Have you thought about geomorphmetrics, such as those carried out by Thomas Cucchi and co.

Thanks Roz, that's exactly what we're doing. We're using semi-landmarks to define and digitize the outline of upper M1 because it a complex shape and preserves very well. After photographing 60 or so individuals of each species some morphological differences are becoming more obvious. The cusps of R. rattus M1's are more inflated or 'bunchy' for instance. I'm hoping that the analysis will reveal more/better morphological differences so we don't need to go through this process to identify every single specimen!