Laboratory bioassays and field experiments were conducted to study the effects of entomopathogenic nematodes on squash vine borer (SVB), Melittia cucurbitae (Harris) (Lepidoptera: Sesiidae). In the laboratory, SVB larvae were exposed to Steinernema riobrave TX strain, S. feltiae SN strain, S. carpocapsae All strain, S. carpocapsae Sal strain, Heterorhabditis bacteriophora Hb strain, and H. sp. Hbl strain. Steinernema riobrave was the bestperforming nematode (overall), producing earlier and greater rates of mortality than the other strains. Species and strains of Steinernema were more effective than either heterorhabditid nematodes. Field experiments were carried out on yellow crookneck squash, Cucurbita pepo, in six trials over the course of 3 years (1997-99) in Clemson, South Carolina. The same nematodes as used in lab bioassays except S. carpocapsae Sal strain and H. bacteriophora Hb strain were compared to an insecticide standard, endosulfan (Phaser), and an untreated check. Two application methods were employed: soil and stem spray, and injection into the main stem. Infection of SVB larvae in the main stem of squash plants ranged from 19-61% in 1997, 9-27% in 1998, and 0-2% in 1999 trials. In the second trial of 1997, the injection treatment did not differ from the untreated check through the entire trial. However, the soil and stem sprays of both nematode species provided control equal to that of endosulfan. In the first trial of 1998, both S. carpocapsae and S. feltiae were equal to endosulfan in minimizing damage. As in 1997, the injection method did not perform well. In the first trial of 1999, S. carpocapsae and S. riobrave were not statistically different from the untreated check. In second trial, S. riobrave differed significantly from the untreated check and provided the same protection as endosulfan. Steinernematid nematodes thus may be promising biological control agents against SVB.