THINK about evolution and you’ll probably conjure up a picture of natural selection sculpting organisms over aeons of time to become adapted to specific environments. Hot climates select animals with heat-dispersing adaptations, such as large ears or the ability to pant. But no matter how much an animal flaps its ears or pants, it is not going to affect the local temperature to any significant extent. Environments shape living things: it rarely works the other way around. At least, that is the conventional view of evolution.
But how does this square with the seemingly innocent observation that the activities of all living creatures bring about changes in their environments? We are not just talking about birds constructing nests, spiders spinning webs and beavers engineering dams. There is much more to this process of “niche construction” than the creations of charismatic animals. Plants change levels of atmospheric gases, modify nutrient cycles, engage in chemical warfare, promote forest fires, create shade and alter wind speeds. Fungi decompose organic matter, weather rocks and extract minerals. Even bacteria and the simplest single-celled creatures leave the world in a different state from how they found it, through decomposition, photosynthesis, nutrient fixation or by initiating ecological processes that allow other organisms to colonise new environments. Surely these changes must influence evolution?
For several years now we have been grappling with this conundrum. Together with our colleague Marcus Feldman from Stanford University in California, we have been trying to assess the full extent of niche construction and its implications for evolutionary biology and ecology. Our studies have convinced us that niche construction should be recognised as a significant cause of evolution, on a par with natural selection. We are arguing for nothing less than a rethink of the evolutionary process itself. By accepting that organisms shape environments as surely as environments shape organisms, evolution is transformed from a linear to a cyclic process. This feedback allows successive generations of organisms to influence their own and each other’s evolution.
Advertisement
If this feedback between organisms and their environments is so crucial to evolution, why hasn’t it been recognised before? The answer is that niche construction is considered in certain limited situations, but rarely in terms that fully capture its evolutionary ramifications. Arguably the most influential description of the interaction between living things and the environments they inhabit comes from Richard Dawkins, in his book The Extended Phenotype. According to Dawkins, “an animal artefact …can be regarded as a phenotypic tool by which that gene could potentially lever itself into the next generation.” He argues that genes build environmental states – the “extended phenotypes” of the title – to their own ends, reaching beyond the bodies of organisms to be expressed in the construction of webs, mounds and bowers.
Like most other evolutionary biologists, Dawkins views niche construction as a product of naturally selected genes, not a part of the evolutionary process itself. Put another way, the only relevant evolutionary feedback from extended phenotypes is to the genes that express them. So when beavers build dams, they ensure the propagation of “genes for” dam building, but that is all.
Yet by constructing their own niche, beavers radically alter their environment in many ways. With dams come calm, protective lakes, safe havens for their artificial island homes or lodges as well as for stores of food in the form of saplings saved for the winter. By cutting down trees, beavers alter the local hydrology, creating wetlands that may persist for centuries and influence plant and animal diversity. All this “engineering” is likely to modify natural selection not only for the genes associated with niche-constructing activities, but also for other genes expressed in quite different traits, such as beavers’ life history, tails, teeth, foraging behaviour, susceptibility to predation and diseases – not to mention the evolutionary effect they have on other animals that share their altered environment.
You need look no further than the humble earthworm to see the power of niche construction. Across the globe, earthworms have dramatically changed the structure and chemistry of soil by burrowing, dragging plant material into the soil, mixing it up with inorganic material such as sand, and mulching the lot by ingesting and excreting it as worm casts. The scale of these earthworks is vast. What’s more, because earthworm activities result in cumulative improvements in soil over long periods of time, it follows that today’s earthworms inhabit environments that have been radically altered by their ancestors.
In other words, some extended phenotypes can be inherited. Through niche construction, organisms not only acquire genes from their ancestors but also an “ecological inheritance” – a legacy of natural selection pressures that have been modified by ancestors and that, in turn, shape subsequent generations. In the case of earthworms, some characters, such as the structure of their epidermis, or the amount of mucus they secrete and the strange physiology of their kidneys – as Scott Turner from the State University of New York in Syracuse describes in his book The Extended Organism – probably co-evolved with niche construction over many generations.
