Don't pack your bags just yet for Gliese 581 g, the potentially habitable extrasolar planet that was announced in September.

No sooner had a team of American astronomers, led by veteran planet hunters Paul Butler and Steven Vogt, announced their discovery than competing groups and independent researchers dove into data sets both public and proprietary to see if they, too, could find the signature of the newfound world. A new such analysis has come up empty in verifying the existence of Gliese 581 g. Only time will tell if it is really there.

If it exists, the planet is a smallish world some three times the mass of Earth and orbits Gliese 581, a dim star about 20 light-years from the sun. Hundreds of worlds are now known that populate solar systems other than our own, but Gliese 581 g made waves because its orbit would place it squarely in its star's habitable zone—the right distance from the star so that liquid water could exist on the planet's surface.

A competing European team reported just weeks after the planet's unveiling that their data, including some that had not yet been published, did not offer any corroboration of Gliese 581 g. And now, Philip Gregory of the University of British Columbia, analyzing data from Butler and Vogt's team as well as the European competitors, has come to an even more dismissive conclusion. His work has been submitted to the Monthly Notices of the Royal Astronomical Society.

Vogt and Butler combined their own 11-year set of observations with 4.3 years of published data from the Europeans and found evidence for six planets in the Gliese 581 system, including the potentially life-enabling Gliese 581 g. (Four of the planets were already known.) But Gregory, applying a statistical analysis to the same data from both the Americans and Europeans, scored a reliable detection of only two planets, or four with a little tweaking. By isolating the European data set, he found the signal of five planets orbiting the star—none of which fits the description of Gliese 581 g. (It is possible that measurement errors in the Americans' data may be underestimated, Gregory speculated.) The six-planet system interpretation of Vogt and Butler is almost certainly flawed, according to the new paper—Gregory calculated that the six-planet detection had a 99.9978 percent chance of being false.

So, Vogt and Butler's analysis suggests one thing, and Gregory's analysis suggests quite another, but the fact remains that both studies rest on the same observational data. The existence (or not) of Gliese 581 g probably won't be resolved until more data become available, starting with the next batch of observations from the Europeans. And by the time the dust settles, which could take years, many more potentially habitable worlds may have been discovered—scientists associated with NASA's Kepler mission are currently racing to do just that.

Artist's depiction of Gliese 581 planetary system: Lynette Cook