In this study we tested the feasibility of coded wire tags (CWTs) as a novel high-efficiency tagging method for Japanese flounder (Paralichthys olivaceus). A total of three size groups were tested. The total lengths for small, medium, and large size groups were (5.92±0.41) cm, (8.92±0.36) cm, and (12.06±0.62) cm respectively. In a short-term experiment, we marked hatchery-reared juveniles with CWTs in the dorsal muscle and examined whether the fish size affected the tag retention and tagging-induced mortality. Tag retention was above 96% in all the three size groups and appeared to be unaffected by the fish size. Tag loss was only observed within 4 days post tagging. Tagging-induced mortality for small juveniles (13%) was significantly higher than that for the medium (4%) and the large (1%) juveniles. The tag shedding rates for the small, medium, and large juveniles were 3.3%, 2.4%, and 0.7% respectively. Linear dependence models were established to describe the relationships of tagging-induced mortality, the total length, and the body thickness. Death of small marked juveniles usually occurred within 4 days post tagging. In order to reduce the post-tagging mortality and biases in stocking assessment, we recommended CWT tagging on juveniles at least 6 cm in body length. The CWT tagging may have no effects on the growth and swimming behavior of the fish, therefore could be an ideal tagging method for Japanese founder juveniles. It is potentially a good tool in large-scale tagging and evaluation of Japanese flounder in the future.