Wednesday, December 17, 2008

The Kennel Club Freak Show

Lionel the Lion Faced Man at right.

People go to dog shows to see pretty dogs, but let's face it they also go to dog shows to look at some very odd-looking dogs.

And so it is with some amusement that I see Kennel Club apologists expressing outrage that the RSPCA's chief vet, Mark Evans, has said that when he goes to Crufts, "What I see in front of me is a parade of mutants. It's some freakish, garish beauty pageant that has nothing frankly to do with health and welfare."

Nor does it have anything to do with working dogs, might I add.

Now to be fair, not all of the dogs entering the ring look like freaks.

But also to be fair, most of the breeds being paraded have serious health issues.

These issues include jaw-dropping rates of cancer, congenital deafness, liver disease, blood diseases, epidemic-levels of epilepsy, painful eye problems, hip problems, back problems, and ... well it goes on forever.

Almost every breed has a serious health issue that is endemic to it. Many are fatal, and frequently they are quite painful.

What's particularly maddening are those cases where breeders are intentionally breeding for a characteristic that they know will cause real pain and misery in a predictably high number of dogs that are born.

Tom Leppard, the Leopard Man of Sky at right.

For example, when people breed spotted and merle-coated dogs that look like the tattooed man at the circus, they are also breeding for a coat pattern they know that in some breeds will predictably lead to high levels of congenital deafness in their litters.

With Harlequin Great Danes, for example, as many as one in four may be born deaf and put to sleep -- something a professional breeder might "take care of" by simply slipping a new-born puppy into the freezer.

Schlitzie the pinhead at right, around 1935.

With Chinese Crested dogs, the gene pool is so toxic that if two hairless dogs are bred to each other, the homozygous offspring is prenatal lethal. Hairless Chinese cresteds are made by mating a "powderpuff" Chinese crested with the hairless variety to avoid perinatal mortalities. All hairless Chinese Cresteds have canine ectodermal dysplasia (CED) which results in teeth loss and a high chance of serious eye and skin malformations and defects.

"Happy Jack" Eckert at right, 1918.

Bulldogs are such genetic wrecks, one does not even know where to start. This is a dog with a head so large that almost all are born cesarean; a breed so poorly formed that it cannot have sex without assistance; a dog which is so achondroplastic that it cannot run; a dog with such a pushed-in face it cannot breathe; and a dog whose insides are so twisted in a knot that it farts from morning to night. In addition, the dog is prone to cherry eye, frequently has hip issues, and is so prone to skin infections that it is common for the pig-like tail of this dog to be surgically amputated after a show career in order to cut down on veterinary bills.

Unknown female sideshow performer with extra skin at right.

Shar Pei's are another breed with numerous serious health issues. This is a dog bred for their freak-show like folds of excess skin. In addition to hard-to-treat skin conditions, Shar Pei's also frequently suffer from entropion, caused by excessive skin over the eye which results in the eyelashes curling inward into the the eyeball. Left untreated it can cause blindness. Another common problem in the breed is "Familial Shar Pei Fever" and "Swollen Hock Syndrome" caused by the dog's inability to process amyloid proteins -- a condition which can lead to renal failure.

Six-toed Lundehund foot at right.

The Norwegian Lundehund is another dog bred for mutation. In this case we have a very undistinguished-looking small Spitz-like dog whose claim to fame is that it has six toes and a rather serious genetic disorder of the digestive tract (Lundehund gastroenteropathy) in which the dog loses its ability to absorb nutrients from food resulting in malnutrition or even starvation in extreme cases.

At left, Welsh giant George Auger (8' 6" tall), with
Tom Sordie (29") poking out through the coat tails.

And of course, we have all those giants breeds and teacup breeds, most of which are beset with serious, life-threatening and painful health problems. Giant breeds suffer from mind-numbing rates of cancer as well as heart problems and joint problems, and routinely die from bloat. Teacup breeds have bones so weak they can snap by jumping off the couch. In addition many teacup and toy breeds have skulls too small for their brains, resulting in open fontanels at the top of their heads. Add to this other common ailments, such as heart problems and dental issues resulting from their jaws being too small for their teeth, and you have a veterinary bill on legs and misery on stilts.

