Many types of optical strain measurement systems have been used for the determination of deformations and strains in soft biological tissues. The purpose of this investigation is to report a new optical strain measurement system developed in our laboratory which offers distinct advantages over systems developed in the past. Our optical strain system has demonstrated excellent performance in calibration and experimental tests. Calibration tests illustrate the system's accuracy to 0.05% strain at 3.52% strain and 0.18% strain at 11.74% strain. Further, this system can measure strains to within 2% measurement error for strains in a 0-11.74% range when 100 microns increments of motion are used for calibration. The resolution of our system appears to be at least as good as the linear micrometer (2 microns) used as a calibrating standard. Errors in strain measurement due to whole specimen rotation or translation are quantified. Rotations about an in-plane axis perpendicular to the direction of strain and translations in/out of the plane of focus result in the largest sources of error. Finally, in an in vitro biomechanical study of the rabbit Achilles tendon, experimental failure strains are 4.3 +/- 0.9% using this system.