Starting in the late 1970s, aspiring evolutionary biologist David Reznick became intent on documenting evolution in action. Although he had learned in school that observable change took place over millennia, the young biologist questioned that notion, and set out to observe genetic adaptation in real time. “Prominent evolutionary biologists were skeptical you could see it happening,” Reznick recalls. “I guess it was a gamble, but it seemed like one that was worth taking.”
He chose Trinidadian guppies as his study system, and as a graduate student at the University of Pennsylvania Reznick flew to the Caribbean island off the north coast of Venezuela in 1978 to observe and collect guppies, and again in 1981 to shuffle the fish between streams. He moved guppies that were living among cichlids and other larger fish to low-predation sites that also lacked other guppies.
“People thought, if we want to understand the process of evolution, we look at the fossil record,” says Reznick, now a biology professor at the University of California, Riverside. But by averaging phenotypic change across tens of thousands or millions of years, the fossil record underestimates rates of change on shorter timescales, so “for a very long time, people were looking at a biased image of how evolution happens.”
Around the same time, researchers were observing similarly high rates of change in other animal populations. As Reznick was puddle-jumping across Trinidad, famed husband-and-wife evolutionary biologists Peter and Rosemary Grant in the Galápagos Islands were documenting changes in the size and shape of finch beaks following environmental fluctuations.3 These natural experiments revealed bursts of change in the birds’ beaks after an exceptionally strong El Niño event in the early 1980s, for example,4 and in response to heightened competition for food between the medium ground finch (Geospiza fortis) and the large ground finch (G. magnirostris), after the latter’s migration to Daphne Major, the Grants’ research site, in 1982.5 Statistical analyses involving pedigrees that demonstrated the high heritability of beak morphology determined that these changes were primarily the result of genetic evolution, not phenotypic plasticity.
These studies weren’t the earliest to document quick shifts in a population’s traits, but “the work came out at just the right time,” says McGill University evolutionary ecologist Andrew Hendry. “People were ready for the idea that evolution occurred rapidly. . . . The finch and the guppy work [served as] the empirical exemplars from which one could say that this is a general phenomenon.”
Adding to those examples, evidence of evolution occurring faster than previously appreciated has continued to accumulate over the past few decades. It is now clear that, while observed rates of change may well be rapid relative to the evolutionary timescales that Darwin theorized about, they’re not at all exceptional. “Within evolutionary biology there really has been an unheralded paradigm shift between 1980 and now,” says Reznick. “Most evolutionary biologists consider it routine to think of evolution as a contemporary process.”
The concept, appropriately termed contemporary evolution, is now well accepted, agrees Stephen Ellner, an ecologist and evolutionary biologist at Cornell University. “At this point, there’s a general understanding that this is happening, and it’s happening all over.” The research has now shifted from documenting this phenomenon to studying its consequences.
How fast is fast?
When discussing evolution, defined as changes in the genetic makeup of a population, the relevant unit of time is the generation: alleles that aid in the organisms’ survival and reproduction are likely to increase in frequency from one generation to the next, while those that reduce fitness will likely become less common. For this reason, it is easier to study evolutionary change in organisms with shorter generation times.
WIKIMEDIA COMMONS/HANS HILLEWAERTThe other key factor dictating the rate of evolutionary change is the strength of selection. The rapid appearance of resistance mechanisms in bacteria exposed to antibiotics is the epitome of what can happen when extreme selection is imposed on a species with short generations. While researchers have long recognized that antibiotic resistance can develop quickly in infection-causing pathogens, the idea that contemporary evolution could be detected in macrofauna species remained on the scientific fringe for years. But, Ellner posits, if it’s happening in bacteria under intense selection, why not in other organisms under other conditions? “A priori there’s no reason [to think] the same sort of thing wouldn’t be happening when generation times are longer and selection isn’t that strong.”
