Francis Quang-Hai Dinh was an undergraduate at McGill. In his free time during his first summer there, he cycled to almost 200 parks around Montreal and recorded what he saw.
Eastern grey squirrels, despite the name, sometimes have black coats, a melanism produced by an incompletely dominant MC1R mutation. Dinh observed over 800 animals, some grey and some black. The resulting analysis, covering 1,734 observations across the metropolitan area, found that the proportion of black squirrels varied sharply from one part of the city to another, and that the variation tracked colder winter temperatures and denser forest cover rather than urbanisation.
The surprising part is the direction. In many North American cities, black squirrels are more common in the most built-up areas. In Montreal they were generally less common there. Major rivers and suburban sprawl appear to shape the pattern alongside local microhabitat.
Allison Roth, the study’s senior author, framed the implication precisely: people often think of cities as a single type of environment, but the conditions animals experience can vary dramatically from one neighbourhood to another. To understand adaptation in urban environments, you have to look past simple comparisons between urban and rural.
The researchers are careful about what this does and does not establish. The study is observational, so the associations between coat colour, temperature and vegetation cannot demonstrate cause. They also lacked direct evidence that black and grey squirrels survive or reproduce differently across these environments.
Take the methodological point out of biology and it becomes one of the most expensive errors in small business.
Market data arrives at the level of the city. Average spend in Manchester. Footfall in Cape Town. Coffee consumption per head in a given metro area. All of it is real and all of it is an average across environments that have almost nothing in common with each other. The street with the hospital on it and the street with the university on it are both in the same dataset and will never behave the same way on a Wednesday.
The practical scale for a single site is not the city. It is the catchment you can actually walk in fifteen minutes, and the thing that determines what happens in it is rarely the thing a report would name. A bus stop being moved. Which side of the road gets the morning sun. A river, or a dual carriageway, or a railway line functioning exactly as the St Lawrence does in this study, as a barrier that makes two adjacent populations behave as though they are in different towns.
The way you find that out is not analysis. It is Dinh’s method. You go and look, repeatedly, across the whole area, and write down what you actually see rather than what the category predicts.
He did it on a bicycle, in his own time, as an undergraduate.
The resulting paper is a study of a city nobody had looked at closely enough.
