Unnatural selection
What have we done to our best friends? (13 minute read)
When my dog Donna was younger, she could run fast. She was also a bit less difficult with other dogs and could go to off-leash parks and make friends. I can still remember when she played a game of chase with a greyhound – with the greyhound chasing her. It didn’t come close to catching her, and it was trying. Running in a straight line, it would gain on her, but then she’d throw in a turn. She could take a corner like a sports car in the hands of a pro, and the greyhound would corner like heavy truck.
Another time, I watched her chase a rabbit. She almost caught it. The reason she didn’t wasn’t her speed – she was clearly faster. It wasn’t her cornering either, because she was every bit as agile. It was her technique. When she caught up to it, she ran straight past then made a lightning turn in front of it. She had no intention of catching the rabbit – she was trying to herd it.
Donna is the product of thousands of years of selective breeding – the process of evolution directed by human hands. For the last hundred or so years, that selective breeding has been on New Zealand farms. Donna is a heading dog, a type of farm dog I’d describe as mostly border collie. The heading dog isn’t considered a true breed, unlike the other New Zealand farm dog, the huntaway, but it’s instantly recognisable to people familiar with farm dogs in New Zealand. Her appearance is irrelevant, though. She was bred to work.
I’ve seen heading dogs work solo, but they are most often part of a team. This has been explained to me in different ways by different farmers, but they all gave the same basic principle. The huntaway’s role is to make the flock of sheep move, the heading dog’s role is to stop them moving, or get them to change direction. I’ve heard huntaways described as the motor or the accelerator, with heading dogs the steering wheel and brake. (If you want to see these dogs working, here is an episode of the quintessential New Zealand television programme A Dog’s Show. It occupied a prime time spot from the late 1970s until the early 1990s.)
Donna has been living in the city as a pet since she was a tiny pup, so when she has met sheep, she’s been confused. It’s obvious that she thinks she recognises them in some way, though. In the absence of sheep, she herds whatever she can. Most often, it’s my very patient and forgiving cats, Bonnie and Marbles.
Dogs were the first animals we domesticated, thousands of years before we domesticated any other animal or, for that matter, plants. There are differing opinions on how long ago that was, but it was probably around 15,000 years ago. At that time, humans were still hunter-gatherers – not that this means we weren’t manipulating our environment in various ways, but we didn’t have flocks of sheep to herd. Archaeologists suggest that dog domestication wasn’t an intentional process. Instead, as we modified our surroundings in various ways, certain animals were attracted to the modified environments. Many of these animals are considered pests, such as pigeons, mice and mosquitoes. Over time, some of these species became so well-adapted to the modified environments that they no longer survived away from human influence.
Dogs, cats and chickens are all thought to have evolved in this way. It’s quite a different process from the domestication of cattle and sheep, which was the result of people managing wild game populations. However, once all of these animals were living closely with us, we could take greater control of their breeding. Although it would be thousands of years before we would understand precisely what we were doing, people recognised that it was a good idea to breed from animals with traits we wanted, and not to breed from animals with traits we didn’t want.
The power of this simple idea is clear every time we see a dog. The diversity within this single species is astounding. Donna is a medium-sized dog, at just over 20 kg (although she still occasionally thinks she’s a lap dog). The smallest dogs, such as the chihuahua, can weigh less than 2 kg. Some of the heaviest, such as mastiffs, can weigh from 50-100 kg, while the tallest, such as Irish wolfhounds, are over 80 cm high at the shoulder.
But dogs differ in more than size. The diversity in shape is remarkable as well, from the long, pointy snout of the greyhound to the squash-faced pug, from leggy saluki to the low-slung dachshund, from the nearly-hairless Chinese crested to the powderpuff Samoyed, and even the dreadlocked puli. And that is only the diversity of appearance, there is a diversity in their minds as well. Donna is obsessed with chasing and herding, but can’t retrieve to save herself. If you throw a stick or ball for her, she just doesn’t get the idea of bringing it back, despite her obvious intelligence.
