The geometric morphometric (GM) analysis of complex anatomical structures is an ever more powerful tool to study biological variability, adaptation and evolution. Here, we propose a new method (combinland), developed in R, meant to combine the morphological information contained in different landmark coordinate sets into a single dataset, under a GM context. combinland builds a common ordination space taking into account the entire shape information encoded in the starting configurations. We applied combinland to a Primate case study including 133 skulls belonging to 14 species. On each specimen, we simulated photo acquisitions converting the 3D landmark sets into six 2D configurations along standard anatomical views. The application of combinland shows statistically negligible differences in the ordination space compared to that of the original 3D objects, in contrast to a previous method meant to address the same issue. Hence, we argue combinland allows to correctly retrieve 3D-quality statistical information from 2D landmark configurations. This makes combinland a viable alternative when the extraction of 3D models is not possible, recommended, or too expensive, and to make full use of disparate sources (and views) of morphological information regarding the same specimens. The code and examples for the application of combinland are available in the Arothron R package.