We investigate the differences in the relation derived recently by Ferrares
e & Merritt and by M-.-sigma Gebhardt and collaborators. The shallower slop
e found by the latter authors (3.75 vs. 4.8) is due partly to the use of a
regression algorithm that ignores measurement errors and partly to the valu
e of the velocity dispersion adopted for a single galaxy, the Milky Way. A
steeper relation is shown to provide a better Dt to black hole masses deriv
ed from reverberation mapping studies. Combining the stellar dynamical, gas
dynamical, and reverberation mapping mass estimates, we derive a best-fit r
elation M-. = 1.30(+/-0.36) x 10(8) M-.(sigma (c)/200 km s(-1))(4.72(+/-0.3
6)).