Donna is a great example of a dog bred to work at a specific job. She’s got the herding instinct, but it’s more than that. She’s tireless. Even now, as she’s approaching 12 years old, people think she’s a young dog. And she pays constant attention to what I’m doing, waiting for a signal, trying to anticipate what I’ll ask her next. When I was teaching her to roll over, for example, it got to a point where I’d ask her to sit, and then she’d sit, lie down, roll over and be up again for a treat before I could give her the second command. At times, it’s exhausting. She was never meant to be a pet.
For thousands of years, we have been shaping dogs to work for us. We had dogs for guarding, hunting, pest control, herding and transport (pulling carts and sleds). There seems no end to the jobs that dogs can do – they are now used as guide dogs, to warn of epileptic seizures, to encourage children to read, to find lost people, alive and dead, and to detect all manner of material from plant and animal products in passenger baggage, to drugs, explosives and invasive species. I taught Donna to find my car keys, and she generalised that to any set of keys even if I have never touched them. We also bred some dogs simply to be our companions.
For almost all of our shared history, we bred dogs mainly for their functions. I’m sure that people did select dogs for their appearance at times, but this wasn’t standardised. It was only in the 19th century that breed standards were developed, and dogs began to be classified primarily by their appearance. Around the same time, people began breeding and training dogs for the show ring in England. Most of what we think today about dog breeds is based on the approach of Victorian dog fanciers.
It may not be fair to blame all the problems of purebred dogs on this modern concept of dog breeds. However, it is now inescapable that our selective breeding of dogs has not served their best interests. We have breeds such as the pug, shih tzu, and French and English bulldogs, with flattened faces. The shape of their noses and nostrils means than many have difficulty with breathing, and they suffer from a number of health problems at much higher rates than dogs with more usual face shapes. We have the dachshund, with back problems at more than ten times the rate of other breeds. We have the wrinkly shar pei, with a host of skin problems, including an awful condition where their eyelids roll inwards, meaning that the surface of the eye is constantly scratched by the eyelashes.
There are two reasons that purebred dogs suffer far more than cross-breeds. The first is that some of the traits we have specifically bred for, such as squashed noses and excess skin, are harmful to health in themselves. If these dogs were still working dogs, it’s unlikely that the traits would have been allowed to become so extreme. The second is because many purebred dogs are inbred. To understand what this is, and why it’s a problem, it helps to understand more about how traits are inherited.
I’ve written previously about genes and DNA, so I won’t go into detail on that here. Instead, I’ll give a summary, and move onto the issue which I want to cover – the inheritance of traits. The short version is that DNA is a complex molecule where the order of the components works as a code. The code tells our cells to make different proteins which perform various functions in our bodies. Sections of the code linked to different functions or traits are known as genes. When I was at school, we learned about genes for things like flower colour in peas and eye colour in humans, and tended to focus on one gene at a time. But many genes can go into complex traits.
There are also genes for less visible traits. It’s pretty clear that Donna’s constant attempts to herd other animals is genetic, because she hasn’t been on a farm since she was a tiny pup. While most of us probably don’t think of human personalities having much to do with genes, it’s also clear that they play some role with us too. I’m a good example of this, as I was adopted at birth yet share obvious personality traits with my biological relatives.
Inside our cells, almost all of us have 46 individual molecules of DNA, which are known as chromosomes (a few people have an extra chromosome, such as in Downs syndrome, or a missing chromosome such as in Turner syndrome). When our cells reproduce, for example a new skin cell is created, they do so by creating a copy of themselves (I’ve linked to a video showing the process here). During this copying process, known as cell division, each of the 46 chromosomes creates a copy of itself. Copying the chromosomes means copying the DNA code, so the new cells contain the same code. This process is broadly the same in every living thing. The main difference is that different animals and plants have different numbers of chromosomes. Dogs have 78 chromosomes. Pine trees have 24. Bread wheat has 42, although the situation is complex, with some varieties having different numbers, such as durum wheat, which has 28 (I may go into this in more detail another time, as it’s also relevant to understanding genetics). Bacteria, though, have only one chromosome.
