Astrobiology: Life in the Universe

c fl1999 by Tony Wong

May 8, 1999

In memory of Carl Sagan, who taught me to love astronomy

ii Contents 0 Introduction 1 1 What is Life? 3

1.1 Fundamental Characteristics of Life . . . . . . . . . . . . . . . . . . . . . . . 3

1.1.1 Non-Equilibrium Thermodynamics . . . . . . . . . . . . . . . . . . . 3 1.1.2 Reproduction Based on Information . . . . . . . . . . . . . . . . . . . 4 1.1.3 Capacity for Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.2 The Chemical Basis of Life on Earth . . . . . . . . . . . . . . . . . . . . . . 5

1.2.1 Proteins . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 1.2.2 Nucleic Acids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.2.3 Energy Utilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.3 Information Content of Life . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 1.4 Mutation and Genetic Diversity . . . . . . . . . . . . . . . . . . . . . . . . . 13 1.5 Viruses and Cell Fragments . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 1.6 Vitalism vs. Mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

2 Constraints on Extraterrestrial Life 17

2.1 Habitat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2.2 The Importance of Carbon . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 2.3 The Importance of Water . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.4 Continuously Habitable Zones . . . . . . . . . . . . . . . . . . . . . . . . . . 21

3 Life Elsewhere in Our Solar System? 23

3.1 The Greenhouse Effect . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23 3.2 Mars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

3.2.1 From Lowell to the Dawn of the Space Age . . . . . . . . . . . . . . . 25 3.2.2 Mariner and Viking . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

iii

iv CONTENTS

3.2.3 Evidence for Liquid Water . . . . . . . . . . . . . . . . . . . . . . . . 26 3.2.4 The Martian Meteorite Controversy . . . . . . . . . . . . . . . . . . . 28 3.3 Venus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 3.4 Europa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 3.5 Titan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 3.6 Other Bodies of the Solar System . . . . . . . . . . . . . . . . . . . . . . . . 33 3.7 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

4 Origin & Evolution of Life on Earth 35

4.1 Historical Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 4.2 The Miller-Urey Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . 37 4.3 Setting the Stage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

4.3.1 Birth of the Solar System . . . . . . . . . . . . . . . . . . . . . . . . 38 4.3.2 The Hadean Earth . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 4.3.3 The Early Atmosphere . . . . . . . . . . . . . . . . . . . . . . . . . . 40 4.3.4 Origin of Organic Compounds . . . . . . . . . . . . . . . . . . . . . . 41 4.4 The Origin of Life: Current Theories . . . . . . . . . . . . . . . . . . . . . . 42

4.4.1 When Did Life Begin? . . . . . . . . . . . . . . . . . . . . . . . . . . 43 4.4.2 Where Did Life Begin? . . . . . . . . . . . . . . . . . . . . . . . . . . 43 4.4.3 How Did Life Begin? The Chicken and Egg Problem . . . . . . . . . 45 4.4.4 The Polymerization Problem . . . . . . . . . . . . . . . . . . . . . . . 46 4.4.5 The Selectivity Problem . . . . . . . . . . . . . . . . . . . . . . . . . 47 4.4.6 Cell Membranes: Encapsulating Life . . . . . . . . . . . . . . . . . . 48 4.4.7 The First Cell . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 4.5 Biological Evolution: Basic Principles . . . . . . . . . . . . . . . . . . . . . . 49 4.6 A Brief History of Life on Earth . . . . . . . . . . . . . . . . . . . . . . . . . 52

4.6.1 The Tree of Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 4.6.2 The Oxygen Revolution . . . . . . . . . . . . . . . . . . . . . . . . . 53 4.6.3 The Cambrian Explosion . . . . . . . . . . . . . . . . . . . . . . . . . 55 4.6.4 Mass Extinctions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 4.7 Life's Effects on the Earth . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56

CONTENTS v 5 Comets & Asteroids: Creators or Destroyers? 59

5.1 Extraterrestrial Sources for Biogenic Elements . . . . . . . . . . . . . . . . . 59 5.2 Organic Molecules from Space . . . . . . . . . . . . . . . . . . . . . . . . . . 60 5.3 Panspermia: Are We Extraterrestrials? . . . . . . . . . . . . . . . . . . . . . 61 5.4 Impacts and Mass Extinctions . . . . . . . . . . . . . . . . . . . . . . . . . . 62

5.4.1 Impact Frustration of Life? . . . . . . . . . . . . . . . . . . . . . . . . 62 5.4.2 The K/T Event . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 5.4.3 Other Mass Extinctions . . . . . . . . . . . . . . . . . . . . . . . . . 63 5.4.4 The Continuing Threat . . . . . . . . . . . . . . . . . . . . . . . . . . 64

6 The Living Galaxy? 65

6.1 The Drake Equation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 6.2 Searching for Extrasolar Giant Planets . . . . . . . . . . . . . . . . . . . . . 67

6.2.1 Astrometric Detection . . . . . . . . . . . . . . . . . . . . . . . . . . 67 6.2.2 Spectroscopic Detection . . . . . . . . . . . . . . . . . . . . . . . . . 68 6.2.3 Other Techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 6.2.4 Properties of the Known Extrasolar Planets . . . . . . . . . . . . . . 70 6.3 Searching for Earth-like Planets and Life . . . . . . . . . . . . . . . . . . . . 71

6.3.1 Pulsars With Planets . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 6.3.2 Direct Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 6.3.3 Finding Evidence for Life . . . . . . . . . . . . . . . . . . . . . . . . 73 6.4 Interstellar Travel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

6.4.1 Propulsion Strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 6.4.2 The Destiny of Civilizations . . . . . . . . . . . . . . . . . . . . . . . 75

7 Intelligent Civilizations 77

7.1 Human Evolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78

7.1.1 The Likelihood of Intelligence . . . . . . . . . . . . . . . . . . . . . . 78 7.1.2 Pre-modern Humans . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 7.1.3 Emergence of Modern Humans . . . . . . . . . . . . . . . . . . . . . 80 7.2 Language and Evolution of the Brain . . . . . . . . . . . . . . . . . . . . . . 81

7.2.1 The Nature of Language . . . . . . . . . . . . . . . . . . . . . . . . . 82 7.2.2 How Did Languange Evolve? . . . . . . . . . . . . . . . . . . . . . . . 83 7.2.3 Consciousness Explained? . . . . . . . . . . . . . . . . . . . . . . . . 85

vi CONTENTS

7.2.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 7.3 SETI: The Search for Extraterrestrial Intelligence . . . . . . . . . . . . . . . 86

7.3.1 The Importance of Longevity . . . . . . . . . . . . . . . . . . . . . . 87 7.3.2 Strategies for Communication . . . . . . . . . . . . . . . . . . . . . . 88 7.3.3 Historical Development . . . . . . . . . . . . . . . . . . . . . . . . . . 90 7.3.4 The Fermi Paradox: Where Are They? . . . . . . . . . . . . . . . . . 92 7.3.5 Rise and Fall of the NASA SETI Program . . . . . . . . . . . . . . . 93 7.3.6 Consequences of Contact . . . . . . . . . . . . . . . . . . . . . . . . . 94 7.4 UFO's and Alien Visits to Earth . . . . . . . . . . . . . . . . . . . . . . . . . 95

8 Life: Nature's Destiny? 97

8.1 The Anthropic Principle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 8.2 Intelligent Design of the Universe . . . . . . . . . . . . . . . . . . . . . . . . 98 8.3 Life and the Future of the Universe . . . . . . . . . . . . . . . . . . . . . . . 100

Preface This book is designed for a one-semester course in astrobiology at the undergraduate level. While putting together a course on the origin and distribution of life in the Universe, it quickly became clear that no textbook contained the breadth and depth of material appropriate for an entire course. This may no be longer the case: recently a new book by Bruce Jakosky, The Search for Life on Other Planets, has been published, with particularly good sections on Solar System exobiology. Also recommended for many topics is Jonathan Lunine's Earth: Evolution of a Habitable World, a new book which takes a refreshingly broad perspective on the evolution of our planet. Still, I decided that I might as well finish what I started, so this is the result.

I assume the reader has a strong background in general science, and can casually browse through an issue of Scientific American. Some abbreviations may be unfamiliar (such as kyr=10

3

years, Myr=10

6

years, Gyr=10

9

years) but in this era of science and technology

you have probably encountered most of them before. No advanced mathematics or even college-level science courses are required, but some familiarity with basic astronomy (at the Astronomy 10 level) and modern biology would be helpful. In short, this is a course for science majors and science enthusiasts.

vii

viii CONTENTS Chapter 0 Introduction This book is an investigation of the nature of life and intelligence, its origin on Earth, and its possible distribution in the Universe. The subject matter that is covered here has been given a variety of names--exobiology, astrobiology, and bioastronomy being the most popular-- but I will stick with "astrobiology" since that seems to be the choice that NASA has made, and unlike "exobiology" it reflects the fact that we are interested in life on Earth as well as extraterrestrial life. Roughly speaking, astrobiology today consists of four major areas: (1) Solar System exobiology, (2) origin and evolution of life, (3) discovery and study of other planetary systems, and (4) search for extraterrestrial intelligence. Note that I have excluded studies of microgravity or human survival in space; this very large body of knowledge I refer to as "space biology," and will not be discussed in this course.

Astrobiology has often been called a science in search of a subject. What is the point of studying things that may not even exist? Yet even if we never discover extraterrestrial life, there is a great deal we can learn about the requirements for life and the process by which it arose on our planet. The knowledge that we are probably alone, at least in our Galaxy, might make us value our planet just a little more. The knowledge that we are not alone, on the other hand, might alter our science and our civilization profoundly. The quest for life in the Universe is a modern-day extension of the restless urge humans have grappled with for millenia: the urge to understand the mystery of existence.

Furthermore, a number of developments over the past few years have lent credence to astrobiology's claim to be a science. Planetary science has been revolutionized by our ability to send space probes throughout the solar system, and to identify meteorites on Earth that were delivered from Mars. Astronomers using optical telescopes have discovered planets orbiting other stars and are developing strategies to search for evidence of life on such planets. In addition, radio astronomers and engineers have continually improved their ability to detect signals from extraterrestrial intelligence, with major new projects planned at Berkeley and elsewhere. Through molecular biology, we are tracing the history of life back to its earliest roots, and biochemists have made substantial progress towards achieving self-replicating biomolecules in the laboratory--the kind of molecules that must have preceded the first living cells.

Certainly this subject has religious and philosophical implications as well, some of which

1

2 CHAPTER 0. INTRODUCTION we will touch upon in the final chapter. In particular, the origin and uniqueness of life has long been considered a subject for religion. Over the past few centuries, though, science has increasingly encroached upon what used to be the domain of belief. Today we consider the Universe to be run by according to certain fundamental "laws", and most scientists believe that the most intricate activities of life and mind can be explained as the interaction of particles obeying these laws. The scientific approach has been so successful in explaining the physical world on all scales that we have every reason to believe that a scientific understanding of life's origins is possible. Yet we should not become overconfident: at this point we lack even a clear definition of what life is. And even if the scientific approach proves to be successful, we may still find the answers it gives unsatisfying: as physicist John Wheeler has said, "The more we understand the Universe, the more it seems pointless."

The philosopher Karl Popper cautioned us that scientific theories had to be falsifiable: they must be subject to tests which can prove them wrong. Although scientists have not always embraced falsifiability in practice, it is a widely accepted ideal, and in our investigations we should try to keep it in mind. In particular, we must be wary of explanations that can be fine-tuned to fit any experimental or observational results. At present we have few observational constraints in many areas of astrobiology: we know of only one place where life has certainly arisen (Earth), and even the record of life's origins here has been erased by geological processes. Many of the ideas we will explore are currently popular for various reasons, but are not yet capable of being tested--and thus must not be mistaken for scientific fact.

A final word of warning: nowadays, there are many scientific results being announced in various fields of astrobiology, and this research is to some extent driven by public excitement about the field. This is wonderful to the extent that it reinforces the value and importance of curiosity-driven science to our civil society. However, it also leads to a situation in which the traditional process of peer review and verification can be sidestepped--the scientist can take his or her case directly to the public through the news media. A primary goal of this course is to make you, as an educated citizen, evaluate what is said in the newspapers by weighing the arguments and evidence.

Chapter 1 What is Life? Although biologists have learned a great deal about the living world, we still lack a clear definition for life. We know that all life on Earth shares a unique biochemistry based on proteins and nucleic acids, but we would like to define life in a way that might encompass other biochemistries as well. Physiological definitions also run into problems. While most living things reproduce, eat, breathe, and grow, not every organism performs all of these activities all of the time--for example, seeds can lay dormant for long stretches of time. In addition, crystals can grow and reproduce under the right conditions, and automobiles can be considered to eat, breathe, move, and consume energy! In this chapter I'll try to identify three necessary conditions that, taken together, distinguish what we would generally agree to be living. Then I'll review some of the basic knowledge we have of one type of life: life on Earth.

1.1 Fundamental Characteristics of Life 1.1.1 Non-Equilibrium Thermodynamics Thermodynamics is a branch of physics that deals with macroscopic (large-scale, as opposed to microscopic) systems and how they exchange energy. The second and most famous of the three basic laws of thermodynamics states the following: in a closed system (one that does not exchange matter or energy with its surroundings), the entropy (a measure of disorder) cannot decrease. Most natural and artificial processes lead to an increase in entropy because they convert "useful" energy, such as motion, into dissipated heat. While the total energy is conserved--according to the first law of thermodynamics--the amount of energy available to do useful work decreases.

Living organisms, on the other hand, are organized entities as the word implies. Ordinary matter is transformed by life into complex structures which would be extremely unlikely to arise from a random agglomeration of molecules. However, this increase in internal order occurs at the expense of greater disorder in the universe outside. Thus, as open systems, living systems do not violate the second law.

3

4 CHAPTER 1. WHAT IS LIFE?

How does life create internal order? Organisms need a continuous supply of energy to perform most of the chemical reactions on which life depends. In general, it is easier from a thermodynamic standpoint to break things down than to build things up: the former reactions are called exergonic because they release energy and the latter are called endergonic. By harnessing the high-energy photons of the Sun, cells can preferentially conduct endergonic reactions and thus build up complex molecules and structures. Without this constant flow of energy from a hot Sun to a (comparatively) cold Earth, the development of life would not be possible.

The Belgian scientist Ilya Prigogine considered life to be one of many "dissipative structures" which maintain themselves in a non-equilibrium state by continually extracting useful energy from their surroundings and disposing of heat energy into it. In living organisms, this energy is used for the synthesis and breakdown of organic compounds in what we call metabolism. But other dissipative structures exist in nature, such as Jupiter's Great Red Spot, a giant atmospheric cyclone which has survived for over three centuries by devouring smaller cyclones.

Scientists call the self-maintenance of living systems autopoiesis. Autopoietic entities maintain themselves by metabolizing continuously. Life uses the energy it consumes to constantly regenerate itself: your body replaces most of its atoms every year. Death is simply the cessation of metabolic activity and a return to thermodynamic equilibrium with one's surroundings.

1.1.2 Reproduction Based on Information Although not all individuals reproduce, all active organisms contain cells which do so. Reproduction is such a fundamental aspect of life that it is tempting to define life as a selfreproducing system. However, there are many examples of "reproductive" processes that are clearly nonliving: for example, the spread of flames in a wildfire, the growth of salt crystals in a supersaturated solution, and the creation of particles from energy in an accelerator. What distinguishes living organisms is that they make copies of themselves using information stored in genes, which (on Earth) are made up of special informational molecules called nucleic acids. Genes are the blueprint for the development of an organism: in multicellular organisms like ourselves, they tell the fertilized egg how to develop into a mature adult. Reproduction based on inherited information is qualitatively different from the growth of crystals or the spread of fires, each of which is directed by environmental conditions and physical law rather than internal instructions.

1.1.3 Capacity for Evolution Not only is life constructed according to internal instructions, but those instructions are flexible in the sense that slight changes to them can still produce viable offspring, and the resulting variations can then be acted upon by natural selection. By natural selection, I simply mean the differential survival and reproduction of individuals competing for the

1.2. THE CHEMICAL BASIS OF LIFE ON EARTH 5 same resources. More organisms are created than can possibly survive, and those that do not survive do not pass on their genes.

Natural selection is effective because random changes, or mutations, do occur in genes and can be transmitted from parent to offspring. If a mutated gene is adaptive (beneficial to survival and reproduction) it is passed on to more offspring, and thus becomes more widespread, displacing its precursor. We will discuss the origin of mutations further in x1.4. For now, it is simply worth pointing out that if reproduction always had to be perfect, nothing interesting would ever happen!

1.2 The Chemical Basis of Life on Earth The rest of this chapter will focus on developing a basic understanding of terrestrial life, as it is the only kind of life we have any information about. I give only a rough outline here; there are many books on biology that can provide greater detail.

The chemistry of life was once called organic chemistry. Since then, however, organic chemistry has expanded to include the study of all types of carbon compounds, including many that are not usually found in organisms. Hence, when we speak of organic compounds we are referring to carbon compounds that may or may not be biological. Many chemists would exclude carbon dioxide (CO

2

), carbon monoxide (CO), and their compounds

from this category; we will adopt that definition here.

For our purposes, we can separate organisms on Earth into two categories on the basis of either metabolism (autotroph/heterotroph) or cellular structure (prokaryote/eukaryote):

ffl Metabolism: Green plants, photosynthetic bacteria (the cyanobacteria, or blue-green

algae), and chemosynthetic bacteria use chemical or light energy from nonliving sources to sustain metabolism, and acquire carbon from nonliving sources such as CO

2

. Such

organisms are called autotrophs. Animals and most kinds of microorganisms derive their energy and supply of carbon from breaking down compounds formed by other organisms, and are called heterotrophs.

ffl Cellular structure: Bacteria have no nuclear membrane separating their DNA from

the rest of the cell, and are therefore classified as prokaryotes. All other organisms are eukaryotes, where the nuclear membrane serves to separate the storage of information in DNA from the process of protein synthesis which occurs in the cytoplasm, the region outside the nucleus. The cytoplasm of eukaryotic cells also contain membrane-bounded organelles such as mitochondria and chloroplasts.

A constant cycling of carbon and oxygen occurs between autotrophs and heterotrophs: plants take in carbon dioxide (CO

2

) and water (H

2

O) to form oxygen (O

2

) and sugars

(C

6

H

12

O

6

), while animals perform the process in reverse.

1

A similar cycle, involving plants,

1

Plants consume oxygen and break down sugars too, but in the daytime they are net producers of these

substances.

6 CHAPTER 1. WHAT IS LIFE? animals, and certain kinds of bacteria, exists for nitrogen. A useful construct for encapsulating these cycles is the biosphere: the set of all living organisms on Earth. While matter cycles through the biosphere, we've already seen that energy flows through it in only one direction: from high-energy solar photons to dissipated heat.

So many molecules occur in living things that categorizing them all seems to be a hopeless task. Fortunately, however, large biomolecules are constructed from a limited number of subunits. In general, relatively simple molecules that can be easily joined together to form larger molecules are called monomers, and the larger molecules they form are called polymers, or macromolecules. An example of a monomer is an amino acid, of which 20 kinds are found in living organisms. Polymers of amino acids are called proteins, of which there are effectively an unlimited number, just as there is no limit to the number of books that can be written using 26 letters of the English alphabet. Similarly, DNA is constructed from 4 basic monomers called nucleotides. In the following pages we will discuss proteins and nucleic acids in more detail. The two other major classes of biopolymers are lipids and carbohydrates, which are used for energy storage and structural support.

1.2.1 Proteins Proteins are involved in every aspect of life. As enzymes, they catalyze virtually all biochemical reactions, often at speeds of 10

10

times faster as they would naturally proceed.

They form the cellular architecture (the cytoskeleton) as well as the motors that allow cells to move. They regulate the entry and exit of substances through the cell membrane and transmit signals to other cells. The list goes on: consult any basic biochemistry textbook.

Protein structure. All proteins are composed of amino acids, arranged in a particular order according to instructions provided by DNA (next section). Although there are many types of amino acids found in nature (even in meteorites), only 20 are used by life. In particular, the same amino acid can appear in "left-handed" or "right-handed" configurations, one being the mirror image of the other, but only left-handed versions (L-amino acids) are found biologically, with very rare exceptions. Having all amino acids possess the same handedness improves the efficiency of many chemical reactions, although the choice of right or left seems to have been arbitrary.

2

Surprisingly, less than 10

5

protein molecules are actually found in living organisms,

compared to the 20

100

possibilities for a string of 100 amino acids, which is a typical length

for a protein. This probably reflects the fact that proteins, like organisms, have been subject to natural selection. The function of a protein is strongly determined by its three-dimensional folded structure, which in turn is highly dependent on the amino acid sequence. For instance, the replacement of a single amino acid in the hemoglobin protein, which carries oxygen in the bloodstream, leads to a deformation of red blood cells known as sickle-cell anemia.

2

Some astronomers have speculated that the cloud from which the Solar System formed had more L-amino

acids because other kinds were preferentially destroyed by polarized starlight.

1.2. THE CHEMICAL BASIS OF LIFE ON EARTH 7 How enzymes work. Enzymes speed up reactions by lowering the activation energy that must be achieved before the reaction takes place. Many reactions are thermodynamically favored (that is, there is a net release of free energy in going from reactants to products) but do not happen spontaneously unless some energy (the activation energy) is added first. Think of a golf ball in a sand trap at the top of a hill: the ball will not roll down unless it is given enough of a push to get out of the sand trap, even though it has more potential energy at the top of the hill than at the bottom. One can give the reactants more energy by heating them, but this is generally not a preferred solution in living cells, where the temperature must be carefully controlled. Instead, the energy barrier itself is lowered by catalysts, which participate in the reaction without being consumed.

The key factor for catalysis is the unique folded structure of each enzyme, which allows specific molecules or parts of molecules to bind to it at certain active sites. For example, an enzyme may link two molecules by attaching to them at its active sites, then assuming a configuration that encourages the two to attach to each other. Catalysis by enzymes has been referred to as a "lock and key" mechanism, since each enzyme catalyzes only one type of reaction. Enzymes are architectural wonders of nature, honed over the eons to allow specific reactions to occur extremely quickly.

Nitrogen fixation. If you've ever tried to grow a plant in a cup of water, you probably found that water, air, and sunlight were not enough to produce a healthy plant: a bit of "plant food" was necessary as well.

3

One reason for this is that the formation of amino

acids requires nitrogen, as the similarity between "amino" and "ammonia" (NH

3

) suggests.

Although nitrogen gas (N

2

) is the principal constituent of air, it does not readily participate

in chemical reactions. Before the accumulation of oxygen (O

2

) in the atmosphere, N

2

could

be converted to more usable ammonia (NH

3

) by solar UV radiation, but the formation

of the ozone (O

3

) layer eventually put an end to this natural process. As a result, life

came to depend on certain bacteria that can convert N

2

to NH

3

through a process called

nitrogen fixation. Although this process is energetically costly and can only take place under anaerobic (oxygen-free) conditions, it is critical to life today.

1.2.2 Nucleic Acids The nucleic acids come in two forms: DNA (deoxyribonucleic acid) and RNA (ribonucleic acid). In modern organisms, DNA is the principal repository of genetic information (the "blueprint" of the cell); RNA serves an accessory role in copying the DNA sequence and transporting it to locations where protein synthesis occurs. Nucleic acids get their name from the fact that DNA in eukaryotes is located within the cell nucleus, as parts of organized structures called chromosomes. DNA and RNA are involved in three principal activities:

1. Replication: The formation of an exact copy of an original DNA molecule. This is the

basis of heredity. The DNA in all of your cells (except your gametes) is identical, and

3

Plants in very nutrient-poor environments sometimes resort to eating insects! Or more precisely, plants

that can eat insects can thrive in nutrient-poor environments.

8 CHAPTER 1. WHAT IS LIFE?

constitutes your genome. Cells themselves take on a variety of shapes and functions as a result of chemical signals that turn "on" and "off" various parts of the DNA blueprint.

2. Transcription: The formation of an RNA sequence that is complementary to part of

a DNA molecule, usually a section that codes for a particular protein. This allows the DNA to stay in the nucleus while its information is carried by "messenger" RNA to other parts of the cell.

3. Translation: Conversion of the nucleotide sequence in the messenger RNA to the

amino acid sequence for a protein. This is accomplished by "transfer" RNA which associates each triplet of nucleotide bases to an amino acid.

Structure and replication. The monomeric subunits of DNA and RNA are called nucleotides. Each nucleotide consists of three parts: (1) a five-carbon sugar (ribose or deoxyribose for RNA or DNA respectively), (2) a phosphate (PO

4

) group, and (3) a nitrogencontaining base. There are four possible bases for a DNA nucleotide--adenine (A), guanine (G), thymine (T), and cytosine (C)--while RNA has the same bases except that uracil (U) replaces thymine. The DNA or RNA molecule then consists of a sequence of nucleotides, with alternating sugars and phosphates forming a molecular "backbone" and the sequence of bases encoding all of the genetic information.

In the case of DNA, there are actually two parallel "backbones", forming a twisted ladder or double helix configuration, as discovered by James Watson and Francis Crick in 1953. The steps or rungs in the ladder are made up of weakly bonded base pairs, with A always paired with T and G always paired with C. It is clear that given half of the ladder and these matching rules, the other half the molecule is determined: thus, each half contains all of the information stored in the molecule. Replication of the molecule is therefore straightforward: the ladder is split ("unzipped") by appropriate enzymes and each half serves as a template from which two complete and identical double helices can be formed by adding the matching nucleotides.

The replication of nucleotides is highly accurate. In the bacterium E. coli, a mistake is made only once for every 10

9

to 10

10

nucleotides added. The "first-try" error rate is actually

about 10

5

times higher; cells have developed elaborate proofreading mechanisms to correct

almost all errors. These mechanisms include special enzymes that catalyze the removal of mispaired bases and undo the linkage of two consecutive T bases which can result from UV irradiation.

An RNA molecule does not generally have two complementary strands, so it does not form a double helix structure. Instead, it forms a complicated folded structure, not unlike a protein. In fact, as we'll discuss in Chapter 4, RNA may have served as a biological catalyst prior to the development of proteins.

Protein synthesis. Nucleic acids are really just instructions on how to make proteins. Three consecutive nucleotides correspond to one amino acid; this correspondence is what

1.2. THE CHEMICAL BASIS OF LIFE ON EARTH 9 we call the genetic code. Using this code, the sequence of nucleotides can be read as a sequence of amino acids. With 4 bases to choose from, there are a total of 4

3

=64 possible

3-letter codes (called codons), more than enough to code for all 20 amino acids--hence, the code is redundant. In many cases, more than one codon corresponds to the same amino acid; some of the other extra codons are flags indicating where the protein begins or ends.

The process of protein synthesis begins with the transcription of DNA to a type of RNA called messenger RNA (mRNA). At first it was believed that this was the only direction in which transcription could occur; in fact, Crick's central dogma was that information flowed in only one direction:

DNA \Gamma ! RNA \Gamma ! protein: We now know of enzymes called reverse transcriptases that can transcribe RNA sequences to DNA; these are found most prominently in certain viruses called retroviruses (like HIV). The ability to go from RNA to DNA is critical because there are strong arguments that RNA preceded the creation of DNA.

The RNA transcript often requires some editing before it can be translated into a protein: in particular, sections of DNA called introns are not used in protein synthesis, and the corresponding RNA segments need to be spliced out. The function of this "junk DNA," common among higher eukaryotes but not bacteria, is unknown. It is possible that the advantages of getting rid of introns were never sufficient to drive evolution in eukaryotes, while there was greater pressure to get rid of them in fast-reproducing bacteria. Walter Gilbert has suggested that it is not the introns themselves but the fact that they separate useful sequences (exons) that has adaptive value: the amino acid chains that exons code for may serve as intermediate building blocks for a variety of useful proteins. In this interpretation, evolution proceeds faster by shuffling exons around than by relying on mutations to change proteins one amino acid at a time.

Finally the edited mRNA is transported to special RNA-protein complexes called ribosomes, where transfer RNA (tRNA) molecules, each one attached to a particular amino acid, deliver free amino acids for protein synthesis. Each tRNA also contains a triplet of bases (the anticodon) which binds to a codon triplet in the mRNA template according to the usual matching rules (G , C, A , U). Once a proper codon-anticodon match has been made,

4

the tRNA releases its amino acid to the growing protein chain. Note that the key

step in translation is the attachment of the correct amino acid to a tRNA molecule, which occurs outside the ribosomes.

1.2.3 Energy Utilization Life requires energy to perform its activities, and the common energy currency used in all cells is a molecule known as ATP (adenosine triphosphate) and close relatives of it. ATP consists of adenine (one of the nucleotide bases), ribose, and three phosphate (PO

2\Gamma

3

) groups,

4

In some cases, only 2 of the 3 bases need to match; this is OK because there are more codons than amino

acids.

10 CHAPTER 1. WHAT IS LIFE? up to two of which can be detached to release energy. The phosphate bonds that must be broken for this to occur are called "high-energy" bonds because breaking them by addition of water (hydrolysis) yields a net energy release. Nonetheless, the bonds are still stable and require some activation energy before they can be broken. For most reactions that require energy from ATP, the bonds are not simply hydrolyzed, as this would release the energy unproductively as heat; instead, a phosphate group is temporarily transferred from ATP to one of the reactants, giving it an energy boost sufficient to allow the reaction to occur.

Redox reactions. The formation of ATP itself generally involves oxidation-reduction, or "redox" reactions, which I'll briefly define here. In such a reaction, electrons are transferred from a donor atom or molecule, which is "oxidized", to an acceptor which is "reduced." Thus both oxidation and reduction happen simultaneously. What can be confusing is that we call a reducing agent one that tends to lose electrons and become oxidized, while an oxidizing agent tends to gain electrons and become reduced. Again, the basic rule is:

ffl Reducing agents want to donate electrons and become oxidized. ffl Oxidizing agents want to accept electrons and become reduced.

Here's a simple example employed by methane-producing bacteria:

4 H

2

+ CO

2

\Gamma ! CH

4

+ 2 H

2

O:

In this case, the carbon dioxide is reduced by the addition of hydrogen, while the hydrogen is oxidized by the addition of carbon dioxide. The reducing agent is H

2

and the oxidizing

agent is CO

2

. In fact the hydrogen doesn't actually lose its electrons, but holds them less

tightly in H

2

O than it did as H

2

. Similarly for carbon, which can hold onto its electrons

more tightly after going from CO

2

to CH

4

. The redox reactions we'll deal with will generally

follow this pattern: the reducing agents are rich in hydrogen while the oxidizing agents are rich in oxygen.

With a suitable choice of donor and acceptor, the transfer of electrons leads to a large enough energy change to drive endergonic reactions such as ATP formation, just as the flow of electrons from one terminal of a battery to the other can drive your portable stereo. The principal pathways for ATP generation are cellular respiration, photosynthesis, and chemosynthesis.