The literature on “rapid evolution” is now 30 years deep. Using what are known as common garden experiments—raising animals from different populations in the same controlled environment (as Reznick did with the guppies)—researchers have observed some astonishing rates. In 1997, Reznick and his colleagues calculated rates of change in his guppy experiments of “up to seven orders of magnitude greater than rates inferred from the paleontological record,” the authors wrote in Science.6 That same year, Harvard University’s Jonathan Losos, then at Washington University in St. Louis, and collaborators published a Nature paper documenting the differentiation of anole populations over a decade and a half following the release of the lizards onto 14 small islands in the Bahamas in 1977 and 1981.7
© ISTOCK.COM/Patrick_Gijsbers The studies caught the attention of Hendry and Michael Kinnison, then fellow graduate students in Thomas Quinn’s lab at the University of Washington. Both young researchers were studying rapid evolution in salmon, but despite the focus of their studies, neither had given much thought to quantifying the rates of change in a way that would be comparable across species or defining what should count as rapid. “A lot of [researchers] were working under the assumption that, if you could see it, it was rapid. If you could see it within a human life span, then it must be exceptional,” says Kinnison, now a professor at the University of Maine. “What is rapid or not rapid should have some measurable frame of reference.”
After the Nature and Science studies came out, Hendry and Kinnison got together to develop a framework for quantifying rates of evolutionary change. They promoted the use of the haldane (a change of one standard deviation in a phenotypic trait per generation; named after evolutionary thinker J.B.S. Haldane) over the darwin (the proportional change in a phenotype per million years; named after you-know-who), and encouraged researchers to provide confidence intervals and measures of statistical significance. If done correctly, “evolutionary rates provide a convenient way to compare the tempo of evolution across studies, traits, taxa, and time scales,” Hendry and Kinnison wrote in 1999.8
In November 2001, the researchers published a meta-analysis of patterns in reported rates of contemporary evolution.9 Mining the literature for usable data sets, Kinnison and Hendry found no shortage of studies documenting change over the course of a study. “There was a lot more out there than we even suspected ourselves,” says Kinnison. “The more and more we dug, the more and more cases that we found . . . all sorts of species spanning all taxonomic breadth.”
Most evolutionary biologists consider it routine to think of evolution as a contemporary process.—David Reznick, University of California, Riverside
In total, the researchers gathered data on 30 different animal species, for a total of 2,151 evolutionary rates calculated in haldanes, and another 2,649 in darwins. The sources included both “genetic” studies—those that performed common garden experiments or used quantitative genetics to infer genetic change in wild populations—and “phenotypic” studies, which just measured change in a trait over time, and thus represented the combined effects of genetic adaptations and phenotypic plasticity. Analyzing all the data together or the genetic data separately yielded similar results. As it turned out, part of the reason so many scientists had been able to document dramatic change over short time frames was precisely because they were limiting the duration of observation; in the short term is when evolution’s at its fastest.
It makes sense, says Kinnison. “If populations are really tracking dynamic environments—year-to-year variation in climate, other species that they depend upon or compete with, or the like—then you would expect their traits to be bumping around pretty rapidly. But over the long term, a lot of these processes average out, [which] flattens those rates down.”
The study was published in a special issue of Genetica, devoted to the topic of contemporary evolution and edited by Hendry and Kinnison. “I guess what Mike and I did was, by bringing these 30 papers about rapid evolution together, we made it clear that these weren’t isolated phenomena that were just weird exceptions,” Hendry says. “They were actually kind of common, and you could use that information to ask questions that weren’t just about guppies or finches.”
Ecological theory evolves
© MESA SCHUMACHERAt the 2005 annual meeting of the Ecological Society of America (ESA) in Montreal, Hendry organized a symposium dedicated to contemporary evolution. “All the main players were there,” Hendry says. At that meeting, Tom Whitham of Northern Arizona University gave a talk about what he called “community genetics,” describing how genetic variation within populations of cottonwood trees had ecological consequences, such as effects on the insect and microbial communities. “One genotype of cottonwood tree will create a very different environment than another genotype,” Hendry recounts. “That was, for me, the realization that rapid evolution wouldn’t just have consequences for the organisms themselves, but for the rest of the community also and for the ecosystem.”
