The impact fracture parameters of blends of nylon 6 and maleated ethylene-propylene rubber (EPR-g-MA) reinforced with glass fibers as a function of glass fiber and EPR-g-MA content were examined. Both the linear elastic fracture mechanics (LEFM) model and a modified essential work of fracture (EWF) model were used to analyze the data. It was found that the addition of EPR-g-MA to unreinforced nylon 6 increased the EWF parameters u(o) and u(d) defined by U/A = u(o) + u(d)l, where U/A is the total fracture energy per unit area and I is the ligament length. Beyond a critical rubber content, which coincided with the ductile-to-brittle transition, there was a large increase in u(d) When glass fiber reinforcement was used without rubber toughening, the EWF model was unable to model the observed fracture response. On the other hand, the LEFM model adequately described the fracture behavior, and it was found that the critical strain energy release rate, G(IC), increased with increasing glass fiber content. When both glass fiber reinforcement and rubber toughening were used, the u(o) increased with increasing EPR-g-MA or glass fiber content; whereas, u(d) increased with increasing ERR-g-MA content or decreasing glass fiber content.