Japanese medaka, Oryzias latipes is widely distributed in the North East Asia including Korea, Japan and east China, and commonly used for freshwater toxicity tests and cytotoxicological studies worldwide. In this study, a series of experiments were conducted to identify the potential of the fish as a standard test species for saltwater toxicity evaluation such as marine receiving waters, ocean-dumped materials and sediment pore waters etc. Hatching, growth and mortality rates of the fish were estimated with the wide ranges of salinity from freshwater to seawater (35 psu). Direct exposure of the fertilized eggs in freshwater to the wide ranges of salinity (from 0 to 35 psu) without pre- acclimation to the saltwater revealed no significant differences in hatching rates by salinities (p =0.24). On the other hand, medaka larvae hatched in freshwater and exposed to saltwater directly showed high mortality at > 25 psu treatment groups (p < 0.0001). However, there was no significant difference in mortality of medaka larvae hatched in 13.8 and 14.2 psu at the wide ranges of salinities ($0~35$ psu). Growth rates of medaka larvae hatched in the above two salinities showed no differences in body length either from 0 to 35 psu treatment groups (p =0.64 for 13.8 psu group and p=0.32 for 14.2 psu group). The number of gill chloride cell in medaka larvae sharply increased when the larvae were exposed to high salinity. Reference tests with zinc chloride revealed 96h $LC_{50}=8.84(7.19~10.87)mg;L^{-1}$ using 7~10 day old medaka larvae. These were comparable or better sensitivity in comparison with the other standard test species such as North American sheepshead minnow Cyprinodon variegatus. Based on the results of these experiments, hatching rates and larvalmortality of medaka must be good toxicity parameters for seawater bioassay and the species seems to be a good standard species for both the freshwater and seawater toxicity test.