This paper presents a theoretical study of the packing of particles with a mixture size distribution. The theoretical treatment of this packing system is first rationalized. It is shown that the porosities of this packing system can be satisfactorily predicted by the linear-mixture packing model. The packing of binary and ternary mixtures of lognormal distributions is then generally discussed. The results indicate that the packing of particles with a mixture size distribution may be quite different from that with a discrete size distribution. The investigation is also extended to multi-component mixtures with particular reference to the packing systems encountered in engineering practice. Finally, the methods of determining the optimum particle size distribution for the maximum packing density of a given packing system are demonstrated by a typical practical example.