How did the Solar System begin? What series of events occurred to form the Sun, planets and Earth? How did the Earth become a suitable place for life to arise? As a result of the exploration of the Solar System by spacecraft over the past 40 years, we have begun to answer some of these questions. Beginning around 1960, our knowledge of the planets underwent an explosive growth. The Ranger, Surveyor, Lunar Orbiter and Apollo Missions to the Moon informed us in great detail about our satellite. This has been reinforced by the recent Clementine and Lunar Prospector Missions. Mars and Venus have been the target of many missions, while the very successful Magellan mapping mission has given us a more precise knowledge of the topography of Venus than we have of the floors of our oceans. Even tiny Mercury, difficult to reach as it is near to the giant gravitational influence of the Sun, has had a brief visit, with more planned. The giant outer planets and their satellites that were mere points of light in our telescopes have been revealed, mostly by the Pioneer and especially by the Voyager
Professor Ross Taylor is an Emeritus Professor at the Australian National University where he is working on a second edition of Solar System Evolution: A New Perspective. He carried out the first analysis of the first lunar sample returned to Earth in NASA, Houston in 1969. He has received the premier awards of the Geochemical Society (Goldschmidt Medal) and the Meteoritical Society (Leonard Medal). He is a Fellow of the Australian Academy of Science mid a Foreign Associate of the US National Academy of Sciences. Asteroid 5670 is named Rosstaylor.
58
and Galileo Missions, to be extraordinary worlds, full of exciting and unanticipated details. Now we have begun to explore the asteroids, with the Shoemaker-Near mission both landing on Eros and sending back information about the composition of its surface.
The present Universe began about 13-15 billion years ago, but it was a long time before the Solar System arose. Four and a half billion years ago, in a Universe that would look quite familiar to us and which had already existed for perhaps 10 billion years, a mass of gas and dust became detached from a larger cloud in a spiral arm of the Milky Way galaxy.
Within about a hundred thousand years, this mass collapsed under its own gravity into a disc. In the centre of this swirling mass of gas and dust, a star that became our Sun condensed. If the disc had been larger or spinning more rapidly, it would have formed a double or triple star system; such as constitute about 80 per cent of all stars. However our disc was just the right size to make one rather than two stars. A little material was left over, from which the planets formed over the next 100 million years.
The primitive disc, called the solar nebula, consisted mainly (98 per cent) of gas (hydrogen and helium), plus about two per cent of elements heavier than helium, which form the remainder of the Periodic Table of the chemical elements. These elements (called 'metals' by astronomers to the annoyance of chemists) are formed by nuclear reactions inside stars and are dispersed as the star dies or explodes. It takes several billion years by these slow processes to make up the two per cent that we and the planets are made of. So rocky planets that are made of oxygen, silicon, magnesium, aluminium, iron and the rest can only arise in a 'mature' Universe. The 'metals' are divided into 'ices' (water,
59
ammonia, methane constituting about 1.5 per cent) and 'rock' (about 0.5 per cent). The amount of 'rock' in the original disc, the solar nebula, was so small that it could be ignored, except that we are standing on some of it. Only the small, inner planets of the Solar System (Mercury, Venus, Earth, Mars) are made mostly of rock. The others are mostly gas.
Once the Sun grew large enough, temperature and pressure conditions in its interior allowed nuclear reactions to convert hydrogen into helium with a great release of energy, and the Sun 'lit up'. Observations on nearby young stars suggest that the Sun underwent violent outbursts as it proceeded on its path towards a stable existence for the next four billion years.
The most popular current theory is that planets are built up `brick by brick' from smaller bodies called planetesimals. This is usually referred to as the planetesimal hypothesis. There is a lot of evidence for the former existence of planetesimals during the formation of the Solar System.
For example, the large tilts of many planets (the Earth's is 23.5 degrees) have been caused by collision with very large bodies well over 1000 kilometres in diameter. The rapid 24-hour spin of the Earth is probably due to a giant impact that probably formed the Moon as a by-product. Venus, in contrast, has almost zero tilt and is rotating very slowly backwards, taking 243 Earth days for one rotation. Perhaps Venus never experienced a giant impact.
The large icy giant, Uranus, 14 times more massive than the Earth, is lying on its side. A body the size of the Earth would be needed to tip Uranus over, while collisions with a body between three and 10 Earth masses would be needed to produce the 27-degree tilt of Saturn. All the planets spin
The ultimate significance is that many chance collisions occur randomly while planets are growing. Many unique and unrepeatable events occurred as the planets were put together. So the formation of planets like the Earth is not an inevitable result that could be repeated like a gigantic computer program, but partly depends on random events during the early history of the nebula.