I could go on for pages, but if I were to only talk about the odd-looking dogs with health issues (and there are many more), I would do a serious injustice to the dogs that look fine from the outside, but which also have serious health problems.

The average Bull Terrier or Boston Terrier, for example, is dead at age 8 or 9 due to cancer, kidney and heart failure.

The show German Shepherd is an animal with such wrecked hips that it is often described as "half frog and half dog."

The Dalmatian suffers not only from congenital deafness, but also from painful uric acid stones.

All of the Setters have cancer rates of about 25% while, the Bernese Mountain Dog struggles with a cancer rate of over 45%!

The average Bloodhound is lucky to make it to age 7 due to serious gastrointestinal issues and cancer.

And so it goes, down almost the entire spectrum of 200 Kennel Club breeds, both old and new.

And so when Mark Evans talks of a "parade of mutants" at the Kennel Club, I suspect he is not only talking about the dogs that look bizarre; he is also talking about the Kennel Club people who profess to love dogs and yet continue to breed and sell animals they know will be in pain and discomfort much of their lives.

There is something seriously twisted here, and it's not all in the double helix of the dog is it?

Barnum's Freaks -- Cigarette insert from 1890 for Ogden's, an English brand.



Pai said...

The 'fatal hairless gene' in Chinese Cresteds is a myth. All Cresteds are heterozygous coated/hairless, with their hairlessness being an incomplete dominant gene. Breeding two hairlesses together does not cause stillborn litters, it just causes any egg fertilized with hl/hl to never develop... at most, 1/4 of the total fertilized eggs. Far from 'toxic', as the Xolo has the same genetic style and they've existed for thousands of years. Now the American Hairless Terrier is different (it's hairlessness is recessive), and maybe they have that issue, but I don't know anything about them.

As a breed Chinese Cresteds are overall very healthy, which comes from their founding lines (probably Xolos) being crossed with various coated toy breeds during their development in the 1930s which gave them better genetic diversity at the start than most.

Pai said...

I also must add, breeding Hairless Cresteds to Powderpuffs is actually a recent thing, for most of the breed history coated Cresteds were not bred from at all. The benefit Powderpuffs have towards Hairless is that when bred together the pups often have better dentition... the only defect truly genetically linked to hairlessness in dogs is that in some individuals all of their adult teeth do not come in.

You should have picked Chihuahuas if you needed a toy breed to point to; any breed standard that mentions having a hole in the skull as a breed feature is way more 'freakshow', in my opinion.

PBurns said...

The Chinese crested is not a very old breed and it does not come from China; it comes from the U.S. in this Century, and its history is pretty well known as a consequence (and never mind the nonsense about the dog coming from Africa, Asia, etc.)

As you note, other hairless breeds do not carry fatal genes; the Chinese Crested does, and that is one reason it is included here.

The main reason the Chinese Crested is included in this list, however, is that this is a dog that is being bred solely for its Freak Show qualities. There is no history to preserve here (as with the Xolo), no work once performed, no "hypoallergenic" qualities that are not existent in other breeds, etc. This dog's purpose is simply to look as odd as possible and nothing else.

As the link (supplied in text) indicates, the Chinese crested has a dominant gene for hypotrichosis (hairlessness) which is "prenatal lethal." "Prenatal lethal" does not mean the dog is sterile. It means that when the egg and sperm unite, they create such a monster that it is invariably born dead. It is true that the monster effect of two hypotrichosis genes coming together does tend to end the game sooner rather than later, but dead is dead whether it is the second week of pregnanacy or the eigth.

Yes, aside from the potentially fatal gene, the bad teeth, patellar luxation issues, two forms of progressive retinal atrophy, and a variety of severe allergy and autoimmune diseases (so severe they are often fatal), and some heart problems as well, the breed is fine and the average lifespan appears to be about 10 years (less than average for all dogs and quite a bit less than average for small dogs).

