I propose a model which accounts for the geometries and sequence in which compartmental boundary lines arise on the different imaginal discs, and on the blastoderm of Drosophila melanogaster; and propose that successive lines are recorded by different binary switches, to create a binary epigenetic code word specifying each disc, and disc compartment. I suppose a biochemical system undergoing reaction and diffusion acts throughout development. As an imaginal disc grows, a succession of differently shaped chemical concentration patterns form at a discrete set of disc sizes. I suppose a specific concentration of one chemical is a threshold. Concentrations above or below threshold switch cells to one or another of two commitments. Then the line across the imaginal disc with the threshold concentration is a predicted compartmental boundary. The sequence and geometries of such lines predict the compartmental boundaries seen on the wing disc, the other discs, and on the blastoderm stage egg. The compartmental lines on the wing disc suggest that a terminal compartment is specified by a combination of binary names recording a sequence of binary commitments: anterior, not posterior; dorsal, not ventral; wing, not thorax; proximal, not distal. Each combination comprises a binary epigenetic code word. Recently I constructed an independent model for transdetermination in Drosophila which proposed a similar binary epigenetic code for the different discs. The clone restriction lines predicted on the blastoderm by my transdetermination model, the chemical pattern model, and analogy with the wing disc, are nearly identical. Several are already confirmed. The resultant binary code scheme correctly predicts many relative transdetermination frequencies and accounts simply for the action of most homeotic mutants as genes which alter a single switch state in one or more discs.