melt
water. But being close enough to get adequate energy from a small star leads
to another problem: tidal lock, the condition where the same side of the
planet always faces the sun. A tidally locked planet is probably unsuitable
for animal life.
What if we increased the number to 1000 planetary systems, so that we
might expect 20 stars of our sun’s size or greater to be born? Even these numbers
are too small to yield a high probability that we will find a truly Earthlike
planet. Perhaps a better way to envision the various odds is to re-create
the scenario that led to the formation of our solar system and then run
through the process once again in a thought experiment. Stephen Jay Gould
used this type of mental reconstruction in his interpretation of the Cambrian
Explosion. In his 1989 book Wonderful Life, Gould described the exercise as
follows:
I call this experiment “replaying the tape.” You press the rewind
button and, making sure you thoroughly erase everything that actually
happened, go back to any time and place in the past—say,
to the seas of the Burgess Shale. Then let the tape run again and
see if the repetition looks at all like the original.
Assessing the Odds
259
A THOUGHT E X P E R I M E N T
In our case, we will replay the tape of our planet’s formation. We begin with
a planetary nebula of exactly the mass and elemental composition that created
our solar system. According to most theorists, this might create a star
identical to our sun—but then it might not. For instance, the spin rate of the
new star might be different from that of our sun, with unknown consequences.
Well, then, 1000 such solar nebulae, perhaps 1000 clones of dear old
Sol. Not so, however, with the planets coalescing out of this mix. If we rerun
this particular tape, we will in all probability not get a repeat of our solar system
with its nine planets, its one failed planet (now the asteroid belt), a
Jupiter and three other gas giants orbiting outside four terrestrial planets; and
a halo of comets surrounding the entire mix. Now we enter the realm of multiple
contingencies. Of the 1000 newly formed planetary systems, none is
likely to be identical to our solar system today—just as no two people are
identical. In a coalescing planetary system many processes, including planetary
formation, may be chaotic.
Planets form in what are known as “feeding zones,” regions where various
elements come together and eventually coalesce into planetesimals,
which finally aggregate into a planet. Recent work by planetary scientists
shows that the spacing of planets will probably be fairly regular. There might
be as few as six planets or as many as ten or even more. James Kasting of Penn
State University believes that planetary spacing is not accidental—that the
positions of planets are highly regulated, and that if the solar system were to
re-form many times, we would get the same number of planets each time. Yet
the observational evidence to date does not back up the theory. The extrasolar
planets that have been discovered exhibit an enormous diversity of spacing
and orbits; their positions are not nearly so orderly as the theory suggests
they should be. Ross Taylor, an astronomer who received the prestigious
Leonard Award in 1998, disputes Kasting’s views. “Clearly,” he maintains,
“the conditions that existed to make our system of planets are not easily reproduced.
Although the processes of forming planets around stars are probably
broadly similar, the devil is in the details.”
R A R E E A R T H
260
No one knows whether a planet the size of Jupiter would always form
or whether there would be a couple of planets like Mars instead. A planet
would probably form in about the position of Earth, but it could be larger or
smaller, somewhat closer to the sun or father away. Would the material
(physical) quantities be essentially the same? Would plate tectonics develop?
Would there be the same amount of water—and would that water end up on
the surface of the planet, rather than locked up in its mantle or lost to space?
Would there be few threats to life from Earth orbit–crossing asteroids? What
is the chance that our Moon would form again, if, as we believe, it is important
in making Earth a stable place conducive to animal diversification?
Even if all of these events occurred more or less the way they have,
would life form again? And given life, would animal life appear once more?
Can there be animal life without the utterly chance events that occurred in
Earth’s history, such as a Snowball Earth or an inertial interchange event, for
instance?
Let us reorganize (and rephrase) this set of questions in the following
way. We might ask: How many of all planets in the Universe are terrestrial
planets (as opposed to the gas giants such as Jupiter, for instance)? What is the
percentage for all planets of the Universe? (In our solar system there are five,
but if we add the larger moons, that number more than triples.) Of the terrestrial
planets in the Universe, how many have enough water to form an ocean
(either as water or as ice)? Of those planets with oceans, how many have any
land? Of those with land, how many have continents (rather than, say, scattered
islands)? Yet these questions are only for the infinitesimally small slice of
time we call the present. All of these conditions are subject to change.
B U Y I N G T I M E : T H E P E R S I S T E N C E O F
OC E A N S AND MO D E R A T E T E M P E R A T U R E
As we have tried to show in the preceding chapters, the most important lesson
from Earth’s history is that it takes time to make animals—long periods
of environmental stability with global temperatures staying well below the
Assessing the Odds
261
boiling point of water. Hence we need to add the time component to
each question. For instance, what percentage of planets that have oceans
keep them for a billion years, or 4 billion years, or 10 billion years, for that
matter?
Of all the factors important in assessing the odds of once again getting
(or finding) a world with animal life, one factor stands paramount: water.
Earth succeeded in acquiring its ark-load of animals and complex plants—and
then keeping them—for more than half a billion years (so far) because it retained
its oceans for more than 4 billion years. Moreover, if our analysis of
the sedimentary record is correct, for the last 2 billion years it maintained the
oceans at average temperatures less than 50°C. Also—at least for the last 2
billion years—the oceans have been maintained at a chemical composition
conducive to the existence of complex animal life: at a salinity and pH favorable
to the formation and maintenance of proteins. The oceans are clearly the
cradle of animal life—not fresh water, not the land, but the saltwater oceans
have spawned every animal phylum, every basic body plan that exists or has
existed on our planet.
Discovering how Earth acquired its supply of water is one of the most
critical concerns of the new field of astrobiology. As we pointed out in an earlier
chapter, water was not abundant in the inner regions of the solar system
when the planets formed. There was far more water in the outer regions of the
solar system than among the inner planets. Where did our water come from?
Although where our oceanic water came from is still the subject of debate,
everyone agrees that it must have arrived during planetary accretion,
with perhaps significant volumes added during the period of heavy bombardment.
