Abstract.-We tagged juvenile upper Yakima River hatchery spring Chinook salmon Oncorhynchus tshawytscha with passive integrated transponder (PIT) and coded wire snout tags in a double-tag study to test the assumptions that tags are not lost and do not affect postrelease survival, behavior, or growth. The average loss of PIT tags was 2.0% (95% confidence interval [CI] ¼ 0.7-3.2%) in juveniles before release and 18.4% in recaptures returning 6 months to 4 years after release (95% CI ¼ 17.2-19.5%). Adult tag losses were not significantly correlated with age of return (analysis of covariance, P ¼ 0.40), indicating that the majority of PIT tag loss had occurred within the first 6 months after release. Smolt-to-adult recruit survival (SARS) of PIT-tagged fish was significantly lower (P , 0.05) than that of non-PIT-tagged (NPT) fish because of tag loss and reduced survival, resulting in an average underestimate of SARS of 25.0%. After correcting for tag loss, we estimated PIT tag-induced mortality to be as great as 33.3% with a mean of 10.3% over all brood years (P , 0.05). Mean lengths and weights of PIT-tagged adults were less than those of NPT adults in all age comparisons. However, only age-4 PIT-tagged adults were significantly smaller than NPT fish of the same age (mean length difference ¼ 1.1 cm; mean body weight difference ¼ 0.1 kg; analysis of variance, P , 0.05). There was no significant difference between migration timing of PIT-tagged and NPT adults within the upper Yakima River (MannWhitney test, P . 0.09). Given the widespread and increasing use of PIT tags, and their use in calculating critical estimators related to salmonid life history of Endangered Species Act populations, the effects of using PIT tags must be quantitatively considered under actual study conditions and, if necessary, be accounted for.