In order to study the mechanism of lameness transfer from fore- and hindlimb lamenesses 2 hypotheses were investigated. Hypothesis 1: Horses with a true supporting limb lameness in one hindlimb show a false supporting limb lameness in the ipsilateral forelimb. Hypothesis 2: Horses with a true supporting limb lameness in one forelimb show a false supporting limb lameness in the contralateral hindlimb. Fourteen horses with fore- or hindlimb lameness were used for this study. Each horse was measured at the trot on a treadmill with standardised speed, before and after diagnostic blocks (9 horses), or with and without induced lameness (5 horses). The head acceleration asymmetry (HAAS) and the sacrum acceleration asymmetry (SAAS) were used for quantification of fore- and hindlimb lameness respectively. Changes were documented by changes of the HAAS or the SAAS. In all 4 horses with a true hindlimb lameness a synchronous false lameness of the ipsilateral forelimb was documented. In 6 of 10 horses with a forelimb lameness a lameness transfer could be assessed according to hypothesis 2. The results of this study show, that horses with a true severe lameness in the forelimb show a false lameness in the contralateral hindlimb, and horses with a true hindlimb lameness show a false lameness in the ipsilateral forelimb. This indicates that the location of the truly lame limb can be deduced from the distribution of 2 lamenesses on a sagittal or diagonal axis.