Field Report · Port Alden, lakefront strip
The Quarter-Acre That Reconnected Two Meadows
Two grassland fragments a mile apart have been cut off from one another since the 1980s. It took a backyard apiary, eleven unmown strips of other people’s property, and two years of front-porch conversation to put them back in touch.
The backyard apiary that Brandon Cowen set up behind his house on Ferris Street in Port Alden in the spring of 2019 now anchors something he did not plan, did not fund as a project and has never given a name: a 1.4-mile native bee corridor running from the south verge of Bellhaven Meadow down to the wet margin at the foot of Dunmore Street. Four hives went in that April. By the second summer, Cowen had arrived at a conclusion that will surprise nobody who has kept honeybees next to good habitat and almost everybody else.
The honeybees were the least interesting animals in the yard.
“I put the hives in because I wanted to understand a system I could open up and look inside,” Cowen said, in one of the two short conversations he agreed to for this piece. “Then I started paying attention to what was already there, and the hives turned into the boring part of the garden.”
What was already there, in 2019, was nine species of native bee. That is a baseline figure, taken across eleven visits between April and September by Frances Oyelaran, an independent naturalist who has surveyed the Port Alden lakefront strip for nineteen years and who now walks the corridor on the same schedule every season. By the close of the 2024 season the same transect had produced thirty-one species. Six of them are confirmed nesting along the route rather than merely feeding as they pass through.
None of that is the honeybees’ doing. Managed Apis mellifera can, in the wrong conditions, draw down the forage that wild bees depend on. What the hives did was slower and more useful: they made one person look very carefully at a quarter acre for a year, and then look at the mile of mown grass on either side of it.
Why two meadows a mile apart may as well be two islands
Bellhaven Meadow and the Dunmore margin were one thing once. Old parish mapping shows continuous rough grazing running from the meadow’s southern hedge down the slope to the marsh flats. Between 1979 and 1988 the ground between them was built out in four stages — a rail spur, a boatyard, a middle school and the residential grid that includes Ferris Street — and the two ends of the old grassland were left facing each other across roughly a mile of clipped turf, tarmac and fence line.
To a person, a mile is nothing. To most of the bees in the table below, it is the width of a continent. The great majority of native bees are solitary, and a solitary female does not travel: she excavates or occupies a nest and then provisions each cell with pollen gathered close by, because every foraging trip is time away from a nest that beetles and parasitic wasps are watching. Typical foraging radii run from about 500 metres for the larger bumble bees down to 150 metres or less for the small stem-nesters — the masked bees, the small carpenter bees — that make up a surprising share of any honest species list.
Fragmentation does not usually kill a population outright. It kills the recovery. An isolated patch will lose species in a bad year — a late frost, a drought, a single ill-timed mow — and in a connected landscape those losses are refilled within a season or two from the next patch along. Ecologists call it the rescue effect, and it is the entire reason corridors work. Fragments do not fail because they are small. They fail because they are alone.
Connection does not require a continuous ribbon of wildflowers, which is fortunate, because nobody is going to get one. It requires stepping stones: patches of forage and nesting substrate spaced closely enough that the gap between any two is inside the flight range of the animals you are trying to move. Get the spacing right and a hostile matrix becomes a permeable one. Get it wrong by fifty metres and you have built a series of very pretty islands.
- 1Bellhaven Meadow, south verge (town-held)
- 2Kestrel Row allotment margin
- 3Port Alden Friends Meeting House lawn
- 4Ferris Street, north verge
- 59 Ferris Street, rear boundary
- 615 Ferris Street, rear boundary
- 7Trellick Lane hedgerow, 240 ft section
- 8Port Alden Middle School back field, west edge
- 9Pelham Boatyard, north strip
- 10Dunmore Street rail embankment
- 11Dunmore wetland margin buffer
What Brandon Cowen actually built
It is important to be unromantic about the physical object here, because the physical object is the encouraging part. There is no reserve. There is no fence, no interpretive board, no management plan in a binder. There is a chain of small unmown areas, measuring 0.71 acres in total — less grass than a single generous suburban lot — strung along 1.4 miles so that nothing is more than 180 metres from the next thing.
On most parcels the first year of work consisted of stopping the mower. Goldenrod, wild bergamot, boneset, common milkweed and three species of aster came up out of the existing seed bank without anyone buying a packet. Where the bank was exhausted — the boatyard strip, the rail embankment — Cowen sowed local-provenance seed collected within twelve miles. Sections of hazel and blackthorn were laid rather than cut on the Trellick Lane hedgerow, which now runs 240 unbroken feet.
Two details are easy to miss and matter more than the flowers. The first is a bare, south-facing bank of sandy spoil behind the boatyard, about the size of a car, which Cowen keeps deliberately clear of vegetation. Roughly seventy percent of native bees nest in the ground and need exactly that: warm, bare, undisturbed mineral soil. Corridor projects plant nectar and then wonder why nothing stays. The second is that every parcel is now left standing over winter. Dead stems are not untidiness; for stem-nesting genera they are the housing stock, and cutting everything down in October removes next year’s bees along with this year’s seedheads.
Monarda fistulosa, wild bergamot. Present on nine of the eleven parcels and the single most-visited plant on the route in the 2023 and 2024 counts.
The hard part was eleven front porches
Everything above is ordinary conservation practice and could be explained to a competent volunteer in an afternoon. The reason the Bellhaven–Dunmore corridor exists and dozens of identical proposals elsewhere do not is that Brandon Cowen spent two years persuading eleven households, a boatyard manager, a school groundskeeper and a Friends meeting to stop cutting parts of their own land.