Venus, Earth and Mars (we have no data for Mercury at present) all appear to have lost volatile elements such as lead, sodium, potassium as well as water. This appears to be typical of the entire inner Solar System. The most plausible model is that early violent solar activity swept away not only the gaseous elements, but also ices and volatile elements that had not condensed.
It is clear from the modelling of planet-forming processes that Jupiter formed very early, 50 to 100 million years before the Earth and Venus. How did such a large planet form so rapidly and so early so far from the Sun?
The current model suggests that as early strong winds from the Sun swept water and volatile elements out from the inner nebula, water ice condensed in the region of Jupiter
60
where it became cold enough. This condensation caused a pileup of ice at such a `snow line'. This acted as a 'cold trap' for other material.
Large masses of ice and rock collected into bodies perhaps of 10 Earth masses. These became large enough to collect an envelope of gas, mostly hydrogen and helium. This occurred within a million years before the gas was also swept away. Probably the cores of the four giant planets grew near this place. Jupiter got a head start and this massive planet dominated the Solar System. It took material from the nearby asteroid belt where another planet should have grown. The region where Mars would later accumulate was likewise depleted so that Jupiter is responsible for the small size of that planet - Mars is only 1/3000 as massive as Jupiter.
The other giant planets, Saturn, Uranus and Neptune, began life near Jupiter, but collected less gas and were kicked out to their present positions by the gravitational forces exerted by Jupiter as it became the dominating planet.
The orbit of Neptune marks the outer edge of the planetary system, but the edge of the Solar System extends out about halfway to the nearest star. Jupiter and the three other large planets behaved like boisterous giants, tossing out these small bodies that they did not seize. Many were sent into the outer Solar System, where they form two clouds. The inner cloud is called the Edgeworth-Kuiper belt. Tiny Pluto, one fifth of the size of the Moon, is the chief member. Demoted from planetary status, it is no longer a dwarf among giants but takes its rightful place as the largest body among this swarm of icy planetesimals. Further out is a spherical cloud (the Con Cloud) of icy bodies. Comets come from both these clouds, sent sunwards by the gravitational pull of the giant planets.
It is interesting that the ancient astronomers, unaware of the true size of Jupiter, nevertheless showed unusual insight in naming it after the chief Roman god. The Roman perception of the importance of Jupiter has a modern significance, for we now understand that we might not be here if Jupiter did not exist.
Now Jupiter acts as a shield against the impacts of comets. Without that mighty gravitational shield, the numbers of impacts on the Earth would be perhaps a thousand times greater than at present. Local areas a few kilometres across
61
would experience a catastrophe several times a year, instead of once in a thousand years. Collisions of the sort that killed off the dinosaurs 65 million years ago might occur every hundred thousand years rather than on time scales of hundreds of million of years.
Such an intense bombardment would have had incalculable effects on the development of life, perhaps stopping it altogether. Even with the protection of Jupiter, this planet has suffered major biological extinctions due to impacts. It seems unlikely that we could have survived such disasters without the protective shield of Jupiter.
This raises the question whether planets like Jupiter that could protect their smaller brethren are common in other planetary systems? If the formation of large gas giants is part of the same lottery of chances that is such a feature in our Solar System, Jupiters might be uncommon. It certainly seems to have been difficult enough to form Jupiter. The timing is exquisite. If the Sun had been bigger, or had a more violent early history, the gas might be swept away before a core big enough to catch it could form. Then one would be left with icy giants, clones of Uranus and Neptune. Suppose that the core forms just too late to catch the gas. We finish up again with a Uranus or Neptune in place of Jupiter. Suppose the `snow line' didn't form. Then we might have a much smaller size planet at Jupiter and a larger version of Mars, perhaps half the size of the Earth. There would be a respectable planet in place of the asteroid belt. Probably all these processes occurred, as well as some that we haven't thought of, around other stars out in the galaxy. This view is reinforced by the variety of recently discovered planetary systems, which fit no simple pattern.
In the region now occupied by the Earth and the other terrestrial planets, the early loss of gas and volatile elements left only rocky bodies. This rubble remaining in the inner nebula grew by collisions into planetesimals of varying dimensions. Some of these reached the size of Mars (about 10-15 per cent Earth mass), before finally being collected by the Earth or Venus. The history of the individual planetesimals is surely complex.
