The purpose of this study was to determine whether delayed gadolinium-enhanced magnetic resonance imaging of cartilage (dGEMRIC) and T2 mapping are accurate techniques for measuring cartilage thickness in the metacarpus3/metatarsus3 (Mc3/Mt3) of Thoroughbred racehorses. Twenty-four Mc3/Mt3 cadaver specimens were acquired from six healthy racehorses. Cartilage thickness was measured from postintra-articular Gd-DTPA2- images acquired using short tau inversion recovery (STIR), and proton density weighted (PDw) sequences, and compared with cartilage thickness measured from corresponding histologic images. Two observers performed each histologic measurement twice at three different sites, with measurement times spaced at least 5 days apart. Histologic cartilage thickness was measured at each of the three sites from the articular surface to the bone-cartilage interface, and from the articular surface to the mineralized cartilage interface (tidemark). Intra-observer repeatability was good to moderate for dGEMRIC where Mc3/Mt3 cartilage was not in contact with the proximal phalanx. Where the Mc3/Mt3 cartilage was in contact with the proximal phalanx cartilage, dGEMRIC STIR and T2 mapping PDw cartilage thicknesses of Mc3/Mt3 could not be measured reliably. When measured from the articular surface to the bone-cartilage interface, histologic cartilage thickness did not differ from STIR or PDw cartilage thickness at the site where the Mc3/Mt3 cartilage surface was separated from the proximal phalanx cartilage (P > 0.05). Findings indicated that dGEMRIC STIR and T2 mapping PDw are accurate techniques for measuring Mc3/Mt3 cartilage thickness at locations where the cartilage is not in direct contact with the proximal phalanx cartilage.