G.W. Marcy (using results from a long-term project at Lick Observatory) explained that the uncertainty in Doppler measurements of spectra of solar-type stars is now down to ± 5 m/s, revealing that five have velocity variations indicating a Keplerian orbit with a definitely planetary (not brown dwarf) companion. The derived minimum masses for the companions are from 0.45 to 7 Jupiter masses. The orbital 3radii2 range from 0.04 AU (much closer to the primary than Mercury is to the Sun) to 2.1 AU, and two of the candidates exhibit non-circular (i.e. eccentric, probably elliptical) Keplerian orbits. Eccentricities are unexpected in close systems and are still to be explained.
Controversy arises from the soon-to-be-published Nature article by David Gray, 3The Star 51 Pegasi Has no Planet,2 in which Gray asserts that the velocity periodicities being detected are associated with stellar oscillations rather than with orbital motions. However, as Marcy points out, stellar oscillations would be expected to cause corresponding brightness changes in the star, and photometry (exact counting of total photons reaching an Earth-bound detector from the candidate star) at the milli-magnitude level shows no such variations. To get an idea of a millimagnitude, the brightest star you can see that is not the Sun has an apparent magnitude of -1, and the faintest that is visible to the average human eye is 6th magnitude. It1s hard to distinguish differences of 0.1 magnitude between two stars with an untrained eye, and 0.01 mag is too difficult even for bare-eyed professionals. Millimag = 0.001 mag difference. In addition, oscillations often have higher harmonics, or overtones (like a bell ringing), and none of those overtones show up in the Doppler periodograms, just the spike at the suspected orbital period of the candidate planet. Finally, the shapes of the 3lines2 in the stellar spectrum would be expected to change periodically if 51 Peg were oscillating. According to a poster paper presented at this meeting by Horner et al., neither 51 Peg not Tau Bootes show periodic variations in their line-profiles.
Dispassionate observers who attended the meeting were heard to remark
that the brightness variations due to oscillations would more likely
be at the 0.1 millimagnitude level, and therefore not observable from
current ground-based systems, so the "photometric variability"
argument still may not be resolved. Space-based platforms will be
capable of resolving this dispute in the near future.
Extrasolar Planets Page (from the Lunar and Planetary Lab at Arizona State U in Tucson)
Penn State University's Physics/Astronomy program has a well-organized page which appears to be quite up-to-date. (If you look at the "51 Peg" candidate, you'll see essentially the same material I covered in "Hot News In Cool Stars," only with more references and written in more-technical language.)
Here is a more-propagandized link to Jeff Marcy's "Extrasolar Planet
Search" group at San Francisco State U