Objective - To ascertain the effectiveness of evaluating ground reaction forces (GRFs) at velocities during walking and trotting in dogs with naturally occurring lameness and determine whether walking would provide sufficient motion to adequately characterize GRFs with respect to trotting. Animals - 29 dogs with a naturally occurring tear of the cranial cruciate ligament. Procedure - Dogs were walked and trotted over a force platform, and GRFs were recorded during the stance phase. Correlation was used to assess the agreement between walking and trotting for GRF. The coefficient of variation was calculated to assess the relative variation of outcome variables among the gaits. Group means for walking GRF were compared between dogs that trotted and that failed to trot. Results - GRFs during walking and trotting were highly correlated. The coefficient of variation was smaller for GRFs during walking than during trotting. Dogs that failed to trot had significantly smaller mean values of peak vertical force and vertical impulse during walking, compared with values for dogs that were able to trot. Conclusions and Clinical Relevance - Either velocity is acceptable for GRF evaluation in dogs. Mean GRF during walking was significantly different between dogs that could and could not trot, principally because dogs with the most severe lameness failed to trot. These dogs would be eliminated from a clinical study, and thus, that study would become biased toward dogs that were less lame. In that situation, differences between interventions may be less pronounced, because they would be evaluated on dogs with less lameness.