Ironically, the volume of water eventually found on Earth may be related
to the formation of Earth’s core. When the iron- and nickel-rich core
formed, most of the water found in the coalescing planet was consumed in
oxidation processes whereby oxygen bound up in water was used to make
iron and nickel oxides. It is the residual water that makes up the oceans. Perhaps
that residual quantity was significantly enhanced by water carried by
comets after Earth’s initial formation, perhaps not. In either case, the oceans
reached approximately their present volume by 3.8 billion years ago. But this
R A R E E A R T H
262
does not mean they were at their present area. Don Lowe of Stanford University
has estimated that before 3 billion years ago, less than 5% of the surface
was land. Until about 2.5 billion years ago, the chemistry of this worldgirding
ocean was controlled largely by interactions with the oceanic crust
beneath it and with Earth’s mantle, whose by-products interacted in the
oceans at mid-ocean ridges and rifting areas. It is estimated that because this
early Earth was much warmer than the Earth we know, the area of this zone
of ocean-mantle contact was as much as six times that found today.
Earth’s atmosphere was also very different from that of today. There
was no oxygen, and there was a great deal more carbon dioxide—perhaps
100 to 1000 times as much as today. Earth’s surface temperature was higher
than it is now because more heat was emanating from the interior and because
of the warming generated by the extensive CO2 and other greenhouse
gases in the atmosphere. Earth’s internal generation of heat was an important
factor; the sun at this time was much fainter, and delivering perhaps a third
less energy, than at the present time.
What would have happened if Earth had stayed a water world? Probably
global temperatures would have remained high or even increased. For animal life
to form, the temperature had to drop from the levels acknowledged to have
been characteristic of Archaean time. A drop in global temperature while the sun
was getting hotter required a drastic reduction of atmospheric CO2—a reduction
of the greenhouse effect. Thus some means of removing CO2 had to be
brought to bear. As we saw in Chapter 9, the most effective way to do this is
through the formation of limestone, which uses CO2 as one of its building
blocks and thus scrubs it from the atmosphere. But significant volumes of limestone
form today only in shallow water; the most effective limestone formation
occurs in depths of less than 20 feet. In deeper water, high concentrations of dissolved
CO2 slows or inhibits the chemical reactions that lead to limestone formation.
There is evidence of deep-water, inorganic limestone formation in very
old rocks on Earth, as demonstrated by John Grotzinger and his team from
M.I.T. These studies showed that the early Earth’s ocean may have been saturated
in the compounds that can produce limestone and thus could have precipitated
limestone in deeper water at that time, removing carbon dioxide from the
Assessing the Odds
263
atmosphere as a consequence. However, Grotzinger points out that occurrences
of carbonate rocks during the Early Archaean—roughly the first billion years of
Earth’s existence—are rare. And this is only partly due to the rarity of rocks of
this age. It looks as though the central mode of removing carbon dioxide from
the atmosphere—the formation of carbonate rocks—seldom occurred.
To form limestone in significant volumes, then, shallow water is needed,
but on a planet without continents, shallow water is in short supply. If the volume
of water on a planet is low enough that significant areas of shallow water
are available even without continents, there is no problem. On Earth and
other planets with significantly deep oceans, however, without continents the
shallow-water regions would not be large enough for the necessary limestone
formation. Thus when planets have too much water on their surfaces—when
their oceans are too deep—there is no natural brake on carbon dioxide buildup.
Water temperatures will rise as planetary temperatures rise.
What about underwater weathering? James Kasting has pointed out to
us that an all-water world can indeed regulate its temperature. He rightly
notes that as oceanic water temperature rises, it eventually causes weathering
of limestone on the bottom of the sea. Although much less efficient than the
weathering of continental material, this mechanism will indeed produce a
feedback mechanism. Yet to heat water temperatures sufficiently to serve as a
global thermostat a planet may well exceed the critical 40°C mark that is the
upper temperature limit of animal life. If plate tectonics on Earth had not created
increasingly large land areas (and, as a by-product of that, massive areas
next to the continents with shallow-water regions where limestone could easily
form), Earth might well have reached global temperatures greater than animal
life could tolerate. And had global temperatures exceeded 100°C, the
oceans would have boiled away, the gigantic volumes of water becoming
steam in the atmosphere. This would have spelled a catastrophic ending of all
life on the surface of planet Earth.
The removal of carbon dioxide is called CO2 drawdown. On Earth, it
was accomplished because of continent formation, which took place during a
relativity brief interval of Earth history. From perhaps 2.7 to 2.5 billion years
ago there occurred a rapid buildup of continental areas. This buildup resulted

R A R E E A R T H
264
in the land surface increasing from perhaps 5% to about 30%. This marked
change had equally profound effects on the atmosphere–ocean system.
With the formation of continents as a result of plate tectonics, ocean
chemistry became dominated by weathering by-products of the continents.
As continents weather, or undergo the chemical and mechanical breakdown
of rock material, river runoff carries enormous volumes of these
chemicals into the sea, where they can greatly affect ocean chemistry and
cause mineralization—such as carbonate formation. Larger continents also,
paradoxically enough, meant larger shallow-water regions, for the emergence
of continents created the shallow continental shelves as well as large inland
seas and lakes. Thus the following sequence unfolded: Large shallow regions
were created; nutrient influx from continental regions increased; the amount
of plant material on Earth (mainly in the surface regions of the shallow seas
and on shallow sea bottoms) skyrocketed; and oxygen production began in
earnest. All of these events opened the pathway to the eventual evolution of
animals.
The critical question is why, on Earth, the volume of water was sufficiently
large to buffer global temperatures, but small enough so that shallow
seas could be formed by the uplifting of continents. If Earth’s ocean volume
had been greater, even the formation of continents would not have produced
shallow seas. To show that there can be great relative volumes of oceans on a
planet, we need only look at Jupiter’s moon Europa, where the planetcovering
ocean (now frozen) is 100 kilometers thick. No Mt. Everest rising
from the sea floor would ever poke through an ocean even half that deep.
There would be none of the shallows necessary for limestone formation and
no continental weathering.
What about the situation where the oceans are lower in volume than they
were on Earth? If the continents covered two-thirds of Earth’s surface (rather
than their present day one-third), would we have animal life? The great mass
extinction of the late Permian almost ended animal life because of high temperatures.
