Already know what you're looking for?

Quick Links

The E Series


"Extreme" Masking and Etching

Recessive Red

Grizzle and Domino (Eg)

The Eh Allele

Quick Summary

Further Info

The E Series

"E" stands for "extension", and the E series is responsible for almost all non-agouti (A series) related eumelanin/phaeomelanin patterning in dogs. In other words, E helps to determine which areas of the coat can (and do) produce eumelanin (black/blue/liver/isabella) and which can (and do) produce phaeomelanin (red).

The dominance hierarchy of the E locus isn't known for sure yet, and there may also be as-yet-undiscovered alleles, but currently it's thought to look something like this:

Em - masked (the mask appears over the top of the A locus pattern - see below)
Eg and Eh - grizzle/domino
E - "normal extension" (no effect on the dog; all areas of the coat able to produce both pigment types, so pattern is then determined by A and K locus)
e - recessive red


Eumelaninistic masks (black/liver/blue/isabella on the muzzle and sometimes the ears) are caused by the Em allele. It is the top dominant of the E series, so that means a dog only needs one Em allele to have a mask, regardless of which other E locus allele it carries. Genotypes for masked dogs in most breeds are, therefore, as follows:

EmE - one mask allele, one normal extension allele
Eme - one mask allele, one recessive red allele
EmEm - two mask alleles

Masks can appear on any dogs that are genetically sable, tan-pointed, saddled or agouti (wolf grey). That means that the presence of a mask on a dog with the mask gene depends on the A and K series. In order to display a mask, a dog cannot be dominant black (KK, KKbr, Kk) on the K locus, because dominant black stops the A series from being expressed, and obviously, a black mask on a black dog is not going to be visible. If a dog is brindle or non-solid black on the K locus, they can express the A locus, so will be sable, agouti, saddled, tan-pointed or recessive black. All of these will show masks except for recessive black (again, a black mask isn't going to be visible on a black dog, even if it is technically there!).

Masks consist of eumelanin pigment, so they can be affected by any gene that changes the colour or intensity of eumelanin. This means a mask will follow the main pigment of the dog and can be liver, blue or isabella, and can also show the merle pattern as well. There is one exception though - sometimes masks aren't affected by greying (at least, not by the greying gene found in breeds such as the Kerry Blue Terrier). This is a bizarre anomaly, and one for which there is currently no explanation.

On a tan-pointed (atat) dog, masks can be detected by looking at the facial points. Tan normally occurs on the sides of the muzzle and above the eyebrows, but a dog with a mask may have all or part of these points covered up by the main coat colour. Usually, however, the tan on the neck is usually still visible.

Masks can vary greatly, covering anything from just the end of the muzzle to the whole of the muzzle, eyebrows and ears.

On sable dogs, black masks may also cause black hairs on the chest and/or the back and tail. Sables with black tipping often have black masks, so it is possible that masks in fact cause some of this tipping. However, there is probably another modifier that helps with this, because not all black-masked sables have extra black, and indeed, many breeds with masks come only in clear or almost clear sable (e.g. Pugs).

Normal extent of variation in masking in dogs.

Examples of Masks

Sable Leonberger with a black mask.

A variety of sable dogs showing the variations in extent of black masks. Note the black extending down the front of the chest on the Tervueren. Any of these variations may also be brindled.

Three dogs with masks fully or partially covering the tan points on their muzzles.

An Akita with white markings covering part of its black mask.

Whippet and Tibetan Mastiff with blue (dd) masks. On the Tibetan Mastiff the mask covers up the tan points on the muzzle.

"Extreme" Masking and Etching

In a small number of breeds, masked dogs may have large amounts of shading on the chest and legs. The genetic basis of this is not known, and it could either be a modifier extending the mask, or a variation of the masking allele itself. The shading occurs on the underside of the dog, making it distinct from the shading on the top/back of the dog typically caused by sable (Ay).

There is a possible connection between extreme masking and "etching", which is when a sable piebald appears to have eumelanin edges to its patches, or when an irish spotted dog has eumelanin "borders" between its white feet/legs and main body colour. This is commonly seen in Akitas, St Bernards, and Anatolian Shepherd Dogs, all of which also often have heavy masking. It's possible that extreme masking causes the eumelanin-producing melanocytes to spread out to the edges of any phaeomelanin areas, which on a dog without white markings will be the legs/feet and chest. On a piebald or irish spotted dog, this will be the edges of the coloured areas.

This Bullmastiff crossbreed shows clear black chest markings on a short-haired dog. Interestingly, these markings follow almost the same pattern as tan points or urajiro. Note the grey and black shading on the legs and feet too.

The three Malinois here, owned by Chelsey Protulipac, show how much the black leg and chest markings can vary. However, even the dog on the left, which has the least amount of shading, has some black on the inside of the hind legs and the underside of the body.

Argos and Riddick the Leonbergers have both mask-related shading (on their chests and legs) and sable-related shading (on their backs). These handsome dogs are owned by Cheryl Heinly and bred by Kathryn Wilson of The Leos Den.

