After a force fi eld has been developed that is suitable for the calculation of molecular structures, it is fairly straightforward to extend it to cover assemblages of molecules, be they in the gas, liquid, or crystalline phase. We will start with the simplest cases fi rst. If we just put two acetone molecules together in the gas phase, with the general planes of the molecules parallel, so that the carbonyl groups are more or less one on top of the other and pointing in opposite directions, a molecular mechanics (or quantum mechanics) calculation will show that the two molecules are held together at substantially less than the van der Waals distances. The individual nonbonded interactions between the two molecules in MM4 calculations are the same (dipole -dipole and van der Waals) as those that would occur intramolecularly. The fact that they are between two different molecules really makes no difference. So if one wants to study gas -phase assemblages of molecules with MM4, in principle one does not have to do anything differently from the way it is done with a study of a single molecule, other than put the molecules together, and then optimize the total structure by minimizing the energy in the usual way. In practice, it may not be quite that simple.For even small assemblages of molecules, such as dimers and trimers, the multiple minimum problem may arise. That is, there may be different ways that one can put the molecules together and obtain minimum energy structures. One will probably want to