Knowing the location of the fracture planes when applying the freeze-fracture technique is of paramount importance when making use of the observations made to deduce the structure of cellular membranes. Freeze-fracture analysis of a number of membranes with very different functions that was performed in our laboratory, had led to the conclusion that there was no unique location of the fracture plane relative to the membranes. The most common type of fracture face appeared to expose the surfaces of membranes and not their interior. This made it justified to subject to a critical examination the commonly accepted concept of a splitting of membranes through their middle during fracturing. This examination led to the conclusion that the experiments that have been referred to as proving the intramembrane splitting can be severely criticized, and in fact can be used to prove that intramembrane splitting has not occurred. The layer that appears after etching can be identified as a layer of contamination formed by solutes that are freed by the sublimation of water. It can be ruled out as being a layer belonging to the membrane. An approach to the problem of locating the fracture faces is described which is based on a refusal to accept any preconceived attitude. Instead, the location of fracture planes should be determined individually in each case on the basis of observations pertaining to that case. Certain rules are presented for locating the fracture faces. Furthermore, the experimental conditions for freeze-fracturing, the interpretation of the topography of the fracture faces, the problem of complementarity, and the limitations of the method with respect to resolution are discussed briefly. The conclusion is that the freeze-fracture technique makes use of only a fraction of the resolving power of the electron microscope and that it is pertinent that other preparatory methods are worked out with the aim of making a further use of the performance of the electron microscope. © 1979 Academic Press, Inc.