Ant Control in Roslyn

Why does a Roslyn house get wet from above and below at once?

Because the village is built on the side of a ravine with a working impoundment at the bottom of it, and those are two different water problems arriving at the same building.

Take them separately, because they produce nests in different places.

From above. The village covers six-tenths of a square mile and falls from roughly 253 feet near Piper Court to sea level at the harbor. That is a steep catchment, and on a steep catchment rain does not soak in where it lands. It runs, it concentrates in the routes the ground and the paving give it, and it arrives at whichever elevation of a building happens to be in its way. Add a century and a half of terracing, driveway aprons, retaining walls and rebuilt gardens, each of which redirected a little of it, and what reaches a house is a channelled flow rather than an even wetting. The consequence upstairs is not obvious at first: water arriving with momentum at the base of a wall splashes, wets the lower courses, saturates the ground beside the footing, and keeps the whole uphill elevation at an elevated moisture content from March to November.

From below. The five-acre mill pond below the grist mill has been impounded since the village was a milling settlement, and it does not draw down in a dry summer the way a seasonal wet patch does. A standing pond sets the height of the water in the ground around it and keeps it there through the year. Timber bearing on rubble or early masonry in that ground wicks moisture out of the wall continuously and never gets a dry spell long enough to arrest decay in it. The grist mill itself is the local proof: a Dutch colonial frame building of 1715 to 1741 whose foundation needed rebuilding in 2024, on ground that has in the past dropped its ground floor well below street level.

Now put a carpenter ant into that. The insect does not eat wood; it excavates galleries in it and carries the spoil out, and it chooses timber that decay has already softened because cutting sound wood is expensive work. The Northeastern IPM Center puts the requirement directly: rotted timber costs a colony far less effort to cut, and the water still held in it keeps the brood from drying out. Penn State Extension gives the threshold — sound wood below fifteen per cent moisture content is not infested.

So a village that supplies wet timber at two separate heights supplies nest sites at two separate heights. On the same house, in the same year, we routinely find one gallery at the sill line where the masonry keeps the plate damp, and another six or ten feet up on the uphill elevation where splashed and channelled runoff has been working on a corner board and a window sill for a decade. Treating either alone leaves the other.

How a colony is organized and what it is actually after is on the carpenter ant profile, with the treatment logic under ant control. The ravine and the district are what this page is for.

What does clapboard do that other cladding does not?

It puts a horizontal joint in the wall every course, and behind it, on this stock, there is very little.

Almost everything in the district is timber frame under clapboard, and the span is enormous: the district’s 1986 nomination lists 111 contributing resources and a period of significance opening around 1680 and closing in 1936. That is three centuries of essentially the same wall assembly, built by different hands with different timber.

Three properties of that construction matter to an insect.

Every course has an exposed lower edge and a joint at its head. A clapboard wall works by shedding — each board laps the one below and water runs off the face. It works well while the boards are flat, the butt joints are tight and the paint film is intact. It stops working at the butt joint where two boards meet in a run, at the end grain where a board dies into a corner board or a casing, and anywhere the board has cupped enough to hold water on the upper face instead of shedding it. Those are small failures and there are hundreds of them per elevation.

Behind the boards there is often nothing. No weather-resistive layer as anyone would understand it now, frequently no insulation, and on the earliest houses no sheathing at all — the clapboard goes straight onto the frame. Water that gets past the cladding therefore lands on structural timber directly, with nothing sacrificial in between.

The wall cavity is a shaft. In frame construction of this age, and particularly in the balloon-framed work of the later nineteenth century, the space between two studs is open for the full height of the wall. That has two consequences. Ants entering at a damp sill can be reported at a second-floor ceiling with nothing to show anywhere in between, so the room where a trail appears says very little about the nest. And an injection into a wall void can be both entirely correct and entirely useless, because the void carries on a great deal further than the person holding the equipment assumes.

Then the roof. Nineteenth-century rooflines in this village carry a great deal of edge per square foot: mansards, dormer cheeks, gable returns, valleys where a rear wing meets a main plane. There is Second Empire work of the 1870s on East Broadway carrying a mansard, and a mansard is effectively a wall clad as a roof: timber-framed, steeply set, and largely unventilated behind the covering. Each of those meetings collects leaf litter, holds water for longer than the open field of the roof does, and sits on a flashing joint made by hand a very long time ago. Behind a good many of them is a boxed return, which is a void with a board across it. On a house that is both steeply pitched and heavily detailed, the high-level nest sites are real and they are the hardest to reach.

Where is the parent colony when the lot is a fraction of an acre?

Off the property, more often than not, and that changes what the job can promise.

Carpenter ant colonies operate as a parent nest plus satellites. The parent needs genuinely wet material because eggs and the youngest larvae will not survive drying out; satellites tolerate much drier conditions and are what turn up in a window sill, a porch post, a soffit return or a length of insulation. Foraging carries a long way from the nest, and the Northeastern IPM Center puts the range at as far as three hundred feet.

