To develop a specific-locus test (SLT) system for environmental mutagenesis using vertebrate species other than the mouse, we first established a tester stock of the fish medaka (Oryzias latipes) that is homozygous recessive at three loci. The phenotypic expression of these loci can be easily recognized early in embryonic development by observation through the transparent egg membrane. We irradiated wildtype males with '"Cs -rays to determine the dose-response relationships for dominant lethal and specific-locus mutations induced in sperm, spermatids, and spermatogonia. Through observation of322,666 loci in control offspring and 374,026 loci in offspring obtained from 0.64-, 4.75-, or 9.50-Gy-irradiated gametes, specific-locus mutations were phenotypically detected during early development. These putative mutations, designated "total mutation," can be recognized only in embryos of oviparous animais. The developmental fate of these mutant embryos was precisely followed. During subsequent embryonic development, a large fraction died and thus was unavailable for test-crossing, which was used to identify "viable mutations."Our medaka SLT system demonstrates that the vast majority of total mutations is associated with dominant lethal mutations. Thus far only one spontaneous viable mutation has been observed, so that all doubling calculations involving this endpoint carry a large error. With these reservations, however, we conclude that the quantitative data so far obtained from the medaka SLT are quite comparable to those from the mouse SLT and, hence, indicate the validity of the medaka SLT as a possible nonmammalian test system. The mouse specific-locus test (SLT), established primarily by W. L. Russell (1, 2), has been used extensively for assessment of genetic risks from environmental mutagens. Attempts to use other species of vertebrates, the guppy (3-5) and the zebrafish (6), for the SLT have been made. However, results obtained were not thorough enough to establish SLTs using these species. We propose that a fish, the medaka (Oryzias latipes), would provide a valid alternative SLT system to study environmental mutagenesis. The medaka is a small freshwater teleost native to Asian countries (Japan, Korea, China, etc.), whose basic biology including classical genetics was well established by Yamamoto (7). Furthermore, our preliminary data (8-15) and a preliminary report (8) on -irradiation-induced mutations in the male medaka at a specific locus, b, suggest that the medaka is a promising candidate for such a test. In our preliminary study (8), we found that many color mutant embryos, whose mutant phenotype could be recognized very early after fertilization, died during subsequent embryonic developmental stages but before hatching. Based on these observations, we proposed the terms "total mutations" for specific-locus mutations that were phenotypically detected during early development and "viable mutations" for hatched viable mutants. It should be noted that the studies on silkworm eggs by Tazima (16) we...