The influence of temperature and strain rate upon the fracture toughness of structural steel is the question considered in this paper. The hypothesis is proposed that fracture toughness, KIc, for initial crack extension is a single-valued function of the rate parameter T ln A/ε˙. Measurements made by Krafft and Sullivan of fracture toughness, KIc, for three steels covering a range of low temperatures and high strain rates are presented as a function of this rate parameter. Two of the materials support the contention that a single-valued relation exists between KIc and the parameter, while scatter in the data for the third steel does not allow a conclusion. The complexity of design against fracture in structural steels is reviewed. For conservative design it must be assumed that a crack is present in the structure. Design against fracture must insure that the minimum fracture toughness at the service temperature, strain rate and stress is sufficient to prevent this crack from extending rapidly through the base metal. A means of using low temperature laboratory fracture toughness tests for estimating a minimum fracture toughness corresponding to the service conditions is discussed.