In 4 separate dominant lethal experiments groups of mice of either Charles River CD1 or Alderley Park strains were fed laboratory diets (Oakes, 41B, PRD, BP nutrition rat and mouse maintenance diet No. 1). The diets were either untreated (negative control diets) or irradiated at 1, 2.5 and 5 megarad and were freshly irradiated, or stored. The animals were fed their test diets for a period of 3 weeks prior to mating. Groups of mice given a single intraperitoneal injection of 200 mg cyclophosphamide per kg body weight served as the positive controls. Freshly irradiated PRD diet fed to male mice of both strains caused an increase in early deaths in females mated to the males in week 7 and to a lesser extent in week 4. The increase due to irradiation was small by comparison with that produced by the positive control compound. The responses for the other irradiated diets showed no significant increases in early deaths although some values for Oakes diet were high. The effect of storage was examined with PRD and BPN diet on one occasion and produced conflicting results. Thus there was some evidence that irradiated PRD diet has weak mutagenic activity in the meiotic and/or pre-meiotic phase of the spermatogenic cycle which appeared to be lessened on storage; the inclusion of such a diet in toxicological studies would therefore need to be carefully considered.