It was a long time ago, and it was a galaxy far, far away, but it's doubtful that any Ewoks, Hutts or Wookies would have had time to evolve there. In fact, the galaxy in question is so far away, and the distance its light must travel to reach Earth so vast, that astronomers see the galaxy as it appeared more than 13 billion years ago, when the universe was just 3 or 4 percent its present age.

The galaxy, which goes by the prosaic name of UDFj-39546284, has just been located in a long-exposure image of the Hubble Space Telescope meant to identify just such faint, distant objects. The find has yet to be confirmed with other observations, but if it stands up, the galaxy would be the most distant object ever identified. A team of U.S.- and Europe-based researchers announced the discovery in the January 27 issue of Nature. (Scientific American is part of Nature Publishing Group.)

When dealing with exceedingly distant cosmological objects, astronomers rely on a measure known as redshift, which quantifies how much an object's emitted light has been stretched to longer, or redder, wavelengths as the object recedes away from us in an expanding universe. Thanks to a 2009 upgrade by NASA astronauts, who delivered a powerful new camera to Hubble, the orbiting observatory has been reaching back to redshifts never before observed. In October a group of researchers announced that they had used a long-exposure Hubble campaign, known as the Hubble Ultra-Deep Field 2009 (HUDF09), along with a ground-based telescope to confirm the presence of a galaxy at redshift 8.55, about 600 million years after the big bang. The newfound galaxy UDFj-39546284, also from HUDF09, has a redshift of approximately 10, placing it about 100 million years closer to the dawn of the universe.

"You're essentially reaching back something like 97 percent of cosmic time, back to the very beginning," says astronomer Rychard Bouwens of Leiden University in the Netherlands, the first author of the new paper. One of the most remarkable features of Hubble's recent finds is that they are dim, somewhat humdrum galaxies, not luminous cosmic beacons such as quasars or gamma-ray bursts, which used to dominate high-redshift astronomy.

The combination of Hubble's perch above the shroud of Earth's atmosphere, the space observatory's new instruments, and its repeated pointing at the same patch for dozens and dozens of hours allowed the researchers to identify the faint smudge of a galaxy. Bouwens and his colleagues used the common technique of looking at an object with a number of wavelength-specific filters to find the point at which the object disappears from Hubble's imagery. If the dropout comes from neutral hydrogen between galaxies absorbing the light before it reaches Earth, researchers can use the known absorption property of hydrogen to infer the object's redshift.

But UDFj-39546284 appears in only one wavelength band, and verifying the galaxy's presence with another observatory may prove impossible until Hubble's successor, the James Webb Space Telescope (JWST), takes flight later in the decade. The researchers calculated that there is a 20 percent chance that the galaxy is not what it seems, an estimate that Bouwens calls "super conservative." UDFj-39546284 could be nothing at all—a mere glitch in the data—or it could be a nearby object masquerading as a distant galaxy. "The source is faint enough that we don't have a crystal-clear spectrum, so a lower-redshift object could look kind of like this candidate if there was a conspiracy in the noise," Bouwens says. "We're pushing the limits, and we're trying to do the best we can," he adds. "From my experience in doing this procedure it looks like this one could be the real thing."

Whether the galaxy proves to be real or not, Bouwens and his colleagues used the relative dearth of visible galaxies found in the HUDF09 at redshift 10 to set some limits on galactic luminosity and rates of star formation at that epoch, important markers of how the universe was developing. For if there were a plethora of bright galaxies 500 million years after the big bang, Hubble should have seen more of them. "Astronomers are often involved in these fishing expeditions, and this is the first time that we thought it might yield some fish," Bouwens says. "That's why the upper limit is meaningful."

The group found that the universe was in the midst of major changes at that time: There was a huge build-up of galaxies and a tenfold boost in star formation rates between redshift 10 and redshift 8, less than 200 million years later. But the rapidity of change raises an eyebrow for Matthew Lehnert of the Paris Observatory, lead author of the October paper announcing the redshift 8.55 galaxy. "That's the blink of an eye, 100 million or 200 million years," Lehnert says. "To have a change that big in that short of time kind of worries me." Perhaps Hubble's field of view was simply a bit barren, Lehnert ventures—if a neighboring patch of the sky contained a few more galaxies at redshift 10, the observed evolution would look to be smoother.

In fact, an earlier version of Bouwens and his colleagues' paper based on a smaller set of Hubble data contained three possible high-redshift galaxies, but with the addition of more observations the new, more compelling candidate emerged. The appearance or disappearance of plausible objects over time "just tells you how difficult it is when you have low signal to noise," Lehnert says, adding that Bouwens and his colleagues have made a reasonable case for the newfound galaxy. "These are really good researchers," Lehnert says. "Rychard really is the guy that does this kind of work."

As redshifts inch ever higher, astronomers are approaching the limits of how far back the current generation of telescopes can reach. "This is one of the deepest images ever taken, and they only have one good candidate, so you can see that we're sort of against the wall," Lehnert says. That only raises the anticipation for JWST, which a congressionally requested analysis recently found to be over-budget and behind schedule. The telescope will not launch before 2015 under the best of circumstances, the report concluded, and the actual launch date could be much later than that.