Respiration. Most cells release energy for the formation of ATP by breaking down carbohydrates (chiefly sugars). For aerobic (oxygen-breathing) organisms, the net reaction can be written as:

C

6

H

12

O

6

+ 6 O

2

\Gamma ! 6 H

2

O + 6 CO

2

+ energy:

However, the reaction does not occur in one step: the uncontrolled release of this much energy would be much too large for the delicate environment of the cell. Instead, an intricate network of redox reactions harnesses that energy to join many ADP (adenosine diphosphate) molecules to phosphate ions, forming ATP.

1.2. THE CHEMICAL BASIS OF LIFE ON EARTH 11

The first phase of respiration is the splitting of one molecule of glucose (C

6

H

12

O

6

) to

two molecules of pyruvate, with a net gain of 2 ATP's. This multi-step process, known as glycolysis, has been conserved throughout the living world, and is the oldest surviving pathway for energy generation. What happens next depends on whether or not the organism makes use of O

2

. Anaerobic organisms do not, converting pyruvate into lactate or ethanol

by the process of fermentation, with no additional gain of energy. Yeast is the best-known example: its conversion of sugar to ethanol (potable alcohol) has been utilized by humans for centuries. In aerobic organisms, pyruvate enters another chemical pathway, the citric acid cycle, where it is oxidized into CO

2

and water with a net gain of up to 36 additional

ATP's. Oxygen, which is combined with hydrogen to form water, serves as the final electron acceptor in this pathway.

Clearly, the ability to utilize oxygen to break down pyruvate allows for a huge increase in the efficiency of respiration. In aerobic prokaryotes, the necessary steps occur in the cell membrane, while in eukaryotes they occur in special "organelles" called mitochondria, which are believed to have once been free-living bacteria that took up residence in other cells (see x4.6.2). But aerobic respiration only became feasible about 2.2 Gyr ago, when significant amounts of oxygen began to accumulate in the atmosphere. In times of physical exertion when not enough oxygen reaches our muscle cells, even humans fall back on the ancient process of fermentation: lactate builds up in the bloodstream, to be regenerated into glucose during the recovery period (with the help of O

2

from some heavy breathing!).

Photosynthesis. The capture of solar energy for metabolic use occurs in two main steps. In the light reactions, so called because they occur only in the presence of light, photons excite electrons in special pigment molecules called chlorophylls. These pigments absorb most strongly in the red and blue parts of the visible spectrum, while reflecting the intermediate wavelengths of sunlight; hence their green color. The excited electrons are released to a series of enzymes that harness their energy to form energy-storage compounds such as ATP. Meanwhile, the electrons lost by chlorophyll are replaced by splitting H

2

O, yielding

O

2

. In the second step, called the dark reactions because they can occur in the dark, the

stored chemical energy is used in the reduction of CO

2

to carbohydrates. The net reaction

is then the reverse of aerobic respiration:

6 H

2

O + 6 CO

2

+ sunlight \Gamma ! C

6

H

12

O

6

+ 6 O

2

:

In plants both kinds of reactions occur in specialized organelles called chloroplasts, which like mitochondria possess some of their own DNA, suggesting that they too were once freeliving bacteria.

It is interesting to note that only UV photons have enough energy to break apart water into oxygen and hydrogen directly. Thus the large amount of oxygen in the Earth's atmosphere is quite unexpected given the small output of UV radiation from the Sun. The trick, of course, is that life has developed a system (photosynthesis) that can split water using the combined energy of several visible photons (four photons for each H

2

O molecule) instead of

one UV photon.

A more primitive form of photosynthesis, which does not produce oxygen, is employed by certain bacteria, such as the "purple bacteria." Electrons are provided not by water but

12 CHAPTER 1. WHAT IS LIFE? by other compounds, such as hydrogen sulfide (H

2

S). This "anaerobic" photosynthesis has

been incorporated into the first stages of oxygen-yielding photosynthesis, indicating that it is more ancient. The switch to using water as the electron donor may have been precipitated by an increasing scarcity of other donors.

Chemosynthesis. One of the great surprises in biology of the past twenty years has been the discovery of ecosystems that flourish around deep-sea hydrothermal vents, where virtually no sunlight penetrates. These ecosystems are dependent on autotrophic sulfur bacteria that acquire energy from the oxidation of hydrogen sulfide (H

2

S):

2 H

2

S + 3 O

2

\Gamma ! 2 H

2

O + 2 SO

2

+ energy:

(Note that this is different from photosynthesis using H

2

S, where the energy source is still

light.) This chemical energy is then used to reduce CO

2

to carbohydrates. In deep-sea

environments, sulfur bacteria are often found in symbiotic relationships with animals such as tube worms, which provide them with raw materials (O

2

, CO

2

, and H

2

S) in exchange for

some of the nutrients produced. Since the O

2

in the oceans ultimately comes from surface

photosynthesis, however, these communities are not truly self-sufficient.

Another common reaction used by chemoautotrophs known as methanogenic (methaneproducing) bacteria is the oxidation of H

2

by CO

2

:

4 H

2

+ CO

2

\Gamma ! 2 H

2

O + CH

4

+ energy:

In principle, such bacteria could be entirely self-sufficient if they had a source of hydrogen, but in practice they generally seem to get their hydrogen from other organisms. In 1995 it was reported that methanogenic bacteria in deep underground aquifers got their H

2

from a

geochemical reaction between basalt and water, but recently that result has been questioned.

5

Regardless of the present situation, the early Earth may have had enough free hydrogen available for this reaction to support simple food chains.

1.3 Information Content of Life The information content in DNA or proteins can be quantified by information theory. Suppose we have a message of length n written in an alphabet with k letters. For definiteness, consider all possible two-letter codes using uppercase letters (n=2 and k=26). The number of possible messages is 26

2

= 676, since there are independently 26 choices for the first letter

(A--Z) and 26 choices for the second. In general, the number of possible messages is:

N = k

n

:

We define the information content of a message as:

H = log

2

N = n log

2

k ;

5

See Science, 270: 377 (20 Oct 95) and 281: 976 (1998).

1.4. MUTATION AND GENETIC DIVERSITY 13 where H is expressed in binary units, or bits. For the example given above, H=9.4 bits. These units may be familiar to you from working with computers: 8 bits equals a byte, a common unit for data storage. H can be thought of as the number of on/off switches needed to encode a message. From this definition, we see that the information content of a message is a measure of how improbable it is. The more letters there are in a message, the harder it is to produce it by chance. A useful property of H is that it is linear in n: a message that is twice as long has twice the information content.

A typical gene contains n=1200 nucleotide bases of k=4 kinds, so in principle it has an information content of 2400 bits. But the protein that it codes for consists of n=400 amino acids of k=20 kinds, and thus has an information content of only 1729 bits. What happened to the other 671 bits? Apparently they were "lost in translation" because the genetic code is redundant: as discussed in x1.2.2, there are 64 possible codons but only 20 biological amino acids.

The genome of the common bacterium Escherichia coli contains about 5\Theta 10

6

base pairs,

which represents about 10 million bits of information (less considering the redundancy of the genetic code). This would easily fit on a floppy disk. The human genome is a good deal more complicated, with about 3 billion base pairs, equivalent to 6\Theta 10

9

bits of information. (Again,

this ignores the redundancy of the code and the presence of introns--95% of human DNA has no discernible purpose.) Still, this is less than 1 gigabyte of data, just a fraction of the disk space on the laptop computer with which I write these words! This in no way trivializes the complexity of life, which takes advantage of not only internal information but properties of the physical world--the particular way in which a protein molecule folds, for instance, or the characteristics of ATP that make it suitable as an energy storage molecule. DNA may contain all the instructions, but it is the enormous possibilities of organic chemistry that make those instructions useful.

1.4 Mutation and Genetic Diversity Changes in the DNA sequence are called mutations and are believed to be ultimately responsible for all of the variety that we see in the living world. The simplest kind of mutations are point mutations that affect individual base pairs: these can arise from damage to the DNA molecule (by cosmic rays, gamma rays, or certain chemicals) which is incorrectly repaired, or from an error in replication of the molecule. Occasionally, a helpful mutation arises by chance, and if it confers sufficient advantages it will become widespread in the population. Moreover, an organism's habitat can change quickly due to changes in the environment, and a seemingly "useless" mutation may suddenly become valuable.

This classical view, that evolutionary change is the result of gradual accumulation of point mutations, is probably incomplete. One difficulty is that almost any innovation requires many nucleotides to be changed, which in turn requires mutations over many generations, but each of those generations would have to survive the (potentially harmful) mutations it had collected up to that point. Sex helps a bit (see below), by allowing you to combine mutations from both parents, but the odds of putting together a useful adaptation in a reasonable amount of time are still miniscule.

14 CHAPTER 1. WHAT IS LIFE?

So how can we explain the incredible diversity and complexity of life, if not by random point mutations? Creationists count this as a strike against Darwinism, but in recent years a number of other mutation processes have been identified. These include the following:

1. Transposable elements: DNA segments, ranging in size from a few hundred to a

few thousand base pairs, that can be detached and re-inserted into new locations in the genome.

2. Variations in mutation rate: There is growing evidence that some organisms can

increase their mutation rates under conditions of environmental stress. In addition, the DNA repair enzymes are more effective on some stretches of DNA than on others, creating "hotspots" of rapidly mutating DNA.

3. Duplication of genes: This can involve a stretch of DNA, or even the entire genome.

The duplicated genes can be inserted as transposable elements into different parts of the genome, where they may take on new functions.

Thus, there is growing consensus that the cell contains many mechanisms for genetic change, and the ability to evolve has itself been selected for by evolution. Nobel laureate Barbara McClintock has called the genome "a highly sensitive organ of the cell, that in times of stress could initiate its own restructuring and renovation."

6

Sex is a peculiar innovation that has been widely adopted by multicellular organisms, although even bacteria occasionally indulge in a form of sexual activity (conjugation--transfer of genes from one cell to another). The disadvantages of sex, in terms of the energy resources that must be consumed in finding a mate, are enormous, so there presumably must be some advantages as well! The advantage is probably the facility with which variation can be achieved by the reshuffling of genes: no offspring is exactly like either of its parents. For larger organisms that reproduce much more slowly than the prolific bacteria, sex is probably the best way to spread helpful mutations quickly through a population, which may be vital as protection against parasites that are also mutating. Sex can also get rid of harmful mutations by producing some offspring that are mutation-free.

1.5 Viruses and Cell Fragments No survey of life on Earth would be complete without reference to the strange entities we call viruses. Viruses consist of nucleic acids encased in a coat of proteins. Once a virus penetrates a cell, however, it springs into action, diverting the cell's resources to creating additional copies of itself and often destroying the cell in the process. No one knows when viruses first appeared, but it is almost certain that they arose after cells because of the fact that they require cells to metabolize and reproduce. For this very reason viruses are generally not considered "alive"--on the other hand, it is really just a matter of definition: only certain autotrophs can survive in complete isolation from all other lifeforms. Perhaps it would be

6

Quoted in Science, 281, 1131 (21 Aug 1998).

1.6. VITALISM VS. MECHANISM 15 better to put viruses on one end of a continuum of life that goes on to parasites (which are dependent on other organisms for nutrients but have their own metabolic apparatus), heterotrophs, and autotrophs.

Infectious agents even more bizarre than viruses exist: the viroids, which are essentially free-floating RNA molecules, and prions, which are proteins. Viroids are responsible for various diseases affecting higher plants, while prions have been implicated in CreutzfeldtJakob disease in humans and "mad cow" disease in livestock, both of which lead to severe deterioration of the brain. These biochemical fragments are like pieces of the cell machinery gone awry. It is conceivable that by studying them, we'll be able to learn more about how molecules interacted to form the first cells.

1.6 Vitalism vs. Mechanism For centuries it was believed that living matter was fundamentally different from normal matter, that it contained some kind of spirit or "vital force." With the rapid development of biochemistry and molecular biology in the past decades, this vitalistic view has been largely replaced by a mechanistic view, according to which the activities of life are reducible to physical processes. In particular we have learned that the processes occurring in living cells are ordinary chemical reactions, but involving very complex molecules. Granted there are properties of a complex system that are not apparent at the level of its individual components (e.g. atoms), which is why chemistry and biology survive and prosper in spite of the fact that they are "just physics." Yet that is very different from postulating a different form of matter or energy that is found only in living things. The mechanistic view opens the door to an abiotic origin for life, and suggests that part of the difficulty in defining life is that the transition from non-living to living systems is a continuous one.

Philosophically, the mechanistic worldview has been criticized as being completely deterministic: leaving no room for free will. Scientists have responded in two different ways: some, like Roger Penrose, suggest that consciousness is a quantum phenomenon, since quantum physics is probabilistic: the range of possible outcomes is predetermined but the actual outcome is determined by chance. Others, like Stuart Kauffman, describe thought as an emergent process, reducible in principle but not in practice to chemical activity. Like a chaotic system, for which even the slightest change in initial conditions leads to completely different behavior down the road, our thoughts and actions may be in a sense predetermined but involve so many inputs that they are effectively unpredictable (at least to ourselves).

16 CHAPTER 1. WHAT IS LIFE? Chapter 2 Constraints on Extraterrestrial Life We now step away from the Earth to consider more generally the necessary conditions for life. This is a difficult problem, not only because we are familiar with only one kind of life, but also because life, along with geological processes, have dramatically altered our planet to the extent that it is difficult to infer the conditions under which it arose.

I should point out that in discussing the requirements for life, I am referring to life that can arise from nonliving matter by chemical processes. A race of self-reproducing metal robots is quite conceivable, but they would have to be created first by other lifeforms, as the spontaneous creation of metallic life from a bunch of rocks seems very improbable.

2.1 Habitat Today life on Earth is found in a wide variety of environments, from geothermal hot springs in Yellowstone National Park (T ? 90

ffi

C) to Antarctic lakes that are still liquid at \Gamma 23

ffi

C,

and from the stratosphere to the bottom of the ocean. Most complex organisms, however, are much more limited in their possible habitats: warm-blooded animals such as birds and mammals have to regulate their internal temperatures quite precisely. Also, the presence of primitive life in extreme environments does not necessarily indicate that it arose there, as we'll discuss in Chapter 4.

What are the minimum requirements for a biological habitat? First, there must be a steady source of energy, such as a nearby star, tidal friction, or leftover heat from the formation of the planet. Second, the habitat should be shielded from high-energy radiation that might be damaging to complex biomolecules. This generally implies a moderately thick atmosphere or a subsurface habitat. Third, there should be an interaction medium in which chemical reactions can occur quickly and which is relatively stable to temperature changes. Such a medium would probably be a liquid, although a very dense gas might suffice. The advantage of a liquid is that it can be more easily contained within a localized region, while gases tend to diffuse away.

Thus, we expect extraterrestrial life to begin on planets or moons with atmospheres and/or liquid oceans. This may seem a little chauvinistic, based purely on our own

17

18 CHAPTER 2. CONSTRAINTS ON EXTRATERRESTRIAL LIFE experience, but it may not be that severe a constraint, as we'll see when we discuss objects in our Solar System. As far as we know, planets and their satellites are the only astronomical bodies that are cool enough to have interesting chemistry while dense enough that those reactions can occur fairly quickly.

2.2 The Importance of Carbon Carbon atoms are the structural backbones of biological molecules on Earth, and they have been selected for that purpose because of their unique ability to form a huge variety of macromolecules. Large molecules are a prerequisite for life because only they can store complex information. In particular, three properties combine to make carbon compounds the best choice for life:

1. Abundance. Carbon is the fourth most abundant element in the universe, after H,

He, and O, and a major constituent of interstellar dust grains. Since the path to the origin of life involves many steps, it is certainly best to start with elements that are easy to find. On the other hand, carbon was relatively rare in the inner solar system at the time of the Earth's formation (x5.1).

2. Bonding flexibility. Since a carbon atom has four outer (valence) electrons, it can

form covalent bonds with 4 other atoms using single bonds, or fewer using double or triple bonds. Moreover, the bonds between C--C and C--H are of comparable strength, so given a simple hydrocarbon (such as CH

4

, methane) one can create a

larger one by replacing an H atom with another hydrocarbon chain! Silicon atoms also have 4 outer electrons, but the bonds between Si atoms are relatively weak, making it difficult to form stable chains. They also lack the ability to form double or triple bonds.

3. Reactivity. Carbon compounds are stable, but not so stable that they cannot be

induced to react further. In particular, large amounts of heat are not necessary for reactions to occur, at least if suitable catalysts are available to aid the reaction. Thus the abiotic

1

formation of complex organic molecules is quite feasible, and going back

and forth between oxidized (CO

2

) to reduced (CH

2

O) forms is relatively easy. Silicon,

on the other hand, tends to form very stable compounds with oxygen (such as quartz, SiO

2

) whenever oxygen is available, whereas the reduced form of silicon (silane, SiH

4

)

is unstable and can be formed only at very high temperatures (?700

ffi

C).

The latter two criteria therefore suggest that silicon, although more abundant on Earth than carbon, would not make a suitable basis for life. It has been argued that it is possible to create long-chain molecules using an alternating sequence of atoms, which would allow us to look for other candidates besides silicon to replace carbon. However, these suggestions have

1

Throughout this book, abiotic processes are those that can occur in the absence of life. Prebiotic

refers specifically to events that occurred prior to the origin of life.

2.3. THE IMPORTANCE OF WATER 19 yet to be elaborated in detail. The point remains that carbon is abundant in the Universe, virtually all of the complex molecules found in meteorites are organic compounds, and we know for a fact that organic compounds can form life. So even if carbon-based life is not the only possible kind, it is surely the most common.

2.3 The Importance of Water Water is one of the most unusual substances we know of, and possesses several properties that make it well-suited to serve as the chemical interaction medium for life.

1. Abundance. Hydrogen is the most abundant element in the Universe, and oxygen is

the third most abundant element, and the most abundant on Earth. The two gases will react violently to form water with the addition of just a little energy. The detection of water vapor in some of the most extreme astrophysical environments known (the nuclei of galaxies) confirms that it is ubiquitous in the cosmos. Note, however, that water, like carbon, may have been absent from the Earth at its formation, a point to which we will return in x5.1.

2. High boiling point and heat of vaporization. While most other liquids vaporize

at temperatures below 0

ffi

C, water remains liquid up to very high temperatures (100

ffi

C),

due to the strong affinity water molecules have for each other. This is a result of hydrogen bonding, in which the oxygen atom in H

2

O strongly attracts the electrons

of the two hydrogen atoms, leaving the molecule with a net polarity (excess negative charge around the O and excess positive charge around the H's). The temperatures which allow liquid water (0--100

ffi

C at standard pressure) are ideal for organic chemistry: warm enough to allow reactions to happen quickly, but not so hot that bonds are broken. Also, this range of temperatures is fairly wide, although not the widest among all liquids.

Water also has the highest heat of vaporization of any known substance--i.e., one must add a great deal of heat to vaporize it. This has two important consequences for life. First, slight seasonal variations do not cause the oceans to boil away--a lethal scenario. Second, evaporation of water has a large cooling effect, which helps organisms regulated their internal temperatures (by sweating, for instance).

3. Heat Capacity. Water has a high heat capacity, in the sense that more heat energy

is required to raise its temperature by 1

ffi

C than is required for most other substances.

This is also a result of strong hydrogen bonding, since more energy is needed to break hydrogen bonds than to break the van der Waals bonds which typically exist between molecules. The large heat capacity aids organisms in thermoregulation, since fluctuations in the amount of heat produced by metabolism have little effect on the overall temperature of the organism. Recall that the high heat of vaporization is also useful for this purpose.

20 CHAPTER 2. CONSTRAINTS ON EXTRATERRESTRIAL LIFE

4. Ice floats. Water solidifies into a very open crystal structure because the H-O-H angle

is almost exactly right for forming a tetrahedral lattice. Virtually all other solids will expand upon melting, but when ice melts, the crystal structure is partially broken and it becomes possible to fit other water molecules into the spaces that were created by the lattice. Thus, the same number of water molecules actually takes up less space in the liquid phase: liquid water is denser than ice.

The impact on life is the following: when lakes and rivers freeze in the winter, the ice collects at the surface because of its lower density. This keeps the water below insulated from freezing and ensures that when the air warms up the ice can melt rapidly. If ice formed at the bottom of lakes, it would tend to accumulate year after year until the lakes were entirely frozen.

2

On the other hand, as Goldsmith and Owen have pointed

out, a liquid with a higher density solid form would have the advantage that (in theory) cells could be frozen without rupturing.

5. Solubility and the hydrophobic effect. Water's love for itself is rivaled only by its

love for other polar molecules, making it an excellent solvent for polar substances such as NaCl (table salt) and organic acids; such substances are called hydrophilic (Greek for "water-loving"). On the other hand, nonpolar molecules such as hydrocarbons (C

x

H

y

) tend to be "caged in" by water molecules because they do not participate in

hydrogen bonding: oil and water don't mix. In other words, the water molecules have such a great affinity for each other that any nonpolar molecules tend to be crowded together, a condition known as hydrophobicity ("water-fearing"). When parts of a molecule are hydrophilic and other parts are hydrophobic, the hydrophilic parts tend to be pushed outward to face the surrounding water while the hydrophobic parts are pushed inward. This effect allows for the spontaneous formation of cell membranes, and gives proteins and nucleic acids their unique three-dimensional shapes. As we've noted, enzymes in particular rely on their shapes to catalyze reactions.

6. Self-shielding. Water has the unique ability to protect itself from UV radiation in

the following way: H

2

O vapor is dissociated (split) into hydrogen and oxygen, with

the hydrogen escaping (since it is so lightweight) and some of the oxygen forming a protective ozone (O

3

) layer, which prevents more UV light from penetrating the

atmosphere. On Earth, this effect is relatively minor compared to the ozone that results from photosynthetic production of oxygen.

Are there other liquids besides water that could serve as interaction media? For stability reasons the liquid should possess hydrogen bonds, limiting our choices to ammonia (NH

3

),

hydrogen flouride (HF), hydrogen cyanide (HCN), and organic alcohols (C

x

H

y

OH). Of these,

only HF is comparable to water in its ability to dissolve polar substances, but it is a poor candidate because of its scarcity in the universe. The other candidates could potentially be useful for life on worlds that are too cold for liquid water. Ammonia, for instance, is actually better than water at dissolving non-polar substances, which could be important in a different biochemistry. However, there are other problems with ammonia, such as the lack

2

This could happen even in temperate regions because oceans and lakes get colder the deeper you go.

2.4. CONTINUOUSLY HABITABLE ZONES 21 of self-shielding and the fact that it sinks upon freezing. In the end, the case for water is strong, but not as overwhelming as the case for carbon.

2.4 Continuously Habitable Zones If water is indeed a likely requirement for life, one can define a habitable zone (HZ) around any star which delineates a region in which liquid water can exist at its surface.

3

In fact, the

identification of such zones is quite a complicated matter, because a planet's atmosphere, not just its distance from its star, affects its ability to possess liquid water. The well-known greenhouse effect, the trapping of heat by certain gases such as CO

2

and water vapor,

is the most important consideration, and is discussed in x3.1. For instance, the average temperature of the Earth would be \Gamma 18

ffi

C--well below freezing--if it were not for the 33

ffi

C

of greenhouse warming provided by the atmosphere. A further complication is that as a star evolves, the amount of energy it gives off (its luminosity) increases, and the habitable zone moves outward. Astronomers therefore define a continuously habitable zone (CHZ) as the region in which liquid water can exist over timescales relevant to the evolution of complex life (several billion years). Again, this is only an approximate concept.

Planetary scientists have attempted to use computer models to determine the boundaries of the HZ and CHZ in our solar system. The results of work by James Kasting and collaborators

4

suggests that the Sun's current HZ ranges from at least 0.95 to 1.37 AU,

5

and

its CHZ from at least 0.95 to 1.15 AU. They note that further refinements to the models are needed, in particular including of the effects of H

2

O and CO

2

clouds. Still, their results

suggest that Earth is the only presently (and continuously) habitable planet: Venus lies at 0.72 AU from the Sun and Mars at 1.52 AU. Other scientists, such as NASA geologist Chris McKay, argue that Mars is within the habitable zone--the only reason it doesn't have water now is its small size. Clearly, there is much about planetary habitability we still don't understand.

What about stars besides the Sun? Stars are classified by their surface temperature along a sequence OBAFGKM,

6

where O stars are the hottest and M stars are the coolest;

our Sun is a G-type star. All other things being equal, stars hotter than the Sun have CHZ's further out, while cooler stars have CHZ's closer in. But stars hotter than F-type also exhaust their nuclear fuel in less than 2 Gyr, so are probably not good candidates for advanced life; they also produce a great deal of ionizing UV radiation which is harmful to life as we know it. At the other extreme, M-type stars have CHZ's so close to the star that any planet there would become tidally locked: rotating once per revolution and thus always showing the same face to the star, just as our Moon always shows the same face to the Earth. Such a planet would have a hot side and a very cold side, and probably no stable

3

We exclude subsurface oceans heated by internal or tidal energy, as may exist on Jupiter's moon Europa.

Life may exist in such environments but we have little hope of detecting it remotely.

4

J.F. Kasting, D.P. Whitmire, & R.T. Reynolds, Icarus 101:108 (1993).

5

The Earth-Sun distance is defined as 1 astronomical unit (AU).

6

The mnemonic here is "Oh Be A Fine Girl/Guy, Kiss Me." Other mnemonics have been suggested, but

they are not as successful because they are not based on sex.

22 CHAPTER 2. CONSTRAINTS ON EXTRATERRESTRIAL LIFE life-supporting atmosphere.

We conclude that F, G, and K stars are the most likely hosts for life-bearing planets. As we've noted, models predict that the CHZ has a fairly narrow range in radius, favoring planets with nearly circular orbits. This makes sense because highly elliptical orbits would lead to huge climate variations.

Chapter 3

Life Elsewhere in Our Solar System? Obviously, the only information we have about other possible planetary environments comes from our Solar System, although it is conceivable that some properties of extrasolar planets will be observable in the not-too-distant future. This is no cause for disappointment, however, considering how much we still have to learn about the "local" planets and their moons. In particular, there is a plausible case for liquid water on one object (Europa), a strong case for liquid water in the early history of Mars, and tantalizing evidence for oceans of hydrocarbons on Titan! Furthermore, it is instructive to consider why Venus and Mars do not have liquid water today--sometimes called the "Goldilocks Problem" because only the Earth seems to be "just right" for life today.

3.1 The Greenhouse Effect Before we consider the present conditions on Venus and Mars, we first must understand the so-called "greenhouse effect." The mechanism by which it works is straightforward, even though its results may be quite complex. The ground absorbs some of the Sun's visible radiation in the daytime, warms up, and radiates infrared (heat) radiation back towards space. Greenhouse gases in the atmosphere, chiefly CO

2

and H

2

O, absorb the outgoing

infrared radiation and re-radiate it in all directions, including back down to the surface. The atmosphere thus acts as an insulating blanket, trapping heat and warming the surface. In contrast, airless bodies like the Moon and Mercury experience huge swings in temperature between day and night, because at night they have difficulty holding on to the heat they absorbed during the day. (This is also why clear nights are often cold, since there is less water vapor in the air.)

Geological evidence in the form of sedimentary rocks (formed at the bottom of lakes and oceans) indicates that the Earth has kept a fairly constant temperature over the course of its history, at least constant enough to allow liquid water at its surface. The likely reason for this is that the greenhouse effect on Earth is self-regulating. This results from a balance between processes that remove CO

2

from the atmosphere (sinks) and those that resupply it

(sources):

23

24 CHAPTER 3. LIFE ELSEWHERE IN OUR SOLAR SYSTEM?

ffl Sinks: Atmospheric CO

2

is incorporated into sedimentary rocks as calcite (CaCO

3

)

as a result of chemical reactions that occur in the presence of water, collectively called weathering reactions. Another important (perhaps now dominant) CO

2

removal process is the formation of carbonate shells by marine organisms, which also end up in rocks when those organisms die. Finally, the carbon in plants comes from CO

2

, and

can be stored in the ground when they die as fossil fuels.

ffl Sources: New CO

2

is injected into the atmosphere by volcanic eruptions, animal

respiration, and our combustion of fossil fuels. The CO

2

released by volcanoes comes

(at least nowadays) from melting sedimentary rocks, while the other two sources come from plants. This completes the CO

2

cycle.

The Earth's temperature can be regulated by feedback effects such as the following. If the temperature rises, the amount of moisture in the atmosphere rises, increasing the rate of rainfall and runoff, which in turn aids the weathering reactions. This removes CO

2

from

the atmosphere and reduces the greenhouse effect. On the other hand, if it gets too cold, the weathering processes slow down, while new CO

2

continues to be outgassed from volcanoes,

increasing the greenhouse effect. In addition to these geological feedback effects there are biological ones as well (x4.7).

Is the greenhouse effect always self-regulating? Apparently not, as attested to by our neighboring planets, Venus and Mars. Venus now has an incredible 500

ffi

C of greenhouse

warming and virtually no water on its surface or even in its atmosphere. It may have experienced a "runaway greenhouse", in which all water was vaporized and could not prevent CO

2

from accumulating in the atmosphere. On the other hand, Mars shows evidence of once

having had liquid water, but is now a cold and desolate world, with its water likely trapped in polar ice caps and permafrost below the surface. It may have experienced a "runaway glaciation" as CO

2

was removed by weathering reactions but no longer replaced following

the cessation of volcanic activity. As for the Earth, an important and controversial issue is whether the rate at which we're pumping CO

2

into the atmosphere and clearing vegetation

is too fast for the Earth's compensating measures to kick in. If so, we may be driving our atmosphere to a new equilibrium--one that is perhaps quite hostile to us.

Indeed, explaining how the Earth has managed to avoid the hothouse conditions of Venus and the deep freeze of Mars throughout its history remains a subject of ongoing debate. The problem is particularly acute for the early history of Earth, when according to stellar evolution models the Sun was 25-30% fainter than it is today, yet there clearly was water. One solution to this faint young Sun problem, proposed by Carl Sagan and his former student Chris Chyba, is the additional greenhouse warming provided by ammonia (NH

3

),

which would be shielded against dissociating solar UV radiation by a layer of organic haze. James Kasting suggests instead that methane (CH

4

), perhaps generated by methanogenic

bacteria, provided the additional warming. An important role for bacteria is reminiscent of the "Gaia Hypothesis" championed by James Lovelock and Lynn Margulis (see x4.7), who claim that life plays a major role in regulating the Earth's climate.

3.2. MARS 25 3.2 Mars The possibility of life on Mars has captured the imagination of both scientists and the public for over a century, and continues to be a driving force behind current and future exploration of the Red Planet. Geological evidence seems to suggest warmer and wetter conditions early in Mars' history, which may have been suitable for the origin (or transplantation) of life.