Unlike Jupiter, which grew within a few million years, the collection of these planetesimals into the four planets, Mercury, Venus, Earth and Mars, is estimated to have taken between 50 and 100 million years. These estimates are fairly firmly based on a combination of astrophysical observations on young stars that are surrounded by dusty discs, by computer modelling, and by the ages established by laboratory measurements on meteorites and lunar samples. So the inner planets must have formed in an essentially gas- free environment, as the gas was driven away within a few million years.
What was the history of the planetesimals prior to their sweepup into the inner planets? Some of the largest, the size of Mars, would have made respectable planets in their own right if fate had taken a different course. Possibly the planetesimals melted and formed metal cores and rocky mantles before they were swept up into planets.
Would the processes that produced our planets form clones of the Earth in other systems? Although we have only one set of planets, three of the giants, Jupiter, Saturn and Uranus, possess regular satellite systems (the capture of Triton was probably responsible for the destruction of any original satellite system of Neptune). These four satellite systems of the giant planets provide some answers to our question. Although these miniature Solar Systems around Jupiter, Saturn, and Uranus probably formed by similar processes and so might have been expected to be similar, they are all quite distinct. They might just as well belong to separate systems. Thus the processes that formed regular satellite systems from disks around the giant planets did not result in a uniform product. A large element of chance has accompanied the formation of the satellite systems as well as of our planets.
Mercury is only five per cent of the mass of the Earth and perhaps only a quarter of the mass of the body that collided with the Earth to form the Moon. It forms an example of a body accidentally left over from planet building, which has survived by reaching a stable orbit. It is peculiar that tiny Mercury has a small rocky mantle, a large iron core, and an inclined orbit so close to the Sun. The current explanation for the origin of the planet is that Mercury was struck by a body about one-sixth of its mass at a late stage in its formation. The collision fragmented the planet, with most of the rocky mantle lost to space. However, the tougher iron core survived to reform the planet. Covered like a beggar with a thin cloak, it collected a little that survived of the rocky mantle.
Mars is a survivor that might just as readily have been swept up into a larger planet. It owes its small size to the loss of material stolen by the earlier formation of giant Jupiter. Between Mars and Jupiter, the multitude of tiny asteroids are all that is left. Although Mars is so small (only about 11 per cent of the mass of the Earth), it is of interest here as the only other planet in our system that we could think about living on.
The crust in the northern hemisphere of Mars consists mainly of monotonous lava plains. The crust in the southern hemisphere is older and heavily cratered and dates from over four billion years ago. Is this composed of granite, like the continental crust of the Earth, or of feldspar; like the crust of the Moon, or of something else? Soils from three different widely separated Martian sites analysed by the Viking and Pathfinder missions are all similar and are basaltic in composition. These soils are derived from planet-wide dust storms and so provide a global average. The crust thus seems to be made dominantly of basaltic lava, like our ocean
62
floors, and there does not seem to be anything on Mars that resembles plate tectonics or our familiar granitic continents. Most of our ore deposits occur on continents and are ultimately due to the operation of plate tectonics. So there will be few ore deposits on Mars.
Venus is the only rocky planet in the inner Solar System that managed to grow to around the size of the Earth. Thus Venus is the most significant planet in our planetary system in so far as it shows what happens when nature tried to duplicate our planet. Venus is a little smaller in radius than the Earth and its density is slightly less than the value for that of the Earth. After correcting for the pressure differences, the uncompressed density is close to that of the Earth. Venus is thus close enough in mass and density to be regarded as a twin planet to the Earth. However, it provides a good illustration that while planets may be similar, they are not necessarily identical.
Venus has no detectable magnetic field, the surface temperature is 470 C, and the only detectable water is the atmospheric content of about 50 parts per million. Venus rotates very slowly backwards, has no satellite, and possesses a thick atmosphere (95 bars, mostly of carbon dioxide). Thus, despite the similarity in size and density, there are major differences between Venus and the Earth, probably the consequence of a differing collisional history from the Earth.
The Magellan spacecraft mission has revealed that the crust of Venus is dominated by basaltic lavas. In contrast to the Earth there is no sign of the operation of plate tectonics. There are no extensive areas of granite analogous to the terrestrial continental crust, and so no likelihood of great ore deposits. The high standing regions of Aphrodite Terra and Ishtar Terra on Venus are apparently crumpled-up basaltic lavas. The extremely dry rock is strong, supporting steep slopes for tens of millions of years in great contrast to the weaker hydrated rocks of Earth. The strength of the crust on Venus seems to be due to the absence of water. Like an armadillo, Venus has encased itself in a strong rigid shell.
