The aim of this study was to evaluate both the repeatability (intraobserver agreement) and reproducibility (interobserver agreement) of Obel grading, the most widely accepted method for describing the severity of equine laminitis. The study comprised two parts. In study A, to test intraobserver agreement, 25 ponies were graded twice by a single blinded experienced equine veterinary surgeon, on the first occasion via clinical examination and on the second occasion via a video recording of the initial clinical examination. In study B, to test interobserver agreement, video recordings of 13 ponies were graded independently by 58 blinded experienced equine veterinary surgeons. A weighted kappa (κw) statistic was used to determine the measure of agreement between gradings. Intraobserver agreement of Obel grading was shown to be moderate (κw=0.54), and interobserver agreement was found to be substantial (κw=0.65). In addition, intraobserver agreement improved from moderate to substantial (κw increased from 0.54 to 0.70) when the Obel grades were grouped and horses were categorised as either sound, or as having mild (grade 1 or 2) or severe (grade 3 or 4) lameness.