Mismark Case Study: Great Dane Revisit and Expansion

The first mismark case study I ever did was of the great Dane. To a large extent that post was incomplete, so I started from scratch. This took me an embarrassingly long time to put together, but I was determined to finish it before doing anything else.

A group of great Danes, two in standard colors (fawn and mantle) and two in non-standard colors (merle mantle and "true" blue merle aka dilute merle). Image is from Flickr.com under a Creative Commons License

Of all breeds that I have looked at, the great Dane breed color standard makes the least amount of sense. While most will restrict a few colors, often recessives or minor aesthetic differences such as eye color, the Dane standard is much more restrictive. Using only the genes that go into making the six "acceptable" colors (fawn, brindle, black, blue, mantle, and Harlequin), there is a significant list of non-standard colors that could occur simple by breeding between those six standard colors. This has lead to color class breeding. For people who breed show Danes, those color classes are a way of life. What are the color classes? According to the Great Dane Club of America, there are four of them. Breeding between these classes is severely frowned upon and there is potential that it could lead to expulsion from the breed club.

Fawn and Brindle

Breeding of fawns and brindles to other fawns and brindles whose pedigrees are free of black, blue, or Harlequin

Blue

Breeding of blues to other blues or to "blue-bred blacks" whose pedigrees are free of fawn, brindle, or Harlequin. A blue or blue-bred black can also be bred to a black-bred black

Harlequin

Breeding between mantles, Harlequins, and "Harlequin-bred blacks." It is also acceptable to breed a dog from this class to a black-bred black. Any combination is acceptable so long as the pedigrees are free of fawn, brindle, or blue. This includes the pairing of two Harlequins.

Black

"Black-bred blacks" come from generations of exclusively black to black breeding, resulting in pedigrees free of blue, fawn, mantle, brindle, or Harlequin.

Even breeding within these classes does not eliminate mismarks, with more than a few breeders being surprised when, for example, they pair a mantle with a five-generation pedigree fitting the Harlequin class to another mantle of the same designation and produce a blue mantle or fawn mantle. This is not even getting into the flaws with Harlequin in and of itself.

The Mismarks:

This great Dane is a merle mismark, one of the most common seen in the breed.

Bolded are standard colors and green are non-standard colors produced ONLY from genes that go into the standard colors. I've omitted Harlequin + mantle and similar as they can be visually indistinguishable from dogs without.

	Black	Fawn	Brindle
add Irish white	Mantle	Fawn mantle	Brindle mantle
add dilute	Blue	Blue fawn	Blue brindle
add Irish + dilute	Blue mantle	Blue fawn mantle	Blue brindle mantle
add merle	Merle	Fawn merle	Brindle merle
add merle + dilute	Blue merle	Blue fawn merle	Blue brindle merle
add merle + Irish	Merlequin	Fawn merle mantle	Brindle merle mantle
add merle + Irish + dilute	Blue merlequin	Blue fawn merle mantle	Blue brindle merle mantle
add merle + Harlequin modifier	Harlequin	Fawnequin	Brindlequin
add merle + Harlequin + dilute	Blue Harlequin	Blue fawnequin	Blue brindlequin
add merle + merle	Double merle replaces single merle. Causes an increase in white. Has a high likelihood of deafness, blindness, and/or eye abnormalities.
add merle + merle + Harl	White. Most whites have virtually no color. Has the same health concerns as double merle.
add liver	Chocolate acts like dilute, replacing black. This includes masks on fawns and stripes on brindles
add liver + dilute	Blue and chocolate combine to make Silver Silver replaces chocolate, dilute, or black
add piebald	Piebald would replace Irish in the above combinations. Both piebald and extreme white piebald are possible

Tan on this Dane is consistent with a seal

This great Dane has no mask

All of these colors are recognized as existing by the American Kennel Club so, unlike with some breeds, the vast quantity of potential mismarks are well known to exist. Fitting (mostly) within the above chart are subsets that would severely penalize or disqualify an otherwise acceptable color from conformation: bad black (aka seal), improper pigmentation (usually a pink-nosed Harlequin) and white on a blue, black, fawn, or brindle (some is allowed, but not preferred). Although unusual, there have been instances of tan pointed and maskless great Danes as well. Ticking sometimes happens, but, for the AKC at least, there is no mention of whether it would be penalized or not.

