Brown Marmorated Stink Bug — The One That Actually Matters

The insect on the windowsill in February

A shield-shaped brown insect appears on the inside of an upstairs window in the middle of winter. Nothing has been open for months, there is no obvious way in, and the natural conclusion is that something is breeding in the house.

It is not. That insect entered the building in October, went dormant in a wall cavity or an attic, and was woken by a warm week. Nothing about the brown marmorated stink bug’s indoor life involves eating or reproducing, and the number in your walls has been falling steadily since the day the last one squeezed in.

Understanding that one mechanism answers nearly every question people bring to us about this animal, and it is why the useful month for doing something is August rather than February.

What it looks like

WSU Tree Fruit describes Halyomorpha halys as one of the larger stink bugs in Washington at 12 to 17 millimeters, marbled gray-brown, and shaped like the heraldic shield that gives the whole family its shape.

Two features do the identifying work, and both are on the front half of the animal.

Pale bands near the end of the antennae. The fifth antennal segment has a pale base and the fourth has a pale tip, which reads at arm’s length as one or two white rings toward the antenna’s outer end. WSU’s own Whatcom County identification guide leads with this.

A smooth shoulder margin. The front edge of the pronotum — the plate behind the head — runs smooth on this species, with a single point beneath each eye. Several of our native stink bugs carry a toothed or spiny leading edge there instead, and running a fingernail along it separates them faster than any amount of squinting at color.

Two further marks confirm it. The edge of the abdomen carries alternating light and dark banding, visible as a striped fringe where the wings do not quite cover the body. And the head and shoulders are pitted with punctations that catch the light green or coppery, which the Whatcom guide records as iridescent green and which is oddly beautiful once you have noticed it.

Color alone will not get you there. Mottled brown describes a good number of insects, several of them native, harmless and doing nothing to anybody.

What it is not

Three insects generate the same photograph in our inbox each October, and the WSU publication FS079E separates them explicitly.

Not a western conifer seed bug. That one is longer, narrower, and carries a flattened leaf-like widening on the hind leg that nothing else here has. It also flies noisily. If it announced itself by hitting a lampshade, it is the seed bug.

Not a boxelder bug. Black with sharp red lines, half the bulk, and gathering in dense groups rather than as scattered individuals.

Not one of our native stink bugs. The rough stink bugs and the consperse stink bug are both established here and both do the same seasonal thing in smaller numbers. They have plain antennae, and several have that toothed shoulder edge. They are not the invasive one and nobody is asking you to report them.

Where it came from and where it is now

The species reached the eastern United States around the mid-1990s, turning up in Pennsylvania in 1996, and arrived in the Pacific Northwest in the mid-2000s with a Portland detection. Washington’s first documented sightings came in the fall of 2010, and at the time FS079E was published they were confined to Clark and Skamania Counties in the far southwest of the state.

That is not where the story stopped. WSU’s tree fruit entomology program, which maintains the state record, had it in 29 counties by November 2020 and 30 by December 2022, most of them strung along the I-5 corridor, with a record 9,500 individuals reported in 2022 alone. WSU Extension in Benton and Franklin Counties now describes the insect as present in almost every county in the state.

For here specifically, WSU’s Whatcom Ag Monthly is the source worth quoting, because it is county-level rather than extrapolated. Its survey found fewer than ten adults and no nymphs in Whatcom County in 2017 — a small number, and a real one. The authors add that established populations are known in British Columbia, which they take as strong evidence that the species is established in both Skagit and Whatcom Counties.

So: present, established, and at nothing like the densities that made this insect notorious in the mid-Atlantic states. That is an honest description of a species still filling in its range, and it is the reason the reporting request below is worth acting on.

Why it comes indoors

The behavior is an overwintering strategy and nothing more.

The adults that reach late autumn have to survive to spring, and their natural refuge is a crevice under bark or in rock. A heated building presents the same thing at a far better temperature, and the insects find it by warmth. As nights cool through late September and October, they gather on the elevations holding the afternoon sun and work inward through openings that already exist — the joint where a window frame meets siding, a lifted siding lap, an open soffit return, a gable vent with tired screening, the gap around a service penetration.

They cannot make an opening. They have no jaws for it. Every route in was already there, put in by construction, settlement or somebody else’s afternoon with a drill.

Once through, they settle into the void and stop. Penn State Extension is unambiguous that they will not reproduce inside a structure or cause damage there, and UC IPM records the same. They do not feed indoors either — their mouthparts are built for piercing plant tissue, and there is no plant tissue in a wall cavity.

In spring the survivors leave under their own power to feed and mate outdoors, which produces one final flurry of indoor sightings that looks like a new problem and is the closing act of the old one.

The smell, and one correction

They are called stink bugs for a reason. Disturbed, handled or crushed, they release a defensive secretion described variously as coriander, burnt rubber or something acrid, and it lingers in a room and in a vacuum canister.