3.2.1 From Lowell to the Dawn of the Space Age Although the famous Martian canali (Italian for "channels" or "canals") were first reported by Giovanni Schiaparelli in 1877, it was the American astronomer Percival Lowell who popularized the idea that they were engineered by intelligent beings. Lowell conducted his own observations of Mars starting in 1894 at an observatory in Flagstaff whose construction he financed himself. Through long nights at the telescope, squinting through the eyepiece and waiting for a brief pause in the atmospheric "seeing" (blurring due to turbulence), he made detailed sketches of a criss-crossing network of linear canals. In 1909, however, the French astronomer Eugene Antoniadi showed that complex shadings in the Martian surface had been mistaken by Lowell (and many others) for canals. Lowell carried his convictions to his death in 1916, even after Antoniadi's interpretation was essentially confirmed by the new 60-inch telescope on Mt. Wilson.

In the years following Lowell's death, interest shifted from the question of whether there was intelligent life on Mars to whether there could exist "simple" life such as vegetation. (In fact, plants are not simple at all, as evidenced by the length of time it took for them to appear on Earth). One argument for vegetation was the seasonal changes in color seen on the Martian surface. Measurements in the 1920's showed that some parts of the surface achieved temperatures above freezing, while early spectroscopy hinted at the presence of water and oxygen in the atmosphere. Even when the atmosphere was found to be dominated by CO

2

,

scientists as late as 1959 were assigning spectral absorption features to vegetation.

3.2.2 Mariner and Viking With the advent of space exploration, NASA made the search for extraterrestrial life a high priority, especially in planning missions to Mars. Enthusiasm was initially dampened by pictures of a dead world relayed to Earth by the Mariner flybys (1964-9) and measurements of an extremely thin atmosphere, with less than 1% of the Earth's air pressure. The Mariner 9 spacecraft, the first to orbit Mars, showed a world devoid of canals and vegetation--the seasonal changes in color were due to dust storms. Nonetheless the planet was fascinating in its own right, with volcanoes and canyons larger than any on Earth. Most notably, some of the smaller canyons resembled dry river beds on Earth, igniting speculation that water once flowed freely on the surface. The possibility of microbial life was compelling enough to place biology experiments aboard the two Viking landers which arrived on Mars in 1976. They performed three types of experiments:

26 CHAPTER 3. LIFE ELSEWHERE IN OUR SOLAR SYSTEM?

1. Labeled release experiment: An aqueous solution of seven simple organic compounds (tagged with radioactive

14

C and

35

S) which bacteria like to eat was added

to some Martian soil, and the surrounding chamber tested for radioactive gases that would be released after digestion.

2. Gas exchange experiment: Water or a nutrient-rich solution was placed underneath

or in contact with a soil sample. The gas surrounding the soil was sampled for evidence of H

2

, N

2

, O

2

, CH

4

, or CO

2

that might be produced by Martians.

3. Carbon assimilation (pyrolytic release) experiment: Radioactively-tagged CO

and CO

2

gas was added to a soil sample under the assumption that any organisms

would have adapted to "breathe" these gases on Mars. The soil was then heated so that any new organic compounds would be broken down (pyrolyzed) to release the tagged CO

2

, indicating that carbon had been taken up from the air by the soil.

The first two experiments assumed that life on Mars would be more or less Earth-like, while the third assumed only that it would be based on carbon. Surprisingly, all three experiments yielded some kind of positive signal, although whether there was biochemistry at work or just exotic soil chemistry was uncertain. For instance, the third experiment showed that some carbon had been converted to organic (but not necessarily living) material by the Martian soil, and the second experiment showed that humidifying the soil yielded both oxygen and carbon dioxide. In the end, the Viking scientists came to the conclusion that the results of the second and third experiments were probably non-biological (positive results still occurred after the soil was heated to over 145

ffi

C), whereas the outcome of the first experiment, which showed the release of radioactive gas following the addition of radioactive nutrients, was deemed ambiguous (its designer, Gilbert Levin, still maintains that the biological explanation is tenable). The main reasons for skepticism were the failure to detect any organic compounds in the soil itself using a mass spectrometer, even down to extremely low levels (a few parts per billion), and the complete lack of liquid water.

Although the Viking experiments were widely interpreted as indicating the absence of life on Mars, we may not have been looking in the right place. Probably the best place to look would be in areas that were likely to have seen liquid water at some point, perhaps even today (e.g., below the surface).

3.2.3 Evidence for Liquid Water Most of the water on Mars now is frozen in its polar ice caps or in a layer of permafrost below the surface; there is very little water vapor in the atmosphere and no liquid water at the surface (the air pressure is too low). Yet there is strong geological evidence that Mars was once a wetter place than it is today:

1. Impact craters on Mars that are older than about 3.5 Gyr show substantial evidence

for erosion, and no ancient craters smaller than about 15 km have been found. This is consistent with erosion by surface water.

3.2. MARS 27

2. Surface features resembling dry river valleys are found on the older parts of the Martian

surface. These heavily branched patterns can only be shaped by running water, not seismic activity or lava flows. Recent images from the Mars Global Surveyor show even clearer evidence that these valleys were carved over long periods of time.

3. Broad channels carved in younger parts of the surface are thought to be due to catastrophic floods, based on their resemblance to the Scablands of eastern Washington, which were produced by a gigantic flood. They may have resulted from the collapse of natural dams or the eruption of underground aquifers under high pressure.

Do these features represent transient flows across Mars' surface, or could liquid water have flowed across the surface for substantial amounts of time? The river valleys suggest that water flowed for thousands or even millions of years (think of the Grand Canyon, carved by the Colorado River) but we don't know for sure yet. One problem with a sustained presence of liquid H

2

O is the huge greenhouse effect needed to keep the surface

warm at the distance of Mars' orbit, especially considering the faint young Sun problem (x3.1). This could probably not be provided by CO

2

, which would form clouds at such

high concentrations and be ineffective for greenhouse warming. Unless other greenhouse gases existed, it seems more likely that early Mars was similar to Antarctica today, where rivers flow briefly in the summer into ice-covered lakes that are sufficiently insulated to stay partially melted year-round.

Could there still be enough internal heat to allow subsurface water today? Evidence that volcanism occurred in quite recent epochs is provided by so-called SNC meteorites which have been found on Earth. These meteorites are thought to be from Mars, based on their oxygen isotope ratios and the composition of gas trapped inside them, both of which resemble that of Mars' atmosphere. Such meteorites may have been debris launched from Mars' surface by an impacting asteroid--debris that eventually slammed into Earth. Studies show that some of these crystallized on Mars from molten rock as recently as 180 Myr ago. Thus it is quite possible that Mars has maintained some of its original store of heat, in which case there may still exist subsurface habitats with liquid water. Images from the Mars Global Surveyor should tell us where the likely sites for such habitats are.

What caused the Martian surface to lose its liquid water? Probably the loss of greenhouse warming, coupled with the loss of atmospheric pressure needed to keep water in liquid form. The main culprit was the small size of Mars, which reduces its ability to hold onto an atmosphere and increases the rate at which it loses its internal heat. Although Mars' gravity is sufficient to hold onto an atmosphere in quiescent periods, the atmosphere would have been greatly depleted after major impacts, and as volcanism died down (due to internal cooling) there were less and less new greenhouse gases to replace those that were lost. If there was initially liquid water on the surface, it would have contributed to the loss of CO

2

by weathering, hastening its own demise.

28 CHAPTER 3. LIFE ELSEWHERE IN OUR SOLAR SYSTEM? 3.2.4 The Martian Meteorite Controversy Although Mars may not be hospitable to life now, it is certainly plausible that it could have harbored life early in its history. A thicker atmosphere, liquid water on the surface, and active volcanism provided a setting that was probably similar to that in which life arose on Earth. As the presence of Martian rocks on Earth indicates, it is even possible that life could have begun on Mars and then been transferred to Earth!

One such Martian (SNC) meteorite, called ALH 84001 because it was found in the Allan Hills region of Antarctica in 1984, captured worldwide attention in 1996 when a team led by NASA geologist David McKay announced that it contained evidence for past biological activity. Their case was based on four principal findings:

1. Evidence for water: Mineral globules containing carbonates (such as CaCO

3

) were

found which are most easily produced in the presence of liquid water. The estimates for the age of these minerals ranges from 1.3-3.6 Gyr; the older estimates are consistent with the photographic evidence of when liquid water was present on the surface (x3.2.3). In contrast, the meteorite was ejected into space only 16 Myr ago, based on the exposure of cosmic rays it received, so the carbonates definitely formed on Mars. One point of controversy is the temperature at which the minerals formed; different groups have come up with numbers from !150

ffi

C up to nearly 700

ffi

C. Only the former would allow

life as we know it.

2. Mineral grains: The carbonate globules also show a layered structure that is associated (on Earth) with the deposition of minerals by bacteria. In particular, there are iron-rich layers that contain small grains of magnetite (an iron oxide) as well as pyrrhotite and greigite (iron sulfides). The presence of all these together requires unusual geological conditions, the authors argued, or biological activity. They noted that magnetite is produced by bacteria as a way of orienting themselves to the Earth's magnetic field (sort of a built-in compass), and the size and shape of the magnetite grains in ALH 84001 are similar to those produced by bacteria on Earth. On the other hand, these minerals are not produced only by life, so their association with life is still tentative.

3. Organic molecules: Certain organic molecules called polycyclic aromatic hydrocarbons (PAHs), were found in the meteorite, although their association with the carbonate globules remains uncertain. The PAHs were found in the inner parts of the rock and not near the surface, suggesting that they did not enter the rock while it was on Earth. Such molecules can be produced by the decay of biological material or the combustion of fossil fuels. However, they are also seen in extraterrestrial environments such as non-Martian meteorites and interstellar gas clouds. Although it is unlikely that these PAHs formed in asteroids and survived the impact onto Mars, we have no way of knowing whether they formed from biological or non-biological processes.

Furthermore, the possibility of terrestrial contamination must be taken seriously. In two reports published in 1998,

1

Jeffrey Bada and colleagues at UCSD found amino

1

J. L. Bada et al. and A. J. T. Jull et al., Science 279, 362 and 366 (16 Jan 1998).

3.3. VENUS 29

acids in ALH 84001 that were predominantly terrestrial contamination, while a group led by Timothy Jull at the University of Arizona showed that there is a significant amount of short-lived

14

C isotope in the meteorite, indicating that most of the organic

molecules are of recent terrestrial origin. On the other hand, the PAHs are such a small fraction of the total organic material that they could still be Martian.

4. Bacteria-shaped objects (BSOs): The most controversial evidence came from electron microscopy of the carbonate globules, which revealed tubular structures resembling bacteria. These "microfossils" are extremely small: only about 100 nm (10

\Gamma 7

m) long and 10-20 nm wide. This is much less volume than even the smallest known bacteria,

2

and is comparable in size to a virus or ribosome. Indeed, many experts

believe that a cell must be at least 200 nm in diameter to be capable of performing the metabolism of life on its own. Although we cannot rule out smaller and simpler lifeforms being possible, one wonders whether Martian bacteria 100 times smaller than terrestrial ones would produce the same sized magnetite grains. The microfossil case is further weakened by McKay's recent admission that the "phalanxes" of BSOs were artifacts of the sample preparation, although he still contends that some of the other features are real.

In short, the past three years have provided no new evidence in support of the Martian microbes, but no conclusive evidence that disproves it either. The likely discovery of organic molecules (PAH's) on Mars is an important one, and dramatically improves the chances that life could have existed on the planet. However, only future missions to Mars will be able to provide clear evidence that life actually did exist, and it may take a long time to rule out the possibility of life, especially below the surface. If life did emerge on Mars independently of the Earth, it would be a strong argument for the widespread existence of life in the universe.

3.3 Venus Earth's sister planet in terms of size, Venus briefly attracted attention as a possible abode for life once it became clear that the Martian canals were an artifact of over-imaginative observers. The presence of thick clouds on Venus suggested a moist planet, perhaps covered in swamps. But later evidence showed that the Venus' atmosphere was mostly CO

2

, which

would maintain an excruciatingly hot surface temperature via the greenhouse effect. In contrast, very little water now exists on Venus: if condensed on the surface, all of the water vapor would form an "ocean" 2 cm deep (on Earth the equivalent depth would be 2 km). Direct measurement of a surface temperature of about 600 K (later found to be closer to 730 K, or 455

ffi

C) in 1956 put to rest any question of present-day life.

The high temperatures on Venus are surprising given that the planet actually absorbs less radiation from the Sun than the Earth does: its thick sulfuric acid clouds reflect 77% of the sunlight that reaches them, whereas the Earth only reflects about 30%. When you do the math, you find that Venus' surface actually gets less sunlight than the Earth's, even

2

Ignoring claims of 20-nm-long nanobacteria, whose biological nature is still extremely controversial.

30 CHAPTER 3. LIFE ELSEWHERE IN OUR SOLAR SYSTEM? though Venus is 30% closer. As a result, Venus' surface would be a frigid \Gamma 45

ffi

C if it were

not for the excruciating 500

ffi

C of greenhouse warming provided by the atmosphere.

How did the greenhouse effect become so strong on Venus? Although the Venusian atmosphere is 100 times thicker than the Earth's, its CO

2

content is actually not much more

than the total CO

2

content of the Earth, when one includes the CO

2

which has been trapped

in the Earth's crust by weathering or is dissolved in the oceans. Thus the loss of water appears to have been crucial in sealing Venus' fate by preventing CO

2

from being locked in

the crust. There is in fact evidence that Venus once had much more water than it does now: the large amount of deuterium (heavy hydrogen,

2

H) in the Venusian atmosphere is consistent

with the preferential escape of light hydrogen (

1

H) into space, following dissociation of water

vapor by UV radiation.

In the conventional scenario for what happened, known as the runaway greenhouse effect, early Venus was too hot to allow water vapor to condense into liquid form. This resulted in a dense steam atmosphere and strong greenhouse effect, since water vapor is an effective greenhouse gas. New water provided by volcanic outgassing or cometary impacts would also be vaporized, increasing the greenhouse effect. The lack of liquid water prevented the weathering reactions that remove CO

2

from the atmosphere, resulting in an buildup of

CO

2

from outgassing and even higher temperatures. Eventually the H

2

O was dissociated

and only the CO

2

(along with some other gases) remained, still adequate to produce a strong

greenhouse effect.

A slightly different model, called the moist greenhouse, is now favored by many scientists, as it doesn't require as strong a solar flux at the beginning of Venus' history. In this case, Venus starts out only slightly warmer than Earth--warm enough for water vapor to reach greater heights in the atmosphere, where it is dissociated by UV radiation. The result is ultimately the same: water is irreversibly lost and CO

2

builds up in the atmosphere,

leading to a strong greenhouse effect. But because the initial temperatures are not so extreme, this model allows for the possibility of stable oceans on Venus that only gradually boiled away. It is a depressing prospect that perhaps Venus lost its water gradually, with enough time for doomed lifeforms to have arisen in its oceans. It is even more depressing to think that our planet will probably end up like Venus someday, as a result of either our own negligence or simply the increasing luminosity of the Sun.

3.4 Europa One of the four large Galilean satellites of Jupiter, Europa has attracted attention ever since it was found to be covered by a smooth layer of water ice--with possibly a liquid ocean underneath. (Remember that water's unique property of being denser than its solid form allows ice to float over water.) But what could keep the water from freezing at Europa's huge distance from the Sun? Probably a combination of two factors: tidal stretching, which gives the surface its fractured appearance, and radioactive heating.

Tidal heating is due to Europa's eccentric orbit around Jupiter. Any satellite orbiting close to a planet has its shape distorted because the planet's gravity pulls stronger on its

3.4. EUROPA 31 near side than its far side. This creates "tidal bulges" on the moon which point directly towards and away from the planet. Since Europa's orbit is eccentric (non-circular), the size of its tidal bulges varies as it goes around Jupiter, and this continual flexing cracks the icy crust and heats the interior. The heat energy ultimately comes at the expense of the moon's orbital energy, and over time the orbit should circularize. But Europa is prevented from settling into a circular orbit because of gravitational interactions with two of the other Galilean satellites, Io and Ganymede.

Radioactive decay of heavy isotopes contained in Europa's rock also produces heat, which can be more or less trapped in the moon's interior, depending on how efficiently heat can be transported outwards. Europa's density is too large (3 g/cm

3

) to be composed entirely

of ice, whose density is 1 g/cm

3

. At least 80% of its mass is therefore made up of rock or a

rock-ice mixture.

Computer models indicate that these energy sources would be sufficient to maintain liquid water on Europa, but not to create it if the moon was ever completely frozen. How deep you'd have to go to find water depends on how well the ice conducts heat from the deeper layers to the surface, where it is radiated away. Ordinary ice conducts heat fairly well, and the resulting ice crust would be at least 10 km thick. But if the ice is a poor conductor (in sort of a frost-like state) then it could "bottle up" the outgoing internal heat and cause the ice to melt at depths of only a few hundred meters. Below that, there could be a liquid ocean roughly 100 km deep!

Is there actually evidence for a liquid ocean on Europa? Observations by the Voyager and Galileo space probes have turned up tantalizing clues, although no conclusive proof.

1. Europa's surface is extremely smooth, with very few impact craters, indicating that it

must be young. Two possibilities come to mind: either liquid water erupted from the interior and froze, resurfacing the moon, or soft ice flows have slowly performed the resurfacing.

2. Some parts of the surface show blocks of ice that are separated but appear to fit

together like a jigsaw puzzle, indicating that they have drifted apart like icebergs, and in some cases been rotated. If this interpretation is correct, the icebergs would have to be lubricated from below by water or slushy ice.

3. Europa has a weak magnetic field indicating that there are electric charges moving

underneath its surface. One simple explanation for this would be a "salty" ocean-- charged ions dissolved in liquid water.

None of these observations, however, require large oceans of water today: the first two could be explained by slushy ice or brief periods of melting in the past, while the last could be the result of some other conducting material. A couple of more direct methods for detecting an ocean have been suggested: using radar to probe beneath the surface, or trying to measure accurately the height of the tidal bulges, which would be larger if there was an ocean. But ultimately, we may simply have to land a probe and start drilling to find out!

32 CHAPTER 3. LIFE ELSEWHERE IN OUR SOLAR SYSTEM?

If there is liquid water, could life have developed on Europa? Although we don't know what the inventory of carbon, nitrogen, and other biogenic elements is, if we assume Europa's composition is more or less comparable to what we find in Jupiter's atmosphere and in meteorites from the asteroid belt, these elements should exist in fairly large amounts. A more difficult problem is finding a source of energy for life: too little sunlight would filter down to any subsurface ocean. On Earth we have found ecosystems thriving in complete darkness near hydrothermal vents, where the root source of energy is the Earth's internal heat. Could the same situation exist on Europa?

Remember that life requires a flow of energy from hot regions to cold regions in order to survive. The hydrothermal vents on Earth are much hotter than the surrounding water, so this requirement is satisfied. But on Europa, the existence of oceans of roughly constant temperature would not be enough to support life. Localized "hotspots" inside Europa, such as undersea volcanoes, might be possible habitats, but whether there is enough internal heating to produce such hotspots is uncertain.

3.5 Titan Saturn's largest moon Titan is remarkable for its thick atmosphere, which produces a pressure 1.5 times that of the Earth's. Like ours, Titan's atmosphere is mostly nitrogen (N

2

); on

the other hand, there are many reduced gases as well, such as hydrogen (H

2

), methane

(CH

4

), and ethane (C

2

H

6

). In this regard, Titan's atmosphere may resemble the conditions

under which life arose on Earth, since making organic molecules seems to require somewhat reducing conditions (x4.2). The one ingredient that is missing is liquid water: with a surface temperature of 94 K (\Gamma 179

ffi

C), all of Titan's water is frozen. Nor is there any substantial

tidal heating as there is for Europa.

Despite the lack of water, it seems that there is quite a lot of interesting organic chemistry happening on Titan. One obvious mystery is the thick layer of haze that obscures the surface from view--since N

2

is transparent, it must be made of something else. The most likely

candidate is some kind of organic "smog" that results from reactions involving sunlight, charged particles in Saturn's magnetosphere, CH

4

, and N

2

. A few years before his death,

Carl Sagan demonstrated such reactions in laboratory experiments, producing a dark organic solid he called "Titan tholin," after the Greek word for "muddy."

3

Oceans on Titan? The presence of methane in the atmosphere is actually somewhat surprising, since methane is easily destroyed by solar UV radiation even as far away as Saturn: the present quantity would be completely lost in 1-10 Myr, a short time by geological standards. There must be a mechanism for resupplying it, such as an ocean of liquid methane at the surface. For various reasons a pure CH

4

ocean is ruled out, but an ocean consisting of

mostly ethane with about 20% CH

4

could be in equilibrium with the atmosphere we observe.

Ethane is actually a likely product of photochemical reactions involving methane, plus we expect it to form clouds and "rain out" from the atmosphere, so this idea is not completely

3

He spoke about this work at the dedication speech for Pimentel Hall, which I attended in 1994.

3.6. OTHER BODIES OF THE SOLAR SYSTEM 33 ad hoc. Radar measurements, however, indicate that Titan's surface is much more reflective at radio wavelengths than would be expected if it were completely covered by an ocean. There are also recognizable surface features in images made with infrared light. Perhaps the methane-ethane liquid occurs in lakes or impact basins, or percolates through cracks in the (water) ice to and from underground reservoirs.

Life on Titan? With possible oceans of hydrocarbons and active organic chemistry, Titan is certainly an interesting target for astrobiology. The lack of water is still a problem: although it is by no means certain that water is necessary for life (Chapter 2), ethane and methane are fairly non-reactive and aren't likely to be the basis for an alien biochemistry. Nonetheless, there will be occasional melting of ice near the surface due to impacts, and early in its history Titan may have retained enough internal heat to allow liquid water to exist in some regions. In any case, the possible insights into prebiotic chemistry that Titan can provide, as the only object in the Solar System with a reducing atmosphere not dominated by H

2

, has helped to motivate the Cassini mission to Saturn, which will send a probe into

Titan's atmosphere in 2004.

3.6 Other Bodies of the Solar System In the previous sections I have focused on objects in the Solar System that are most likely to have experienced conditions suitable for life as we know it, either now or at some time in the past. The remaining objects appear far less promising. Mercury and the Moon are clearly dry, barren worlds. Io, Jupiter's volcanic moon, has been the subject of some exobiological speculation, but despite the abundance of available energy (it receives much stronger tidal heating than Europa), the lack of water and the extreme conditions on Io make this seem a bit of a stretch. The other moons of the giant planets are probably too cold for liquid water to exist, and lack atmospheres that would allow other kinds of liquid oceans to exist. However, new discoveries in the coming years may revive interest in some of these objects.

It has been suggested that life could exist in the atmospheres of the giant planets themselves, particularly Jupiter, whose atmosphere has a large source of energy for chemical reactions: the heat released by Jupiter's internal contraction. In fact, the most spectacular lightning storms we know of occur there. Jupiter's atmosphere is also rich in hydrogen (H

2

),

methane (CH

4

), and ammonia (NH

3

), which laboratory simulations show can be used to

make amino acids (x4.2). However, there are several serious objections to this idea. First, there are no oceans or solid surfaces for concentrating molecules, making it hard to imagine how the series of reactions needed to build up complex molecules could occur. Second, any Jovian lifeforms would have to immediately evolve mechanisms to keep from being dragged down to the lower parts of the atmosphere, where they would be cooked. Finally, the Galileo atmospheric probe found virtually no trace of water vapor or complex organic molecules in Jupiter's atmosphere.

34 CHAPTER 3. LIFE ELSEWHERE IN OUR SOLAR SYSTEM? 3.7 Conclusions The fundamental requirements for life as we know it are organic molecules, liquid water, and energy to drive endergonic reactions. A habitat that doesn't kill you right away is also desirable. On Venus, Mars, Titan, and Europa, these conditions could all have existed at some point in the past, although only on Mars and Europa is there any hope for finding life now (perhaps in underground water). However, the conclusion that the ingredients for life automatically produce life is incredibly na"ive: as we'll discuss in the next chapter, we still don't understand how life began here on Earth. There may have been conditions unique to the Earth that were crucial for the origin of life.

Chapter 4

Origin & Evolution of Life on Earth The origin of life has remained such a perplexing problem in part because of the difficulty of separating what is contingent (determined by chance) from what is necessary in living systems. The particular biochemistry of life on Earth seems so well-suited to its purpose that some scientists believe it must be shared, with minor variations, by all living things in the Universe. But until we discover extraterrestrial life that clearly has a separate origin from life on Earth, we will have little indication of how random--or predestined--our biochemistry is. Since the early days of the Space Age, this has been the primary motivation for many exobiologists: extraterrestrial life, they believe, holds the key to uncovering fundamental "laws" of biology that will supersede our knowledge of terrestrial biology and tell us in how many ways molecules can organize themselves into living systems.

In the meantime, though, there are other ways to investigate the origin of life. One approach (the synthetic approach) is to try to simulate as realistically as possible the conditions of the early Earth in the laboratory, in the hope that some kind of self-replicating molecule can be created. Success may be hard to come by, considering the fact that nature had millions of years and all the world's oceans to work on this problem, and most government funding agencies are not going to be that patient. On the other hand, if life is a robust phenomenon that pops up quickly when the conditions are right, it is not too far-fetched to believe that we may someday succeed in producing it.

The other approach (the historical approach) is to try to work our way back towards life's origins by pushing the boundaries of what we know about the earliest forms of life, both from the fossil record and the genetic history of contemporary organisms. Recent interest has focused on the revolutionary technique of molecular phylogeny, which allows biologists to determine the ancestry of organisms--the so-called "tree of life"--from their DNA or RNA sequences. It has been suggested that certain bacteria found in hot underwater environments are the most primitive in existence, igniting speculation that these environments served as the cradle for life.

35

36 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH 4.1 Historical Background Over the course of history, there are four basic hypotheses that have been proposed for the origin of life.

1. Life was the result of a supernatural or freak event, beyond the explanatory powers of

science.

2. Life arrived on Earth from outer space, soon after the Earth's formation. 3. Life arises spontaneously from nonliving matter in a short amount of time, even today. 4. Life arose by a process of chemical evolution over a significant period of time.

The first idea is advocated by creationists, and even some scientists who consider the origin of life to be a singular historical event--a lucky accident. It is not inconsistent with what we know, but provides no basis for further discussion or investigation. In particular, scientists generally reject any hypothesis that involves supernatural forces, which by definition are beyond scientific verification.

1

Recently there has been a movement to revive the

creationist hypothesis, based on the assertion that is life is too complex to have emerged by a process of natural selection (see e.g. Darwin's Black Box by Michael Behe). We will address this further when we discuss evolution (x4.5) and again in the final chapter.

The second idea has been associated with the eternal or "Steady-State" universe advocated by the cosmologist Fred Hoyle. Hoyle and his colleague Chandra Wickramsinghe popularized an earlier suggestion by the chemist Svante Arrhenius called panspermia: life floats in "spores" through interstellar space, laying dormant until it comes upon a suitable planet. If the Universe has always existed, perhaps life has as well, in which case it had no origin and simply travels from star to star. Since then, the Steady-State Universe has been largely discredited by modern cosmological data: we now believe the Universe had a beginning, the Big Bang. Still, there are modern versions of the panspermia hypothesis which are under discussion (x5.3).

The third idea was popular in ancient times: as late as the 17th century, it was perfectly natural to assume that insects, worms, and other household pests arose spontaneously. The Flemish alchemist Jan Baptiste van Helmont even circulated a recipe for making mice from dirty underwear. Perhaps the first person to experimentally test the idea of spontaneous generation was the Florentine physician and poet Francesco Redi (1626-1697). He found that meats placed in sealed jars did not develop maggots, and concluded that life sprang only from "the seeds of the plants or animals themselves." On the other hand, later researchers pointed out that bacteria still seemed to appear spontaneously: even the sealed meat rotted. Only after the experiments of Louis Pasteur (1822-1895) was it clearly established that this was not the case. Pasteur found that a nutrient solution, boiled in a flask with a curved neck so that dust and other particles could not reach the liquid, stayed sterile for months.

1

This is admittedly a philosophical prejudice, but one that has served science well over time.

4.2. THE MILLER-UREY EXPERIMENTS 37

Once it was accepted that the living and non-living worlds were distinct, the question of how living matter first arose became acute. Charles Darwin, in a speculative mood, presciently suggested that life began in a "warm little pond, with all sorts of ammonia and phosphoric salts, light, heat, electricity, etc." This is the essence of the fourth idea, which guides current scientific thinking on the subject: that life arose by a process of chemical evolution, starting with interactions of simple molecules and culminating in the formation of a self-reproducing system. For the remainder of this chapter we will focus on this idea, first expounded in detail by A. I. Oparin and J. B. S. Haldane in the 1920's.

4.2 The Miller-Urey Experiments The hypothesis that the ingredients of life could have arisen spontaneously on Earth was put to the test by a graduate student, Stanley Miller, working with Harold Urey at the University of Chicago in 1953. A highly reducing atmosphere (rich in compounds containing hydrogen) was simulated by a mixture of CH

4

, NH

3

, H

2

, and H

2

O, connected in a sealed system to

a pool of water representing the early ocean. A heater and condenser allowed water vapor to circulate through the system, mimicking the water cycle in our atmosphere. The use of a reducing atmosphere was reasonable assuming it was gravitationally captured from gases in the solar nebula. However, most geologists now believe that the early atmosphere was largely a result of outgassing from the interior and the vaporization of impacting comets, and thus may have been primarily CO

2

and N

2

(x4.3.3).

What sources of energy were available to drive endergonic reactions? The largest source of energy was probably solar UV radiation, but organic molecules synthesized in the upper atmosphere might be destroyed by the same UV radiation before it could reach the Earth's surface. This led Miller to use electrical discharges to simulate lightning. The advantage of lightning is that it occurs closer to the surface, so newly formed molecules can reach the oceans before they are destroyed. In addition, electrical discharges are more efficient at producing hydrogen cyanide (HCN), a building block of amino acids, than UV radiation is.

The results of the experiment were stunning: after a few days, amino acids and other organic compounds--including formaldehyde (CH

2

O), a building block of sugars--were produced in significant quantities. Subsequent experiments using long-wavelength UV light, performed by Carl Sagan and colleagues, also yielded amino acids. However, virtually no organic molecules could be produced in an oxidizing atmosphere: if this was the path to the origin of life, it must have occurred under somewhat reducing conditions. This is simply because organic molecules contain lots of hydrogen, and it's easier to get hydrogen from hydrogen-rich molecules.

Miller later showed that one process by which the amino acids could form was the so-called Strecker synthesis. In an example of this process, HCN and CH

2

O--two of the

"intermediate products" produced by sending an electric discharge through the gas mixture-- are combined in the presence of NH

3

and H

2

O to make the simplest amino acid, glycine:

CH

2

O + HCN + NH

3

+ H

2

O \Gamma ! CH

2

(NH

2

)CN + 2 H

2

O \Gamma ! CH

2

(NH

2

)COOH + NH

3

38 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH Interestingly, the proportions of other amino acids produced relative to glycine were the same in Miller's experiment as had been found for the Murchison meteorite, indicating that this process probably occurred in the parent bodies

2

of meteorites as well. Also, both lefthanded and right-handed amino acids appeared in equal numbers, unlike biological amino acids which are all left-handed. This confirmed that the amino acids were not a result of contamination in the laboratory, but also raised the question of how handedness arose in the first place.

