Replies to This Discussion

Permalink Reply by Beth on December 29, 2009 at 5:52pm

What you are talking about is line-breeding.

Line-breeding is meant to preserve traits, though sometimes I personally think it gets too close and hence some of the systemic genetic problems we have in some pure-bred dogs.

My (limited) understanding is that an optimal "out-cross" to your line, once it gets heavily line-bred, would be to breed your line-bred bitch to a totally unrelated line-bred stud, or vice-versa. Your example of breeding two dogs that share a mother (half-siblings) would probably be too close for most breeders, UNLESS perhaps you are either developing a brand-new breed, or trying very hard to add a rare trait, or remove a ubiquitous trait from a breed where a problem is being rectified, and you need your close relatives in order to make a sudden dramatic shift in a certain direction.

I have Maddie's pedigree here. Her sire's sire (paternal grandsire) is the same as her sire's dam's maternal grandsire. Hmm, hard to explain. If you look at HER sire's pedigree, his father is also his mother's grandfather. In a four-generation pedigree, therefore, the dog that is #2 on the top of one side is also #4 on the bottom of that same side. If that makes any sense at all.

And her paternal great-grandsire is the same as her maternal grandsire, so a dog that is in the #2 position on the bottom half is the same as the # 3 spot on the top half. Which means she is somewhat line-bred top and bottom, but since the bottom is not terribly line-bred at all there is still a lot of diversity.

Since Maddie is already somewhat line-bred, when she herself was bred, she was bred to a totally unrelated stud, who himself has line-breeding in his own pedigree.

That's my limited understanding; it's similar in Thoroughbred racehorses, which I used to follow more closely than I follow dog pedigrees.

• ▶ Reply

Permalink Reply by Alice on December 29, 2009 at 6:45pm

I wonder where the line between carrying on certain traits and developing genetic problems falls. I know I saw a breeding pair that shared the same mother (half siblings) which would also mean they'd each have one set of grandparents the same. I also saw some of what you talked about where the grandmother of one (on the Mom's side) is also the great grandmother of the other (on the Father's side). It seems a little risky to me. Maybe not when talking about great grandparents but anything more makes me wonder.

• ▶ Reply

Permalink Reply by Beth on December 29, 2009 at 9:28pm

I think that often line-breeding is not so big a problem on a small scale. It's when it gets carried out across a breed so the entire gene pool shrinks that you have a problem.

If you look at the Thoroughbred horse, whose entire ancestry has been documented since the development of the breed, you can trace them all back to three foundation sires and a slightly larger pool of similar-type mares.

This was well-documented and posed few problems (thoroughbreds having been relatively healthy) for a couple hundred years.

Then recently you had what Kerry mentions below--- popular sire syndrome and a handful of sires placed an inordinate influence on the breed, not just in tail-male lines (sire to sire to sire) but as broodmare sires as well. Why this happened is debatable, but it seemed to be when yearlings from certain lines began bringing in millions in the sale ring and people started breeding to sell rather than breeding to race. Genetic diversity rapidly declined, and now much to the horror of quite a few bloodstock specialists, two of the three sire lines have been almost completely lost. In tail-male line, something like 95% of all racing Thoroughbreds now go back to the one foundation sire, the Darley Arabian (through his descendent, Eclipse). Now, the other lines are still present in the distaff side, but not so much on the sire's sire's sire line. And the result is troubling.

My understanding is there is still a lot of genetic diversity in Corgi lines, and I think that is because you still have a lot of smaller breeders with just a few dogs, each working on their own lines. But trouble can come up quickly if diligence is not maintained. Some breeds are more or less doomed because serious genetic problems are so rampant; in those cases it may be advisable to open up the stud book and allow some fresh blood back in.

• ▶ Reply

Permalink Reply by Joanna Kimball on December 30, 2009 at 12:16am

Dr. Bell, as you may recall, makes no recommendations as to how to breed "most correctly." The ten-generation COI is a decent rule of thumb because it lets you get far enough back to see where your dog fits in the breed but doesn't go so far back that you're into the founding individuals (which would start to elevate COI again) but it's not the ONLY way to do things. It depends on when your breed came into the US and how long the generations typically are and so on. In Danes, where typical first litters are at age 2, a ten-generation COI is only going to get you back between 20 and 30 years. I'd like to go further back in that to see where the dogs start to intersect again. In Salukis, where a typical first litter is at age five to seven, going back ten generations could put you right into the founding imports and then all the COIs are going to look high. It's something that you need to figure out based on your breed and its breeding culture.

