The Alberta Infant Motor Scale (AIMS) is a norm-referenced measure of infant gross motor development. The objectives of this study were: (1) to establish the best cut-off scores on the AIMS for predictive purposes, and (2) to compare the predictive abilities of the AIMS with those of the Movement Assessment of Infants (MAI) and the Peabody Developmental'Gross Motor Scale (PDGMS). One hundred and sixty-four infants were assessed at 4 and 8 months adjusted ages on the three measures. A pediatrician assessed each infant's gross motor development at 18 months as normal, suspicious, or abnormal. For the AIMS, two different cut-off points were identified: the 10th centile at 4 months and the 5th centile at 8 months. The MAI provided the best specificity rates at 4 months while the AIMS was superior in specificity at 8 months. Sensitivity rates were comparable between the two tests. The PDGMS in general demonstrated poor predictive abilities.