Dogs came from wolves, a very long time ago, and they belong to a family of animals called canids, along with wolves, coyotes, foxes and jackals. Dogs have lived with people for at least 14,000 years. Yet people still argue about when, where and how dogs were domesticated: scientists have worked out the time, place and ancestry of nearly every other domesticated animal, from sheep to chickens, but not dogs. Some archaeologists say there were domestic dogs by at least 15,000 years ago; how much earlier is argued, with some saying 30,000 years. Where dogs first appeared is hard to say: it could have been in the Middle East, China, Siberia or several places at once. Most genetic evidence says dogs emerged in the farming era, but some dog fossils are much older, dating as far back as about 33,000 years ago.
The scientist Dmitri Belyaev proposed that calmer wolves hung around people because scraps gave them steady food, and that as these calmer wolves interbred they became less afraid of people. It is an idea, not settled fact. In 2013 a study of 60 dogs from 14 breeds and 12 wolves found that dogs have more copies of AMY2B, a gene for an enzyme that breaks starch into sugars, than wolves do. The researchers on that 2013 study said both brain changes and diet changes were probably needed for wolves to become dogs. A later, larger study found that some wolves also carry extra AMY2B copies, and that dingoes have two copies and Siberian huskies only three or four. A population geneticist said the later data make a case against starch digestion playing a key role in taming dogs.
The scavenger hypothesis says wolves that were less afraid of people lived near villages and ate scraps from garbage piles, and over time became dogs. Other ideas say people trapped wolf pups on purpose to train them as sentinels or hunters, or that wolves and people helped each other, with people learning to hunt by watching wolves. The older idea that Stone Age people risked their lives to steal wolf cubs as guards, workers or companions makes less sense, Belyaev and Trut argued, than the idea that calm wolves came to people. An evolutionary biologist who studies dog origins said there are as many scenarios for how dogs began as there are people working on it, because it was so long ago. One article says dogs worked for people as shepherds, guard dogs and hunters from the beginning of their domestication, and calls the pet-and-family bond a relatively recent thing.
A 2013 study found that the wolves it compared were all equally related to dogs, so none could be called the dog ancestor. The authors suggested an unknown, possibly extinct, wolf was the source. Interbreeding between dogs and wolves after domestication made some wolves look more closely related to dogs than they really are, which blurs the search for where dogs began. A 2020 study sequenced 72 ancient wolf genomes from Europe, Siberia and North America spanning the last 100,000 years. That study found dogs are more closely related to ancient wolves from eastern Eurasia than to western Eurasian wolves, which suggests dogs were domesticated in the east. Some dogs also carry ancestry from Middle Eastern wolves, so researchers suggested dogs might have been domesticated twice, or that Asian dogs interbred with wild wolves in the Middle East as they moved west. Many experts assumed the wolf population that gave rise to dogs went extinct because no living wolves match dog ancestry, but wolf genes blend across continents, which could hide the match.
Wolves at the Wolf Science Center in Austria, after living and working with scientists for seven years, still keep an independence of mind and behavior that dogs do not. Eleven pure-bred wolves raised by people did not make regular eye contact with their owners. Among wolves, eye contact is a threat, so wolves rarely look directly at each other. A Duke study of 44 dog puppies and 37 wolf puppies, 5 to 18 weeks old, tested how well puppies read human gestures. The dog puppies followed a person's pointing at hidden food without training, but the wolf puppies did no better than chance. The wolf puppies had near-constant human contact from just after birth, and were hand-fed, so the difference was not due to a lack of time with people. The study's authors said the results support the idea that domesticating dogs from wolves changed the way dogs think.
Dmitri Belyaev asked Lyudmila Trut to lead an experiment with foxes, which are closely related to dogs, to see how domestication unfolded and which evolutionary forces were at work. Trut worked on the fox experiment from 1959, and continued it after Belyaev died in 1985. It began hidden inside a larger program to breed better foxes for the fur industry. The idea they tested was that domestication syndrome, which means floppy ears, curly tails, youthful faces and mottled fur, came from always breeding the calmest, friendliest animals in the early stages of domestication. Many of the tame foxes kept childlike traits into adulthood, like floppy ears, curly tails and baby faces, that their wild cousins outgrew. After about 15 generations of selective breeding, the experiment had foxes that could live with people. The results supported Trut and Belyaev's ideas and changed how scientists understand domestication.
