Ants are incredible little creatures, known especially for their complex societies and teamwork, but we've never quite seen them work together like this. In this video, the speed of the internet meets the deliberateness of the scientific enterprise and reveals the limitations of each.

Earlier this week, a viral video on the site LiveLeak revealed an unexpected sort of teamwork among ants. In order to haul their millipede prey back to the nest, a group of ants formed a daisy chain, with each ant biting onto the rear of the ant in front of it.

Click here to view this embed.

When ant expert Alex Wild came across the video, he did what any good scientist would do: turn to the literature to see if the curious behavior had ever been documented. It hadn't:

...the clip appears to show an Asian Leptogenys daisy-chaining their bodies in parallel lines to haul away a large millipede. I have spent the morning searching the technical literature for mention of this unusual behavior, and am coming up empty. Some Leptogenys species, including L. diminuta, L. nitida, and L. processionalis, are known to forage in groups and transport prey "cooperatively" (source, source). What is meant by "cooperative" is often vague. (For more, see this excellent recent review of cooperative transport by Helen McCreery). Yet I didn't find any explicit description of workers linking up, mandible to abdomen, to pull together.

What Wild did find, however, was another video of a similar behavior on Youtube, uploaded by a Cambodian beekeeper. Just because the scientific world isn't aware of the phenomenon doesn't mean that the internet isn't.

Iain Couzin, an expert in animal collective behavior, was also stunned:

Finally, researcher Christian Peeters confirmed that he had seen the daisy-chain behavior in Cambodia as well. He just hasn't had enough observations in order to write about it in a scientific paper. In an email to Wild, he explained:

The behaviour was very stereotyped: mandibles grab preceding ant's gaster (between first and second segment)...

...The millipedes were 130mm long, identified as order Spirostreptida (Diplopoda). Ant is 16mm long.

Back then I reviewed the literature and found no other record of chain behaviour in Ponerinae. No record of millipede predation in Leptogenys. Specialized hunting on millipedes is restricted to Thaumatomyrmex, Probolomyrmex and Gnamptogenys, but these are solitary hunters on a very different kind of millipedes (polyxenids).

I started writing a ms on this behaviour (formation of chains in ants through a self-assembling behaviour) but sadly I have not been able to get further observations. It seems to happen at certain times of the year only.

By an amazing coincidence, two days ago I finished fieldwork in northern Thailand and came across the same Leptogenys species. There were cleaned out ring segments of big millipedes outside entrances. Unfortunately I did not observe any raids.

The entire episode reveals one of the myriad ways in which the internet has fundamentally changed scientific research in animal behavior and behavioral ecology. Indeed, in many ways the internet has changed the way science is conducted across many fields. Armed with nothing but an internet connection, anybody can upload a video to the internet. It can then pass across the desktops of researchers who are equipped with the knowledge that they're looking at something special, and who have the resources - if they want - to pursue the question further.

But it also reveals the tremendous disconnect between formal scientific knowledge and the collected knowledge distributed across thousands of individual scientists. When Wild turned to the typical sources of scientific information, he found nothing. But it turned out that while other researchers had documented the behavior, that knowledge had not yet made it into the formal record. How many "new species" are in fact well known to individual researchers? How many "novel behaviors" are, in truth, old hat to those who spend time around the animals in question?


Meanwhile, the internet (like any other tool) is not free of problems either. With any viral video, as Wild points out, multiple copies find their way onto new platforms, often without any attribution. When that happens, it is sometimes difficult to track down the original photographer or videographer, making any kind of serious scientific data collection nearly impossible. Stealing content doesn't just rob content creators of their work; it can also rob the scientific community of valuable information.

While it's good to have some filters in place - not just any old scrap of information warrants inclusion in a scientific journal - perhaps some enterprising researcher will come up with a strategy for organizing all those scraps of "maybe interesting" and "potentially novel" observations. Maybe even something leveraging the power of the internet.

[Myrmecos via Nerdist]