The surface of Venus displays a range in ages, averaging about 750 million years. No ancient heavily cratered surfaces, as on Mars, Mercury or the Moon, have been discovered.
63
Most of the craters appear relatively fresh and uneroded. The surface of Venus is thus relatively young and a geologist on Venus would find little evidence of any truly ancient rocks.
Venus and the Earth most likely have a similar internal production of heat from radioactive potassium, thorium and uranium, but their geological histories are totally distinct. A few hundred million years ago, Venus covered its surface with a great outpouring of lava, obscuring any older crust. Since then geological activity hasn't produced more than a trickle of lava. Venus, producing barren basaltic plains, has thus had a very different history to the Earth with its diverse landforms. So it is not sufficient only to make a suitable planet; the subsequent geological history is also a crucial factor. The absence of abundant water is probably the crucial difference between the two planets and Venus and the Earth are similar only in a Jekyll- Hyde sense.
Earth. Large collisions in the final stages of accretion, including the Moon-forming event, are likely to have removed any primitive atmosphere which might have formed. The present atmosphere and hydrosphere of the Earth appear to be entirely secondary in origin. The terrestrial amount of water actually contained in the Earth, although uncertain, is a tiny proportion of the Earth mass. It is about a thousand times less concentrated than the amount of water in the primitive solar nebula. This is an amount so small that it could be ignored except that we are here on account of it. Water is also responsible for the operation of plate tectonics, which produces continents and the ore deposits that are so crucial for our technical civilisation.
In the current model for the evolution of the solar nebula, water is driven out very early from the inner nebula. So water, necessary for life, arrived on the Earth late, perhaps supplied from comets but more likely the tiny amounts of water actually in the inner planets arrived as ice particles drifting back Sunwards from near Jupiter.
In summary, nature made four rocky planets in our system, each one so distinctive that they could just as well reside in separate planetary systems. The bottom line is that all are different and owe their particular composition to the operation of random accumulation processes. Albert Einstein made a famous comment that `God does not play dice'. However, the evidence for the importance of chance events in forming the Earth indicates either that He does, or that He was not involved and we need some other ultimate philosophical explanation for the origin of the Universe.
Our Moon is a unique satellite: the satellites of the other outer planets are mainly mixtures of rock and ice, formed by accretion around their parent planets, or by subsequent capture of bodies left over after the major planets formed. As discussed earlier, a giant collision of a Mars-sized body with the Earth is the most likely explanation for the origin of the Moon.
This model accounts both for the high angular momentum of the Earth-Moon system and the strange lunar orbit that lies neither in the plane of the Earth's equator nor in the plane of the Earth's orbit around the Sun but at 5 degrees to the latter. Such unique geochemical features as the bone-dry nature of the Moon, the extreme depletion of very volatile elements and the enrichment of refractory elements in the Moon are also consistent with the very high energies and
64
heating involved in this stupendous event that also melted the Earth.
What were the implications for the Earth of the single-impact origin of the Moon? It may be responsible for the 24-hour rotation period and the 23.5 degree tilt of our planet. These are both significant factors that have affected biological evolution throughout time. If there were no Moon the tilt of the Earth would be chaotic with large variations reaching more than 50 degrees in a few million years and even, in the long term, more than 85 degrees. Thus our satellite is 'a climate regulator for the Earth'.
Any primitive atmosphere was removed during the impact event. So two questions without answers are how crucial is the presence of the Moon in producing this habitable planet and how likely are such events to happen in other planetary systems?
The Earth is about the right distance from the Sun to make it an agreeable and habitable planet. This question is often referred to as the Goldilocks problem. Venus is too hot, Mars may be too cold, but the Earth, like Baby Bear's porridge, is just right. However, this is a simplistic view, as much more than distance is involved. Thus the surface temperature of the Earth is maintained by a greenhouse effect. Without water and carbon dioxide in the atmosphere, the surface temperature would average - 18 C and the world would resemble Siberia in the depths of winter. Also, the nearly circular orbits of low inclination that we mostly take as a given, are much more difficult to achieve than originally supposed. Observations of extra-solar planets and model simulations of inner planetary systems appear to favour highly eccentric orbits. The orbits of the extra-solar planets are
similar to the eccentric orbits of binary stars. Thus the nearly circular orbits of planets in our system are perhaps exceptional. Estimates of the width of the zone around the Sun in which a habitable planet can reside are quite narrow, ranging from about a tenth of an astronomical unit to about half an AU around the orbit of the Earth (AU - the mean distance between the Sun and the Earth). This works for nearly circular orbits as in our Solar System. If the orbits were eccentric, the planet would drift in and out of the 'habitable zone'.
