OBJECTIVE The aim of the study was to investigate differences in the shape of the palatal rugae in two populations of India and develop discriminant function to identify the populations based on rugae shape. METHODS Thirty plaster casts from each population group, equally distributed between the sexes and belonging to similar age-group, were examined for rugae shape (straight, wavy, curved, unification and circular) and their incidence recorded. Association between rugae shape and population as well as rugae shape and sex were tested using chi-square analysis and discriminant function developed using SPSS 10.0 statistical package. RESULTS Wavy and curved were the most prevalent rugae shape in both groups, followed by straight rugae. Unifications were few in number while circular rugae were not observed. Chi-square analysis for association between rugae shape and population groups showed significant differences in straight and curved forms. No significant sex differences were observed. Three rugae shapes - straight, wavy and curved - contributed to the discriminant function which enabled population identification with an accuracy of 70%. CONCLUSIONS Palatal rugae shape revealed significant differences between the two Indian populations and also confirmed previous reports of lack of sex dimorphism. Discriminant function analysis allowed moderate differentiation of the populations and it is inferred that discrete variables such as rugae shape are better suited for the purpose than continuous variables such as rugae length. However, these interpretations are precluded by the small sample size and further work on larger samples is required to validate the findings.