Adventures among Ants. Mark W. Moffett

Such operations are vaster than any raid. Colony members that normally wouldn’t venture from the nest—every egg, larva, and pupa, every swollen and cowering replete, every delicate callow worker—join a caravan that proceeds as far as 80 meters to a new nest site. The enterprise involves a staggering protective force of workers exploring almost to the span of my hand from the trail flanks. Two to six nights are required, with the convoy taking a break during daylight hours.

      Only once have I seen the queen in a migration, and that was in the Malayan species Pheidologeton silenus, which is similar in many ways to P. diversus, the marauder ant. It was near midnight. I had been sitting for six hours in a particularly water-saturated corner of dense rainforest at Gombak Field Station in peninsular Malaysia, watching ants hauling their brood. Suddenly, there she was, part of the convoy, marching along with her stout body and strong legs as if she were designed for a life on the run. Escorting her was a tight retinue of several hundred minor workers. Some of them rode on her body; others ran in a mass a couple of inches ahead and behind her and on each side. The emigration column swelled as she passed, with the entourage flowing at exactly her pace. So quickly I had no time to pull out my camera, she disappeared where the trail led into the dripping brush.

      Why move? Changing house can be a time-consuming chore. The honeypot ants of the southwestern United States, who laboriously carve nest chambers into tough desert clays, seem to never move: perhaps their expenditures on home construction are too high. For others, migrations occur only after a dire circumstance, such as the flooding of the nest or attacks by a vertebrate predator. But with the marauder, when I expected a migration to occur, it often didn’t, and vice versa. I documented migrations of colonies that were eating well (in one case, dining on daily servings of bird seed supplied by me) but then inexplicably moved to a barren area. Conversely, colonies often stayed put even after I had dug up part of the nest for study.

      The frequency of marauder colony migrations remains a mystery. My best guess is that colonies move a few times a year on average, but because I couldn’t watch colonies around the clock, I could not be sure the colony at a site was the same or had changed since I had last been there. Several times after observing one colony migrate, for instance, I saw another move into its abandoned nest, which made me wonder if the ant colonies were like human families upgrading their homes. One colony moved 8 meters and then two weeks later relocated to its original location.¹¹

      Similar to marauder ants, though at the opposite extreme from homebody honeypot ants, are the nomadic army ants, which have been characterized as unique for the frequency, predictability, and organization of their migrations. Describing the transient domiciles of African army ants, the Reverend Thomas Savage reminded his readers in 1847 that “a man’s dwelling indicates the nature of his employment.”¹² While the large colonies of other ants require intricate nests, and like large human populations are hard to move, army ants avoid investing in substantial shelters so that they are as prepared to change locations as are Bedouins with their tents. Many army ants use abandoned chambers under objects or beneath the ground, as the marauder ant does.

      New World Eciton burchellii army ants take this trait to an extreme: their nests are called bivouacs, because the only physical structures are the bodies of the ants themselves, as a half million or more gather under a low branch and form a hanging, bushel basket–sized mass of interlinked bodies. (Other army ants form similar chains within their underground enclaves.) As Reverend Savage might have predicted from the simplicity of Eciton burchellii encampments, the colonies of this species and a few others migrate with great regularity, every day for weeks at a time. It’s thought that as their armies became more effective in the ancient past, army ants tended to exhaust the supplies of prey near their nest, forcing the evolution of such roaming behavior in response to a recurrent need for fresh hunting ground.¹³ No surprise, then, that one army ant species has been shown to migrate more often if the colony is underfed.¹⁴

      The plainly nomadic Eciton burchellii has been a research favorite and has dominated our perceptions of army ant behavior. The evidence suggests, however, that other army ants vary in their nest movements and that the species that relocate in a regular migratory cycle represent the minority. Some early naturalists, who had the luxury (rarely afforded modern researchers) of remaining for years at a site, recorded army ant colonies staying put for many months.¹⁵

      If my assessment is correct, many army ants may be no more nomadic than the many other ant species that migrate periodically, and sometimes on a regular basis.¹⁶ For species with colonies of a few individuals, relocations can commence, as they do with Bedouins, at any provocation, as appears to be true of Myrmoteras trapjaw ants whose nests in crannies in forest litter require little construction. At least some ants seem to be driven to pull up stakes by food requirements or outright hunger. A nomadic mushroom-eating ant in Malaysia, for example, changes its nest most often when its food runs low.¹⁷ With the marauder ant, the connection between migration and foraging remains uncertain, but in all likelihood this species doesn’t eat itself out of house and home as often as the more predatory army ant, permitting extended residences in one place. As with other wayfaring ants, when the need for a migration arises, the establishment of reliable thoroughfares to the new nest is the cornerstone of its success.

      CREATING A NEW COLONY

      Marauder ants and army ants share a common strategy for mass foraging and to some extent a proclivity for moving nests. But they have very different strategies for establishing new colonies. In most garden-variety ants, the young queen flies from her mother colony to mate. The foundress snaps off her wings upon arriving at her destination and then digs a nest chamber. In it, she rears her first crop of workers unassisted. (In some species the queen forages at this stage, but more commonly she doesn’t leave her chamber and survives off her body fat.) As soon as these few ants mature, they take over all the labor and begin the first tentative foraging expeditions, leaving the queen to lay eggs, basically the only task she will accomplish for the rest of her life.

      The queen of an army ant colony, in contrast, does not grow wings or fly away. Instead, through a system known as fission, one of the queen’s daughters inherits half the colony and takes it as her own; the other half goes its own way with the original queen. Because even a start-up colony has thousands, if not hundreds of thousands, of workers, army ants never have to deal with problems of a meager labor supply. From its inception, a colony can always count on a huge contingent for its raids.¹⁸

      How marauder colonies get their start, however, is a mystery. This much I know: their queens are tough, and they are excellent runners during migrations, but otherwise they don’t resemble army ant queens in that they do grow wings and fly from the mother nest. After mating, they dig a nest chamber and attach their eggs to a hairless patch on the underside of their abdomens, a behavior unknown for any other ant. They carry the eggs around by tucking the abdomen partially under their bodies, which forces them to stand awkwardly high on their legs.

      Unfortunately, despite my repeated, frustrating attempts to observe a colony’s establishment, the queens I managed to follow didn’t survive to rear workers. Nor did I find a marauder “starter” nest, or any nest with fewer than tens of thousands of workers—again, despite many long searches. So this part of my fieldwork remains tantalizingly incomplete. I’m eager to find out how a juvenile colony, lacking multitudes of ants, gets its food. While mass foraging becomes obviously advantageous when there’s a labor force that vastly outnumbers its quarry, perhaps droves aren’t necessary for success beyond that achieved by an ant foraging alone. If ants use a buddy system, even two workers, or at least a small group, might travel together to gang up on prey. Whether the developing marauder ant colony employs such a strategy is at present pure conjecture.

      My failure to locate small nests suggests that extremely few new marauder ant colonies survive. Indeed, I often saw young queens, after their mating flight, being killed by workers from an established marauder nest. A queen’s survival likely requires her family to grow swiftly until it has a safely large population—a rare event. Fortunately, we can gain clues as to how colonies mature quickly from the marauder’s relative silenus. At Gombak Research Station, I excavated silenus nests, put the nest soil and all the stray workers in a bag, froze the bag, shook it up, counted all the