Contact: Steve Bradt
University of Pennsylvania

In the most accurate picture yet of the makings of our universe, astronomers have determined that a measly 5 percent of its mass comes from the ordinary matter that makes up planets, stars and gases. The finding, by scientists at the University of Pennsylvania, the Institute for Advanced Study in Princeton, N.J., and the University of Colorado at Boulder, is scheduled for publication next month in the journal Physics Review D.

“Our universe is a very strange cosmic cocktail,” said lead author Max Tegmark, assistant professor of physics and astronomy at Penn. “The 95 percent of the universe thatís not matter like we see around us is matter that canít be seen at all ñ matter of a type that still mystifies astronomers and cosmologists.”

The report by Tegmark and his colleagues draws upon careful readings of light emanating from the cosmic microwave background, the faint afterglow of the Big Bang. This light comes from an opaque, ever-expanding wall of hydrogen and other matter spewed forth by the Big Bang, which delineates the observable universe and has been racing inexorably outward ever since our universe’s birth 14 billion years ago. The glowing inner surface of this wall of primordial matter holds many clues to the universeís origins.

The groupís determination that the universe is only about 5 percent ordinary matter confirms an earlier prediction based upon the manner in which light elements like helium and hydrogen scattered in the minutes immediately after the Big Bang.

“We now have two completely independent ways of coming up with the 5 percent figure,” said collaborator Matias Zaldarriaga of the Institute for Advanced Study, “one based on theories about the newborn universe and one relying on an understanding of the universe hundreds of thousands of years later.”

The 95 percent of the universeís mass thatís not ordinary matter is a stew of curious ingredients, all of it dubbed “dark” because astronomers canít yet see it. Tegmark and his collaborators suspect that roughly 33 percent is cold dark matter, a class of slow-moving matter that can be detected at this point only by the presence of its mysterious gravitational pull. Hot dark matter, primarily neutrinos ñ speedier, chargeless particles that also pass right through ordinary matter ñ appears to contribute a scant 0.1 percent of the universeís mass.

Most of the remaining 62 percent of the universe is apparently an even more puzzling type of matter known as dark energy. Like the two types of dark matter, dark energy canít be seen or touched and is known only by its gravitational pull. But unlike dark matter, which is thought to appear haphazardly throughout the universe, dark energy is believed to be uniformly distributed and is thought to be responsible for our universeís accelerating growth.

The first evidence of dark energy came only two years ago, when the behavior of certain supernovae suggested this accelerating expansion of the universe. This latest work is the strongest independent suggestion that dark energy actually exists.

“A few years ago, it was widely believed that the universe ran a ëbudget deficití and that this 62 percent of the cosmic energy budget wasnít even there,” said co-author Andrew Hamilton of the University of Colorado. “But according to Einsteinís theory of gravity, such a budget deficit would curve space much more than has been observed, so that possibility is now excluded.”

According to Einstein, any curving of space should reflect the amount of matter in the universe. Sphere-like curvature, which magnifies distant objects, would suggest the presence of relatively more matter; saddle-like curvature would indicate the opposite. Recent microwave background observations revealed that space was essentially flat, suggesting that our universe’s energy budget is, in fact, balanced.

Tegmarkís team used a three-dimensional map of the galaxy distribution in a sphere 4 billion light-years in diameter. The galaxies within this colossal sweep of space were scanned by the NASA/NIVR/SRC Infrared Astronomical satellite and a team centered at the University of Edinburgh.

The group fitted the resulting data to 11 cosmological parameters, calculating theoretical predictions for billions of different models on Penn computers. They developed a new way of making these theoretical predictions about 1,000 times faster, but without any loss of accuracy.


Tegmark, Zaldarriaga and Hamiltonís research was funded by the National Aeronautics and Space Administration, the National Science Foundation and the Penn Research Foundation.

NOTE: Tegmarkís paper is available on Physics Review Dís web site,; when prompted, enter volume number 63, page number 043007.