When an animal or plant grows, it does so through this process of cell and chromosome copying. But when an animal or plant reproduces, the process is different. I’ll focus back onto dogs for the sake of simplicity, because the terminology isn’t the same for every living thing. In producing eggs or sperm, the chromosomes reproduce in a different way.
You might have noticed that all the numbers of chromosomes I mentioned, apart from for bacteria, are even numbers. That’s because chromosomes occur in pairs. When cells divide to produce eggs or sperm, the they end up with only half of the number of chromosomes, so for dogs, which have 78 chromosomes, the eggs or sperm will have 39. The process is considerably more complicated than when cells copy themselves. I won’t describe the full process, but there’s a good video here if you want to know more. But there are two important things to know. The first is that this process means that each pup gets half their chromosomes from one parent and half from another, and the same applies to humans and other mammals). The second is that there’s a step early in the process where the pairs of chromosomes swap some material. This means that chromosomes aren’t passed on unchanged from generation to generation and results in much more variation than there would be otherwise.
The pairing of chromosomes means that every cell has two copies of every gene. This is crucial, because accidents can happen when copying chromosomes – such accidents are known as mutations. Sometimes, these mutations don’t matter. Very occasionally they are beneficial – it is these beneficial mutations which are the basis for evolution. But, often these mutations are damaging.
If a cell had only one copy of every gene, then every mutation could potentially affect every cell. It usually won’t, because cells perform different functions depending on which genes are switched on or off – if it was a mutation which affected the eyes, for example, the cells made to produce skin, or muscle or blood cells would be fine. But, because there are two copies, in many cases the mutation can just sit there causing no problem while the healthy gene does its job.
It might be easier if I use a specific example, so I’m going to talk about a male dog with a mutation which affects certain cells in the eye. However, this dog has only one copy of the mutation and the other gene is healthy. So he has good eyesight into old age. Because one copy of the gene was healthy and one damaged, when this dog produced sperm, 50% of the sperm carried the mutation and 50% did not. As long as this dog bred with a female which didn’t have the mutation, everything was fine. Over many generations, this mutation was passed on but never seen, because each dog only had one copy.
But what happens when a dog which has one copy of the damaged gene breeds with another dog which carries the damaged gene? Because each parent has a 50% chance of passing on the mutation, some puppies will get lucky. One in four will have two healthy eyesight genes and have good eyesight until old age. Two out of four will have a single copy of the mutation and the other gene will be happy, and they will also have good eyesight until old age. And one in four will get two copies of the mutation, and will slowly begin to go blind at some point between the ages of three and nine.
This kind of mutation can spread unnoticed for many generations, masked by the healthy genes. But if two dogs which are related are bred together, there’s a much greater chance that some of the puppies will end up with two copies of the mutation. Because there has been a lot of breeding between closely-related dogs in the process of developing modern breeds, many diseases linked to this kind of mutation show up.
And our dogs pay the price.
There is a lesson in what we have done to our dogs. We have developed breeds with many health problems using conventional breeding techniques. These same breeding techniques resulted in dogs which are superbly adapted to their roles, such as the heading dog which is so effective as a herder. These techniques also resulted in all of our food crops – wheat, maize, apples, grapes, cabbages – and our livestock. The techniques themselves aren’t the problem, but sometimes we make poor decisions in what we use these techniques for.
This is an important question when it comes to new genetic technologies. There are two separate questions we need to consider. The first is the question about the risk of the technology, and I will look into this over the next couple of months. The second is the question of what the techniques are used for, and whether we are making good decisions. Perhaps we can learn from our best friends, and avoid making some of the same mistakes.
I’ve noticed a huge increase in small dogs being led around the streets