Traditionally, ecologists modeling communities and biomes have treated species as static organisms, assuming that evolution takes place on timescales that are far too slow to influence ecosystem dynamics, which operate on the order of weeks, months, and years. But as many studies have now shown, organisms are not constant at all, and “the rate at which they change is comparable to the rate at which ecological interactions are happening,” says Reznick. “It can have a profound effect.”
Two years before the ESA meeting, working in collaboration with the lab of Cornell colleague Nelson Hairston, Ellner and colleagues published the first experimental demonstration of this effect in the lab, showing that the rapid evolution of algae (prey) in response to oscillating densities of rotifers (predators) substantially altered the overall predator-prey dynamics.10 “Often when you look at natural populations of predators and prey, you get cycles of abundance,” says Reznick. “What they showed was that those oscillations change in a very dramatic way if the prey are evolving as part of the cycle; the whole structure of the cycle is different.”
Studying such eco-evolutionary dynamics in natural systems “is vastly harder to do,” Hendry says. But a handful of studies make a strong case that the dynamics researchers observe in the lab are also operating in nature. In 2011, for example, Reznick worked in collaboration with Martin Turcotte’s group at the University of Pittsburgh to demonstrate that evolving field populations of green peach aphids grew significantly faster and reached higher densities than control populations that could not evolve because they harbored no genetic variation.11,12 In 2014, Tim Farkas, a postdoc at the University of Connecticut, reported similar dynamics in experimental populations of stick insects, where the relative fitness of two morphs—equally represented at the outset—varied according to the density of the founding populations.13 And in the streams of Trinidad, Reznick’s team has observed that guppy evolution triggers a ripple of change through the ecosystem. Guppies play an important role in limiting algal growth in the streams, for example, and the fish prey on young killifish, creating a complex predator-prey dynamic. “Guppies have always been the victim—killifish eat baby guppies—but guppies eat baby killifish,” Reznick says, and the two species compete. As the guppies evolve, changes in them can affect all of these interactions, which can further affect the environment and all the organisms evolving within it.
“We’ve got all of these interesting examples of what you would call eco-evolutionary effects,” says Kinnison—“places where evolution is feeding [back on] some ecological aspect.”
RON BASSARFrom field studies like these that validate laboratory findings and theoretical work, researchers know that individual genotypes—and diversity itself—have ecological consequences, says Hendry. And the general rule is that more individual genotypes equates to better ecological function. In fact, he says, “evolution within a community will tend to maintain diversity in that community, on average . . . generally enhancing ecosystem function, productivity, nutrient cycling, things like that.”
The complexity of both ecosystems and evolutionary theory make these dynamics very challenging to dissect, however. In addition to the logistical limitations of the research, there are theoretical challenges. For example, most studies of contemporary evolution follow a focal species, when in reality, every species in a community is evolving at the same time, Hendry says. And it’s very possible that eco-evolutionary dynamics are yielding “cryptic” effects. “The classic measure of eco-evolutionary dynamics is a [phenotypic] change,” says Hendry. “But what evolution often does is it generates stability, so you see no change. So you have this paradox where the lack of apparent dynamics actually probably reflects a huge underlying component of eco-evolutionary dynamics, and that’s just super hard to study.”
In 2005, Ellner, Hairston, and colleagues outlined ways of assaying how important evolutionary change is for ecological dynamics.14 And in 2011, the researchers published an updated protocol, emphasizing the data and statistical analyses needed to determine whether observed changes are the result of evolution or of differing ecological conditions.15 “By hook or by crook, do the full factorial of the population before and after the evolutionary change in the environmental conditions. Then you can parse out the contributions,” says Ellner.
