An extended one-generation study is proposed to replace the standard two-generation study, and would eliminate breeding F1 animals unless triggered by effects on parental reproduction or pup development. Nine two-generation studies in rats were reviewed to determine whether reproductive or developmental toxicity was more pronounced in the F2 generation. Three studies lacked reproductive effects, one had reproductive toxicity only in parental animals, three had mostly equivalent effects between generations, and two showed greater toxicity in the second generation. In each study with relevant effects, criteria proposed by Cooper et al. [Cooper RL, Lamb IV JC, Barlow SM, Bently K, Brady AM, Doerrer NG, et al. A tiered approach to life stages testing for agricultural chemical safety assessment. Crit Rev Toxicol 2006;36:69-98.] were applied to determine whether additional breeding would be triggered in the extended one-generation design. Additional mating was triggered for all but one study with more pronounced F1 reproductive toxicity. Thus, while the extended one-generation design may be a useful substitution for the two-generation study, more stringent criteria must first be developed to determine whether additional mating is actually needed. (C) 2009 Elsevier Inc. All rights reserved.