Pavement Ant — Tetramorium immigrans

The craters give it away before the ant does

Most species in this library are identified from the insect. This one is usually identified from the dirt.

Walk the edge of a driveway, a garage apron, a concrete path or the joint where a patio meets a house on any dry morning between April and September, and look at the cracks. Where this ant is working you will find small conical heaps of fine soil pushed up out of the joint — a few teaspoons at a time, loosely piled, with a hole somewhere near the middle. Sweep one away and it is back within a day or two.

That is excavated material. The colony is in the ground underneath the slab, and every gallery it digs produces spoil that has to go somewhere. The nearest crack is where it goes. Utah State Extension describes the same thing from the other direction: nests in soil along the edges of and in cracks around pavement, patios, driveways, sidewalks and foundations, often marked by a small mound of excavated dirt.

Once you have seen a few of them the pattern is unmistakable and it is diagnostic on its own. No other ant we deal with in Whatcom County housing produces neat craters along a hard edge in that particular way. Moisture ants throw up soil mounds too, but in open ground and around timber rather than along a concrete joint. Carpenter ants produce fibrous shavings and no soil at all.

What it looks like under a lens

Get one onto white paper and the ant itself is unremarkable at arm’s length — small, dark, and easy to file mentally under “the little black ants.”

The measurements are consistent between sources. Utah State puts workers at two and a half to three millimeters and dark brown; Penn State gives a slightly wider range at two and a half to four millimeters, dark brown to black; UC IPM records them at three-sixteenths of an inch. All three agree the workers are monomorphic, meaning every one in the column is the same size as every other. That single observation rules out carpenter ants immediately, since a mature carpenter ant nest turns out workers of visibly different sizes from the same queen.

Three structural features settle it, and a hand lens or a phone camera at full zoom will show all of them.

Two nodes on the waist. The pedicel between thorax and abdomen carries two distinct raised segments rather than one. Penn State states it plainly, and it is the check that separates this species from most of the small dark ants that walk across a Whatcom County counter.

A pair of spines on the rear of the thorax. Two short backward-pointing projections, visible in profile. Female reproductives keep them; Penn State notes that the males do not have them.

Grooves on the head and thorax. This is the field mark worth learning. Utah State calls it heavy sculpturing forming parallel grooves; Penn State describes distinctive parallel furrows. Under magnification the head looks finely combed rather than smooth. Nothing else likely to be on your kitchen floor here has that.

Winged reproductives are roughly twice worker size and carry the same furrowed head and thorax, which is useful in spring when a scatter of small winged ants turns up on a windowsill and nobody can tell what produced them.

One naming note, because it causes confusion when people go looking. Older extension material — including the UC IPM key cited on this page — calls this insect Tetramorium caespitum. Newer material, including the Pacific Northwest handbook, calls it Tetramorium immigrans. It is the same animal on your driveway. The name moved, not the ant.

The sidewalk wars

This is the behavior people actually ask us about, and it is worth the space because the answer is reassuring and almost nobody gives it.

Some time in spring — usually a mild dry afternoon in April or May here — a householder walks out and finds a seething black mass on a sidewalk, a patio or a driveway seam. It can be a foot across or several feet long, thousands of ants deep, and it looks like an emergency. People photograph it. People call about it. A reasonable number of people assume something has gone badly wrong with their property.

Nothing has. Two neighboring colonies have met along a boundary and are settling where the line runs. UC IPM describes it directly as aggressive territorial behavior, with spring conflicts between neighboring colonies known as ant wars. Utah State records the same thing: large numbers of ants massed together where one colony is fighting another, over resources and territory.

The useful facts about it are these. It is between the ants and has nothing to do with people, pets or the building. It is short-lived — a day or two, sometimes a few hours — and then the survivors go back to work with the border redrawn. It is a sign of a healthy population of a common outdoor insect rather than a sign of an infestation. And it does not need treating, will not respond usefully to being treated, and would be a strange thing to spend money on.

If you want to do something, hose it off. If you want to do nothing, that is the better answer.

What it does tell you is that you have at least two established colonies within a few yards of each other, which is worth knowing if ants have also been coming into the house. The war is not the problem. It is a free survey of where the nests are.

Where the nest actually is

Hard surfaces are the point. A slab holds heat into the evening, keeps rain off the soil beneath it, gives a stable roof over a gallery system, and provides a clean edge to travel along. From the ant’s side of it, a residential driveway is close to ideal habitat.

The places we find them, in rough order of frequency on properties here:

Expansion joints and cracks in a driveway or garage apron. The single most common, and the easiest to confirm from the craters.

The seam where a concrete path meets a lawn, particularly on the sunny side of a house.

Under patio slabs, stepping stones and pavers, including the ones set into gravel in newer landscaping.

Along a foundation, in the strip of soil between the concrete and whatever has been planted against it.

Under stones, boards, timber offcuts and anything else left lying on soil for a season. UC IPM lists nesting in lawns or under stones, wood or boards alongside the pavement habit.

