Decades ago, a Soviet geneticist purposely bred foxes to make them extra tame, an experiment that produced a host of unanticipated physical changes in the animals. It’s one of the most famous experiments in genetics, but it might not have gone down in the way we were told. A new opinion paper argues these foxes weren’t wild to begin with and that the “domestication syndrome” associated with the changes doesn’t exist—a claim that’s stirring up controversy among some biologists.
As the story goes, geneticist Dimitry Belyaev, who worked at the Institute of Cytology and Genetics in Novosibirsk, USSR, took 30 male and 100 female wild foxes (Vulpes vulpes), and, over the course of the next few decades, bred only the most human-friendly individuals. The experiment began in 1959, and by the late 1970s—10 generations later—Belyaev’s foxes were exhibiting the desired behaviour, showing affection toward humans in a manner eerily reminiscent of dogs.
At the same time, however, the foxes acquired a host of unanticipated and unintended physical characteristics that distinguished them from the source population, such as floppy ears, turned-up tails, piebald coats, and wider faces, among other traits.
None of these physical characteristics were selected for, but Belyaev believed these traits were tied to the selected behavioural change (i.e. tameness), which somehow influenced the rise of unexpected traits.
The Russian Farm-Fox Experiment, as it’s now called, has since been used by biologists to showcase the sweeping influence of domestication on a species. It also invigorated a term used to describe the phenomenon: domestication syndrome, as these sorts of physical changes have been documented in other domesticated animals, such as dogs, horses, and cows.
But as a new opinion paper published in Trends in Ecology & Evolution points out, a critical part of this story isn’t actually true: the original foxes used in the experiment weren’t actually taken from the wild.
Moreover, and perhaps more controversially, the authors, who include Elinor Karlsson, a biologist from the University of Massachusetts Medical School, and Gregor Larson, a paleogeneticist from the University of Oxford, contend that domestication syndrome is a half-baked concept that’s probably not even a real condition.
That Belyaev’s foxes weren’t originally wild seems to be the case. The authors provided evidence—much of it already publicly available—showing that Belyaev acquired the foxes from Soviet fur farms, which in turn had acquired their foxes from Canadian breeders, specifically fox farms in Prince Edward Island. Canadian entrepreneurs had been domesticating foxes since the late 19th century, selecting for both appearance and behaviour, according to the paper. So by the time Belyaev got his hands on them, these foxes were already going through domestication.
And in fact, Belyaev himself admitted as much, describing the founding population as “fur-farm foxes,” but because he referred to them as “wild controls,” he unintentionally created a misconception.
“The story we’d heard was that the Russian scientists had started with a wild population of foxes, selectively bred the least fearful foxes, and as a result of that selection, also gotten foxes with white spots, curly tails, and other changes,” Karlsson told Gizmodo. “But up in Canada, they had foxes that were not fearful and already had white spotting (we don’t know about curly tails)—decades before the project started. And then we found the fox project in Russia didn’t start with wild foxes, but with fur farm foxes originally from Canada. It totally changes the way I think about cause and effect in the project,” she said.
Karlsson said her team concluded that Belyaev was continuing a domestication process that had already started many decades earlier in Canada.
“We’re not saying they are indistinguishable,” said Karlsson of Belyaev’s later generations of foxes and the founding group from the Soviet fur farms. But she said her team doesn’t think that “any of the changes since the Russian project started would suddenly qualify the foxes as now being ‘domesticated.’”
It’s entirely possible, she said, that Belyaev created “even more friendly foxes,” but even before this experiment began, “we know the foxes were already ok being around humans—and some of them were pretty happy about it, at least according to the photos and stories—and were breeding in captivity, which are the essential elements that are used to qualify them as being domesticated.”
But the larger issue described in the paper is that there “isn’t really any good definition for domestication,” Karlsson said, which “made this hard to write about!”
Karlsson and her colleagues argue against the existence of a so-called domestication syndrome, which they define as a “suite of behavioural and morphological characteristics consistently observed in domesticated populations.” The term was coined by botanists in the early 20th century, but Charles Darwin hinted at its existence in his 1868 book, The Variation of Animals and Plants Under Domestication. It was eventually applied to mammals in the 1980s, and its usage “has risen dramatically” since the 1990s, according to the new research.