4.3 Setting the Stage The relevance of the Miller-Urey type of experiments depends on whether they accurately reflect the conditions on the prebiotic Earth, before life arose. Unfortunately, due to the action of plate tectonics, the constant resurfacing of the Earth due to convection (overturning motions) in the Earth's mantle, the geological record of these conditions appears to be lost to us forever. The oldest rocks on the Earth's surface are only about 3.8-4.0 Gyr old, much younger than the oldest Moon rocks (4.5 Gyr). As a result, we must rely for guidance on theories of planet formation, as well as indirect evidence from rocks that we do have.

4.3.1 Birth of the Solar System The standard theory for the Solar System's formation involves the collapse of a rotating gas cloud (the solar nebula) into a disk. This collapse is governed by the force of gravity, which also causes most of the material in the disk to be drawn (accreted) to the center. As a result of this accretion, 99.9% of the material in the Solar System now resides in the Sun. But since the disk began with a certain amount of angular momentum (rotational inertia) that was not easily lost, some of the material stayed in orbit around the Sun: this material was collected together by gravity to form the planets and other solid debris of the Solar System. Observations of disks of gas and dust around other young stars suggests that the formation of planetary systems is a quite general phenomenon; further evidence of this has come from the recent discovery of extrasolar planets (Chapter 6).

The formation of the planets was a long and agonizing process: solid grains of dust and ice slowly accumulated into rocks and snowballs, which gently collided and stuck together to form ever larger bodies. How quickly the planet building blocks (planetesimals) grew depended on how far away they were from the Sun. In the inner Solar System, where orbital speeds were higher, frictional effects heated the disk more. This heat, along with the heat being produced by the proto-Sun, made it much too hot for water ice to form: only heavy elements like iron and silicon could solidify. On the other hand, the outer planets and their moons could form from a combination of rock and ice. The availability of water ice allowed the "embryos" of the outer planets to grow rather quickly, soon acquiring enough gravity to capture some of the original hydrogen and helium of the solar nebula and become gas

2

Meteorites are believed to be fragments of asteroids that were broken apart by collisions; the original

asteroids they came from are called parent bodies.

4.3. SETTING THE STAGE 39 giants. The inner "terrestrial" planets, lacking ice, never got large enough to capture and hold on to the light nebular gases, which were blown outward by the solar wind; their present atmospheres are secondary, meaning that they formed after the gas in the solar nebula had been dispersed.

How long did it take to form the planets? Computer simulations suggest the small, kilometer-sized planetesimals formed quickly, in about 10,000 years, but once they were in separate orbits around the Sun it took a long time (up to 100 Myr) for them to merge to form the terrestrial planets. On the other hand, Jupiter and Saturn must have formed much faster, since they were able to capture significant amounts of gas from the solar nebula, which, as observations of gas disks around young stars indicate, only lasted about 10 Myr before being blown away. With more material available to start with, it is conceivable that the giant planets experienced a kind of "runaway growth" that the terrestrial planets did not. As a result, the gravitational influence of rapidly growing Jupiter probably disturbed the orbits of planetesimals in its vicinity, preventing the formation of a planet where the asteroid belt lies today and stunting the growth of Mars.

Radioactive dating of meteorites shows that their parent bodies solidified about 4.56 Gyr ago, so presumably the planets were well along in their formation at this time. Once the giant planets had grown to an appreciable size, they began to deflect the orbits of nearby planetesimals into collision courses with the other planets. The resulting collisions were far more powerful than those that had assembled the planets: some were strong enough to blast material off the planets, or to force dramatic changes in the tilts of their rotation axes. The leading theory for the Moon's formation involves the impact of a Mars-sized body onto the young Earth, throwing a chunk of Earth's mantle into orbit where it coalesced to form the Moon. Grim reminders of this era of heavy bombardment--the impact craters--scour the surfaces of the Moon, Mercury, and parts of Mars, as well as the satellites of the giant planets. Attempts to date the craters, while somewhat uncertain, suggest that planetary bombardment continued up to about 4.0 Gyr ago--although the frequency of impacts was decreasing rapidly throughout this period as material got swept up by planets or ejected from the Solar System.

The leftover planetesimals of the Solar System, referred to as comets when they are made mostly of ice and asteroids when they are made mostly of rock, probably retain the best record of its early history. This is by virtue of their small size: they assembled quickly, and most did not produce enough internal heat to substantially transform their interiors. While most astronomers accept the general scenario I have described above, much will be learned about the Solar System's formation once we are able to retrieve samples from comets and asteroids to study in detail. We will discuss the role of comets and asteroids in the evolution of life in Chapter 5, and consider the question of whether our planetary system is typical in Chapter 6.

4.3.2 The Hadean Earth Geologists refer to the period that stretches from the Earth's formation to the formation of the oldest surviving rocks, 4.6-3.9 Gyr ago, as the Hadean eon. The ancient Greek

40 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH term for hell (Hades) is probably appropriate: the Earth was red hot and subject to frequent catastrophic impacts. The large number of craters on the Moon dating from this time implies that the Earth suffered even greater bombardment, given its larger size and stronger gravity. Unlike the Moon, however, the Earth has erased all of its ancient craters by the incessant forces of plate tectonics and erosion by wind and water.

The enormous heat energy deposited by impacting bodies melted the Earth's interior and led to its reorganization, a process known as differentiation. Most of the iron sank to form the core, with silicon and magnesium rocks forming an overlying mantle and a thin, even lighter crust riding on top. Volatiles, the most easily vaporized materials such as CO

2

and H

2

O, bubbled up through the molten Earth and were outgassed by volcanoes to form

the early atmosphere--although it has been argued that this was a minor contribution to the final atmosphere (x4.3.3). The settling of heavy materials to form the core produced even more internal heat--like the impacts of planetesimals, it transformed gravitational energy into kinetic energy and then into heat energy. This heat kept the interior molten and allowed further settling to occur. Today impacts are rare and core formation is long since over, but the Earth retains a source of internal heat because of radioactive decay and the solidification of the core, processes that also release heat. The continued geological activity of the Earth has been crucial to the stabilization of the greenhouse effect, as discussed in x3.1.

4.3.3 The Early Atmosphere One of the major unknowns about the environment in which life arose is the composition of the atmosphere. The most basic question is whether the atmosphere was reducing or oxidizing. Recall from Chapter 1 that a reducing agent is one that tends to give up its electrons to other substances, and often contains hydrogen, while an oxidizing agent is one which tends to take electrons, and often contains oxygen. The present atmosphere of Earth is highly oxidizing, with O

2

the major constituent after nitrogen, but this is largely the result

of life (x4.6.2), and is therefore not a good guide for what the prebiotic atmosphere would have been like.

The absence of free oxygen at this time had another important consequence: no ozone shield existed. Presently most of the solar UV is blocked by the ozone (O

3

) layer, but O

3

comes from O

2

, and there was only a small amount of O

2

available from the photodissociation

of H

2

O. This may have been both a blessing and a curse: with more UV radiation reaching

the surface, there was an additional energy source for prebiotic chemistry, but also the potential to destroy organic molecules once formed (x4.2).

In the 1950's, Harold Urey proposed that the early atmosphere was a reducing one similar to Jupiter's: rich in methane (CH

4

), ammonia (NH

3

), and hydrogen (H

2

). Such an

atmosphere would have favored the formation of complex organic molecules, as had been demonstrated by Urey and Miller (x4.2). However, methane and ammonia are susceptible to destruction by solar UV radiation. Furthermore, we expect that the intense solar wind and bombardment associated with planetary formation drove off most of the hydrogen-rich gases that the Earth accreted from the solar nebula. In particular, the collision that formed the Moon would have completely vaporized the atmosphere. Direct evidence for loss of the

4.3. SETTING THE STAGE 41 original (primary) atmosphere is provided by the fact that the proportions of noble gases in the present atmosphere differs markedly from the proportions found in the Sun.

There is some debate about where the secondary atmosphere came from. Many geophysicists believe that it was mainly outgassed from the interior--derived from volatiles which were originally trapped in the planetesimals which formed the Earth. Many astronomers argue instead that the atmosphere was almost entirely brought by comets vaporizing upon impact in the late stages of the Earth's formation. The cometary hypothesis has several advantages: comets are much richer in volatile elements (C, H, N, O, P, S) than inner-Solar-System planetesimals were, the cratering record suggests high bombardment rates, and the impact which formed the Moon would probably have blown away the Earth's atmosphere and most of the gases trapped in its interior. On the other hand, this last example illustrates the fact that large impacts can remove more volatiles than they add! What is generally agreed by both sides is that the secondary atmosphere was not nearly as reducing as the primary one, being composed mostly of CO

2

, CO, and N

2

.

Still, there may have been processes at work that could produce reducing gases, at least locally. Even if the volatiles were mainly brought by comets rather than outgassed, some of them would have been dragged into the mantle by overturning motions (convection) or injected into the mantle by large impacts, only to be released later by volcanism. While passing through the mantle, the volatiles could be reduced by reacting with minerals such as ferrous iron (FeO), a common form of iron at that time. For instance, water reacting with FeO can produce ferric iron (Fe

2

O

3

) and H

2

. Thus, mantle rocks may have served

as reducing agents, and the volcanoes and deep-sea vents where outgassing occurred would have been a source of highly reduced (hydrogen-rich) gases. In this model, gases released by volcanoes today are mostly oxidized because the reducing power of the mantle has been depleted: the FeO has been oxidized and the H

2

lost to space. A similar evolution from

a more reduced to a more oxidized state occurred on Venus and Mars as well, since they too have been continually losing hydrogen. If life was extinguished today, it would probably never start again, a sobering thought.

4.3.4 Origin of Organic Compounds It's very difficult to make hydrogen-rich compounds like glucose from oxidized gases like CO and CO

2

. Just ask a plant, which gets its hydrogen by breaking apart water with sunlight.

So if the early atmosphere was somewhat oxidized, where did the organic compounds needed to produce life come from? There are essentially three possibilities:

1. They were delivered to Earth by comets, meteorites and meteoritic dust. This is plausible given the fact that amino acids and other complex organic molecules have been found in meteorites and interstellar clouds. In particular, their presence in meteorites indicates that they can sometimes survive the impact onto Earth. We will discuss this possibility in greater detail in Chapter 5.

2. They formed by the reduction of CO

2

in a non-biological process which is uncommon

today. It has been suggested that hydrogen liberated by the formation of pyrite (FeS

2

)

42 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH

from FeS and H

2

S, or by the photo-oxidation of Fe

2+

dissolved in water, could be used

to reduce CO

2

, but this has yet to be demonstrated experimentally. FeS and Fe

2+

are

uncommon today because they have long since been oxidized by the copious O

2

in the

atmosphere.

3. They formed in a Miller-Urey type of synthesis from reduced gases including CH

4

.

Early volcanism would have provided such gases, at least in localized regions (bear in mind that volcanism was probably much more widespread on the hot young Earth), and researchers have reported modest success at producing amino acids in simulated hydrothermal environments.

All in all, it seems that organic molecules should have been available on the early Earth, although in what concentrations is uncertain. The big stumbling blocks come later, as will be discussed in x4.4.

4.4 The Origin of Life: Current Theories Life must have started out simple. Recall from x1.3 that the DNA of a bacterium encodes about 6 million bits of information. The creation of that much information by chance is virtually impossible, even given all the oceans of Earth (or all the worlds in the known universe). Fred Hoyle has likened the problem to a whirlwind assembling a 747 from a junkyard: it simply couldn't happen in the real world. An estimate made by Edward Argyle of the maximum number of amino acid sequences which could be generated at random in the Earth's oceans, given 500 Myr, is roughly 10

58

, corresponding to just 194 bits of information.

Expanding our sample to a billion planets only increases this to 224 bits.

Of course, once a system becomes capable of evolution, it can generate new information very rapidly, as it no longer has to depend on blind luck. But a fairly accurate replication system must exist before evolution is possible, since evolution only works on inherited features. So assuming the first reproducing molecule had to emerge by chance, it could have contained at most 194 bits of information--in other words, less than 100 nucleotides or 50 amino acids. Here we ignore the possibility of extremely improbable events occurring ("miracles") or chemical reactions strongly favoring the formation of large self-replicating molecules ("loading the dice"). Still, there are an enormous number of possible sequences using just 20 amino acids (20

20

ss 10

26

)--we've by no means exhausted our options.

Thus the first lifeforms were information-bearing molecules, capable of reproducing individually or as part of a community. They were far smaller than the molecules found in modern organisms, and probably were heterotrophic--relying on pre-formed organic molecules for energy, as autotrophy would have required a sophisticated metabolism. So what were these molecules, and how, where and when did they arise? These are the fundamental questions facing origins-of-life research.

4.4. THE ORIGIN OF LIFE: CURRENT THEORIES 43 4.4.1 When Did Life Begin? The second major eon in Earth's history, the Archean, lasted from 3.9 to 2.5 Gyr ago. The oldest rock samples date from roughly the beginning of this period: the Isua rocks of western Greenland, for instance, are 3.8 Gyr old. Rocks probably existed prior to that time, but none have been found intact: they were probably destroyed by resurfacing or impact events that were common in the Hadean eon. One interesting aspect of these oldest rocks is that they appear to be metamorphosed from sedimentary rocks: oceans of water must have already existed at this time.

Surprisingly, some of the oldest rocks already show evidence for life. Fossil imprints of filamentous bacteria have been found in 3.5-Gyr-old rocks from Australia by J. William Schopf, who has noted a strong resemblance to photosynthetic bacteria. Support for this interpretation comes from stromatolites of comparable age found in Australia and southern Africa, thought to be fossilized versions of bacterial colonies found today in coastal regions. These colonies are topped by a layer of photosynthesizing cyanobacteria that provide the food for lower layers of heterotrophic bacteria. The resemblance between modern and fossil stromatolites suggests that oxygen-yielding photosynthesis, already quite an accomplishment, may have arisen over 3.5 Gyr ago.

Even earlier evidence for life, back to 3.85 Gyr ago, has been reported by Stephen Mojzsis and colleagues based on the isotopic ratio of

13

C to

12

C in the Isua rocks. This

ratio is different from that seen in non-biological carbon samples and is what we expect from the preferential uptake of

12

C by organisms. However, the rocks have been metamorphosed

(partially melted), removing any fossils, and the metamorphism itself can affect isotope ratios, so this evidence is somewhat controversial.

In any case, the data seem to point to a (geologically) short time between when the Earth became stable enough for life to exist (about 4.0 Gyr ago, based on the lunar cratering record) and when fairly sophisticated organisms begin to appear in the fossil record. Life seems to have originated in less than 500 Myr, perhaps in 100 Myr or less. This is especially startling given the large number of steps needed to get from simple molecules to self-sufficient cells. Many scientists have drawn the conclusion that the origin of life must have involved a series of probable events: if even some of the steps had been improbable, the whole process would have taken vastly longer. On the other hand, maybe life had to emerge quickly if it was going to emerge at all--the reducing atmosphere, if it existed, was probably short-lived. Then we might be an accident of history after all, extremely rare in the cosmos. This is a less satisfying option, though, because if life was a fluke we'll probably never understand its origin.

4.4.2 Where Did Life Begin? The question of where life began on Earth is a crucial one for understanding how it began. It also has implications for life elsewhere in the Solar System: if life can originate in deep subsurface environments, it could have existed on many worlds when they were young, and might still exist on Europa and Mars today. On the other hand, if life can only begin at or

44 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH near the surface, the number of possible habitats is substantially reduced.

At first glance, surface environments where water and land meet seem ideal because there exist places which are alternately wet and dry: water is needed to bring chemicals together, but then dehydration is needed to get them to join. In the early Archean, the Earth may have been covered by a global ocean, as the formation of continental crust, which is less dense than oceanic crust, occurred only after the repeated melting of crustal rocks. Still, there were probably some irregularities in the Earth's surface, due to volcanoes, large impact craters, etc., that poked out of the oceans as dry land. The ocean tides were also much stronger because the Moon was much closer to Earth, so a given spot would see the oceans rise and fall by tens of meters. Thus it is conceivable that life could have begun in a "warm little pond" as Darwin envisioned, or perhaps a tidal pool or lagoon which was periodically filled and drained. Organic molecules, delivered by comets or formed by MillerUrey type reactions, would be concentrated in shallow water and would dehydrate to form polymers when the water receded. Eventually the pool would be filled with a primordial soup of organic molecules of ever-increasing complexity. From this soup, a molecule would arise that was able to make crude copies of itself--the ancestor of life.

This orthodox view, espoused early on by Haldane and Oparin, has recently been challenged by another proposal: that life originated in a hydrothermal system, where the source of energy is geothermal heat rather than lightning or sunlight. These systems occur deep below the surface, as hot springs or mid-ocean ridges where new crust is being created. They all involve the circulation of water through volcanically heated rock. The hydrothermal model has several advantages over the primordial soup model:

ffl As mentioned earlier (x4.3.3), the early atmosphere is no longer believed to have been

strongly reducing. There was, however, another source of reduced chemicals: volcanoes and mid-ocean ridges, which power the hydrothermal systems. The upper mantle was probably less oxidized than it is today, and chemical reactions occurring near vents generate reduced molecules even now.

ffl The Earth's surface may have been a dangerous place even after it cooled enough for

oceans to form, as gigantic impacts continued to sporadically frustrate the development of life. UV radiation would also have been damaging to fragile chemical bonds.

ffl Studies of RNA sequences in the ribosomes of different kinds of bacteria indicate that

the bacteria lying closest to the "last common ancestor" of all life are thermophilic, living in high-temperature environments (x4.6.1).

The major uncertainties in this model are whether molecules can polymerize with so much water available (x4.4.4) and whether the temperatures in these environments are too high to allow large molecules to be stable. Also, would the organic molecules stay concentrated enough to react with each other, or would they just drift away into the open ocean? These are important questions that remain to be addressed.

4.4. THE ORIGIN OF LIFE: CURRENT THEORIES 45 4.4.3 How Did Life Begin? The Chicken and Egg Problem In Chapter 1 we introduced the basic processes of life: metabolism and reproduction. A twoway relationship exists between these processes, with metabolism allowing the replication of hereditary molecules (DNA) and DNA determining what reactions occur in metabolism. The relationship between metabolism and heredity boils down to the relationship between proteins and nucleic acids: each seems to be necessary for the other. Useful proteins do not seem to form spontaneously, but are produced on RNA templates. RNA and DNA, on the other hand, cannot reproduce without the action of protein enzymes. This is the classic "chicken and egg" problem.

How did such a tightly integrated system ever emerge? Scientists have long favored a "genes first" hypothesis, in which metabolism was initially based on chemical reactions already occurring in the environment rather than controlled by enzymes. Somehow this proto-metabolism created the first replicating molecules, which only later gained the ability to synthesize proteins. But thus far, no one has been able to get nucleic acids to replicate without the help of proteins, although we may be getting close.

The RNA World A major breakthrough occurred in 1983, when Thomas Cech and Sidney Altman, working independently, discovered that RNA can sometimes behave like an enzyme--catalyzing reactions involving other RNA molecules. For this work they received the Nobel Prize in Chemistry in 1989. At first the known catalytic abilities of these ribozymes (RNA enzymes) were limited to cleaving an RNA molecule and joining it back together. Since then, however, ribozymes have been induced to perform many of the steps in RNA replication, such as stringing short segments of RNA together, and some have been endowed with the ability to cleave peptide bonds (bonds linking amino acids). The idea of an RNA world, where RNA could evolve and replicate independently of proteins, is now widely accepted as the solution to the chicken-and-egg problem and the first step in the evolution of life.

So why RNA rather than DNA? Because RNA, like proteins, can form a variety of 3-dimensional shapes depending on the nucleotide sequence. DNA, on the other hand, occurs mainly in the double helix form--making it extremely stable and error-resistant, but hindering its ability to behave like an enzyme. The idea that RNA preceded DNA as the hereditary molecule of life is also attractive for various reasons: DNA nucleotides are harder to make than (and indeed, derived from) RNA nucleotides, and it is RNA, not DNA, that is involved in the all-important process of protein synthesis. In this picture, the more stable DNA molecules eventually supplanted RNA as the repository of genetic information.

Proteins first? The abilities of ribozymes, although impressive, are meager in comparison to what modern proteins can do. Some opponents of the RNA world hypothesis, such as Sidney Fox, argue instead that proteins preceded nucleic acids. In 1958, he and his collaborators discovered that a dry mixture of amino acids heated to 170

ffi

C yielded polymers

roughly 50 amino acids long. Moreover, these "proteinoids", as he called them to distinguish them from true proteins, are not randomly structured, but possess a high degree of order,

46 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH and even some catalytic properties. This is not entirely unexpected, since amino acids have complex 3-dimensional structures that fit together better in some ways than in others. Still, Fox's conclusion--that life could arise from the spontaneous self-ordering of amino acids, independent of any other informational molecule such as DNA or RNA--was greeted with skepticism.

Fox's critics pointed out that his proteinoids could not reproduce, and that the dry heating he applied to the amino acids, followed by return to aqueous solution, didn't correspond to any realistic prebiotic environment. Still, the process of forming amino acids from a Miller-Urey type of process and then joining them together to make short proteins is probably easy in comparison to what is needed to make RNA, as we'll discuss below. There remains the possibility that proteins, even if they can't catalyze their own reproduction, can form a network where each member catalyzes the reproduction of another. As a group, then, this autocatalytic network would be capable of self-reproduction. The feasibility of such networks remains a subject of current debate.

4.4.4 The Polymerization Problem If life began as RNA, where did that RNA come from? Each ribonucleotide is composed of three parts--a five-carbon sugar called ribose, a phosphate group, and a nitrogenous base-- and an RNA molecule is a polymer composed of many nucleotides. Getting all these pieces to come together in the absence of enzymes remains an elusive goal for biochemists. Synthesis of ribonucleotides has been achieved to some degree in the laboratory, but the likelihood of the process occurring on the early Earth is controversial, especially since biologically usable phosphate appears to be uncommon and difficult to produce. Furthermore, the phosphate and base must bond with ribose in the right configuration or else further polymerization--the joining of nucleotides to make RNA--will be hindered.

The deeper problem, as mentioned in x4.4.2, is that polymerization usually requires the removal of water, either by a catalyst (a condensing agent) or by simple drying, such as would occur in an evaporating pond or tidal pool. Theories of life's origin in oceans or hydrothermal vents thus face the basic difficulty that water tends to break apart (hydrolyze) large molecules and inhibit them from joining together again. Attention has therefore focused on whether chemical agents or mineral surfaces can act as catalysts to induce polymerization, even in the presence of large amounts of water. An attractive feature of mineral surfaces is that they can concentrate reactants: with only two dimensions to move in, it's much easier for molecules to bump into each other and join, and much harder for water to attack the resulting bonds. Recently, a German group has succeeded in getting amino acids to join in simulated hydrothermal vent conditions using iron and nickel sulfide mineral grains.

3

Chemical catalysts, on the other hand, cannot concentrate reactants, and are themselves susceptible to dilution.

Once the polymerization problem is solved, it becomes possible to create information, stored as sequences of nucleotides. But for evolution to begin there must have been a way to replicate this information. Much recent effort has been focused on trying to build an

3

C. Huber and G. W"achtersh"auser, Science, 281, 670 (31 Jul 1998).

4.4. THE ORIGIN OF LIFE: CURRENT THEORIES 47 RNA molecule that is capable of self-replication--to recreate the RNA world in the laboratory. Not surprisingly, the catalytic ability of ribozymes depends on the specific sequence of nucleotides. As a result, an RNA catalyst with a "good" sequence may be able to piece together a complementary RNA molecule (with the usual matching rules, so ACU is complementary to UGA) but the complement will generally not have the ability to generate a replica of the original molecule. Perhaps someday we will discover a pair of ribozymes that are complementary to each other, so that each would be able to generate copies of the other. Alternatively one can imagine a ribozyme that is able to make a complement of its complement: a complement of your complement is a copy of yourself! If our laboratory experiments were to stumble upon one of these solutions by chance, it would give us confidence that Nature could do the same.

4.4.5 The Selectivity Problem There is one other serious difficulty with the RNA world, and the chemical evolution model in general, that needs to be addressed. Nearly all laboratory simulations of the origin of life share a common weakness: the types and concentrations of reactants (chemicals you begin with) have been carefully controlled. In the primordial soup or whatever medium life began in, this could not have been the case--there would have been all kinds of "junk" around, ready to interfere with the reactions. The purification steps that are used in the lab to get rid of the "junk" and concentrate the good stuff are unlikely to occur in nature. One example is the formation of ribose: the most straightforward process for doing so also makes lots of other sugars that would get in the way of making RNA. No abiotic process has been found which can yield ribose as its main product; similarly any process for making nucleotide bases will make more than just the biological ones. Yet another illustration of the need for selectivity, mentioned in x4.4.4, is the tendency for incorrectly shaped nucleotides (with the phosphate and base in the wrong place) to form alongside the correctly shaped ones.

Understand how life came to use only left-handed (L) amino acids and right-handed (D) sugars is a particularly vexing problem. Any nonbiological processes for making amino acids and sugars will make roughly equal numbers of both right- and left-handed varieties, and without enzymes to supervise the process, these will tend to participate in the same polymerization reactions. But inserting even a single amino acid or sugar of the wrong handedness into a protein or RNA molecule will render it useless. Scientists are aware of this and design their experiments to include only building blocks of the correct handedness, but how this situation would arise in nature is completely unclear.

The need for selectivity can be met by the existence of some catalyst or template that enforces it--that somehow favors the "correct" ingredients being used, like the enzymes of today. It has been suggested that the surfaces of clays could have performed this function, since clay crystals have ordered structures that bind with some molecules better than others. However, this idea has not been demonstrated in the laboratory. In the absence of such mechanisms, many researchers believe that a more primitive genetic system must have preceded RNA. Proposals for such a system include peptide nucleic acid, where the ribose is replaced by a unit containing the amino acid glycine. Further work is needed to determine whether such molecules could form in a prebiotic setting and catalyze their own replication.

48 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH 4.4.6 Cell Membranes: Encapsulating Life Suppose Nature has overcome these difficulties and has managed to create the RNA world. How then can one form a living cell? I have intentionally avoided discussion of cell membranes, which are today one of the universal features of life, because in the RNA world such membranes may have been a hindrance rather than an asset. One reason cells exist is to keep the materials needed for metabolism together, but these materials ultimately have to be acquired from the environment. The existence of a cell membrane would have trapped and starved RNA molecules unless mechanisms for transporting substances across the membrane existed (today this is done by special proteins embedded in the membrane).

On the other hand, one can imagine scenarios where some kind of surrounding membrane (generically called a vesicle) would have been useful if not mandatory. For example, the need to keep all the molecules of an autocatalytic network in close proximity to one another would favor a closed environment. Regardless of how life began, there must have come a time when it became advantageous for molecules to be concentrated, and to work together (the way nucleic acids and proteins do today) instead of competing for resources. Moreover, many energy utilization pathways, including photosynthesis and aerobic respiration, depend critically on membranes.

Living cells today are contained within membranes structured as phospholipid bilayers. A phospholipid can be thought of as a sperm-shaped molecule that has a hydrophilic end (the "head") and a hydrophobic end (the "tail"). A bilayer is formed from two parallel sheets of phospholipids making a sandwich. The heads of the molecules all face outwards (towards the surrounding water) while the tails all point inwards (towards the middle of the sandwich). This structure emerges quite spontaneously because the hydrophobic tails want to avoid contact with water, so they hide behind the heads. With a limited number of molecules, the sandwich tends to curve around on itself to produce something akin to a soap bubble; the resulting "empty cell" is called a liposome.

However, the earliest life probably could not use phospholipids for its membranes. For one thing, their formation requires first making lipids (the main ingredient in fats and oils), which seem to require high temperatures and large amounts of H

2

to form. Perhaps

hydrothermal vents provided a suitable environment for these reactions. But the main problem is the relative impermeability of modern cell membranes, which makes it hard to get food in and throw wastes out. David Deamer and his colleagues have had success in creating more permeable membranes from shorter organic acids, similar to some found in the Murchison meteorite. This approach has the potential to solve both the formation problem (assuming these molecules are generally found in meteorites) and the permeability problem.

4.4.7 The First Cell If the spontaneous formation of vesicles was occurring everywhere, it is likely that they existed alongside the first self-replicating molecules, and that some of those molecules occasionally got trapped in a vesicle. If the membrane was somewhat leaky, the monomers needed for replication might still have been able to move in and out of the vesicle while the

4.5. BIOLOGICAL EVOLUTION: BASIC PRINCIPLES 49 polymers would be kept inside. Thus, the first living molecules would have found a cozy home and still been able to survive.

Once RNA could survive inside cells, it could start doing other things besides copying itself. For instance, ATP, a ribonucleotide with two extra phosphate groups, could be used to store energy. The most important innovation was the ability to make specific proteins using information in nucleic acids, what we now call translation. The primitive protein translation system was probably based on chemical attractions between certain RNA sequences (the ancestors of tRNA) and certain amino acids, since by assumption protein enzymes weren't around to supervise the process.

With a simple protein translation apparatus, cells could begin to compete with one another based on how useful their proteins were. Natural selection would favor the ability to translate RNA to protein reliably, and the general affinities between RNA sequences and amino acids would solidify into a genetic code. It is likely that less than 20 amino acids were originally used, as few proteins really use all 20--this is especially true for the more ancient proteins. Thus the genetic code could have been much simpler, in accordance with speculations that RNA may have started out with only two kinds of nucleotides. Initially, different cells may have employed different genetic codes, and the cells with the "best" codes produced more descendants. In fact, the modern genetic code does not appear to be arbitrary, but rather a product of natural selection. Recent studies indicate that compared to most possible codes, it is highly resistant to errors: in other words, an error in the DNA will often leave the protein being coded for unchanged, or at least similar in structure.

These speculations about the first living cells should be taken with a grain of salt, since we still don't know the exact order in which basic structures and processes (cell membranes, protein synthesis, use of ATP, creation of DNA, etc.) arose. Of one thing, however, we can be fairly certain: if the first forms of life were heterotrophs, there would have been a huge advantage for a cell that was autotrophic, especially as the "soup" of complex organics got eaten up and the supply of reduced compounds (provided by the atmosphere or mantle) dwindled. This raises an interesting question: on how many worlds did life eat itself to death by not developing autotrophy fast enough?

4.5 Biological Evolution: Basic Principles The theory of evolution by natural selection, first put forward by Charles Darwin and Alfred Russell Wallace in 1858, is one of the great achievements of science. The fact that evolution had occurred--that organisms have changed over time--had long been apparent from the fossil evidence, but it was Darwin and Wallace who hit upon natural selection as the explanation: that in any population there is variation among individuals, that these variations are often inherited, and that beneficial variations will become more widespread in succeeding generations. Far more individuals were born into the world than could possibly survive, leading to a "struggle for existence," as the philosopher Thomas Malthus had prophesied would happen in human populations. Darwin expounded this theory in his masterpiece, The Origin of Species, and as a result it is often called "Darwinian evolution" or "Darwinism."

