Planets may, in principle, exist around any type of star, but some stars are better than others as places to look just because they are amenable to our techniques. Stars substantially more massive than the Sun tend to have broad lines, for example, precluding radial velocity searches. They also tend to exhibit intrinsic variability. Stars much less massive than the Sun are very faint, limiting the attainable signal-to-noise for even the nearest examples.
A census of the nearby solar-type stars is in progress. It will soon use
colors and parallaxes from Hipparcos to define a volume-limited sample
of ~5,000 G dwarfs within 50 pc (i.e., down to a V magnitude of about 9). A
preliminary estimate of which stars these are has been made, based on existing
information. Additional observations are being made to determine which stars
are in binary systems, and it is also possible to estimate ages. Stars that
have stellar companions within ~10 AU are not promising targets for planet
searches, nor are the very young stars. Both qualities - duplicity and age -
weed out about 1/2 of the starting sample, meaning that ~1/4 of the stars are
suitable targets. That is about 1,000 stars, roughly, or 1 µSagan.