Of course, it is fine only so long as you never once forget to keep the heat up and keep it out of the sun ....


Lisa said...

It is not the Merle pattern itself that causes deafness but rather lack of pigmentation. Therefore, selection for *any* trait that results in excessive white around the head can have the same effect.

I can understand that some people have a problem with "deliberately" producing puppies that may have a defect. But, there *are* situations where doing so is not about "color" but rather about working ability.

Many of the finest working Australian Shepherds in the breed's history are the result of MxM crosses. Given the complex inheritance of herding ability, I'd sooner see breeders cull 1/4 of a litter for excessive white than make a cross to a lesser dog simply because of its color.

Eliminating merle dogs from breeding programs is impractical and would simply further decrease genetic diversity and working ability.

Lisa Kucharski

PBurns said...

Actually, the merle gene is very associated with deafness, as is the gene for dapple, and spotting and blue eyes (a common merle trait) and white in general. I have never read anywhere that it has a thing to do with white on the head per se. See >> Most border collies have white on the head somewhere, but not many are deaf!

As for herding dogs with working abilities, there are lots and lots of dog with working abiltiies that are not merle-coated (entire breeds, actually), and if you are not finding them, you are not looking too hard I suspect, or else you are treating health as a very casual thing ("Hey, who cares, they're just dogs, we can always cull the deaf ones"). That seems to be your attitude and it is pretty sad.

Of course some people simply stand behind "the breed standard," and never mind if it results in defective dogs. The real health of the dog is subordinate to the theory and the ribbon. If a large percentage of defective or deformed dogs end up in the mix, so what? That seems to be your thesis too, as it does not appear that you actually work your own dogs to anything more advanced than an AKC agility course. Is this what you think deaf dogs need to be bred for? Blue ribbons?


YesBiscuit! said...

Regarding the hairless gene, from

“Clearly this gene is critical to development, since we know that a double-dose of the mutated gene is lethal during early embryonic stages,” said Karlsson. “We just don’t know the mechanism yet.”

Didn't want to quote the whole durn thing but it's an interesting read.

Cat, Tessie, & Strata said...

I'm also having an issue with the merle/deafness link you make. While I am not arguing with the fact that the merle gene is connected to deafness there ARE breeds that, despite hosting merle dogs, have an very low percentage of dead dogs. In fact, deafness is so uncommon in Shelties that I had trouble finding hard numbers online. Exactly one dog has been BAER tested with the results sent to OFA (and the dog tested normal).

Deafness in Sheltie has a lot to do with merle-merle breedings done by irresponsible breeders. BYBs will mate two blue merles and get "double merles", dogs that are almost totally white and are often deaf, blind, or both. This color is not recognized by any registry and most reputable breeders no longer dabble in merle-merle breedings. (They did 25-50 years ago, but that was when they were trying to figure out what was up with merle-merle matings and double merles.)

So in short, no, at least Sheltie breeders are not "also breeding for a coat pattern they know will predictably lead to high levels of congenital deafness in their litters". Other breeds? Absolutely. And I'm not suggesting that Shelties are uber-healthy, but it's one problem we've pretty much avoided.

Anonymous said...

What we need is a kind of dog fancy perestroika. We need to restructure the way we breed dogs and value them.

Right now, we are valuing "purity of blood" and novelty. We are not valuing good genetic diversity and healthy conformations.

Reputable zoos that breed endangered species are always concerned about genetic diversity and whether conformation is being compromise. Mexican wolves, for example, after generations of being bred in captivity are developing inappropriate conformation for survival in the wild, becoming too short in the muzzle and too short in the leg. Zoos are trying to breed them back to a more natural type so they can be successfully reintroduced to the wild (actually, this reintroduction has been a bit of a failure, in part because they've been released in the wrong kind of habitat).

