Tiny brown dots cemented to the glass, usually blamed on tar or insects. They were fired from your mulch bed by a fungus aiming at the brightest thing it saw.
Small brown dots that won't wash off, on the north side of the house, near a mulch bed. Here's the whole thing:
And the reason it's your house and not your neighbour's is usually that you re-mulched in March, and they didn't.
They look like somebody flicked a paintbrush loaded with tar at the house.
Small dots — a millimetre or two, brown going on black, slightly domed. They're on the siding, on the soffit, on the downspout, on the leaves of the hostas, and on the windows. They will not wash off. They do not respond to the thing that took the bird mess off. And they are almost always, in the first instance, blamed on something that didn't do it: tar spatter, insect droppings, scale insects, or a passing truck.
None of those. What you're looking at was fired at your house, deliberately, by a fungus living in your mulch bed — and the reason so much of it is on the glass is that the glass is what it was aiming at.
The organism is Sphaerobolus stellatus. The genus name is Greek for sphere thrower, which is the single most honest bit of nomenclature in this article.
It's a wood-decay fungus — a Basidiomycete, the same broad family as the mushrooms you eat, and a close relative of the bird's nest fungi and the earth stars. It lives on well-rotted wood: bark mulch, wood chips, the soft decomposed layer underneath a bed that's been topped up every year for a decade. It is not a disease. It is doing the job fungi do, which is turning dead wood back into dirt, and it is extremely common across the United States and especially in the East.
The problem is entirely in how it reproduces.
The fruiting body is tiny — about 2mm — and you will almost certainly never see one, which is why the dots seem to arrive from nowhere. It opens into a small cup. In the bottom of that cup, sitting in liquid, is a mass of spores about 1mm across called the gleba. Underneath the gleba is a layer of cells that takes on water and swells, and for roughly five hours after the cup opens, pressure builds.
Then the layer turns inside out, all at once, and throws the spore mass up to six metres.¹
Michigan State's diagnostic sheet estimates the discharge at about one ten-thousandth of a horsepower. That number is worth sitting with for a second. It's a rounding error of an engine — and it is enough to put a milligram of spores twenty feet in the air, because the thing being thrown weighs nearly nothing and the launch is nearly instantaneous. Pound for pound it is one of the more violent events in biology, performed by something with no muscles, out of a cup the size of a pinhead, in your flowerbed, several thousand times a spring.
The gleba is coated in an adhesive. Whatever it hits, it sticks to, and then it dries — and dried is where your problem starts.
Here is the part that every fact sheet about this fungus skips, and it's the part that matters if you clean windows.
The cup aims. It is strongly phototropic: before firing, it orients toward the brightest light it can detect. Outdoors that's the sun — or, when the sun isn't directly available, whatever is reflecting it hardest. Michigan State names the alternatives outright: the sun, or highly reflective surfaces such as glass and light-coloured walls. The well-worn field observation that a white car parked next to a dark one collects nearly all the spots is the same phenomenon, and it's the reason this fungus has a reputation among people who park under trees.
Now picture your house from ground level, from inside a mulch bed, through the eyes of something that can only detect brightness.
The siding is matte. It's mid-toned. It scatters.
The windows are mirrors.
**So the glass isn't collateral damage. The glass is the target.**² This is why the spots cluster on and around windows and why they show up on the north elevation first — on a north wall there's no direct sun to aim at, so the brightest object in the fungus's entire field of view may well be your own window, and the north side is also the side whose mulch never dries out enough to stop fruiting in the first place.
Every extension sheet, every landscaping blog, every contractor page about artillery fungus is written about siding. The genuinely useful fact — that the organism is preferentially targeting the one surface on the building that is hardest and most cleanable — appears essentially nowhere, because the people writing about it don't clean glass.
Before you take a blade to anything, confirm it. Four things get called artillery fungus and two of them wash off with water, so the diagnosis is worth ninety seconds.
Check the height gradient. This is the one that settles it. The fungus fires from ground level, and six metres is its ceiling. So the density falls off as you go up, and somewhere around eighteen or twenty feet it stops dead — a hard upper limit, because nothing is throwing anything higher than that. Ground-floor windows are peppered, first-floor windows have a scattering, and the second storey is clean. Nothing else on this list produces that pattern. Road tar doesn't thin with height. Insects don't respect a twenty-foot ceiling. If you have a clean band above and a dense band below, you're done — that's ballistics, and it has one source.
Then check the direction. Trace it back. Artillery fungus needs an organic bed within firing range, so the spots should get denser as you approach a mulched area and disappear on elevations that don't have one. A wall with no bed near it and spots all over it is telling you to look somewhere else.
Then scrape one open. This is the confirmation. The shell is brown, darkening toward black as it ages — but the inside is off-white, finely granular and gummy. That interior is diagnostic and nothing else in the lineup has it.
Now the impostors:
One more confirmation if you want it and the season is right: go look at the bed on a damp spring morning. The fruiting bodies are only about 2mm, but on the top layer of an old, rotted, north-facing mulch bed in April, a close look will often find the cups — tiny pale rings, open, aimed at your house.