The field is young, and more than anything, it needs more time, he adds. “In part, it’s just a matter of waiting for the data to accumulate. . . . One generation is the speed limit.”
The human factor
Although researchers aren’t entirely sure how evolutionary and ecological forces interact, these newly appreciated dynamics demand consideration in a number of different contexts, perhaps most obviously, conservation. Organisms are likely evolving in response to the degradation of their habitats, for example, not just dying off randomly such that the population’s genetic mix remains the same even as it dwindles toward extinction.16
For example, the occurrence of rapid evolution could have implications for the world’s fisheries. “When we are not targeting the small juveniles and babies that natural predators would, but we are actually taking out the big adults that are in their reproductive prime, it makes sense to every evolutionary ecologist that we are causing evolutionary changes, genetic changes,” says Silva Uusi-Heikkilä, an evolutionary biologist at the University of Turku in Finland.
Indeed, Kinnison and his colleagues have calculated that populations harvested by fishermen, hunters, or plant cultivators evolve some three times faster than organisms subjected to other types of selective forces.17 In addition to the severity of harvest—with people effectively clearing out all individuals of a particular size range, for example—the practice also exerts a very consistent selective pressure, in contrast to the instability of many environmental factors, Kinnison explains. “Harvest is the champ,” he says. “Harvest drives evolution faster than everything else.”
It makes sense to every evolutionary ecologist that by fishing we are causing evolutionary changes.—Silva Uusi-Heikkilä, University of Turku
In the laboratory, Uusi-Heikkilä and her colleagues have found that size-selective harvesting of fish can indeed drive their evolution. Starting as a graduate student at the Leibniz-Institute of Freshwater Ecology and Inland Fisheries in Berlin, she raised zebrafish in large tanks, periodically scooping them out to measure them all and removing 75 percent of the largest fish, to mimic the situation in recreational and commercial fishing. “We were able to show that only five generations of size-selective harvesting caused genetic changes in these populations,” Uusi-Heikkilä says. Those populations that were “fished” were smaller as adults and had lower reproductive output. Females produced fewer eggs and spawned less frequently. The zebrafish were also less active, less explorative, and less bold.
But the importance of evolution has remained controversial in the fisheries industry, says Uusi-Heikkilä, largely because it’s difficult to demonstrate that evolutionary changes in wild populations are the result of size-selective harvest and not some other factor. And fisheries biology is a contentious field. “It can be difficult; there’s a lot at stake,” Uusi-Heikkilä says. In addition to being a scientific endeavor, “this is an economic [and] political issue.”
Contemporary evolution also throws a wrench into the use of conservation hatcheries, captive-breeding programs initiated to help repopulate wild fish stocks. In 2008, Mike Blouin’s group at Oregon State University, along with NOAA researchers at the Northwest Fisheries Science Center in Washington State, published a study that provided evidence that fish were adapting to captivity in ways that made them less fit when released into the natural environment. The team collected steelhead trout from the Hood River, a tributary of the Columbia River a few hours’ drive north of the university, and bred them in captivity for two generations. The researchers then bred the captive fish with wild-caught fish and raised those offspring in the hatchery until they matured, alongside the offspring of two wild fish to control for the effects of the rearing environment. They then released this fish to breed in the wild, and for three years sampled wild trout for genetic analyses to determine if they were descendants of the fish raised in captivity. They used this information to calculate the reproductive success of the captive-reared trout, and compared fish whose genomes came entirely from wild trout with those that had one captive-raised parent.18
“[It was] a really powerful experiment,” Kinnison says. “They estimated that the loss of fitness in the wild from just one generation of having parents bred in captivity was pushing upwards of a 40-percent reduction in performance. It was shocking.”
How eco-evolutionary dynamics could inform the development of sustainable harvesting practices remains an open question. But the importance of considering these forces when developing fisheries-management strategies is slowly being recognized by the fishing industry. Five years ago, an international group of researchers suggested the implementation of evolutionary impact assessments “as a structured approach for assessing the evolutionary consequences of fishing.”19 And in November 2015, researchers published the first peer-reviewed evolutionary impact assessment, that of the North Sea plaice fishery.20 “I think it’s really now starting to catch,” says Kinnison.