So, what causes all of that, anyway? 

As can be seen in the chart above, a very large number of mismarks can occur when only looking at the genes that make the six standard colors.

This puppy is a Brindlequin mismark

Perhaps the biggest indicator of color in great Danes is the K locus, which codes for black and brindle. Black (K) is most dominant, followed by brindle (kbr), and then non-black/non-brindle (k). Since a black dog only needs one copy of the dominant K gene to be black, it can easily carry either brindle or fawn as Kkbr or Kk, respectively. When the right two dogs come together, a litter that was expected to be all standard black mantles could throw something unexpected. When homozygous for the non-black/brindle allele (kk), the agouti locus will show through unobstructed. For most Danes, this will make them fawn, but this would also be why the tan pointed or even brindle pointed Dane appears. In the case of brindle points, the agouti locus is still peaking through, but it is obscured by the brindle overlay.

Referring back to the chart, it's quite clear that most of the standard Dane colors are black-based. Black, mantle, and Harlequin all have obvious black on them, while blues are genetically dominant black with the extra quality of also being diluted. Due to the black base of all of these colors, if a dog inherits a gene combo that does NOT code for black (kbrkbr, kbrk, or kk) then it would have either brindle or fawn instead of black. This includes Harlequins (turned to brindlequin or fawnequin) and mantles (turned to brindle mantle/brantle or fawn mantle/fawntle).

This dog is a heavily marked blue Harlequin

Blue/dilute is a simple recessive gene inherited on the D locus, with the dog having to receive two copies (dd) for it to show. Dilute affects all black in a dog's coat, turning it to some shade of gray. This includes masks and stripes in brindles. Even if trying to only breed the dominant variant of a gene, there is no guarantee that the recessive form will disappear. This is why diluted variants such as blue Harlequin (aka porcelain) continue to pop up from parents whose pedigrees may have not seen a blue for decades.

This Dane is a fawn mantle mismark

The A (agouti) locus is probably the most varied locus in the dog world, but in great Danes there appears to be only two variations, with the breed being nearly fixed for one of them. Being fixed for a gene means that no other genetic variants are seen at that locus. Sable (Ay), which causes Fawn, is the most dominant of the agouti variants and tan point (at) one of the recessive variants. Considering the low incidence of tan point, great Danes can, for the most part, be treated as if they were fixed for sable due to there not being many tan point carriers.

Fawn and brindle great Danes are also known for having black masks. This is inherited independently on the E locus, aka the extension locus. The most dominant of the extension genes is, indeed, the mask gene (Em). Considering that maskless Danes do exist, some other variation has to exist, but it can be difficult to tell whether that would be recessive red (e, which strips away all black in a coat, leaving it entirely red) or the "null" allele of non-masked, non-everything-else (E). There is some evidence for both but, like with the agouti locus, the breed appears to be near fixed for one allele. In this case, that's the mask gene.

This dog is a dark blue mantle mismark

Mantle comes from recessives on the S (spotting) locus. White spotting varies significantly from virtually no white to virtually all white. Blue, black, fawn, and brindle Danes must have at least one copy of the solid (S) allele. If carrying Irish white (si), piebald (sp), or extreme white piebald (sw) a dog will likely have at least a small amount of white due to the semi-dominant nature of the spotting genes. This is usually on the chest and/or toes, but may also lead to a white tail tip or some facial white. A dog with white, however, does not necessarily carry any of the recessive variants. Residual white is quite common in genetically solid (SS) dogs due to how color migrates during very early development. It essentially spreads from the spine to the extremities in the womb, so if the color doesn't spread all the way, a dog will be born with some amount of white. As with many breeds some white is allowed on otherwise solid Danes, but too much is penalized.

This great Dane is a piebald merle

To be mantle, a dog's phenotype must have more white than what is possible from a homozygous solid dog. The white required by the breed standard is essentially Irish white, a pattern most people associate with breeds such as the border collie. This appearance can come from several variations of the spotting locus. As expected, a dog can be homozygous for the Irish white variant (sisi), but a dog carrying piebald (sisp) will look similar, though possibly with a little more white. Since the standard allows white to break the main body of the dog, some Danes who fit the standard appear to be minimal piebalds (spsp).  In addition, pseudo Irish white, caused by either a solid carrying extreme white (Ssw) or what would have been a low-white Irish carrying extreme white (sisw), will look like a homozygous Irish dog.

