Abstract. We have been investigating a set of genes, collectively called mups, that are essential to striated body wall mu...__scle cell _positioning in Caenorhabditis elegans. Here we report our detailed characterization of the mup-2 locus, which encodes troponin T (TnT). Mutants for a heat-sensitive allele, called mup-2(e2346ts), and for a putative null, called mup-2(upl), are defective for embryonic body wall muscle cell contraction, sarcomere organization, and cell positioning. Characterizations of the heat-sensitive allele demonstrate that mutants are also defective for regulated muscle contraction in larval and adult body wall muscle, defective for function of the nonstriated oviduct myoepithelial sheath, and defective for epidermal morphogenesis. We cloned the mup-2 locus and its corresponding cDNA. The cDNA encodes a predicted 405-amino acid protein homologous to vertebrate and invertebrate TnT and includes an invertebrate-specific COOH-terminal tail.The mup-2 mutations lie within these cDNA sequences: mup-2(upl) is a termination codon near NH2 terminus (Glu94) and mup-2(e2346ts) is a termination codon in the COOH-terminal invertebrate-specific tail (Trp342). TnT is a muscle contractile protein that, in association with the thin filament proteins tropomyosin, troponin I and troponin C, regulates myosin-actin interaction in response to a rise in intraceUular Ca 2÷. Our findings demonstrate multiple essential functions for TnT and provide a basis to investigate the in vivo functions and protein interactions of TnT in striated and nonstriated muscles. STUDIES of many organisms, including vertebrates, indicate that the stable attachment of muscle to the skeleton requires muscle sarcomere assembly, muscle contraction and extracellular matrix formation. Caenorhabditis elegans offers unique advantages for investigations of the cellular mechanisms influencing the establishment and maintenance of muscle attachment. C. elegans has a simple cellular anatomy of only 959 adult somatic ceils and contains a small number of muscle types (for review see Waterston, 1988). The cell divisions and migrations giving rise to these muscle types have been fully described at the cellular level (Sulston et al., 1983). Since the spatial relationships of individual muscle cells are easily visualized and are invariant, the attachments of muscle cells can be precisely determined in wild-type and mutant worms. Given these attributes, it is possible to study specific gene mutations to