On a village lot in a district of six-tenths of a square mile, three hundred feet is not a yard. It is the pond bank, Roslyn Park, the street planting, a neighbor’s stump, the timber edging in somebody else’s terrace, and the wooded slope above the road. So the honest structure of a job here is often: we can find and treat what is in your building, we can dry the timber that made it habitable, and we cannot remove a parent colony sitting in a mature tree on land nobody involved controls.

That is not a counsel of despair. It changes the emphasis, and it changes it toward the building. If the property is going to face continuing pressure regardless, then the durable work is making the structure unattractive and hard to enter rather than eliminating a colony we cannot reach. That means dry framing first, then closed entry points at the sill, the casings, the utility penetrations and wherever planting touches the cladding.

Two things follow practically. First, the survey has to leave the house. A flashlight walk after dark along the wall, up the porch posts, across any timber steps or retaining structure, and around the base of every mature tree within reach shows trails that are simply not visible at midday — these insects forage between dusk and midnight. Second, the timber in the yard is on the list. Terracing on a ravine hillside gets built in timber more often than in masonry, and a timber wall that has been retaining a bank since the 1990s is soft where it meets the earth, whatever it was treated with when it went in. So are the sleepers around a bed, a fence post at grade, an old set of steps and a woodpile that has spent two winters on bare earth.

Which repair needs the board’s permission, and which does not?

Most of the repairs that actually end an ant problem do not, and that is a useful thing to know before anything is priced.

The village runs a Historic District Board that issues certificates of appropriateness. Any construction, alteration, change, move or demolition within the Historic/Scenic Overlay District needs approval before a permit can issue, and the criteria the board applies take in the materials and the color of the work, the shape of the roof, and what any altered opening does to the elevation. The exemption is the part that matters here: ordinary maintenance or repair, with like materials of similar quality and color, does not require approval.

Read that against the list of things a carpenter ant job on this stock actually requires.

Cutting out a rotted run of clapboard and replacing it with clapboard of the same profile and exposure is maintenance. Splicing a new section into a decayed sill in the same dimension is maintenance. Rebuilding a porch post or a beam end in like material is maintenance. Rebedding a failed flashing, repointing an open mortar joint in matching mortar, refitting a dropped soffit board, clearing and repairing a gutter — all maintenance. Which is convenient, because those are precisely the items that decide whether the colony has somewhere to come back to.

What does need a certificate is anything that changes what the building looks like: a new vent where none existed, a visible metal closure on a street elevation, a different cladding material, an altered opening, a changed roof profile. Some of that is occasionally the right answer — a foundation vent to dry a chronically damp crawl, for instance — and where it is, it goes through the board rather than around it.

The point is not that preservation review is an obstacle. It is that on a historic house the correct durable repair happens to be the one that keeps the building looking like itself, and the shortcuts that would trigger review are mostly shortcuts we would not want anyway. Establishing which category a proposed repair falls into is ten minutes of work and it saves weeks. It is also worth checking which Roslyn a property is actually in. The incorporated village is small, and Roslyn Heights, Roslyn Harbor and Roslyn Estates are different jurisdictions with different boards — as is a good deal of surrounding unincorporated ground that carries the postal name.

Where the closing side of the work is substantial — the sill line, the penetrations, the soffit returns — it is scoped as structural exclusion in Roslyn, which runs the same like-materials logic through the whole envelope.

What about the apartments above Main Street?

A different insect, a different route, and a cellar at the lowest point of the valley.

The village is not only frame houses. Only 44.5 per cent of its roughly 1,423 units are detached, with 238 units in buildings of fifty or more and a further 165 in three- and four-unit buildings, and there are apartments above the trade premises on Main Street and Old Northern Boulevard.

Carpenter ants do turn up in those buildings, and when they do it is generally the roof edge or the joist ends pocketed into a masonry wall rather than anything at eye level. But the more common complaint is a trail of small ants, and the mechanism is vertical rather than structural. Put a shop at street level, a cellar underneath it and two flats over the top, and the three are tied together by a stair, a flue and a run of pipework passing through floors that nobody ever closed behind them. At the valley floor that cellar is damp because everything down there is damp. The colony sits in the damp part of the route and forages upward toward the food.

Odorous house ants are the usual identification. Their nests are shallow and they shift them easily, so a trail can turn up within a day of a storm — and on this relief a storm floods a shallow nest fast, because the water arrives from uphill rather than falling on it. Crushed, they give off a faintly rotten smell. The odorous house ant profile carries the identification, and the operational rule is not to spray the trail: a colony with many queens answers pressure by splitting, and the fragments set up further inside the building than the original ever was.

Pavement ants come up through joints in a cellar slab, at the edge of a sidewalk or between flags in a service yard, leaving small craters of excavated grit at the joint. See the pavement ant profile.

Pharaoh ants deserve naming for the larger heated buildings, because they live permanently indoors, they bud aggressively when treated with anything repellent, and getting the first move wrong with them does lasting damage. The pharaoh ant profile explains it.

Where a food premises is involved the trail is a compliance matter as well as a nuisance, and that side runs through restaurant pest control and the commercial program. Where the building is managed, property management pest control sets out the structure.