In 2015 a study found that when a dog gazes at its owner, the owner's oxytocin level rises, which prompts more social contact. Owners who got more gazes looked back, touched and talked to their dogs more, and dogs that got the most attention in return also had the biggest oxytocin spikes, so each raised the other's oxytocin in a loop. A researcher suggested that dog evolution probably selected for eye-gazing, a behavior that gives people a body response that helps them bond. Brain scans show that some of the same brain regions become active when human mothers look at pictures of their children or their dogs. Other scientists said it was too early to say the gaze loop evolved with domestication, because the same loop might also form between puppies and any caregiver, including a dog mother.
About 12 percent of the air a dog breathes in detours into a recessed area at the back of the nose that is only for smelling, and the rest goes to the lungs. In that smelling area, odor-laden air passes through scroll-like bony structures called turbinates, and receptors in the tissue that lines them recognize odor molecules by their shape and send signals to the brain. When a dog breathes out, the air leaves through slits at the sides of its nose, which helps bring in new smells and lets dogs sniff almost continuously. A dog's nostrils each reach only a small area, which helps it tell which nostril a smell came in through and so work out where the smell is coming from. Dogs also have a vomeronasal organ, also called Jacobson's organ, a second smell system we do not have, built for chemical communication.
In a study in Norway, a hunting dog with its head high into the wind sniffed in a continuous stream of air for up to 40 seconds, across at least 30 breaths. In one study, dogs led at right angles to a trail that a person had just walked could tell which direction the person went from as few as five steps. A dog researcher wrote that a dog could detect a teaspoon of sugar in a million gallons of water, about two Olympic-sized pools. The vomeronasal organ's pheromone signals go to their own part of the brain, so they do not get mixed up with ordinary smells. A wet nose helps a dog smell because scent particles stick better to damp surfaces.
A wagging tail or a crouching body does not always mean a dog is friendly. One study showed dogs videos of other dogs wagging their tails: dogs that watched a dog wag to the left had faster heart rates and looked anxious, but those that watched a dog wag to the right stayed relaxed. The same research team had found that dogs wag to the right when they feel positive emotions and to the left when they feel negative ones; the researchers do not think the dogs are necessarily meaning to tell other dogs. Before you pet a dog, ask its owner if it is okay. If a grown-up says you can, let the dog sniff your closed hand first. If a dog you do not know comes up to you, stay still and quiet instead of running.
Brachycephalic dogs are flat-faced breeds with shorter skulls, like pugs and bulldogs; people bred them to have these features and found the look attractive. BOAS, short for brachycephalic obstructive airway syndrome, is a chronic disease of short-skulled dogs in which narrowed airways cause noisy breathing and can limit exercise, sleep and coping with heat. From 2021 to 2024 a team in England studied nearly 900 dogs of 14 short-skulled breeds, and found 12 of the 14 breeds had some breathing abnormality. In the study, 89 percent of Pekingese and 82 percent of Japanese Chin were affected by BOAS, rates like those of pugs, French bulldogs and bulldogs. Risk of BOAS varied a lot by breed, and boxers and Staffordshire bull terriers rarely had breathing issues.
The researchers found dogs with wider, shorter head shapes were more likely to have BOAS. A study author said the link between muzzle length and BOAS risk is more complex than commonly assumed. From 2006 to 2016 the number of registered bulldogs in the U.S. went up by 60 percent and French bulldogs by 476 percent. In 2023 the French bulldog topped the American Kennel Club's list of most popular breeds. The most severely affected flat-faced dogs struggle constantly to get air, but many owners think this is normal for the breed and do not seek treatment. A veterinary professor argues that breeders and vets should be trained to grade the fitness of dogs before breeding to avoid producing offspring with the worst problems.
Words to know
- domestication: taming over time
- genomes: full gene sets
- hypothesis: a testable idea
- turbinates: scroll-like bones
- brachycephalic: short-skulled