The reason that typically only one or two individuals in a litter are shown has absolutely nothing to do with expendability. It is the same reason that a supermodel can have a plain-looking sister. When you combine two individuals, sometimes the genes come together extremely fortuitously and sometimes they don't. This is the case whenever a breeder of any species is breeding for any reason - not all guide dogs make the grade, not all cows produce well, not all horses can jump. The dogs, cows, horses, etc. that are not exceptional are NOT therefore expendable. They are put into homes where they will have less asked of them, with the goal of a long and happy life.

It's one more reason you have to be so careful when you choose a breeder. Every breeder has a goal in mind, whether that's show or herding or field trials or guide dog or whatever, and they will keep and breed (to further that goal) the puppies that most closely approximate the ideal. As a pet owner you're getting that breeder's "failures." As a breeder, my goal is to make sure that I may fail on head type or tailset or color or coat, but I consistently produce a sound, happy dog who can work. Every time you breed you compromise; every dog has something you don't like. I won't breed to an unsound dog, no matter how worthy he is in some other way, because while I may get a sound puppy to move on with, her siblings may not be. I WILL breed to a dog known to produce "bad" color or coat or with an ugly head, because ending up with a puppy that has those faults won't hurt its ability to live a long life with its owners.

If, as a breeder, your goal is nothing but "nice personalities" or something, the failures you're putting in pet homes won't even have that, let alone bodies that will last them for fifteen years without pain or a brain that makes them worthy of their breed name.

• ▶ Reply

Permalink Reply by Joanna Kimball on December 29, 2009 at 11:51pm

Here is my series of blog posts on inbreeding/linebreeding (some go back a ways): http://blacksheepcardigans.com/ruff/?s=inbreeding&x=0&y=0

As a show breeder, here's the basic story:

Linebreeding is just a flavor of inbreeding. Generally the distinction we make is that inbreeding concentrates every dog in the pedigree, without favoring any. A full brother-sister mating would be an inbreeding, because now the two halves of the pedigree are identical. A linebreeding tries to have a single dog or a few dogs on both halves of the pedigree; we say we are "linebreeding on" that dog. A "tight" linebreeding is when that shared dog or dogs appear very close to the front of both halves of the pedigree, for example a half-brother/half-sister mating or a granddaughter-grandfather breeding. A "loose" linebreeding is when that shared dog or dogs appears further back on the pedigree, for example if the two dogs share a great-grandfather.

There are no real rules to linebreeding or outcrossing. There are jealously guarded "secrets" and long-time breeders will write books about how you have to do this many generations of this type of linebreeding and then outcross to this extent or with that dog, but the fact is that all you're getting is what they think worked for THEM, and it's as likely to be a fluke as it is to be real. Since we don't have production data for dogs (we don't have feed utilization numbers or growth rates or milk production, etc.) there's actually almost no real science involved, except that you can make fairly accurate statements about the fact that higher-inbred dogs, across a population, will be more predictable in type and ability than lower-inbred dogs. "More predictable" doesn't mean better, it doesn't mean that individual dogs won't excel or fail; it just means that the inbred group will be more alike than the non-inbred group.

- It's relatively uncommon in the show world to do full sibling matings. It's an ancient technique to flush out any problems in the pedigree - for example, if you think there may be an adrenal disease in the pedigree, you might breed brother-sister because that is the breeding most likely to produce the issue and that lets you confirm that it's there and then work to breed it out. Or you may do it once, to produce an inbred dog that you then use on completely unrelated (outcross) bitches. As genetic testing for diseases becomes more useful and available, it's likely that we'll see fewer and fewer full-sib matings.

- Some people linbreed (now we're talking about concentrating one or a few dogs in a pedigree) because they don't know what they're doing. They had somebody tell them that the optimal breeding is to breed back to the maternal grandfather and so they do it. Thankfully, this is relatively rare.

- Some people linebreed in order to give themselves a greater chance of getting a particular conformation trait in the puppies. Every breed has some traits that are easily lost in random mating and are extremely difficult to get predictably. Across the world of dogs, getting a good shoulder layback (critical in fluid, easy movement, which is supposed to be our number-one goal) is SUPER hard. Shoulders tend to want to go straight up and down. In Cardigans, getting the proper front (where the upper arm curves around the ribcage and then the "wrist" drops straight down) is VERY hard. Fronts tend to want to turn out too much. So if you know that your dog's grandfather had a wonderful front, or a great shoulder layback, you try to get two doses of him in the litter instead of just one. Some people will actually breed back to him; some will instead try to find a dog related to him. Either way the goal is the same.

