Abstract Two methods that determine the critical J- integral ( J c ) for crack initiation in the presence of remote energy dissipation are compared. The locus method of partitioning the energy for crack propagation from the total deformation energy is based on energy release rate interpretation of J- integral . This method is a reliable means to measure J c value in the presence of remote energy dissipation. On the other hand, the extrapolation method is based on an assumption that the total energy absorbed can be thought of as the sum of the energies absorbed in the crack tip region and the region away from it, and thus the effect of remote energy dissipation on the critical J- integral value can be eliminated by evaluating the value using the crack tip energy only. However, there exists a theoretical problem in the extrapolation method, and the problem is discussed. A comparison of these two methods is attempted based on the same load-displacement records of a highly deformable thermoplastic rubber, and advantages of the locus method are presented.