It is already known that about 10% of central stars of planetary nebulae (PNe) are very short-period binaries (hours to days), which are detected through photometric variations. These must have been formed through common-envelope interactions in initially wide binaries, accompanied by ejection of the envelope and its subsequent photoionization as a PN. Radial-velocity observations by ourselves and others are now suggesting that an even larger fraction of planetary nuclei may be spectroscopic binaries, making the total binary fraction very large. However, we have not as yet been able to rule out the possibility that the apparent velocity changes are actually due to stellar-wind variations. Pending follow-up spectroscopic observations with large telescopes, it presently appears plausible that binary-star ejection is the major formation channel for planetary nebulae.