Beneath a slab-on-grade floor, where the colony is under the building itself. Utah State specifically names sub-slab heat ducts as a nest location, which is worth knowing because that puts the nest inside the thermal envelope and makes activity possible in months when the outdoor colonies have gone quiet.

Foraging range is modest. Utah State gives workers ranging up to about thirty feet from the colony, which means that when a trail comes into a kitchen the nest is genuinely close — under the path outside that wall, or under the slab you are standing on, rather than somewhere across the yard.

Why it is always the ground floor

The way this species enters a building follows directly from where it lives, and it produces a pattern that is worth recognizing because it narrows the search enormously.

Utah State’s list of entry routes is short and specific: cracks in foundation walls, under the frames of sliding glass doors, behind window frames, and along heating ducts. Every item on that list is at or below the level of the ground floor.

So the trail you find will run along a baseboard, up out of the joint where the kickboard meets the floor, in at the threshold of a slider onto a patio, or from behind the trim of a ground-floor window. It will not start in an upstairs bathroom. If ants are appearing on a second story with nothing downstairs, the identification is worth checking again, because that is not how this species behaves.

Two further habits are worth having. UC IPM notes that they typically forage at night and leave visible trails between food sources, which is why a line that was not there at nine in the evening is dense by seven the next morning. And UC IPM also records that they travel through plumbing and electrical runs, which explains the trail that appears under a sink with no visible route in from outside.

Food preference is broad enough to be unhelpful as a diagnostic. Penn State’s summary is that nearly any morsel that falls to the floor gets eaten, and lists sugars, fruit, grease, dead insects and seeds. UC IPM adds honeydew and pet food. In practice, indoors, the resource is usually pet food, spilled crumbs under a toe-kick, or grease behind a range.

One queen, and what that changes

There is a structural fact about this species that decides most of the practical advice, and it is the opposite of the situation with the odorous house ant.

UC IPM states that pavement ants usually have only one functional queen per colony. Utah State agrees, describing colonies of thousands of workers with a single queen. That single reproductive is the whole colony’s future, and it means a fragment of workers cut off from her is a group of ants that will die rather than a new colony.

The consequences run in three directions.

Treatment is more forgiving. A repellent product applied by a householder to part of the trail is still a poor idea for the reasons given further down, but it does not carry the specific documented risk of turning one nest into four, which it does with a multi-queen species.

The colony is findable. One nest, one queen, one place. A network of connected sites moving brood between them is not what you are dealing with here, so tracing a trail to its source is a realistic afternoon’s work rather than a guessing exercise.

Bait reaches the whole thing. Material carried inward and shared has one target rather than dozens, which is a large part of why this species responds faster to correct baiting than the odorous house ant does.

New colonies get started by mating flight. UC IPM puts those in early spring generally, with the qualification that pavement ants can swarm at any time of year, and Utah State adds that swarming happens mainly in late spring but can occur whenever conditions suit. Penn State notes that in a heated building swarming can happen at any season, which is the explanation for the occasional winged ants at a January window in a house with a warm slab.

What they actually cost you

Very little, and it is worth being plain about that on a pest control company’s website.

They do not excavate structural timber. They do not chew wiring. They do not carry a household disease of consequence. They are not aggressive and although Penn State records a stinger in the final abdominal segment, being stung by one is not something that happens in the course of ordinary life.

What they do is walk across food preparation surfaces in numbers, get into open packaging, and persist. That is a real nuisance and it is a legitimate reason to deal with it. It is not a structural finding and nobody should be selling it to you as one.

The soil they bring up has a second-order cost that occasionally matters. Where a colony works under a slab for years it moves a measurable quantity of material out from beneath it, and on thin or poorly bedded concrete that can contribute to settlement at a joint. This is a slow, minor effect rather than a structural threat, and it is far more often blamed on ants than caused by them. Frost heave, tree roots and a base that was never compacted properly account for most of what people show us.

Why this page says “depends”, and on what

The verdict above is genuinely split, and the line between the two sides is easy to draw.

Usually nothing needs doing. Craters along a driveway with no ants indoors is a landscape observation rather than a pest problem. A mass fight on a sidewalk in April is an event to watch rather than a job. Ants on a patio in July, on a deck, around a barbecue or along a garden path are a normal part of a functioning yard in this county, and treating a whole property to remove them is expensive, indiscriminate and pointless. We would rather say that here than after somebody has paid for a visit.

Sometimes it is genuinely worth doing. The picture changes when a trail is running into the building and staying there — a line that reforms within hours of being cleaned, activity in a pantry or a food cupboard, several entry points along one wall, activity continuing into the cold months, which usually means a nest under a heated slab, or a commercial kitchen where any ant traffic across a preparation surface is a problem regardless of species.

The other case that decides itself is a food premises or a shared building. Where units share a slab, a colony under it serves all of them, and treating one tenancy while the others are left alone is not a plan.

