A material test carried out at high speed may be markedly influenced by plastic-wave propagation effects. In such a case, a variation of strain occurs along the test specimen, and the stress-strain relation cannot be determined from measurements made on the specimen as a whole. If average values are taken, it is shown that a spurious strain-rate influence will be deduced when propagation effects first begin to appear as the testing speed is increased. The effect is due not to a true material strain-rate dependence, but to the appearance of strain variations along the specimen, and its magnitude depends upon the dynamics of the whole test arrangement. The theoretical plastic-wave analysis of a particular test arrangement is also given. The range of speed is determined which permits satisfactory interpretation without the need for detailed analysis of plastic-wave propagation. Application to other test arrangements is discussed, and it is pointed out that a theoretical analysis can, in general, be made to predict the permissible speed range in planning a high-speed testing program.