On phasing the small-angle x-ray diffraction pattern from nerve myelin.

Using a method they developed, Stamatoff and Krimm (1976) have phased swelling data from nerve myelin. Although most phases agree with those I determined previously, there are a few differences. In this letter the two different phasings, theirs and my own, are used to compute the corresponding electron-density profiles, which are then closely compared. For both phasings, small differences are seen in the membrane profile at different degrees of swelling. The explanation that these differences are due simply to errors in measuring intensity is shown to be quite improbable; thus the differences indicate a real change in the profile. It follows that the assumption of a constant membrane profile appears to be invalid in the case of myelin swelling. The differences therefore are assumed to indicate a real change in the profile. It is shown that this change can be attributed consistently to interdigitation of protein molecules at the surfaces of neighboring membranes, while the membrane structure itself remains unchanged. In this case, valid phases still can be determined by swelling, but the phases determined by Stamatoff and Krimm are not valid.