Reply to cond-mat/0111504

Abstract

We show that the choice of the sign of the hopping matrix in our impurity band model for disordered III-V diluted magnetic semiconductors [PRL 87, 107293 (2000); cond-mat/0111045] is justified: with this choice, the impurity band is placed inside the gap and it has a mobility edge, as expected for a disordered system. The other sign choice, suggested in cond-mat/0111504, leads to an unphysical description of the occupied states of the impurity band (extremely long tail, no mobility edge, no bulk ferromagnetism).