Even supposed working conformation must be taken with a grain of salt, the Lundehund you mention in this post was bred with the extra toes, because the extra toes were supposed to help them climb cliffs and catch puffins. I've read that their nutrient absorption issues are genetic in nature, and that the genes associated with the extra toes are actually associated with poor nutrient absorption. So breeding for polydactyly in dogs might cause poor nutrient absorption. Have you read anything on this linkage?

PBurns said...

Cat --

You make a good point, and I have added two or three words to the merle squib to reflect that.

Let me see if I can sum this up (I was not intending to write about Merle coats and deafness, just put up some pictures of freaks, LOL):

There are two pigment genes associated with deafness in dogs: 1) Piebald and 2) Merle.

Piebald (present in Dalmatians, Bull terriers, Jack Russell terriers, and several other breeds) is a recessive gene which produces areas of white by suppressing pigment cells (melanocytes). Both sides of the mating have to have the piebald gene for it to be triggered in the coat of the dog.

Merle (present in Shelties, Aussies, Great Danes and some other breeds) is a dominant gene, which means only one side of the mating needs to have the gene for it to be expressed in a Merle coat color.

As you note, the Merle gene is not destructive in all breeds of dogs for reasons that are not well understood, but may have to do with the physical construction of the Merle gene itself. More on that in a moment.

What is known is that in some breeds, even heterozygous merle dogs can have visual and auditory problems, and Great Danes are one of those breeds as are Ausssies. In other breeds, however, such as the Catahoula Leopard dog, homozygous merle breedings do not result in known health defects.

What's going on?

Scientists are not sure, but they are working on it.

In 2006 the Merle gene in dogs was isolated for the first time, and since then scientists have been looking at three different parts of the Merle gene (the head, the body, and the tail) to determine whether the relative length of each part (especially the tail) has an impact on health problems.

It should be noted that the health problems associated with Merle genes are not limited to deafness, but also include a wide vartiety of eye issues that fall into the folder of "merle ocular dysgenesis."

Since a good dog is never a bad color, I have a simple bit of advice for most folks looking for a dog: Avoid Merle. Is that too simple? Sure, but it works. ;)


PBurns said...

Retrieverman -

I have to say I am not saluting the malarkey that the Lundehund needed 6 toes to catch puffins.

For one thing, puffins are not that hard to catch (most folks use a net on a long pole), and any small dog can go down a hole to pull a damn bird out. Trust me on this: I have dirt dogs that climb rock cliffs to battle raccoon and fox, and I have the pictures to prove it!

In additon two small additional vestigial toes do not enable a dog to fly -- they are a very slight marginal advantage, especially when they are as far up the foot as they are in the picture shown above.

I suspect what we have here is a simple "sport" of the kind that often happens in inbred island populations of animals. Look at Ernest Hemmingway's six-toed cats for example. If Papa Hemmingway had been a cat peddler, he would have said it was a "rare breed designed to walk on soft beach sand," and perhaps sold a few cats extra out the back door to keep himself in booze. In truth, of course, Hemmingway's 6-toed cats were just genetic sports that made what had been a common house cat a "rare and special cat". Hemmingway's 6-toed cats are now a tourist attraction in both Key West and Cuba, by the way...

As to the notion that the Lundhund's digestive issues are tied to the toe, who knows? What is clear is that the whole breed is tied to a VERY narrow gene pool that is expressing itself through negative mutation. Mother Nature loves diversity and self-prunes all genetic trees that do not fork early and often. All the Lundehunds in the world today are descended from just six dogs left in 1962. Breeding out from just six dogs (or six people) in a closed registry will get you "Deliverance" pretty darn quick, if you know what I mean ;)


Anonymous said...

I would add one correction to Cat's discussion of double merle Shelties: double merle breeding, in Shelties at least, is not limited to the activities of the "backyard" breeder. There are some breeders active in AKC conformation who continue to perform merle-merle breedings, presumably in an effort to produce double merle males that will be of sufficient conformational quality to be frequently-used sires of blue merle dogs. While these dogs can't be shown because they are >50% white, they will produce nothing but blue puppies when bred to black bitches - something that makes them valuable to breeders who place value on color.

