The End

(3 of 9)

That lumpiness, reasoned theorists, must have evolved from some original lumpiness in the primordial cloud of matter that gave rise to the background radiation. Slightly denser knots of matter within the cloud--forerunners of today's superclusters--should have been slightly hotter than average. So some scientists began looking for subtle hot spots.

FIRE OR ICE?

Others, meanwhile, attacked a different aspect of the problem. As the universe expands, the combined gravity from all the matter within it tends to slow that expansion, much as the earth's gravity tries to pull a rising rocket back to the ground. If the pull is strong enough, the expansion will stop and reverse itself; if not, the cosmos will go on getting bigger, literally forever. Which is it? One way to find out is to weigh the cosmos--to add up all the stars and all the galaxies, calculate their gravity and compare that with the expansion rate of the universe. If the cosmos is moving at escape velocity, no Big Crunch.

Trouble is, nobody could figure out how much matter there actually was. The stars and galaxies were easy; you could see them. But it was noted as early as the 1930s that something lurked out there besides the glowing stars and gases that astronomers could see. Galaxies in clusters were orbiting one another too fast; they should, by rights, be flying off into space like untethered children flung from a fast-twirling merry-go-round. Individual galaxies were spinning about their centers too quickly too; they should long since have flown apart. The only possibility: some form of invisible dark matter was holding things together, and while you could infer the mass of dark matter in and around galaxies, nobody knew if it also filled the dark voids of space, where its effects would not be detectable.

So astrophysicists tried another approach: determine whether the expansion was slowing down, and by how much. That's what Brian Schmidt, a young astronomer at the Mount Stromlo Observatory in Australia, set out to do in 1995. Along with a team of colleagues, he wanted to measure the cosmic slowdown, known formally as the "deceleration parameter." The idea was straightforward: look at the nearby universe and measure how fast it is expanding. Then do the same for the distant universe, whose light is just now reaching us, having been emitted when the cosmos was young. Then compare the two.

Schmidt's group and a rival team led by Saul Perlmutter, of Lawrence Berkeley Laboratory in California, used very similar techniques to make the measurements. They looked for a kind of explosion called a Type Ia supernova, occurring when an aging star destroys itself in a gigantic thermonuclear blast. Type Ia's are so bright that they can be seen all the way across the universe and are uniform enough to have their distance from Earth accurately calculated.

That's key: since the whole universe is expanding at a given rate at any one time, more distant galaxies are flying away from us faster than nearby ones. So Schmidt's and Perlmutter's teams simply measured the distance to these supernovas (deduced from their brightness) and their speed of recession (deduced by the reddening of their light, a phenomenon affecting all moving bodies, known to physicists as the Doppler shift). Combining these two pieces of information gave them the expansion rate, both now and in the past.

DARK ENERGY