We derive the rates of capture { \dot { N } } of main sequence turn off stars by the central massive black hole in a sample of galaxies from Magorrian et al .
1998 .
The disruption rates are smaller than previously believed with { \dot { N } } \sim 10 ^ { -4 } -10 ^ { -7 } per galaxy .
A correlation between { \dot { N } } and black hole mass { M } is exploited to estimate the rate of tidal disruptions in the local universe .
Assuming that all or most galaxies have massive black holes in their nuclei , this rate should be dominated by sub- { L _ { * } } galaxies .
The rate of tidal disruptions could be high enough to be detected in supernova ( or similar ) monitoring campaigns—we estimate the rate of tidal disruptions to be 0.01 - 0.1 times the supernova rate .
We have also estimated the rates of disruption of red giants , which may be significant ( { \dot { N } } \ga 10 ^ { -4 } \hbox { y } ^ { -1 } per galaxy ) for { M } \ga 10 ^ { 8 } { M _ { \sun } } , but are likely to be harder to observe—only of order 10 ^ { -4 } times the supernova rate in the local universe .
In calculating capture rates , we advise caution when applying scaling formulae by other authors , which are not applicable in the physical regime spanned by the galaxies considered here .