I have just got back from fieldwork in the Bolivian Amazon and witnessed this strange phenomenon of ants swirling around a kernel of living ants holding eachother in a ball. They kept going in the same direction even when disrupted and there was no hole or food on the ground below. Does anyone know what and why they are doing this?
There's an explanation here.
Sounds a bit dodgy to me though. I can find several references to it on the Internet, so if it happens that often, and they die as it says in the video, why has it evolved? But then why do whales get stranded on beaches etc.?
Alex, a great video of such strange and rare ant behaviour!
I can't find accurate sources on it, but I do believe it is due to some ants getting separated from their colony and following each others pheromones aimlessly. It is known that some ants act 'lost' when they loose their colony as this is what they live for. Note that in some species workers can still lay eggs that turn into flying male ants (drones) in the abscence of a queen.
I will post the video on some ant groups and see what feedback we get.
Tim from antark.net
That seems a good explanation to me (jag's video).
If such a loop is started, by chance, as just one pheromone-trail circuit, and it happens to contain a lot of ants, it seems likely that:
A1. by their random ventures either side of the trail itself, short-cuts back to the circuit will be created
A2. by virtue of the curvature of the original trail, those short-cuts are likely to be more-successful on the inside than on the outside
A3. those short-cuts will be reinforced by other ants following (again more on the inside than the outside)
A4. repetition of that short-cutting will tend to make the circuit into a spiral
A5. towards the centre of the spiral, once ant-number-density gets to a certain point, the pheromone trails probably form a layer or desnse mesh rather than lines, allowing the ants to stop spiralling and just be nudged on in the same direction, thus forming a disc (as opposed to a problematically deep pile of ants)
B1. (further to A1) random ventures away to the outside may result in some expansions of the 'event horizon', but smaller ant numbers mean they will be less likely to be reinforced by followers
B2. random ventures that do not join-up will result in some ants being lost from the system - but they will not appear in the system we see, giving a slightly false impression of the perfection of the system
C. a critical number/density of ants may be needed to catalyse the process; if that is not met, a spiral/circuit may form but may not be as obvious and may not persist as long - both making such events less likely to be noticed -- resulting in greater attention and surprise when the full phenomenon is observed (this is an issue of human observation and suboptimal -objectiveness)
Processionary moth caterpillars also follow trails, and it has been shown in experiments (I don't know about in the wild) that they can be induced into a loop (this was actually done on the rim of a bucket). In that situation they were observered to crawl until they died.
- oak processionary (Thaumetopoea processionea)
- pine processionary (Thaumetopoea pityocampa) - maching: http://upload.wikimedia.org/wikipedia/commons/thumb/3/3e/Thaumetopea.pityocampa.01.jpg/240px-Thaumetopea.pityocampa.01.jpg
With both the ants and the caterpillars, I wonder what mechanism/sense/factor/event normally decides (when they are not in-circuit) when the marching ends. Without knowing that, my thoughts above are at risk of being off the mark. That is, at the end of a normal march, the behaviour changes, and maybe it is a mistake in the march-end signalling that results in the spiralling/circling.
Just my thoughts.
Thanks for the thoughtful reply. I am not 100% convinced of looping trails as this must happen all the time amongst army ants where they have a moving and living nest with workers going out all the time. Surely their trails would loop all the time??