Acoustic telemetry was used to obtain the movement histories of 915 juvenile fall-run Chinook salmon (Oncorhynchus tshawytscha) through the lower San Joaquin River and Sacramento-San Joaquin Delta, California, in 2008. Data were analyzed within a release-recapture framework to estimate survival, route distribution, and detection probabilities among three migration pathways through the Delta. The pathways included the primary route through the San Joaquin River and two less direct routes (Old River and Turner Cut). Strong inferences about survival were limited by premature tag failure, but estimates of fish distribution among migration routes should be unaffected by tag failure. Based on tag failure tests (N = 66 tags), we estimated that only 55-78 percent of the tags used in this study were still functioning when the last fish was detected exiting the study area 15 days after release. Due to premature tag failure, our 'survival' estimates represent the joint probability that both the tag and fish survived, not just survival of fish. Low estimates of fish-tag survival could have been caused by fish mortality or fish travel times that exceeded the life of the tag, but we were unable to differentiate between the two. Fish-tag survival through the Delta (from Durham Ferry to Chipps Island by all routes) ranged from 0.05 +or- 0.01 (SE) to 0.06 +or- 0.01 between the two weekly release groups. Among the three migration routes, fish that remained in the San Joaquin River exhibited the highest joint fish-tag survival (0.09 +or- 0.02) in both weeks, but only 22-33 percent of tagged fish used this route, depending on the week of release. Only 4-10 percent (depending on week) of tagged fish traveled through Turner Cut, but no tagged fish that used this route were detected exiting the Delta. Most fish (63-68 percent, depending on week of release) migrated through Old River, but fish-tag survival through this route (0.05 +or- 0.01) was only about one-half that of fish that remained in the San Joaquin River. Once tagged fish entered Old River, only fish collected at two large water conveyance projects and transported through the Delta by truck were detected exiting the Delta, suggesting that this route was the only successful migration pathway for fish that entered Old River. The rate of entrainment of tagged juvenile salmon into Old River was similar to the fraction of San Joaquin River discharge flowing into Old River, which averaged 63 percent but varied tidally and ranged from 33 to 100 percent daily. Although improvements in transmitter battery life are clearly needed, this information will help guide the development of future research and monitoring efforts in this system.