The objectives of the present study were to develop methods for conducting chronic toxicity tests with juvenile mussels under flow-through conditions and to determine the chronic toxicity of copper and ammonia to juvenile mussels using these methods. In two feeding tests, two-month-old fatmucket (Lampsilis siliquoidea) and rainbow mussel (Villosa iris) were fed various live algae or nonviable algal mixture for 28 d. The algal mixture was the best food resulting in high survival (???90%) and growth. Multiple copper and ammonia toxicity tests were conducted for 28 d starting with two-month-old mussels. Six toxicity tests using the algal mixture were successfully completed with a control survival of 88 to 100%. Among copper tests with rainbow mussel, fatmucket, and oyster mussel (Epioblasma capsaeformis), chronic value ([ChV], geometric mean of the no-observed-effect concentration and the lowest-observed-effect concentration) ranged from 8.5 to 9.8 ??g Cu/L for survival and from 4.6 to 8.5 ??g Cu/L for growth. Among ammonia tests with rainbow mussel, fatmucket, and wavy-rayed lampmussel (L. fasciola), the ChV ranged from 0.37 to 1.2 mg total ammonia N/L for survival and from 0.37 to 0.67 mg N/L for growth. These ChVs were below the U.S. Environmental Protection Agency 1996 chronic water quality criterion (WQC) for copper (15 ??g/L; hardness 170 mg/L) and 1999 WQC for total ammonia (1.26 mg N/L; pH 8.2 and 20??C). Results indicate that toxicity tests with two-month-old mussels can be conducted for 28 d with >80% control survival; growth was frequently a more sensitive endpoint compared to survival; and the 1996 chronic WQC for copper and the 1999 chronic WQC for total ammonia might not be adequately protective of the mussel species tested. However, a recently revised 2007 chronic WQC for copper based on the biotic ligand model may be more protective in the water tested. ?? 2007 SETAC.