Cyatta abscondita, a new fungus-growing ant

Cyatta abscondita Sosa-Calvo, Schultz, Brandão, Klingenberg, Feitosa, Rabeling, Bacci, Lopes & Vasconcelos 2013. Yes, that’s the full name. Modified from Sosa Calvo et al 2013 Figure 1.

Myrmecology continues apace! This week saw the publication of a particularly interesting new ant genusCyatta, from Brazil.

Why the excitement over this discovery? First, Cyatta is an attine fungus-growing ant, and attines are a multi-species system and a rich model for studies of co-evolution across microbes, fungi, animals, and plants. Attines are farmers, cultivating a specialized underground fungus from bits of detritus or, in the case of the spectacular leafcutters, from live vegetation. The ants also foster an array of organisms that live on their bodies, some of which produce agrochemicals that protect their gardens from weeds. Any new species of attine enriches our ability to study this system.

An excavated Cyatta fungus garden, showing the cultures suspended from a chamber ceiling. (modified from Sosa-Calvo et al 2013 Fig. 6).

But Cyatta is not just another Trachymyrmex. This new ant occupies an unusual space in the attine tree. Cyatta, along with its sister Kalathomyrmex, doesn’t share recent ancestry with other attines, instead tracing its origin to near the origin of the whole tribe. As such, it will provide another perspective from which to triangulate our inferences of how ant agriculture developed.

Here is the molecular tree:

Figure 7 from Sosa-Calvo et al 2013. This phylogeny is based on 4 nuclear protein-coding genes.

For example, Cyatta gardens resemble those of Kalathomyrmex and Mycocepurus, strengthening our inference that simple suspended gardens were the form used by the ancestor of all neoattines. And the presence of larval anchor hairs employed in other genera to hang larvae along the sides of the nest chamber (see Clint Penick’s research), suggests that brood-hanging may have been present in the early attines but was subsequently lost.

One final gripe- because I always have a gripe- is that all authors of the paper are also listed as authors of the genus and species. This makes the formal name for the new ant an impressive:

Cyatta abscondita Sosa-Calvo, Schultz,
Brandão, Klingenberg, Feitosa, Rabeling,
Bacci, Lopes & Vasconcelos 2013


This moniker will be a handful for people who handle taxonomic databases, or for taxonomists who will need to write about this ant. I doubt all authors contributed equally to the written description embedded in the paper; surely a separate, smaller authorship for the description would have made for a less cumbersome name.

In that vein, does anyone know if there is a longer authorship for any animal species? This is the largest I’ve seen.

Sosa-Calvo J, Schultz TR, Brandão CRF, Klingenberg C, Feitosa RM, et al. (2013) Cyatta abscondita: Taxonomy, Evolution, and Natural History of a New Fungus-Farming Ant Genus from Brazil. PLoS ONE 8(11): e80498. doi:10.1371/journal.pone.0080498

How to tell the difference between Atta and Acromyrmex leafcutter ants

Three pairs of promesonotal spines indicates Acromyrmex, while two pairs indicates Atta (photos from

The iconic leafcutter ants of the New World tropics and subtropics are currently split into two similar genera: Acromyrmex and Atta. What’s the difference?

In an evolutionary sense, the answer isn’t clear. A recent molecular study suggests Atta may be no more than a derived lineage within a larger Acromyrmex, and that our distinction is artificial.

But what if you just want to key a specimen to one or the other? That’s easier. Count the spines on the front of the thorax- the promesonotum- you’ll find that Acromyrmex sports three pairs, while Atta has just two.

As an exercise, see if you can identify the ants in the following images:




Most “Carpenter Ants” aren’t carpenter ants

A major worker Camponotus ephippium guards the entrance to her underground nest. (Victoria, Australia).

Since we’re on the topic of common names, a peeve of mine is the use of “Carpenter Ant” to refer to the genus CamponotusWhile it’s true some Camponotus– including the common North American wood-destroying C. pennsylvanicus– do carve chambers in wood, a great many others nest in the ground where woodworking is not much of a possibility.

An accurate common name for Camponotus is not an arcane problem. This is the single most widespread and abundant ant genus worldwide. If I were to recommend just one genus that non-specialists should recognize, it is this one. Camponotus is found nearly everywhere, often in abundance, and the genus is arguably the most species-rich in the world. With the possible exception of Pheidole big-headed ants, of course.