Uusi-Heikkilä knows without more solid fieldwork there’s still a steep hill to climb to convince skeptics; laboratory studies alone just won’t cut it. “I have to be very careful when I mention zebrafish and fisheries in the same sentence,” she admits. “But what we are able to do with this model system and experimental studies is disentangle the plastic and evolutionary effects to show that it is possible. Because I think previously people didn’t even believe this is possible. I mean, evolution can’t happen in a few generations, it’s something that takes millions of years.”
- D. Reznick, J.A. Endler, “The impact of predation on life history evolution in Trindadian guppies (Poecilia reticulata),” Evolution, 36:160-77, 1982.
- D.N. Reznick, “The impact of predation on life history evolution in Trinidadian guppies: The genetic components of observed life history differences,” Evolution, 36:1236-50, 1982.
- P.R. Grant, B.R. Grant, “Unpredictable evolution in a 30-year study of Darwin’s finches,” Science, 296:707-11, 2002.
- B.R. Grant, P.R. Grant, “Evolution of Darwin’s finches caused by a rare climatic event,” Proc R Soc B, 251:111-17, 1993.
- P.R. Grant, B.R. Grant, “Evolution of character displacement in Darwin’s finches,” Science, 313:224-26, 2006.
- D.N. Reznick et al., “Evaluation of the rate of evolution in natural populations of guppies (Poecilia reticulata),” Science, 275:1934-37, 1997.
- J.B. Losos et al., “Adaptive differentiation following experimental island colonization in Anolis lizards,” Nature, 387:70-73, 1997.
- A.P. Hendry, M.T. Kinnison, “The pace of modern life: Measuring rates of microevolution,” Evolution, 53:1637-53, 1999.
- M.T. Kinnison, A.P. Hendry, “The pace of modern life II: From rates of contemporary microevolution to pattern and process,” Genetica, 112-113:145-64, 2001.
- T. Yoshida et al., “Rapid evolution drives ecological dynamics in a predator-prey system,” Nature, 424:303-06, 2003.
- M.M. Turcotte et al., “Experimental assessment of the impact of rapid evolution on population dynamics,” Evol Ecol Res, 13:113-31, 2011.
- M.M. Turcotte et al., “The impact of rapid evolution on population dynamics in the wild: Experimental test of eco-evolutionary dynamics,” Ecol Lett, 14:1084-92, 2011.
- T.E. Farkas, G. Montejo-Kovacevich, “Density-dependent selection closes an eco-evolutionary feedback loop in the stick insect Timema cristinae,” Biol Lett, 10:20140896, 2014.
- N.G. Hairston Jr. et al., “Rapid evolution and the convergence of ecological and evolutionary time,” Ecol Lett, 8:1114-27, 2005.
- S.P. Ellner et al., “Does rapid evolution matter? Measuring the rate of contemporary evolution and its impacts on ecological dynamics,” Ecol Lett, 14:603-14, 2011.
- C.A. Stockwell et al., “Contemporary evolution meets conservation biology,” Trends Ecol Evol, 18:94-101, 2000.
- C.T. Darimont et al., “Human predators outpace other agents of trait change in the wild,” PNAS, 106: 952-54, 2009.
- H. Araki et al., “Genetic effects of captive breeding cause a rapid,cumulative fitness decline in the wild,” Science, 318:100-103, 2007.
- A.T. Laugen et al., “Evolutionary impact assessment: Accounting for evolutionary consequences of fishing in an ecosystem approach to fisheries management,” Fish Fish, 15:65-96, 2014.
- F.M. Mollet et al., “Evolutionary impact assessment of the North Sea plaice fishery,” Can J Fish Aquat Sci, 73:1126-37, 2016.