Skip to main content
Macmillan Childrens Publishing Group

Darwin Comes to Town

How the Urban Jungle Drives Evolution

Menno Schilthuizen





Thousands of animal species have evolved to cohabit with ants in their “cities.” Here, a Lomechusa rove beetle is being cuddled by its ant host.

Some 20 miles west of the city of Rotterdam lie the coastal sand dunes of Voorne—an extensive area (at least, by Dutch diminutive standards) of rolling, vegetated dunes, though increasingly consumed from the north by Rotterdam’s expanding port. You can sit there, with your buttocks on a carpet of mosses and lichens, eating a sandwich among the rare yellow-wort and marsh helleborines, while in the distance gigantic heaps of iron ore and coal are shifted around, the cling and clang wafting in and out on the incessant wind.

It is here that I spent almost every Saturday as a schoolboy, hunting beetles for my expanding collection. My juvenile-naturalist friends and I, sometimes accompanied by our indefatigable biology teacher, would cycle along the Meuse, take the ferry across the river, zigzag among the oil-storage tanks and daunting chemical installations of the refineries, and then spend a whole day in the dunes, botanizing and entomologizing. Sundays would then be devoted to sorting, pinning, and identifying the booty, and conscientiously penciling everything into notebooks, an oasis of bliss before the dreary school week began again on Monday morning.

There are about 4,000 species of beetle in the Netherlands, and I had set myself the task of finding as many of those as possible in Voorne. After two or three years, in the racks of mothballed insect drawers in my room, I had amassed more than 800 different species, some never found before in the country.

The first few hundred of those species were easy: common widespread ones that I would simply bag as they ambled across the path or sat perched on the tip of a leaf. But as my list of catches grew, more advanced collecting techniques were called for to add the more elusive species from so-called “special habitats.” Such as myrmecophiles—animals whose place in nature is inside ant nests. My entomology handbook told me that the best time to find these was in the middle of winter, when all the inhabitants of an ant nest would be huddled up in the deeper reaches, and—more importantly—would be too hypothermic to be bothered to bite me.

So, one frosty winter morning, I tied a large spade to the frame of my bicycle and headed for one of the stands of pine in the inner dunes where I knew there were large, dome-shaped nests of the red wood ant, Formica rufa. The mounds were still there, covered with the dried-out stems of stinging nettles that had sprouted on top of the ammonium-rich sites. I plunged the spade deep into the ant mound. Heaving up spadefuls of pine needles mixed with ice crystals, I finally reached the frost-free depths where the ants were hiding. I took out my seasoned beetle sieve, a clever time-honored contraption of German design consisting of a strong cloth bag with a sieve and a funnel, and passed handfuls of the nest material into it, shaking vigorously to separate the insects from the larger debris, and finally placing the flow-through into a large white plastic sorting tray. Then, I sat down and waited.

Before long, the undercooled ants slowly began unfolding and stretching their legs and unsteadily started walking around on their plastic floor. But they were of no interest to me. What I was after was what I spotted scattered in between the ants. Here, a small brown clown beetle, with its legs held tight against its round glossy body, looking for all the world like a seed. There a ditto rove beetle, its abdomen curled up in alarm. These were the ones I was after! Myrmecophilous beetles, never seen outside of ant nests. I put the beetles in my killing jar (an old jam jar with tissue paper and a few drops of ether), took them home, and carefully pinned them, adding to the pin a card with a specimen of the ant glued onto it (as recommended in my authoritative beetle book). Then I took out my identification keys to confirm that I had indeed found a whole series of beetle species that I would never have seen had I not taken the trouble to dig up an ant nest in the middle of winter.

In their hefty, definitive volume The Ants, esteemed ant-specialists Bert Hölldobler and Edward O. Wilson devote an entire chapter to the animals that shack up with ants. They provide a “summary” table that goes on for fourteen pages and covers not just beetles, but also mites, flies, butterfly-caterpillars and spiders. Woodlice, pseudoscorpions, millipedes, springtails, bugs, and crickets … In almost any group of creepy-crawlies, there are species that have crept and crawled their way into the ant society and found tricks to eke out a living there.

Those tricks are of two kinds. The first is to blend in. Ants live in a largely chemical world. Communication within an ant society happens with a whole bouquet of scents and smells, with which ants pass messages to one another that are the pheromonal equivalents of a simple “Howdy,” a comforting, “Fine, fine, everything is hunky-dory,” an excited, “Ooo, nice food two leagues west of nest,” or a frantic, “SAVE YOURSELVES!!! SOME BASTARD IS STICKING A BLOODY SPADE INTO THE NEST!!!”

The ants’ chemical language also functions as a social immune system: it distinguishes “self” from “foreign.” Any creature that does not smell like a fellow colony member is mercilessly attacked. So, to invade a nest, myrmecophiles (even those that do not mean the ants any harm) have needed to break the ant’s identification code. They have evolved to speak “ant” to avoid detection. Many myrmecophiles have special glands on their bodies that produce their host’s signal molecules (especially “appeasement” signals), which are wafted into the air via tufts of hair. Some myrmecophiles, such as the rove beetle Lomechusa, are even bilingual: in winter, Lomechusa lives in a nest of the red stinging ant Myrmica and chemically chats along with them happily. But in spring, it leaves Myrmica and takes up summer residence in a red wood ant nest and somehow, seamlessly switches its chemical vocabulary to Formica.

