Martin Rees describes how cosmologists aim to set our solar system in a grand evolutionary scheme by describing a cosmic history probably beginning with a'big bang'.

Whilst this planet has been cycling on according to the fixed law of gravity, from so simple a beginning forms most wonderful . . . have been and are being evolved." These are the concluding words of Charles Darwin's On the Origin of Species. Cosmologists aim to go back before Darwin's "simple beginning" to set our solar system in a grand evolutionary scheme stretching back to the formation of the Milky Way galaxy - right back to a so-called "big bang" that set our entire observable universe expanding, and imprinted the physical laws that govern it.

About 4.5 billion years ago our sun, a typical star, condensed from an interstellar cloud, and contracted until the centre became hot enough to ignite fusion of hydrogen into helium. This process will keep it shining until, after another 5 billion years, the hydrogen runs out. The Sun will then flare up, becoming large enough to engulf the inner planets and to vapourise all life on earth. After this "red giant" phase the inner regions will contract into a white dwarf - a dense star no larger than the earth, though nearly a million times more massive.

We are quite confident about these calculations because the relevant physics has been well studied in the lab - atomic and nuclear physics, Newtonian gravity and so forth. Astrophysicists can just as easily compute the life cycles of stars with half the Sun's mass, twice, four times and so on. Heavier stars burn brighter and trace out their life cycle more quickly.

ÖйúAƬ

Stars live so long compared with astronomers that we are granted just a single "snapshot" of each one's life. But we can test our theories by looking at the whole population of stars. Trees can live for hundreds of years. But even if you had never seen a tree before, it would take no more than an afternoon in a forest to deduce their life cycle: from looking at saplings, fully grown specimens and some that had died.

In, for instance, the Orion Nebula, new stars are even now condensing within glowing gas clouds. The best "test beds" for checking such calculations are globular clusters - swarms of a million different stars, held together by their mutual gravity, which all formed at the same time.

ÖйúAƬ

But not everything in the cosmos happens slowly; sometimes stars explode catastrophically as supernovae. The closest supernova of the 20th century occurred in 1987. Its sudden brightening and gradual fading have been followed not only by optical astronomers but by those using other techniques - radio, X-ray and gamma-ray telescopes -that have opened new "windows" on the universe.

In about 1,000 years, it will look like the Crab Nebula, the relic of a supernova witnessed and recorded by Chinese astronomers in 1054 ad. Now, nearly 1,000 years later, we see the expanding debris from the explosion. The Crab nebula will remain visible, gradually expanding and fading, for a few thousand years; it will then become so diffuse that it merges with the very dilute gas and dust that pervade interstellar space.

Supernovae fascinate astronomers. But why should anyone else care about explosions thousands of light years away? Because, were it not for supernovae, there would be no planets, still less any complex evolution on them.

Of the 92 chemical elements that occur naturally, some are vastly more common than others. For every 10 atoms of carbon, you would find, on average, 20 of oxygen, and about five each of nitrogen and iron. But gold is a million times rarer than oxygen and others - platinum and mercury, for instance - are rarer still.

Why are carbon and oxygen common, but gold and uranium so rare? This everyday question is not unanswerable - but the answer involves ancient stars that exploded in our Milky Way more than five billion years ago, before our solar system formed.

Stars much heavier than the sun evolve in a more complicated and dramatic way. After they have used up their central hydrogen (and turned into helium) gravity squeezes them further. Their centres get still hotter, until helium atoms can themselves stick together to make the nuclei of heavier atoms - carbon (six protons), oxygen (eight protons) and iron (26 protons). A kind of "onion skin" structure develops: where the hotter inner layers have been transmuted further up the periodic table.

When their fuel has all been consumed (when their hot centres are transmuted into iron) big stars face a crisis. A catastrophic infall squeezes their centres to the density of an atomic nucleus, triggering an explosion that blows off the outer layers. This explosion manifests itself as a supernova of the kind that created the Crab Nebula.

The debris contains the outcome of all the nuclear alchemy that kept the star shining over its entire lifetime - a lot of oxygen and carbon, plus traces of many other elements. The calculated "mix" is gratifyingly close to the proportions now observed in our Solar System.

ÖйúAƬ

The Milky Way, our home galaxy, resembles a vast ecosystem. Pristine hydrogen is transmuted, inside stars, into the basic building blocks of life - carbon, oxygen, iron and the rest. Some of this material returns to interstellar space, thereafter to be recycled into new generations of stars.

A carbon atom, forged in an early supernovae, might wander for hundreds of millions of years in interstellar space. It might then have found itself in a dense interstellar cloud, which collapsed under its own gravity to form stars. It might have joined one of the less massive stars, each surrounded by a spinning gaseous disc that condenses into a retinue of planets. One such star could have been our Sun. The same carbon atom may have found itself in the newly forming Earth, perhaps eventually in a human cell.

Each atom has a pedigree extending back far earlier than our solar system's birth. We are literally the ashes of long-dead stars.

The Big Bang

But how did our galaxy itself emerge? And where did the basic hydrogen come from? Did everything really start with a so-called "big bang"?

The idea goes back to Belgian Catholic priest Georges Lematre in 1930 but the clinching evidence for the theory came in 1965, when Penzias and Wilson found excess microwave noise, coming equally from all directions and with no obvious source, in their antenna at the Bell Telephone Laboratory. This has momentous implications - intergalactic space is not completely cold - it's about 3 degrees above absolute zero. That may not seem much, but there are about a billion quanta of radiation - photons - for every atom in the universe.

This "cosmic background" causes some 1 per cent of the background "fuzz" on a television set. It is an "afterglow" of a pregalactic era when the entire universe was hot and dense and opaque. After expanding for about half a million years the temperature fell below 3,000K; the primordial radiation then shifted into the infrared. The universe literally entered a dark age, which persisted until the first stars in the first galaxies, and maybe also the first quasars, formed and lit space up again. The expansion cooled and diluted the radiation, and stretched its wavelength. But it would still be around - it fills the universe and has nowhere else to go.

But we have got firm grounds for believing that temperature was once billions of degrees, not just thousands - hot enough for nuclear reactions. The rapid expansion did not allow time for everything to be processed into iron, as in hot stars. However, about 25 per cent would have turned into helium. The rest would still be hydrogen apart from traces of deuterium and lithium.

What is remarkable is that the proportion of helium in old stars and nebulae - now pinned down with 1 per cent accuracy - turns out to be just about what is calculated. As a bonus, so are the proportions of lithium and deuterium too.

Over the past few years, the case for a "big bang" has had several boosts - the COBE (Cosmic Background Explorer) satellite showed that the background radiation had the expected spectrum, to a precision of a part in 10,000; and there have been better measurements of cosmic helium and deuterium. Moreover, there are several discoveries that might have been made, which would have invalidated the hypothesis, and which have not been made - the big bang has lived dangerously for 25 years and survived.

The theory boasts fervent believers. The great Soviet cosmologist Zeldovich once claimed that the big bang was "as certain as that the Earth goes round the Sun" (even though he must have known his compatriot Landau's dictum that cosmologists are "often in error but never in doubt").

I would bet at least 90 per cent on the general concept that we can extrapolate back to when the universe was a second old.

ÖйúAƬ

The Future

Cosmic timespans extend at least as far into the future as into the past. Suppose America had existed for ever, and you were walking across it, starting on the east coast when the Earth formed, and ending up in California ten billion years later, when the Sun was about to die. To make this journey, you would have to take one step every 2,000 years. All recorded history would be three or four steps, just before the halfway stage - somewhere in Kansas perhaps. In this perspective, we are still near Darwin's "simple beginning" of the evolutionary process. The progression towards diversity has much further to go. Even if life is now unique to the Earth, there is time for it to spread from here through the entire galaxy, and even beyond.

In about five million years the Sun will die; and the Earth with it. At about the same time (give or take a billion years) the Andromeda Galaxy, already falling towards us, will crash into our own Milky Way, merging to form a single amorphous elliptical galaxy.

But will the universe expand for ever, attaining some asymptotic "heat death"? Or will it, after an immense time, recollapse to the big crunch? The ultra-long-range forecast depends on how much the cosmic expansion is decelerating. The deceleration comes about because everything in the universe exerts a gravitational pull on everything else. It is straightforward to calculate that the expansion will eventually go into reverse if the average cosmic density exceeds about three atoms per cubic metre. But space seems even emptier than that: if the atoms in all the stars and gas in all the galaxies were dispersed uniformly, they would fall short of this "critical" density by a factor of at least 50.

At first sight this seems to imply perpetual expansion, by a wide margin. But the case is not so straightforward, because there seems to be at least ten times as much material in "dark" form as we see directly.

What could this dark matter be? Maybe it is faint stars whose centres are not squeezed hot enough to ignite their nuclear fuel; or black holes - remnants of big stars that were bright when the galaxy was young but have now died.

But there are other options. The hot early universe may have contained not just atoms and radiation, but other particles as well. In particular, there should be huge numbers of neutrinos - about a billion for every atom in the universe. So even a very tiny individual mass would make the cumulative gravitational effects of neutrinos important. But do neutrinos have any mass at all? Recent experiments at Los Alamos suggest that they do. But these have so far been announced in the New York Times, and not yet in a proper scientific journal, so we would be prudent to suspend judgement. If the claimed mass is right, neutrinos contribute more gravitating stuff than all the stars and gas in the universe.

At least we know neutrinos exist. But particle theorists have a long shopping list of particles that might exist, and (if so) could have survived from the early phases of the big bang. If such particles pervade our galaxy, there would be 100,000 of them in every cubic meter, most passing straight through the Earth without interacting. But their cross-section for colliding with ordinary atoms, though tiny, is not quite zero, and sensitive experiments are being set up to detect the rare events when this happens.

The equipment must be placed deep underground, to reduce other types of background signal. A group in the United Kingdom is building such an experiment down a mine in Yorkshire. It is a difficult experiment, but a positive result would not only reveal what 90 per cent of the universe is made of, but also discover new types of particle that could never be detected in other ways.

We should not be surprised that there is dark matter. There is no reason why everything in the universe should shine. The challenge is to decide among many candidates. Dark matter's dominance may demote our cosmic status still further. Copernicus dethroned the Earth from a central position. Hubble showed that the Sun was not in a special place. But now particle chauvinism may have to go. We ourselves, and all the stars and galaxies would then be trace constituents of a universe whose large-scale structure is controlled by the gravity of dark matter of a quite different kind - we see, as it were, just the white foam on the wave crests, not the massive waves themselves.

People often wonder how the universe can have started off in thermal equilibrium - a hot, dense fireball - and ended up manifestly far from equilibrium. Temperatures now range from blazing surfaces of stars to the night sky only three degrees above absolute zero. Although this seems contrary to thermodynamic intuitions that temperatures tend to equilibrate as things evolve, it is actually a natural outcome of cosmic expansion, and the workings of gravity.

Gravity has the peculiar tendency to drive things further from equilibrium. When gravitating systems lose energy they get hotter. A star that loses energy and deflates ends up with a hotter centre than before. To establish a new and more compact equilibrium where pressure can balance a (now stronger) gravitational force, the central temperature must rise.

And gravity does something else. It renders the expanding universe unstable to the growth of structure, in the sense that even very slight initial irregularities would evolve into conspicuous density contrasts. Theorists are now carrying out increasingly elaborate computer simulations of how this happened.

