Paracrystalline lattice distortions in the nerve myelin membrane system lead to the distortion of the Q function of the membrane stack and the Qo function of the unit cell. The Q function can be expressed as a convolution polynomial of the electron density distribution and the distance-statistics functions between neighbouring membranes. Fitting this model Q function to the experimental Q function in a non-linear least-squares refinement procedure, one obtains the electrondensity profile of the myelin stack and the distance-statistics functions. Elimination of the distance statistics permits the calculation of the undistorted Qo function, the deconvolution of which makes possible a unique determination of the electron density distribution of the unit cell.

Varying degrees of distortion of the Q functions of myelin samples in different states – native, fixed, swollen, dehydrated etc. – suggest that disorder in the membrane system may appropriately be classified into intrinsic and nonintrinsic components. Quantitative estimates of the paracrystallinity in such biological systems (especially in the native state) from small‐angle X‐ray and neutron scattering experiments may not be directly correlated with similar measurements from electron microscopy. This is because preparative procedures for electron microscopy such as fixation and dehydration not only cause additional disorder but also induce in the membrane system structural changes which are not yet fully understood.

Recent investigations point to the presence of paracrystalline lattice disorder in the nerve myelin in the swollen, fixed and to some extent also in the native state. This leads to the distortion of the Q-function of the membrane stack and the Q0-function of the unit cell, which can be ex­perimentally isolated through swelling.Representing the distance statistics between neighbouring membranes and the electron density distribution of the single membrane with Gaussian functions, the Q-function can be expressed analytically as a convolution polynomial of the electron density, its mirror image and the distance statistics functions. Fitting this model (^-function to the experimental Q-function obtained by the inverse Fourier transformation of the scattered intensity, one can determine the optimal parameters of the electron density distribution and the distance statistics functions. The elimination of the distance fluctuations between neighbouring membranes permits the calculation of the undistorted Q-function or the Q0-function. The Fourier Analytical Deconvolution of the undistorted Q0-function enables a unique determination of the double membrane profile, thus a unique phase determination.