CHARLES DARWIN was fascinated by the many breeds of domesticated birds and mammals. His cataloguing of the unusual and sometimes bizarre varieties of pigeon was more than a hobby: he was an avid pigeon breeder, and devoted many pages of On The Origin of Species to documenting how readily the birds change their form when selected for unusual traits such as webbed or feathered feet.
The study of domesticated animals was a key pillar in Darwin鈥檚 argument for natural selection. The changes he observed in pets and farm animals showed that artificial selection by humans 鈥� whether for webbed feet in pigeons or more milk in dairy cows 鈥� can quickly produce heritable changes from the 鈥渨ild type鈥� state. This provided a compelling argument that analogous 鈥渘atural selection鈥� by Mother Nature could produce similar changes and eventually lead to new species.
Darwin didn鈥檛 limit himself to pigeons. Through correspondence with hundreds of breeders, he compiled information about virtually every domesticated species of the time, from chickens and ducks to dogs, cats, pigs, cows and horses. He published his research in 1868 in the massive two-volume Variation in Animals and Plants Under Domestication, which even today remains the most comprehensive work on the subject.
Advertisement
Darwin鈥檚 detailed research uncovered a remarkable regularity, which has recently been dubbed the domestication syndrome. In all domesticated mammals, a bizarre collection of traits shows up time and again. Besides a general docility, Darwin noticed that coloration is modified (for example, black and white coats appear), teeth and brains get smaller, and snouts shorter. In many species, the tail may become curly or reduced, or ears flop over. Why should this unusual set of features appear together in different species that were domesticated at different places and times? Although Darwin speculated that some traits might have been specifically selected for (black-and-white coloration, for instance, might make lost livestock easier to find), he never came up with a convincing solution for the whole set of traits.
Recently, together with biologists Adam Wilkins and Richard Wrangham, I proposed (Genetics, vol 197, p 795). Our hypothesis hinges on the fact that virtually all of the traits involved in the domestication syndrome are derived from the same source: an unusual type of cell called the neural crest.
Neural crest cells arise during early embryonic development, when the brain and spinal cord are forming. These cells originate on the embryo鈥檚 back and migrate to form the adrenal glands and parts of the nervous system, along with pigmentation cells and major portions of the skull, teeth and ears.
Our hypothesis suggests that selection in any newly domesticated species is focused on tameness. Animals that become fearful and agitated when encountering humans are unlikely to breed in captivity. Fearful animals are also more likely to bite, kick or otherwise injure their human caretakers. Thus docility and a lack of fearfulness are among the most important traits to appear when a wild species is first domesticated.
How does tameness arise, physiologically? A key change is that the adrenal glands and sympathetic nervous system, which are jointly responsible for the 鈥渇ight-or-flight鈥� response to scary events, mature late. When young animals are first exposed to their human custodians, they do not have a full-blown physiological capacity to feel afraid. Thus they remain relatively calm during these encounters, and by the time their fear response matures they have already been habituated to humans. This is precisely what it means to be 鈥渢ame鈥�.
Wolves, for example, have a brief window of time after their eyes and ears start working before they are capable of mounting a mature fight-or-flight response. Anyone wishing to tame a wolf must expose it repeatedly to humans during this period, which lasts until the age of about 1陆 months. In contrast, many experiments have shown that this 鈥渟ocialisation window鈥� lasts until the age of 4 to 10 months in dogs, depending on the breed, after which time dogs exposed to people for the first time will remain fearful regardless of how much they interact with humans.
Now comes the interesting part. We hypothesise that the late maturation and general under-functioning of the adrenal glands and sympathetic nervous system 鈥� leading to docility 鈥� derives from reduced numbers and delayed migration of neural crest cells in the embryo. Because the cells are the precursors of teeth, pigmented skin, snouts, ears etc, these are also smaller or develop late. Rather than humans selecting for black-and-white coloration, we suggest that pigmentation changes and other traits of the domestication syndrome are just unintended side effects of selection for tameness.
鈥淧igmentation and other visible traits are side effects of selection for tameness鈥�
Strong evidence in favour of our hypothesis comes from experimental studies of domestication. The most famous were carried out with foxes in Siberia over many decades, beginning in the late 1950s. The researchers attempted to create a new domesticated species by selecting only for tameness. They started with unselected, fearful foxes, and carefully tested their young for fearfulness and friendliness. Only the friendliest offspring were chosen to breed.
After a surprisingly short time 鈥� less than 10 generations 鈥� the Siberian team bred a strain of foxes that is amazingly tame and unafraid of humans (see photo above). What鈥檚 more, these foxes showed most of the other traits of the domestication syndrome, such as reduced and delayed adrenal function and a longer socialisation window (as required by our hypothesis), plus pigmentation changes, floppy ears and shortened snouts. These experiments have been repeated in rats and mink, with similar results.
Tameness is the only attribute selected for in this work. Nonetheless, the traits of the domestication syndrome pop up much more frequently than expected if they were random.
There is other evidence. We predicted that neural crest development changes should be pervasive in domesticated species compared to wild counterparts. A study published just last month confirms this for multiple neural crest genes in domesticated cats ().
Our hypothesis leaves several questions open. The most obvious concerns the brain: the neural crest cells make no direct contribution to the central nervous system, so it is unclear whether cognitive changes in domesticated species can be fully explained by the same mechanism, or whether interactions between the neural crest and the developing brain are important in brain reduction as well.
Further research is needed, but our hypothesis is consistent with available data. We think it finally provides a coherent explanation for the characteristics of domestication that have puzzled biologists since Darwin.
This article appeared in print under the headline 鈥淗ow pets got their spots (and floppy ears)鈥�
