Fully claustral or not: how your queen founds a colony

Dean Hart · 5 August 2026
Most queens found after a nuptial flight are fully claustral: she seals herself into a chamber and raises the first workers on her own flight muscles and fat, taking no food from outside. Semi-claustral queens have to forage while they do it, and a socially parasitic queen cannot found alone at all.

Do I feed her?

It is the first thing a new keeper asks, usually at eleven at night with a queen in a test tube and a phone in the other hand. The answer is genuinely different for different ants, and not in a matter-of-opinion way. The same treatment saves one queen and kills another.
There are three founding strategies. Most people who pick a queen off a pavement in summer are holding the first one.

Fully claustral: she is the food

A fully claustral queen seals herself into a chamber and does not come out until her first workers do. No foraging, nothing entering from outside, for weeks.
What she runs on is her own body. She has mated and shed her wings, the flight muscles packed into her thorax have no job left, and they get broken down and spent on brood along with the fat she loaded before takeoff. Wheeler and Buck gave a 1996 paper on this the fairly blunt title Depletion of reserves in ant queens during claustral colony founding, which is the whole strategy in six words. The term for a queen who has dropped her wings, incidentally, is dealate, which is where this app got its name.
So for a Lasius niger queen, the black garden ant that most of Britain and Europe is holding in July, correct husbandry is close to doing nothing at all. Somewhere dark, undisturbed, room temperature of about 20 to 24 °C (68 to 75 °F), and water she can drink from without falling in. Putting food in with her achieves nothing and grows mould, and mould in a small damp sealed space is a real way to lose a queen.
Same instruction, different ant, wherever you are. In North America that is the pavement ant Tetramorium immigrans, a 7 to 8 mm queen, call it 0.3 inches, flying somewhere across the continent between March and September, and the black carpenter ant Camponotus pennsylvanicus, a much more imposing 16 to 20 mm, roughly 0.6 to 0.8 inches, and no hungrier for it. In Australia it is Camponotus consobrinus, the banded sugar ant behind Sydney's December flying-ant nights, at the same 16 to 20 mm. All of them seal in, and none of them want feeding.

Semi-claustral: she has to go out

Some queens cannot fund the whole operation from reserves and forage their way through it instead.
The clean published case is Pogonomyrmex californicus, a Californian seed harvester. Robert Johnson described its semi-claustral founding in Oecologia in 2002 and set it against the claustral founding of its relatives, which is the part worth taking away: one genus of harvester ants contains both strategies. Two harvester queens off the same desert road can want opposite things from you.
The Australian version is harder to miss. A bull ant queen, Myrmecia, hunts her way through founding rather than sealing herself in, which is the pattern Caryl and Edna Haskins set out for Myrmecia regularis as far back as 1955. A M. brevinoda queen is 28 to 30 mm, well over an inch, and she is not going to sit quietly in the dark waiting for her thorax to turn into eggs.
In Britain and Europe the example is quieter and the arithmetic is the same. Myrmica rubra, the red ant that runs August into September and can actually sting you, is listed as semi-claustral in the antkeeping.info care data this app bundles. She is a smaller animal than a garden ant queen, 6.5 to 7.5 mm against 8 to 9 mm, call it a quarter of an inch against a third, with proportionally less in the tank. Seal her in and wait and you may be waiting for something that cannot happen.
Feeding one is undramatic. A small insect, a little sugar water, somewhere she can reach it without the whole setup being opened and flooded with light every day.

The queen who cannot found at all

The third case catches people out, because she looks healthy and behaves normally and still gets nowhere. A socially parasitic queen founds by getting into an established colony of another ant and letting its workers raise her first brood. Karl Hölldobler published observations of Lasius umbratus doing this in 1953. In a test tube on her own there is no path to a first worker, and no amount of careful keeping opens one.
Now the awkward arithmetic for anyone in Britain or Europe. A Lasius umbratus queen is 8 to 9 mm. A Lasius niger queen is 8 to 9 mm. Umbratus flies from mid August into late September, well inside the stretch that niger is still using, so a queen found on paving in late summer can be either, out of the same sort of ground, with opposite instructions attached. The umbratus queen is gingery and hairier than the dark brown niger, which is a real difference and not one you notice unless you already know to look for it.
North America runs the same trap with different actors. The citronella ants, yellow Lasius of the Acanthomyops group that smell of lemon when you disturb them, are socially parasitic too, and Cover and Sanwald titled their 1988 chapter Colony founding in Acanthomyops murphyi, a temporary social parasite of Lasius neoniger, which saves anyone the trouble of summarising it. Lasius neoniger is the cornfield ant, the species behind the enormous late-summer flights across the Northeast. So the host is the ant an American keeper actually wants, and the parasite that needs it is in the same genus, flying in the same weeks.

Genus is not enough

Those two pairs are the whole argument. "She's a Lasius" is compatible with a beginner's first colony and with an animal that has no future in a tube, on both sides of the Atlantic, and there is no cautious middle setting that covers both.
The species is what unlocks the advice. Everything short of it is a guess wearing a lab coat.