With greater continental area, we might expect temperature swings
to have been even greater, and the prospects for continued existence of at least
Assessing the Odds
265
land animals far lower, because large land areas create very high and very low
seasonal temperatures. Large land areas also reduce CO2 drawdown, because
carbonate formation takes place almost exclusively in oceans. On landdominated
worlds, opportunities for life to thrive would thus be reduced.
It appears that Earth got it just right. Without continents there seems a
strong likelihood that a planet will become too hot (especially because
main-sequence stars such as the sun increase their energy output through
time, and planets cannot move away from this increasing heat source). With
too much continental area, the opposite is likely to happen, as continental
weathering draws down carbon dioxide so much that glaciations ensue.
Earth may have been headed down the path toward a global mean temperature
so high as to boil away its oceans, or perhaps still cool enough to retain
its oceans but yet too warm for complex metazoans to evolve. Animals are
not thermophiles.
How much land area is “just right,” and how much is too little or too
much? The answer probably depends on the given planet’s distance from the
sun. A planet whose orbit dictates that it receives less energy from its star
than Earth does from the sun might need a greater amount of ocean cover (assuming
that an increased sea surface creates warmer planetary temperatures
because of greater greenhouse effects from CO2 buildup).
The relative areas of land and sea affect more than just planetary temperature.
If plate tectonics is not operating, there will be no continents, only
large numbers of seamounts and islands (whose number will be dictated by
the amount of volcanicity, itself a function of a planet’s heat flow). And without
continents, a planet’s ocean may never achieve a chemistry suitable for
animal life. Sherwood Chang of NASA cites an example of this. In 1994
Chang proposed that without substantial weathering (which can occur only
when there is substantial land area to weather), the early ocean of an Earthlike
planet would remain acidic—a poor environment for the development of
animal life. It seems that water worlds might be quite fecund habitats for
short periods of time but might not achieve the long-term temperature or
chemical stability conducive to animal life.
R A R E E A R T H
266
T H E I M P O R T A N C E—AND S H E E R CH A N C E
OC C U R R E N C E—O F OU R L A R G E MOON
Although many scientists have been doggedly pursuing the various attributes
necessary for a habitable planet—Michael Hart, George Weatherill, Chris
McKay, Norman Sleep, Kevin Zahnlee, David Schwartzman, Christopher
Chyba, Carl Sagan, and David Des Marais come to mind—one name stands
out in the scientific literature: James Kasting of Penn State University.
Kasting notes that whether habitable planets exist around other stars
“depends on whether other planets exist, where they form, how big they are,
and how they are spaced.” Kasting stresses, as we do, the importance of plate
tectonics in creating and maintaining habitable planets, and he suggests that
the presence of plate tectonics on any planet can be attributed to the planet’s
composition and position in its solar system. But one of Kasting’s most intriguing
comments is related to our Moon. Kasting notes that the obliquity
(the angle of the axis of spin of a planet) of three of the four “terrestrial” planets
of our solar system—Mercury, Venus, and Mars—has varied chaotically.
Earth is the exception, but only because it has a large moon. . . . If
calculations about the obliquity changes in the absence of the
moon are correct, Earth’s obliquity would vary chaotically from 0
to 85 degrees on a time scale of tens of millions of years were it not
for the presence of the Moon. . . . Earth’s climatic stability is dependent
to a large extent on the existence of the Moon. The
Moon is now generally believed to have formed as a consequence
of a glancing collision with a Mars-sized body during the later
stages of the Earth’s formation. If such moon-forming collisions
are rare . . . habitable planets might be equally rare.
We have accumulated a laundry list of potentially low-probability
events or conditions necessary for animal life: not only Earth’s position in the
“habitable zone” of its solar system (and of its galaxy), but many others as
well, including a large moon, plate tectonics, Jupiter in the wings, a magnetic
Assessing the Odds
267
field, and the many events that led up to the evolution of the first animal. Let
us explore what these conditions might mean for life beyond Earth.
T H E OD D S O F A N I M A L L I F E E L S E W H E R E ,
AND O F I N T E L L I G E N C E
In the 1950s, astronomer Frank Drake developed a thought-provoking equation
to predict how many civilizations might exist in our galaxy. The point of
the exercise was to estimate the likelihood of our detecting radio signals sent
from other technologically advanced civilizations. This was the beginning of
sporadic attempts by Earthlings to detect intelligent life on other planets.
Now called the Drake Equation in its creator’s honor, it has had enormous influence
in a (perhaps necessarily) qualitative field. The Drake Equation is simply
a string of factors that, when multiplied together, give an estimate of the
number of intelligent civilizations, N, in the Milky Way galaxy.
As originally postulated, the Drake Equation is.
N*  fs  fp  ne  fi  fc  fl  N
where:
N*  stars in the Milky Way galaxy
fs  fraction of sun-like stars
fp  fraction of stars with planets
ne  planets in a star’s habitable zone
fi  fraction of habitable planets where life does arise
fc  fraction of planets inhabited by intelligent beings
fl  percentage of a lifetime of a planet that is marked by the presence of
a communicative civilization
Our ability to assign probable values to these terms varies enormously.
When Drake first published his famous equation, there were great
uncertainties in most of the factors. There did (and does) exist a good estimate
for the number of stars in our galaxy (over 300 billion). The number of
R A R E E A R T H
268
star systems with planets, however, was very poorly known in Drake’s time.
Although many astronomers believed that planets were common, there was
no theory that proved star formation should include the creation of planets,
and many believed that the formation of planetary systems was exceedingly
rare. During the 1970s and later, however, it was assumed that planets were
common; in fact, Carl Sagan estimated that an average of ten planets would
be found around each star. Even though no extra solar planets were found
until the 1990s, their discovery seemed to vindicate those who believed
planets were common. But is it so? A new look at this problem suggests that
planets may indeed be quite rare—and thus the presence of animal life rarer
still.
A R E S T A R S W I T H P L A N E T S A N O M A L O U S ?
