Because of our relatively low spectral resolution, we compare our observations with Gray's line bisector data by fitting observed line profiles to an expansion in terms of orthogonal (Hermite) functions. To obtain an accurate comparison, we model the emergent line profiles from rotating and pulsating stars, taking the instrumental point spread function into account. We describe this modeling process in detail. We find no evidence for line profile or strength variations at the radial velocity period in either 51 Peg or in Tau Boo. For 51 Peg, our upper limit for line shape variations with 4.23-day periodicity is small enough to exclude with 10 sigma confidence the bisector curvature signal reported by Gray & Hatzes; the bisector span and relative line depth signals reported by Gray (1997) are also not seen, but in this case with marginal (2 sigma) confidence. We cannot, however, exclude pulsations as the source of 51 Peg's radial velocity variation, because our models imply that line shape variations associated with pulsations should be much smaller than those computed by Gray & Hatzes; these smaller signals are below the detection limits both for Gray & Hatzes' data and for our own. Tau Boo's large radial velocity amplitude and v*sin(i) make it easier to test for pulsations in this star. Again we find no evidence for periodic line-shape changes, at a level that rules out pulsations as the source of the radial velocity variability. We conclude that the planet hypothesis remains the most likely explanation for the existing data.

Spectroscopic observations of 51 Pegasi and tau Bootis show no periodic changes in the shapes of their line profiles; these results for 51 Peg are in significant conflict with those reported by Gray and Hatzes (1997). Our detection limits are small enough to rule out nonradial pulsations as the cause of the variability in tau Boo, but not in 51 Peg. The absence of line shape changes is consistent with these stars' radial velocity variability arising from planetary-mass companions.

We report the discovery of near-sinusoidal radial velocity variations of the G0V star rhoCrB, with period 39.6 days and amplitude 67 m/s. These variations are consistent with the existence of an orbital companion in a circular orbit. Adopting a mass of 1.0 M(Sun) for the primary, the companion has minimum mass about 1.1 Jupiter masses, and orbital radius about 0.23 AU. Such an orbital radius is too large for tidal circularization of an initially eccentric orbit during the lifetime of the star, and hence we suggest that the low eccentricity is primordial, as would be expected for a planet formed in a dissipative circumstellar disk.