xii, 107 p. : ill. (some col.) A print copy of this thesis is available through the UO Libraries. Search the library catalog for the location and call number. / Non-muscle myosin II generates contractile forces for processes such as cytokinesis, motility, and polarity. Contractility requires assembly of myosin molecules into bipolar mini-filaments through electrostatic interactions between coiled-coil tail domains of the heavy chains. Analyses of myosin II from various organisms have revealed "assembly domains" within the C-terminal portion of the tail domain that mediate filament formation. However, it has been unclear precisely how assembly domains interact with one another, or otherwise contribute to tail-tail interactions, to form the bipolar mini-filament structure. To understand tail domain interactions, we first identified a 90-residue region (1849-1940) of the Drosophila non-muscle myosin II tail domain that was necessary and sufficient for filament formation, using salt-dependent solubility and a novel fluorescence energy transfer assay. We identified residues within this "assembly domain" that were critical for filament assembly by analyzing the effect of point mutations. We found that single point mutations in specific positively charged regions completely disrupt filament assembly. Surprisingly, none of the negatively charged regions within the assembly domain are required for assembly. Most of the mutations in positively charged residues that disrupted filament assembly clustered within a 15-residue segment (1880-1894) that appears to form a critical interaction surface. Using this information, along with known geometrical constraints and electrostatic calculations, we constructed a structural model of the bipolar mini-filament. This model features one favored anti-parallel tail overlap and multiple slightly less stable alternative overlaps. The ability of the positive segment to interact with multiple negative regions explains the lack of required negatively charged residues in the assembly domain. To our knowledge, this structural model of the non- muscle myosin II bipolar filament is consistent with all physical observations and provides a framework for understanding the detailed mechanism by which this fundamental cellular structure is generated. This dissertation contains previously published and unpublished co-authored material. / Committee in charge: Tom Stevens, Chairperson, Chemistry; Kenneth Prehoda, Advisor, Chemistry; J. Andrew Berglund, Member, Chemistry; Christopher Doe, Member, Biology; Karen Guillemin, Outside Member, Biology