The velocity dispersion of galaxies on scales of r similar to 1 h(-1)
Mpc, sigma(12)(r), may be estimated from the anisotropy of the galaxy-
galaxy correlation function in redshift space. We present a reanalysis
of the CfA1 survey in which we correct an error in the original analy
sis of Davis & Peebles. We give a detailed breakdown of how the value
of sigma(12)(r) depends on various aspects of the way in which correct
ions for infall into the Virgo Cluster are applied. We find that sigma
(12)(r) is extremely sensitive to the details of these corrections. We
conclude that a robust value of sigma(12) cannot be obtained from thi
s survey.