Ecological inheritance does not depend on the presence of biological replicators, but only on the persistence of physical changes made by ancestral organisms in the environments of their descendants. This kind of inheritance, which has been largely overlooked by evolutionary biologists, has more in common with the passing of property from generation to generation than that of genes. We believe it can be found throughout nature wherever generations of organisms progressively alter the environment in which their descendants live – not least in our own species, where children are born into a world elaborately created by the planet’s virtuoso niche constructor. And ecological inheritance has important implications for understanding human evolution. For example, it challenges the idea that humans are adapted to an ancestral environment that existed from about 1.8 million years ago on the African savannah, an idea argued most vociferously by evolutionary psychologists.
Close inspection of the niche-constructing activities of countless organisms, from bacteria to humans, has convinced us that this is an important evolutionary process. But we are aware that we are not going to persuade others to abandon the conventional evolutionary perspective unless we can show that the feedback from niche construction is substantial. To that end, we have carried out mathematical analyses using population genetics models. They reveal that niche construction is far from inconsequential.
We found that niche construction forges new evolutionary pathways for species, sometimes allowing apparently damaging mutations to become incorporated into populations, at other times removing characters that seem well adapted, or creating new equilibria between alternative genetic forms. What’s more, ecological inheritance can generate time lags in a population’s response to selection. These manifest as curious “momentum effects” (where populations continue to evolve in the same direction after selection has stopped or reversed), “inertia effects” (where there is no noticeable evolutionary response to selection for several generations), and unpredictable or catastrophic responses to selection. In other words, niche construction can dramatically change the evolutionary dynamic.
Niche construction is not simply a product of prior natural selection of genetic traits, as most evolutionary biologists contend. It is a separate process that creates evolutionary pressures in its own right. This is particularly obvious when you look at genetic changes that have come about as a result of niche-constructing activities which are not themselves guaranteed by specific genes. The woodpecker finch of the Galapagos Islands, for example, uses a cactus spine to peck for insects under bark. It has no “grubbing genes” per se, but instead uses a more general and flexible adaptation – the ability to learn. Nonetheless, the birds’ learned activities have apparently created selection pressures favouring a bill capable of wielding tools, rather than the long, pointed bill characteristic of woodpeckers.
Another example of a niche-constructing behaviour leading to genetic change is found in our own species. Around the world, the ability of adults to consume dairy products without becoming sick – lactose tolerance – reflects their ancestors’ history of dairy farming. Recent genetic analysis reveals that dairy farming emerged before the spread of the genes that allow adults to digest lactose (the energy-rich sugar in milk), indicating that the practice of dairying almost certainly created the natural selection pressures that favoured these genes.
Incorporating such feedback loops into evolutionary models will certainly create some headaches, but the new perspective offers some major benefits. In particular, it has the potential to unify evolution and ecology in a way that has not been possible until now.
At first sight, ecology appears thoroughly evolutionary. But take a closer look and you find that not all ecologists use evolutionary methods. Ecosystem ecologists paid a high price for accommodating Darwinism, namely the division of their field into two approaches: “population-community ecology”, which takes an evolutionary perspective, and “process-functional ecology”, which does not. Superficially, these two sub-fields appear to be separated by little more than their choice of subject matter, but in practice they have proved remarkably difficult to integrate.
Typically, process-functional ecologists ask questions about the composition of ecosystems, their scale and their boundaries, their structural and functional design, and their regulation. One of their principal goals is to understand how energy and matter flow through both living and nonliving parts of an ecosystem. So they often grapple with complete biogeochemical cycles such as the carbon or nitrogen cycle. In contrast, population-community ecologists concentrate primarily on organisms. For them, living creatures are the ecosystem, and the nonliving components (soil, water and so on) are largely viewed as backdrops to the tapestry of life.
