The maternity calendar that decides when an attic can be sealed

The roofline, the attic and the chimney · reviewed August 2026

Why does a calendar decide an attic job at all?

Because the standard method for getting an animal out of a roof cavity relies on the animal walking out under its own power, and for a large part of the year some of the occupants cannot walk.

The method is a one-way device: a fitting over the animal’s own hole that opens outward and does not open back. It is a good technique and it is the humane one, but it is completely indifferent to what is behind the hole. It cannot tell a solitary adult from a female with a nest of blind, hairless young four feet inside the soffit. The device does exactly what it was built to do, the mother leaves at dusk to feed, and the building’s roofline is now closed against her.

What follows inside the cavity is the reason this article exists. The young are not fed. They are too small to travel and too young to see, so they do not follow. They vocalise for a day or two, then stop, and then decompose inside an inaccessible assembly between a plaster ceiling and a layer of blown insulation. The homeowner rang about scratching and now has an odor they cannot locate, moving through the house on the heating, for something between three and eight weeks depending on the weather and the size of the litter. Flies and dermestid beetles arrive to work on the carcasses and appear in the bedroom below. With a raccoon there is a further chapter, because a female who can hear her litter will spend several nights taking a roof apart to reach them, and the second hole she makes is invariably larger than the first.

So the calendar is not a courtesy extended to the animal. It is the difference between a job that finishes in two visits and a job that turns into ceiling demolition, deodorisation and an insect problem, and it is the single most common way this trade converts a manageable situation into an expensive one.

New York states the principle directly rather than leaving it to trade custom. The Department of Environmental Conservation’s guidance on raccoons tells homeowners not to permanently seal entrances without first verifying that all the animals are out of the den, and to look and listen for animal noises, especially in spring. That is a state agency writing down the sequencing rule.

What follows is the specific version of it, species by species, because “spring” is much too coarse to schedule from.

When do gray squirrels have dependent young?

Twice a year, and the two windows between them account for a remarkable proportion of the calendar.

Cornell’s New York wildlife material sets out the biology precisely. Gestation runs about forty-four days. Females in good condition often breed a second time in June, so most produce two litters, and the first and second litters are born in March and August respectively. Litter size ranges from one to nine, averaging about 2.5 in the first litter and 3.2 in the late-summer one.

The number that governs scheduling is not the birth date, though. It is the dependency period. Young gray squirrels do not leave the nest for the first time until they are six to seven weeks old, and they are weaned and capable of foraging with adults at eight to ten weeks.

Run those figures forward from the birth months and the practical picture emerges. A March litter is immobile inside the cavity through most of April and is not foraging independently until some point in May. An August litter is immobile through September and not independent until October. Add the animals that breed slightly early or slightly late, and a house with an established squirrel population has dependent young above the ceiling for a substantial part of the late winter, most of the spring, and again through the early autumn.

This is why “we will do it in the spring” is one of the least useful sentences in wildlife work on Long Island. Spring is a peak dependency period for the most common attic animal in this market, not a clear window.

The compensating fact is that squirrel young are comparatively mobile once they start moving, and by the time they are exiting the nest with the adult they will use a one-way device themselves. So the good windows are real: broadly the stretch after the first litter has weaned and before the second is born, and again once the late-summer litter is foraging, running into the winter. What the calendar demands is that somebody establishes which litter, and how old, before anything is fitted — which is done by listening at the ceiling and by watching the entrance at the right hour, not by consulting a chart.

What about raccoons?

One litter a year, born late, and dependent for far longer than people expect.

The Department of Environmental Conservation records litters of one to seven young born in April and May, with females seeking den sites in late winter. The young open their eyes at about three weeks, nurse for about six weeks, and then leave the den to follow the mother — but they stay with her until September or early October, when they disperse.

That last clause is the one that catches people out. A raccoon litter is not “done” in June. It is mobile in June, which is a different thing, and a mobile juvenile will leave through a device and can then be excluded successfully. But between April and roughly mid-June the animals in an attic are physically incapable of leaving, and any closure fitted in that period produces the worst version of the outcome described above — the largest carcasses, the most determined mother, and the greatest damage during her attempts to re-enter.

Raccoons make this worse than squirrels in three practical ways. They are much heavier, so a nest sitting on a ceiling assembly does structural damage on its own. They establish latrine sites, concentrating waste in one location until it soaks through the ceiling material below. And they are dexterous enough to defeat almost anything short of a proper mechanical closure, which means a hurried repair fitted to buy time until the window opens is likely to fail and to fail expensively.

There is also a genuine urgency exception, and it deserves saying plainly rather than being hidden behind a rule. A raccoon inside living space, or one in a chimney flue with occupants in the building, is a public safety situation that is judged on its own facts by the licensed operator handling it, and it is not resolved by pointing at a calendar. That judgment belongs to the person holding the license. The general case — an animal in an attic, discovered because of noise — is not that situation, and treating it as though it were is how bad decisions get made in April.

