Permeability‐temperature curves of polycrystalline Mn0.41Zn0.42Fe2.16O4+xare interpreted in terms of theoretical models using the intrinsic magnetic properties as measured on a single crystal having virtually the same composition. The temperature dependence of &mgr; can be fitted by the relation &mgr;∼M/K1/2, as expected for cylindrical wall bulging, but the absolute value of &mgr; is an order of magnitude below the theoretical value. Spherical bulging of a totally pinned wall is shown to yield a &mgr; which is an order of magnitude smaller than the observed value due to the stiffness of the wall due to demagnetizing and anisotropy energy. It is suggested that a finite pinning of the walls at the grain boundary might explain our data. After annealing at pressures near the phase boundary, the theory does not fit the &mgr;‐Tcurves. It is suggested that the bulk composition of the grains of the polycrystalline material is different from that of the single crystal due to preferential segregation at the grain boundary even before a second phase is formed.