I've met a dog much like this extreme white Dane

Due to the vast amount of variation in the potential genotypes that would lead to the acceptable mantle phenotype, it isn't surprising that piebalds and extreme white piebalds are known to pop up in the breed. As extreme white is more common, it's likely there are a lot of pseudo Irish dogs out there.

In addition to the vast quantity of potential for mismarks seen above, there is the illogicality of the nature of breeding for Harlequins. Harlequin can, by its nature, NEVER breed true. It is a color caused by a very specific combination of two problematic genes. One of them is merle (M), which is well known to cause serious issues when a dog inherits a double dose of the gene. Harlequin (H) is, in some ways, worse than merle. While double merles are generally viable despite the high incidence of deafness and/or blindness, double Harlequins are lethal in utero. Since no living dog can be homozygous for Harlequin, this explains why merle dogs are to be expected in every litter with at least one Harlequin parent. The worst part of Harlequins, however, is that breeding a Harlequin to another Harlequin is perfectly acceptable under color class breeding. This is a serious problem.

Here is a Punnet square for a Harlequin/Harlequin cross:

	HM	Hm	hM	hm
HM	HHMM double double lethal!	HHMm double Harlequin merle lethal!	HhMM double merle Harlequin possibly deaf/blind	HhMm Harlequin
Hm	HHMm double Harlequin merle lethal!	HHmm double Harlequin lethal!	HhMm Harlequin	Hhmm black or mantle carrying Harlequin
hM	HhMM double merle Harlequin possibly deaf/blind	HhMm Harlequin	hhMM double merle possibly deaf/blind	hhMm merle or merle mantle
hm	HhMm Harlequin	Hhmm black or mantle carrying Harlequin	hhMm merle or merle mantle	hhmm black or mantle

This white Dane is deaf.

When crossing two Harlequins together, the assumption is that the breeder wishes to produce more Harlequins. Considering this, a 25% chance of a Harlequin is pretty bad when the price to pay is the likelihood one-in-four of the puppies will never even have a chance to live and another 18.75% of embryos have a very high chance of being blind and/or deaf. That's almost half the litter! Some breeders cull double merles, but this is a shameful practice, sweeping poor decisions under the rug in the form of dead puppies. The alternative, however, isn't exactly wonderful. Due to their sheer size, only so many people are willing to adopt a great Dane. When that size comes with impaired senses, the number of available homes plummets. A dog that is deaf and/or blind is significantly more difficult to train than a dog with full use of its senses, and a huge, poorly trained dog can be very difficult to deal with. I once met a deaf great Dane at my local dog park who would not stop humping my Labrador. My poor dog barked at him in frustration, but, being deaf, the dog had no awareness of such social cues. If my dog wasn't so even-tempered the Dane could have easily been injured due to his poor social skills. Since the Dane's owner wasn't leaving, I understandably removed my dog from the situation.

Many assume that white great Danes are albinos, but their lack of pigment is caused by a very different set of circumstances. As breeding from a merle is not acceptable in color class breeding, whites come from Harl/Harl crosses like the one above and, as they are double merles, they have a high risk of sensory issues. Some may be very lightly marked Harlequins, but this is unlikely. I have also seen at least one dog that appears to be an extreme white piebald Harlequin, but this is also not the most likely phenotype. What patches you do see on white great Danes will usually be solid black, but just as fawn mantle and blue Harlequin can occur, you can also get double merle Harlequins of a different color as well.

In contrast to the Harl/Harl cross, here's a Punnet for a Harlequin/black cross where the black does not carry the Harlequin modifier:

	HM	Hm	hM	hm
hm	HhMm Harlequin	Hhmm black or mantle carrying Harlequin	hhMm merle or merle mantle	hhmm black or mantle

This cross has the same 25% chance of producing a harlequin, but there is a 0% chance of color-related non-viable embryos and a 0% chance of double merles, so the breeder will not have to find homes for deaf and/or blind dogs. There will still be merles, but since the Harlequin modifier is lethal in a double dose, this is an inevitable consequence of Harlequins.

