Abstract Instrumented impact testing has been used to measure the dynamic fracture-initiation toughness of some precipitation hardened aluminum alloys. The results of these tests have been calculated by the energy method GIc = ω/ηA and by the K calibration formula. The difference in origin and interpretation between the initiation toughness GIc = ω/ηA and specific fracture energy ω/A has been considered. Double-cantilever beam (DCB) specimens of these alloys have been used to measure point values of fracture toughness by the compliance method and average propagation toughness by the specific surface energy method. Fracture propagation is of the semistable type in these non-rate-sensitive alloys, and the toughness values are found to be similar by both methods. An exception is when crack tunnelling develops during crack propagation in thick DCB specimens. Then KIc from the surface energy method tends to be lower than that from the compliance method.