A new technique for characterization of the three‐dimensional morphology of murine articular cartilage is proposed. The technique consists of a novel application of confocal laser scanning microscopy (CLSM), where the objective was to develop and validate it for cartilage measurements in murine joints. Murine models are used in arthritis research, because they are well‐described for manipulating the disease pathophysiology, facilitating our understanding of the disease, and identifying new targets for therapy. A calibration and reproducibility study was carried out to provide a consistent testing methodology for quantification of murine joints. The proximal tibial condyles from male C57BL/6 mice were scanned using a CLS microscope with an isotropic voxel size of 5.8 μm. Measurements and analyses were repeated three times on different days, and in a second step the analysis was repeated three times for a single measurement. Calculation of precision errors (coefficient of variation) for cartilage thickness and volume was made. The bias of the system was estimated through comparison with histology. This technique showed good precision, with errors in the repeated analysis ranging from 0.63% (lateral thickness) to 3.48% (medial volume). The repeated analysis alone was robust, with intraclass correlations for the different compartments between 0.918 and 0.991. Measurement bias was corrected by scaling the confocal images to 32% of their width to match histology. CLSM provided a fast and reproducible technique for gathering 3D image data of murine cartilage and will be a valuable tool in understanding the efficacy of arthritis treatments in murine models. Microsc. Res. Tech. 2009. © 2009 Wiley‐Liss, Inc.