Objective  To assess the accuracy， precision， repeatability， and reproducibility of biometric measurement of the mouse eye made by extended scan depth spectral domain optical coherence tomography （SD-OCT）. Methods  It was an experimental study. An SD-OCT with an extended depth range of 7.3 mm， A-scan speed of 28 kHz， axial resolution of 9.0 μm and transversally 19 μm was developed. Five 40-day-old C57BL/6 mice （10 eyes） were imaged with a custom built， extended scan depth SD-OCT in 2 sessions. Two scans were performed by operator 1 and a following scan by operator 2 at the first session， to determine repeatability and interoperator reproducibility. Intersession reproducibility was assessed by an additional scan performed by operator 1 2 days later. The data were compared to those from time-domain OCT （TD-OCT） by independent t test. All experiments were performed between 2：00 PM and 4：00 PM. The repeatability and reproducibility were evaluated by coefficient of repeatability/reproducibility， intraclass correlation coefficients （ICCs）， Wilcoxon matched-pairs test and Bland-Altman graph. Results  High-quality OCT images of the entire mouse eyes with high resolution were acquired. The ocular dimensions from TD-OCT measurements were close to these obtained by SD-OCT. The overall coefficients of repeatability/reproducibility for anterior chamber depth， lens thickness， vitreous chamber and axial length were all less than 3% and relevant intraclass correlation coefficients （ICCs） were nearly greater than 0.75， whereas these were less than 5% and ICCs were smaller for corneal thickness and retinal thickness. Wilcoxon analysis showed there was no statistically significant difference between any 2 groups for all parametres except corneal thickness of repeatability （Z=-2.310， P<0.05）. From the graph of differences against mean for the axial lengths， it can be seen that 100% of the values fall within 2 SDs of the mean. The overall coefficients of repeatability/reproducibility for curvatures were less than 5%. The ICCs for anterior curvature of the cornea and anterior/posterior curvatures of the lens were almost greater than 0.75， while it was greater than 0.45 for posterior curvature of the cornea. Graph of differences against means of average curvatures of reproducibility showed 5% data were without 2 SDs of the mean. Conclusion  SD-OCT is quite repeatable and reproducible for ocular dimension measurement of mouse eye. The novel method is helpful for monitoring the growth of the eye in normal mice and in strains that have developmental conditions such as myopia.