Don’t crush them. There are three good reasons and one bad one, and the bad one is repeated so widely that it is worth taking apart.

The good reasons: the smell is unpleasant and hangs around, the fluid marks some surfaces, and Penn State notes that a bug crushed against skin can produce dermatitis at the point of contact in a small proportion of people.

The bad reason is the claim that crushing one calls in others. The National Pesticide Information Center — run jointly by Oregon State University and the US Environmental Protection Agency — states the opposite in plain words: killing a stink bug does not attract more stink bugs, and the odor released will not attract other bugs. We have seen that claim on a great many pest control websites and we are not going to repeat it here, including where it appears in older writing of our own.

What is true, and is probably where the confusion started, is a different chemical story entirely. NPIC also records that when one of these insects finds a suitable winter shelter, it secretes an odor that attracts other stink bugs to it. That is an aggregation signal produced by a settled insect, not a distress signal produced by a squashed one, and it explains something the crushing myth does not: why the same building collects them year after year while the identical house next door does not.

So the advice stands and the reasoning changes. Vacuum rather than crush, empty the canister outside because they are not necessarily dead, and keep a cheap machine for the job or fit a stocking inside the hose so the smell never reaches the motor.

The part that genuinely matters

Everything above is a nuisance. The reason this species has a research program behind it, a state distribution map and a public reporting address is agricultural.

WSU Tree Fruit rates apples as a high-risk crop, with pears and peaches similarly vulnerable and documented damage on cherries, apricots and plums. The feeding mechanism is a proboscis pushed into developing fruit; in apples a brownish depression forms around the puncture, often becoming more obvious in storage than at harvest. Peaches can show gummosis and internal injury. WSU Hortsense describes the equivalent on peas — sunken lesions on the pod, whitish or brown spots inside, seeds deformed or failing to develop.

None of that kills a plant. It makes the produce unsaleable, which for a grower is the same outcome arriving by a different route.

Two features make the species worse than the natives it resembles. Its host range is enormous, so it moves between crops and ornamentals and wild hosts across a season rather than staying put. And WSU records at least two generations a year where our native stink bugs typically manage one, with both nymphs and adults feeding on fruit.

This county grows berries, tree fruit, seed crops and vegetables at scale. That is the actual reason a Bellingham homeowner should care about this insect beyond the windowsill, and it is why we would rather you knew what it was than treated it as one more brown bug.

If you are farming and finding them in numbers, that conversation belongs with WSU Extension. The monitoring — pheromone traps using the aggregation lure and a synergist, beating tray sampling, the whole apparatus — is theirs, it is good, and a residential pest control company should not be improvising alongside it. There is also encouraging news on that side: an adventive population of the samurai wasp, Trissolcus japonicus, a parasitoid with a very high rate of parasitism on this species, has now been found in Washington and Oregon without anybody releasing it.

Please report it

This is the one species in our library where an ordinary sighting has value to somebody.

WSU’s tree fruit entomology group builds the Washington distribution map from public reports. They ask for a photograph, a general location — inside or outside the house, in a car, a backyard, a park — and the address where it was found, sent to tfrec.reportBMSB@wsu.edu. The Whatcom Ag Monthly guidance adds that a suspect insect can be taken to a local extension office.

Given that the county-level record here rests on a survey that found fewer than ten adults, a photograph from a Ferndale kitchen window is not a trivial contribution. It costs a minute and it improves a map that growers use.

What we do, and what we will not

The verdict at the top of this page says it depends, and it depends on two things: the month, and whether the building is a house or a growing operation.

Late August into early September is when treatment does something. An exterior application to the elevations they gather on, in place before the aggregation starts, measurably reduces how many get inside. That is a real service and we provide it. One honest caveat: it is a residual sitting on siding, and siding here gets rained on through autumn, so its useful life is shorter than it would be in a dry climate. It is the sealing that keeps working after the product has washed away.

The same treatment in late October is largely spent. By then a good share of the population is behind the siding rather than on it.

Once they are in, we will not treat the void. Not because it is difficult, but because it is the wrong thing to do. Very little reaches an insect wedged dormant in a cavity, and a treatment that succeeds leaves a mass of dead insects in an inaccessible space. Penn State spells out the consequence: carpet beetles feed on the dead stink bugs and then go after woolens and stored dry goods. You would be trading a harmless nuisance for a destructive one.

Interior sprays are not worth buying. UC IPM calls indoor sprays generally ineffective, NPIC says the same, WSU Extension says pesticide treatments applied indoors typically prove ineffective, and Minnesota Extension does not recommend them at all. Four sources, one answer. These insects are not feeding and not moving across treated surfaces in any consistent way, so there is nothing for a residual to do except sit in your house.

So the honest shape of a plan made in January is: vacuum what appears, put the sealing work in the calendar for late summer, and expect a genuinely different winter afterward.

