A brood kept in a colony with no queen caste at all — the queenless ants and their gamergates, mated workers who look exactly like their sisters, every one of them born with working ovaries and a spermatheca that can store sperm, and the one who breeds chosen not as a larva but by rank, contest and policing among adults
R910 closed with a door locked in the larva: a honey-bee worker divided from her queen by food in her first three days, built without the apparatus to mate or succeed, so that no death in the hive could ever promote her. Its closing turn asked for the colony where that lock was never fitted — no queen caste to be locked out of, every worker totipotent, and succession settled among grown adults. This page takes that turn underground, to the queenless ants.
No separate reproductive body. In the species gathered under the name gamergate — “married worker”, a term coined in the early 1980s — the queen caste has been lost entirely, or a short-lived founding queen is followed by workers who take her place. What makes it possible is equipment the honey-bee worker never got: workers in these species keep a functional spermatheca, the sperm reservoir, and ovaries that can work. A gamergate is not a different kind of animal from her sisters. In the queenless species she is morphologically identical to them; the difference between the one who breeds and the many who do not is strictly functional — rank, mating, and ovarian activity, nothing written into the body before adulthood.
Everyone equipped, almost no one breeding. That openness creates the problem this page is about. If any worker can in principle mate and lay fertilised eggs, then unregulated, many would — and a colony of rival reproductives works less and coheres less. What the queenless ants run instead is a dominance hierarchy. In Dinoponera quadriceps, a large Brazilian ponerine whose colonies average around eighty adult workers in the studies followed here, a single gamergate holds the alpha rank, with a small set of high-ranking workers — three to five — beneath her, and the beta, the second-ranked, best placed to replace her. The rest keep the colony: foraging, tending, defending, raising a brood that is mostly not theirs, while their own intact equipment stays unused. The keeping here is not a missing organ. It is a queue enforced between complete adults.
The contest, and how a challenge is answered. Rank is not merely waited for; it can be contested. A high-ranking worker in Dinoponera may challenge the gamergate outright. The gamergate’s answer is one of the strangest punishments in the animal literature: she does not win the fight herself. She smears secretion from her Dufour gland onto the challenger — a chemical mark only a gamergate produces in that form — and the low-ranking workers do the rest, seizing the marked pretender and immobilising her, sometimes for days, until she loses her high rank. The alpha keeps her place by a signal; the colony supplies the force. Policing, here, is not a metaphor layered over the hierarchy. It is the mechanism that makes a hierarchy among equals hold.
What winning does to a body. The Indian jumping ant Harpegnathos saltator shows how deep an adult decision can go. Its colonies begin with a winged founding queen; when she dies, younger workers fight for dominance in ritualised antennal duels, over weeks, and a few high-ranking winners become gamergates. The change that follows is not a badge but a rebuilding: ovaries expand to fill the abdomen, the brain shrinks by roughly a quarter, dopamine levels run two to three times higher than in ordinary workers in the work followed here, and life expectancy jumps from a matter of months to several years. Read that against R910 and the shape of the contrast sharpens. The bee’s future was closed before she emerged, by what a larva was fed. The ant’s future reopens in adulthood, at the cost of weeks of contest — and the promotion, when it comes, rewrites organs, chemistry and lifespan in a body that was, until the duels, an ordinary worker.
Succession without a cradle. So a queenless colony replaces its breeder in the opposite direction from a hive. The hive, queenless, must go backwards to a larva young enough not to have been decided yet, and fails if none remains. The ant colony goes sideways, to an adult who was never decided against: when a gamergate dies or fades, a formerly subordinate worker — often from a younger cohort, often the one already standing next in rank — rises, mates, and begins ovarian activity. No caste is unlocked, because none was locked. What had to be settled was never the body. It was the ordering of bodies that were all, from the start, capable — and the colony’s whole apparatus of duels, marks and immobilisation exists to settle exactly that, again and again, for as long as the colony lasts.
One colony, in order: a queenless colony of workers every one of whom carries working ovaries and a functional spermatheca, with no queen caste and no separate reproductive morphology; a dominance hierarchy forms in which a single alpha worker mates and breeds as gamergate while a few high-rankers wait beneath her; the rest keep the colony with their own intact equipment unused; a challenger is answered not by the alpha’s own force but by her chemical mark, which sets the low-rankers to immobilise the pretender until her rank is lost; and when the gamergate dies, an adult who was never decided against rises, mates, and succeeds her — in the jumping ant’s version of this transition, ovaries expand, the brain shrinks by about a quarter, and a lifespan of months becomes years. Schematic, compressed from the published record.
Walk it in order.
Schematic only: the stages compress work across species, studies and settings into one colony. Labelled cautions: Dinoponera quadriceps and Harpegnathos saltator are different arrangements and this page holds them side by side deliberately — the first is queenless throughout, the second begins with a founding queen and turns to gamergates after her death — so do not merge them into one species’ story; the colony size of around eighty and the three-to-five high-rankers are the Dinoponera figures from the studies followed here, not constants for gamergate species generally, which range widely and may hold one gamergate or several; “any worker can become a gamergate” is a statement of potential, not of equal odds — age, rank and cohort strongly shape who actually succeeds, and new gamergates often come from younger workers; the brain-shrinkage, dopamine and lifespan figures belong to Harpegnathos work specifically, and lifespan claims in that literature range from at least a doubling to several-fold, so the months-to-years contrast is the claim, not any single multiplier; and do not present the immobilisation of a challenger as the gamergate’s own violence — the mark is hers, the force is the colony’s, and that division is the finding.