It is also ironic that the 'habitable zone' in our present Solar System occupies a region in the inner nebula that was originally depleted in water, carbon and other volatile elements essential for life. Thus the very elements critical for life are in short supply in the only regions where it might take root and flourish. If these elements, vital for life, come from accidental trapping in planetesimals or have to be reimported from the asteroid belt or beyond, this makes the creation of a habitable planet and the emergence of life even more dependent on random processes. As we can see, making a habitable planet depends on a complex set of factors, of which distance from the Sun is only one.
Moreover, the Earth has maintained a rather even climate for four billion years. The Sun at that ancient epoch would have produced about one-quarter to a third less light than it does now. This is the famous 'faint early Sun' problem. The astronomical theory seems robust enough, firmly based on the physics of nuclear fusion of hydrogen to helium. One might therefore expect that the early Earth would have been a frozen waste, which warmed up slowly through the ages as the Sun increased its output.
In contrast, the geological evidence is quite definite that running water, eroding the surface and producing water-laid
66
sediments, has been present throughout these vast epochs. Various explanations have been offered to explain how the Earth managed to maintain an even climate despite the faint sunshine that the astronomers insist upon but none has reached general acceptance. Usually some kind of greenhouse effect is invoked. Methane, ammonia and carbon dioxide have all been suggested. However, the early atmosphere rich in carbon dioxide that is mostly called for can have the reverse effect. Clouds of dry ice are likely to form in the high atmosphere, reflecting the Sun's rays and leading to a 'snowball' Earth that might become a permanent condition. Although there is good evidence for at least one such event in the late Proterozoic Epoch on Earth, the buildup of carbon dioxide due to volcanism continued unabated and thawed the Earth. Clearly a delicate balance has enabled the survival of the benign conditions that allowed life to continue.
The existence of other planetary systems used to be a hypothetical question. Now we have some preliminary answers. At least 70 extra-solar planets have been discovered orbiting other stars. The immediate reaction to the new discoveries, like the report of possible life on Mars, was to raise the hopes of finding intelligent extra-terrestrial life, the 'little green men'. Do these new discoveries have the philosophical significance of Galileo's observation of the phases of Venus and of four satellites in orbit around Jupiter? His discovery effectively destroyed the Ptolemaic System that had Earth at the centre of the Universe and led eventually to the acceptance of the ideas of Copernicus that the planets orbit the Sun. However, the differences between these new planets and our familiar Solar System do not encourage hope of finding clones of this planet. Meanwhile, like most other new discoveries in science, these bodies have raised more questions than answers.
The most successful method of detecting bodies in orbit around other stars relies on the Doppler principle. Planets, lost in the glare from the star, are detected by their gravitational pull on the star. In our own system, Jupiter causes a wobble in the rotation of the Sun of 13 metres per second. As a star moves towards the Earth, the spectral lines in its light move towards the shorter blue wavelengths. As a star recedes from the Earth, the lines move towards the longer red wavelengths. This forms a tiny example of the large 'redshift' effect that we observe for faraway galaxies and that tells us that the Universe is expanding.
Other possible techniques involve detecting the planet as it transits in front of the disc of a star. Thus a Jupiter-sized body moving across a solar-sized disc would dim its light by about one per cent. This sensitive technique has successfully determined the mass, radius, density, eccentricity and inclination of a planet transiting the disc of a star. The planet is a gas giant like Jupiter, but is so close that it orbits its star in just three and a half days, whereas Jupiter takes 11 years to go around our Sun.
Astronomers have found that about five per cent of stars like the Sun have planets, although so far very few have been examined. These planets are mostly much closer to their star than Mercury is to the Sun and have orbital periods of a few Earth days. Whether this is typical or merely a consequence of the search procedure remains to be discovered. Even Mercury in our system, 58 million kilometres away from the Sun, with surface temperatures on the illuminated side reaching 470 C, takes 88 days for one orbit. In contrast, one of the massive planets orbits around Tau Bootis in 3.3 days and is only seven million kilometres distant from its star. At that distance, the temperature is estimated to be 1500 C, too high for most minerals to condense.