Upturning the concept is a big deal, because domestication syndrome has inspired many related ideas, including the neural crest hypothesis and the pedomorphosis hypothesis. The neural crest hypothesis suggests neural crest cells—a specific class of stem cells—are a common factor in influencing the biological cascade that leads to multiple unanticipated physical changes in a species.
The pedomorphosis hypothesis, also known as neoteny, suggests some of the changes introduced by domestication have a distinct juvenile quality and that selecting for tameness and other attributes effectively maintains a species, or at least some of its attributes, at an underdeveloped level. (In fact, some scientists argue that humans have tended to select mates with more juvenile features, leading to the hypothesis that humans exhibit certain features consistent with self-domestication.)
The authors looked to various domesticated mammalian species in order to further scrutinize the concept of domestication syndrome. In addition to domesticated foxes, the authors examined the characteristics of other species of domesticated animals, including dogs, cats, goats, pigs, rabbits, rats, and mice.
The researchers charted their anomalous traits, such as shorter jaws, curled tails, drooping ears, changes in coat colour and patterning, earlier sexual maturation, decreased brain size, and other attributes typically associated with domestication syndrome. Their comparative analysis revealed many gaps and inconsistencies among the species studied.
Their main complaint is that domestication syndrome has no standard definition that is applicable to all domesticated species. “These hypotheses assume that the domestication syndrome exists, but with little supporting data,” wrote the authors. “The defining characteristics vary widely and have not been observed in most domesticated species. Many studies fail to distinguish traits that accompanied domestication from those only in modern breeds, and some traits are reported anecdotally without any accompanying frequencies or measurements.”
The researchers devised a list of three essential criteria consistent with their interpretation of domestication syndrome, namely:
1. Onset: A trait must appear…in conjunction with the onset of selection for tameness.
2. Frequency: A trait must be significantly more common in the selected population.
3. Association: A trait must be associated with tameness in individuals, not just at the population level
When the researchers applied these criteria to domestication syndrome, they were “unable to identify a single species for which all three criteria were met.” The authors concluded that the Russian Farm-Fox Experiment is overstated as a model for understanding the effects of domestication, while adding that traditional conceptions of domestication need to be re-evaluated and re-defined.
“Rather than focus on the domestication syndrome, we should instead consider how domesticated species have changed, and are still changing, in response to human-modified environments,” wrote the authors. “This effort will provide a robust framework to investigate the cultural and biological processes that underlie one of the most important evolutionary transitions.”
In terms of what’s happening to the animals, “a lot of it might just be something we call ‘genetic drift’, which basically comes down to random chance,” Karlsson told Gizmodo.
Perhaps unsurprisingly, the new opinion paper is not going over well with some biologists.
David MacHugh, a professor of functional genomics at University College Dublin, wasn’t directly involved with the research but said he discussed Belyaev’s experiment “at length” with Larson, a co-author of the paper, prior to publication. As MacHugh told Gizmodo in an email, he was “convinced” by the team’s arguments regarding the Canadian provenance of the source population. As for the authors’ takedown of domestication syndrome, he was less persuaded.
“It is important to note that as data accumulates from genome-scale functional and comparative analyses of domestic animals and their wild [ancestors], we should eventually have sufficient data to fully test the domestication syndrome hypothesis,” said MacHugh.
He also said it should be possible to figure out if the biological basis of domestication syndrome is tied to genetic disruptions, or “perturbations,” that affect the development of various tissues derived from neural crest cells.
In other words, MacHugh believes the jury is still out on the neural crest hypothesis, but future genetics research will be able to finally settle the score. What’s more, he believes ongoing research into ancient DNA will allow paleogeneticists to better chronicle the history of domestication, pointing to a new study he co-authored on this exact subject.
Adam Wilkins, an evolutionary biologist from Humboldt University in Berlin, was less charitable, saying the opinion piece was “deeply problematical.”
As Wilkins told Gizmodo in an email, he’s probably biased, since he was part of the team that “put the term ‘domestication syndrome’ on the map,” with respect to mammals (the term had previously been used for plants), in reference to a 2014 paper he co-authored with biologists Richard Wrangham and Tecumseh Fitch.