A thatch nest of Camponotus rufipes (Minas Gerais, Brazil).

Australians call theirs “Sugar Ants”. Camponotus novaehollandiae is the “Northern Common Sugar Ant“, for example. Insofar as all Camponotus like sugar that moniker is decent, but a sweet tooth is hardly a unique behavior among formicids. Still. It’s more accurate than “Carpenter Ants”.

Camponotus punctulatus happy digging away in the sand, no wood to be seen (Parque Nacional El Palmar, Argentina).

Unlike for Polyergus, where I have a decidedly unpopular suggestion, I don’t really know where to start for Camponotus.

Any thoughts?

Polyergus moves forward, and a modest proposal for kidnapper ants

Workers of Polyergus mexicanus return from a successful raid with kidnapped Formica subsericea brood. Urbana, Illinois.

This morning saw the publication of James Trager’s much-awaited taxonomic revision of the parasitic ant genus Polyergus. These insects are commonly known as “slave-raiding ants”, a controversial moniker I’ll discuss in a bit. But first, the new paper.

James, who often comments on this blog, is known for thorough, methodical taxonomic papers on particularly difficult genera of ants. In the 1980s, for example, he revised the North American Nylanderia and Paratrechina. As James is no longer employed as a researcher, his publications are infrequent. But his patience and attention to detail lend James’ works particular longevity. This new revision fits the mold, so I expect it will be the standard reference for identifying Polyergus to species for the next decade and beyond.

(An aside: A taxonomic revision is what it sounds like. The existing taxonomy is reevaluated based on new data, new species are described if needed, old mistakes are corrected, and inadequate taxa are sunk into more appropriate ones. The result is a more stable arrangement of species.)

The old scheme for Polyergus crammed an uncomfortable level of variation into just a handful of described species. Yet consistent morphological differences among Polyergus that attack different species of host Formica, particularly in the F. pallidefulva group, suggested a much finer division. The new scheme is more sensible in light of host/parasite biology, elevating a number of subspecific taxa to full species while describing five as entirely new. The paper also contains fascinating observations on Polyergus natural history.

And what a spectacular natural history these ants have!

Colonies of Polyergus cannot function without a large contingent of workers from another ant genus, Formica, that care for the brood, maintain the nest, and forage for food. In fact, Polyergus workers themselves do little other than kidnap immature brood from nearby Formica nests. The raids are usually in late afternoon, in the summer, and can be spectacular to watch.

At the entrance of a Polyergus mexicanus colony, a Formica subsericea worker carries excavated soil from the nest. The parasitic Polyergus workers do not perform those sorts of tasks. (Urbana, Illinois)

The dependence of Polyergus on stolen labor has lent a bondage metaphor to the common name: slave-raiding ants. To the limited extent that a human analogy can apply to an insect, the comparison is reasonably apt.

Recently the slave-raiding name has become controversial. For good reason. Myrmecologist Joan Herbers (2006, 2007) observes that references to slavery can make public communication about these ants unduly difficult:

In the United States, we are technologically dependent yet scientifically illiterate, and using jargon that discourages even one individual from learning more about science is simply irresponsible. I find it hard to imagine a young black student being attracted to a discipline that calls parasitized insects “slaves” and “negro ants.”

Herbers proposed substituting “pirate ant” for “slave-raiding ant”, yet I’ve never liked that solution. A couple years ago I explained:

I don’t think Herbers’s solution is workable, though. Piracy is a terrible parallel to what ants like Protomognathus and Polyergus do. Pirates take things. Slave-raiding ants aren’t primarily pillaging the food stores of other colonies. They don’t lay in wait along trails to steal forage. No. The brood parasites take actual, living ants whose labor they use for their own benefit. That is slavery. If I call these ants pirates, I am not communicating accurately about their biology. Piracy, for me, is out.

While I still don’t like the piracy metaphor, I’ve come around to Herber’s perspective that the slave-raiding comparison, while apt, is not ideal for those of us trying to introduce myrmecology to its broadest possible audience.