We now know that planetary formation outside our own system does indeed
occur. The recent and spectacular discovery of extrasolar planets, one of the
great triumphs of astronomical research in the 1990s, has proved what has
long been assumed: that other stars have planets. But at what frequency? It
may be that a substantial fraction of stars have planetary systems. To date,
however, astronomers have succeeded in detecting only giant, “Jupiter-like”
planets; available techniques cannot yet identify the smaller, rocky, terrestrial
worlds. Now that numerous stars have been examined, it appears that only
about 5% to 6% of examined stars have detectable planets. Because only large gas-giant
planets can be detected, this figure really shows that Jupiter clones close to
stars or in elliptical orbit are rare. But perhaps it indicates that planets as a
whole are rare as well.
The evidence that planets may be rare comes not so much from the
direct-observation approach of the planet finders (such as the Marcy/Butler
group) but from spectroscopic studies of stars that appear similar to our own
sun. The studies of those stars around which planets have been discovered
have yielded an intriguing finding: They, like our sun, are rich in metals. AcAssessing
the Odds

269
cording to astronomers conducting these studies, there seems to be a causal
link between high metal content in a star and the presence of planets. Our
own star is metal-rich. In a study of 174 stars, astronomer G. Gonzalez discovered
that the sun was among the highest in metal content. It appears that
we orbit a rare sun.
Other new studies also require us to question the belief that planetary
systems such as our own are common. At a large meeting of astronomers held
in Texas in early 1999, it was announced that 17 nearby stars had been observed
to be orbited by planets the size of Jupiter. Astronomers at the meeting
were also puzzled by an emerging pattern: None of the extrasolar planetary
systems resembles the sun’s family of planets. Geoff Marcy, the world’s
leading planet finder, noted that “for the first time, we have enough extrasolar
planets out there to do some comparative study. We are realizing that
most of the Jupiter-like objects far from their stars tool around in elliptical orbits,
not circular orbits, which are the rule in our solar system.” All of the
Jupiter-sized objects either were found in orbits much closer to their sun than
Jupiter is to our sun, or, if they occurred at a greater distance from their sun,
had highly elliptical orbits (observed in 9 of the 17 so far detected). In such
planetary systems, the possibility of Earth-like planets existing in stable orbits
is low. A Jupiter close to its sun will have destroyed the inner rocky planets.
A Jupiter with an elliptical or decaying orbit will have disrupted planetary orbits
sunward, causing smaller planets either to spiral into their sun or to be
ejected into the cold grave of interstellar space.
It is still impossible to observe smaller, rocky planets orbiting other
stars. Perhaps such planets—which we believe are necessary for animal
life—are quite common. But perhaps this is a moot point. We have hypothesized
that animal life cannot long exist on a planet unless there is a giant,
Jupiter-like planet within the same planetary system—and orbiting outside
the rocky planets—to protect against comet impacts. It may be that Jupiters
like our own, in regular orbits, are rare as well. To date, all tend to be in orbital
positions that would be lethal, rather than beneficial, to any smaller,
rocky planets.
R A R E E A R T H
270
P L A N E T F R E Q U E N C Y AND
T H E DR A K E E Q U A T I O N
All predictions concerning the frequency of life in the Universe inherently
assume that planets are common. But what if the conclusions suggested by
emerging studies—that Earth-like planets are rare, and planets with metal
rarer still—are true?
This finding has enormous significance for the final answer to the Drake
Equation. Any factor in the equation that is close to zero yields a near-zero final
answer, because all the factors are multiplied together. Carl Sagan, in 1974, estimated
that the average number of planets around each star is ten. Goldsmith
and Owen, in their 1992 The Search for Life in the Universe, also estimated ten planets
per star. But the new findings suggest greater caution. Perhaps planetary formation
is much less common than these authors have speculated.
To estimate the frequency of intelligent life, the Drake Equation hinges
on the abundance of Earth-like planets around sun-like stars. The most common
stars in the galaxy are M stars, fainter than the sun and nearly 100 times
more numerous than solar-mass stars. These stars can generally be ruled out because
their “habitable zones,” where surface temperatures could be conducive
to life, are uninhabitable for other reasons. To be appropriately warmed by
these fainter stars, planets must be so close to the star that tidal effects from the
star force them into synchronous rotation. One side of the planet always faces
the star, and on the permanently dark side, the ground reaches such low temperatures
that the atmosphere freezes out. Stars much more massive than the
sun have stable lifetimes of only a few billion years, which might be too short
for the development of advanced life and evolution of an ideal atmosphere. As
we noted earlier, each planetary system around a 1-solar-mass star will have
space for at least one terrestrial planet in its habitable zone. But will there actually
be an Earth-sized planet orbiting its star in that space? When we take into
account factors such as the abundance of planets and the location and lifetime
of the habitable zone, the Drake Equation suggests that only between 1% and
0.001% percent of all stars might have planets with habitats similar to those on
Assessing the Odds
271
Earth. But many now believe that even these small numbers are overestimated.
On a universal viewpoint, the existence of a galactic habitable zone vastly reduces
them.
Such percentages seem very small, but considering the vastness of the
Universe, applying them to the immense numbers of stars within it can still
result in very large estimates. Carl Sagan and others have mulled these various
figures over and over. They ultimately arrived at an estimate of one million civilizations
of creatures capable of interstellar communication existing in the Milky Way galaxy at
this time. How realistic is this estimate?
If microbial life forms readily, then millions to hundreds of millions of
planets in the galaxy have the potential for developing advanced life. (We expect
that a much higher number will have microbial life.) However, if the advancement
to animal-like life requires continental drift, the presence of a
large moon, and many of the other rare Earth factors discussed in this book,
then it is likely that advanced life is very rare and that Carl Sagan’s estimate
of a million communicating civilizations is greatly exaggerated. If only one in
1000 Earth-like planets in a habitable zone really evolves as Earth did, then
perhaps only a few thousand have advanced life. Although it could be argued
that this is too pessimistic, it may also be much too optimistic. Even so, we
cannot rule out the possibility that Earth is not unique in the galaxy as an
abode of life that has just recently developed primitive technologies for space
travel and interplanetary radio communication.