“The root of our disagreement lies, I think, in that we mean something different by ‘syndrome’ than they do,” Wilkins told Gizmodo. “They seem to believe that something can only be called a ‘syndrome’ if the affected individuals all display the exact same set of traits. Whereas we argue that if domestication is accompanied by a range of unselected traits, which might differ somewhat but often overlap, it counts as a syndrome.”
Wilkins said he and his colleagues never claimed the existence of an identical set of traits across all domesticated mammals, which he says is implied in the new opinion piece.
“Furthermore, under our neural crest cell hypothesis—which the authors refer to briefly, twice, but do not explain or discuss—what you see is exactly what the hypothesis predicts,” said Wilkins.
Specifically, that reductions in neural crest cells—the result of different mutations in the large set of neural crest cell genes—produce a “range of affected traits,” he said, pointing to a recent Development and Evolution paper he wrote that provides more colour to this claim.
Wilkins also took great exception to the authors’ assertion that the two fox populations—the farm-fox population that Belyaev bred from and the foxes he produced—were essentially the same.
“No!” wrote Wilkins. “This shows a lack of understanding of what an evolutionary process is. Belyaev clearly increased the frequencies of those mutant alleles [alternative forms of a gene] and brought them together and with that, he created a new population with distinctive properties. Now, that is evolution, as Belyaev recognised but [the authors] do not,” he said.
This point is consistent with a 1999 review of the Belyaev experiment written by biologist Lyudmila Trut from the Institute of Cytology and Genetics of the Siberian Department of the Russian Academy of Sciences. In the paper, Trut describes the “remarkable transformations” seen in the foxes over a 40-year period, in an experiment that, at the time of her paper, involved some 45,000 foxes and somewhere between 30 and 35 generations.
“The founding foxes were already tamer than their wild relatives,” wrote Trut. “Foxes had been farmed since the beginning of this century, so the earliest steps of domestication, capture, caging and isolation from other wild foxes had already left their marks on our foxes’ genes and behaviour.”
Despite this, the breeding program produced “an array of concrete results,” she wrote. The foxes “are unusual animals, docile, eager to please and unmistakably domesticated. When tested in groups in an enclosure, pups compete for attention, snarling fiercely at one another as they seek the favour of their human handler.”
What’s more, the unexpected physical changes weren’t seen until around the eighth to 10th selected generations, as Trut wrote:
The first ones we noted were changes in the foxes’ coat colour, chiefly a loss of pigment in certain areas of the body, leading in some cases to a star-shaped pattern on the face similar to that seen in some breeds of dog. Next came traits such as floppy ears and rolled tails similar to those in some breeds of dog. After 15 to 20 generations we noted the appearance of foxes with shorter tails and legs and with underbites or overbites.
Wilkins raised several other issues with the new paper, including the authors’ failure to define domestication after claiming that Belyaev’s experiment wasn’t a true example of domestication. He also wasn’t happy that they looked at just seven domesticated species instead of the 26 documented in the scientific literature.
“If they had, they would have seen a lot more domestication-associated traits… including many that almost certainly were not deliberately selected by breeders,” Wilkins told Gizmodo. “They want to attribute all these changes to selection, however; at least, that is what they imply at the end [of the article].”
Wilkins believes the array of anomalous traits seen in domesticated mammals—whether these traits are common or not across species—cannot be explained away so easily and that domestication syndrome best describes this biological phenomenon.
Deficiencies in our understanding of Belyaev’s experiment notwithstanding, the authors of the opinion piece raise a very good point—domestication syndrome as it’s understood today is not a fully formed concept. But that doesn’t mean it doesn’t exist.
Pleiotropy, in which a single gene influences multiple traits, is very much real, so it makes sense that you could get some unexpected surprises—like floppy ears when you were only selecting for friendliness—by messing with a multifunctional gene.
It’s not immediately obvious that many of the unexpected physical characteristics seen in domesticated animals are truly the result of selection (whether those traits were consciously bred into them or not) or genetic drift. Moving forward, scientists will need to better elucidate the underlying cause of each identified trait.
And if these accidental byproducts or side-effects can be indisputably identified as being the unanticipated consequence of domestication, then we have something that can only be called one thing: a syndrome.