So. Down with ant slavery! Instead, I humbly propose that Polyergus and other cleptergic species be called kidnapper ants.

As a vivid vernacular, kidnapping is as accurate as slavery. Polyergus raids target immature forms of its hosts, after all. Yet kidnapping, though terrible as a human atrocity, is not so culturally encumbered as slavery.

I tried out the phrase on a local church group recently. Given how quickly attendees sat forward in their seats on mention of the mysterious Kidnapper Ants, I’ll stick with the newer metaphor.

Anyway. Regardless of what you think of the common names, check out James’ revision. It’s good.

Herbers, JM (2006) The loaded language of science. Chronicle of Higher Education. 52: B5
Herbers, JM (2007) Watch Your Language! Racially Loaded Metaphors in Scientific Research. BioScience 57. doi 10.1641/B570203
Trager, JC (2007) Collected thoughts on “pirate” ants and “leistic” behavior. Notes from Underground, online.
Trager, JC (2013) Global revision of the dulotic ant genus Polyergus (Hymenoptera: Formicidae, Formicinae, Formicini). Zootaxa 3722 (4): 501–548.

Army Ants of Darkness: Eciton burchellii parvispinum

Eciton burchellii
An Eciton burchellii parvispinum worker carries an ant pupa back from the raid front. Belize.

When I first saw a milling swarm of these dark army ants pushing across the road last month at our lodge in Belize, I did a double-take. Their behavior and form were reminiscent of the familiar Eciton burchellii army ant I’ve seen in Panama and across South America, yet these ants were completely black. I’m used to swarm-raiding army ants having distinctly reddish backsides, like so:

Eciton burchellii, Ecuador.

I vaguely recalled Jack Longino having written about black Eciton burchellii in Costa Rica, so when I arrived home I looked it up:

There are numerous subspecies of E. burchellii. In Costa Rica there are two, E. b. foreli and E. b. parvispinum. The workers of foreli have a red brown metasoma, such that minor workers appear bicolored in the field, while the workers of parvispinum have a black metasoma, and the minor workers are entirely black. I can find no other morphological or behavioral feature that correlates with the color difference. The color forms could easily rest on a very small genetic difference, perhaps a single gene. However, the males also exhibit a difference. Males from the Atlantic slope, corresponding to the range of foreli, have a number of long flexuous setae on the scutellum; males from the Pacific slope and the range of parvispinum have the scutellum bare. In most cases this character difference is discrete, but I collected one male that was intermediate. Among a series of males from Monteverde, which is close to the zone of contact of the two forms, one had long setae on one half of the scutellum and the other half was bare (differing across a sagittal plane). All the rest of the males and all of the workers I have seen from Monteverde have the parvispinum phenotype.


The distributions of foreli and parvispinum are very sharply parapatric and do not seem to correlate with particular habitats or thermal environments. Eciton b. parvispinum has the broadest distribution, occurring across all of the Pacific slope, from the dry forests of Guanacaste to the rainforest of the Osa Peninsula, from sea level up into the mountains, and dropping a short distance down onto the Atlantic slope, where it meets foreli. For example, on the Barva Transect of Braulio Carrillo National Park, foreli is common from La Selva up to 500m elevation. Above that elevation Eciton burchellii has lower density, but I have three collections from 1400-1500m elevation, about 10km further up-slope, and they are parvispinum. In the Peñas Blancas Valley, east of Monteverde, only foreli occurs at Refugio Eladio, at 800m elevation. At Refugio El Aleman, 950m elevation and 5km further up the valley, I have only collected parvispinum. West of El Aleman, continuing up to Monteverde, only parvispinum is found. Like other examples of step-clines in morphological or genetic characters, it begs the question of what mechanism is maintaining the sharp boundary between these two forms, especially for such large, nomadic organisms.


Apparently I had seen my first Eciton burchellii parvispinum! The majors and submajors still had light-colored heads while retaining the dark metasoma:

Eciton burchellii
Ection submajor workers are a specialized prey-carrying caste.

If you’ve spent time trying to identify ants, you’ll be familiar with the caution applied to using color to diagnose species. The lack of confidence in such an obvious trait can be frustrating to ant newbies (“But this one is orange! That one is black! Why are they the same species?!”), and could be just another case where a single species encompasses substantial variation. Not unlike our own species, of course.