This great Dane is a merle mantle mismark

Since merle is never going away, the absurdity of making it a disqualification is rather remarkable. In most breeds, colors that pop up with the high frequency of merles in great Danes are generally accepted by the standard. In addition, why does the Great Dane Club of America allow Harlequin/Harlequin cross when crossing a merle with a black that's been tested positive as a carrier for the Harlequin modifier is frowned upon? Both have the potential to produce Harlequins, but the merle cross, like breeding a Harlequin to a non-carrier black, doesn't have the issues that a Harlequin/Harlequin cross does. Mantle became an accepted breed color because of Harlequin breeding. I don't see why merle shouldn't be treated the same way.

This puppy is a chocolate fawn

This portrait is believed to include a Dane

The last color determinant I have not discussed is the B locus. Brown/liver acts much like the blue dilution. It's a simple recessive (so all dogs expressing it are bb) that turns any black in the dog's coat from black to some shade of brown. Like dilute, any dog with black can have liver coloration instead. In great Danes, liver is known as chocolate. It can also act in combination with blue, diluting down to Isabella (also known by names such as silver, gray, ash, pearl, lilac, and fawn). Weimeraners are well known for their Isabella color. In great Danes, it appears that Isabella is known as either silver, lilac, or dilute chocolate. There is some evidence that chocolate was once an acceptable color in the breed that since fell out of favor, and the seventeenth century portrait at right is interesting in relation to this possibility. If the dog is in fact a very early great Dane or Dane relative, it isn't any color that would be found in a modern show ring. It appears to be an Isabella mantle.

Conclussions

Of all the breeds I have looked at, the great Dane is most problematic in terms of its color standard. It's one thing for a breed to have a number of recessive colors that show up on occasion, but on top of that Danes have mismarks that are guaranteed to occur when breeding one of its six acceptable colors. Also problematic are the color classes, which essentially create breeds within breeds. All purebreds have limited gene pools due to closed registries. Color classes take that genetic variation and limit it again. Since closed registries prevent any new genetic variation from being added, what there is cannot be replaced once it's gone. Genetc variation is essential to a healthy animal, allowing their immune system to function effectively. The less variation, the less functional the immune system.

With the limited variation that comes with closed registries, selective breeding is like taking a weed-wacker to a spindly bush. It can only take so much before there isn't enough of it left to survive. That is essentially what is occurring in modern dog breeds, where effective populations sizes are worse than what is considered at risk of immediate extinction in endangered species. Since the great Dane has such major restrictions in color acceptance, there is a massive amount of potential for loss in genetic variation that could have otherwise been saved.

Images in this post are from Flickr.com or Wikimedia Commons with the exception of the heads used in the Punnett squares, which were created by me. Everything is under a Creative Commons license and source links can be found beneath each image. 

Mystery Leg Markings: Some Theories

Ebon, showing some of the brown that can be seen on his legs

For the longest time I thought nobody else noticed the fact that Ebon has distinct rusty markings on his legs. Then, this past November I came home from the vet and found this on Ebon's shot records:

Color: Black and Brown. The vet did once list him as Black, but some time during the many years they've been caring for him it was changed. I apparently wasn't the only one to notice.

Ebon at 21 months, Hellon at 8 years. Click to view larger

Recently I also found out that Ebon's mother, Hellon, apparently had red markings on her legs as well. I honestly never looked close enough to tell. I hunted through the pictures that I can find of her and yes, in the pictures to the right you can see what appears to be some hints of lighter color on her front legs. Considering that none of the pictures I can find are of her in bright sunlight, and many are from more recent years when her legs had heavy greying, this is probably the best I'm going to be able to find.

So, this is my conundrum. What exactly is Ebon? In my Guess the Genotype posts I try to take a stab at determining the likely genotypes of dogs who sometimes have some very unusual markings. I would place Ebon into the somewhat unusual category. Mismark Labrador retrievers happen. Some of them are quite striking. Dogs with tan points, with or without brindle, have been popping up in purebred litters for many, many years. It does seem that, to some extent, this mismark is more common in working lines since there is generally little to no real concern over what the dog looks like, as long as it can retrieve. However, usually the mismarks that are seen are far more obvious than what is seen on Ebon or his mother. Why in the world are their markings so subtle?

