Background: To determine the relationship between clinically significant tooth size discrepancies (TSD) and archform classification in orthodontic patients. Material and Methods: Two hundred and forty consecutive sets of pre-treatment orthodontic study models were scanned and landmarked. All models had permanent teeth erupted from first molar to first molar in both arches. Sixty sets of images were classified into two groups of 30 according to the presence (group 1) or absence (group 2) of a clinically significant overall or anterior TSD (>2 SD from Bolton’s original means). Mean upper and lower archforms were created for each group using a fourth degree polynomial curve. Upper and lower archforms in each group were classified as square, tapering or ovoid; their distribution was analysed using the Fisher test with a 5% level of significance. To evaluate the intra-operator error when determining archform type, the 60 archforms were re-classified by the same operator two weeks later. The unweighted Kappa statistic at 95% confidence intervals was used to determine the similarity of the classification on the two occasions. Results: Reproducibility of the classification of archform was very good (unweighted Kappa statistic of 0.83 with a 95% confidence interval of 0.73, 0.93). There was no statistically significant difference in the distribution of archform type between group 1 and group 2 for the upper ( p =0.3305) or lower ( p =0.6310) arches. Conclusions: The presence of a clinically significant TSD and archform classification do not appear to be related.