What competent work looks like

Identify first. The two waist nodes and the grooved head take under a minute and they determine everything after. Getting this wrong in the other direction — treating a pavement ant as though it were an odorous house ant, or the reverse — is the most common reason an ant job needs doing twice.

Find the nest, which here is realistic. Follow the trail out of the building and along the hard edges. The craters do most of the work for you. Within thirty feet of the entry point there is usually a joint, a paver or a strip of soil that resolves it.

Bait, and keep it stocked. Utah State’s warning is the one that matters: underbaiting yields subpar results. Bait has to be available continuously for as long as the colony is taking it, and the common household failure is putting down one small quantity, watching it disappear in a day, and concluding that the product did not work. Penn State points to slower-acting formulations for exactly this reason — the material has to be carried home before it acts.

Place it beside the trail rather than on it, near the entry, away from children, pets and food surfaces.

Take the competition away while the bait is down, because bait only works if it is the best thing available. The pet bowl, the crumbs beside the range and the unrinsed recycling all matter for the fortnight it takes.

Use non-repellent materials where a product is needed, so workers move through a treated zone and carry it inward instead of turning back at the edge. The Pacific Northwest handbook is direct that over-the-counter sprays are highly repellent to ants and can scatter them through a structure, and that advice holds here even though the budding risk is lower.

Treat the nest where it is exposed. Utah State describes direct application to an exposed nest as a professional option, and where a paver can simply be lifted that is a short job with a definite end.

Then close the route. Once activity has stopped, seal the crack at the threshold, the gap behind the window trim, the joint at the kickboard. Sealing while a colony is actively feeding indoors just moves the entry a foot along the wall.

The things people have already tried

A retail aerosol along the trail. Kills what it touches and erases the pheromone route that would have led to the nest. Less damaging with this species than with a multi-queen one, but it removes the best evidence in the room and buys about a day.

Boiling water down the crack. Occasionally effective on a small, shallow, directly accessible nest under a paver. Useless against anything under a slab, where the gallery system runs well beyond the reach of what you poured in, and a poor idea near anything you would rather not damage.

A perimeter spray around the house. Better placed, and against this species it can genuinely help as part of a plan. On its own, applied without knowing where the nest is, it mostly redirects traffic to a different crack.

Sealing the driveway. People do this hoping to close the ants out. What it actually does is remove the spoil route, and the colony finds the next joint. It is worth doing for the concrete’s sake and it is not ant control.

Cinnamon, vinegar, chalk, coffee grounds. These interfere with a pheromone trail, which is why a line appears to break up for an afternoon and then re-forms a few inches over.

Ultrasonic devices. No mechanism, no effect.

The year, as this species runs it here

February and March. Colonies come out of winter slowly. The first indoor sightings are usually a handful of scouts along a baseboard on a mild day, and they mean very little on their own.

April and May. Mating flights, and the territorial fights that go with the season. Craters start appearing along driveway joints as excavation resumes. This is when most of the alarming phone calls happen and most of them need no treatment.

June through August. Peak activity, peak foraging, peak indoor pressure. The trails are easiest to follow now, which makes it a good stretch to work a problem properly. Warm dry spells push ants indoors looking for water rather than food, which is why a trail sometimes ends at a condensating pipe with nothing edible along it.

September and October. Outdoor resources fall away and pressure on buildings rises as the rain returns and shallow soil nests take on water.

November through January. Outdoor colonies go quiet. Anything active indoors now is being sustained by warmth — a heated slab, a sub-slab duct run, a boiler cupboard — and that changes the search rather than the treatment.

Pavement ant or odorous house ant

These two account for most of the small dark ants in Whatcom County kitchens, and the difference decides how patient you have to be.

The waist. Pavement ants carry two nodes. The odorous house ant carries a single node that is flattened and tucked out of sight beneath the abdomen, so from above it appears to have no waist segment at all. This is the definitive check and it survives a squashed specimen.

The head. Grooved and sculptured on a pavement ant, smooth on an odorous house ant.

The spines. Present on a pavement ant’s thorax, absent on the other.

The smell. Crush half a dozen. A strong rotten-coconut odor means odorous house ant. Pavement ants produce nothing distinctive.

The dirt. Craters at a concrete joint are pavement ants. Odorous house ants build no mounds.

The queens. One functional queen here against many in an odorous house ant colony, which is why one responds to bait in days and the other takes weeks, and why a repellent spray is far more dangerous applied to the second.

And where the ants are large, vary in size within the same column and are running at night around a window frame, neither of these is the answer — that is a carpenter ant question, and the stakes are different.

The short version

Small, uniformly sized, dark, two waist nodes, a grooved head, a pair of spines, and a heap of fine soil at a crack in your concrete. It will not hurt your building. The spring battles on the sidewalk are normal and self-resolving. Outdoors it is part of the landscape and worth leaving alone.

Indoors, if it keeps coming back after the crumbs are gone, it is a bait and perimeter job with a findable nest and a short timeline — which makes it one of the more satisfying problems in this library, and one of the cheapest to get right.