Ebon's left forepaw in bright sunlight, the most red it ever appears

I have broken down Ebon's markings before, and theorized that he might be a "bad black" over black and tan. Bad black, aka seal, has been shown to be genetically dominant black when tested, seen in a study on Italian greyhounds. However, unlike normal dominant black, some of the underlying Agouti locus is able to show through. In most breeds where seal is a known color, tan point and other recessive alleles on the Agouti locus are either rare or completely absent. The Italian greyhound is one such breed, with tan point being nearly absent from the population. This could explain why seals usually have red that is patterned much like a shaded sable. Ebon, on the other hand, has red in an almost identical pattern to dogs who are tan point, such as Rottweilers and Doberman pinschers. Labrador retrievers are quite possibly fixed or near fixed for the tan point gene. The mismarks that are seen are never sable or the like, but tan point pops up on a fairly regular bases. If Ebon were bad black, he would have to be bad black covering tan point, otherwise the "bad black theory" would not work.

However, this is not the only possible explanation for Ebon's rusty legs. There are, in fact, two more.

The first is that he's a very minimal black and tan. Though highly unusual, there are dogs that occasionally pop up in breeds that are normally tan pointed that appear to have little to no red on them. This seems to be somewhat prominent in Rottweilers, as some people seem to prefer the low amount of tan. If you do a little hunting around the Internet, it isn't that difficult to come across dogs with almost no tan at all. The same is true for Dobermans, though the dogs with less tan usually have more than the near-black Rotties. However, this is not always the case. There are some dogs that seem to have absolutely no tan at all, but those dogs usually appear to not be purebreds. The "all black Rotties" usually look like Labrador or molosser mixes and the "all black Dobes" usually look like they have some great Dane or greyhound in them. The fact of the matter is, dogs who are simply black and tan never have as little red as Ebon does. There has to be something else at work.

You can draw a line between the solid black and the red-laced

The final theory is that he's a black and tan with heavy brindle. When I first discovered his red markings, this was my thought. However, I was hesitant to say this was the case because I couldn't see any distinctive stripes on his legs. Despite this initial thought, the more I stare at his legs the more I start to think that he does indeed have stripes. For one thing, brindle is a highly variable pattern, with the stripes so randomly placed that a dog can have lots of stripes on one side of their body and virtually none on the other. Ebon's legs don't match, and the color is not consistent. His right foreleg has more red than his left, including a quarter-sized patch near his wrist that has more red than the surrounding area. Black and brindle is a known mismark in Labradors, with nearly as many black and brindle dogs popping up as black and tan dogs. This, combined with the mask gene to explain why he has virtually no red on his face, could very well be the truth behind Ebon's rusty legs.

So, all in all, I find the black and brindle theory to be the most plausible. Black and tan doesn't fit very well at all and the bad black theory is based on a variant that may not even be present in the Labrador breed. Much of the reason why I went with the bad black theory is his Guess the Genotype post was purely for the sake of argument. It is still possible that Ebon's coloring could be explained this way, but black and brindle is more likely. I only wish I could afford to test him to abate my curiosity.

Mismark Case Study: Cavalier King Charles Spaniel

According to the AKC, Cavaliers come in four allowable colors, three of which are seen here: Blenheim, tricolor, and ruby. Black and tan is not shown.

The Cavalier King Charles spaniel is a popular pet and is one of the largest of the breeds in the toy group. Though the breed comes in four colors, two are more commonly seen: Blenheim and tricolor. There are also several mismarks, some of which are due to breeding within the allowable breed colors:

• Incorrect white
• Too much white on the face of a Blenheim or tricolor
• Any white on a ruby or black and tan
• Ticking 
• On a Blenheim or tricolor
• Incorrect tan 
• Seen on tricolors and black and tans 
• Incomplete pigment 
• Of the nose or eye-rims on a Blenheim or tricolor
• Blue eyes 
• Usually only seen on a Blenheim or tricolor