Perhaps we can suggest a new equation, which we can call the “Rare
Earth Equation,” tabulated for our galaxy:
N*  fp  ne  fi  fc  fl  N
where:
N*  stars in the Milky Way galaxy
fp  fraction of stars with planets
ne  planets in a star’s habitable zone
fi  fraction of habitable planets where life does arise
R A R E E A R T H
272
fc  fraction of planets with life where complex metazoans arise
fl  percentage of a lifetime of a planet that is marked by the presence of
complex metazoans
And what if some of the more exotic aspects of Earth’s history are required,
such as plate tectonics, a large moon, and a critically low number of
mass extinctions? When any term of the equation approaches zero, so too
does the final result. We will return to this at the end of this chapter.
If animal life is so rare, then intelligent animal life must be rarer still.
How can we define intelligence? Our favorite definition comes from Christopher
McKay of NASA, an astronomer, who defines intelligence as the “ability
to construct a radio telescope.” Although a chemist might define intelligence
as the ability to build a test tube, or an English professor as the ability
to write a sonnet, let us for the moment accept McKay’s definition and follow
the lines of reasoning he sets out in his wonderful essay “Time for Intelligence
on Other Planets,” published in 1996. Much of the following discussion
comes from that source.
McKay points out that if we accept the “Principle of Mediocrity” (also
known as the Copernican Principle) that Earth is quite typical and common,
it follows that “intelligence has a very high probability of emerging but only
after 3.5 billion years of evolution.” This supposition is based on a reading of
Earth’s geological record, which suggests to most authors that evolution has
undergone a “steady progressive development of ever more complex and sophisticated
forms leading ultimately to human intelligence.” Yet McKay
notes—as we have tried to emphasize in this book—that evolution on Earth
has not proceeded in this fashion but rather has been affected by chance
events, such as the mass extinctions and continental configurations produced
by continental drift. Furthermore, we believe that not only events on Earth,
but also the chance fashion in which the solar system was produced, with its
characteristic number of planets and planetary positions, may have had a
great influence on the history of life here.
McKay breaks down the critical events in the evolution of intelligence
on Earth as shown in the accompanying table.
Assessing the Odds
273
When It Happened How Long It Possible
on Earth (millions Took to Complete Minimum Time
Event of years ago) (millions of years) (millions of years)
Origin of life 3800–3500  500 10
Oxygenic photosynthesis
 3500  500 Negligible
Oxygen environments 2500 1000 100
Tissue multicellularity 550 2000 Negligible
Development of animals 510 5 5
Land ecosystems 400 100 5
Animal intelligence 250 150 5
Human intelligence 3 3 3
We can certainly quibble with some (or all) of his numbers, especially
his estimate of when life first arose on Earth, for we think it occurred far earlier
than 3800 to 3500 million years ago. Yet these estimates are probably not
off by orders of magnitude. McKay’s point is that complex life—and even
intelligence—could conceivably arise faster than it did on Earth. If we accept
McKay’s figures, a planet could go from an abiotic state to the home of a civilization
building radio telescopes in 100 million years, as compared to the
nearly 4 billion years it took on Earth. But McKay also concedes that there
may be other factors that require a long period:
What is not known is whether there is some aspect of the biogeochemical
processes on a habitable planet—for example, those
dealing with the burial of organic material, the maintenance of
habitable temperatures as the stellar luminosity increases gradually
over its main sequence lifetime, or global recycling by
tectonics—that mandates the long and protracted development of
the oxygen-rich biosphere that occurred on Earth. Other important
unknowns include the effect of solar system structure on the
origin of life and its subsequent evolution to advanced forms.
R A R E E A R T H
274
His inference is that plate tectonics has slowed the rise of oxygen on Earth.
But it also may be necessary to ensure a stable oxygenated habitat, just as having
the correct types of planets in a solar system is important as well.
In their 1996 essay “Biotically Mediated Surface Cooling and Habitability,”
Schwartzman and Shore tackle this same problem and reach a different
conclusion: They believe that the most critical element in determining
the rate at which intelligence can be acquired is a potentially habitable
planet’s rate of cooling. Their point is that complex life such as animals is extremely
temperature-limited, with a very well-defined upper temperature
threshold. Although some forms of animal life can exist in temperatures as
high as 50°C or sometimes even 60°C, most require lower temperatures, as
do the complex plants necessary to underpin animal ecosystems. A maximum
temperature of 45°C is probably realistic. It is thus the time necessary for a
planet to cool to below this value that is critical, according to these two authors.
Many factors affect the time required, including the rate at which a star
increases in luminosity through time (which works against cooling), the volcanic
outgassing rate (which also works against cooling, because such outgassing
puts more greenhouse gases into a planetary atmosphere), the rate at
which continental land surface grows (as continents grow, planets usually
cool), the weathering rate of land areas, the number of comet or asteroid impacts
and their frequency, the size of a star, whether or not plate tectonics exists,
the size of the initial planetary oceans, and the history of evolution on
the planet.
With this in mind, let us return to our Rare Earth Equation and flesh it
out a bit by adding some of the other factors featured in this book.
N*  fp  fpm  ne  ng  fi  fc  fl  fm  fj  fme  N
where:
N*  stars in the Milky Way galaxy
fp  fraction of stars with planets
fpm  fraction of metal-rich planets
ne  planets in a star’s habitable zone
Assessing the Odds
275
ng  stars in a galactic habitable zone
fi  fraction of habitable planets where life does arise
fc  fraction of planets with life where complex metazoans arise
fl  percentage of a lifetime of a planet that is marked by the
presence of complex metazoans
fm  fraction of planets with a large moon
fj  fraction of solar systems with Jupiter-sized planets
fme  fraction of planets with a critically low number of mass
extinction events
With our added elements, the number of planets with animal life gets
even smaller. We have left out other aspects that may also be implicated:
Snowball Earth and the inertial interchange event. Yet perhaps these too are
necessary.
Again, as any term in such an equation approaches zero, so too does the final product.
How much stock can we put in such a calculation? Clearly, many of
these terms are known in only the sketchiest detail. Years from now, after the
astrobiology revolution has matured, our understanding of the various factors
that have allowed animal life to develop on this planet will be much greater
than it is now. Many new factors will be known, and the list of variables involved
will undoubtedly be amended. But it is our contention that any strong
signal can be perceived even when only sparse data are available. To us, the
signal is so strong that even at this time, it appears that Earth indeed may be
extraordinarily rare.
