My search-fu may be weak today, so apologies if this has been asked and answered before.
I’m not clear exactly on how dog (or any animal with lots of breeds, I guess) breeding works–when I was growing up, my folks had a couple of “pure-bred” basset hounds. They had AKC papers that, as I understand it, permitted them to (if they cared) trace the lineage of the dog back n number of generations. It would seem to me that strict adherence to this would mean that the gene pool is constantly narrowing, unless new blood can be added. Can it? If I wanted to create a new basset hound from existing non-basset dog breeds, and succeeded in producing something that looked the same, I presume there would still be some kind of folderol about it not being the same thing, like there are with projects that presume to “breed back” to extinct species?
I started thinking about this looking over the Wikipedia article on tamaskan dogs, which says “The last foundation dog was added to the breed in 2006, when the doors were closed on any new bloodlines.”. Can somebody familiar with dog breeding explain to me what this means? I would think that without the regular infusion of new stock, the dog becomes progressively more inbred. That’s not necessarily a problem, I suppose, if the original dogs don’t have any marked defects. And since, modulo certain instances of undesirable traits cropping up in heavily-bred dogs (which I would imagine have more to do with unscrupulous breeders than inbreeding per se), breeds don’t seem to be collapsing, I presume it’s not a problem?
So my question is: does breed standardisation and lineage tracing inherently lead to shrinking gene pools? Considering everything you hear about “hybrid vigor”, is this a problem for the health of the breed? Or is there some mechanism, bureaucratic or biological, that keeps a breed–if not its individual members–stable?
Yes, it’s a huge problem. Many breeds suffer from health defects of various kinds that are made more prevalent through inbreeding. And it’s not just dogs and the artificial breeding they go through. Cheetahs went through a near-extinction event in relatively recent history that left a small surviving gene pool, and they’re suffering in some of the same ways because of it.
Right, I understand that there are problems that have been created or exacerbated by unscrupulous and overzealous breeding… I guess my real question is if there is an inherent problem with the system of dog breeding (that is, am I right in thinking that the gene pool is constantly narrowing?), or whether–if handled responsibly, no small order–breeds created under this system could endure indefinitely.
That is why responsible dog breeders go to such lengths to select their pairings. I know a number of breeders, including one of a particularly uncommon breed (nothing so rare as the tamaskan), and they work internationally to get the right pairings. Artificial insemination, of course; nobody would send a dog from Minnesota to New Zealand for a breeding. But you will often hear of the stud being from a different country from the dam, to widen the gene pool.
Puppy mills just swap 'em around. Genetic problems? Sell it anyway, unless it’s too obvious, and then you have a pond out back, don’t you?
Its a massive, massive problem. The RSPCA (British equivalent of ASPCA) recently and the BBC withdrew their support of Crufts (the premier British pedigree dog show) because of it …
I speak as a geneticist that has no particular expertise on dog breeding.
Inbreeding will always have these sorts of problems. Purebreds are just that - inbred lines. Now, careful crosses within a breed can minimize some of these problems, but even then the gene pool for each breed is very small, compared the the gene pool of all dog breeds.
Outcrossing a purebred will solve most, if not all, of these problems. This is probably a major reason for the popularity of mixed breeds, since the genetic defects of one purebred parent won’t line up with the genetic defects of the other, and the resultant offspring will be healthy. Some breeders are trying to outcross their purebred animals, by bringing in healthy genes from other breeds, and then selecting for the traits of the original purebred. This ideally would result in essentially the same breeds we have already, but without the genetic defects.
Here’s a specific case, where breeders are attempting to remove a specific genetic defect from Dalmations by backcrossing to similar breeds. The Dalmation Club of America refused to register these as purebreds, even though the resulting dogs had all of the outward traits of dalmations, minus the genetic defect. Recently, they’ve at least decided to reconsider.
IMO, purebreeding is a bad idea. Mutts are almost always healthier dogs. Now, I can understand breeding for specific traits, but the desire for a “pure” bloodline seems antiquated and unnecessary, and in the worst cases bordering on cruelty. A breed really should be outcrossed every several generations at the very least. And canine morphology can only be pushed so far before the changes are pathological.
We have a Guide Dog breeder (a Golden) and yes there are problems. The bloodlines of the guide dogs go back 60 years. They try to deal with the problem by getting studs from other clubs. The last time our dog went into season they tried to impregnate her with semen from a dog from Australia, but it didn’t work. The time before that it was a dog from Ireland, which did.
There have been some attempts at cross-breeding within the guide population. Her second litter was a Golden/Lab cross, but it didn’t work very well.