White on a black and tan

When it comes to incorrect white, both of the forms I describe are really to be expected when breeding for the allowable colors within the breed. The piebald gene (sp), which results in Blenheim and tricolor dogs, is recessive to the solid gene (S) on the Spotting locus. However, in most cases a dog who is heterozygous for these genes (Ssp) and thus carrying piebald will have some amount of white. This is often as little as a bit of white on the chest and/or toes, but this may extend up the chest and even result in some white on the face and tail. The problem is: according to the standard, no amount of white is allowable on ruby or black and tan dogs. This means it is unlikely that the solid dogs are bred to the piebald dogs, leading to color class breeding. This is problematic, as it basically creates breed within breeds, narrowing overall genetic diversity to two smaller pools, and thus raising the potential for problems that result from increased homozygosity. Also, only breeding solid dogs to other solid dogs will not eliminate white markings. It's far from uncommon for a genetically solid dog to end up with some white on their chest and/or toes. This is thanks to normal developmental variation in the womb.

Too much white, lacking pigment

In addition to the above, "incorrect" facial white on piebald dogs is also far from uncommon. Facial white is controlled by as-of-yet unknown modifiers. Selection can favor one form of facial white, but it's almost impossible to eliminate the ones that would be considered "undesirable." Since this is true, dogs with more or less than the preferred amount of white shouldn't be surprises. Also, since selection in Cavaliers favors dogs with rather a lot of facial white, this likely increases the likelihood of a dog being produced that has too much white. In relation to this, dogs with a lot of white on the face are far more likely to lack pigment of the nose or eye-rims, or to have one or more blue eyes. Since the standard favors dogs with so much white on the face, this makes it quite likely that these dogs will appear.

Heavy ticking, too much white

As for ticking, there is a bit of a conundrum in the breed. Some ticking is fine, and a lozenge between the ears, which may or may not be a large ticking spot, is preferred. But, if a dog has too many ticking spots, it's not allowed. So, it's likely that the dogs that are allowed are heterozygous for the ticking gene (Tt) and the dogs that would have too many spots are homozygous (TT). So, why allow ticking at all if you're only going to penalize dogs that have too many spots?

Incorrect tan on the face

Another mismark that is based on normal color variation in incorrect tan markings. Taking the standard as it is stated, any tan on the outside of the ears is not to be allowed and any tan between the marking above the eyes and on the cheeks would also be incorrect. However, these things are not uncommon in tan pointed dogs. This is another case of minor differences being enough to make a dog "undesirable."

All of the mismarks that are seen in the Cavalier King Charles spaniel are really rather minor things that are to be expected from normal canine color inheritance. As such, they are nothing more than silly aesthetic preferences and do not reflect on the quality of the dog. Since this breed is a very common pet and, indeed, bred to be a companion, temperament and health should be far more important than what the dog looks like. Looks should never take precedence over these other, far more important traits. As such, color standards are really rather silly things based on nothing more than minor aesthetic preferences of the few people that wrote the standard. As the differences I have mentioned have absolutely no effect on the quality of a dog, eliminating a dog from the gene pool simply because it does not fit the color standard is ridiculous. Also, as purebreds are bred in closed registries, dropping any dog from the pool of potential mates makes it quite possible, and indeed likely, that a negative loss of genetic diversity will result. Since the color standard has also lead to color class breeding in the Cavalier, this is even more problematic.

Images are from Wikimedia Commons and Flickr.com under Creative Commons licenses: one, two, three, four, five.

Mismark Case Study: Gordon Setter

A Gordon setter in the breed's only acceptable color.

The Gordon setter is one of a small number of setter breeds, which also includes the English setter, Irish setter, and Irish red and white setter. Though the Gordon setter now only comes in one acceptable color, the breed's history included a number of other colors that are now considered to be mismarks. Part of why these colors are in the breed is due to its relationship with the other setters. So, what are these mismarks?

• Incorrect tan markings
• Dogs with more or less than than required in the standard
• Liver
• Inherited on the Brown locus, a dog must be bb to be liver
• Red
• Inherited on the Extension locus, a dog must be ee to be recessive red
• Too much white
• Inherited on the Spotting locus, a dog with a variety of genotypes can have too much white

Too much white

Looking at these mismarks, they are all recessively inherited except in dogs that are genetically solid but have too much residual white. All of theses colors were well known when the breed was young. Much like the Irish setter, the predominant color in the early years of the breed is actually not what you think it would be when looking at modern dogs. Gordons were once mostly tricolor with some dogs being solid black and tan, liver, or red, but the white markings and other colors fell out of favor and led to the production of the breed you see today.

