We do not believe the calculations of Auerbach and Loc in Ref. N
predicts the sign of Mf/Mgt for 47K. The simple schematic model of analog-antianalog mixing has an overall relative possible minus sign between Eqs. 1 and 2.
Although the sign between the antianalog fraction of the bound 1/2+ state
and the rest of the wf should be given by RPA, RPA has not been done for
this nucleus.