Planet discovery retracted.

When astronomers invited Andrew G. Lyne to lecture on his dramatic report last July that a planet orbits a Milky Way pulsar called PSR1829-10 (SN: 7/27/91, p.53), they figured he would give a standard review of his work. That's what Lyne himself thought until about seven days before the talk, when he rechecked his latest radio data and uncovered an error.

At a meeting in Atlanta last week of the American Astronomical Society, the radioastronomer from the University of Manchester in England shocked an audience of his peers. Periodic delays and advances in the arrival times of radio waves from the pulsar, which had seemed to indicate that a planet with a six-month period orbited the object, were in part due to incorrect accounting for Earth's motion around the sun, he announced.

With that motion properly accounted for, the planet simply "evaporated," he said. A retraction appears in the Jan. 16 NATURE, the journal in which Lyne and his team initially reported their results.

Two errors conspired to create the appearance of a telltale, six-month variation in the radio signals, Lyne says. The pulsar's exact position proved difficult to pinpoint and researchers failed to insert the actual position of the object, once they had discovered it, into their calculations. That in turn magnified a normally negligible error in their analysis, which assumes that Earth has a circular orbit, rather than its actual elliptical path around the sun. Lyne notes that such an assumption had never caused an error in analyzing radio data from 300 other pulsars his team has observed.

Astronomers say that the retraction does not cast doubt on a more recent report that two, or possibly three, planets orbit another Milky Way pulsar (SN: 1/11/92, p.20). These radio signals have a complex, quasiperiodic pattern that Earth's motion cannot mimic. In addition, two radiotelescopes independently measured the position of this pulsar, minimizing the likelihood of errors.