50 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH

Two important concepts in evolutionary thought are the notions of a species and a population. A species is a group of individuals that can successfully breed with one another to produce fertile offspring, but cannot breed with members of other groups. A population is a group of individuals of the same species that occupy the same region and can mate with each other freely.

Natural selection is often called "survival of the fittest," but is perhaps better described as "reproduction of the fittest." As the work of Gregor Mendel in the 1800's demonstrated, heredity is controlled by interactions among discrete units called genes (now known to be comprised of DNA) that are passed down from one generation to the next. Natural selection simply says that genes which lead to traits aiding survival and reproduction will be more strongly represented in subsequent generations than genes which are neutral or detrimental. The tendency for harmful traits to be removed from the gene pool (the set of all genes in a breeding population) is known as selective pressure, and traits that improve the chances of survival and reproduction are called adaptations.

The biochemical similarities among living things suggest that they are derived from a single common ancestor. Thus, not only can species change over time, but they must sometimes split into multiple descendant species. The process by which this occurs (speciation) is still poorly understood. In sexually reproducing organisms, it is thought to occur in small populations that become geographically isolated from the main group. A small group will tend to evolve in a different direction since it possesses less genetic variation and can yield unlikely gene combinations by inbreeding. If communication is later re-established with the main group, the isolated population may have changed enough that it no longer interbreeds with the original population: a new species has been created. Well-documented examples of this process are found on oceanic islands, including the finches that Darwin studied on the Gal'apagos Islands.

As critics of Darwinism have eagerly pointed out, there are still many questions about how evolution works that have not been fully answered. These include the following:

ffl Is natural selection the only force in evolution?

Here the answer is clearly no, although the extent to which other factors are important remains controversial. New variations are constantly being created by mutation, and many of those variations have no immediate effect on the "fitness" of the organism. These neutral mutations are especially common at the molecular level: many changes that have occurred in proteins over time have had little effect one way or the other on their functionality. In another process, known as random genetic drift, certain genes become more or less common in the gene pool simply as a result of statistical fluctuations. This effect is especially strong in small populations. Finally, chance appears to play a big role in evolution, especially when it comes to extinction. Hence, we must beware of the tendency to look at every feature of organisms as perfected by natural selection--some characteristics have no adaptive value, or may become useful only after the environment changes.

ffl How do complex adaptations arise?

4.5. BIOLOGICAL EVOLUTION: BASIC PRINCIPLES 51

The origin of complex biological adaptations is clearly an important and unsolved problem. These characteristics have clearly been perfected by natural selection, but how did they arise in the first place? A common criticism of Darwinian theory is that it relies on the accumulation of small changes to produce major innovations. In particular, it is sometimes argued that many complex adaptations, such as flight and vision, must have been acquired in a single step because an incomplete version of a wing or an eye would be useless. The conclusion is then that some kind of miracle had to be involved.

However, on closer inspection this argument is hardly convincing. One can imagine situations where the ability to glide from one tree to another or the ability to detect light with a primitive eye would still be quite valuable. Certainly primitive eyes are better than no eyes at all. In situations where an incomplete structure appears to be useless, perhaps there were once auxiliary parts of the structure which have since been lost. Given that we are still discovering new and powerful sources of mutation (x1.4), it seems premature to decide what can and cannot be achieved by natural selection.

ffl Does evolution proceed gradually or in jumps?

The traditional view that new species arise from pre-existing ones by gradual change governed by natural selection is not borne out by the fossil record: instead, species seem to abruptly appear and disappear, with few if any transitional forms between species. This has led Stephen Jay Gould and Niles Eldredge to propose that speciation occurs in brief episodes, followed by long periods of stasis in which little change occurs. This model is known as punctuated equilibrium, or "evolution by jerks" to its opponents. It has been popularized by Gould in a number of books.

Gould and Eldredge argue from the fossil record that most adaptive change occurs not within a species, but only when new species are created. Although they seem to think this is quite a revolutionary idea, it is actually not in conflict with standard evolutionary theory: in a large stable population, mutations are always occurring but their effects are diluted by the large size of the gene pool and the lack of strong selective pressures in one direction or another. As a result, large populations evolve hardly at all. On the other hand, small, isolated populations can experience different environmental pressures and begin with a smaller variety of genes than the larger population. This allows mutations to have a stronger effect and promotes both adaptive change and speciation. If most of the evolutionary "action" occurs in small populations that evolve quickly, we are unlikely to find transitional forms in the spotty fossil record.

A more extreme version of punctuated equilibrium, called saltationism, argues that speciation occurs not just quickly, but instantaneously as a result of macromutations that are fundamentally different from the mutations which lead to variations within a species (micromutations). Such a macromutation might affect a gene that regulates other genes, so its effects would be greatly magnified. Needless to say, instant speciation has not been observed in living organisms (e.g. a lion giving birth to a cougar), but it may have been more common in the past (for instance, at the time of the Cambrian explosion, x4.6.3).

52 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH

ffl What is the role of chance in evolution?

The role played by chance events in every aspect of evolution has attracted growing attention in recent years. Not only are the sources of genetic variation often unpredictable, but the extinction of species seems to depend much more on blind luck than was previously believed. Well-adapted groups like the dinosaurs were wiped out by giant impacts and other global cataclysms, while poorly-adapted species like koalas have been able to survive in "safe havens" where there are few predators. The flightless birds called dodos which once lived on oceanic islands were certainly well-adapted to their habitats, but were driven to extinction by an unexpected change in their environment: European sailors.

In this regard, it is worth emphasizing that evolution does not always lead to "progress" in the sense of improved fitness or higher intelligence. Where evolution does seem to lead to improvement over time, this improvement is often a result of competition between predator and prey: such an "arms race" forces you to keep running just to stay in the same place. But often, as occurs in parasitic organisms, evolution leads to a loss of complexity. One of the most exciting fields in evolutionary biology is the study of the "evolution of evolvability," where the emphasis is not on becoming perfectly matched to your environment but having the genetic tools to adapt to sudden changes in that environment or to exploit new opportunities.

4.6 A Brief History of Life on Earth 4.6.1 The Tree of Life The "family tree" of life can be determined in many different ways. For many years the standard method was to compare anatomical characteristics between species: these indicated, for example, that mice and gophers were more closely related to each other than to aardvarks. One must be careful, however, with this approach. We find, for instance, that the superficial resemblance between sharks and dolphins is a result of convergent evolution: dolphins actually are descended from shrew-like animals like all mammals, but have adopted a form that allows for efficient swimming. On the other hand, shared features that have little to do with functional necessity, such as having five digits on each limb, are much more likely to indicate a common ancestor.

In recent years it has become possible to isolate and sequence nucleic acids from cells quickly and efficiently, and this has led to a revolution in our understanding of life's evolution. One particularly ancient RNA molecule is the 16S ribosomal RNA, which is possessed by all organisms and is involved in protein translation. By comparing 16S rRNA sequences between different kinds of organisms, it is possible to construct a phylogenetic tree (a genealogy showing evolutionary descent) under the assumption that the more similar the sequences are between two species, the more recent was their last common ancestor (LCA). This technique of molecular phylogeny, pioneered by Carl Woese of the University of Illinois, is subject to much less ambiguity than classification by appearances.

4.6. A BRIEF HISTORY OF LIFE ON EARTH 53

One startling result of this analysis is that plants, animals, and fungi have nearly identical 16S rRNA: they have evolved little from their LCA. Virtually all of life's diversity is in the microbial world of bacteria and protists. In particular, three distinct groups seem to stand out, all very ancient: (1) the archaeabacteria or simply archaea, (2) the eubacteria or simply bacteria, and (3) the eucarya, which includes us. The eukaryotes (plants, animals, protists, fungi) therefore make up just one of the three branches, and the prokaryotes are revealed as two very distinct groups, the archaea and the "true" bacteria. (See the article by Norman Pace in Science, Vol. 276, p. 734, 2 May 1997.)

Another surprising feature is how similar all forms of life are genetically, in spite of their different appearances and lifestyles. The possession of essentially the same molecules, protein-synthesis machinery, and genetic code, and the use of similar basic genetic sequences like 16S rRNA, strongly indicate that all life today is descended from a single universal ancestor. Life may have originated several times, but only one of those lineages survived. This of course is an assertion about history that can never be proven, but it could be disproven at any time by simply finding a radically different form of life.

The exact form of the phylogenetic tree is somewhat uncertain. A common phenomenon in the microbial world is horizontal gene transfer, in which one organism transfers some DNA to another. The resulting organism is then difficult to classify. Imagine having fly DNA put into your genome: would you be classified as a mammal or an insect? For eukaryotes the introduction of foreign DNA is almost always disruptive (consider the damage caused by viruses), but in prokaryotes it is often not, and the biological species concept of a welldefined community capable of interbreeding only with each other sometimes breaks down. Another complication is the merger of two organisms into a single cell: the mitochondria and chloroplasts in eukaryotes arose from such an endosymbiosis, and it has happened other times in the history of life too.

But in spite of these uncertainties, we have some indication of what properties the universal ancestor possessed and lacked. For instance, photosynthesis is only found in bacteria or chloroplasts (which were once free-living bacteria), so it probably arose in the bacterial line after it split from the others. Researchers have found that the most primitive bacteria and archaea (primitive in the sense that they have evolved little from the universal ancestor) are all thermophilic, preferring very high-temperature environments (near 100

ffi

C),

and chemosynthetic, deriving their energy from inorganic compounds such as H

2

and H

2

S

(x1.2.3). This has led UCB microbiologist Norman Pace and others to conclude that the universal ancestor also possessed these properties, and to favor an origin of life in deep hydrothermal environments. On the other hand, even if the universal ancestor lived in such an environment, it could have been descended from surface life that was wiped out by a giant impact.

4.6.2 The Oxygen Revolution The fact that some prokaryotes require oxygen, some can live with or without it, and some survive only in the absence of oxygen indicates that the Earth's atmosphere experienced a transition from being oxygen-poor to oxygen-rich at a time of abundant prokaryotic life. The

54 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH geological record is also consistent with an early oxygen-free atmosphere. Continental rocks about 2.7 Gyr old contain fragments of uraninite (UO

2

) and pyrite (FeS

2

), which would

have been chemically altered in the presence of an oxygen-rich atmosphere. In addition, the sedimentary iron-rich rocks called banded iron formations (BIFs), which make up over 90% of the world's commercial iron supply, could not form today because there is too much oxygen dissolved in the oceans. Most of the BIFs are 2-3 Gyr old, with a few older examples. By 2 Gyr ago, we begin to see the appearance of a new type of iron-rich rocks called redbeds that require oxygen for their formation; these overlapped with the BIFs for a few hundred Myr before supplanting them entirely.

How is this early lack of oxygen consistent with the existence of stromatolites (colonies of photosynthesizing bacteria) 3.5 Gyr ago? One possibility is that these bacteria did not produce oxygen, but relied on more primitive sulfur-yielding photosynthesis. Alternatively, stromatolites may not have been widespread at such early times, and what little oxygen they produced reacted with iron in the oceans or surface rocks rather than accumulating in the atmosphere. Even today, oxygen would disappear from the atmosphere in roughly 6 Myr if it was not continually replaced by photosynthesis.

In any case, by about 2.2 Gyr ago cyanobacteria were releasing copious amounts of oxygen into the atmosphere, which could no longer be balanced by inorganic processes. At the same time, the volcanic gases released by the mantle were becoming more oxidized as the escape of hydrogen continued. This oxygen buildup must have led to a severe crisis for life on Earth, as the gas tends to react with and destroy organic molecules. Some of the bacteria sought refuge from the air and remain anaerobic to this day. Others, including (not surprisingly) the cyanobacteria themselves, synthesized antioxidants such as vitamin C that could get rid of the dangerous free radicals produced by oxygen in the cell, and developed the means to take advantage of O

2

in respiration. As noted in Chapter 1, the use of oxygen

as an electron acceptor allows respiration to proceed much further, yielding 36 ATP's in addition to the 2 provided by fermentation.

One of these aerobic bacteria, the mitochondrion, became a permanent resident inside one of the eucarya, and the modern eukaryotic cell was born.

4

The cell provided the mitochondria with a constant supply of food in return for mopping up oxygen and producing ATP. The once radical idea that eukaryotic cells arose from a symbiosis with bacteria, suggested by Lynn Margulis in 1967, is now widely accepted. In particular, mitochondria and chloroplasts contain DNA of their own (although not enough to be self-sufficient), and some aspects of their metabolism closely resemble that of bacteria. Although clear fossil evidence is lacking, this symbiosis probably emerged from 1.5 to 2 Gyr ago.

Armed with the ability to use free oxygen with the help of mitochondria, eukaryotic cells became larger and more complex than anything that had existed up to that point. Increased fidelity in copying DNA allowed much more genetic material to be put into a single cell. Some formed colonies that would develop into the first multicellular organisms, between 600 and 700 Myr ago. Multicellularity was a tremendous advance, relying on regulatory genes

4

I use eukaryotes to refer to the traditional definition, with nuclear membranes, organelles, etc., and

eucarya to also include the ancestors of modern eukaryotes--sometimes called urkaryotes--that may not have had all these features but were still distinct from archaea and bacteria.

4.6. A BRIEF HISTORY OF LIFE ON EARTH 55 that could control which genes were activated and which were silenced: this allowed different kinds of cells to be produced from the same genetic information. It is important to bear in mind, however, that long before plants and animals came along, our one-celled ancestors had perfected most of the biochemical apparatus needed for life. The old notion that life "didn't do much" in the first 3 billion years of its existence seems foolish when we consider how important the advances made during that time were.

4.6.3 The Cambrian Explosion A critical period in the evolution of life occurred about 530 million years ago, when a huge variety of new animals emerged, inaugurating essentially all of the animal phyla that survive to this day. For the taxonomists, the phylum represents the level of biological classification just below the kingdom (plant, animal, etc.) and distinguishes between basic body plans. For instance, making up one of the largest phyla are the arthropods, creatures with hard exoskeletons that include the insects and spiders of today, while we belong to the chordates, creatures with a spinal column that include the vertebrates of today. The sudden and almost simultaneous appearance of all of these phyla has been called the Cambrian explosion, as it occurs at the beginning of the Cambrian period within a timespan of less than 10 Myr. The most famous fossils from this period come from the Burgess Shale, sediments that are located today in the Cascade Mountains of British Columbia.

There are two remarkable features of the Cambrian explosion that remain a puzzle for biologists. First, it appears to have been a "top-down" event, in the sense that the phyla emerged first and then small variations led to the creation of new species within each phylum. This runs contrary to the traditional Darwinian view, in which differences between high-level classifications like phyla are the result of differences between species that accumulate over long periods of time. Second, nothing like the Cambrian explosion has occurred since: in fact, the fundamental diversity of animals has barely increased, if at all.

Although it is unclear what led to the Cambrian explosion, it may have been a combination of a sudden rise in oxygen levels in the atmosphere, which allowed for more efficient metabolism, and evolutionary innovations including the development of internal circulatory fluid (blood) and sex. The dearth of fossils prior to the Cambrian must be due in part to the fact that the animals of that time were soft-bodied and didn't fossilize readily, but what fossils do exist show little foreshadowing of the diversity of Cambrian fossils. It may be that the Cambrian explosion was an event that could happen only once in the history of life, when few predators or competitors existed and more flexible genes allowed evolution to test out many different patterns of development, including the tetrapod model we're familiar with (head, torso, and four limbs).

4.6.4 Mass Extinctions The more complete fossil record of animals since the Cambrian has revealed the occurrence of several mass extinctions, where a large fraction of life's diversity has been wiped out in a relatively brief amount of time. One such mass extinction occurred at the end of the

56 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH Permian Period (225 Myr ago) and ushered in the Mesozoic era, the age of the reptiles. The most famous extinction event, although less severe than the Permian, occurred at the end of the Mesozoic (65 Myr ago) and wiped out the dinosaurs; this ushered in the Cenozoic era, the age of the mammals. In the case of this Cretaceous-Tertiary (K/T) extinction, there is now convincing evidence that it was caused by the impact of an asteroid about 10 km in size into the Yucatan peninsula of Mexico (x5.4). Not only would such an impact release a huge amount of energy (equivalent to over a million H-bombs), but it would have severe and long-lasting effects on the atmosphere and climate.

Clearly these mass extinctions, whether caused by astronomical, geological, or other phenomena, have had a strong effect of the recent evolution of life. Of course, this is a somewhat biased perspective: most of the bacteria would have barely noticed, aside from the fact that there was more dead stuff to eat. The "dominant" organisms that are high up on the food chain, though, are quite susceptible to extinction, especially when the food they depend on becomes scarce. This is probably what happened to the dinosaurs, and could happen to many groups of mammals (including us) when the next event occurs. The higher intelligence of mammals, made possible by their faster metabolic rates (warm-bloodedness), was probably not what gave them the edge over the dinosaurs, and will probably not be of much use when the next asteroid hits.

The many niches and environments vacated by the dinosaurs allowed for a rapid diversification of the mammals after the K/T event. This led to the development of a number of new forms, including the primates. But only one of those species--Homo sapiens--would develop a technological civilization. What steps were involved in the evolution of intelligence? We will discuss this issue further in Chapter 7.

4.7 Life's Effects on the Earth There is no question that the atmosphere of the Earth has been strongly affected, indeed reshaped, by life. In fact, life has been involved in the production or removal of most of the gases in the atmosphere.

1. The most obvious effect of life is molecular oxygen, which makes up 21% of the present

atmosphere. As discussed in x4.6.2, rocks did not become exposed to large quantities of O

2

until about 2 Gyr ago, indicating that oxygen-yielding photosynthesis had become

widespread by that time. Chemical reactions involving O

2

in the upper atmosphere

in turn created an ozone (O

3

) layer, which shields the ground from the Sun's UV

radiation. In a sense, life was able to pave the way for itself to colonize the land.

2. Aside from oxygen, the major constituent of the atmosphere is nitrogen (78 percent).

Although N

2

comes from volcanic sources, its presence in conjunction with O

2

is also

largely a result of life. The reason for this is that nitrogen and oxygen combine to form nitric oxides when lightning passes through the air, which collect in the soil and oceans. Certain "denitrifying" bacteria play a major role in breaking these compounds back down into N

2

.

4.7. LIFE'S EFFECTS ON THE EARTH 57

3. While the Earth has much more O

2

in its atmosphere than Venus or Mars, it also

has much less CO

2

. CO

2

dissolved in seawater becomes trapped in rocks as calcium

carbonate (CaCO

3

), a process that has been greatly accelerated by marine life, which

make their shells from this material. Soil microorganisms also remove a great deal of CO

2

from the air and promote weathering reactions in rocks. As discussed in x3.1, the

lowering of CO

2

levels since the Archean has reduced the greenhouse effect, balancing

the growing luminosity of the Sun and maintaining the Earth's habitability.

4. Finally, life is responsible for a major disequilibrium in the atmosphere: methane

(CH

4

, natural gas) is present at a level 10

35

times higher than one would expect in an

atmosphere with free oxygen. That's because oxygen and methane react very easily:

CH

4

+ 2O

2

\Gamma ! 2H

2

O + CO

2

(This reaction is used to burn natural gas, but can happen more gently too.) The high level of methane, it turns out, is maintained by methane-producing bacteria (such as in cows' guts) that are constantly injecting the gas into the atmosphere.

Based on this and other evidence of life influencing its environment, James Lovelock has argued that the Earth's atmosphere is regulated by life so as to keep the planet habitable. For instance, the constant production of oxygen maintains the ozone layer, which in turn blocks UV radiation and protects water vapor from dissociation (recall that this is how Venus lost all its water). Lovelock has compared the Earth to a living organism--dubbed Gaia after the Greek goddess of the Earth--that maintains itself the way any organism does. Most geologists seem to be uncomfortable with this idea, pointing out that non-biological mechanisms, such as the weathering of rock by CO

2

, could maintain the Earth's temperature

in the absence of life. Clear evidence that biological activity was able to reverse an ice age or some other global calamity would lend support to the Gaia hypothesis.

58 CHAPTER 4. ORIGIN & EVOLUTION OF LIFE ON EARTH Chapter 5 Comets & Asteroids: Creators or Destroyers?

Chapter 4 presented an overview of current thinking about the origin and evolution of life, from the perspective of chemistry and biology. In this chapter we examine these events from a more astronomical standpoint. In particular, we will examine the suggestion that life, or at least complex organic molecules, could have originated in space and then been brought to Earth. We will also look at the effects that impacting bodies (comets and asteroids) had on the evolution of life once it got started.

5.1 Extraterrestrial Sources for Biogenic Elements The four major biogenic elements (chemical elements used by life) are carbon, hydrogen, oxygen, and nitrogen. Except for oxygen, which is a component of silicate (SiO

2

) rock, none

of these is very abundant on Earth, despite being very common in the Universe at large. The reason is that they tend to be found in volatile forms such as gases, which make up only a tiny fraction of the Earth's mass. Thus, while the atmosphere and hydrosphere (oceans) are composed mostly of these elements, they are nowhere near as large a percentage of the Earth's mass as they are for the Sun or Jupiter. The terrestrial planets, being too small to hold on to massive atmospheres and too warm for volatile substances to occur in solid form, are severely lacking in biogenic elements.

This raises an interesting dilemma: the biogenic elements are found mostly in the Sun or the outer solar system, exactly where liquid water can't exist! In fact, this segregation was probably even more severe early on, when the Sun's strong wind cleared the inner solar system of gases and blew away the atmospheres of the terrestrial planets, leaving only exposed rocks. Whether there were some gases still trapped in the interiors of the planets is debated (x4.3.3), but at least in the case of the Earth, the giant impact that created the Moon probably ejected any such gases into space as well. The habitable zone probably started out uninhabitable!

Comets, being very rich in volatiles such as water ice, could have resupplied the inner

59

60 CHAPTER 5. COMETS & ASTEROIDS: CREATORS OR DESTROYERS? solar system with biogenic elements during the era of heavy bombardment. These leftover planetesimals beyond Jupiter's orbit had their orbits disrupted by the gravitational forces of the giant planets. What happened when a comet hit one of the planets depended on the relative masses of the two bodies. A small comet hitting a large planet would inject most of its volatiles into that planet's atmosphere, helping to build it up. On the other hand, a large comet hitting a small planet would produce a huge vapor plume that would carry most of the cometary gases out into space, along with some of the planet's existing atmosphere. Calculations by Chris Chyba have indicated that this process of impact erosion would have depleted the atmosphere of an object smaller than Mars. The Earth, however, was large enough to acquire its oceans and atmosphere by accreting comets.

5.2 Organic Molecules from Space What about reduced organic compounds, which are particularly interesting for the origin of life? The first evidence for an extraterrestrial origin of such organics came from meteorites, rocks which fell to the earth from space--in particular, certain kinds known as carbonaceous chondrites. The most famous example is the Murchison meteorite which fell in Australia on 28 September 1969. It was found to have equal amounts of left-handed and right-handed amino acids, unlike biologically produced amino acids, and its isotope ratios confirmed that the organic material was not terrestrial contamination. Since then, other meteorites have been found to contain amino acids, as well as a variety of other organic compounds (alcohols, aldehydes, hydrocarbons, ketones, carboxylic acids, etc.).

The discovery of many organic compounds in carbonaceous chondrites, combined with detection of some in Halley's Comet (although no amino acids), suggests that a substantial quantity of organics were delivered to Earth during the bombardment era by both comets and asteroids, planetesimals which formed inside Jupiter's orbit and are therefore less icy than comets. However, the principal sources of extraterrestrial organics may have been the interplanetary dust particles (IDP's), tiny grains with masses ranging from 10

\Gamma 12

to

10

\Gamma 6

grams. These particles, mostly evaporated off of comets, constantly rain down onto

Earth and are gently decelerated as they enter the atmosphere, so they do not burn up as meteors. In contrast, while the impacts of asteroids may have contributed much more carbon to the Earth, the huge energies released would have destroyed all but the simplest of organic molecules.

1

Dense interstellar clouds, from which stars and planets form, themselves contain a rich supply of biologically interesting molecules, including water, ammonia (NH

3

), hydrogen

cyanide (HCN), and formaldehyde (H

2

CO, or often ionized as HCO

+

). Such molecules

are detected by the unique radio frequencies they emit, which can be observed using radio telescopes. Much larger molecules, such as polycyclic aromatic hydrocarbons (PAH's), are thought to be responsible for infrared absorption features seen in low-density interstellar gas--one of the most common PAH's is naphthalene (C

10

H

8

), the active ingredient in

1

On the other hand, amino acids found at the K/T boundary layer suggest that the impact energy can

be used to synthesize fresh amino acids.

5.3. PANSPERMIA: ARE WE EXTRATERRESTRIALS? 61 mothballs. Finally, recent radio observations have revealed tentative evidence for glycine (NH

2

CH

2

COOH), the simplest amino acid, and a firm detection of acetic acid (CH

3

COOH),

the main ingredient in vinegar. Since comets and the smaller asteroids formed from gas and dust in the solar nebula that was minimally altered, their organic molecules may have originated, at least in part, from interstellar clouds.

The importance of extraterrestrial sources for organic molecules depends on how reducing the early atmosphere was. In a highly reducing atmosphere, such as the one Miller and Urey simulated in their pioneering experiments, such sources would be negligible compared to the amounts produced by reactions on Earth. In a mostly CO

2

atmosphere, however, the

extraterrestrial contribution can rival the terrestrial one.

5.3 Panspermia: Are We Extraterrestrials? Life appeared early in the history of Earth, indicating that it either arose quickly or originated elsewhere and was brought to Earth by impacting bodies (comets, asteroids, and interplanetary dust). The latter hypothesis is called panspermia, a term coined by Nobel prize-winning chemist Svante Arrhenius at the beginning of the 20th century. Another Nobelist, Francis Crick, has stated that the Earth could have been intentionally seeded with life by an intelligent race, an idea referred to as directed panspermia. Although unpopular for many years, recent evidence for the rapid appearance of life soon after the Earth cooled, perhaps in 100 Myr or less, and the putative Martian microfossils have reawakened interest in panspermia--although it is debatable whether 100 million years was really too short a time for life to begin.

One of the drawbacks of the panspermia idea is that since the Universe had a beginning (the Big Bang), life had to originate somewhere, and it seems that the Earth's surface was as good a place as any for this to happen. Furthermore, organisms would have to survive a long journey through interstellar space, exposed to ultraviolet radiation and cosmic rays, then arrive on Earth at a location suitable for their survival and reproduction. On the other hand, transfer of material between Mars, Venus, and Earth would be much less problematic, and has undoubtedly occurred (as the Martian meteorites tell us). Perhaps life began first on Mars, and we are all Martians! Future exploration of the other planets will allow us to explore this possibility.

Could life have begun on comets? One problem is that life on Earth requires liquid water--there are no known organisms that can survive on pure ice. We do not know if comets possess enough internal heat to have liquid water at their cores; reasonable arguments suggest that radioactive heating could have provided this heating early in the Solar System's history. However, this radioactivity may also have been damaging to fragile organic molecules. If life on Earth really has a cometary origin, it might be impossible to tell from meteorites that land on Earth, because you would not be able to distinguish Earth life from comet-borne life. Only a careful analysis of several comets in space will be able to answer this question.

The question of where life arose has enormous implications for what kinds of life we can expect to find elsewhere in the Universe. If life on Earth was seeded from interstellar space,

62 CHAPTER 5. COMETS & ASTEROIDS: CREATORS OR DESTROYERS? we can expect that extraterrestrial life is common and based on the same biochemistry as ours. Note that this does not imply that other humanoid civilizations exist, contrary to the beliefs of science fiction writers and alien abductees! If, on the other hand, life arose independently on Earth, extraterrestrial life could be substantially different--although we should not be surprised if it is also based on carbon and water.

5.4 Impacts and Mass Extinctions 5.4.1 Impact Frustration of Life? As mentioned in Chapter 4, several extremely devastating impacts probably occurred in the Hadean eon, prior to 3.9 Gyr ago, when the Earth was still accreting debris from the solar nebula. The heat from such impacts would have vaporized the oceans and led to a hot greenhouse atmosphere persisting for thousands of years. For this reason, we usually assume that life originated less than 4 Gyr ago, although it is quite possible that life started several times but was completely wiped out by ocean-vaporizing impacts. The evidence that the last common ancestor to present-day life lived in a hydrothermal environment may indicate that only in such a subsurface environment was life able to survive this early bombardment.

5.4.2 The K/T Event More recently, the mass extinction that occurred between the Cretaceous and Tertiary periods, 65 Myr ago, is now widely believed to have been the result of a giant impact. This idea has a long history, but is now supported by several lines of evidence:

1. An excess of the metal iridium was found in a thin clay layer marking the CretaceousTertiary (K/T) boundary. Iridium is a rare element which is much more common in meteorites than in the Earth's upper mantle or crust. This discovery led Luis and Walter Alvarez to revive the impact hypothesis in 1979. Confirmation of this iridium excess (a factor of up to 1000 times the normal level) has been found in K/T layers around the world.

2. The K/T layers also show particles of shocked quartz and droplets of once-molten

rock, often found near impact craters and thought to result from the intense heat and pressure of the blast.

3. In many locations, graphite (a form of carbon) in the K/T layer suggests widespread

burning of forests. These fires could be ignited by hot debris ejected by the impact and re-entering the atmosphere.

4. A huge (300 km diameter) crater in the Gulf of Mexico, near the tip of the Yucatan

peninsula, has been identified as the likely remnant of the impact. As expected, the thickness of the K/T layer and the amount of shocked quartz becomes greater the closer you get to the crater site (called Chicxulub). Radiometric dating of shocked material

5.4. IMPACTS AND MASS EXTINCTIONS 63

in the crater gives an age of 65 Myr: hence, we seem to have found the "smoking gun." Parts of North America show effects of tidal waves that were kicked up by the impact.

The general scenario then goes as follows: a comet or asteroid about 10 km in diameter hit the earth, throwing a huge quantity of pulverized rock into the atmosphere and causing widespread devastation. The smaller bits of dust, consisting of material from both the impactor and the crust, would remain suspended in the upper levels of the atmosphere (the stratosphere) for months, reducing the amount of sunlight reaching the surface. Global temperatures would have fallen dramatically and photosynthesis, the basis for most food chains, would have been severely curtailed. This "impact winter" is a more severe version of the "nuclear winter" that might follow a full-scale nuclear war.

One can get a simple estimate of the amount of energy released by such an impact using the basic physics formula:

Kinetic energy =

1

2

M v

2

where M is the object's mass and v is its speed. For a 10 km diameter asteroid with a density of 3 g/cm

3

, the total mass of the asteroid is 1:6 \Theta 10

15

kg. Assuming an impact velocity of

30 km/s (the possible range is roughly 11-72 km/s), the total kinetic energy is 7 \Theta 10

30

ergs,

which is equivalent to over 100 million megatons of TNT. This would be a catastrophe far worse than even the largest conceivable exchange of nuclear weapons (a typical H-bomb is equivalent to a few megatons of TNT).