All kinds of claims are made that this practice is "perfectly acceptable" when undertaken by "experienced breeders". However, since the double-merle genes and a preponderance of deaf/blind offspring cannot be separated, I'm not sure exactly what sort of "experience" will make the inevitable euthanasia of large numbers of defective puppies acceptable. After all, how litters of double merle puppies need to be produced to yield a dog that is a) male, b) conformationally superior, and c) healthy enough to show off to other breeders? I'm guessing it's a pretty big number.

As you pointed out, Patrick, in all the world of dogs (even if people stay within one breed), it would seem possible to make a suitable match between two quality dogs who are not both merle. The knowing production of what are guaranteed to be cull puppies seems a really miserable way to go about dog breeding.

Pai said...

Known genetic diseases in Jack Russel Terriers: Lens Luxation, PRA, Cataracts, Deafness, Cushings, Ataxia, Luxating Patella, Legg Perths. Cherry picking the lowest age ranges I can find from the variety of statements I can find on Google, and I'm already halfway to being able to present JRTs as basket cases just as bad as the Chinese Crested. Crested lifespan is 10-14 years old... quite a range. To only point to the lowest bracket and apply it to the breed as a whole is misleading.

That kind of bias and skewing of data is not intellectually honest. There is a lot more to the health of a breed than just lists of all potential genetic flaws that have been recorded in that breed out of context, and making unfounded statements about their quality of life based on their looks.

If you research '% of fertilized eggs that develop' in regards to humans, a whopping 35-40% of fertilized human eggs are figured to never develop into embryos... so human genes should be called toxic too, by your measure.

It's clear you have very low opinions of nonworking breeds, toys especially, and scorn people who breed a dog for looks other than a job. But none of those things together or separately automatically equate inhumane breeding of dogs or diseased dogs.

I don't mind people saying Chinese Cresteds are useless, ugly dogs (I'm quite used to it), but to accuse them of being a defective, sickly breed is just simply not true.

Caveat said...

Interesting thread. I've always heard you should never breed a merle to a merle or bred a recessive colour such as fawn or blue, to the same colour.

I'm no dog breeder but I do like to absorb information.

Dogs such as collies, Aussies, etc also seem to be vulnerable to epilepsy and other neurological afflictions. It could be the coat colour or it could be the 'champion' syndrome, ie, too few studs fathering too many puppies.

As I say, I'm not a breeder.

Oh, and Patrick this post was great, I loved the graphics. And I say that as a lover of weirdo dogs like Toys, among other :>)

Anonymous said...

In Australia we have a type of working, herding dog called a Koolie ( in which merle coats are common. The breed is not a kennel club breed at all so the coat is not for decoration, it simply comes with the dog and bestows the risk of deafness. It seems that suitable genetic tests would make it possible to avoid the problem of deaf puppies.


PBurns said...

Pai, what I said about Chinese Crested is simply true: the dog is bred for its extreme looks. It looks like a freak and that is its primary attraction for most of the people who acquire the breed.

It is also true that a homozygous hairless version of this breed, crossed to another hairless homozygous version of this breed has a 100% fatal genetic defect; there are NEVER any live pups born. Ever. It is also true that this dog is a very recent breed. Sorry if those facts cause injury, but there they are, and your opposition to them does not change their reality.

As for the Jack Russell, it is the least inbreed dog in the Kennel Club (see ), and unlike the Chinese Crested this dog is not only several centuries old, it is actually bred for a function and it is still worked across the world today. There is nothing very extreme in the looks of the Jack Russell, and it is famous for its overall health. Yes, there are some problems with deafness among those who gravitate to white dogs and dogs with a lot of ticking (a show ring fixation among a very small number of kennels) and there is some lens luxation in some lines as well. No denying it. That said, the average Jack Russell is living quite a few years longer than the average Chinese Crested. And yes, there actually is real data on this. Go here >> and you will see that "The median age at death for the Chinese Crested breed was 10 years
and 1 month." The median age for Russell Terriers (all varities) is listed at 13.6 years (see In short, the lifespan of a Rusell terrier averages about 35% longer than that of the Chinese Crested, is longer than "traditional terriers" in general, and is among the longest of any of the dog breeds, second in the terrier class only to the Border Terrier (my other breed) at 14 years.