Messengers
from the
Stars
Our planet is not in a special place in the solar system, our
Sun is not in a special place in our galaxy and our galaxy is
not in a special place in the Universe.
—Marcello Gleiser, The Dancing Universe
Some things have to be experienced firsthand; for some wonders, no written
description or photograph can substitute—the birth of one’s child,
for example, or the first music heard from an actual orchestra, love, sex,
standing before a Monet canvas. One such revelation not so often experienced,
is one’s first glimpse of the starry night through a telescope.
VI E W I N G T H E UN I V E R S E
We have all seen photographs of endless star fields, galaxies, and nebulae. But
no matter how great their beauty, the stars in photographs are lifeless, and no
view of the night sky with the unaided eye, even in the clearest atmosphere,
277
13

R A R E E A R T H
278
is like the first view through a small telescope. If looking at the Milky Way
with the unaided eye is akin to snorkeling on a coral reef, then adding a telescope
is like strapping on scuba tanks: We are no longer tied to the surface
but can roam the depths of the star fields and see unimagined splendors amid
stars whose numbers are beyond belief. Even with a low-power telescope a
new vision emerges; the uncountable pinpricks of light now revealed are
seemingly alive, in no way diminished by their passage through corrected
lenses. In fact, the stars gain strength, color, and clarity. But the greatest and
most lasting impression is the increase in their sheer numbers. The superb
double-star cluster in Perseus changes from a dull, unresolvable glow to
bountiful diamonds sprinkled on black velvet; the globular cluster in Hercules
is transformed from a tiny smudge to scattered grains of light. With
time and experience, even greater vistas open up. We discover the joys of
other deep-space objects, galaxies and nebulae. And eventually, in the
Northern hemisphere, we inevitably find ourselves slowly moving through
the crowded star fields of Sagittarius on a dark summer night, the light from
this luminous expanse of stars sweeping the senses like a wind, nebulae and
galaxies an endless visual melody punctuated by staccatos of brighter suns.
Those in the Southern hemisphere witness even more dramatic vistas: the
two great Magellenic Clouds looming so close overhead. It becomes spectacle,
overwhelming, and ultimately—diminishing. The myriad stars overcome
us, so utterly do they trivialize (marginalize? minimize?) our small planet and
we who stare out.
The Universe seems to be finite; there are not an infinite number of
planets circling the vast number of stars in the ocean of space. But the numbers
are immense beyond understanding. We are one of many planets. But as
we have tried to show in this book, perhaps not so many as we might hope—
and perhaps not so many that we will ever, however long the history of our
species, find any extraterrestrial animals among the stars surrounding our sun.
That is a fate not foreseen by Hollywood—that we may find nothing but bacteria,
even on planets orbiting distant stars.
If the Rare Earth Hypothesis is correct—that is, if microbial life is common
but animal life is rare—there will be societal implications, or at least
Messengers from the Stars
279
some small personal implications. What will be the effect if news comes back
from the next Mars mission that there is life on Mars after all—microbial to
be sure, but life. Or what if, after astronauts voyage repeatedly to other planets
in our solar system, or even to the dozen nearest stars, we find nothing
more advanced than a bacterium? What if, at least in this quadrant of the
galaxy, we are quite alone, not just as the only intelligent organisms but also
as the only animals? How much of our striving to travel into space is the hope
of discovering—and perhaps talking to—other animalia?
VI E W S O F E A R T H THROUGH
HUMAN HI S T O R Y
Since the time of the Greeks, science has tried to make sense of the Universe
and of our place in it. More than two millennia ago, a Greek named Thales of
Miletus, credited by many as the founder of Western philosophy, was among
the first to leave a record of his musings about the place of Earth in the cosmos.
Thales thought that the cosmos was an organic, living thing, and in that
he may not have been far wrong if bacteria or bacteria-like organisms are as
common as we believe them to be in the Universe. Thales’s student Anaximander
was among the first to place Earth at the center of the cosmos, postulating
that Earth was a floating cylinder with a series of large wheels with
holes in them rotating around it. The Pythagoreans tried to break from this
central-Earth motif, proposing that Earth moved in space and was not the
center of the Universe. But Earth’s centrality was restored by members of
Plato’s school and became exalted by the students of Aristotle. Eudoxus
placed Earth at the center of 27 concentric spheres, each of which rotated
around it. Soon two schools of thought competed: the “sun-centered” model
of Aristarchus and the Earth-centered model of Ptolemy. The latter held sway
through the Middle Ages.
During the Middle Ages, Earth was not only regarded as the center of
the Universe but was again believed to be flat. St. Thomas Aquinas made
Earth a sphere again but codified its place as the center of the Universe. It was
R A R E E A R T H
280
Nicholas Copernicus who finally shattered the notion of an Earth-centered
Universe and put the sun at the center of all orbits. But even with this great
leap forward, the sun remained at the center of the Universe as well, according
to Copernicus in his revolutionary book of 1514, Commentariolus.
Copernicus forever destroyed the myth that our Earth lay at the center
of the Universe, with the sun and all other planets and stars revolving around
us; his work eventually led to the concept of a “Plurality of Worlds”—the idea
that our planet is but one among many. This has now been described as the
“Principle of Mediocrity,” also known as the Copernican Principle. Yet an
even greater blow came with the invention of the telescope. There is still debate
about who built the first optical telescope, although Dutch optician Johannes
Lippershey obtained the first official license for construction of a telescope
in 1608. The device was an immediate sensation, and by 1609 this
revolutionary new instrument found its way into the hands of Galileo, who
built his own soon after hearing of the concept. Before Galileo, telescopes
had been used to assess the terrestrial world (and for various military applications),
but Galileo pointed his into the heavens and forever changed our understanding
of the cosmos.
Galileo quickly surmised that there are far more stars in the sky than
anyone had guessed. He discovered that the Milky Way is made up of uncountable
individual stars. He observed the Moon, discovered satellites revolving
around Jupiter (and in so doing showed that our Earth could just as
conceivably orbit the sun). Earth’s central place in the Universe, the fervent
belief of Aristotle, was now observationally shown to be wrong. Copernicus
had dealt with theory; Galileo and his telescopes dealt with reality. Galileo’s
message, published in his booklet Siderius nuncius, or “Messenger from the
Stars,” was about the truth told by the stars: that Earth is but one of many cosmic
objects. To illustrate his point, he noted the presence of faint patches of
light just visible to the unaided eye—objects called nebulae. Even with his
primitive and tiny telescopes, Galileo could see these curious objects far better
than anyone before. He thought them to be great masses of stars, made indistinct
by their very distance.
