Peering at black holes: an eventful look.

Twisting space-time and devouring anything--even light--that gets too close, black holes rank among the most bizarre objects in the cosmos. Theory predicts that these gravitational beasts are surrounded by a so-called event horizon, a one-way threshold through which things can fall in but nothing can get out.

Two studies reported at last week's meeting of the American Astronomical Society (AAS) in San Diego provide new evidence for event horizons.

Analyzing a trove of data collected nearly a decade ago by the Hubble Space Telescope, astronomers observed what seems to be the last gasp emitted by gaseous material spiraling into Cygnus X-1, a suspected black hole 6,000 light-years from Earth.

Joseph F. Dolan of NASA's Goddard Space Flight Center in Greenbelt, Md., and his team found that blobs of hot gas orbiting Cygnus X-1 radiated pulses of ultraviolet light that grew fainter and more rapid and then disappeared.

These fading pulses are just what astronomers expect from gas about to enter an event horizon. Light emitted from the gas grows dimmer because the black hole's gravity shifts the light to longer and longer wavelengths, so the radiation vanishes from view before it actually reaches the event horizon. In contrast, when material crashes into the surface of a star, it releases all its energy and becomes brighter rather than fainter.

The gravitational bending of light causes the radiation to appear pulsed, like a lighthouse beacon, as the gas orbits the black hole thousands of times a second. As the gas spirals ever closer to the hole, it speeds up and its pulses of light quicken until they die out.

William R. Stoeger, a Jesuit priest and astrophysicist now at the University of Arizona's Vatican Observatory Research Group in Tucson, predicted such a scenario in 1980. Although the new study corroborates Stoeger's work, Dolan cautions that his team found just two trains of pulsed ultraviolet light among the myriad data gathered by Hubble's high-speed photometer during the summer of 1992. Astronauts removed the detector in 1993 to make room for the optics required to correct for Hubble's flawed mirror.

"What we have is more in the nature of a lead" than an actual fingerprint of an event horizon, Dolan said at the AAS meeting. Random fluctuations in the Hubble data could mimic the expected swan song of matter falling into a black hole, he notes. Follow-up studies with an ultraviolet detector on Hubble and on craft yet to be launched could firm up the findings, Dolan adds. He and his colleagues are also searching for pulses in data taken by NASA's Rossi X-Ray Timing Explorer.

In another study reported at the San Diego meeting, researchers used NASA's Chandra X-Ray Observatory to study two types of X-Ray-emitting novas. One type consists of a suspected black hole siphoning gas from a sunlike star that orbits it. In the other type, an ultracompact star called a neutron star devours the nearby gas.

Although both types of nova consumed matter at a similarly slow rate, those suspected of harboring black holes radiated only 1 percent as much light as novas with neutron stars did.

"Seeing just this tiny amount of energy escape from the [candidate] black hole is like sitting upstream, watching water seemingly disappear over the edge," says study collaborator Ramesh Narayan of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Mass. He rates this scarcity of radiation as the strongest evidence yet for event horizons.

Roger D. Blandford of the California Institute of Technology in Pasadena is cautious about both studies. "Although I personally don't doubt that we are dealing with black holes in these systems, I think there are other interpretations of what is going on," he says. For instance, the regions surrounding black holes in the novas could be faint because of some process that expels most of the approaching gas before it reaches the event horizon, Blandford notes.