The County Mows in July and That Is the Whole Story

County Road 12, west side, between the culvert and the section line: on 28 June I counted blooming stems in ten one-metre quadrats and got a mean of 41, dominated by common milkweed, bergamot and two goldenrods. On 19 July a county crew mowed it to four inches, full width, fence to ditch. On 24 July the same ten quadrats gave a mean of 0.

That stretch is 6.4 miles. The county maintains 812.

I have been going to meetings about this for three years and I have stopped leading with the pollinators, because it does not work and because I have come to think it was never the strongest thing I had. The strongest thing I had was an acreage figure, and I did not compute it until year two.

Here it is. I pulled the right-of-way widths off the county's own parcel layer and totalled them against the maintained-mileage record the road department files with the state every year. A 30-foot managed verge along 812 miles is about 2,950 acres. Every conservation easement, township park, state preserve and wetland restoration inside this county comes to a little under 1,900, and I have checked that number twice against two different sources because I did not believe it the first time.

So the county's largest single block of herbaceous habitat is the road network, by a margin of more than a thousand acres, and unlike everything else on the list it is contiguous. The preserves are islands. The right-of-way is the only connected structure in the county and it touches every land cover type we have.

Contiguous is doing some work in that sentence and it deserves a qualification I left too late in the original. The right-of-way is interrupted at every culvert, every driveway and every bridge, and how much of a corridor that leaves for anything that does not fly is a question the literature does not answer for this landscape.

Which means the mowing calendar is the largest habitat management decision made here, and it is not recorded anywhere as a decision. There is no document. I looked. It is not in a management plan, it is not in the comprehensive plan, and the road department's own files hold it as a scheduling practice rather than a policy — a thing the crews have done in the third week of July for as long as the current superintendent has worked there, which is nineteen years. The single most consequential piece of land management in this county exists in the record as a habit.

That is the part that connects it, for me, to everything else I spend my time on. It is the same failure as an uncorrected datum: the data required to see the thing exists, in more than one office, and no one has ever joined it, because joining it is nobody's job.

Now the part I got wrong for two years, and I got it wrong in the way that people who show up with a slide of a monarch reliably get it wrong. I assumed the July date was inertia and that the fix was to explain the ecology to somebody. It is not inertia. I sat down with the superintendent for two hours and he walked me through a calendar that is downstream of three separate constraints, all of them real.

The first is sightlines at intersections and curves, which is a safety matter with crash data behind it. I asked for the analysis and he sent it, and it is not a pretext. I would not touch it.

The second is the drainage ditches, which stop draining within about three seasons if woody growth establishes, and a blocked ditch is a washed road, which costs more than everything else in this essay put together.

The third is labour. He has a seasonal crew in July. In September the same people are doing something else, and the something else is also mandatory.

All three are legitimate and none of them is an argument for what actually happens, which is 812 miles taken to four inches, full width, at peak bloom. Sightline mowing is a triangle at each intersection — I measured a sample and it is on the order of two percent of the mileage. Ditch maintenance is the ditch and the near slope, not the back slope where the milkweed is. What produces the full-width July pass is that one crew making one trip is the cheapest way to satisfy all three constraints at once, and nobody had ever priced the alternative because in nineteen years nobody had ever asked for it in a form he could act on.

So I asked. Two passes — a July sightline-and-ditch pass, a full-width pass after 15 September — came back at about eleven percent more in fuel and hours, an estimate he produced in a day and a half from his own records.

Eleven percent, on a line item I can find in the published county budget. That is the number the whole argument turns on. I spent two years on milkweed.

The limit of the claim, stated plainly, because I have sat through presentations by people on my side of this who overreached and lost the room. I do not know that a September mow produces more pollinators. The right-of-way literature is thin, mostly from three states, and the effect sizes vary by more than I am comfortable citing. What I know is the seed-set timing, which is not in dispute, and that mowing a plant before it sets seed prevents it setting seed. Everything past that is inference and I say so at the meetings, which costs me some persuasive force and is the only version I am willing to give.

It goes to the commission in November. I have no prediction. What I keep returning to is that the superintendent had the eleven percent the entire time, in a filing cabinet, and that the reason nobody had it was that the habitat question and the roads budget have never once been asked in the same room — the same missing join, in a different building, with a different set of people who are all doing their jobs correctly.

The survey stake at the culvert end of my transect has been there since 2023. On 12 August this year I counted 4 blooming stems inside the quadrat next to it, all of them bergamot, all of them regrowth, none of them going to set anything before frost.

It went to the commission in November and it did not pass. The vote was three to two and the reason given by one of the three was that the second pass would need a budget amendment mid-year, which is true and which I had not costed against the amendment process. The superintendent has put it in next year's request. I am told that is the normal route and that I should have started there.