What is the bat maternity window, and where does it come from?

It is published by the New York State Department of Health, it runs from late May to mid-August, and it is the clearest statement of any of these constraints.

The Department’s fact sheet on bat habitat inspection and batproofing instructs that buildings are not to be batproofed during the period from late May to mid-August, in order to avoid trapping baby bats inside the building. Its public material on bats and rabies restates the same window — do not batproof between late May and mid-August so that baby bats are not trapped in the roost — and adds two facts that are just as useful for planning. First, that the house bats people actually encounter in buildings here, the big brown and the little brown bat, use those buildings as nurseries from early June through late August. Second, that the best time to batproof is late autumn through winter.

That gives a whole-year shape rather than a prohibition. Late autumn and winter is the window to aim for. Spring, before the colony forms, is workable. Late May to mid-August is closed.

The Department of Environmental Conservation’s material on the northern long-eared bat lands in the same place from the conservation side, advising against activity near roosts until after the pup rearing season in mid-August. The species matters here because federally endangered species that occur in New York carry endangered status under the state’s own Endangered Species Law, and because DEC treats Nassau and Suffolk separately in its guidance rather than folding Long Island into the statewide position. A practice working on this island does not get to reason from the statewide default.

Two further details from the health department’s inspection guidance are worth carrying up the ladder, because they change what an inspection is looking for. Bats are described as using openings larger than half an inch by half an inch, and long thin slots larger than a quarter of an inch by two inches. That second dimension is the one that defeats most surveys. Nobody scanning a roofline for “a hole” finds a two-inch slot at a ridge course or the gap where a fascia board has cupped away from a rafter tail, and a roof assembly can present dozens of candidates of that description.

Which is precisely why bat work is a survey job with a seasonal deadline attached, and why the survey is worth doing during the closed period rather than after it.

How do you tell whether there are young up there?

By combining three things — the species, the date, and what the building is telling you — and by treating the answer as provisional until all three agree.

Start with the sound and the hour. Time of day separates the two commonest attic animals before anyone climbs anything. Gray squirrels are diurnal, so their activity concentrates in the first hours after sunrise and again in the late afternoon. Raccoons are nocturnal and heavy, and the noise they make is thumping and deliberate rather than skittering. Bats are neither: what people hear is a faint rustling or squeaking at dusk, and much more often they see the animals leave rather than hear anything at all.

Then listen for the litter directly. This is the most reliable single indicator available and it costs nothing. Dependent young vocalise, and they vocalise from a fixed point rather than moving around, which is the tell. Squirrel kits produce a high, rapid chittering; young raccoons make a distinctive chirring and mewing that people frequently mistake for birds; a bat maternity roost produces a continuous high squeaking that is loudest in the warm part of the afternoon. The right time to listen is when the adult is out — mid-morning for a raccoon, the middle of the day for a squirrel — because the young are quiet while they are being attended to and noisy afterwards.

Watch the entrance and count. An adult that leaves at dusk and does not return until dawn is behaving like an animal with nothing to come back for. An adult making repeated short trips, in and out several times in an evening, is provisioning something. That pattern in a nocturnal animal during April or May is close to conclusive on its own.

Read the cavity if it can be read safely. Nesting material concentrated and shaped at one point rather than scattered, heavy urine staining in a single area, a latrine site, or a rate of guano accumulation that has climbed sharply over a few weeks all point the same way. On a bat roost the accumulation rate is the useful measurement, because a maternity colony deposits at several times the rate of the few animals that were there in April.

Then apply the calendar as a floor, not a ceiling. If the date falls inside a dependency window for the species identified, the working assumption is that young are present until evidence says otherwise — not the reverse. That is the posture the Department of Environmental Conservation’s own guidance takes when it tells homeowners to look and listen for animal noises before sealing an entrance, and to be particularly careful in spring. Absence of evidence in a roof cavity is very weak evidence, because roof cavities are dark, deep, insulated and largely inaccessible, and the part of one that a flashlight reaches from a hatch is a small fraction of it.

The cost of being wrong is asymmetric, which is what settles the decision. Waiting three weeks on a job where the attic turned out to be empty costs three weeks. Not waiting on a job where it was not costs a ceiling.

Who is allowed to do the removal, and what does that change?

The removal is licensed work in New York, and it belongs to a licensed nuisance wildlife control operator.

The Department of Environmental Conservation issues the license for an individual to take or trap, transport and release wildlife when the animal becomes a nuisance, destroys property or threatens public safety, and only the commercial license holder may charge a fee for it. Several categories sit outside that license and need something further: an endangered or threatened species license from DEC’s Special Licenses Unit; a federal migratory bird permit from the U.S. Fish and Wildlife Service, obtained by the homeowner requesting removal; and a regional depredation permit for a restricted list that includes white-tailed deer, black bear, beaver, otter, mink, muskrat, fisher, marten, bobcat, wild turkey and all native snakes and turtles.

