The dependence of switching rate on the crystal thickness has been measured by Merz and explained by him in terms of a surface layer which has a low dielectric constant, and is about 10−4cm thick. While not explicitly stated in Merz' original arguments, the layer must have a reversible polarization. If a layer without a reversible polarization is assumed, instead, and the discontinuity of the normal component of polarization at the interface between the layer and the bulk is taken into account, then a much thinner layer (∼1 atomic thickness) will explain the thickness dependence. This layer can be taken to be very lossy, so that it has a relaxation time (for the disappearance of electric fields) short compared to the switching time, and yet the layer will still be completely effective in slowing down domain wall motion.