Background: The morphogenic movements that characterize embryonic development require the precise temporal and spatial control of cell-shape changes. Drosophila dorsal closure is a well-established model for epithelial sheet morphogenesis, and mutations in more than 60 genes cause defects in closure. Closure requires that four forces, derived from distinct tissues, be precisely balanced. The proteins responsible for generating each of the forces have not been determined. Results: We document dorsal closure in living embryos to show that mutations in nonmuscle myosin 11 (encoded by zipper; zip/Myoll) disrupt the integrity of multiple tissues during closure. We demonstrate that Myoll localization is distinct from, but overlaps, F-actin in the supracellular purse string, whereas in the amnioserosa. and lateral epidermis each has similar, cortical distributions. In zip/Myoll mutant embryos, we restore Myoll function either ubiquitously or specifically in the leading edge, amnioserosa, or lateral epidermis and find that zip/Myoll function in any one tissue can rescue closure. Using a novel, transgenic mosaic approach, we establish that contractility of the supracellular purse string in leadingedge cells requires zip/Myoll-generated forces; that zip/Myoll function is responsible for the apical contraction of amnioserosa cells; that zip/Myoll is important for zipping; and that defects in zip/Myoll contractility cause the misalignment of the lateral-epidermal sheets during seam formation. Conclusions: We establish that zip/Myoll is responsible forgenerating the forces that drive cell-shape changes in each of the force-generating tissues that contribute to closure. This highly conserved contractile protein likely drives cell-sheet movements throughout phylogeny.