The problem was that they’ve been bred for the qualities that make them good guides, so they can’t introduce traits from outside or risk getting dogs that won’t make it through training.
BTW, they have tissue samples for every dog who has ever been the program, so when the dog genetic code gets cracked, they are going to be able to correlate genetic disorders with actual DNA from the dogs having them.
Inbreeding in and of itself doesn’t create problems - it just makes it more likely that harmful alleles will be expressed. In theory, you could have a perfectly healthy inbred population. In practice, it’s all but impossible to actually do it. If you wanted to create a new breed that didn’t suffer from this problem, you could do it by somehow weeding out every negative allele in your breeding stock, or by having an exceptionally large number of “founders”. Obviously, it’ll be tricky to get a large number of distantly-related or unrelated animals to look similar enough to be a “breed” and still breed true.
So I’d say, from a practical perspective, it’s a pretty inevitable problem.
Inbreeding can’t produce any new traits, just cause previously-hidden recessives to crop up. The theoretically ideal purebred is homozygotic on every gene (other than the sex chromosomes, I suppose), so there are no hidden recessives, and breeding two such animals will just get you another one exactly like its parents.
In the real world, of course, this is an impossible ideal, so you will occasionally get unexpected defects showing up. A responsible breeder will cull out those defects (killing or at least neutering the animals that show it), so the defective gene gradually dwindles out of the gene pool, but of course, not all breeders are responsible.
Inbreeding can’t produce any new traits, just cause previously-hidden recessives to crop up. The theoretically ideal purebred is homozygotic on every gene (other than the sex chromosomes, I suppose), so there are no hidden recessives, and breeding two such animals will just get you another one exactly like its parents.
In the real world, of course, this is an impossible ideal, so you will occasionally get unexpected defects showing up. A responsible breeder will cull out those defects (killing or at least neutering the animals that show it), so the defective gene gradually dwindles out of the gene pool, but of course, not all breeders are responsible.
A large part of the problem (as I understand it) is breed standards. Border collie breeders have been fighting the AKC to prevent the breed from being listed because most of them are breeding them as working dogs, and once a “breed standard” is decided on, the dogs are mostly bred for looks and demeanor, which (as the OP notes) can introduce a persistent genetic defect into the breed.
I’d say that even theoretically, you couldn’t avoid these problems in an inbred population.
For humans, it’s been estimated that every individual has about a dozen novel recessive allele which would be harmful if expressed. The same holds for dogs. Even if you could find an individual or small group that completely lacked harmful recessive alleles (probably impossible), their progeny would pick up a few from new random mutations, and several generations down the road you’d have the exact same tendency towards genetic diseases.
You’d really need a very large “founder” population, and enough genetic variation that you probably couldn’t get the desired traits consistently.
Sorry, I have to refute this. If you cross a purebred Rottweiler with a purebred Labrador, and both parents carry genes predisposing them to hip displaysia, then the mongrels you’ve created are just as likely to suffer from the condition.
And ‘mixed breed’ dogs are popular because they’re easily available, bred in huge numbers by people who have no idea what they are doing, but think their dog is cute, and wouldn’t it be fun to have puppies? (Sorry, bit of a personal soapbox…!)
Some breeds, like my own favourite, basenjis, are very open to new blood being introduced. AKC registered basenjis are all descended from a limited pool, around 30 individual dogs, imported from central Africa in the 30s. In the 1980s, American basenji breeders went to Sudan and brought back thirteen dogs ‘of basenji type’ which the AKC allowed into the breed registry. The new blood has had immediate positive effects on the breed, as well as introducing a new coat colour, brindle. More here.
Thanks for the link about the Dalmatians, lazybratsche - lets hope some lessons are learned there.
So, modulo backcrossing or a willingness on the part of breeders to admit stock from outside the bloodlines, the breeds of dogs as we know them today are inherently doomed to extinction? Does anyone know (this isn’t directed specifically at you; I know you said you’re not a dog breeder yourself) how recent the notions of bloodlines and registries–purebred dogs, in other words–are? Before the popularisation of eugenics, evolution and genetics in the last hundred and fifty years or so, would it’ve been sufficient if my long-eared, long-legged floppy dog just looked enough like a bloodhound?
That’s very interesting–on both counts. From a layman’s perspective, as far as I can tell a “breed” is just the sum of some dimensions of dogness–colour, shape, behaviour, size, etc–and if you recreated a dog that had those traits you’d basically be creating the requisite breed. On the other hand, I suppose I can appreciate that even if you put the same parts, the same kind of case and the same operating system together there are people who would reject what you’d built as being a “Macintosh”.
