This set of files includes a fortran listing of the program epmod.f, written by A. Bothner-By in the spring of 1998. The program has been compiled and run on a Sun Sparcstation1, and the Unix-based computers of the Andrew system at CMU, and has performed thus far without glitches. It accepts as input, the number of atoms, identification of each atom by its atomic number, and the Cartesian coordinates of each for some real or imagined organic structure such as a molecule, ion, complex or fragment. The program then determines the bonding pattern between atoms, based primarily on interatomic distances, and the nature of the bonds, based on bond lengths and the number of bonds to each atom. It then dissects the molecule into groups which have well determined anisotropic magnetic ssusceptibilities, and adds the anisotropic tensors together with due regard to the spatial orientation of each to obtain the anisotropic tensor for the structure, and its principal axes. Finally it calculates, given the proton spectrometer frequency and sample temperature, the value of the dipolar splitting expected for each NH or CH bond (as specified by the user), and writes a file of all results. At present the constituent atoms are limited to C,H,O,N, and S. Data on the anisotropic susceptibilities of P are almost non-existent, so it could not be included. Halogens, by virtue of their high polarizability, have extremely variable isotropic and anisotropic susceptibilities. I plan to add the capability to calculate HH dipolar splittings, perhaps only for hydrogens which are in close proximity, and to calculate the quadrupole splitting for deuterated structures. A more nebulous plan,is to add the capability to calculate the dynamic frequency shift observed for (N15-H) and (C13-H) pairs. This set of files also contains several sample input and output files, which can be used to check the operation of the program, and as models for the format of the input files. It is important to enter realistic spatial coordinates in the input file; larger or smaller bond-lengths than correspond to the type of bond will cause the bondor subroutine to go crazy and produce nonsense. I would be glad to receive any comments or criticism of this program. A. A. Bothner-By Dept of Chemistry Carnegie Mellon University Pittsburgh, PA 15213 ab6d@andrew.cmu.edu