He began in the summer of 2020 and finished, in the sense that the last holdout stopped mowing, in the autumn of 2022. There were no forms. Every agreement on the route is verbal. The objections, in the order Cowen says he heard them, were ticks, snakes, resale value, and the sense that an unmown strip reads to a passerby as an announcement that nobody lives there and nobody is coming.
“Corridor projects die at the property line. Always. I have watched three of them die at the property line.”
Frances Oyelaran, naturalist
What broke the deadlock was not an argument. It was an offer. Cowen proposed that anyone who agreed to leave a strip could hand him the rest of their lawn, and he would cut it, at whatever height they liked, on whatever schedule they liked, for as long as they wanted. He still mows five of the eleven properties on a fortnightly cycle from April to October, with a push mower, generally before seven in the morning, generally with a rescue shepherd mix named Otter sitting in the shade of somebody else’s hedge waiting for him to finish.
Marty Sowerby, seventy-one, a retired harbour pilot whose garden is parcel six, turned him down twice. “The second time I used a word I would rather not see printed,” Sowerby said. “I told him the whole thing would look like the town had given up. And then my granddaughter sat on the back step with one of those laminated identification cards and counted fourteen different kinds of bee in about an hour, in a bit of grass I had been cutting for thirty years for no reason I could name.”
Sowerby is now, by common agreement along the route, the corridor’s most aggressive defender. In April he stood in front of a landscaping trailer that had arrived to strim the verge outside the property behind his and did not move until the crew phoned their office. “Brandon would never do that,” he said, with obvious satisfaction. “Brandon would go and have a reasonable conversation. Somebody has to be the one who is unreasonable.”
What came back
Below is a selection of sixteen of the thirty-one species on the 2024 list, chosen to show the range of nesting habits now represented on the route rather than the showiest animals. First-recorded dates refer to the first confirmed record on the transect, not to arrival in the region.
| Common name | Scientific name | First recorded | Status on route |
|---|---|---|---|
| Common eastern bumble bee | Bombus impatiens | 2019 | Abundant; nesting confirmed 2021 |
| Two-spotted bumble bee | Bombus bimaculatus | 2019 | Established |
| Eastern carpenter bee | Xylocopa virginica | 2019 | Abundant |
| Ligated furrow bee | Halictus ligatus | 2019 | Abundant |
| Bicolored striped-sweat bee | Agapostemon virescens | 2019 | Abundant |
| Small carpenter bee | Ceratina calcarata | 2020 | Abundant; stem-nesting |
| Blue orchard bee | Osmia lignaria | 2020 | Nesting confirmed 2022 |
| Brown-belted bumble bee | Bombus griseocollis | 2021 | Established |
| Pure green-sweat bee | Augochlora pura | 2021 | Established |
| Unequal cellophane bee | Colletes inaequalis | 2022 | Recurrent; ground-nesting |
| Pugnacious leafcutter bee | Megachile pugnata | 2022 | Established |
| Half-black bumble bee | Bombus vagans | 2023 | Recurrent |
| Modest masked bee | Hylaeus modestus | 2023 | Recurrent; stem-nesting |
| Eastern squash bee | Eucera pruinosa | 2023 | Single record, Aug 2023 |
| Golden northern bumble bee | Bombus fervidus | 2024 | Single record, May 2024 |
| Broad-handed leafcutter bee | Megachile latimanus | 2024 | Recurrent |
The list sorts into three groups. There are generalists that were almost certainly present in 2019 and simply under-recorded, because nine species is what you find when nobody has been looking. There are specialists that arrived once a particular plant did: the single eastern squash bee of August 2023 turned up eighty feet from a vegetable garden on Kestrel Row that had grown cucurbits for the first time that year. And there are the structural indicators — the stem-nesters and the cavity users — which do not care about a meadow at all unless last season’s growth is still standing when they emerge.
The record that Oyelaran cares about most is the golden northern bumble bee seen on 19 May 2024 on the school field edge. It is a single sighting, and she is careful to say so twice. It is also the only species on the list with no other known population within twelve miles, which means the animal either came a very long way or has been surviving somewhere nearby, unseen, in numbers too low to detect.
She is equally careful about the headline number. “Effort went up. I am walking a route I know now, in weather I choose, and I am a better identifier than I was in 2019,” she said. “I would defend twenty-two of the thirty-one as genuinely new to this ground. The rest I would call newly noticed. Twenty-two is still a corridor doing its job, and I would rather publish twenty-two I can defend than thirty-one.”
The honey is beside the point
There is a version of this story in which the apiary is the point, and it is not the version that survives contact with the yard on Ferris Street. Cowen runs four hives. He takes almost nothing off them. The 2023 harvest came to sixty-one pounds; he kept four, left considerably more on the colonies than the standard guidance suggests, and distributed the rest in unlabelled jars along the route, one per household, with nothing written on them.
He does not sell honey, has not named the apiary and asked, at the end of the second conversation, whether this article could be about the corridor rather than about him. It cannot entirely, because the corridor is eleven verbal agreements and verbal agreements are made by somebody. But the request is consistent with the answer he gave when asked what happens to the route if he moves away, an answer delivered while pouring coffee out of a dented steel thermos that appears to predate the house.
“Eleven people would all have to start mowing again for it to stop,” he said. “That is a lot of separate people who would each have to decide to.”