What does the color difference mean? Is the dark form, as Jack mentions, just a single allele that’s become fixed in some Central American populations? Or is it just the tip of a larger genetic iceberg, such that the dark and light forms are separate, non-interbreeding groups?

The two forms do not appear to coexist. Taxonomists typically interpret this lack of sympatry as indicative of a single species that varies across space, but in this case I’m not so sure that conclusion is appropriate. I am not an army ant taxonomist, so take this with a truckload of salt, but my opinion is that E. burchellii parvispinum is a good biological species and should be elevated to E. parvispinum.

My reasoning is as follows:

1. Genetic studies on E. burchellii in Mexico and Panama show that males of the species disperse well enough to maintain genetic diversity across reasonably large areas. Thus, even if the flightless queens are limited in their dispersal, their genes are broadcast widely on the wings of males. I doubt these forms are completely geographically isolated.

2. Queens are polyandrous- that is, they mate with many males- and if E. burchellii is just a single, interbreeding species we should find at least some colonies of mixed color variations in regions where dark and light forms occur. The two species don’t coexist locally, but they do broadly overlap across large swaths of Mesoamerica, enough so that males should jump populations at least occasionally.

3. Jack’s observation that males of E. b. parvispinum  and E. b. foreli are consistently different in other traits suggests a more complete genetic separate than just a single color difference in workers.

The lack of local co-existence among forms might have another explanation beyond mere genetic isolation by distance. The raiding biology and dietary habits of both forms are so similar that the two would certainly enter into intense competition, so one or the other ends up winning out in a particular patch of habitat. A forest might just not be big enough for the both of them.

In any case, I’m just speculating. A proper study would involve a great deal more genetics and the measuring of male bits.


Berghoff SM, Kronauer DJC, Edwards KJ, Franks NR (2008) Dispersal and population structure of a New World predator, the army ant Eciton burchellii. J Evol Biol 21:1125–1132. doi:10.1111/j.1420-9101.2008.01531.x

Jaffé R, Moritz RFA, Kraus FB (2009) Gene flow is maintained by polyandry and male dispersal in the army ant Eciton burchellii. Population Ecology 51: 227–236. doi:10.1007/s10144-008-0133-1

Longino, JT (2007) Eciton burchelli parvispinum Forel 1899 (website) – accessed online 10/11/2013.

How to recognize Apoidea

One last post on wasps, while I’m still on the topic.

Bees and spheciform wasps, forming the superfamily Apoidea, are hugely diverse in form, size, color, and habits. The rich variety within Apoidea can make the group difficult to recognize, but most share one particular morphological trait that, with practice, can reliably be used to diagnose these insects from other types of wasps.

(Adapted from Scott & Stojanovich “Stinging Hymenoptera, Pictoral Key To Some Common United States Families”, via the CDC)

The pronotum is the first big dorsal plate on an insect’s thorax, and in apoids this plate ends at the sides in a distinctly rounded lobe that does not touch the tegula, a small plate associated with the wing base. Bees, crabronids, sphecids, ampulicids, and other spheciforms have this distinctive pronotal lobe; similarly sized and colored wasps in the Vespoidea do not.

Hylaeus, a bee, has the rounded pronotal lobe.
Can you spot the lobe in this cicada killer wasp?

In contrast, here is Euodynerus, a vespoid wasp showing the non-apoid pronotum without the distinctive lobe:

In vespoid wasps, the pronotum typically touches the tegula in a sharp angle.


Be cautious when using for identifications

And now, a public service announcement.

While I am flattered that many of you use my photo galleries to identify mystery ants, please be aware my site has limitations as a diagnostic tool. In particular, I am missing a lot of species, even some common ones, and even in North America.

Of the 33 North American species of the common genus Lasius, I have only 11:


That’s typical. Here, as an additional example, is how I am faring with Pogonomyrmex harvester ants:


Thus, when browsing my galleries looking for a match, bear in mind there’s more diversity than what I’ve posted. Often, much more. I’m just a guy with a camera, and progress happens only as time and budget permit.

If you’ve got ants needing reliable ID, you’ll do better consulting a site designed for species diagnosis like Or better yet, use the primary literature.