This dog appears to be liver

The current breed standard for the Gordon only allows for black and tan dogs with specific tan markings. Dogs that are anything other than black and tan are disqualified and anything more than a small bit of white on the chest is not allowed. A dog with more or less than the required tan would be penalized, despite the fact that ten markings can vary greatly on dogs that are all genetically atat tan pointed. So far, it is known that there are modifiers that control this amount of tan, but it isn't known where they are or how they are inherited.

Too much tan

This is a case where color standard is based on, basically, fashion. What once was popular was no longer liked by those who wrote the breed standard, and thus those other colors faded into obscurity. However, since the colors are basically all recessively inherited, they continue to pop up on occasion in litters that are born today. These past decisions are really problematic when looking at the breed's history and what this holds for the future.

Too little tan

Since Gordon setters were originally hunting dogs, the main thing that they should be bred for is working ability. This was at least true of setters past. However, modern Gordons have a lot more emphasis placed on type and show potential. It remains true that Gordons were developed as working dogs and color has no affect on working ability. It also has no affect on health or temperament, which are both important to people who wish to have a happy life living with a dog, particularly if the dog is meant to be a pet. If breeders wish to emphasize the Gordon setter's place as a working breed and its ability to do field work, why should they place any emphasis on color?

Another major concern in the Gordon setter and any other breed is a loss of genetic diversity due to selective breeding. Since Gordons of only one color were favored, a lot of diversity was lost rather early on in the breed. Continued shunning of off-color dogs only drops more potential sources of genetic diversity from the breed and risks loss of some very good, sound individuals based on a silly cosmetic preference.

Sources are the American Kennel Club, Gordon Setter Club of New South Wales, and Gambit Gordon Setters. Images are from Wikimedia Commons or Flickr.com under Creative Commons licenses: one, two, three, four, five.

Mismark Case Study: Boston Terrier

Boston terriers come in three acceptable colors: black, brindle, and seal, all with white.

This dog would have too little white

The Boston terrier is a breed that came from crossing a number of different breeds together. In its ancestry are such breeds as the French bulldog and bull and terriers types, crosses between bulldogs and terriers that are the ancestors of such modern breeds as the bull terrier and pit bull. The reason why modern Bostons are so small is that they have since been bred down in size from the original stock. This ancestry also explains all of the colors that pop up on occasion in the breed. What are these mismarks?

• Too little white
• Lacking the required white chest, muzzle and blaze between its eyes
• Too much white
• Excessive facial white or "splash white"
• Any base color other than black, brindle, or seal 
• Liver, blue, lilac, fawn, gold, or tan point
• Not enough skin pigment
• The nose must be completely black
• Blue eyes
• Even a touch of blue disqualifies

Too much white and a blue eye

To begin, let's look at variations in the amount of white on the dog. Facial white is forever a conundrum in the canine world. In every breed that had facial white, there is a great amount of variation. This variation is controlled by as-of-yet unknown modifiers and is likely polygenic and inherited mostly independent of the Spotting locus that controls body white. This is the issue with having a rather strict requirement for facial white: there is no real guarantee that your breed will always have the same exact amount. This is why in breeds like the Boston terrier, too much white on the face happens so often. Too little white also happens, but in the Boston this appears to be a more unusual occurrence.

A splash rescue

The "splash" white is completely different. Piebald and extreme white can both be sources of the splash pattern, and both are recessive alleles on the Spotting locus. Most Bostons express the Irish white phenotype, which can vary from the minimal markings required by the breed standard up to what's often called collared Irish, which has white extending all the way around the neck and prominent white on the legs. Irish white is dominant to both piebald and extreme white, and as such it's very possible for dogs who conform to the standard to carry these alleles. Though not always the case, it's possible that some of the heavily marked collared Irish dogs have as much white as they do thanks to the recessives they carry. Extreme white especially can cause a dog to have more white than it would otherwise have. Minimal piebald can also mimic collared Irish markings, and so it's possible the occasional minimal piebald may slip into the show ring.