Graduate’s arrangement is stated the same way on every page that touches this work. The removal is carried out by licensed nuisance wildlife partner firms we coordinate. The exclusion — the permanent closure of the building — is ours, and it is the reason to call us.

The reason that division matters to a discussion about timing is that it puts the calendar judgment in the right hands twice over. The licensed operator decides whether and when an animal can lawfully and humanely be moved. We decide, and can start on, everything that does not involve touching an animal. Those are different questions with different constraints, and a firm that answers both by pointing at its own schedule is answering neither.

It also removes an incentive worth being honest about. A business whose entire revenue on a wildlife job comes from the removal has a commercial reason to set a device this week. A business whose work is the sealing does not, because the sealing is worth the same in November as it is in May and is worth considerably more when it is done once rather than twice.

What can actually be done during a closed window?

A great deal, and this is the part of the argument that turns a delay into an advantage.

The mistake is to treat a maternity window as a period in which the file sits closed. What it actually is, is the only period in which you can do envelope work on a building while the animals are still using a known entrance and are therefore not opening new ones. That is a genuinely useful condition and it never exists again on the same job.

Reduce the roofline to one opening. Every secondary gap on the structure gets surveyed and closed while the occupants continue to use their main entrance. Soffit returns on the other three corners. The gable louver at the far end. The lifted ridge cap. The fascia gap over the porch. Done in this order, the animals are not displaced anywhere, because their route is untouched, and at the end of it the building has one hole in it instead of nine.

Design the closure rather than improvise it. Deciding the substrate, the fabrication and the fastening in advance is what makes the final visit a single afternoon. Custom-fabricated 26-gauge galvanised sheet metal is cut and formed to the detail it has to close. Where the substrate is masonry the answer is cement and mortar. Vents that must keep moving air get copper or stainless mesh or hardware cloth from behind, sized to the species, because blocking a functioning vent trades an animal problem for a condensation problem in the roof deck. Xcluder fill fabric and door sweeps close the openings that have to stay serviceable. Everything is mechanically fastened. Foam is not used in any position — not as a filler, not as backing, not as a cover over something better — because a squirrel is through cured foam in minutes and a raccoon is through it in one movement, and the opening is by then recorded as closed on somebody’s paperwork.

Protect the interior and limit the damage. Where an animal is traveling above a ceiling, containment of the traffic route, protection of stored contents and assessment of chewed wiring can all proceed. Damaged wiring in particular should be flagged to an electrician immediately rather than at the end of the job.

Get the roofer in first if one is needed. Rotted fascia, a failed flashing detail or soft decking cannot be sealed against; there is nothing sound to fasten to. Discovering that in the last week of the window is how a job slips a season.

Book the sequence. The removal visit, the watching period, the closure and the return check are separate appointments. Arranging them during the wait is the difference between excluding in the first fortnight of the open window and excluding in the last.

How does the permanent closure differ from a repair?

By where it stops. A repair closes the hole the animal used. An exclusion closes every detail on that building constructed the same way.

This is the distinction that decides whether the calendar problem recurs next year. A house that had a squirrel in one soffit return has three or five other soffit returns framed identically by the same carpenter on the same afternoon in 1974. A house where a raccoon pulled one section of vinyl soffit out of its channel has forty more feet of the same channel. Closing the one opening the animal happened to choose leaves the building in the same condition that produced this year’s job, and the next occupant will be along on the next breeding cycle — at which point the maternity calendar becomes a problem all over again, on the same house, for the same reason.

There is a scent dimension too, and it is more consequential than it sounds. A cavity that has raised a litter goes on broadcasting that fact to every animal of the same species passing along the roofline, and a roof void with an occupancy history behind it is taken up again far more readily than one without. So the removal of fouled and compressed insulation, and the treatment of the affected cavity, is not housekeeping at the end of the job. It is part of the exclusion.

Then verification, which on this work means going back. A newly closed roofline is tested hard by the animals in the surrounding acre, and anything under-specified reveals itself quickly rather than slowly. Checking after the first stretch of weather and the first period of real pressure is how a closure earns the description permanent.

The whole of the roofline material — soffit returns, ridge closures, chimney crowns, gable louvers and the survey that finds them — is indexed in this cluster on the blog, and the service behind it is wildlife management. The species detail sits in the pest library on the eastern gray squirrel, raccoon and big brown bat profiles, and the underlying trade is structural exclusion.

If you are hearing something above a ceiling, the two useful things to note before you call are what time of day you hear it and whether it is skittering or thumping. Those two facts usually settle the species, and the species plus the date usually settles whether this is a job for next week or a job to be prepared for now. The consultation is free and most of it can be done on the phone.

The work behind this piece is Wildlife Management, and everything else in this cluster is indexed on the blog hub.