Faculty of Orthodontics School of Health Sciences 81/0315/8 Miltstein@global.co.za / The incidence of relapse following orthodontic treatment has always been of paramount interest to clinicians and research workers. Many investigations have been carried out at various Universities to assess the long-term stability of orthodontic treatment. This study assessed the stability of 31 cases at least ten years out of retention treated by post-graduate students in the Department of Orthodontics at the University of the Witwatersrand, Johannesburg. Associations between certain clinical parameters and postretention relapse were also evaluated. Occlusal changes which had taken place were assessed by evaluating data from models taken prior to treatment and comparing these with similar parameters at the end of treatment and, on average, just less than 13 years following the end of retention. Eleven parameters were recorded and measured according to a scoring system proposed by Sadowsky and Sakols. A total malocclusion score was obtained for each study model by calculating the algebraic sum of all the scores. The total malocclusion scores thus indicated the severity of malocclusion presented by each study model at each period. A further examination of each of the study models was carried out in order to establish the Irregularity Index. The anatomic points at which contact would be made were defined on each surface of the six mandibular anterior teeth. The distance between the contact points of the adjacent teeth were measured using a digital vernier caliper calibrated to the nearest 0,05mm. The sum of these 5 measurements provided the Irregularity Index. Values for mandibular irregularity were then grouped into minimal, moderate and severe categories. v Changes in the intercanine and intermolar widths were correlated with overbite, overjet and anterior mandibular crowding parameters. The study also determined whether the presence or absence of third molar teeth and/or premolar extraction therapy were associated with relapse. At postretention 54.8 percent of the sample displayed some degree of relapse in at least one of the variables studied. However, all of the variables studied showed a greater degree of stability than relapse. Canine relationship showed the most postretention stability, whilst deep bite, overjet and molar relationships showed the least. With a few exceptions, occlusal abnormalities which were considered to be severe at the commencement of treatment demonstrated a high degree of stability following corrective therapy, whereas the relapse of mild to moderate features prior to treatment was not as predictable. The small amount of change noted in intercanine and intermolar widths at all three stages of treatment were not significant predictors of relapse at the 5 percent significance level of the parameters assessed. Furthermore the presence or absence of third molar teeth was not a significant predictor of poor postretention malocclusion score, overjet or overbite. Similarly extraction therapy was not a significant predictor of lower incisor alignment relapse (p=1.0). The mean irregularity index of the whole sample following retention indicated that the lower incisors displayed minimal crowding (2.7 mm). Following at least ten years out of retention, treated cases in the Department of Orthodontics, University of the Witwatersrand, Johannesburg showed a high degree of stability for the parameters assessed.