PBurns said...

YES BISCUIT really did put up a fascinating link.

It is here >> and is worth a read.

The bottom line is that they have sequenced the gene for hairlessness in Chinese Cresteds, and:

"Performing their own studies, the team discovered that the Foxi3 gene was indeed active in developing hair and teeth, supporting its possible role in hair and teeth abnormalities in hairless dogs. ... To identify the exact mutation in FOXI3, the researchers examined the precise sequence of DNA in the 102,000-letter stretch shared by the hairless Mexican, Peruvian, and Chinese crested dogs. One peculiarity was found in the DNA of all 140 hairless dogs and in none of 87 coated dogs tested: seven letters of repeated DNA in the FOXI3 gene. Because the genetic code operates by a rule of three — three letters of DNA encode a single amino acid in a protein — the addition of seven letters to the template completely shifts how the DNA is made into protein. Even though they come from different parts of the world, all three breeds share exactly the same seven-letter DNA change, suggesting they are all are descended from the same, ancestral hairless dog. ... Dogs with both copies of their gene mutated, who make no normal protein, usually do not survive until birth. The researchers found that all living hairless dogs in the three breeds they studied are actually heterozygous for the mutation. It seems that having one normal version of the gene may give the hairless dogs sufficient normal protein to survive, but not enough to generate a full coat of hair. The researchers don’t yet know exactly how the FOXI3 mutation leads to hairlessness, though their continuing work may yield an explanation soon. “Clearly this gene is critical to development, since we know that a double-dose of the mutated gene is lethal during early embryonic stages,” said Karlsson. “We just don’t know the mechanism yet.”

Full text of the paper in the Sept 2008 issue of Science is here >> (there is a cost).

Anonymous said...

There do not seem to BE any homozygous (HH) hairless Chinese cresteds, as any pup inheriting two hairless genes from its parents, each of which carries one dominant hairless gene and one recessive coated gene (Hh), simply does not develop. I have been breeding cresteds over 15 years (yes, and showing them), and I have had ONE stillborn puppy, and it was a powderpuff, carrying no hairless genes at all.
You are certainly entitled to your opinion about cresteds, but they do exactly the job they were bred for, companions. They don't leave hair all over the house, they rarely harbor fleas or ticks (nowhere for vermin to hide!!), and they have delightfully entertaining personalities. I wouldn't be without them.

PBurns said...

Yes, or as it says on one Crested sites puts it: "The gene, which produces hairlessness, is an incomplete dominate lethal. The unlucky puppy that inherits a double dose of the dominant lethal will die in the uterus or soon after birth, due to abnormalities and will not survive."


Pai said...
This comment has been removed by the author.
PBurns said...

Pai, you have fallen off the edge. You write "there are NO HETEROZYGOUS CRESTEDS"?

This is simply demonstrably wrong even if it is written in full caps.

Try again. Maybe you meant something else?

Tell you what. Go back and read what I have written. I have certainly read what you have said. I think you agree with me that homozygous hairless genes results in neo-natal mortality. Your big point is: "So what?"


Anonymous said...

Yes, homozygous hairless genes result in fetuses that never develop. And I couldn't have said it better myself--"SO WHAT??"

PBurns said...

Hmmm .... Kim seems new to the word "neo-natal."

Let's let her look it up.

Or is a dead pup still a "so what?"

No doubt that's the answer. So what!

Ditto on the wrecked teeth of the dog. So what!

So long as the check clears! So long as the ribbon comes for that long ride to the show!