- Some people linebreed in order to create dogs that will reproduce consistently. This isn't actually the same as the above. A linebred dog, because he has less genetic variability, will tend to produce puppies that look a lot like him no matter what bitch he is bred to. He can consistently fix conformation problems and make better puppies than the mom. There are a couple of dogs I can instantly think of that are so prepotent (that's the word for consistently producing his own qualities) that you can immediately tell whenever one of his children or even grandchildren walk by you. They all have the same distinctive look and share the same virtues.

- Some people linebreed because their goal is the greatest number of show puppies in each litter. This is not a goal I share, but for some breeders success is having every puppy (or as close to it as possible) get a championship. Linebreeding allows them to have a very predictable result.

- Some people linebreed because they have no choice. Many breeds have high COIs (coefficients of inbreeding) across the board and you'd have to sacrifice quality in order to get a lower COI. That's the case with the breeding I just did - if you look at the pedigree: http://blacksheepcardigans.com/?page_id=2539, you can see a dog named Stormy (Ch. Tafwnr Coedwig Blackstorm) three times in the five-generation pedigree and a dog named Ch. Sisterwood Triathlon twice. I did this breeding because I believe the quality of the resulting puppies will be very high, and I don't just mean in the show ring. Stormy, who is the dog who contributes most heavily to the puppies, is the sire of a herding champion (by that I mean the actual title; he achieved his HC, not just that he was a champion who herded) and was known for his sound structure and great work ethic. The breeding allowed me to keep sizes relatively small, since neither dog is oversize, and they're both structurally correct and have herding instinct. In order to go to a lower COI, especially since I am not allowed to breed her to anything but a black dog (tri or tri-brindle), would have meant sacrificing not just show potential but genuine soundness. As much as I value low COI, that would not be worth it.

Linebreeding is not just done by show-dog people. I've actually had more field breeders than show breeders argue to me that a high COI is essential to judging quality. Even in production animals (cows, goats, sheep) it's common to do a couple of generations of inbreeding to end up with a very prepotent sire who is then used on a lot of females. Dogs (and import-bred Arab horses) are I think the only group of breeders who does it for generation upon generation (which I do have a problem with), but it's not something "owned" by show breeders in particular.

YOU ABSOLUTELY MUST look at pedigrees before purchasing a puppy. That's not because anybody expects you to know a magic number of shared ancestors or to calculate a ten-generation COI in your head. It's because the pedigree is the key to what the breeder's goals are, and you MUST be comfortable with your breeder's goals. An absolute MINIMUM qualification of a decent breeder is to be able to discuss why they did the breeding they did and why it was a better breeding than average. You can breed two champions together like an idiot or breed two unfinished dogs together like an artist; ditto with any field title or herding or working title. A good breeder knows what's behind his dogs and he will talk about qualities that have nothing to do with personality; it should be ability and soundness and quality.

• ▶ Reply

Permalink Reply by Alice on December 30, 2009 at 11:44am

Thanks Joanna and everyone else, you've provided some great information. I know the basics of inbreeding and the possible repercussions and it is interesting to hear about linebreeding.

I personally always look at pedigrees, even when looking at dogs I have no stake in (purely curiosity). Someone who is looking to purchase a puppy and has no intention of showing or breeding the dog may not take the pedigree into account though which is why I thought I would bring it up. I looked at the pedigree of Finn's parents but I was not looking for inbreeding or linebreeding, I was looking to see what dogs were in there to learn more about the breeders intentions as Joanna mentioned. Now looking at their pedigrees I do see a couple names popping up multiple times but apparently that's perfectly alright. Finnigan's breeder later decided to open his lines and retired several dogs because he suspected too much linebreeding or inbreeding though so there must be a point at which it becomes cause for concern.

Just out of curiosity Joanna, what are some of the prepotent dogs that came to mind? I know what you mean though. I've come across that as well. Actually, what started all this was that I saw a dog who looked so much like another that I checked the pedigree and sure enough he was in there along with a half sibling breeding, hence all my questions.

• ▶ Reply