Most of the other new planets, sometimes referred to as 'Hot Jupiters' are also very close in to their stars and orbit around them with periods of a few days.
Others are further away from their star. At least one planet is in a highly eccentric orbit around one member of a double star system.
All these properties were unexpected. Although no current theory allows for the formation of giant planets in such close orbits, these orbits are stable. All the extra-solar planets that are not close to their star have very eccentric orbits. The highly eccentric orbits observed among the new planets add further complexities to the problem of producing planets with orbits close to circular, within 'habitable zones' around stars.
Dusty discs from which planets might form around other stars are relatively common, over 100 having been observed. Perhaps half of all young stars are surrounded by dusty discs. No current model allows for the formation of the observed gas giants so close to their stars. The early violent evolution of a star will drive the gas, water and volatile elements needed to form the planet far out in the nebula. Various models have been suggested to overcome this paradox posed by the `Hot Jupiters'. Some suggest the trapping of gas around dry rocky cores close in to the star, but how the gas remains when the ice has gone seems unresolvable. The temperatures so close to the star are too high even for refractory elements to condense, so that growth of rocky cores seems difficult if not impossible.
The first direct measurement of planetary mass and density has confirmed that the planet orbiting a star named HD 209458 is indeed a gas giant. It seems likely that the other planets with similar masses, mostly larger than Jupiter, are also gas giants and not entirely made of rock and ice.
A basic problem is that there is not enough of the rock and ice components available to construct such large planets. In our system, there was only enough rock and ice to form the small cores of the giant planets as well as our familiar rocky inner planets.
67
The most realistic model is that the giant planets indeed formed out at several AU in a similar manner to the giants in our Solar System. While there was still some gas left, the giant could migrate inwards. When the gas had gone, the giant planets were left stranded at various distances from the star just like a whale that ventured too close to the shore and was left high and dry by the tide going out. Other plausible studies begin from the forming of two or three gas giants in the nebula. These giants interact by collisions and ejection that may result in one planet in an eccentric orbit far from the star and another in a close but stable orbit.
The arrival of a Jupiter-like giant in the region where our rocky planets formed would create the sort of mayhem that accompanied the capture of Triton and devastated the inner satellites of Neptune. No rocky planets in its path are likely to survive. The giant would collect any surviving rocky debris, toss it into the star, or into the outer reaches of the nebula. Some interesting planets or asteroids formed from this debris might result in the outer reaches of such a planetary system.
In summary, our hard-won theories for forming the giant planets in our Solar System seem to be robust, but tidal evolution may produce spacings of the giant planets that seem bizarre to us. Although the processes of forming planets around stars are probably broadly similar, the devil is once again in the details.
The production of giant planets has been simulated on computers. A wide variety of stable planetary systems was generated. The number of giant planets produced ranged from one to seven, only a few bearing a general resemblance to our system.
These new discoveries reinforce the message from our own system. Nothing resembling our Solar System has been discovered. The conditions that existed to make our set of planets are not easily reproduced elsewhere. Indeed, no two planets in the Solar System are alike. Likewise, the 80-odd moons are also odd characters that defy efforts to put them into pigeonholes. So it should have come as no surprise that when nature tried elsewhere to build planets the end result was different. We are left with the conclusion that attempts to find some general formulae for recreating the detail of the Solar System are likely to be on the wrong track. Local accidents have predominated over general theories, just as some overlooked detail of the landscape may ruin the course of a battle that was planned according to the best principles of military strategy.
Beatty, J. K. et al., (Editors) The New Solar System (fourth ed.) Cambridge University Press, 1999
Taylor, S. R., Solar System Evolution: A New Perspective (second ed.) Cambridge University Press, 2001
Taylor, S. R., Destiny or Chance: our Solar System and its Place in the Cosmos, Cambridge University Press, 2000
Weissman, P. R. et al., (Editors) Encyclopedia of the Solar System. Academic Press, 1999
[...]
(Taylor, R., "The Solar System: An Environment For Life?," in Walter, M., ed., et al., "To Mars and Beyond: Search for the Origins of Life," Art Exhibitions Australia: Sydney & National Museum of Australia: Canberra, Australia, 2001, pp.57-67. Emphasis original. Figures omitted)