When people call about problems they are having with the dogs related to canine ectodermal dysplasia, she no doubt gives them the "so what?" answer too.

Hey, that's what the breed is all about. So what.

Caveat emptor.


Anonymous said...

Actually the study you link to in your article says 'prenatal' lethal. "The homozygote HrHr is a prenatal lethal, hence the Chinese crested dog is an obligate heterozygote."

Cat, Tessie, & Strata said...

Anonymous, I think you've proved my point in your post.

I personally would not define anyone making an intentional breeding that had an extremely high risk of stillbirths, deafness, blindness, and too-small eyeballs as a "reputable breeder".

The breeders I bought my Sheltie from and now handle under would not engage in such a thing despite being asked several times. If you can imagine it, there are pet people asking for double-merle dogs. We got a request for one just a few weeks ago. (I guess they're hoping for a little freak to enter in the World's Ugliest Dog contest, something to rival those Chinese Cresteds.) Needless to say it was a wish we refused to grant. As far as we are concerned, that is right up there with asking for a "toy" Sheltie or requesting a dog with a very bad bite for whatever reason.

I understand the genetic reasoning some breeders use for purposely creating double-merles, but that doesn't justify it in my eyes. The kennel I work for does do a lot with "AOACs" (non-sable dogs) and we have lots of blacks and LOTS of blues without resorting to double-merle breeding. It is definitely possible to get lots of blues (and good blues!) without doing merle-merle breeding and breeders who justify culling all of those puppies or dumping them into rescue are simply making excuses for taking the easy route out.

PBurns said...

homozygous hairless = "prenatal lethal" = DEAD

homozygous hairless = "perinatal mortality" = DEAD

homozygous hairless = "neonatal mortality" = DEAD

homozygous hairless = "dominant lethal gene" = DEAD

homozygous hairless = "the mix of two Hairless genes is fatal in unborn puppies" = DEAD

"not heteroygous hairless" = DEAD


Anonymous said...

regarding merle, it seems to be a fairly old mutation as it occurs in a large number of breeds, including French hunting breeds, "old" German Shepherds (Altdeucher -- one of the parent breeds of the modern GSD), collies, etc.

In herding dogs, if you had a good dog, people bred to it. It was easy, if the dog happened to have distinguishing characteristics, to be "sure" that dog X was the sire of the litter. Historically, up until registries were created, for many breeders "little blue dogs" (merles) were "Australian Shepherds" and any non merles were "English Shepherds" (see the Hartnagle's excellent history of the Aussie for some of this). The same logic applied to the double dewclaw found in Beaucerons and their related cousin the Briard. For most ranchers, it didn't matter if the dog produced some pups with iris columbas (see the ASCA write up on this being linked to merle or even just google merle + eye defects) -- dogs that had "issues" simply didn't make it to being herding dogs. You could tell "old Blue" sired a litter by finding "little Blues" in it. Keep in mind, the AVERAGE age of a working ranch dog is 6 or so -- like wild wolves, it's a hard job and one kick from one steer can end a dog's life in a second. So mostly, working dogs with genetic "time bombs" didn't live to express it. Border collies were mostly black with SOME white -- there was a very large prejudice for decades regarding merle (which occurs), red or mostly white BCs (all of which have been ISDS winners)-- PRA is a problem that crops up in BCs-- most of the working registries have some kind of program to eliminate it, but again, before modern medicine, most dogs that had progressive PRA simply didn't live long enough to express it.
If it doesn't look like a wolf, it's been modified SOMEWHERE in it's history. Some breeds are lucky enough that the modification came with minimal side effects. Some aren't. The thing is, that denial of the side effects does nothing to eliminate or minimize them and breeding for "I just want it" without consideration of the possible issues is irresponsible. The fox domestication project yielded particolored foxes, so it's entirely possible that colors (including merle) may be so tied to particular behavior or talents that elimination is not desireable. BUT, one should not do it "just because I can" and dismiss the consequences.

Peggy Richter