In dairy cattle locomotion score (LS), based on gait and posture, is related to allodynia, as measured using nociceptive threshold (NT). The effect of lameness treatment on this relationship has not been properly evaluated. As part of a study evaluating lameness treatment protocols, longitudinal measurements of NT and LS were made to allow such an evaluation. Data were collected from 149 dairy cattle with claw horn lesions from nine farms. On presentation, LS was recorded (1 to 5 scale) and NT measured with a pneumatically-activated pin. After corrective paring, cows were randomly allocated to one of six treatment groups. NT and LS were recorded 3, 8, 28 and 100 days after treatment. A PROC MIXED model was developed with square root transformed NT as the outcome variable. Without time in the model, NT and LS had a significant inverse relationship (r2=0.26). Treatment had no significant effect on this relationship. For example mean NT for cattle with a LS of 1 was 39.8 (0.54) KPa, 28.0 (0.86) for a LS of 3, and 21.5 (1.99) for one of 5. Inclusion of time altered the relationship. For cows with LS of £3, earlier visits were associated with a lower NT than later ones. Example on Day 0, the mean NT of cattle with a LS of 2 was 28.5, whereas on Day 100 it was 41.3 KPa. For cows where LS remained ³4 this was not so; mean NT was 23.5 on day 0 and 28.7 on Day 100. Progress in LS was not necessarily related to improvement in NT and vice-versa. As NT is a more objective measure of lameness-associated pain, with less inter and intra-observer variation than LS, it is probably a better measure of treatment response.