A recent paper by C. H. Henry and D. V. Lang claims that the adiabatic approximation breaks down in the neighborhood of the intersection of the adiabatic potential curves for the two electronic states, between which multiphonon transitions occur. It is shown that their claim is not justified; what they consider to be the sign of failure of the adiabatic approximation is no more than an indication of the fact that owing to the uncertainty principle, there is a finite neighborhood around the above mentioned point of intersection, throughout which multi-phonon transitions can occur. Direct calculation of the multi-phonon transition probability on the basis of the adiabatic approximation gives in fact a result identical with the result obtained with their version of the theory. Further discussions contend that in more general situations Henry and Lang’s formulation of the theory will no longer be applicable and a proper adiabatic approximation treatment by first lifting the degeneracy at the intersection point will be necessary.