Dolichoderus attelaboides: the long-necked ant of the Amazon

Dolichoderus attelaboides

I give numerous reasons for why I like ants, but chief among them is my fascination with just how many ways there are to be an ant. Taxonomists have formally named some 13,000 ant species and expect to discover at least another 5,000 or so. There are ants smaller than the head of a pin and ants larger than my thumb. Their societies may number several million, or just half a dozen. Some ants glisten like polished metal, others are furrier than a woolly bear. Some farm, some hunt, some scavenge. Some glide from tropical treetops, some swim in pitcher plants.

Distribution of D. attelaboides (modified from MacKay 1993).

One unusual species I was happy to encounter in Ecuador this January was Dolichoderus attelaboides. This is an ant that never outgrew its awkward adolescent physique. Dolichoderus attelaboides is more gangly than an ant should be, its limbs slightly too long. Its eyes boggle from a head supported by a false neck too thin for its size.

Dolichoderus attelaboides is a common insect in lowland South American forests. It is one of the larger species you might find wandering about in the undergrowth of a tropical rain forest. Yet like most tropical ants, little is known of its biology, and nothing is known about why it has evolved such a charmingly clumsy form.

Like most members of its subfamily (Dolichoderinae), D. attelaboides often tends hemipterous insects for honeydew. And like most members of its genus, it sports a rough, armored exoskeleton.

I will venture my own hypothesis for the strange appearance of this ant, however. I suspect the shape helps Dolichoderus attelaboides escape dismemberment by Eciton army ants.

When ants fight each other, the highest incidence of physical damage is body parts severed at their weak points, the junctions between the hardened plates. Heads detach most easily when cut at the neck, for example. If we look at the two points of unusually narrow constriction on the body of D. attelaboides– partway along the ant’s pseudo-neck and in the middle of the mesothorax- we’ll see they are adjascent to more vulnerable sutures. Any large ant attempting to bite a D. attelaboides will find their mandibles settling into grooves in the middle of armored plates rather than the membranous tissue between them. Essentially, they are false joints. The remaining vulnerability- the waist- is protected by a set of spines.

Most species of Eciton are predators of other ant species. Dolichoderus attelaboides is a sizable ant living in large colonies- a great source of protein for army ants if the colony’s defenses could be breached. Given that both ants are common in the same habitats, and that the size of both ants is similar, I think it entirely reasonable that the gawky pencil-neck form and heavy armor of D. attelaboides is a response to the army ant threat.

Of course, this is speculation. Someone will have to feed dozens of Dolichoderus workers to army ants to test whether the predators’ mandibles actually land in the false joints, and whether those joints provide any real protection from bites.


source: MacKay, W.P. 1993b. A review of the New World ants of the genus Dolichoderus. Sociobiology 22: 1-148. [(31.xii).1993.]


17 thoughts on “Dolichoderus attelaboides: the long-necked ant of the Amazon”

  1. Sorry, Alex, but I do have to say every now and then how much your pictures, and nonetheless your remarks, are appreciated here!
    Every day I hope you have something new, and often enough I reread “old” stuff to get over the day 🙂
    Regarding the speculation, I think you’ve got something there, though I believe the “neck” itself could also, perhaps, have to do with some hunting preference of their own as well – they seem to be quite agile too, with their long legs: probably good for escaping, as well as for hunting…
    Just a thought.

  2. Plain, down to earth, post – very nice!

    The shape of the body reminded me of some insect, who I could not at first remember. I’m still not sure of which one, but after some googling I must say that the bell-shaped joints of the Tricondyla sp. (Cicindelinae) I saw on Borneo is quite similar. (a quick googlage of “Tricondyla” finds many good pictures)

    How this relates to the (supposed?) mimicing of Camponotus gigas by Tricondyla might be worth a thought…

    Any ideas?

  3. I think the answer can be both of them, a natural “blank space” and/or lack of collection effort. Apparently this Dolichoderus is distributed inside rain forests, like Amazon Forest and on Atlantic Forest (the forest that goes along Brazilian Atlantic coast). Between those tropical rain forest, you can find a kind of savanna (called Cerrado, wich has some forests that share some plant genera and families with Atlantic Forest), and a xeric scrubland called Caatinga in the Northeast of the country. The soil of bioma Cerrado is acid, so forests that occur there are not so dense as a rain forest (although there are considerable dense forests that goes inland along rivers). Thus, even if Dolichoderus just can be found inhabiting rain forests, maybe it might not be found in the riparian forests due to collection artifact — actually you can see some collection events in the center of Brazil. In some areas like Caatinga, the information about ants is still incipient.

    1. I live in Cuiaba (cerrado-land) for 8 months a long time ago (mid-1980s) and never found this ant around there. I was going to write something like this, Flavia, but you stated it with more authority.

      1. That explain your Portuguese! I was born and had lived for many years in Minas Gerais (mostly Cerrado-land), and never saw Dolichoderus attelaboides there.

  4. Interesting idea on the morphology. But Eciton kill adult ants with venom, not by biting them into bits. All they need is a firm grasp to slip in the sting. Check any picture of prey transport in Eciton and you will see whole adult ants being dragged back like rag dolls. They only lose body parts once the venom proteases kick in the the literally fall apart from the inside out.

    The morphology for general defense against other aggressive ants patrolling vegetation/in the canopy sounds right on though.

    1. Well, crum. Another perfectly good hypothesis meets the cold reality of facts and logic!

      Thanks for your insightful comments, as usual, Scott.

  5. Don’t some other completely unrelated ants exhibit very similar body plans? It seems somewhat similar to Leptomyrmex, perhaps Anoplolepis, and definitely other ants out there. There are also even Aphaenogaster spp. with ‘necks’ (http://www.antweb.org/specimen.do?name=casent0103338&shot=p&project=). It would be interesting to see how well this hypothesis holds up with ants worldwide…if so it would only further attest to the great importance of army ants in communities.

    1. Yeah, there is Pheidole (gallatrix) gallatrix in Madagascar, and if I am not wrong, there is no army ants in such place, but a great diversity of Cerapachys…

    2. Yes- there are many other species with similar false necks. It’s not the elongate form by itself, though. Leptomyrmex and Anoplolepis gracilipes are gracile but lack the constriction of the head to form a narrow tube with a posterior flange.

  6. Pingback: ants! « El escarabajo azul

Leave a Reply