An environmental catastrophe such as the K/T impact would be an "equal opportunity" killer: as Stephen Jay Gould has written, extinction in such circumstances may be a result of bad luck rather than bad genes. Of course, certain organisms would have an advantage--the small size and underground lifestyles of mammals probably allowed them to survive when the dinosaurs were wiped out--but only after the fact. The small size of mammals probably had more to do with their inability to compete with the dinosaurs (they coexisted for over 100 Myr) than any considerable advantage (prior to the impact) to being small.

5.4.3 Other Mass Extinctions Were there other mass extinctions caused by impacts? Here the evidence is less conclusive. The only events since the Cambrian

2

that are universally agreed to have been mass extinctions occurred at the end of the Ordovician, Devonian, Permian, Triassic, and Cretaceous periods (about 440, 365, 250, 215, and 65 Myr ago respectively). Recently, a crater has been found off the southern tip of South America that may be the remnant of an impact at the end of the Permian. Further evidence linking mass extinctions with craters may be found in the coming years.

In addition to the major extinctions, perhaps a dozen or more minor extinction events may have occurred during the past 250 Myr, although it is debatable whether they were mass extinctions, i.e. occurring in a short time around the world. For instance, it is very

2

Mass extinctions prior to the Cambrian are difficult to deduce because the fossil record is so sparse.

64 CHAPTER 5. COMETS & ASTEROIDS: CREATORS OR DESTROYERS? difficult to tell whether extinction of different species occurs simultaneously or over a period of a few million years, since larger animals are generally less common in the fossil record, so their youngest remaining fossils may have preceded their extinction by a substantial period of time. Recently, these minor extinctions have been the focus of considerable controversy.

In 1984, David Raup and John Sepkoski presented evidence that there is a periodicity in the extinctions of families of marine organisms over the past 250 Myr, with mass extinctions happening roughly every 26 Myr. Various astronomical triggers have been suggested, including a faint, distant companion star to the Sun ("Nemesis") and encounters with giant molecular clouds as the Solar System passes through the midplane of the Galaxy every 30 Myr. Both scenarios assume that gravitational encounters will disturb the Oort Cloud of comets surrounding the Solar System and cause comet showers. Unfortunately, there can be severe uncertainties in determining when families go extinct--uncertainties in age dating and in distinguishing extinction from evolutionary change. Thus, it is not universally agreed that the mass extinctions claimed by Raup and Sepkoski are real. It has also been claimed that a similar periodicity exists in the cratering record, which would support the astronomical scenarios, but the ages of most craters are rather poorly known.

The role of impacts in shaping the evolution of life, therefore, remains uncertain. Many if not most mass extinctions may be the result of climate change induced by continental drift or other geological processes. At one extreme is the Shiva hypothesis, in which periodic or occasional impacts control the evolution of life, leading to extinction of some species and subsequent proliferation of others. (Shiva is the Hindu god of destruction and rebirth.) Whether or not they has been a dominant force, it is certainly an interesting possibility that comets have played a dual role in the evolution of life: helping to get life started but occasionally acting to snuff it out.

5.4.4 The Continuing Threat An impact of the magnitude of the K/T event, if it occurred today, would probably destroy human civilization and might well lead to the extinction of our species. It is true that we have the advantage of a more flexible diet than the dinosaurs had, but on the other hand we are so dependent on modern agriculture that most of us wouldn't know how to live without it. Although the chances of a giant impact in the near future are extremely small, it is only a matter of time, and there is certainly good reason to develop the technology to detect and deflect such impactors before they hit.

An interesting point raised by Carl Sagan is that civilizations which never acquire technology will always be vulnerable to extinction by an impact. Thus, there may be a natural limit to the lifespan of an intelligent species that does not wield technology.

Chapter 6 The Living Galaxy? We now turn our attention to the possibility of discovering life outside our Solar System. It may seem surprising that such an important topic is given such limited discussion in this book, but this is mainly an indication of how little we know. Our observations of the universe are limited to stars and other luminous objects; only in the past few years have we found any firm evidence for extrasolar planets (planets outside our Solar System) and we still cannot study these planets directly. But these are exciting times: plans are now on the drawing board for much larger telescopes that will actually be able to see Earth-like planets and determine if they contain oxygen and water in their atmospheres!

6.1 The Drake Equation Our ignorance about the prevalence of extraterrestrial life is neatly summarized in a formula given by Frank Drake in the early days of SETI, the search for extraterrestrial intelligence. The goal was to define the relevant quantities needed to estimate the number of civilizations we could communicate with in our Galaxy. In a statistical sense, one would like to know how many long-lived stars there are, what fraction of those have planets, what fraction of planets have life, what fraction of inhabited planets develop intelligent civilizations, and what fraction of those civilizations want to communicate. This will give us the total number of communicating civilizations that ever arose. Finally we multiply by the ratio of a typical civilization's lifetime to the age of the Galaxy, to get the number of civilizations that exist now. These considerations led to the formulation of what is commonly called the Drake equation:

N

c

= N

\Lambda

\Theta f

p

\Theta n

e

\Theta f

l

\Theta f

i

\Theta f

c

\Theta f

L

where N

c

= the number of communicating civilizations

N

\Lambda

= the number of long-lived stars in the Galaxy

f

p

= the fraction of stars with planets

n

e

= the average number of habitable planets per system

65

66 CHAPTER 6. THE LIVING GALAXY?

f

l

= the fraction of habitable planets on which life arises

f

i

= the fraction of planets with life where intelligence has developed

f

c

= the fraction of intelligent civilizations interested in communicating

f

L

= the fraction of the Galaxy's age that a typical civilization lasts

The Drake equation is best considered a tool for informed speculation; we have no basis for predicting the values for many of these parameters. The only way we could determine their true values would be to take a census of the entire Galaxy, after which there would be no point in using the equation anyway. In this sense, the Drake equation is tautological, rather than prescriptive as the laws of physics are. The first three terms, however, can be determined in principle by astronomers:

ffl N

\Lambda

: A rough estimate of the number of stars in the Galaxy is 300 billion (3 \Theta 10

11

).

In addition, we would like to include in N

\Lambda

all stars that have died already but could

potentially have yielded civilizations. But most of the stars that have died already are either short-lived or lacking in "metals" (an astronomer's term for heavy elements) needed to make planets. As a first approximation, we therefore ignore the dead stars, and consider only stars shining today which have lifetimes of more than about 5 Gyr. This is about half the stars in the Galaxy, so let's take a rough estimate of N

\Lambda

= 10

11

(100 billion).

ffl f

p

: We are starting to do surveys of nearby stars to look for planets, but such surveys

are still in their infancy, so we can only detect large planets relatively close to their stars. As of 1998, the Lick survey led by Geoff Marcy had detected 6 planetary companions around 107 Sun-like stars over a timespan of 11 years, so f

p

? 0:06. Given that their

survey would not have detected Jupiter yet, it is likely that they will find additional stars with planets among this group, so perhaps a more realistic estimate of f

p

is 0.1

or even higher. For instance, roughly half of young stars appear to be surrounded by dusty disks of material.

ffl n

e

: We cannot detect terrestrial planets like the Earth around other stars yet, so we

don't know what the typical number of habitable planets in a system is. We can use our own Solar System as a guide: the CHZ ranges from 0.95-1.15 AU, according to Kasting's models (x2.4), so it is 0.2 AU wide. The typical spacing of the terrestrial planets in our Solar System is 0.4 AU, so if planets are laid down at random with this spacing, roughly half of solar systems should have a planet in the CHZ: n

e

= 0:5. If we

find life outside our CHZ (like on Europa) we will probably have to raise this estimate; on the other hand, if our Solar System turns out to be a freak (as the extrasolar planets found so far suggest) then this estimate may need to be lowered.

Although it is common to plug in some values for the remaining parameters, this is meaningless because you can come up with any answer you want (Jakosky gives 10

\Gamma 13

as

a "pessimistic" value for N

c

and 5 billion as an "optimistic" value). Personally, I am an

optimist when it comes to f

l

, f

c

, and f

L

and a pessimist when it comes to f

i

(the likelihood

of developing intelligence). Using my favorite values, along with the numbers given above, I get N

c

= 5.

6.2. SEARCHING FOR EXTRASOLAR GIANT PLANETS 67 6.2 Searching for Extrasolar Giant Planets Finding extrasolar planets, even giant ones like Jupiter, is extremely difficult because they are so close to their stars and so much fainter. If you could block out the star's light, a planet circling a nearby star would be detectable using the best telescopes, but in practice you must separate the planet's light from the much stronger light of the star. A rough analogy is how you must raise your voice in order to be heard in a noisy restaurant. The Sun is a billion (10

9

) times brighter than Jupiter at visible wavelengths and 10,000 (10

4

) times brighter at

infrared wavelengths. This suggests that the best hope for observing planets directly is to look in the infrared, a strategy that will likely be pursued in the 21st century (x6.3).

As a result of these technical problems, all of the successes achieved so far (aside from a controversial result announced in May 1998, discussed further below) have come from indirect searches, where the planet's presence is deduced from its effects on its star. Two basic methods can be employed, the astrometric and spectroscopic techniques. Both rely on an important principle embodied in Newton's Third Law: the force exerted by the star on the planet equals that exerted by the planet on the star. Thus, although the planet appears to orbit the star, both the star and the planet should be viewed as orbiting a common center of mass that is closer to the more massive body (the star). Jupiter, for instance, causes the Sun to execute a small orbit (roughly the size of its own radius) around their center of mass, with a speed of about 12.5 meters per second. This regular motion of the Sun, having the same period as Jupiter's orbit (about 12 years), betrays the presence of a planet.

1

The period of the planet (P ) can be translated into its average distance from its star (given by the semimajor axis, a) using Kepler's Third Law, a

3

= kP

2

, where k is a constant

that depends primarily on the mass of the star (which can be estimated from knowledge of its spectrum, see x6.2.2). The law predicts that more distant planets take longer to complete an orbit; in fact, a planet twice as far away takes more than twice as long because it moves more slowly in its orbit.

6.2.1 Astrometric Detection The first indirect detection technique, astrometry, involves trying to measure very small angular motions in the sky. It can be used to predict the appearance of the constellations millions of years into the future or the past--the Big Dipper, for instance, used to look more like a spear. Careful astrometry can also reveal the presence of unseen companions to stars: such a companion will cause the star to execute a zig-zag motion through the sky, as its motion around the center of mass is superposed on the motion of the entire system through space. However, the precision required for this kind of observation is very difficult to achieve. In 1963 Peter van de Kamp announced that Barnard's Star showed exactly this kind of wobble, which he attributed to a planet of about 1.6 Jupiter masses, but improved observations failed to confirm this result, and it is now assumed that van de Kamp was misled by small instrumental errors.

1

Saturn causes an additional wobble of the Sun by 2.7 m/s, with the remaining planets having much

smaller effects.

68 CHAPTER 6. THE LIVING GALAXY?

Another possible detection of a planet by this method was announced by George Gatewood in 1996. It was especially exciting because the planet, orbiting the star Lalande 21185, has about the same mass as Jupiter and an orbit roughly the size of Saturn's. However, because the period is so long, it will take many years of observations to confirm this detection.

In principle, the astrometric method is better at detecting giant outer planets than the spectroscopic method discussed below, since the amount of the wobble (size of the star's orbit) is proportional to the size of the planet's orbit for a fixed planet mass. On the other hand, the long periods of such planets (decades) requires careful monitoring of any instrumental changes that can cause errors, as van de Kamp's work illustrates.

6.2.2 Spectroscopic Detection Spectroscopy is the separation of light into its component wavelengths.

2

It works by taking

a glass prism or equivalent dispersing device and passing a star's light through it: the different wavelengths of light are spread out as a rainbow of colors. The rainbow, of course, is seen only for visible light, but spectroscopy can be done with any form of light (radio, infrared, ultraviolet, X-rays, or gamma rays), although the techniques used are different. The dispersed light of a star is called its spectrum, and is valuable because atoms in the cooler outer layers ("atmosphere") of the star will tend to absorb light at particular wavelengths, producing dark stripes across the colors of the spectrum. The pattern of stripes is unique for each element, so in this way the chemical composition of the star's atmosphere can be determined. Moreover, if the star has some motion toward or away from the observer, its light, as well as all of its absorption lines, will be shifted in wavelength toward the blue (if the star is approaching) or the red (if the star is receding) end of the spectrum: this is the well-known Doppler effect. The absorption lines in the spectrum provide sharp features with which one can measure the amount of the Doppler shift, and thus determine the star's velocity along the "line of sight" to the observer.

For a star in a circular orbit around the center of mass, the line-of-sight or radial velocity traces out a sine curve over time, as the star moves towards you for half its orbit and away from you for the other half. The spectroscopic method of planet detection relies on our ability to construct such velocity curves from Doppler shifts in the star's absorption lines measured over months or even years. Velocities of 10 m/s must be reliably measured to detect the equivalent of Jupiter; in practice this requires that errors in the measurement not exceed 3 m/s on average. Such precision has recently been achieved by Paul Butler, Geoff Marcy, and their colleagues using the Lick telescope. Greater precision is possible in principle but would probably not be useful, given the fact that motions of gas on the star's surface can be as high as a few meters per second.

Improved observational techniques over the past decade have allowed this technique to come to fruition: the first confirmed planet around a normal (hydrogen-burning) star, 51 Pegasi, was announced by Michel Mayor and Didier Queloz of the Geneva Observatory in 1995. Since then, roughly twenty additional planets have been discovered. The limited

2

We can characterize light by either frequency or wavelength; the two are interchangeable because their

product is always the speed of light.

6.2. SEARCHING FOR EXTRASOLAR GIANT PLANETS 69 period over which these observations have been going on (about 12 years) means that we are not yet able to detect long-period planets with this technique. The Doppler signal is also strongest for massive planets that orbit quickly, so it is not surprising that most of the planets found so far are giant planets close to their stars. What is surprising is that there exist such planets at all (see x6.2.4).

A fundamental uncertainty of the the Doppler technique results from the fact that you can only measure the velocity along the "line of sight" (towards you or away from you). If you are looking at the orbit almost face-on (see figure), you will measure a very small Doppler shift, even for a large planet, because most of the star's motion is perpendicular to the line of sight and therefore produces no Doppler shift. Thus the signature of a small planet in an edge-on orbit is identical to that of a large planet in a more face-on orbit. The mass that is usually quoted is the minimum mass and would only be correct for an edge-on orbit. Mathematically, the minimum mass is written as M sin i, where M is the true mass and i is the inclination of the orbit, which could range from face-on (i=0

ffi

) to edge-on (i=90

ffi

). The

inclination cannot be determined using the spectroscopic method alone.

i = 80 (Nearly) Edge-on ViewFace-on View i = 0

If all we can measure is the minimum mass, how do we know that the unseen companions discovered by this technique are not dozens of times larger than Jupiter, in which case they are more likely to be failed stars known as brown dwarfs? There are several key arguments, each of which can be used to counter this claim:

1. If most of our planet candidates were brown dwarfs in nearly face-on orbits, then there

should be other brown dwarfs in more edge-on orbits, and thus possessing a larger minimum mass. (Remember the closer the orbit is to being edge-on, the closer the minimum mass is to the actual mass.) In fact, because the same object is easier to detect in an edge-on orbit, we should be biased towards finding companions in such orbits. However, surveys find very few objects with minimum masses in the brown dwarf range (10-70 M

J

, where M

J

is Jupiter's mass), even though such objects would

be much easier to detect than planets! This contradiction implies that few (if any) of the planet candidates can be brown dwarfs with actual masses over 10 M

J

.

70 CHAPTER 6. THE LIVING GALAXY?

2. We can also test the idea that an individual planet candidate (rather than the ensemble

as a whole) could be a brown dwarf in a face-on orbit. To produce a noticeable Doppler signal this companion would have to be quite massive--massive enough to be detected using the astrometric technique. This test has been applied to some of the stars with candidate planets and no telltale zig-zag motion of the star has been detected, placing a limit on how face-on the orbit can be.

3. In some cases we can estimate the true inclination of the orbit. The star's absorption

lines are not only shifted as the star moves around, but are widened by the rotation (spin) of the star: one half of the star is approaching you and the other half is receding from you. The amount of this "line broadening" can be measured, and depends on how fast the star is spinning and the angle of the star's equator to the line of sight. If we assume that the planet's orbit is in the same plane as the star's equator (as is true for our Sun), and have an independent measure of the star's rotation speed (e.g. by monitoring flux variability), then we can estimate the angle of the orbit (or the star's equator) to our line of sight.

6.2.3 Other Techniques Another suggested method for planet detection is photometry, which involves looking for a periodic dimming of the star's light as the planet passes in front of (eclipses) it. It requires a nearly edge-on orbit, which will occur less than 4% of the time. However, it would be more direct proof of a planet's existence than provided by the astrometric or spectroscopic methods, and if combined with the spectroscopic method would yield the planet's exact mass (since you know the orbit's orientation) and size (from the duration of the eclipse), and therefore its density, which would tell you if the planet is rocky or gaseous.

Finally, we return to the question of whether planets can be observed directly. In principle, we already have the capability to take pictures of young giant planets, since they still emit a great deal of infrared light as they radiate away the energy of their formation. In May 1998 Susan Terebey and collaborators claimed such a discovery using the Hubble Space Telescope. The candidate planet was not next to a star (it was purportedly ejected from a nearby stellar system), hence the usual problem of distinguishing its light from its star's light was not a concern. However, the whole interpretation depends on knowing the distance to the "planet," since a faint object may be intrinsically faint (hence a planet) or only appear faint because it is very distant. Terebey argues that the object is nearby because a bright filament seems to connect it with a young star known to be quite close to us. But it is not clear why an ejected planet should leave behind such a filament; it is quite possible that this association is just a coincidence. A spectrum of the object should be able to settle the issue by telling us whether it shows the spectral features of a star or a planet.

6.2.4 Properties of the Known Extrasolar Planets With a growing sample of extrasolar planets (over 20 in April 1999, including a 3-planet system orbiting Upsilon Andromedae), some interesting trends have started to emerge. In

6.3. SEARCHING FOR EARTH-LIKE PLANETS AND LIFE 71 particular, many of the planets have turned out to have completely different properties from what we'd expect based on our Solar System.

1. There is a large group of planets, dubbed hot Jupiters, that have masses comparable

to Jupiter but orbit their stars more closely than Mercury orbits the Sun. This flies in the face of the traditional scenario for giant planet formation (x4.3.1), which relies on the extra material in the outer Solar System provided by ice. It is not surprising that hot Jupiters were among the first planets detected, since their short periods and high masses make them easy to find, but it is surprising that Nature has found a way to make such planets: perhaps they migrate in from larger distances. Furthermore, there don't seem to be many "warm Jupiters" at about 1 AU (the Earth's distance from the Sun), even though they too would have been detected.

2. For several extrasolar planets, the line-of-sight velocity does not vary like a sine curve

as expected for circular orbits. These planets are in highly eccentric (oval) orbits: for instance, the distance from 16 Cygni B to its planet varies from 0.5 to 2.7 AU, giving it an eccentricity (on a scale of 0 to 1) of 0.69. In our Solar System, the orbits of the planets have eccentricities of 0.2 or less, with the exception of Pluto, which many astronomers no longer consider a genuine planet. Note, however, that eccentric planets are easier to detect by the Doppler method, given the same semimajor axis, so we should be careful about making generalizations.

3. As mentioned earlier, there is a notable lack of planets with masses greater than about

10 Jupiter masses and orbital periods less than a dozen years. Supergiant planets or "brown dwarfs" can be easily detected by present surveys, so their absence is significant.

4. The stars with detected planets seem to have above average quantities of heavy elements

(carbon, oxygen, iron, and so on). A reasonable explanation is that the presence of more heavy elements allows planets to grow bigger, increasing the detection rate.

Various theories have been proposed to explain the first two properties, but it's been difficult to account for the features of both our Solar System and the others within a single theoretical framework. Is our Solar System then a freak? Not necessarily: perhaps the "hot Jupiters" are formed by a different process than "cold Jupiters." Bear in mind that while these "oddball" planets are the easiest to detect, they have only been found orbiting !5% of Sun-like stars. As it becomes possible in coming years to detect solar systems resembling our own, we may discover that our Solar System is the norm rather than the exception.

6.3 Searching for Earth-like Planets and Life While the techniques discussed above may be useful for finding giant planets, completely different techniques will probably be needed to detect terrestrial (Earth-like) planets.

72 CHAPTER 6. THE LIVING GALAXY? 6.3.1 Pulsars With Planets Surprisingly, the first extrasolar planets detected were not much bigger than the Earth! They orbit a pulsar, which is a rapidly rotating stellar remnant (a neutron star) that appears to emit regular pulses of light, just as a rotating lighthouse beacon appears to flash on and off. The planets were discovered in 1992 by Alex Wolszczan and colleagues around the pulsar PSR 1257+12, and have minimum masses of 2.8 and 3.4 Earth masses; there is also possible evidence for a third planet the size of the Moon. Their detection relied on the ability to determine the arrival times of radio pulses to accuracies of a few tens of microseconds: as the pulsar is pulled toward us by the planet, the time between pulses shrinks, while as the pulsar moves away from us the time between pulses grows. A second pulsar, PSR 0329+54, has a companion of at least 2 Earth masses. This may be the most sensitive indirect method of detecting planets.

Pulsars, however, are not likely to support habitable planets. They emit huge amounts of high-energy radiation, and moreover the supernova explosion that produces a neutron star is sterilizing to all planets in the system--and would probably destroy the smaller planets. Many astronomers think the pulsar planets formed after the explosion, out of the gas ejected by the star. If we're interested in finding habitable terrestrial planets, we should be looking around normal, long-lived stars like the Sun.

6.3.2 Direct Detection The traditional indirect techniques, using astrometry or spectroscopy to observe the star, would not be adequate for finding Earths. Even with perfect instruments, we would be limited by the instability of stars themselves: motions of gas on the stellar surface would swamp any Doppler shift. Direct detection, which requires only that we separate the light of the planet from the light of the star, seems to be the best way to proceed. Resolved imaging--actually producing maps of the surface of the planet--would be much more difficult, requiring telescopes or arrays of telescopes the size of the Moon.

Perhaps the best way to directly detect Earth-like planets is with a nulling interferometer. There are two basic techniques involved. First, interferometry allows two telescopes that are separated by a distance D to achieve much higher resolution (image sharpness) than they could individually; in effect, mimicking a single telescope of diameter D that would be impractical to build. High image resolution is key because at too low resolution, the planet's light will be blended into the star's. The second technique is the use of nulling, which involves combining the light rays from the telescopes in such a way that all light from a known location in the sky (in this case, the star) is precisely cancelled.

The interferometer would be designed to work at mid-infrared (mid-IR) wavelengths around 10 microns (10 _m, where 1 _m = 10

\Gamma 6

m), in order to minimize the contrast

between the planet and star and to allow spectroscopic study of the planet's atmosphere. Some of the most interesting atmospheric gases from a biological standpoint, including CO

2

,

H

2

O, O

3

, and CH

4

, have absorption lines around these wavelengths (see x6.3.3).

However, the technical challenges are still enormous. In order to observe at mid-IR

6.3. SEARCHING FOR EARTH-LIKE PLANETS AND LIFE 73 wavelengths, the whole system will have to be put in space, preferably in the outer Solar System where there is less IR radiation from interplanetary dust. The planet-star contrast, although much lower than at visible wavelengths, is still about 10 million (10

7

) for terrestrial

planets and 10

4

for giant planets. More than two telescopes (probably at least four) would

be needed for accurate nulling, and the distances between them should be adjustable and go up to at least 75 meters. Most likely this would require placing them on different spacecraft and flying them in formation, with their relative positions monitored to within fractions of a micron. Such a space-based nulling interferometer, called the Terrestrial Planet Finder (TPF), is being designed by Roger Angel and Neville Woolf at the University of Arizona, and could be built within 15-20 years assuming the necessary technology is developed.

6.3.3 Finding Evidence for Life At present, the only way to look for extraterrestrial life beyond our Solar System is to listen for intelligent signals. Note, however, that life has existed for at least 80% of the Earth's history, while radio signals have only been sent out for a hundred millionth (10

\Gamma 8

) of that

time, and it's unclear how much longer we'll be doing it. A more promising strategy may be to look for evidence of O

2

in the atmosphere: this has been produced by life in large

quantities for about half the planet's history. In fact, there is a strong absorption line for ozone (O

3

) in the mid-infrared spectrum at 10 _m wavelength, which can be used as a

sensitive tracer for O

2

since O

3

is made from O

2

. Just as spectroscopy can be used to find

out what stars are made of, it can be used to determine what's in a planet's atmosphere.

But would oxygen alone be an indicator of extraterrestrial life? Although neither Venus nor Mars has much atmospheric oxygen, this may be partly a result of their individual histories. A non-biological O

2

atmosphere might occur, for instance, as a result of ultraviolet

light breaking water into H

2

and O

2

. The hydrogen would escape into space but the heavier

oxygen might remain in the atmosphere. Venus has very little free O

2

because it emits

volcanic gases that combine readily with it, but a smaller Venus with less volcanism might have detectable amounts of O

2

. Thus, a more convincing indication of life would be detection

of oxygen in the presence of liquid water, since water greatly accelerates rusting processes that remove oxygen. Liquid water could be inferred by detecting water vapor absorption lines (strong around 7 and 20 _m) and showing the planet is in the habitable zone of its star.

Perhaps the best evidence for life would be the simultaneous detection of the 10 _m ozone line and the 7.65 _m line of methane (CH

4

), since oxygen and methane cannot coexist for long

periods of time. This is exactly the kind of thermodynamic disequilibrium we expect from life: on Earth, methane and oxygen are constantly being pumped into the atmosphere by bacteria and plants. However, the methane absorption line is weak and probably undetectable outside the Solar System using the telescopes we can currently conceive of building. Thus, the only practical strategy at present seems to be searching for ozone and water vapor.

74 CHAPTER 6. THE LIVING GALAXY? 6.4 Interstellar Travel Even if we didn't see evidence of oxygen in a planet's atmosphere, it would be impossible to rule out life entirely: perhaps it doesn't rely on oxygen-yielding photosynthesis to the extent that life on Earth does. The only way to know for sure whether there's non-communicating life on extrasolar planets is to go out and visit them.

People have long dreamed of travelling to the stars, but for now it remains a distant goal. The problem is not just building a sufficiently powerful engine. A conventional rocket engine, burning continuously, could accelerate a spacecraft to very high speeds given enough time. The main problem is that there are no gas stations in space! If you had to lift off with all the fuel needed for the entire trip, you'd never get off the ground in a conventional rocket. (There's no way you could scrounge up that much rocket fuel on Earth anyway.)

If interstellar travel is ever to become practical, it must be possible to reach average speeds of at least 0.1c (where c is the speed of light) without requiring a prohibitively large amount of fuel. Such speeds would allow us to travel to nearby stars in about 50-100 years. Even that seems like a long time; several ways have been suggested to make this more tolerable for humans, and have been toyed with by science fiction writers:

ffl Huge "world ships" could be built containing hundreds of people, living out their lives

and procreating in the darkness of interstellar space. Although an ocean cruise that lasts your entire life may not sound like much fun, the passengers might be able to pass the time living in virtual reality.

ffl We could develop a method of placing the crew in "suspended animation" until they

arrive at their destination, refreshed and ready for exploration. Although this is an enticing idea, we are nowhere near being able to implement it: the limitations are physiological as well as technological.

From experience, we expect that the first missions will be robotic probes like the Voyagers. This would give us an indication of the unforeseen dangers and allow us to economize on fuel by sending a much smaller payload. Even a robotic mission would have to be carefully targeted--we would probably want firm evidence that there are planets orbiting the star.

6.4.1 Propulsion Strategies The question remains: how could speeds of even 0.1c be achieved, given that our fastest spaceships travel at only 0.00003c? If you must bring all your fuel with you, lightweight, energy-efficient fuels are the best way to go. Since mass can be converted to energy and vice versa (E = mc

2

), the most efficient engine would be able to turn all of the fuel's

mass into energy--in other words, it would be powered by matter-antimatter annihilation. Antimatter would be very expensive to create, though, and storing it safely is beyond our present capability.

6.4. INTERSTELLAR TRAVEL 75

A more practical energy source is nuclear fusion, since the basic fuel would be hydrogen, the most common element in the Universe. However, fusion is much less efficient, with a little under 1% of the rest mass being converted into energy, and the fusion of pure hydrogen as occurs in stars would be difficult to achieve. Some success at controlling other fusion reactions, however, is at least conceivable in the near future, unlike antimatter engines.

Another option is not to rely on fuel at all; have a laser beam from Earth (or somewhere else in the Solar System) push you into space. The spacecraft would then be designed as a giant light sail. The amount of energy needed to push even a small payload, however, would be enormous, several times larger than the annual world energy production. Given that most of our energy sources are non-renewable, it is probably not a good idea to waste them flying interstellar kites. The laser would therefore have to be powered by sunlight.

3

Also, the sail would have to be durable enough to withstand the long journey, and there needs to be a way to decelerate the spacecraft once it reaches its destination, unless you're willing to settle for a flyby.

Finally, you could try to use the extremely rarefied gas between stars as fuel. An interstellar ramjet would have a huge frontal scoop that gathers hydrogen atoms from interstellar space and feeds them into a fusion reactor to generate thrust. The scoop would have to be enormous, about the size of the Earth for an acceleration of 10 m s

\Gamma 2

(Earth

gravity). This effectively rules out a solid scoop, so a "virtual" scoop composed of magnetic fields seems to be necessary. Critics have pointed out, however, that magnetic fields are rather poor at collecting particles, and a huge amount of energy would need to be expended to simply maintain the field.

6.4.2 The Destiny of Civilizations The technical problems with interstellar travel, although daunting to us in the 20th century, might not seem so huge in the 30th century, and might well be solved by the time the 100th century rolls around, if not sooner. Yet this would still be an insignificant amount of time compared to the age of the Galaxy or Solar System. Some people have argued, therefore, that even civilizations marginally more advanced than we are should have mastered interstellar travel. Traveling at a tenth the speed of light, the entire Galaxy could be explored in a few million years. So where are they? Why aren't they here now?

This question, known as Fermi's paradox, is especially puzzling given what we know of our own civilization: it has experienced explosive growth in both population and per capita energy consumption. These pressures may ultimately compel us to colonize the Solar System and then other planetary systems, quite apart from any interest in science or tourism. Even if "mature" civilizations are able to rein in their appetite for resources, they face the eventual prospect of their star using up its hydrogen fuel and dying. Better to find some habitable planets elsewhere, and start looking as soon as possible in case they're hard to find.

We discuss possible solutions to Fermi's paradox in x7.3.4. One explanation is that advanced civilizations are rare, either because intelligence is difficult to evolve or civilizations

3

Sunlight itself could not be used to push the sail because the Sun's rays are constantly diverging, so they

are too weak to accelerate spacecraft at great distances.

