A high performance concrete (HPC) bridge was instrumented for short-term and long-term behaviors of prestressed beams. The bridge is a twin-bridge over Pistol Creek in Blount County, Tennessee. The beams in one bridge lane were of high strength concrete (HSC) with a specified compressive strength of 10,000 psi and the ones in the other lane were of normal strength concrete (NSC) with a specified compressive strength of 6,000 psi. The primary objectives of the research were to experimentally study the time-dependent behaviors of HSC and NSC bridge beams and cast-in-place concrete deck and diaphragm during various stages of construction and service and to investigate the optimal erection schedule for HSC bridge beams. The selected bridge was instrumented for measurements of concrete temperature, strains of prestressed beams, strains of cast-in-place decks and diaphragms, and cambers of prestressed beams during curing and other stages. Material properties of both concretes were obtained through laboratory tests. A live load test was conducted for the evaluation of distribution factors of live load moment. Based on the test results and analytical studies several conclusions and recommendations were drawn for the design and construction of HSC bridges. The deformation of HSC beams developed rapidly during the early age and stabilized after 30-45 days in the study. No significant changes were observed in strain and camber after 45 days. Therefore, the HSC beams could be erected after having stayed in a producer's yards for 45 days. Use of lower curing temperature for HSC beams was a successful practice in this study. All of HSC beams used in the project achieved the required strength of 10,000 psi at 28 days. The current method of deck construction may need some improvements for HSC bridge because a large differential shrinkage between deck concrete and HSC beam concrete may cause tensile strains in HSC bridge deck. A fogging system could be used to prevent rapid evaporation of moisture from fresh deck concrete and reduce the early shrinkage of concrete. The distribution factor of live load moment calculated from Henry's method was the closest one to finite element analysis and live-load test results.