The United States Endangered Species Act (ESA) requires a biological assessment of any activity that is authorized, funded, or carried out by a federal agency and likely to affect a federally listed endangered species or its critical habitat. We developed a standardized survey protocol for biological assessments of the effects of bridge replacements on 2 federally listed endangered freshwater mussels, Epioblasma torulosa rangiana and Pleurobema clava, found in the Allegheny River, Pennsylvania. The protocol combines qualitative sampling to determine species present with quantitative sampling to estimate density. Data on species present satisfy the minimum requirement of a biological assessment, whereas estimates of density are needed to assess the number of individuals that would die as a result of bridge replacement. Some excavation of substrate is necessary for unbiased population estimates because of species and sex-specific differences in detection at the substrate surface. We reduced the amount of excavation and cost of the survey by using a statistical sampling technique called double sampling, which uses counts from excavating a subset of quadrats to calibrate counts from searching the substrate surface of all quadrats. We applied the survey protocol to the Allegheny River at West Hickory where E. t. rangiana was the 3^rd and P. clava was the 4^th most abundant mussel at the site. Only 31% of P. clava and 52% of E. t. rangiana (80% of females, 45% of males) were detected at the substrate surface. We estimated that 9173 (95% CI: 6309–13,336) E. t. rangiana and 7010 (95% CI: 4462–11,013) P. clava lived within 50 m of the existing bridge and would be affected immediately by bridge construction. (Population estimates did not include mussels too small to be retained on a 6.35-mm-mesh sieve.) Application of the protocol is not limited to biological assessment under the ESA, but is appropriate where site-specific status of freshwater mussel populations is required.