The received wisdom on artillery fungus is uniformly grim, and on siding it's correct. On painted and vinyl surfaces the spots are effectively permanent: the adhesive grips harder than the paint does, so anything aggressive enough to shift the spot takes the finish with it. Pressure washing at a setting that works also strips the surface. There's a reason this fungus is discussed in the same tone as structural damage.
Glass is a different material and it deserves a different verdict.
The spore mass is organic. It's dried plant-derived gunk with a granular, gummy centre — scrape one open and you'll find it off-white inside under the darkened shell. Glass is a hard, chemically inert, non-porous surface, harder than anything in that dot. The adhesive is stuck to the glass; it hasn't gone into it, because there's nowhere to go.
Which means the tool that fails on siding works here: a new razor blade, held at a low angle, on wet glass. Same technique as paint spatter, same discipline, and the same hard rules apply without exception:
Soak the dots first if they've been there a while — plain water, ten minutes, a wet cloth draped over them. Fresh spots sometimes surrender to a scrub brush and soap alone. Old ones need the blade, and old ones will often leave a faint shadow where the adhesive has held a microscopic film — that usually polishes out.
On the siding, honestly: manage expectations. On the glass, it's a Saturday.
If you do one thing from this article, do this one, because it's the difference between solving the problem and renewing it.
**The spore masses stay viable for more than ten years.**³
They cannot grow on glass, siding, concrete or any other inert surface — there's nothing there to eat. But they are not dead. They are dormant projectiles waiting for an organic substrate, and if they fall into mulch, they germinate and reinfest it.
So consider what happens when you scrape a wall clean and let the debris fall where it will. You have taken several hundred viable propagules off an inert surface where they could do nothing forever, and planted them, by hand, in the exact bed the problem came from. You haven't cleaned anything. You've re-seeded it, with a decade of shelf life, and in two springs you'll be back.
Lay a tarp. Under the window, under the working area, wide enough to catch drift. Collect everything — what comes off the glass, what comes off the siding, what you rinse down — and dispose of it well away from any bed. On a route, this is the line between fixing a customer's problem and booking yourself a callback to the same address in eighteen months to do the same work again for free.
Worth saying plainly, because this one has a trap in it.
You arrive, the north elevation is peppered, and the customer's expectation is that a window cleaner washes windows and these are on the window. But the fix here is a blade, and a blade is the one tool on your belt that can convert a customer's fungus into a customer's scratched pane — on tempered glass, through no fault of your technique at all. That risk doesn't disappear because the spots are ugly and the customer is watching.
So price it and scope it separately. This is not washing; it's spot removal, it's slow, it is charged by the hour or by the elevation, and it comes with the same coated-and-tempered conversation you'd have before razoring anything else. Get the answer before the blade comes out. If the pane is coated, the honest answer is that the spots stay — and saying so on the doorstep costs a great deal less than saying it afterwards.
And the part that earns you the next job: tell them about the mulch. You can clean that glass beautifully and be back in eighteen months to identical spots, because the bed is still there and still loading. A cleaner who explains the source — and mentions that gravel within a metre of the foundation ends it, or that mushroom compost blended in suppresses it — has just told a customer something their landscaper didn't. That's worth more than the invoice.
It also reframes the callback before it happens. When the spots come back — and over a live bed they will — you're the person who predicted it, rather than the person whose work didn't hold.
There is no registered fungicide for artillery fungus, and there isn't going to be one.⁴ The reason is structural: the organism lives in the bed, not on the wall, and the thing stuck to your window is not alive in any way a spray could reach — it's spent ordnance. Spraying the spots accomplishes nothing. Spraying the bed means broadcasting fungicide across your soil to suppress an organism whose entire crime is decomposing wood.
What actually works:
Blend mushroom compost into the mulch. This is Penn State's finding, out of their Department of Plant Pathology and Environmental Microbiology, and it's the best non-chemical control available — roughly 40% is the figure most commonly cited, with other work reporting substantial reduction at around a third. The mechanism appears to be competition: spent mushroom compost carries a microbial community that's antagonistic to Sphaerobolus and crowds it out.
Or stop feeding it. Gravel, stone and plastic don't rot, so nothing fruits in them. Against the foundation — specifically the strip within firing range of the windows — inorganic mulch ends the conversation permanently. You don't have to re-landscape the property. You have to change about a metre of it.
And watch the calendar. This fungus runs against the grain of every other mulch organism: it likes cool and wet, so it peaks in spring (April–May) and again in autumn (September–October), while everything else in the bed waits for summer heat. Two seasons a year, both of them mild and damp, both of them exactly when a Northeast route is busiest.
Which is the whole thing in one sentence: the mulch you laid down in March is what your windows look like in October.
Derek is an editorial team contributor covering the Northeast corridor, with a focus on pre-1945 glazing and pollen-season route scheduling. Articles bylined by Derek are researched and reviewed in collaboration with the Giordano Inc. editorial team. Artillery fungus is a Northeast problem with two seasons a year, which makes it a scheduling problem before it is a cleaning one.