How do you tell an ant from a termite here, and what happens then?

By the waist, the wings and the contents of the workings — and then by calling the right firm.

The ant carries a sharp constriction between thorax and abdomen, an elbow in each antenna, and a leading wing longer than the one behind it. The termite carries none of the three: an unbroken outline from head to tail, antennae like a short string of beads with no bend in them, and four identical wings that come away and drift into a window track or a spider’s web. If all that survives is wings, that answers it on its own.

The building evidence is even cleaner. Carpenter ants push their spoil out and keep the gallery empty, so the workings are smooth-walled and clean and there is a pile of stringy pale material below the opening. Subterranean termites bring soil in, in shelter tubes on masonry and in the workings themselves. Spoil on the floor means ants. Earth in the tunnel means termites. The full comparison, with the swarm calendars and what each insect leaves inside the timber, is on the eastern subterranean termite page.

The reason to be careful in this village specifically is that the owners of very old houses expect the worse answer. Three centuries of grade built up against rubble masonry beside a permanent impoundment is a genuinely favorable set of conditions for a subterranean colony, and quarter-inch shed wings at a sash in April are a real sign. But the same conditions produce softened frame timber in quantity, and on clapboard-sheathed houses with damp sills the more likely answer is the ant.

Where it is the ant, that is our work and everything above applies. Where it is genuinely the termite, the position is short: Graduate does not treat termites. There is no partial scope, no inspection service and no version where we take a piece of it. That is a specialist trade and you should hire into it, taking your evidence with you — which wall, what height, whether anything was alive when you opened a section of tube, and when the wings appeared.

What does the work look like on a house in the district?

Diagnosis, then water, then wood, then treatment — with the treatment last because it is the part that does the least on its own.

Read the site before the building. Where does the runoff come from, and what has been built across its path? Follow every leader to its actual termination. Find the terrace, apron or wall that interrupted the original fall. Then look at the corner nearest each of those for staining, soft trim and spoil.

Then find the spoil. It falls straight down out of a slit, so a drift of coarse pale material — stringy rather than powdery, with insect fragments through it — on a cellar floor, a windowsill, a baseboard top or a porch beam is aimed straight up at the gallery that produced it. Sweep it and date it. If it returns within the week the gallery above is being worked now.

Then sound and measure. Run a tool handle along a sill, a beam or a casing and listen for the note to fall away where the timber behind has gone; an awl then finds how far the soft zone runs, and a meter over the same member narrows the search faster than looking ever does. Painted trim on a house like this can be immaculate on the face and saturated behind.

Then correct the water. Extend the leaders somewhere with fall. Clear and prove the gutters with a hose rather than a glance. Re-flash the porch and the roof-to-wall junctions. Take the soil and the planting back off the sill so air reaches the bottom of the wall and somebody can inspect it. Where the runoff genuinely has to be intercepted uphill, do that.

Then replace what has lost section. In like materials, for the reasons above. Killing a colony does not put wood back into a member. Once section has gone, the site stays exactly as attractive as it was — to this species, or to the fungi that softened it in the first place.

Then treat, where the nest is. Bait placed on active trails and carried back, or material put straight into a gallery once its position is actually known. A spray along a visible line kills the workers on that line, which is the smallest and most replaceable part of a colony.

Then check the following spring. These colonies work slowly, and a quiet July says a great deal less than a quiet April does.

When in the year does the ravine give up its information?

Each part of the job has a season, and they are not the same season.

March and April. Activity resumes. Ants indoors while there is still nothing happening outside are living inside the heated envelope, which is the single most useful sighting available and is available only now. This is also when winged reproductives appear and get misidentified.

May and June. The best weeks for reading the water, before the planting closes in and hides the base of every wall in the village. Anything that failed over the winter on the uphill elevation is visible now.

June to August. Foraging is at its peak and spoil is accumulating fastest, which makes these the weeks to hunt a nest in. They are also the only honest test of a wet member: a sill that a meter still calls damp in a dry August week is being fed, and near the pond it usually is. Night trail-following is at its most productive.

September and October. The repair window. Gutters, flashing, clapboard runs, sill splices, porch work and drainage correction all go in while the ground and the roof are still workable, and the season’s own rain then checks the result.

November to February. Nothing is moving, so a silent house between December and February is not evidence of anything. What winter is good for is diagnosis. On a cold, still morning the warm air escaping an uninsulated wall marks the same open joints that let water in during a storm, and on this stock those joints are the whole of the problem.

For the rest of what this village generates across every service, see pest control in Roslyn; the rodent version of the same sill-line argument is at rodent control in Roslyn. At the other end of the same harbor, Glen Cove runs a very different housing mix; Sea Cliff sits on the bluff between them with an older stock again; and west of here Manhasset and Port Washington put the same insect into different construction. Everywhere else we work is listed on the Long Island hub.

If you have found stringy spoil on a sill or a cellar floor, leave it where it is and tell us which wall it is under.

Related

• Ant Control — how we approach it
• All pest control in Roslyn
• Rodent Control in Roslyn
• Structural Exclusion in Roslyn