The second trick that myrmecophiles have evolved to maintain themselves in ant society is to find a niche where they can be happy and safe. The ants’ obsessive-compulsiveness helps this. Whenever we accidentally snatch a peep into one when lifting a rock in the garden, the inside of an ant nest may seem a chaos of criss-crossing ants and randomly strewn brood. However, it is actually a highly structured society with dedicated areas for the different services that make the society tick—not unlike a medieval city. There are refuse areas where the colony’s waste is dumped; peripheral nest chambers and guard nests where the nest’s defensive troops reside; storage chambers for keeping supplies; brood chambers with separate sections for pupae, larvae, and eggs; the queen’s private quarters …

Some ants have stables where they keep the aphids they milk or vegetable plots for growing edible fungus or for germinating tough seeds so that they can be eaten. And then there are the different parts of the nest’s transportation system: foraging trunk routes, thoroughfares within the nest, peripheral branches, even an endlessly branching system of roads connecting the nest with its hinterland; without any central planning or budgets, ants are able to build sophisticated travel networks that human urban planners often cannot match.

Each of these many different substructures of the ant nest and its surroundings has its own specialized myrmecophiles. This already starts on the access roads leading in and out of the nest. The European jet ant (Lasius fuliginosus) has its main transportation routes up and down tree trunks, and this is where the beetle Amphotis marginata hangs out. These beetles are true highwaymen. By day, they hide in shelters along the trail, but at night they come out and stop passing ants that are returning to the nest with food. The beetle uses its short, powerful antennae to tap the ant’s mouth and drum rapidly on top of its head. This mimics, in a rather persuasive way, the begging behavior of ants in the nest, and the startled ant will void its crop, the regurgitated food being quickly lapped up by the beetle. The ant, however, often realizes its mistake and then tries to attack the vagabond. But Amphotis is flat and big and heavily armored, and it just cowers, withdraws its appendages and is as impregnable as a tank, so that the duped worker ant soon gives up and returns to its nest empty-handed.

Inside the nest of the jet ant we find another beetle plying its trade. The larvae of the rove beetle Pella funesta are the nest’s garbage men. They live in the nest’s refuse heaps where they consume dead ants, staying out of sight by feeding from below or even getting inside the ant corpses. When a worker ant attacks them, the larvae lift their abdomen which carries glands with chemicals that instantly relaxes or confuses them—like some sort of “antnip.” The adults of Pella funesta scavenge on dead ants, too, but in addition, they also hunt live ants, sometimes in a group. Like a pride of lions, the beetles give chase and one of them will try to climb on an ant’s back, get its jaws into the ant’s neck and sever its nerves and throat. These attacks often fail, but if successful, the whole pack of beetles will feast on the prey communally.

The nest’s Eldorado, however, is the brood chambers. Here, the ants bring their highest-quality food (freshly-killed insects, for example) for their newborn larvae. Many myrmecophiles have found their dream niche there, either begging food from the ant workers by chemically pretending to be ant larvae, or preying on the larvae themselves. But brood chambers are also heavily defended. Any interloper discovered there will be killed instantly. So the myrmecophiles who have evolved brood chamber specialization have also needed to evolve very sophisticated techniques to evade the ants’ enemy detection. The peculiar beetle Claviger testaceus is one of them. It bears the hallmarks of millions of years of adaptation to living inside ants’ nests. It is pale, with a curious elongated, eyeless head, strange, club-like antennae, and thick tufts of golden hair on its back. Once again, the secret lies in those tufts of hair. Underneath lie glands that produce chemicals that apparently give off the smell of death. Of insect cadavers, that is. An ant worker coming across a Claviger beetle will take it for freshly killed prey (further fooled by the beetle playing dead), pick it up by its conveniently stalk-like forebody, and then carry it to the brood chamber, where all the most tasty morsels go. There, it may dump additional bits of decaying meat on the beetle, cover the pile in puked-out saliva with digestive enzymes, and move on to other chores—thinking it has done the developing larvae a favor. But, in fact, as soon as the Claviger scrambles from underneath the pile of insect remains, it will start feeding on the ant eggs, larvae, and pupae.

Claviger testaceus, Pella funesta, and Amphotis marginata are just three of the 10,000 or so different myrmecophile species that scientists think exist, belonging to at least a hundred different families of invertebrate animals. This evolutionary explosion of myrmecophily has probably been going on for as long as there have been ant societies—at least some 75 million years. The reason being that ants belong to that elite corps of movers-and-shakers that ecologists call “ecosystem engineers.”