76 CHAPTER 6. THE LIVING GALAXY? destroy themselves quickly. If this is true, and humans manage to avoid self-destruction (that's a big if), it may be that our descendants will eventually colonize much of the Galaxy. Such colonies would be too far apart to allow easy travel, and would probably evolve as independent societies, and eventually independent species (remember that new species arise by geographic isolation!).

One wonders whether these species would be friendly to each other, or whether tribalism will prevail as it has so many times in human history. Perhaps the interstellar warfare found in science fiction will be waged in the real universe someday by our descendants, as they compete for dwindling energy sources in the Galaxy. It's hard to imagine running out of energy in the whole Galaxy, but then again a few years ago it was hard to imagine needing a gigabyte of disk space.

Chapter 7 Intelligent Civilizations A fundamental question in the search for extraterrestrial life is the prevalence of intelligent life. But what do we mean by intelligence? It's often said that before we go looking for extraterrestrial intelligence we should look for evidence of intelligent life on Earth! Of course, when we speak of intelligence we are generally referring to the cognitive capacities that seem unique to our species, including the following:

1. The ability to even ask questions like What is intelligence? (abstract thinking). 2. The ability to modify our environment voluntarily (technology). 3. The ability to plan for unexpected or hypothetical circumstances (foresight). 4. The ability to use symbolic language. 5. The ability to adopt another's point of view, the basis for morality. 6. The sense of identity and control over our actions (self-consciousness).

Although this list is by no means exhaustive, and several of these abilities may be dependent on each other (for instance, it is hard to conceive of abstraction without symbolic language, or vice versa), they illustrate some of the expectations we might have for intelligent life elsewhere, based solely on our own example. Perhaps not all intelligent species will possess all of these features, but they should possess several of them, in order for us to have any hope of recognizing them as intelligent.

In this chapter I begin by giving an overview of human evolution, with an emphasis on the view that symbolic language is the key ability that separates us from the rest of the living world. I will then switch gears to discuss the search for extraterrestrial intelligence--how it is being pursued, and how it has been defended in the face of strong criticism.

77

78 CHAPTER 7. INTELLIGENT CIVILIZATIONS 7.1 Human Evolution

Our understanding of human evolution is based on the sciences of paleontology, the study of fossils, archaeology, the study of ancient human artifacts, and anthropology, the study of human societies. Recently, the ability to compare DNA sequences extracted from modern human populations and fossils has added a new dimension to this study, that of molecular genetics. New insights are also coming from researchers who study psychology and linguistics.

7.1.1 The Likelihood of Intelligence Was the emergence of human-level intelligence a likely development? The answer to this question is the key to assessing the likelihood of extraterrestrial intelligence. Most evolutionary biologists, from Alfred Russell Wallace to Ernst Mayr, have tended to see the evolution of humans as a vastly improbable event. Mayr goes further to argue that intelligence of any sort is hugely improbable: of the billions of species that have existed on Earth, only one has created a civilization, whereas other useful attributes such as flight, eyes, and the ability to electrocute prey have evolved several times independently. Moreover, while life itself emerged quite rapidly (in 500 Myr or less), it took over 3.5 billion years to produce an intelligent species, and the Earth will remain habitable for only another 1-3 billion years, according to recent estimates. From our self-centered perspective, we often see ourselves as the goal of evolution. But dim-witted creatures like plants and beetles are far more pervasive than we are, while the bacteria still account for virtually all of life's diversity.

Indeed, we have paid a heavy evolutionary price for our abilities. Our large brains require a huge amount of energy (20% of the entire body's) to keep running, and complex thermoregulation systems to keep from overheating. Our ability to speak requires that the larynx be low in the throat, hence if we try to breathe and eat at the same time we choke (babies don't develop this problem until about the age of 2, before which they can neither choke nor speak). The large size of our skulls requires us to be born early in a helpless state, and even so childbirth is painful and risky for the mother. Our long lifetimes are balanced by a strikingly long period of childhood. Thus, it may not be surprising that no evolutionary lineage has followed the path of massive brain growth to the extent that ours has.

On the other hand, we see that over life's history some animals have taken the route of increasing behavioral complexity, since life, having started out simple, naturally expands its range of complexity. In addition, "arms races" between predators and prey have often led to bigger brains in each just to keep up with the other. Therefore, we might expect that many planets with life, given enough time, will evolve lifeforms that display complex behaviors. Perhaps intelligence emerged just once because evolution needed time to achieve a proper starting point for its development, and the first intelligent species may prevent the emergence of intelligence in other lineages. In other words, if the dinosaurs had developed a civilization that survived, we might not be here to marvel at the fact.

7.1. HUMAN EVOLUTION 79 7.1.2 Pre-modern Humans DNA sequencing shows that the living species which are phylogenetically closest to modern humans are the chimpanzees (both common and pygmy species), which differ from us in about 1.6 percent of their DNA. In fact, chimps are closer to humans than they are to either gorillas (2.3 percent difference) or monkeys (7 percent). Assuming a relatively constant rate of mutations, the last common ancestor (LCA) of humans and chimpanzees lived about 6-8 Myr ago. This was a hugely controversial result when it was first announced thirty years ago by UC Berkeley biochemists Allan Wilson and Vincent Sarich (they used antibody reactions rather than DNA itself and got an age of 5 Myr), but has been confirmed by more sophisticated techniques.

To see how the evolution toward humans progressed, we must rely on a very incomplete fossil record. "Lucy," the oldest fairly complete hominid (ape that walked on two legs) skeleton, was discovered in 1974 in the Hadar area of Ethiopia, and dated to roughly 3.2-3.4 Myr ago. Her species has been assigned the name Australopithecus afarensis. An even older set of hominid footprints dated to 3.6 Myr ago has been found in Tanzania, presumably also due to A. afarensis. The emergence of bipedal locomotion--note that chimps can move short distances on two feet but are more comfortable walking on feet and knuckles--was a crucial development in human evolution, freeing the hands for manipulation of tools and allowing for a much more varied lifestyle.

The immediate reason for adopting bipedal locomotion is unclear, but is probably related to a shift in ecological habitat from dense forests to more open grassy woodlands in East Africa. This was due largely to the opening of the Great Rift Valley by tectonic forces, which may have also formed a physical barrier between the hominids and their ancestors, allowing them to evolve in different directions. The hominids spent less time in trees and more time traveling between isolated patches of forest, and while they were not better runners than many four-legged animals, they were clearly superior to other apes in this regard.

The successor to the australopithecines was Homo habilis, the first hominid species classified under our genus and the first that carries an archaeological record of crafted stone tools. Originating about 2-2.5 Myr ago, H. habilis clearly had intelligence beyond today's chimpanzees, as evidenced by the deliberate methods required to produce stone flakes and the fact they they often transported their tools great distances. The toolmakers were also predominantly right-handed, pointing to a reorganization of the brain relative to apes, and had a more slender body structure which was better suited to running. The brain volume, about 700 cm

3

, was significantly larger than it was in the australopithecines (400-500 cm

3

).

Approximately 1.5 Myr ago, a new species, Homo erectus, emerged, accompanied by a more sophisticated repertoire of stone tools, including carefully fashioned hand axes. H. erectus was the first hominid to migrate out of Africa, occupying parts of Europe and Asia by about 1 Myr ago. Over time, brain size increased further (from about 900 cm

3

in early

fossils to over 1000 cm

3

) and geographically separated populations developed some noticeable physical differences, leading to "Peking Man" in East Asia and "Neanderthal Man" in Western Europe. Around 400 kyr ago some of these groups began to routinely use fire. Collectively, these groups are often referred to as "archaic Homo sapiens," but they are still anatomically distinct from modern humans.

80 CHAPTER 7. INTELLIGENT CIVILIZATIONS

It should be noted that the exact shape of the hominid family tree is still a subject of great controversy--especially the question of how many species coexisted and for how long. The sequence of H. habilis ) H. erectus ) H. sapiens reflects what appear to be the most prevalent species in each of their respective eras, but should not be taken to indicate direct evolutionary descent.

7.1.3 Emergence of Modern Humans There are two competing ideas for the origin of anatomically modern humans. The multiregional evolution hypothesis posits that H. erectus populations in Africa, Europe, and Asia evolved in concert towards H. sapiens. In this model, the growth of the brain was driven by cultural or linguistic development, and enough interbreeding between populations occurred to keep the species intact. Anatomical continuity between features of Peking Man and those in present-day Asian populations has been cited in support of this model, although that evidence has been disputed.

The alternative "Out of Africa" hypothesis states that Homo sapiens was not the descendant of H. erectus around the world, but represents a second wave of expansion out of Africa beginning around 100 kyr ago. Modern humans completely replaced archaic species, and racial differences emerged relatively recently. The evidence for this point of view is now fairly compelling:

1. In Europe, there is no gradual transition from Neanderthals to modern humans. In

fact, the two groups overlap in time, with moderns appearing in the fossil record about 40 kyr ago and the last traces of Neanderthals around 27 kyr ago. When DNA was extracted from a Neanderthal fossil in 1997, it was found to differ significantly from modern humans, further demonstrating that Neanderthals could not have been our immediate ancestors. No "hybrid" fossils exist which would indicate that the two groups were capable of interbreeding.

2. If modern humans had co-evolved from archaic ones, they should all have appeared at

roughly the same time. In fact, the fossil evidence indicates that they appeared first in Africa about 120 kyr ago, spreading to the Middle East around 100 kyr ago and to Asia and Europe much later.

3. To end up with the genetic similarity between modern humans requires, in the multiregional model, a huge amount of gene flow (interbreeding) between widely separated populations. This is difficult to achieve using realistic models for migration and population genetics.

4. The mitochondrial DNA (mtDNA), thought to be inherited only from the mother,

can be traced back in time using assumed mutation rates to indicate the date and location of the LCA. When applied to people today, researchers find that all human mtDNA is extremely similar, indicating an LCA in Africa between 100 and 200 kyr ago. Although all humans can trace their mtDNA from this individual (the so-called mitochondrial Eve), this does not mean that our nuclear DNA can be traced back

7.2. LANGUAGE AND EVOLUTION OF THE BRAIN 81

to a single couple--Eve was probably one of many individuals we are descended from, all of whom contributed to our nuclear DNA.

1

Thus, the weight of the evidence now seems to favor a relatively recent origin (120- 200 kyr ago) for Homo sapiens in Africa. It is notable that the appearance of modern humans in Europe (about 40 kyr ago) coincides with a major advance not only in the level of technology but in the rate at which it changes. Prior to H. sapiens, each new hominid species had an improved toolkit over the previous one, but there had been little change in these tools once they were established. Our species showed the first indications of regional differences in technology--a strong indicator of culture--and the first appearance of art objects, cave painting, and burial goods. These achievements have been referred to as the Upper Paleolithic Revolution or the "Great Leap Forward," as there is no precedent for them in the archaelogical record. It is still debatable whether these signals constitute a true revolution or whether earlier artifacts were made of materials that have not been preserved, but there is no question that these people were really us. The rest of the story--from the invention of agriculture 10,000 years ago and writing 5000 years ago to the technological civilization of today--is merely an extension of that cultural awakening.

7.2 Language and Evolution of the Brain The two strongest trends that are seen in human evolution are an increase in brain size, which occurred fairly continuously over the past two million years, and an increase in the pace of technological and cultural evolution, most of which appears to have occurred only in the past 40,000 years. In fact, the burst in creative output that occurred in our species followed a leveling off of brain size: our brains are actually a bit smaller on average (1350 cm

3

) than the Neanderthals' were (1550 cm

3

). How can these two very different rates of

change be explained?

Maybe technology does not correlate with brain size because brain size is irrelevant. After all, people with big heads are not necessarily smarter than those with small ones (despite the pejorative term "pinhead"). However, the growth in brain size is too dramatic to be dismissed, especially when we consider that the part of the brain which grew the most was the prefrontal cortex at the front of the brain, which is involved in most "higher" brain functions. In order to outweigh the many disadvantages of large brains, there had to be some increase in cognitive ability going on, and it must have had tremendous survival value. Some of the more plausible suggestions have included the demands of social living, which require constant scheming against rivals, the need to adopt more flexible lifestyles during the Ice Ages, and the development of language. Although any or all of these hypotheses could be correct, I focus here on the idea that language was the driving force in the evolution of the brain, as has been argued by Derek Bickerton, Terrence Deacon, and others.

1

We get our mtDNA from just one of our grandparents (as with our surnames) but our nuclear DNA

from all four.

82 CHAPTER 7. INTELLIGENT CIVILIZATIONS 7.2.1 The Nature of Language If language has been the main driver behind human evolution, then it must be far more ancient than agriculture, the wheel, and other cultural artifacts. But at least superficially, languages seem to differ markedly from one culture to the other. In English, word order is generally used to indicate the function of a word in a sentence (he gave me the book, not he gave the book me). Latin and German make much heavier use of word endings, allowing word order to be used for emphasis instead. Chinese has no word endings, even for singular and plural, relying entirely on auxiliary words and context.

Yet despite these differences, children seem to have an innate ability to acquire any language, and not simply by trial and error like they learn other skills. Noam Chomsky pointed out many years ago that the difficulties a child faces in learning a language are so monumental--especially in the absence of formal training in grammar or even correction by adults--that the feat could not be accomplished without some innate "Universal Grammar" giving children prior knowledge of how language works. Steven Pinker has elaborated on this idea with his notion of a "language instinct" that has been honed by natural selection. The underlying commonalities among grammars of unrelated languages, the spontaneous creation by children of new languages such as creoles, the rapidity with which children acquire language, and the fact that even "primitive" cultures have languages every bit as complex as "advanced" ones argues strongly for a biological basis for language, independent of culture.

There are three fundamental properties of language that distinguish it from other forms of natural communication. The first is symbolic reference, the ability of words to refer to each other rather than to environmental stimuli. The second is syntax, the ability of words to convey different information depending on how they are combined. The third is the fact that language can be employed in virtually any context, and is often used to transmit information. In contrast, animal calls are generally intended to express the caller's emotional state or to influence the behavior of others. In this sense, they are more analogous to expressions of human emotion (laughter, crying, clenched fists, etc.) than true language. Animal calls are also discrete: their meanings remain the same if they are used in combination, and they are often repetitive and contagious within a group, like laughter is for us. But repeating word-for-word what someone else says is merely annoying.

Symbolic reference is a more sophisticated way of defining words than indexical reference, which depends on a correlation in time and place between the word and its meaning. If you say "dinnertime" before feeding your dog, this word will be associated with food in the dog's mind. However, if you stop using "dinnertime" in this context and start using it when you go out for the evening, your dog will learn to interpret the word differently. On the other hand, a symbolic reference preserves its meaning irrespective of whether there is any physical association. In the story of the Boy Who Cried Wolf, the "wolf" cry was an indexical reference for danger that gradually faded for the townspeople, but the meaning of the word "wolf" itself did not fade. In fact, we have words for objects such as unicorns and quarks that have never been reinforced by everyday experience!

The persistence of symbolic words derives from the fact that their meanings are defined in terms of other words. How else can concepts like "nothing" be understood? The brain

7.2. LANGUAGE AND EVOLUTION OF THE BRAIN 83 is constantly forging connections between these symbols, sometimes horizontal (between equivalent terms) and sometimes vertical (between subsets and supersets). The cognitive "leap" that occurs when we move from memorizing to understanding something can be thought of as forging new symbolic links between mental concepts.

The difficulty animals have with human language is that they focus on indexical associations between words and actual objects or events rather than symbolic associations between words themselves. Consider the fact that many words in English have multiple meanings, such as converse and life, and so can only be understood in the context of other words. How strange this would seem to any other animal! Furthermore, we invent words such as fruit that are generalizations of other words such as banana, apple, and watermelon. In acquiring language, we must often suppress the relationship between a word and its external object and focus on "higher-order" relationships between the word and other words. One result is that we can place words into mental categories (fruits, countries, universities) so that recalling one item will immediately bring to mind similar items that are completely unassociated in time and space. In contrast, animal calls are separate entities that bear no systematic relationships to one another.

So what good is symbolic language? One clear advantage is the ability to free oneself from the constraints of time and space imposed by indexical reference. Not only can we talk about things outside of immediate experience, but we can think about such things because language gives us the mental symbols to construct alternative realities. Another benefit is the ability to convey information in great detail, particularly once syntax is available to explain "who did what to whom." A burst of laughter indicates that something is funny, but not what. Language allows us to frame our thoughts and then share them with others. On the other hand, a certain number of symbols must be acquired before the relationships between them become apparent. The "instant gratification" provided by indexical relationships makes that kind of learning much more natural for animals: there are few contexts, aside from artificially designed experiments, for them to acquire and use symbols.

7.2.2 How Did Languange Evolve? If language is indeed a biological adaptation, it must have evolved by a process of natural selection. Presumably this was a step-by-step process, with each refinement conferring some reproductive advantage to those who possessed it. Unfortunately, it is difficult to trace this evolutionary process, as there is no "simple" human language in existence today. We have created simple languages to interact with computers, but these are designed for very limited worlds, while any human language must be able to deal with the complex world which humans live in and create. In the preceding sentence roughly half the words are grammatical in function (to, with, but, while, etc.) and this is true to varying degrees in any language.

Is there a simpler version of human language in the animal world? Although no natural form of animal communication appears to use symbolic reference, with some effort we can teach our closest relatives (chimpanzees) to interpret and use simple sign language or geometric symbols on cards. One subject, Kanzi, was exposed to language at a particularly young age and was later able to understand basic English syntax. Thus, it seems that the

84 CHAPTER 7. INTELLIGENT CIVILIZATIONS mental capacity for language existed in our primate ancestors, and the subsequent development of the brain focused on being able to "grease the wheels," so to speak: to make the use of symbolic reference more natural and effortless. Significantly, it has been impossible to teach apes to use symbolic language vocally. This fact, along with the fact that the human "body language" of gestures still exists alongside spoken language, indicates that symbolic language was a supplement to animal communication, not a replacement for it.

In the presence of cultural or environmental pressures favoring the use of symbols, a tendency for symbolic learning could eventually be encoded genetically by what has been called "genetic assimilation" or the Baldwin effect, the ability to develop more efficient anatomy for pre-existing behaviors. Unlike the discredited theory of Lamarckism, where helpful genetic traits emerge by an individual's striving, this effect is still governed by natural selection over many generations: individuals with the right mutations have more offspring. A simple example is lactose intolerance. Most mammals lose the ability to digest lactose once they are weaned, since it isn't needed. But adult humans descended from societies that have a long tradition of herding animals have evolved to digest lactose, since milk could serve as a major source of nutrition for their ancestors. Similarly, capacities like language which are strongly favored by natural selection can lead to physiological changes (descent of the larynx and growth of the prefrontal cortex) and behavioral changes (the language instinct) if and when the right mutations occur.

The initial reason for adopting symbols is unclear, but Terrence Deacon suggests that the emergence of monogamy in the context of group living, not communicative utility, may hold the key: symbols were invented as a medium for social contracts (promises made in public), the most important of which, to this day, is marriage. Hunting expeditions could be coordinated quite well without symbolic language, as lions demonstrate, but a promise is a statement about future behavior that can only be made symbolically. In this rather unorthodox view, the need for mothers to have access to a reliable supply of meat required pair bonding plus communal living, but for such an arrangement to remain stable, mating had to be enshrined as a symbolic relationship. All of this may have happened long before spoken language appeared.

The first stage of language may have been what Derek Bickerton calls protolanguage, the "language" of trained apes, children under two, and pidgin speakers.

2

Protolanguage is

like language without formal grammatical structures: More cookie, Want Mommy read, Go open door. If you've ever tried to survive in a foreign country on a Berlitz phrasebook, you've probably produced it. But protolanguage still employs symbolic words, and in conjunction with gestures it is possible to express simple ideas.

In Bickerton's theory, the first two million years of human evolution were driven by improvements in protolanguage, whose use was limited to simple expressions but still provided important advantages in negotiating social interactions and communicating about food sources. Such utterances might have included dead mammoth, no water there, don't touch her, and so forth. The stunning advances that begin with our species were driven by the

2

Pidgin is used for communication by adults from different cultures who live together without a common

language. One historical example occurred in Hawaii in the late 19th century, when sugar plantation workers were brought in from various parts of Asia.

7.2. LANGUAGE AND EVOLUTION OF THE BRAIN 85 acquisition of syntax, which enabled logical reasoning and foresight (If I were to place a barb on my spear, then even if the beast jumped about, the spear would remain lodged in its flesh, and eventually it would bleed to death) and therefore a rapid pace of cultural evolution. Bear in mind that innovation and creativity depend on the ability to simulate what might happen in your mind, and then turn imagination into reality--trial and error are simply too inefficient. The absence of innovation prior to our species suggests that our predecessors were incapable of stringing together thoughts into logical sentences: in other words, they lacked syntax.

7.2.3 Consciousness Explained? What about that other mystery of the human mind, our sense of self-awareness? Bickerton has some interesting ideas about this problem as well. First, he notes that consciousness probably exists in varying degrees throughout the animal kingdom. Like other creatures, we use data from our sense organs to construct a mental representation of the world. Some animals process this data more than others: your dog can recognize you by sight or smell, whereas your hamster probably can't. The continuous monitoring of the external environment (and one's own body) can be thought of as the first level of consciousness. The part of the brain used for real-time processing of sensory information is called the primary representation system (PRS) by Bickerton.

Humans have a secondary representation system (SRS) in addition to the PRS which other animals have. Like cats, we have a bunch of neurons that fires when we see a dog, allowing us to recognize it as a dog. But we also have a bunch of neurons that fires whenever we think of a dog, or hear the word "dog", and calls up a somewhat abstracted concept of "dog" (some of these may also fire when we see a dog, but they don't have to). This concept of dogness must be abstract because there is no way your brain could store every little detail of every dog you've ever seen. Thus, one of the basic functions of the secondary representation system is to store and recall abstracted versions of things we have seen, allowing us to think about them even when not directly confronted by them.

The SRS, being disconnected from direct sensory input, also provides a second level of consciousness by making us aware of the first level. As Bickerton writes, "the creation of human consciousness depended on the creation of areas of the brain from which the workings of animal consciousness could be objectively perceived." Artificial intelligence pioneer Marvin Minsky makes a similar point: "Divide the brain into two parts, A and B. Connect the A-brain's inputs and outputs to the real world--so it can sense what happens there. But don't connect the B-brain to the outer world at all: instead, connect it so that the A-brain is the B-brain's world."

3

Here the A-brain contains the PRS, and the B-brain

contains the SRS.

Why would natural selection favor the development of consciousness? The awareness of being aware does not offer any immediate survival advantage--more likely than not, it could lead to fatal hesitation. As the example above illustrates, the most likely answer is language, or even protolanguage (which lacks syntax but contains symbolic words). Again,

3

Marvin Minsky, The Society of Mind, New York: Simon & Schuster (1985), p. 58.

86 CHAPTER 7. INTELLIGENT CIVILIZATIONS the distinguishing characteristic of symbolic reference is that it does not rely on external reinforcement. In order for hominid brains to communicate about things outside of their immediate situation, they needed a place to store that information, which the SRS provides. Although original purpose of the SRS was probably linguistic, as a by-product it led to our unique experience of consciousness. It also leads to our sense of agency ("free will"), in that our behavior is influenced not just by external forces, but by our own symbolic representations.

In arguing that other animals have access to only their senses, however, Bickerton may be going too far. We certainly get the impression that animals such as dogs experience emotional states similar to our own: are we to conclude that they are not really "aware" of those emotions? Chimpanzees have been observed to engage in tactical deception, in which there is clearly an ability to anticipate the behavior of others. How can this be done without a sense of one's own motives? They are also able to recognize their own reflections in a mirror: does that not indicate a notion of self? It therefore seems questionable whether consciousness can really be divided into discrete levels, or whether a more continuous description is appropriate.

7.2.4 Conclusions So what was the purpose of this rather long digression into language origins? It is simply to point out that the search for how we became who we are must involve an understanding of how the huge gap between animal communication and language was bridged by our ancestors. In the picture I've described, protolanguage and consciousness began to develop in early hominids, driving brain growth by the Baldwin effect, and a recent reorganization of the brain (without further growth) enabled syntax, giving our species foresight and logical reasoning. On the other hand, many anthropologists prefer to interpret the achievements of the Upper Paleolithic as indicating a sudden and late development of spoken language rather than a long period of language development. With very little historical evidence to guide us, the choice among these or other alternatives may rely mostly on philosophical preference.

How often might this have occurred elsewhere in the Universe? Given our present knowledge, it is difficult to say how many steps are needed to achieve our level of intelligence, or what environmental conditions are needed to favor those steps. But these are questions we should be able to answer in principle, at least with regards to our own history. Even a satisfactory theory of human evolution, however, might have limited applicability to the rest of the Universe. When we are done theorizing, perhaps we should just go out and do the experiment.

7.3 SETI: The Search for Extraterrestrial Intelligence Scientists involved in SETI have always had to struggle with outsiders' skepticism for their research. This is probably true to some extent for any endeavor where the chances of success appear slim, but the situation has been exacerbated by the association of SETI (in the minds of many) with science fiction and UFO's. A number of scientists have derided SETI programs

7.3. SETI: THE SEARCH FOR EXTRATERRESTRIAL INTELLIGENCE 87 as a waste of time and money, and politicians who control research appropriations bills have mocked it as "The Great Martian Chase," among other things. It is conventional wisdom in astronomy departments that SETI is a pastime for tenured professors, not something a graduate student should risk her career on.

7.3.1 The Importance of Longevity The Drake equation is not quite as meaningless as suggested in x6.1. There is one important constraint that can be deduced as follows. Choose the most optimistic numbers imaginable: N

\Lambda

=10

11

, f

p

=0.5, n

e

=0.5, f

l

=0.5, f

i

=1, f

c

=1, f

L

=L/(12 Gyr), where L is the average

lifetime of a communicating civilization. If you multiply the numbers out, you find that

N

c

= N

\Lambda

f

p

n

e

f

l

f

i

f

c

f

L

ss L ;

in other words, the number of civilizations in the Galaxy is equal to the average lifetime of a civilization in years. Since this is an extremely optimistic case, we take it as an upper limit: N

c

^ L.

This poses an interesting problem for communication. If the lifetime of civilizations is short, say 100 yr, then there are at most about 100 civilizations in the Galaxy, and they are probably spaced very far apart from each other. The time it would take for a radio message to go from one civilization to another would exceed the lifetimes of both! For instance, the center of our Galaxy is 30,000 light years away, so it would take 60,000 years to receive a reply to our transmission. In such a situation, interstellar conversations would be impossible, even at the speed of light.

One can actually work this out mathematically with a crude model. Suppose the Galaxy is a cube 100,000 light-years on a side (a light-year is the distance traveled by light in one year). Then the volume of the cube is 10

15

cubic light years. If there are N =L civilizations

(best-case scenario), then if they are more or less evenly spaced they should each have a subcube of about 10

15

=L cubic light years to themselves. The distance between civilizations

will then be roughly the length of a side of a subcube, or 10

5

=L

1=3

light years. In order to

have even a single exchange of messages, this distance must be less than L=2 light years, so L should be at least 10,000 years. In other words, if the average lifetime of a civilization is much less than 10,000 years, then chances are that no two-way communication will be possible. We may still be able to have a conversation with an unusually long-lived civilization, but we too would have to be unusually long-lived.

Of course, it may not be possible to speak of an "average" lifetime for civilizations. Our civilization has only been capable of communication for about 50 years, and it's likely that unless we solve our problems (overpopulation, pollution, global warming, nuclear proliferation, and so forth) we will last only another century or two. However, if we do solve these problems, our lifetime might be much longer. So it may be typical that many civilizations go through a period of adolescence, in which they must learn to live with their technology or perish, and those which survive live much longer. In that case, a "typical" lifetime for a civilization might be either 100 or 1 billion years, and a straight average would not be very informative.

88 CHAPTER 7. INTELLIGENT CIVILIZATIONS 7.3.2 Strategies for Communication A fundamental premise of SETI is that most civilizations will be much more advanced than we are. This follows from the fact that our civilization is extremely young, and if most civilizations only lived to about our age, there would be too few out there to give us much hope for contact. Since our civilization has only recently acquired the technology for radio communication, we expect that other civilizations will have developed such techniques to a much higher degree than we have. Thus, we are better off listening for transmissions from other, more advanced civilizations than trying to initiate a two-way conversation. Conversations are also impractical given the huge distances involved and the finite speed of light, and may well be impossible if the average lifetime of civilizations are only a few thousand years. Finally, there is the worry that other civilizations may not harbor good intentions--after all, the most expansionist societies will be the most likely to intercept our signals.

Nonetheless, we have sent out a few intentional transmissions, in addition to the stray "leakage" radiation that has continually left the Earth's surface since the dawn of television some 50 years ago. In 1974 the largest radio telescope in the world, the 300-meter Arecibo telescope, beamed a message containing 1679 bits of information toward the globular cluster M13, about 25,000 light-years away. An intelligent civilization, recognizing that 1679 is the product of two prime numbers (23 \Theta 73), would find that the message encodes a blackand-white image of 23 columns and 73 rows, with some important information about DNA, humans, the Earth, and the Arecibo telescope, if they are clever enough to decode it.

Much more slower-moving messages are carried by the Pioneer 10 and Voyager 1 and 2 spacecraft that explored the outer solar system are now heading out into interstellar space. Pioneer carries an engraved metal plaque while the Voyagers carry gold-plated phonograph records. These spacecraft would take about 100,000 years to reach the nearest star, although none of them are even headed towards nearby stars.

When it comes to receiving signals, there are a number of strategies that can be pursued.

ffl We can listen for "leakage" radiation similar to what we put out daily, or we can search

for intentionally beamed signals. The latter will be stronger and thus detectable from greater distances, but there may not be many civilizations that are intentionally trying to communicate (maybe they're all listening like we are).

ffl We can search a number of nearby stars that we think are more likely to host planets

with life, or we can search blindly across the heavens. The former is called a targeted search and the latter an all-sky survey.

In fact the two sets of strategies are linked: leakage radiation is best sought using a targeted search, while intentionally beamed signals are best sought using an all-sky survey. In order to understand why, we need to understand two important scientific principles:

1. The strength of an electromagnetic signal decreases as the square of the distance from

the emitter. This is because as the signal moves away from the source, its energy is spread out over a sphere of surface area 4ssd

2

, where d is the distance from the source.

7.3. SETI: THE SEARCH FOR EXTRATERRESTRIAL INTELLIGENCE 89

So if you move twice as far away from the source, the signal is four times weaker. This inverse square law for light propagation means that we can only detect strong signals from distant transmitters, but we can detect both strong and weak signals from nearby transmitters.

2. In any receiver there is a certain amount of "noise" resulting from the electronics and

from background radiation emitted by the earth and atmosphere. This noise gets added to the signal we are trying to detect. The signal-to-noise ratio (SNR), which is a measure of how easily the desired signal can be separated from the background noise, increases as the square-root of the integration time:

4

SNR / t

1=2

:

Thus a fourfold improvement in the SNR requires 16 times as much integration time (i.e. the observation must last 16 times longer).

