Abstract
A nonthermal mechanism for driving coronal hole mass loss is proposed. Three empirical results are noted, namely (1) that polar plumes with short-lived X-ray bright points (XBP) at their base are sites of matter flow sufficient to account for the total solar mass flux, (2) that solar wind densities are positively correlated with the number of XBP in coronal holes, and (3) that XBP are associated with newly emerged magnetic flux. It is noted that since the dynamical development following the onset of nonequilibrium is a hitherto unsolved problem, subsequent features in the scenario described here are relatively speculative. It is proposed that bubbles of matter ejected from magnetic reconnection sites in polar plumes drive the solar wind in coronal holes.