The term “ecosystem engineer” was coined in an article in the journal Oikos in 1994 by three ecologists: Clive Jones, John Lawton, and Moshe Shachak. They write: “Ecosystem engineers are organisms that […] modulate the availability of resources to other species, by causing physical state changes in biotic or abiotic materials. In so doing they modify, maintain and create habitat.” To put it bluntly: ecosystem engineers create their own ecosystems. It is easy to see how ants fit this definition. Ants branch out into their environment, and, by virtue of their advanced levels of self-organization, concentrate resources in their nest. The inside of the nest is a novel ecosystem with a constant influx of energy in the form of food carried in by ants, that may be exploited by other species. Those 10,000 myrmecophiles are the new species that have evolved to make use of the opportunities that the ants’ engineered ecosystem offers. But even species that do not qualify as myrmecophiles may be affected by the ants’ modifications of their environment. Such as those stinging nettles growing on the nitrogen-rich patch around the red wood ant nest that I excavated.

Many organisms besides ants are important ecosystem engineers as well. Think of other animals that create structures much larger than themselves, such as termites or corals. And ecosystem engineers need not be tiny. Take beavers, for example. There is no better hydrological engineering team than a family of beavers. They chew down trees and use these, together with rocks, to build dams up to hundreds of yards long. In slow-flowing water, they will build a straight dam, but in a faster-flowing river, the dam will be curved to better withstand the push of the water. The dams cause the stream waters to slow down and widen, creating swamp land that is less easily traversed by the beavers’ predators, such as wolves, and maintains a steady supply of beaver food (water plants and tree saplings) during winter. The animals dig canals to transport logs that are too heavy to drag over land, and they build lodges: large hut-like dwellings constructed from branches, twigs and grass, and solidified with mud, bits of wood, and bark. Because of all this environmental enhancement, beavers have such an overriding impact on their environment that they create new niches for whole swathes of other species. Even after beavers abandon an area and the dams they have erected decay and are breached, the resulting flood allows the development of meadows that can persist for decades after the beavers have left.

One area where beavers had such an impact in the past is a large island on the east coast of North America, in the estuary of the Muhheakantuck river. The elongated island has gently rolling elevations and depressions—its local Lanape name means “island of many hills.” Until a few hundred years ago, most of those were abundantly clad in chestnut, oak, and hickory forest, which sucked up the abundant rainwater and only released it piecemeal, allowing 62 miles of slow-flowing creeks and streams to develop all over the island. Beavers were aplenty in this fine beaver habitat. In one spot in the southern part of the island, two creeks converged on a gently depressed valley. Beavers dammed the creeks, and the valley transformed into a red maple swamp, slowly colonized by other animals that feel at home in such a habitat, like wood ducks, green frogs, and brown bullhead. Besides the red maple, there was northern water plantain and marsh blue violet. We know all this because of a study—groundbreaking in more than one sense of the word—led by landscape ecologist Eric Sanderson of the Wildlife Conservation Society in New York. Using information on the island’s climate, soil types and topography, early Dutch and English records of its landscape and wildlife, and computer modeling of the entire food web of that part of North America, they were able to reconstruct what the landscape, and all the life it supported, looked like four hundred years ago.

Today, nothing remains of what was there. For that island is Manhattan, and Eric Sanderson’s work is also known as the Mannahatta Project. The project’s purpose was to create a website with a navigable map of today’s Manhattan where any location could be chosen, stripped from all its human structures to reveal, in full color and detail, the model’s best estimate of that location’s habitat and abundant wildlife before Europeans set foot. “[Four hundred] years of development have rendered this earlier abundance as difficult to imagine to us as perhaps our modern roads, skyscrapers, and wealth would be to those first European colonists and their Native American neighbors,” Sanderson writes. His aim was reached by September 12, 2009, the quadricentennial of the day when Henry Hudson, sailing in on a ship of the Dutch East Indies Company, first set eyes on it, and scribbled in his log, “[A]s pleasant a land as one can tread upon.”

Indeed, when you visit the project’s interactive map on, it’s as if Google Earth has directed you to one of the few remaining unspoiled wildernesses on earth. Coast-to-coast forest cover, only interrupted here and there by meadows, swamps, brooks, some Lanape settlements, and a few beaches and rocky bluffs along the shore. A paradisiacal place. But click on the button “STREETS,” and the modern street plan appears on top of all that verdure. Suddenly you realize that that lush brook you have been staring at is actually in what today is Harlem, or Greenwich Village. For example, that confluence of two creeks, where beaver ecosystem engineering had created a red maple swamp, lay smack in the middle of what now is Times Square, with one stream appearing from the New York Post building, the other from under the Jacqueline Kennedy Onassis High School.

By now, you may have an inkling of where this narrative has been heading. By clicking on the buttons on the Mannahatta Project’s interactive map, we are toggling between the work of one ecosystem engineer and another. The beavers of Mannahatta are gone, but they have been replaced by what we could call nature’s ultimate ecosystem engineer: Homo sapiens—running around in modern-day Manhattan, the ecosystem it has engineered for itself, like ants in an anthill. And, as with any good ecosystem engineer, in so doing it has created niches for cohabiting animals and plants. Not myrmecophiles, but, if you will, anthropophiles. It is those anthropophiles and the niches they carve for themselves in the human-engineered ecosystem that we will discover in this book.

Copyright © 2018 by Menno Schilthuizen