Sealing, which is the half that lasts

The openings that matter on a Whatcom County house are mostly at height, which is why this problem survives on buildings where somebody has carefully sealed everything they could reach from the ground.

Soffit and fascia runs, gable and attic vents with coarse or damaged screening, the frame-to-siding joint on the sunny elevations, the open ends of vinyl siding and its J-channel, the junction where an addition’s roof meets a wall, service penetrations at height, and any disused duct terminating in open air.

Because these insects are small, slow and incapable of chewing, sealant and insect-grade screen are sufficient. This is one of the few exclusion jobs on this site where hardware cloth and sheet metal are not the answer.

It is also worth doing the interior side. Sealing around a recessed light housing, a bathroom fan surround, an attic hatch or the top of an interior wall does not stop them entering the building, but it stops them entering the room — and for most households that is the part experienced day to day.

Book it for August, when it overlaps with the rodent sealing window and one trip up a ladder solves several problems at once.

Around this county

The agricultural corridor — Lynden, Everson, Nooksack, Sumas. Properties beside berry fields, orchards and large gardens see the highest autumn numbers, because a population that built up over the growing season is looking for shelter and yours is the nearest building.

Newer subdivisions on open ground around Ferndale, Lynden and the Bellingham edge developments. Large unshaded south and west elevations are exactly what these insects locate from a distance.

Older city stock — the Lettered Streets, Columbia, Sunnyland, York. Smaller walls, far more gaps, so fewer land and more get in.

Seasonal and waterfront property at Blaine, Birch Bay and Semiahmoo, where nobody is present in October to see the wall or in February to notice the windowsill, and the owner returns in spring to a collection on the sills.

Along the I-5 corridor generally, which is where WSU’s map is densest and where this species has done most of its traveling.

A year in the life

Knowing the whole cycle makes the indoor phase far less mysterious, and it takes a paragraph each.

Spring. Adults leave their overwintering shelters as the weather steadies — Penn State puts emergence in the eastern states between late March and June — and disperse to plants to feed. There is a fortnight or so of feeding before mating begins.

Late spring and summer. Eggs go down in neat masses on the undersides of leaves; the Whatcom guide records twenty to thirty in a cluster, white to pale green. First-instar nymphs are bright orange and black and stay huddled around the hatched eggs. Later instars turn spiny — including on the shoulders, which is a useful mark — and already carry pale antennal banding.

Summer into early autumn. Five nymphal stages, then new adults. Where the season is long enough for a second generation, this is when the numbers really build, and it is the second brood that supplies the autumn wall.

October. Everything above converges on your siding.

November to March. Dormancy, punctuated by whatever warm spells the winter provides.

Nothing in that cycle happens indoors. The building is a refuge with a roof on it, occupied for one season by insects that arrived fully grown and will leave in the same condition.

Why your building and not the neighbor’s

Households take this personally, and the profile of an affected building is specific enough to be worth setting out.

Orientation first. South and west elevations hold the afternoon sun and radiate it into the evening, so that is where the insects concentrate; a north wall is largely unvisited. Pale siding warms readily and shows them plainly, which is partly a real effect and partly a matter of what you can see against a light background.

Height matters, because a taller wall presents more warm surface and more openings nobody has ever examined. Vinyl siding is a particular case: its construction leaves open channels at the J-channel and at panel ends, and those sit exactly where the insects are gathering.

Then there is what surrounds you. A property beside a hedgerow, an orchard, a berry field or a large ornamental planting sits next to a season’s worth of production.

And finally the aggregation signal described above, which is the part that makes this persistent rather than random. A building that has hosted them before is chemically marked as somewhere that worked, and the following autumn’s insects have a reason to prefer it. That is the mechanism behind the complaint we hear most — that it is the same house every single year — and it is also the argument for sealing rather than waiting to see whether this year is different. It will not be.

If you rent

These calls reach us from tenants at least as often as from owners, and the answer is structural in both directions.

Nothing a tenant does inside changes the outcome. The gaps are in the building envelope, generally at height, and closing them is exterior maintenance rather than housekeeping. A tenant cannot screen a gable vent or seal a siding lap, and a tenant buying a product from a hardware store to spray a windowsill is spending money on the one intervention every extension source agrees does nothing.

What a tenant can usefully do is vacuum, keep the machine’s canister emptied outside, and describe the problem accurately to whoever manages the building — that these are overwintering insects entering through the envelope, that the work belongs in late summer, and that it is the same category of job as any other exterior repair.

For an owner, the encouraging part is that it is bought once. The elevation stays sealed, and the same afternoon’s work reduces cluster flies and rodents through the same openings.

The short version

It got in last October. It is not eating anything, not breeding, and not damaging your house. It cannot bite you. Crushing it does not call in reinforcements, but it does smell, so use a vacuum.

The insect matters — genuinely, and more than most things in this library — but it matters to a berry field rather than to your drywall. Photograph one for WSU, vacuum the rest, and call somebody in August about the gaps.