To determine the positive and negative classification error rates associated with the HTA in our laboratory, F1 sons of TEM-exposed CD-1 male mice were evaluated by the sequential fertility method with subsequent cytogenetic analysis. Males who sired three litters of size 10 or less when mated to primiparous females from either the B6C3F1 or the BCF1 strain were classified as partial steriles. When meiotic chromosome analyses revealed the presence of at least two cells containing multivalent figures, males were classified as translocation heterozygotes. When the fertility evaluation and the cytogenetic analysis were compared, normal fertility was observed on 5 of 83 (6.02%) translocation-bearing F1 males mated to B6C3F1 tester females and on 3 of 83 (3.61%) F1 males mated to BCF1 tester females. Thus, the false-negative error rates were 6.02% and 3.61% with these two tester strains. Multivalent figures were not observed in the meiotic chromosomes of 410 F1 males. Of these, 12 (2.93%) had reduced fertility when mated to the B6C3F1 tester strain as did 7 (1.71%) mated to the BCF1 strain. Thus, the false-positive error rates with these two tester strains were 2.93% for the B6C3F1 strain and 1.71% for the BCF1 strain. Our results indicate that non-zero error rates, both false-positive and false-negative, are associated with the sequential mating method HTA. In addition, the magnitude of these error rates was influenced not only by the tester female strain but also by the genotype of the F1 male.