Messengers from the Stars
281
The decentralization of Earth continued in relentless fashion. In 1755
Immanuel Kant theorized that a rotating gas cloud would flatten into a disk
as it contracted under its own gravity. Kant was familiar with the numerous
nebulae of the night skies, the faint glowing patches of luminosity scattered
through the heavens. All the early astronomers knew of the faint cloud in the
constellation of Andromeda. He knew these objects to be one of many distant
groups of stars he called “island Universes.” But Kant didn’t stop there:
He theorized that the sun, Earth, and other planets might have formed in this
swirling mass of gas. This concept was taken a step further by Pierre-Simon
de Liplike, who speculated in detail about how planetary systems might form
from nebular origins. He invoked a dynamical mechanism for the formation
of stars and their planets. Earth and the solar system became one of many
such systems all formed in the same way.
But how far away were these island Universes? Was there only a single
galaxy in the Universe, of which our star was part, or were there many? This
debate was not resolved until the early twentieth century, a time when gigantic
new telescopes were being constructed and outer space was being
probed as never before. The conflict came to a head on April 26, 1920, when
Harlow Shapley from the Mount Wilson observatory in California and
Heber Curtis from the Allegheny Observatory in Pittsburgh met before the
members of the National Academy of Sciences, a clash that came to be
known as the Great Debate. The debate ended inconclusively, because it was
not yet possible to assess the distance of the nebulae. That soon changed,
however, thanks to the efforts of astronomer Edwin Hubble. Using a newly
constructed, 100-inch reflecting telescope, Hubble was able to make observations
that proved conclusively that the island nebulae were not associated
with our Milky Way but were far-distant objects. Even the closest, the Andromeda
galaxy, was found to be at least 2 million light-years from Earth and
similar in shape to our Milky Way galaxy. The debate was over. The Milky
Way is one of a vast number of scattered and widely separated galaxies floating
in space. We became even more trivialized—now our galaxy was but one
of many.
R A R E E A R T H
282
Two millennia of astronomers and philosophers removed Earth from
the center of the Universe and placed us orbiting a sun that is but one of hundreds
of billions in a galaxy itself but one among billions in the Universe. And
it was not only astronomers who changed the world view. Einstein showed
that there is no preferred observer in the Universe, and quantum mechanics
told us that chance is king. Charles Darwin and his powerful theory of evolution
demoted humans from the crown of creation to a rather new species on
an already animal-rich planet, the chance offspring of larger-scale evolutionary
and ecological forces. Nothing special. And yet . . .
The great danger to our thesis (that Earth is rare because of its animal
life, the factors and history necessary to arrive at this point as a teeming,
animal- and plant-rich planet being highly improbable) is that it is a product
of our lack of imagination. We assume in this book that animal life will be
somehow Earth-like. We take the perhaps jingoistic stance that Earth-life is
every-life, that lessons from Earth are not only guides but also rules. We assume
that DNA is the only way, rather than only one way. Perhaps complex
life—which we in this book have defined as animals (and higher plants as
well)—is as widely distributed as bacterial life and as variable in its makeup.
Perhaps Earth is not rare after all but is simply one variant in a nearly infinite
assemblage of planets with life. Yet we do not believe this, for there is so
much evidence and inference—as we have tried to show in the preceding
pages—that such is not the case.
OU R R A R E E A R T H
Let us recap why we think Earth is rare. Our planet coalesced out of the debris
from previous cosmic events at a position within a galaxy highly appropriate
for the eventual evolution of animal life, around a star also highly
appropriate—a star rich in metal, a star found in a safe region of a spiral galaxy,
a star moving very slowly on its galactic pinwheel. Not in the center of the
galaxy, not in a metal-poor galaxy, not in a globular cluster, not near an active
Messengers from the Stars
283
gamma ray source, not in a multiple-star system, not even in a binary, or near
a pulsar, or near stars too small, too large, or soon to go supernova. We became
a planet where global temperatures have allowed liquid water to exist for
more than 4 billion years—and for that, our planet had to have a nearly circular
orbit at a distance from a star itself emitting a nearly constant energy output
for a long period of time. Our planet received a volume of water sufficient
to cover most—but not all—of the planetary surface. Asteroids and comets hit
us but not excessively so, thanks to the presence of giant gas planets such as
Jupiter beyond us. In the time since animals evolved over 600 million years
ago, we have not been punched out, although the means of our destruction by
catastrophic impact is certainly there. Earth received the right range of building
materials—and had the correct amount of internal heat—to allow plate
tectonics to work on the planet, shaping the continents required and keeping
global temperatures within a narrow range for several billion years. Even as the
Sun grew brighter and atmosphere composition changed, the Earth’s remarkable
thermostatic regulating process successfully kept the surface temperature
within livable range. Alone among terrestrial planets we have a large moon,
and this single fact, which sets us apart from Mercury, Venus, and Mars, may
have been crucial to the rise and continued existence of animal life on Earth.
The continued marginalization of Earth and its place in the Universe perhaps
should be reassessed. We are not the center of the Universe, and we never will
be. But we are not so ordinary as Western science has made us out to be for
two millennia. Our global inferiority complex may be unwarranted. What if
Earth is extremely rare because of its animals (or, to put it another way, because
of its animal habitability)?
The possibility that animal life may be very rare in the Universe also
heightens the tragedy of the current rate of extinction on our planet. Earlier,
we suggested that the rise of an intelligent species on any planet might be a
common source of mass extinction. That certainly seems to be the case on
Earth. And if animals are as rare in the Universe as we suspect, it puts species
extinction in a whole new light. Are we eliminating species not only from our
planet but also from a quadrant of the galaxy as a whole?
R A R E E A R T H
284
To understand the rates of extinction on Earth today, one has only to
examine the plight of tropical rainforests. Forests have been a part of this
planet for more than 300 million years, and although the nature of species has
changed over that long period, the nature of the forests has changed little.
