The EPR zero-field splitting D for Fe3+ in LiTaO3 crystal has been calculat
ed from the microscopic spin-orbit coupling mechanism and the empirical sup
erposition model. It is found that to reach the good fit between the calcul
ated and observed values of D, the Fe3+ ion does not occupy the exact Li+ s
ite, but is required to move about 0.24 +/- 0.11 Angstrom in the direction
of the -c axis in LiTaO3 crystal. The result is very similar to that for Fe
3+ in the isomorphous LiNbO3 crystal obtained from the X-ray standing waves
experiment. The reasonableness-of the Fe3+ displacement is discussed.