Even 15 years ago, "everyone agreed that the universe was expanding," he explains. Evidence for that expansion came via observations dating to the 1920s that no matter where astronomers turned their telescopes, galaxies were receding. The more distant the galaxy, the faster the pace.
Like Magic Marker dots on an expanding balloon, galaxies were being carried along by the expansion of space-time itself.
Early on, this led to the idea that the universe began in an enormous release of energy, dubbed the Big Bang. Over the years, the idea has been refined, but the general outline has been upheld by increasingly sophisticated observations, which put the event at some 13.8 billion years ago.
As scientists filled in the picture of the universe's beginning, however, its future was still open to debate, which centered on two possible paths.
If the universe held just the right density of matter and energy, the combined gravity of everything in the cosmos would allow the universe to expand forever, but at pace that would continually slow, though it would never reach zero. The other view posited that gravity ultimately would win out and throw the expansion into reverse. The universe would contract until everything collided in a cataclysmic "Big Crunch."
For a variety of reasons, most cosmologists tended to back the idea of an endless slowdown in the expansion rate. An accelerating universe was not on the cosmological radar screen.
Two teams tried to resolve the question by observing the light from exploding stars, known as supernovae, appearing at the farthest distances their telescopes would allow them to observe. Dr. Perlmutter led one team. Dr. Schmidt led the other, which included Dr. Riess.
One type of supernova in particular, known as a type 1A supernova, shows a common peak brightness wherever it pops up. By studying the spectra and the brightness of some 50 type 1A supernovae at enormous distances, the two teams independently found that the supernovae were dimmer than should be the case if the universe's expansion rate were slowing. After crunching the numbers, the data showed that the expansion was speeding up.