The term dominant lethal may be defined as death of the heterozygote arising through multiple chromosomal breaks. The assay is generally conducted by treating male animals, usually mice or rats, acutely (1 dose), subacutely (5 doses), or over the entire period of spermatogenesis. Animals treated acutely or subacutely are mated at weekly intervals to females for a sufficient number of weeks to cover the period of spermatogenesis. Those treated for the entire spermatogenic cycle are mated for 1 or 2 successive weeks at the termination of treatment. Females usually are killed at 14 days of pregnancy and examined for the number of total implantations in the uterus, the number of implantations classified as early deaths, and, in some cases, the number of corpora lutea. The category of early death is the most significant index of dominant lethality. A total of 249 papers were reviewed and 140 chemicals were evaluated. Of the 140 chemicals, 65 were positive by the criteria used by the Work Group in evaluating each publication. The category of "positive" includes those responses of a borderline nature. 99 chemicals were declared negative. There is considerable overlap of chemicals in both categories, which accounts for the incongruity in the total number of chemicals tested and the number considered positive and negative. A total of 44 animal carcinogens have been tested in the dominant lethal assay, 26 of which were positive and 18 negative for a correlation of 59%. The role of the assay should be that of confirming positive results from lower tier chromosomal aberration-detecting systems (confirming in the sense of indicating the ability of the chemical to penetrate gonadal tissue and to produce cytogenetic damage). The dominant lethal assay should not be used as a risk assessment method.