Never Perceived

Keeping with the border never crossed because it was never noticed — a slave-maker that walks through a castle’s single door wearing chemistry that keeps its hosts from recognising it, and carries off brood, and sometimes adults, while the defenders fail to begin defending

R956 closed with keeping undone from inside: Temnothorax longispinosus captives tending a parasite’s larvae faithfully, then, at pupation, detecting a foreign cuticle signature and destroying the pupae — recognition arriving once, too late for the captives to leave but in time to strike. This page takes the turn it pointed to, to the system that stops recognition earlier still, at the door. The parasite is Temnothorax pilagens, the pillage ant, described as a new species in 2014 by Seifert, Kleeberg, Feldmeyer, Pamminger, Jongepier and Foitzik. Its name comes from the Latin for plucking, plundering, pillaging. It is a Nearctic slave-maker, about two and a half millimetres long, known from three sites in the north-eastern United States — Niquette Bay State Park in Vermont, the E. N. Huyck Preserve at Rensselaerville in New York, and Sleeping Bear Dunes National Lakeshore in Michigan — with possible occurrence in south-eastern Canada. In all three populations it was found enslaving the two relatives Temnothorax longispinosus and Temnothorax ambiguus, many nests holding slaves of both.

First, the targets are real castles, and the raiders are very small. The hosts live in hollow acorns, hickory nuts and sticks: thick walls, a single entrance hole about one millimetre across, too small for any larger enemy ant to enter. The raiders’ range of action is only a few square metres of forest floor, and their raiding parties are minimal — on average four slave-hunters including the scout that found the target, and sometimes the scout alone. Because the raiders are so small they slip through the single door that keeps everyone else out. An Amazon-ant campaign may send thousands of warriors; this one sends four, and it works.

Second, in the peaceful raids, the door is crossed because nothing registers as wrong. In several of the raids the researchers watched, the attacked ants did not defend, and let the robbers freely carry away brood — and, in six of the eleven observed raids, adult host workers too, to be integrated into the slave workforce. The authors’ reading, summarised in the Mainz group’s own releases, is that specific neutralising chemical components on the raiders’ cuticles prevent the hosts from recognising them as enemies, so the little or delayed response the hosts show is the absence of recognition, not a choice to stand down. This is the rare stratagem the literature calls chemical insignificance: recognition cues muted or absent, so the intruder is never classed as a foreigner. It carries a stated cost — the same surface hydrocarbons that carry recognition also waterproof the insect, so the chemically quiet risk drying out — and it buys the most complete kind of keeping: hosts that do not defend can be raided again next year, and the raiders’ accounts name the survival of the raided nest as the ideal outcome.

Third, when the chemistry fails, the raider has a second method, and it is not subtle. In other raids the camouflage was less effective, perhaps where a host population was strongly imprinted to a more specific blend of surface chemicals — a defence response was more likely where the attacked colony held a queen, whose presence is associated with stronger imprinting of recognition cues. Where the hosts did recognise the invader, the pillage ant fought with its sting, aimed with what the authors describe as rapier precision at the one soft-skinned spot between the slave ant’s head and thorax, stinging from behind to cause paralysis and quick death. The morphology is matched to the move: strongly developed flexor muscles in the petiole and postpetiole let the gaster bend easily for the aim. A defending host’s chance of winning such a fight is described as nearly nil, and none of the attackers were lost. Host casualties across raids ran from five per cent to one hundred per cent of the attacked nest’s population — raids could be relatively peaceful or highly aggressive, the peaceful kind more often found against queenless host nests — while the raiders went home whole. The authors’ two-method summary is chemical camouflage plus the fencing sting: unperceived entry when it works, lethal precision when it does not.

Last, keeping here has a direction. R955’s system stole the workforce before it could work, pupae carried home so no recognition question could ever be asked; R956’s let the question arrive at pupation, against the wrong brood. Here the question is prevented at the border itself — not by a forged profile passed to the guards, as in R953’s butterfly caterpillar, but by the absence of anything for the guards’ chemistry to catch. The contrast the authors draw is with Temnothorax duloticus, the close congener whose raids look similar in their sting use but whose hosts do recognise it, fight back, and inflict casualties on the attackers. One of the two wins by never being noticed. The other wins in the open, and pays for it.

A walk from the door to the raid. Five stages: a castle with one millimetre door; a party of four; an entry that registers as nothing; brood — and sometimes adults — carried out unopposed; and, where the chemistry fails, the precise sting at the soft neck. Schematic, compressed from the published accounts.

Walk it in order.

Schematic only: the stages compress a species description and the raid observations reported with it, so no single raid is promised this exact sequence. Labelled cautions: do not name the neutralising cuticle components as identified compounds — the published accounts describe their effect, recognition prevented, not a solved chemistry. Do not say the hosts never defend — the record is little or delayed response in the peaceful raids, dragging the invaders out occasionally, and aggressive response mainly when attacked; the camouflage’s failure is part of the account, not an exception to it. Do not present the queen-linked imprinting as a proven cause of the failed raids — it is the authors’ reading of where defence was more likely. Do not merge the six-of-eleven adult-capture figure with any other study’s raid counts — eleven raids are the observed set here. Do not extend the range beyond the three known sites and the possible south-eastern-Canadian note. Do not call the species Temnothorax ambiguous — the host is Temnothorax ambiguus. And do not dress the raiders’ success as a strategy of choice — which raids stay peaceful depends on the hosts’ response as much as on the raiders’ method.