The key sticking point is that nonliving components do not evolve. Take the carbon cycle. Carbon typically enters a living community when atmospheric CO2 is fixed by plants and bacteria in photosynthesis. Later it becomes available for consumption by herbivores as cellulose and sugar, and subsequently by carnivores, as fat and protein, before being released back into the environment through respiration, decay and other processes. In the soil, it may be taken up again by organisms or washed away by rain.
Living, evolving organisms obviously play a huge role in the carbon cycle, but the carbon in the atmosphere, soils, rivers and oceans is effectively out of bounds for evolutionary theory. As a result, ecologists are left with a tough choice. They can pretend that evolution has stopped and incorporate the nonliving components, allowing them to study chemical cycles such as the carbon cycle, but only in terms of flows of energy and matter. Or they can incorporate evolution, but at the cost of editing out the physical environment. This division makes it difficult to build a complete picture of some of the central players in ecosystems. But the niche construction perspective offers a way out of this impasse. By emphasising the role of organisms in modifying and controlling environmental resources, it provides evolutionary methods that apply to nonliving resources too.
While most ecologists do not yet think organisms can orchestrate ecosystems, a growing number are seeing the benefit of incorporating these ideas into their thinking. They include Clive Jones of the Institute of Ecosystem Studies in Millbrook, New York, and his colleagues John Lawton from the Centre for Population Biology at Silwood Park, UK, and Moshe Shachak from Ben-Gurion University at Sede Boqer, Israel. They have been studying “ecosystem engineering” in a wide variety of species, including three species of snail of the genus Euchondrus, which eat lichens that grow under the surface of rocks in the Negev desert in Israel. One consequence of this unusual form of herbivory is that the snails play a major role in eroding rock.
It’s not that these snails consume large amounts of lichens – they don’t – but rather that they have to ingest rock to get at the lichens underneath. They later excrete this as faeces, which becomes soil. The researchers estimated that the annual rate of biological weathering of these rocks by snails is approximately 1 tonne per hectare, which is sufficient to affect the whole desert ecosystem. By converting rock to soil at this rate, the snails are major agents in both soil formation and nitrogen cycling, and critical to the establishment of higher plant communities.
Studies like this show that a focus on niche construction provides ecologists with fresh tools for linking species to ecosystems. The potential of this approach is considerable. For instance, it might allow ecologists to predict which species will invade a community and what their impact will be, depending on the species’ ability to tolerate the effects of niche construction by other populations and the nature of their own engineering effects. It also suggests a host of new empirical methods, such as tracing the niche construction of populations round entire ecosystems, and identifying particular modifications of the environment associated with specific genes.
Niche construction may also shed light on some of ecology’s biggest questions. For instance, nature seems so harmonious and coordinated that some have wondered whether ecosystems are super-organisms. In his Gaia hypothesis, James Lovelock went so far as to use the metaphor of a living entity with powers of self-regulation to describe Earth. But understanding niche construction allows you to see ecosystems not as mysterious super-organisms, but rather as super-constructions created by the collective activities of their constituent organisms. The impacts of niche construction thread entire ecosystems, binding them together – which explains their impressive structural and functional integration and underlies the illusion that they are alive. Yet, since not all niche construction imposes order, it also explains why other ecosystems are not functionally integrated but are the messy, uncoordinated outcomes of several species pulling in different directions.
Nowhere is this more apparent than in some man-made environments. Understanding the impact of our own activities on ecosystems and on evolution could give us new insights into the legacy we are creating. It might even help us find ways to reduce the damaging consequences of our actions. For example, conservation efforts might be more effective if they ensured the survival of the key niche constructors within any ecosystem. Researchers studying niche construction and ecosystem engineering are devising new tools for identifying these.
But their future work may provide an even more radical approach to conservation because, in some cases, the best way to preserve an ecosystem may be to preserve the niche-constructing effects of particular animals or plants in it, rather than the species themselves. So niche construction could offer a new kind of medicine for sick ecosystems.