As for the dogs with off-standard base colors, there are really quite a number of possibilities. Here's the full list of possible colors, what they're called in the breed and the genes that cause them. Wherever I say black, this is basically interchangeable with brindle and seal. All three colors are caused by the K-black locus, with brindle (kbr) being recessive to black (K) or seal (K, bad black, possibly recessive to regular black).

Masked fawn puppy

• Sable/fawn 
• Recessive to black on the K-black locus, allows the Agouti locus to show through: Ay- kk results in sable/fawn
• Dog may or may not have a black mask and may be liver, blue, or lilac instead of black
• Tricolor/tan points
• Also on the Agouti locus: atat kk results in tan point; may be liver, blue, or lilac instead of black

Liver pigment

• Non-black pigment that would otherwise be black
• Liver/brown, recessive to black on the Brown locus: bb results in liver pigment
• Blue/gray, recessive to black on the Dilution locus: dd results in blue pigment
• Lilac/champagne, combination of the blue and liver genes: bb dd results in lilac

Recessive red

• Gold/honey/cream/blonde
• Caused by recessive red on the Extension locus: ee results in recessive red/clear red
• This gene hides all black body pigment, so any of the above, including blue/liver/fawn, but skin and eyes can still be effected by the liver and blue genes

Liver pigment and too much white

The major thing to note about all of the above is that they're all recessive. Due to this fact, you won't know your dog is a carrier until it throws a mismark puppy. Even if you remove those mismark puppies from your breeding program, its siblings have a two-in-three chance of being carriers. Why is that? It comes down to basic genetics: mismark puppies can only show up in litters bred from carrier parents or those expressing the gene. If breeding from two carriers, the expected ratio in the puppies is 1:2:1. That is, about one-forth will be non-carriers, half of the litter will be carriers, and the last forth will express the recessive mismark. If you remove the mismarks, you're left with a ratio of 1:2, or one-third non-carriers, two-thirds carriers. This means there's a rather high chance of the gene persisting, particularly if a breeder then chooses to linebreed or do other forms of inbreeding. Inbreeding only increases the likelihood that the recessive mismark will appear. This is why so many breeds have persistent mismarks.

Two dogs, three blue eyes

The two mismarks that I haven't yet mentioned are blue eyes and noses lacking in pigment ("Dudley nose"). Both are disqualifications according to the standard, and both are natural consequences of breeding for dogs with lots of white. White around the nose makes it quite possible that a dog will end up with a pink spot on its nose, or even a completely pink nose. So, expecting all dogs to have a fully pigmented nose when you are breeding for prominent white markings is a bit silly. It's basically the same with blue eyes. Though they can be independently inherited, the more white a dog has, the more likely it is to have one or more blue eyes. If the white touches a corner of the eyes, this also increases the chances. This is why so many dogs with a lot of white on their face have one or more blue eyes. Since the Boston standard desires very prominent white, blue eyes shouldn't be a surprise.

Liver pigment

As with every dog, color should be the last thing a breeder should worry about. There are so many more important things that a breeder should be striving to produce, like dogs with even temperaments and ones that are healthy. By disallowing certain colors, especially ones that were perfectly acceptable at one point in the breed's history (like splash in Bostons), there's a very high likelihood that perfectly good, healthy dogs are being dropped from the breeding population for extremely superficial reasons. This does nothing but narrow the gene pool and damage the breed as a whole. However, this doesn't mean that one should purposefully breed for mismarks. On the contrary, breeding for recessive is always bad because it usually means inbreeding, and potentially very heavy inbreeding at that. Inbreeding in any form and by any name (such as linebreeding) is very bad as it narrows gene pools and makes it more likely that deleterious recessives pop up. This is why so many purebred dogs have inherited health problems. Color should not matter. Purebred dogs have enough problems.

Apparently, there are some people out there selling merle Bostons. This color has been introduced through mixing with another breed. Since merle has so many negatives, there is absolutely no good reason to out-cross to purposefully introduce merle into a breed. 

Images are from Wikimedia Commons or Flickr.com under Creative Commons licenses: one, two, three, five, six, seven, eight, nine, ten. Image four is copyright to Rescue Furdaddy on Flickr.com.