If one had to summarise what has been happening since the big bang the best answer might be: "Ever since the beginning, gravity's 'antithermodynamic' effects have been amplifying inhomogeneities, and creating progressively steeper temperature gradients - a pre-requisite for emergence of the complexity that lies around us ten billion years later, and of which we're part."

The Ultra-early Universe

Cosmic history can be divided into three parts: part one is the first millisecond, a brief but eventful era spanning 40 decades of logarithmic time. This is the intellectual habitat of the high-energy theorist and the quantum cosmologist.

The second stage runs from a millisecond to about a million years. It is an era where cautious empiricists like myself feel more at home. The densities are far below nuclear density, but everything is still expanding in an almost homogeneous fashion. The relevant physics is firmly based on laboratory tests, and theory is corroborated by good quantitative evidence - the cosmic helium abundance, the background radiation and so on. This stage, though it lies in the remote past, is the easiest to understand.

The tractability lasts only so long as the universe remains amorphous and structureless. When the first gravitationally bound structures condense out - when the first stars, galaxies and quasars have formed and lit up - the era studied by traditional astronomers begins. We then witness complex manifestations of well-known basic laws. Gravity, gas dynamics and feedback effects from early stars, combine to initiate the complexities we see around us and are part of.

But we then realise that the few basic numbers that determine how the universe has evolved are all legacies of the uncertain physics of the first phase. Even the physical laws themselves may have been imprinted in the ultra-early universe. (This is speculative territory.) First, what about the initial expansion rate? This has to be very precisely tuned. The two eschatologies - perpetual expansion or recollapse to a "crunch" - seem very different. But our universe is still expanding after 10 billion years. Had it recollapsed sooner, there would not have been time for stars to evolve - indeed, if it had collapsed after less than a million years it would have remained opaque, precluding any thermodynamic disequilibrium. On the other hand, the expansion cannot be too much faster than the critical rate. Otherwise gravity would have been overwhelmed by kinetic energy and the clouds that developed into galaxies would have been unable to condense out.

In Newtonian terms the initial potential and kinetic energies were very closely matched. How did this come about? And why does the universe have the large-scale uniformity which is a prerequisite for progress in cosmology?

The answer may lie in something remarkable that happened during the first 10-36 seconds. Ever since that time, the cosmic expansion has been decelerating, because of the gravitational pull that each part of the universe exerts on everything else. But theoretical physicists have come up with serious (though still, of course, tentative) reasons why, at the colossal densities before that time, a new kind of "cosmical repulsion" might come into play and overwhelm "ordinary" gravity. The expansion of the ultra-early universe would then have been inflated, homogenised, and established the "fine-tuned" balance between gravitational and kinetic energy when it was only 10-36 seconds old.

This generic idea that the universe went through a so-called inflationary phase is compellingly attractive. The fluctuations from which clusters and superclusters form, and the even vaster ones whose imprint on the background radiation spreads right across the sky, may be the outcome of microscopic quantum phenomena from an ultra-ancient epoch when the universe was squeezed smaller than a golfball. (The two foundations of 20th-century physics are Einstein's theory of gravity on the one hand, and the quantum uncertainty principle on the other. But there is no overlap between these two concepts. Gravity is so weak that it is negligible on the scale of single molecules, where quantum effects are crucial. Conversely, gravitating systems like planets are so large that quantum effects can be ignored in studying how they move. But right back at the beginning of the universe, the densities could have been so high that quantum effects were important for the whole universe.) We do not, of course, know the physics that prevailed at this ultra-early time. But there is a real prospect of discovering something about it. Specific models of how the inflation is driven make distinctive predictions about things we can observe - large-scale clustering, and small non-uniformities in the background radiation over the sky. We shall soon be confronting these speculations about the ultra-early universe with real empirical tests.

Martin Rees is a Royal Society professor, Cambridge University, and the Astronomer Royal. This is an edited extract from his Darwin lecture.

ÖйúAƬ