Impressed by the widespread misunderstanding of the issue , we return to the old question of the location of the inner edge of accretion disk around black hole .
We recall the fundamental results obtained in the 1970 ’ s and 1980 ’ s by Warsaw and Kyoto research groups that proved , in particular , that the inner edge does not coincide with the location of the innermost stable Keplerian circular orbit .
We give some novel illustrations of this particular point and of some other fundamental results obtained by Warsaw and Kyoto groups .
To investigate the flow dynamics of the inner edge of accretion disk , we carefully solve the structure of the transonic flow and plot the effective potential profile based on the angular-momentum distribution calculated numerically .
We show that the flow does not have a potential minimum for accretion rates , { \dot { M } } \gtrsim 10 L _ { E } / c ^ { 2 } 
( with L _ { E } being the Eddington luminosity and c being the speed of light ) .
This property is realized even in relatively small viscosity parameters ( i.e. , \alpha \sim 0.01 ) , because of the effect of pressure gradient .
In conclusion , the argument based on the last circular orbit of a test particle can not give a correct inner boundary of the super-critical flow and the inner edge should be determined in connection with radiation efficiency .
The same argument can apply to optically thin ADAF .
The interpretation of the observed QPO frequencies should be re-considered , since the assumption of Kepler rotation velocity can grossly over- or underestimate the disk rotation velocity , depending on the magnitude of viscosity .