The goal of this work is to understand how complex patterns of differentiation are established in epithelial cells during Drosophila development. Much of the focus is on the middle period of embryogenesis, from germ band extension to the onset of cell differentiation. During these stages, the segmental pattern established at the blastoderm stage has its first consequences on cell behavior. The maintenance of segmental pattern during these stages depends on cell communication mediated by the segment polarity genes. Homeotic identity of the individual segments must also be established during these stages so that cells can show their segment specific morphologies at differentiation. A substantial fraction of our work will focus on the segment polarity gene armadillo (=arm). arm is the homolog of plakoglobin, a vertebrate protein found in a variety of adhesive junctions. AT mid-embryonic stages arm protein staining accumulates in regions of the epidermis that express another segment polarity gene, wingless. We will investigate the cellular mechanisms responsible for this accumulation, and determine whether it represent increased levels, modifications, or changes in intracellular distribution. We will use heat shock constructs to manipulate the level of arm protein, and determine the consequences on the final embryonic phenotype. We will investigate the requirement for arm in those embryonic cells that do not show high arm staining and will extend our studies of the arm/wg signalling pathway to development of the larval epidermis and the imaginal discs. We will analyze the earliest cellular responses of the embryo to blastoderm pair-rule expression, using cell intercalation during germband extension. We will identify the pair-rule genes that affect germ band extension and use a variety of morphological techniques to characterize the effects of cell behavior in embryos mutant for those genes. To identify the downstream genes that mediate the cellular responses, we will examine the phenotype of homozygous deletion embryos and analyze enhancer trap lines which express betagal in pair-rule pattern during germ band extension. We will complete the cloning of extradenticle (=exd) which have shown to be necessary for the normal morphological consequences of homeotic gene expression. We will determine the expression pattern of exd in the embryo and in imaginal discs. We will begin a molecular analysis of the interactions of that gene with Ubx protein. We will also investigate the role of exd in patterning in the imaginal discs.