The East Bay Plain, on the densely populated eastern shore of San Francisco Bay, contains an upper aquifer system to depths of 250 feet below land surface and an underlying lower aquifer system to depths of more than 650 feet. Injection and recovery of imported water has been proposed for deep aquifers at two sites within the lower aquifer system. Successful operation requires that the injected water be isolated from surface sources of poor-quality water during storage and recovery. Hydraulic, geochemical, and isotopic data were used to evaluate the isolation of deeper aquifers. Ground-water responses to tidal changes in the Bay suggest that thick clay layers present within these deposits effectively isolate the deeper aquifers in the northern part of the study area from overlying surficial deposits. These data also suggest that the areal extent of the shallow and deep aquifers beneath the Bay may be limited in the northern part of the study area. Despite its apparent hydraulic isolation, the lower aquifer system may be connected to the overlying upper aquifer system through the corroded and failed casings of abandoned wells. Water-level measurements in observation wells and downward flow measured in selected wells during nonpumped conditions suggest that water may flow through wells from the upper aquifer system into the lower aquifer system during nonpumped conditions. The chemistry of water from wells in the East Bay Plain ranges from fresh to saline; salinity is greater than seawater in shallow estuarine deposits near the Bay. Water from wells completed in the lower aquifer system has higher pH, higher sodium, chloride, and manganese concentrations, and lower calcium concentrations and alkalinity than does water from wells completed in the overlying upper aquifer system. Ground-water recharge temperatures derived from noble-gas data indicate that highly focused recharge processes from infiltration of winter streamflow and more diffuse recharge processes from infiltration of precipitation occur within the study area. However, recharge of imported water from leaking water-supply pipes, believed by previous investigators to be a large source of ground-water recharge, was not supported on the basis of oxygen-18 and deuterium data collected as part of this study. Based on tritium/helium-3 ages, most water in the upper aquifer system is relatively young and was recharged after 1952; however, water in the lower aquifer system is older and does not contain detectable tritium. Carbon-14 ages interpreted for water from wells in the lower aquifer system and underlying partly consolidated rocks range from 500 to more than 20,000 years before present. The greatest ages were in water from wells completed in the partly consolidated deposits that underlie the northern part of the study area. Ground water from wells in the lower aquifer system near the proposed Bayside injection/recovery site was recharged about 9,400 years before present and appears to be isolated from surface sources of recharge and ground-water contamination.