The forests are the great Noah’s ark of species on this planet. Although the
land surface of our globe is only one-third that of the oceans, it appears that
80% to 90% of the total animal and plant biodiversity of the planet inhabits
the land, and most of that diversity is found in tropical forests. As we destroy
these forests, we destroy species. It has been estimated that between 5 and 30
million species of animals live in the tropical rainforests and that only about
5% of these are known to science. The fossil record tells us quite clearly that
the world has attained the highest level of biological diversity ever in its history.
There are also disturbing and unmistakable signs that this plateau in the
number of species on Earth has been crested and the biodiversity of Earth is
diminishing.
There appear to be several forces driving a reduction of biodiversity—a
destruction of biodiversity, to be less delicate. The most important seems to be
the rapid increase in human population. Ten thousand years ago there may
have been at most 2 to 3 million humans scattered around the globe. There
were no cities, no great population centers. There were fewer people on the
globe than are now found in virtually any large American city. Two thousand
years ago the number had swelled to perhaps 130 to 200 million people. Our
first billion was reached in the year 1800. If we take the time of origin of our
species as about 100,000 years ago, it seems that it took our species 100,000
years to reach the billion-person population plateau. Then things sped up
considerably. We reached 2 billion people in 1930, about 1000 times faster
than it took to reach the first billion. But the rate of increase kept accelerating.
By 1950, only 20 years later, we had reached 2.5 billion souls. In 1999,
we hit 6 billion. There will be approximately 7 billion people by 2020 and
perhaps 11 billion by 2050 to 2100.
Rainforest conversion, which changes forest to fields, and then (usually)
to overgrazed, eroded, and infertile land within a generation, is perhaps the
most direct executioner of biodiversity. It appears that 25% of the world’s topMessengers
from the Stars
285
soil has been lost since 1945. One-third of the world’s forest area has disappeared
in the same interval. The result is species extinction. A thousand years
from now, when humanity reflects on the world that was, and looks out at the
desert surrounding the rare and notably less diverse animals that remain,
whom will it hold responsible?
President Theodore Roosevelt closed off the Yellowstone region to development
in forming the first national park in the United States. Wouldn’t it
be ironic if some alien equivalent had done the same thing for our planet? Astrobiologists
have suggested this—it is known as the Zoo Hypothesis. The
joke would be on us: We are somebody’s national park, our rare planet Earth
stocked with animals for safekeeping. Perhaps that is why we have yet to hear
any signals from space. A big fence surrounds our solar system: “Earth Intergalactic
Park. Posted: No trespassing or tampering. The only planet with animals
for the next 5000 light-years.”
***************************************************************************
Twenty thousand years ago, Earth was locked in the glacial grip of the
last ice age. Wooly mammoths and great mastodons, ground sloths, camels,
and saber-toothed tigers roamed North America; people didn’t. Humans
were still thousands of years from crossing the land bridge from what would
someday be called Siberia to the place we now know as Alaska. Humans were
still 10,000 years from mastering agriculture. On some given yet forever
anonymous day in that long ago time of 20,000 B.C., a distant neutron star in
the constellation Aquila, part of the Summer Triangle so familiar to star
watchers in the Northern hemisphere, underwent a violent cataclysm of some
type and belched hard radiation into space, hurling an expanding sphere of
poison at the speed of light in all directions. For 20,000 years it sped through
space. It hit Earth over the Pacific Ocean on the evening of August 27, 1998,
as it continued ever onward, its energy diminishing with each mile it traveled
from its original source.
For 5 minutes on that late summer day, Earth was bombarded by gamma
rays and X-rays, the lethal twins generated by thermonuclear bombs as well
R A R E E A R T H
286
as by the interiors of stars. Even after traveling 20,000 light-years, the energy
was sufficient to send radiation sensors on seven Earth satellites to maximum
reading or off scale. Two of these satellites were shut down to save their instruments
from burnout. The radiation penetrated to within 30 miles of
Earth’s surface and then was dissipated by the lower regions of our planet’s atmosphere.
This event was the first time that such high energy from outside
the solar system was detected to have had a measurable effect on the atmosphere.
But in all probability, it was not the first time Earth has been buffeted
by energy from interstellar space. Perhaps a closer neutron star, or some
other stellar demon not yet known to us, caused one or more of the mass extinctions
in Earth’s past. Perhaps we have only begun to see the demons surrounding
us as we take our first tentative peeks through our planetary bedroom
window into space.
Astronomers believe that the 1998 event was caused by the surface disruption
of a kind of star that had only been theorized to exist: a magnetar. A
type of neutron star, a magnetar is perhaps 20 miles in diameter but is more
massive than our sun. It is estimated that a thimbleful of its material would
weigh 100 million tons. It is matter compressed far beyond the point of
human comprehension. The star has a surface of iron, but iron of a type never
found in our solar system. The star spins, as all neutron stars spin, and the result
is the formation of an intensely powerful magnetic field. For reasons we
can only guess at, the surface of this star—20,000 years ago—underwent a
massive disruption, sending energy into space as a consequence.
Energy dissipates with distance. Had the magnetar in question been
only 10,000 light-years away, the energy reaching Earth would have been
four times stronger—perhaps strong enough to damage the ozone layer.
Did this particular event sterilize worlds within a light-year or less? Were
there civilizations existing on worlds that were seared out of existence by
gamma rays and a magnetic field pulse sufficient to tear the very molecules of
living matter apart? Was another Earth sterilized? Perhaps life can flourish
only in neighborhoods far from magnetars. Have magnetars—as well as so
much else we have seen in the pages of this book—made animal life rare in
the Universe? And what else is out there, lurking in the dark?
Messengers from the Stars
287
The discovery of a phenomenon such as the magnetar is an object lesson
that suggests a great deal more than life’s rarity: There is still so much
more to learn about the heavens surrounding us. We humans are like 2-yearolds,
just beginning to comprehend the immensity, wonder, and hazards of
the wide world. So too with our understanding of astrobiology. It is clearly
just beginning.

The Mathematical Probability Of Life On Other Earth-Like Planets

http://www.science20.com/news_releases/the_mathematical_probability_of_life_on_other_earth_like_planets