Fossils and DNA tell different stories about ant evolution. Or do they?

Did modern ants evolve up from subterranean ancestors? Or did they diversify from above-ground species?

This straightforward question about ant history does not have a straightforward answer. If we look at early ant fossils, most sport the long limbs and large eyes typical of surface-dwelling species. Consider the Cretaceous ant Haidomyrmex:

Haidomyrmex scimitarus (via

If the fossils tell a tale of large-eyed, surface-dwelling ancestors spawning the modern ant fauna, genetic data from modern species give an apparently conflicting story. A recent paper by Andrea Lucky and others in PLOS ONE took the known habits of modern ants and triangulated back over an evolutionary tree to infer the ancestral state. The odds of a subterranean ancestor were more than 90%:

Modified from Figure 1 of Lucky et al 2014.


Which story do we believe?

Possibly, both. The existence of a conflict depends on how the extinct, large-eyed species are related to modern ants. Many of them are in the subfamily Sphecomyrminae, and their phylogenetic position is not known with certainty. If sphecomyrmines are merely another ant lineage contained within the other known ants, as beautifully illustrated in the below diagram that took  at least 2 minutes to sketch, then we indeed have a conflict.


If, instead, sphecomyrmines are a separate lineage that diverged earlier in ant evolution, then there isn’t  a conflict at all. tree

In this latter scenario, one particular lineage went underground (where they were less likely to be preserved as amber fossils) and from there radiated into the ants we know and love today. Meanwhile, their sphecomyrmine sisters persisted, large-eyed and above ground, until extinction. It is entirely possible that the diversity and success of modern ants traces to modifications forced by this subterranean existence.

Finally, the underground result of Lucky et al depends on particular assumptions of how traits evolve. Specifically, that above- and below-ground species go extinct at similar rates. If subterranean ants tend, on average, to go extinct less often than their above-ground relatives, then many of the surviving members of older lineages will be subterranean and we might infer a subterranean ancestor as an artifact.

Leptanilla is an ancient lineage of blind subterranean ants.

source: Lucky A, Trautwein MD, Guénard BS, Weiser MD, Dunn RR (2013) Tracing the Rise of Ants – Out of the Ground. PLoS ONE 8(12): e84012. doi:10.1371/journal.pone.0084012

10 thoughts on “Fossils and DNA tell different stories about ant evolution. Or do they?”

  1. Innocent question: how much variation of life style is found inside ant families or genera? In plants, at least, it sometimes happens that closely related species have very different habits. There are herbs that are closely related to trees. Are there living subterranean ants that are closely related to surface-dwelling ants? How hard is it for an ant species to evolve into a species with longer legs and bigger eyes?

      1. Perhaps Lasius could also be considered in this way. The parasitic species like Lasius claviger are completely subterannian while their hosts Lasius neoniger and Lasius alienus forage above ground.

        1. Jim Harrison and others: When one considers that Myrmecocystus is the sister group of, and very closely related genetically to, Lasius, habitat diversification in that lineage becomes even more remarkable, to include long-legged, large-eyed, as well as distinct diurnal and nocturnal, arid land inhabitants.

  2. Pardon any ignorance about the state of knowledge concerning fossil ants below. But is it at all possible that conclusions drawn are the simply the result of the relatively small numbers of known fossilized ants and the sampling skew resulting from that essentially anecdotal picture ?

    I recognize that we work with what we have, but sometimes the tendency to draw conclusions beyond the data is just too tempting. 91.45 ????? Seems like this ant kung fu is weaker than hundreths percentile. Of course we desperately need the standard deviation of that number, or is 91.54167 % simply out of the question ?

  3. Is it possible that some change in the habitability of the dominant plants of the era is a factor here as well?

    That might open up a few other possibilities, something along the lines of. . .
    1) Ants are a semi-arboreal species nesting in ancient plants, needing large eyes to see and making galleries in the plant.
    2) Said plant habitat dies out/is supplanted by other plants which are ant-unfriendly or have anti-ant defenses
    3) Some ant species are able to adapt their nesting strategy to the leaf litter and eventually begin migrating downward.
    4) Eventually some ants start to figure out these new, weird plants and explore arboreal ecosystems again.

    I mean, I just totally made that up, and it could be completely wrong. (likely is) but I didn’t find a good ant-evolution timeline to put up next to a plant one (I could just be having some sort of keyword brain fart and it might be a really easy search, admittedly)

    1. In fact, I’m going to go ahead and make up a crazy theory having absolutely no idea what I’m talking about but the power of the internet and one of the recent benefits of living in Washington state 😉

      Perhaps ants were a branch off of a proto((proto dauber)/(proto-bee)) that lived in plants like cycads and plants like Williamsonia (no relation), many species nested in the bases of those trees and some species tunneled into the root system to protect their young but otherwise were power-players in the zones where it’s painful to put your hands.

      Then the K-T event hit and their primary natural environment went with it, so those that could thrive in the leaf litter and had more sophisticated methods of tunneling underground began to dominate, with specialists like queens (or sometimes entire families!) losing the once lovely eyes because the view was just that boring (evolution may have been a factor).

      The trees come back, but they no longer have nearly as many hand-hurty zones, but that’s okay for the ants, because they’ve learned a whole bunch of new tricks and found a new niche . .. and many of them do manage to find new zones to thrive in with clever adaptations.

      Would that explain the sequence of events better?

      1. Possible, except for the queens with small eyes part. Most species with small-eyed workers still have “normal”-eyed queens, except in the army ant (dorylomorph) branch of Formicidae.

        1. Ooh, good call! I hadn’t thought that bit through properly.

          I was just trying to incorporate the change in environment and see if there might be a scenario that fit, which actually ended up a little easier than I expected with the K-T boundary right there.

  4. Pingback: Morsels For The Mind – 31/01/2014 › Six Incredible Things Before Breakfast

Leave a Reply to MrILoveTheAnts Cancel reply