That’s something else I was kind of thinking about. I have friends who have working dogs–Border collies and Aussies, mostly–who have been… at best nonplussed and at worst resentful of the efforts of the AKC to register them. Anecdotally, it seems to me AKC Borders have substantially less-varied body types than the working dogs I’ve seen, and I seem to percieve differences between AKC greyhounds at the dog shows I’ve visited and the retired racing dog my parents used to have.
That makes a lot of sense, and I agree that would seem like a good way to keep the breed healthy. Looking at the website, it looks like other dogs were brought in prior to the 80s. I wonder if the AKC or the breed organisation authorises that on a dog-by-dog basis, or what?
A lot of people kind of snicker at some of the newer designer dog mixed breeds–you know, of the labradoodle (poodle and labrador), schnoodle (poodle and schnauzer), cockapoo (poodle and cockatiel) variety. I wonder if this is slightly misplaced–or if the high demand for dog hybrids has just compounded the problem when they’re drawn from already heavily-bred purebreds. Come to think of it, especially with cockapoos–I’m to understand “spaniel rage” has been discredited, but all the cockers I’ve met have been completely insane and a couple of owners told me they thought it was genetics.
I think you mean poodle and cocker spaniel. A cockatiel’s genes bred into a dog would certainly be quite an outcross!
I can’t help but wonder if part of the designer dog mixing comes from people thinking up Xtra Qte names, and then figuring out which dogs to breed together.
I was fascinated by the Dalmatian website. I didn’t know that there were lemon flavored (OK, lemon SPOTTED) Dalmatians.
I was debating whether or not to add some aside about not really thinking they were bird-dogs but I hoped my reference to cockers in the following sentence would make it clear I did understand the cross… in the event I admit myself to finding the whole thing kind of silly. I understand that it’s not quite like people say, you know, “hey, you don’t have a German Chusky you got a durned mutt, that’s what”–that there’s something more to it than that–but I do think the designer dog trend is a bit strange, and the fact they all have names like that is just part of the whole weird vibe.
One of my first memories as a child is being menaced by a lemon dalmatian. Or possibly it was liver–a food item, anyway. She was not a pleasant creature.
No. Your typical human being has a dozen or so unexpressed bad recessives, but that’s just because humans haven’t been extensively inbred and culled. I’ll use hip dysplasia as an example, since that’s one that actually shows up in dogs.
Say you have a bunch of German Shepherds, and let’s say that the hip dysplasia gene is absolutely rampant in the population, half of all of the hip genes, let’s say. This means that we’ll have about a quarter of our dogs who express the trait, half of them who carry the gene but don’t express it, and a quarter who don’t have it at all. Well, if you’re a responsible breeder, you immediately cull all of the animals who have the trait. That leaves you with two thirds of your dogs who carry the gene, and one third who do not: The total incidence of the gene is now down to 1/3, instead of 1/2. Breed those dogs, now, and the next generation will have 1/9 of the animals expressing the trait, 4/9 who carry the trait, and 4/9 who don’t carry it. Cull the dysplasiac dogs again, and the incidence of the gene is reduced further. After a while, you can start seeing patterns in which dogs do and don’t have dysplasiac puppies, and cull out some of the carriers, too, even without them showing the trait. Eventually, if the breeders are all responsible, you can completely eliminate the bad gene. And this is just using classical animal husbandry methods: With genetic testing, you could find the ones who have the bad gene and cull them before they had any puppies at all.
Hmm… I’m not sure I buy that. (ETA: I mean that I don’t think that sound breeding strategies can counteract all of the problems associated with inbreeding). There are going to be new deleterious mutations cropping up all the time (many each generation), and every time you cull one out, you’re reducing the genetic diversity of your breed. Essentially it’s a game of genetic whack-a-mole, except that the current techniques for identifying mutations won’t be of use for newer mutations. You can’t genotype a new allele until it’s obvious enough to be worth studying over several generations.
Probably it comes down to a set of trade offs. The smaller the breeding stock, the smaller the genetic diversity to select from, and culling will cut that even further. Now, obviously, this is what natural selection does on a larger scale, but species with low genetic diversity don’t tend to survive for long. So there will be a spectrum, on average, with the smallest populations plagued by major heritable defects, medium populations have lesser defects at a lower frequency, and large populations have minor defects and a very low frequency.
I don’t think I buy that… In any stable population, the total number of bad genes would be in equilibrium, so that the number of new bad genes that mutate in each generation would be the same as the number of old bad genes that die away each generation, and I just can’t see how that many old bad genes could die off per generation.