Shortly after arriving in the Gorongosa National Park in Mozambique I witnessed a puzzling phenomenon: while exploring the network of roads in the woodland savannas of the park our local driver would barely slow down to avoid hitting antelopes and warthogs, but immediately slammed on the brakes if he noticed a long column of large black ants that were streaming from one side of the road to the other. We couldn’t quite get the exact explanation for his reluctance to drive over the insects (although we were very happy about it), but eventually gathered that driving over them could bring great misfortune. In most places on Earth killing a bunch of ants carries the moral equivalency of blinking, and one might wonder why in Mozambique people would show such respect for these insects. But having met these particular ants before I immediately understood that, like so many seemingly irrational cultural oddities, this one also had a very rational explanation.

I first encountered Matabele ants (Pachycondyla analis) a few years earlier in Africa, and it was a memorable experience. They are named after a particularly fierce tribe of Zulu warriors, and fully deserve this designation – a single ant delivers one of the most painful stings I have ever experienced, and just a few of them can put you out of business for most of the day. So, yes, crossing paths with Matabele ants will certainly bring great misfortune.

But other than their propensity for overreacting to being crushed under one’s foot, these insects are truly amazing creatures. Matabele ants are specialist termite hunters, and do so in a very sophisticated way. It all begins with a single scout leaving her underground nest and embarking on a mission of discovery. The scout wanders, seemingly aimlessly for about an hour in all directions, searching for a nest of termites of the genera Odontotermes and Microtermes. Often she finds nothing, and returns to the nest empty handed. But if she finds a termite mound her behavior changes immediately – she runs back along the shortest possible path back to the nest, leaving behind a trail of pheromones that will guide her nestmates to the source of prey. The pheromonal trail is produced by two glands in the ant’s abdomen: the pygidial gland, the secretion of which has a powerful recruitment effect on her nestmates, and the venom gland, which paints longer lasting orientation cues on the trail. Once back in the nest, it takes as little as 60 seconds to get the entire worker cast of the nest mobilized and streaming along the chemical path left by the scout towards the termite mound, which can be as far 100 m away (in human terms, it compares to walking to a grocery store located 10 miles away.)

Upon arrival at the termite mound, the ants pour in through an opening marked by the scout, and begin to slaughter the termites. Within minutes, several thousand termites are dead. Each Matabele ant then stuffs her mandibles with as many limp bodies as she can carry, and the entire column runs back to the nest. In most cases the ants suffer no casualties during the raid, and the entire colony acts almost as a single, large, predatory organism, equivalent in its impact on the termite colony to a small pangolin.

As I watched the column streaming in a tight formation towards the termite mound during one of many night raids that I encountered in Gorongosa, I could clearly hear the constant chatter of hundreds of workers. Bert Hölldobler and his colleagues (Hölldobler et al. 1994. J. Ins. Physiol. 40: 585-593) concluded that the sound made by the ants serves only as a warning to potential predators, and has no role as a cohesion signal to help keep the column together. But when I recorded these signals with an equipment of much greater sensitivity and frequency range than what was available to these scientists in 1994, I realized that the ants produced not one, but two types of sound, one of frequencies an order of magnitude higher than the warning signals typically heard in insects. One kind of sound was produced by rubbing together segments of the gaster, whereas the other was probably made by the ant’s mandibles. You can here the sound of the marching column here; a call of a single ant, slowed down 10 times, can be heard here – notice the knocking sound made by the mandibles, and the scraping sound made with the gaster.
In the light of the recent discoveries of acoustic communication in ants, I think that I will revisit and test in Gorongosa the idea that Matabele ants use sound to locate the column if a worker becomes separated from the rest of her nestmates. Ants may have evolved a sophisticated language of chemical compounds but when you need to shout nothing beats the good old sound waves.