The $\dot{M} - M$ relationship in pre-main sequence stars

We examine the recent data and analysis of Natta et al. concerning the accretion rate on to young stars as a function of stellar mass, and conclude that the apparently steep dependence of accretion rate on mass is strongly driven by selection/detection thresholds. We argue that a convincing demonstration of a physical relationship between accretion and stellar mass requires further studies which, as is the case for Natta et al., include information on upper limits, and which quantify the possible incompleteness of the sample, at both low and high accretion rates. We point out that the distribution of detections in the ($M, \dot{M}$)-plane can in principle be used to test conventional accretion disc evolutionary models, and that higher sensitivity observations might be able to test the hypothesis of accelerated disc clearing at late times.