Abstract
It is shown that the tunneling conductance curves of Woolf and Reif for quench-condensed In-Fe alloy films do not agree with values which are calculated from the theory of Shiba and Rusinov, which is supposed to describe magnetic impurities that interact strongly with the conduction electrons. Possible reasons for the discrepancy are presented. It is also suggested that the theory should be modified to take strong electron-phonon coupling into account in relating the pair-breaking parameter to the impurity concentration for Pb-Mn alloy films. This would not affect the good agreement between the observed tunneling curves for Pb-Mn films and the theory, which Chaba and Nagi have found previously.