A blog on “insects, science, and photography,” Myrmecos, tells the story of severe phylogenetic conflicts in a genus of tropical and subtropical ants called Pachycondyla. As Myrmecos reports:

Pachycondyla, among the most common ants in tropical regions worldwide, turns out to be a motley assortment of unrelated species. While the taxonomy of the world’s 12,000 or so ant species is obviously still a work in progress, I don’t think any of us had seen a case where ant names showed such a non-relationship to their genealogy. We knew before that Pachycondyla wasn’t really a natural group. But this? This was bad.

What was so “bad” was a diagram that myrmecologist Chris Schmidt showed at a recent ant conference. This diagram shows a DNA-based phylogenetic tree of ant species, with members of Pachycondyla highlighted in red:

According to Myrmecos, when this diagram was shown at the ant conference, “the whole room broke into laughter.” Why is that? As can be seen in the diagram, the ant species that were once thought to belong to Pachycondyla group instead into different, distantly related groups, when one considers their DNA sequence data. Evolutionary biologists would not have expected these ants, once gathered together in a single genus, to be what Myrmecos called “a motley assortment of unrelated species.”

This is yet another example of how morphology, DNA, and even biogeography can lead to wildly different ideas about phylogenetic trees, showing that the methods of phylogenetic reconstruction — used to bolster common descent — often yield results at variance with key pieces of evidence.