Teak(Tectona grandis L.f.) is a popular hardwood species native to South and South-East Asia. The possible association of amplified fragment length polymorphism(AFLP) marker with morphological variables of eleven important characters viz. girth at breast height(GBH), height of tree, bole height, branch knots, presence of fluting, spiral stem, leaf hair, leaf length/breadth ratio, branching pattern, bark colour and petiole shape for nine natural populations comprising 180 genotypes of teak(9 populations × 20 trees). The phenogram constructed using Euclidean distances for the eleven morphological characters showed that the populations were not grouped according to their geographical origin. The Mantel’s test for pairwise correlation between Euclidean distances of different morphological variables and genetic distances from AFLP data revealed that only petiole character(r =0.269; p =0.046) and height of tree(r =0.200; p =0.001) were significantly correlated with that of AFLP data matrix. The nine populations in this study covered a geographic area of about 1000 km stretch along the Western Ghat of South India. A test of correlation between genetic and geographic distance matrices revealed a significant positive correlation(r =0.475; p =0.009). The lack of perfect congruence between morphological and molecular data except for geographic distance, tree height and petiole character suggested that the morphological system might be useful for the morphotypes management but not appropriate to study the genetic structure of the teak populations. The possible association of amplified fragment length polymorphism (AFLP) marker with morphological variables of eleven important characters viz. Girth at breast height (GBH), height of tree, bole height, branch knots, presence of fluting, spiral stem, leaf hair, leaf length / breadth ratio, branching pattern, bark color and petiole shape for nine natural populations including 180 genotypes of teak (9 populations × 20 trees). The phenogram constructed using Euclidean distances for the eleven morphological characters showed that the populations were not grouped according to their geographical origin. The Mantel’s test for pairwise correlation between Euclidean distances of different morphological variables and genetic distances from AFLP data revealed that only petiole character (r = 0.269; p = 0.046) and height of tree (r = 0.200; p = 0.001) were significantly correlated with that of AFLP data matrix. The nine populations in this study covered a geographic area of ​​about 1000 km stretch along the Western Ghat of South India. A test of correlation between genetic and geographic distance matrices revealed a significant positive correlation (r = 0.475; p = 0.009). The lack of perfect congruence between morphological and molecular data except for geographic distance, tree height and petiole character suggested that the morphological system might be useful for the morphotypes management but not appropriate to study the genetic structure of the teak populations.