Consequences of the choice of landmarks for visualization and biological interpretation of shape changes in 2-D geometric morphometrics: a study case

We compare shape changes in mandibles of marsupials, represented by 2-dimensional coordinates, using different subsets of landmarks. We show that use of Type III landmarks, often not recommended in geometric morphometric techniques, might permit a better interpretation of shape changes than use of Type I landmarks alone, or a combination of Types I and II. We suggest that the use of as many landmarks as necessary to cover the whole configuration of the studied structure, regardless of the type, will allow a better visualization of the biological forms without leading to a less clear discrimination among them.