Background: Two-point discrimination (TPD) is an assessment of tactile acuity. People with multiple sclerosis (MS) can have reduced foot sole tactile acuity, which has been linked to impaired balance. Objective: To quantify the test–retest reliability of TPD on the sole of the foot in people with MS. Approach: 41 participants (32 females), with mean (SD) age of 60 (9) years, and Expanded Disability Status Scale of <7.5, had their TPD measured at the head of the first metatarsal and the heel on two occasions, 2–14 d apart. Mean systematic change, within-subjects SD, limits of agreement (LOA), coefficient of variation and the intraclass correlation coefficient (ICC) were quantified as point estimates (95% CI). Main results: Systematic learning effects were evident. The within-subjects SD at the metatarsal and the heel was 6.7 mm (5.5–8.6) and 8.3 mm (6.7–10.8), and the LOAs were 18.6 mm (15.2–24.) and 23.7 mm (18.7–30.1), respectively. ICCs for metatarsal and heel was 0.87 (0.76–0.93) and 0.90 (0.80–0.95), respectively, but these were likely inflated by sample heterogeneity. Significance: In people with MS, TPD on the sole of the foot has an adequate test–retest reliability for research purposes, but there is substantial measurement variability for individual patients.