Using these principles, it is clear that intrinsically weak emission such as leakage radiation can only be detected from nearby sources, and only by spending a long time looking in the right direction. This is the motivation for targeting a few nearby Sun-like stars. On the other hand, powerful beacons, although presumably rare, might be observable in a relatively short integration time, allowing us to cover much more of the sky. Given our ignorance of where they might be, the best way to search for such beacons would be an all-sky survey, possibly biased towards the Galactic plane where there are more stars.

In addition to deciding where to look, we also have to decide at what frequency or wavelength. (Recall that the frequency and wavelength of light are interchangeable concepts, being related by the formula ** = c, where * is the wavelength and * is the frequency.) The electromagnetic spectrum is a way of ordering the different types of light by wavelength: in order of increasing wavelength, they are radio, infrared, optical, ultraviolet, X-rays, and gamma rays. Different instruments and techniques are needed to detect each of these. Among the considerations involved in choosing an observing frequency for SETI are the following:

ffl Astronomical background radiation. No matter what direction you look in, a

certain amount of radiation is emitted by astronomical objects--stars, active galactic nuclei, interstellar dust, ionized interstellar gas, etc. There is also the leftover radiation from the Big Bang, now redshifted into the microwave part of the spectrum. These "background" radiations are a major subject for astronomical study, but they can also overwhelm the weak signals being searched for.

ffl Interstellar absorption. Signals can be attenuated, and brief pulses broadened in

time, by gas and dust between the stars. These effects vary strongly with frequency; in general, there is more absorption at higher frequencies until the ultraviolet, beyond which the absorption goes down again.

4

This is because averaging the signal leaves it unchanged, assuming it has constant amplitude, while

averaging the noise reduces its amplitude by

p

N , where N is the number of samples averaged.

90 CHAPTER 7. INTELLIGENT CIVILIZATIONS

ffl Atmospheric absorption. All frequencies above that of visible light are blocked

by the atmosphere, and infrared radiation is strongly attenuated by water vapor and oxygen. Hence only optical and radio transmissions or searches are possible from the ground. Of course, this is only a consideration for the transmitters if they are signaling from the surface of a planet or are interested in communicating with planet-bound civilizations.

ffl Cost-effectiveness. This is a consideration both to the party producing the signal

and the party trying to detect it. In general, since higher frequency photons have higher energies (from Planck's Law, E = h*), it is cheapest to produce radio photons. However, there are other factors as well, including the ease with which the signal can be amplified and the narrowness of the light beam that can be achieved.

ffl Terrestrial sources. Earth-based SETI must continually deal with man-made signals

which can mimic interstellar communications. Unfortunately this is most severe in the radio part of the spectrum, and the situation in worsening. Light pollution is also severe at optical wavelengths, although there are still places on Earth that get quite dark.

Most of these considerations point to using radio waves, particularly with wavelengths close to the 21-cm line of neutral hydrogen, to search for signals. The use of optical or infrared light might be cost-effective if very narrow beams could be produced (such as with a laser), and might even be advantageous because more information could be transmitted in the same amount of time. However, such a beam would have to be intentionally directed towards us if we are to have much of a shot at receiving it. Radio waves diverge much more than lasers do, and therefore any radio signal sent in our general direction will reach us. Also, we can search for radio signals both day and night, and the light from any parent star close to the transmitter would be relatively faint at radio wavelengths.

Since natural phenomena tend to produce radio signals with large bandwidths (about 1000 Hz or greater), an artificial signal could be best identified as one having a narrow bandwidth, i.e. employing only a small range of frequencies, perhaps only about 1-10 Hz wide. Such signals require less energy to produce anyway. Some kind of pulsation in the signal might also be expected, although most searches are designed to detect either a pulsed or continuous-wave signal. In order to detect narrow-bandwidth signals, we must first divide the spectrum into very narrow frequency channels. But since we don't know in advance what frequency will be used, we literally need to search through millions or billions of these channels, a challenge that modern electronics can now deal with. The BETA system at Harvard, for example, can search through 80 million channels at once!

In this remainder of this section, we will briefly examine the tumultuous history of SETI, as well as the arguments posed by some of its detractors.

7.3.3 Historical Development The idea of communicating with extraterrestrial civilizations across vast distances is not a new one: radio pioneers Nikola Tesla (1856-1943) and Guglielmo Marconi (1874-1937) both

7.3. SETI: THE SEARCH FOR EXTRATERRESTRIAL INTELLIGENCE 91 claimed to have received unexplained signals that they attributed to alien intelligence. In 1924, astronomer David P. Todd (1855-1939) even convinced the U.S. military to observe a period of radio silence on August 22 and 23, when Mars was at its closest approach to Earth, so he could listen for radio signals. However, the technology for receiving and sending radio signals over interstellar distances did not fully emerge until radio technology came of age after World War II.

Modern approaches to SETI can be traced back to two pioneering studies dating from 1959, one theoretical and the other observational. The theoretical study was published by two physicists at Cornell University, Giuseppe Cocconi and Philip Morrison. Cocconi had been studying gamma rays that were emitted by the Cornell synchrotron, and it occurred to him that this radiation could be produced artificially and used to communicate across interstellar distances. He and Morrison discussed the idea, and ultimately decided that the radio part of the spectrum would be more appropriate for practical communication: there was less natural interference and radio waves were easier to make. In particular, they suggested that the emission line of neutral hydrogen, at a wavelength of 21 cm, would be a good place to look, since hydrogen is the dominant constituent of the Universe and any civilization doing radio astronomy would have built receivers at this wavelength early on. The paper suggested a program to observe some of the nearest Sun-like stars, and ended with the aphorism that SETI proponents continue to live by today: "The probability of success is difficult to estimate; but if we never search, the chance of success is zero."

5

That very same year, Frank Drake, a young astronomer at the National Radio Astronomy Observatory in Green Bank, WV, pointed out that the new 85-foot telescope being completed there, when outfitted with the latest receiver technology, could detect radio signals comparable to those produced by humans out to a distance of 10 light years--more than twice the distance to the nearest star. Luckily for Drake, his bosses were supportive of the idea of searching for signals, and Project Ozma was born. A dual-use receiver was built that could be used for astronomical studies of the 21 cm line as well as SETI, circumventing any criticism of the project on scientific grounds; in fact, only about $2000 was spent on parts unique to Ozma. The two nearest solar-type stars, o/ Ceti and ffl Eridani, were observed for over 200 hours starting in April 1960; no signal of extraterrestrial origin was detected. Still, news of the project sparked the popular imagination.

The following year, an informal conference on interstellar communication was organized at Green Bank, attended by Cocconi, Morrison, and Drake, as well as a handful of other scientists including Carl Sagan, then a young assistant professor at Harvard. It was at this meeting that Drake first presented the equation that would later bear his name (see x6.1), as a way of organizing discussion of the parameters relevant to assessing the likelihood of SETI's success. Based on what they considered their best guesses for the values of these parameters, the participants concluded that the number of communicating civilizations in our Galaxy could range from less than 1000 to 1 billion. (These numbers are now considered a bit too optimistic, even by SETI proponents. But of course nobody knows.)

Meanwhile, scientists in the Soviet Union were also taking an interest in SETI. One of the 5

G. Cocconi and P. Morrison, "Searching for Interstellar Communications," Nature, 184 (September

1959), 844.

92 CHAPTER 7. INTELLIGENT CIVILIZATIONS leading proponents of SETI there was Iosif S. Shklovskii, who with Carl Sagan wrote what would be the definitive work on the subject for many years, Intelligent Life in the Universe. A former student of his, Nicolai Kardashev, went so far as to propose a classification scheme for technological civilizations:

ffl Type I: a civilization with technology and energy consumption comparable to present

human civilization.

ffl Type II: a civilization capable of harnessing most or all of the energy from its star,

e.g. with a "Dyson sphere" surrounding the star.

ffl Type III: a civilization capable of harnessing the energy of its entire galaxy.

7.3.4 The Fermi Paradox: Where Are They? In 1975, two articles appeared in the scientific literature which cast serious doubt on the entire SETI enterprise. The authors, Michael Hart and David Viewing, argued independently that the absence of extraterrestrials on Earth strongly suggested that other advanced civilizations are rare. (Like most scientists, they did not give credence to claims that the Earth is presently being visited--see x7.4.) In essence they revived an argument that the great physicist Enrico Fermi had posed casually over lunch in 1950: if there are extraterrestrials, where are they-- i.e., why aren't they here now?

Hart's argument is disarmingly simple: if interstellar travel is feasible at speeds of 0:1c, where c is the speed of light (he argues that nuclear propulsion could achieve this in principle) then visits to nearby stars could be made in a few decades. Once colonies were established, they could send out there own expeditions, and so on. The limits of colonization would then be a sphere expanding out from the home planet at roughly 0:1c. This sphere would encompass the Galaxy in less than a million years--a negligible amount of time compared to the age of the Galaxy. Unless by some remarkable coincidence all other technological civilizations are roughly as young as we are, we should have been visited by now.

Frank Tipler has gone on to argue that even if the rigors of space travel were considered too costly by a civilization, it could send out robotic probes capable of traveling to nearby star systems and utilizing available resources to build copies of themselves, which could then be sent on to other systems. Such "universal constructors" or von Neumann probes, even traveling at 10

\Gamma 4

c (the speed of the Voyager spacecraft), could explore the entire Galaxy

in 300 million years at very little cost--since only a few probes need to be made at the beginning--communicating the results of their explorations to the home planet. Probes, he argues, are a superior means of initiating contact with other civilizations, since there is no ambiguity about if and how other civilizations communicate, what frequency they use, etc.

The interstellar travel/colonization argument has been met by SETI proponents with a variety of responses:

1. Technical limitations: Perhaps there are unseen technical problems that discourage

or prevent space travel at speeds of 0:1c, or that complicate the construction of von

7.3. SETI: THE SEARCH FOR EXTRATERRESTRIAL INTELLIGENCE 93

Neumann probes. Given the progress we have made in 50 years, it would be surprising if these problems were insurmountable over thousands of years. Still, in terms of costeffectiveness, photons are a lot easier to send than spaceships.

2. Control Hypothesis: Perhaps civilizations do not build von Neumann machines

because they are afraid that they will lose control of them. After all, such "universal constructors" might go out of control and turn all the solid material in the Galaxy into von Neumann machines. If such machines are possible, this explanation requires that no civilization be willing to take the risk of building them.

3. Contemplation Hypothesis: Perhaps the aliens are simply not interested in space

travel, and would rather sit at home. However, this would have to be the case for every civilization in the Galaxy at all times after they reach technological maturity. In particular, there are at least three rather persuasive reasons to colonize other worlds: growth in population as mortality declines, the need for additional energy resources, and the aging of the home star. It would seem that any civilization capable of interstellar travel would have reason to attempt it, if only to stake out worlds for later settlement.

4. Zoo Hypothesis: Perhaps the aliens or their probes are here, but are intentionally

keeping us ignorant of their presence. Again, this would have to apply to all civilizations that have visited our Solar System. Furthermore, it is not clear why advanced civilizations would want us to be unaware of their existence, and whether all parties would be willing to participate in such a deception.

5. Temporal explanations: Perhaps they have not had time to reach us, or perhaps

they have come already and left without a trace. The first explanation could work if, for instance, civilizations expanded at a much slower rate than they could--only moving outward when the youngest colonies became overcrowded, and not sending out von Neumann probes. The second explanation is impossible to test, but again forces us to assume that no civilization that passed through bothered to stick around.

In summary, there are three general solutions to the Fermi paradox: (1) there are severe technical limitations to space travel and the creation of von Neumann machines, (2) all civilizations that could have visited or colonized us have chosen not to, or (3) long-lived technological civilizations are rare in the Galaxy.

7.3.5 Rise and Fall of the NASA SETI Program The arguments put forward by Hart, Tipler, and others helped to convince Shklovskii, the father of Soviet SETI, to abandon the search. However, activity in the U.S. actually increased in subsequent years. Philip Morrison and others argued that theorizing was no substitute for actually doing the observations, echoing the closing mantra of the Cocconi & Morrison paper. In that spirit, ongoing SETI programs were launched at Ohio State University in 1973 and Harvard in 1983, the latter funded in part by the Planetary Society. A UC Berkeley project

94 CHAPTER 7. INTELLIGENT CIVILIZATIONS called SERENDIP (Search for Extraterrestrial Emission from Nearby Developed Intelligent Populations) began operation in "piggyback" mode on various telescopes--meaning that the part of the sky searched depends on whatever direction the telescope happens to be pointed at for some unrelated science project.

Although NASA was associated mainly with spaceflight, there had been enough interest in SETI by some NASA scientists to undertake a design study of possible projects in 1971. The result, known as "Project Cyclops," was led by Bernard Oliver and envisioned an enormous array of roughly 1000 telescopes dedicated to SETI. Oliver advocated searching in a wavelength range between the H and OH emission lines (at around 21 cm and 18 cm respectively), a region he dubbed the "water hole" since H+OH=H

2

O. Not only are problems

of interference reduced in this region, but water seems essential for life and the symbolic importance of this region should be apparent to other civilizations as well.

Cyclops was too ambitious for the political and economic realities of the day, but by 1979, when NASA sponsored a major conference on "Life in the Universe," there was significant support for a SETI program that would apply the latest detector technology to existing telescopes. There would be two components to the program: an all-sky survey at frequencies of 1-10 GHz (wavelengths of 3 to 30 cm), to be directed by NASA's JPL facility at Caltech, and a targeted search of about 1000 stars within roughly 100 light-years at frequencies of 1-3 GHz, to be directed by NASA's Ames Research Center in Mountain View, CA. The project, called the High Resolution Microwave Survey (HRMS), was launched on 12 October 1992 to mark the 500th anniversary of Columbus' arrival in the New World. The Arecibo dish launched the Ames Targeted Search, while a 34-m antenna in the Mojave Desert launched the JPL All-Sky Survey. Both projects were expected to take about a decade to complete.

The HRMS was to be short lived. In a shocking turn of events for the SETI community, the program was singled out by Congress for termination in September 1993, after 15 years of planning, $60 million of research and development spending, and less than 1 year of observations. Nevada Senator Richard Bryan dubbed it "The Great Martian Chase" and introduced a successful amendment to NASA's appropriations bill that ended all NASA funding for SETI. Many of the researchers that had been supported by the project organized the nonprofit SETI Institute in Mountain View and began seeking private funding for a scaled-down version of the Targeted Search. Aptly named "Project Phoenix," the SETI Institute's program, led by Jill Tarter, has rented time on a number of major telescopes, including the 47-m telescope in Green Bank, WV and the Arecibo dish, while sharing in the cost of a new array of small radio dishes being constructed in Hat Creek, CA with the UC Berkeley Radio Astronomy Laboratory. Meanwhile, the university-sponsored SETI programs continue at Berkeley and elsewhere.

7.3.6 Consequences of Contact Skepticism towards SETI is based not only on the perceived chances of success, but on the question of what would be gained by establishing contact. Our technological civilization is so young that any civilization we contact in the foreseeable future is likely to be far more

7.4. UFO'S AND ALIEN VISITS TO EARTH 95 advanced than we are. Would we really be able to learn anything useful from them? Our experience on Earth is not encouraging: whenever an advanced culture has come into contact with a primitive culture, the result has generally been the destruction of the primitive culture (and sometimes its people as well), regardless of the intentions of the advanced one.

6

If we

enter into contact a more advanced race, will we really want to become more like them-- adopting their science, art, political and ethical systems?

On the other hand, the simple knowledge that we are not alone would be an important discovery in and of itself. It would put to rest forever the notion that we are somehow special or unique in the Universe. Moreover, the existence of even simple life outside our Solar System will be very difficult to confirm with confidence by telescopic observations. Thus, it could be that our first knowledge of any kind of extraterrestrial life will occur via SETI. In time, we might discover how similar (or different) biological evolution is on different planets, even if for cultural reasons it proves impossible to ever have a meaningful dialogue with the extraterrestrials.

7.4 UFO's and Alien Visits to Earth The assumption of most SETI researchers is that interstellar communication would occur primarily via radio waves, since actual interstellar travel would be much more expensive, in both time and energy. Yet the idea that civilizations would meet and physically interact was continually raised by SETI skeptics (x7.3.4) and would remain a principal theme in science fiction, perhaps catering to the deep-seated human need for personal contact. Witness the enormous popularity of Star Trek, in which an intrepid crew of explorers journey through the Galaxy on a ship equipped for faster-than-light ("warp") travel, with its stated goal being to "seek out new worlds and civilizations."

But when does science fiction become science fact? Flying saucers and unidentified flying objects (UFO's) represent one of the most persistent areas of controversy between scientists and the public. Ironically, while the prevalence of ETI became a popular idea among many scientists, their skepticism toward UFO claims has become stauncher than ever. As the arguments surrounding Fermi's paradox indicate, the possibility of extraterrestrials visiting the Earth, at present or at some point in the past, cannot be dismissed. Why, after substantial research in the 1960's by the Air Force and the National Academy of Sciences, is there no longer any serious scientific exploration of this possibility?

The lack of incontrovertible physical evidence is probably the main reason, even though scientific prejudice has certainly played a role. No extraterrestrial device or spacecraft has ever been recovered. Conspiracy theorists often accuse the government of covering up the evidence--this seems implausible given the difficulty of covering up anything, especially something as momentous as extraterrestrial visitation, for a significant amount of time. A reliable photograph has still not been obtained, in spite of the huge number of photographs of Earth that have been taken from space over the years. We are generally left with eyewitness

6

Of course, if there is no physical contact but only radio communication, these problems might be avoided,

or at least forestalled.

96 CHAPTER 7. INTELLIGENT CIVILIZATIONS testimony, which as anyone knows is the most unreliable kind of evidence there is. Many hoaxes have been uncovered, which add to the disrepute of the UFO case.

Among the natural phenomena that are often reported as UFO's are the planet Venus, meteors or space junk breaking up in the atmosphere, military aircraft, and weather balloons. Although UFO proponents admit that most UFO's are misidentified natural phenomena, they argue that there is a significant set of sightings that cannot be explained. Still, the lack of an explanation does not indicate an extraterrestrial origin--not all crimes are solved, after all. In fact, statistical tests have shown that the sets of explained and unexplained UFO's are indistinguishable when it comes to duration of occurrence, time of day, number of witnesses, etc. Furthermore, the percentage of unexplained UFO's stays roughly the same, even when the number of reports changes dramatically. If there are eight times as many spaceships visiting this year than last, why are there eight times as many mistaken sightings of Venus, etc.?

In summary, we can't rule out the possibility that a small number of UFO's are alien spacecraft, but in the absence of any physical evidence it seems much more plausible to attribute them to hoaxes, hallucinations, and natural phenomena, as the majority of UFO's can be identified with. As Robert Sheaffer as written, "The burden of proof always rests on the person who asserts the existence of an anomaly."

7

One final thought: if aliens wanted

to study the Earth without being detected, why would they enter the atmosphere at all? One can make extremely detailed observations even from orbit. On the other hand, if they wanted to communicate with us, why do it in the middle of rural America? Why not drop in on Manhattan or the Super Bowl?

7

Robert Sheaffer, "An Examination of Claims that Extraterrestrial Visitors to Earth are Being Observed,"

in Extraterrestrials: Where Are They?, 2nd ed., ed. Ben Zuckerman and Michael H. Hart (1995), p. 20.

Chapter 8 Life: Nature's Destiny? The growth in complexity of the Universe over time, as the result of the organization of matter into stars, planets, and ultimately life, seems to demand an explanation. The Universe started out smooth and featureless, but, governed by the laws of physics, it has gradually evolved away from that state. We are starting to learn how it did this, but what about the question of why? Is our Universe special?

Of course, there are many deep issues here, such as why it is even possible to understand the Universe and describe it with laws that have a simple mathematical form. This is not the appropriate venue for such a discussion. What I want to examine in this final chapter is the question of whether the Universe is in some sense "fine-tuned" for our existence. This could result from either a self-selection effect (the so-called anthropic principle) or from intentional design.

8.1 The Anthropic Principle The basic idea of the anthropic principle is that the existence of intelligent life in some way defines the properties of the Universe. As John Barrow and Frank Tipler write in their treatise, The Anthropic Cosmological Principle, "The measured values of many cosmological and physical quantities that define our Universe are circumscribed by the necessity that we observe them from a site where conditions are appropriate for the occurrence of biological evolution and at a cosmic epoch exceeding the astrophysical and biological timescales required for the development of life-supporting environments and biochemistry."

At a trivial level, we understand that the reason we find ourselves on a planet rather than floating in interstellar space is that life must originate on planets. A similar question is why the Earth's orbit has the right size to allow liquid water at its surface. Most scientists would say that this is simply a coincidence of the Solar System's formation. The reason we happen to live on such a planet is that life requires such a planet in order to develop.

A more serious application of the anthropic principle has been to explain the fundamental constants of nature, such as those labeled by physicists as c (the speed of light), h (Planck's constant), G (the gravitational constant), and e (the unit of electric charge). It turns out that

97

98 CHAPTER 8. LIFE: NATURE'S DESTINY? tweaking these constants even a little would have quite detrimental effects on the structure of matter. A fine balance between nuclear and electromagnetic forces, for instance, makes possible the production of carbon in stars. Since we have no theory to explain the values of these constants, the suggestion is that we live in "the best of all possible worlds": our Universe must have been extremely lucky.

The anthropic principle explains this observation as self-selection. One way that selfselection could work is if there are many (perhaps infinitely many) parallel universes, or isolated parts of a single Universe, each with a different set of physical laws and fundamental constants. In such a many-worlds interpretation, most universes are barren--devoid of stars, planets, and life--but those with just the right combination of laws and constants allow the development of life. As living organisms, we would automatically find ourselves in such a "lucky" universe. Of course, there is the thorny issue of what it means for other universes to exist at all, or how we could verify this.

The problem with anthropic explanations is that it is difficult to define the circumstances under which they apply. In principle you could look at any property of the natural world and say, "well, it just had to be that way or we wouldn't exist." An alternative is to say that maybe the constants and laws are the way they are because of underlying physical principles we just haven't discovered yet. In other words, we may be wrong to think that all possible values for the constants are permissible. In fact, many apparent "coincidences" noted by ancient peoples, such as the regularity of the Moon's phases and the seasons, now have simple physical explanations.

8.2 Intelligent Design of the Universe In contrast to the many-worlds interpretation, in which the apparent design of our Universe is a result of self-selection, the theistic principle argues that there is just one Universe, and its suitability for life is a result of intelligent design by a Creator. Naturally, this interpretation is favored by theologians, most of whom would argue that God constructed the Universe with us in mind.

Clearly, it is not possible to disprove the existence of a Creator. Any feature of the Universe, in principle, could be explained as a product of intelligent design, including the constants and laws of physics. Even if we do find more fundamental laws of physics that can explain the cosmic "coincidences," we would still wonder where those laws came from (irrespective of whether that is a meaningful question). Many scientists reconcile their scientific and religious beliefs by assuming that their job is to uncover the laws of nature and their consequences, not to explain how the laws came about.

On the other hand, there have been recent attempts to prove the existence of a Creator by demonstrating that features of the Universe cannot have been created except by intelligent design. This, I think, is a bit dangerous, because it suggests that we should stop looking for scientific answers to some of the most compelling mysteries in science today: the origin of life, the nature of mind, and so forth. Here are some of the so-called design arguments, and why I think they are misguided:

8.2. INTELLIGENT DESIGN OF THE UNIVERSE 99

ffl The cosmological coincidences. The suitability of the Universe for life as we know

it

1

can be explained by a many-worlds interpretation and the anthropic principle, by

more fundamental (and as yet undiscovered) laws, or by a cosmic Designer. The idea of a Designer does have a certain appeal, since we routinely design things ourselves for specific purposes. But the coincidences do not prove the existence of a Designer, they only suggest the possibility.

ffl Life as a violation of the 2nd Law. Some have claimed that the maintenance

of order in living systems violates the 2nd Law of Thermodynamics, which requires that entropy not decrease in closed systems. As we've seen, this is just plain wrong: organisms are open systems, exchanging matter and energy with their environment. The entropy of the combined Earth-Sun system is increasing as the Sun burns its nuclear fuel.

ffl The origin of life. As we discussed in Chapter 4, the information content of a

bacterium is far too large to be created spontaneously by a chance event. This has been used to argue that life could not have arisen except by design. The assumption, of course, is that no form of life simpler than a bacterium is possible. We don't yet know if this is true. It is also possible that the origin of life was not entirely a chance event, i.e. that some kind of self-organization process was at work, as Stuart Kauffman suggests. I referred to this as "loading the dice" in x4.4.

ffl Complex adaptations. The biochemist Michael Behe writes in Darwin's Black Box

that many biological systems, from the organismic to the molecular level, are irreducibly complex: they require all their parts to function properly. Among the examples he cites are cilia and flagella, which cells use to move around, and the process of blood clotting. Such systems, he argues, cannot be created by natural selection, which depends on gradual improvements on existing forms, since an incomplete system would be non-functional.

Admittedly, these arguments can be quite compelling. Behe makes the valid point that although we may not be able to verify the existence of a Designer experimentally, we also cannot verify what mutations took place to produce some biological form. In the end, the best we can hope for is a plausible story to get from where we started (which itself may be unobservable) to what we see today.

However, even Behe admits that certain biological structures and pathways can be explained by natural selection, and therein lies the problem. Systems that cannot yet be explained by natural selection are assigned to intelligent design, yet there is no guarantee that no Darwinian explanation will ever be found. An irreducibly complex system may have arisen as a result of increasing specialization: components that initially were self-sufficient evolved together to become dependent on one another. Alternatively, a new system may have "taken over" the functions of an older system that life initially depended on. Consider a stone arch, which is nearly impossible to build by an incremental process of adding stones. Instead, you first build a scaffolding which supports the stones, then remove the scaffolding once the stones are able to support each other.

1

It is difficult to guess if some other kind of life would be possible in some other kind of Universe.

100 CHAPTER 8. LIFE: NATURE'S DESTINY?

Note that I am not arguing that every aspect of the Universe has a scientific explanation. Certain things, such as the creation of the Universe, may need to be accepted as brute facts. When it comes to the origin and evolution of life, though, I don't think this is necessarily the case, and a major motivation for astrobiology is trying to understand these events scientifically. Scientists should always be hesitant to apply the anthropic or theistic principles willy-nilly, because neither explanation has a well-defined realm of applicability: they can be invoked to explain anything. A meaningful scientific theory should not be a last resort for cases in which no other explanation is available. For if the past is any guide, many apparently intractable problems will eventually be solved.

8.3 Life and the Future of the Universe Perhaps the most speculative application of the anthropic principle is the final anthropic principle, which states that "intelligent information processing must come into existence in the Universe, and, once it comes into existence, it will never die out." Is there a limit to how long life can exist? Our Sun will eventually burn out, but new stars are being born which might support our descendants after that. There is a limit, however, to how much useful energy can be extracted from stars: each generation of stars has less hydrogen to work with, and eventually there will be too little hydrogen for further star formation. This will occur roughly a trillion (10

12

) years from now. The Universe will then consist of old stars, dead

planets, stellar remnants, and black holes. Of course, it is possible by this time that the Universe will have reversed its expansion and already recollapsed.

Although carbon-based life is ultimately doomed, perhaps computers can continue to exist into the far future--a computer being a generic term for an information-processing device. Gradually more and more of the Universe could be transformed into information, until life engulfs everything. What that information would say, and whether it would be "conscious" in some sense, I hesitate to imagine.

Bibliography

[1] Avers, Charlotte J. Process & Pattern in Evolution. New York: Oxford U. Press, 1989. [2] Barrow, John D. & Tipler, Frank J. The Anthropic Cosmological Principle. Oxford U.

Press, 1986.

[3] Bickerton, Derek. Language and Human Behavior. Seattle: U. of Washington Press,

1995.

[4] Brockman, John. The Third Culture. New York: Simon & Schuster, 1995. [5] Cairns-Smith, A.G. Seven Clues to the Origin of Life: A Scientific Detective Story.

Cambridge U. Press, 1985.

[6] Deacon, Terrence W. The Symbolic Species: The Co-Evolution of Language and the

Brain. New York: W. W. Norton, 1997.

[7] De Duve, Christian. Vital Dust: Life as a Cosmic Imperative. New York: Harper Collins,

1995.

[8] Dick, Steven J. The Biological Universe: The Twentieth-Century Extraterrestrial Life

Debate and the Limits of Science. Cambridge U. Press, 1996.

[9] Dick, Steven J. Life on Other Worlds: The 20th-Century Extraterrestrial Life Debate.

Cambridge U. Press, 1998.

[10] Goldsmith, Donald & Owen, Tobias. The Search for Life in the Universe (2nd ed.).

Reading, MA: Addison-Wesley, 1992.

[11] Jakosky, Bruce. The Search for Life on Other Planets. Cambridge U. Press, 1998. [12] Kauffman, Stuart. At Home in the Universe: The Search for the Laws of SelfOrganization and Complexity. New York: Oxford U. Press, 1995.

[13] Leakey, Richard. The Origin of Humankind. New York: BasicBooks, 1994. [14] Lehninger, Albert L., Nelson, David L., & Cox, Michael M. Principles of Biochemistry

(2nd ed.). New York: Worth Publishers, 1993.

[15] Lewin, Roger. The Origin of Modern Humans. New York: Scientific American Library,

1998.

101

102 BIBLIOGRAPHY [16] "Life in the Universe." Scientific American (special issue), Vol. 271 No. 4, October 1994. [17] Lunine, Jonathan I. Earth: Evolution of a Habitable World. Cambridge U. Press, 1999. [18] Maddox, John. What Remains to be Discovered. New York: Free Press, 1998. [19] Marcy, G. W. & Butler, R. P. "Detection of Extrasolar Giant Planets." In Annual

Review of Astronomy and Astrophysics, Vol

. 36, p. 57, 1998.

[20] Regis, Edward, ed. Extraterrestrials: Science and Alien Intelligence. Cambridge U.

Press, 1985.

[21] Ridley, Mark. Evolution. Boston: Blackwell Scientific, 1993. [22] The Search for Life's Origins: Progress and Future Directions in Planetary Biology

and Chemical Evolution. Space Studies Board, National Research Council, Washington D.C., 1990.

[23] Smith, John Maynard & Szathm'ary, E"ors. The Major Transitions in Evolution. Oxford:

W.H. Freeman, 1995.

[24] Tattersall, Ian. Becoming Human: Evolution and Human Uniqueness. New York: Harcourt Brace, 1998.

[25] Thomas, Paul J., Chyba, Christopher F., & McKay, Christopher P., eds. Comets and

the Origin and Evolution of Life. New York: Springer-Verlag, 1997.

[26] Woolf, Neville & Angel, J. Roger. "Astronomical Searches for Earth-like Planets and

Signs of Life." In Annual Review of Astronomy and Astrophysics, Vol. 36, p. 507, 1998.

[27] Zubay, Geoffrey. Origins of Life on the Earth and in the Cosmos. Dubuque, IA: Wm.

C. Brown, 1996.

[28] Zuckerman, Ben & Hart, Michael H., eds. Extraterrestrials: Where Are They? (2nd