If you cannot tell yet

Offer rather than impose. A drop of sugar water she can reach without you dismantling her chamber daily costs her nothing if she ignores it, and take it away after a couple of days if it goes untouched, before it goes furry.
Photograph her early, from directly above first, then the side, then the head. That's the order that makes an identification possible later, when you are not also trying to keep her calm. Note the date, the time of day and where you found her too. Month and hour narrow a species down faster than most amateur photographs do.

Where we stay quiet

Founding strategy varies inside a genus, so filling a gap with a genus-level average would mean handing out an instruction about a living animal on the strength of a guess. When the care data carries no founding value for a species, the honest answer is to say nothing about founding and show what is known. Messor barbarus, the harvester ant half of Iberia keeps, is a real example of exactly that hole.
New Zealand is the sharpest version of it. The southern ant, Monomorium antarcticum, is native, widespread and still being pulled apart taxonomically, to the point that people publish on how to split the complex rather than on how its queens found. The recent work on it is about where it nests as soil moisture changes. Nobody should be handing a keeper in Dunedin a confident founding instruction for an animal whose species boundaries are an open question, and the app does not.
That is the same principle behind the forecast itself. Dealate scores the odds of a flight near you from live weather, species by species, so you know which evenings are worth standing outside for. A reason to look up, not a guarantee.
If she is already in a tub and you have not got as far as the founding question, you caught a queen this week, now what covers the first few days. The second wave covers which British species are still flying into August and September, including both the Myrmica and the parasitic Lasius above, and the flight timetable covers which hour of the day to be out there.

The papers

  • Wheeler, D.E. & Buck, N.A. (1996). Depletion of reserves in ant queens during claustral colony founding.
    Insectes Sociaux
    43(3): 297–302.
  • Johnson, R.A. (2002). Semi-claustral colony founding in the seed-harvester ant
    Pogonomyrmex californicus
    : a comparative analysis of colony founding strategies.
    Oecologia
    132(1): 60–67.
  • Haskins, C.P. & Haskins, E.F. (1955). The pattern of colony foundation in the archaic ant
    Myrmecia regularis
    .
    Insectes Sociaux
    2(2): 115–126.
  • Hölldobler, K. (1953). Beobachtungen über die Koloniengründung von
    Lasius umbratus umbratus
    Nyl.
    Zeitschrift für Angewandte Entomologie
    34(4): 598–606.
  • Cover, S.P. & Sanwald, R. (1988). Colony founding in
    Acanthomyops murphyi
    , a temporary social parasite of
    Lasius neoniger
    .
    Advances in Myrmecology
    , 405–417.
  • Raczkowski, J.M. & Luque, G.M. (2010). Colony founding and social parasitism in
    Lasius
    (
    Acanthomyops
    ).
    Insectes Sociaux
    58(2): 237–244.
  • Lamb, R., McCombe, C., Lawrence, K., Macwan, T., Mayer, S. & Jandt, J.M. (2020). Subterranean nesting behaviour in response to soil moisture conditions in the southern ant,
    Monomorium antarcticum
    .
    New Zealand Entomologist
    43: 33–43.
  • Hölldobler, B. & Wilson, E.O. (1990).
    The Ants
    . Harvard University Press, for the founding strategies generally.
  • Per-species founding, size and flight months from antkeeping.info, CC BY 4.0. British flight windows from BWARS.
Frequently asked questions

Do I need to feed a founding queen ant?

Usually not, if she is fully claustral, which covers most of the common species people find after a flight. She raises the first workers on her own flight muscles and stored fat. Semi-claustral queens are the exception and do have to be fed through founding.

What does fully claustral mean?

It means the queen seals herself into a chamber after mating and raises her first brood without leaving or taking in any food. The energy comes from the flight muscles she no longer needs and the fat she built up before the nuptial flight.

What is a semi-claustral queen?

One who cannot cover the whole founding period from her own reserves and forages during it. Australian bull ants (Myrmecia) hunt their way through founding, some American seed harvesters such as Pogonomyrmex californicus do too, and Myrmica rubra, the red ant that flies in Britain from August into September, is usually listed as semi-claustral.

Why can't my queen start a colony on her own?

She may be socially parasitic. Lasius umbratus in Europe, and the American citronella ants such as Lasius murphyi, found by entering an established colony of another ant and having its workers rear the brood. Alone in a test tube there is no route to a first worker, however well she is kept.

Is knowing the genus enough to know how to keep her?

No. Lasius niger is fully claustral and Lasius umbratus is a social parasite; both queens are 8 to 9 mm and both are on the wing in late summer. North America has the same problem with Lasius neoniger and the citronella ants that parasitise it. You need the species, not the genus.

Know the day, and the hour

Dealate scores each species separately and gives you the best window within the day, plus a rolling 7-day outlook and a live map. Every recurring plan starts with a 7-day free trial. A season pass is $12.99 and ends in September.
Start a free 7-day trial →
← Everything we've written