In some cases, the shading may fade as the dog ages. Kiwi, above, shows this nicely. Kiwi's photos were submitted by Elaine Senez Ableidinger.

Extreme masking can also "hide" tan points. These photos of Astro, submitted by Oona Pesonen, show what this may look like. Astro looks almost solid black, but you can just about make out tan markings on his legs and underside. It's probable that the tan markings that should be on his chest and face have been covered up by the kind of extreme masking seen in the Malinois above.

This St Bernard shows etching around the edges of its patches, which may be caused by extreme masking. Leonbergers, which were bred from St Bernards, display extreme masking, so it's very likely that it's present in St Bernards too.

Recessive Red

The other two E locus alleles found in many breeds (other than sighthounds and Cocker Spaniels - see below) are E (normal extension) and e (recessive red). Normal extension simply allows a dog to express its other genes on other locii normally. To put it simply, it has no effect on the dog. Recessive red, however, produces a much more visible effect.

A dog that is homozygous for recessive red, so has the genotype ee, will be completely red. Its nose will remain black (or liver, blue or isabella, whatever its eumelanin pigment colour is), and so will its eye rims and lips, but the rest of its coat will be solid red (with or without white markings and ticking). This is because the recessive red gene, in effect, gives the skin cells a disability (but not a harmful one!) - it stops them from being able to produce eumelanin. It is therefore impossible for a recessive red dog to have any black (or liver, blue or isabella) in its coat.

Recessive red, although recessive in its own series, is essentially dominant over almost all other locii. Dominant black, sable, tan points, wolf grey, merle, and any other pattern with black in it will be turned to solid red by the recessive red gene. This, in a way, is the danger of recessive red - it can mask so much.
It's impossible to know from looking at a recessive red dog whether it carries sable, tan points, brindle, or any other A or K locus gene, and so impossible to know what it will pass down to its puppies unless you know its family history. A recessive red dog could even be merle while still appearing solid red, and this is a problem that has arisen in Pomeranians, among other breeds. Merle is a recent addition to the Pomeranian breed, and it is entirely possible for double merles to be bred by accident because recessive red is so common in the breed. The only way to tell if a recessive red dog is merle is if it has blue or partially blue eyes (which not all merles do), so it would be easy to accidentally breed two merles together if you weren't aware of how merle can be hidden. See the merle pages for information on why merle to merle breedings are dangerous.

Recessive red only occurs in particular breeds, but many of these breeds also carry sable. The problem is that a clear, unmasked sable can look identical to a recessive red. This means that solid red dogs must undergo genetic testing to determine whether they are sables or recessive reds; there is no other reliable way to tell the two genes apart. However, if a breed always comes in solid red with no black hairs at all, and never comes in shaded or tipped sable or with a mask, you can be relatively sure that the breed carries recessive red (although remember a recessive red dog can still carry sable, just not express it! There is debate about this though - Little, whose work most modern dog genetics is founded on, believed that a puppy carrying sable and recessive red would miscarry. To my knowledge, this has not yet been proven or disproven). Golden Retrievers and Labradors are two such breeds, and genetic testing has shown that they do indeed carry only recessive red (along with a few other gundog breeds). There are a number of breeds, though, that carry and express both sable and recessive red, such as the Pomeranian and the Dachshund.

There has been a hypothesis suggested (by Sue Ann Bowling - that recessive red and sable can be distinguished by whisker colour. A recessive red dog would have red whiskers according to this hypothesis and a sable dog would have black. This has not been proven, but seems to make sense. However, this wouldn't necessarily help with a liver-pigmented red dog as the whisker colour could potentially be too close to the red coat to be able to be distinguished easily.

All three of these dogs are probable recessive reds (and the Labrador probably carries an I locus allele as well, which has turned its red into cream). We know the Labrador must be genetically dominant black, because all Labradors are (see the Labrador case study in the introduction section). The recessive red gene overrides the dominant black, turning it to red.

The Dachshund is a breed that comes in both sable and recessive red. It's therefore impossible to tell which these three dogs actually are! The first dog has a little black on its tail and the third dog does appear to show a few black hairs on its ear, in which case both are sables, but the middle dog has no visible black hairs at all and is potentially a recessive red.

Although the recessive red allele does not directly affect the colour of the nose, eye rim, lip and nail pigment, it is common for this pigment to fade from black to grey or pink as a recessive red dog ages. This may well be another indicator that a dog is recessive red rather than sable - although as sables may also sometimes have poor pigment, it is not foolproof.

Above are examples of (presumably) recessive red dogs with pigment loss. Note how in all the examples, the lip pigment seems to stay darker than the nose or eye rims. In some dogs the whole pigment fades to a grey/blue, and others develop a "dudley" nose, where there is a strip of pink in the centre. It can be easy to mistake some of these dogs for blues, but when compared to a true blue (see Dilution/D-locus page), the difference should become clearer. Of course, it is entirely possible for a recessive red dog to also be a dilute - but in this case the dog would be born with blue pigment.

Liver (bb) recessive red dogs show more significant pigment loss than black (BB) recessive red dogs. We know this Bracco Italiano is a liver because the breed does not come with black pigment at all. Note how there is very little nose pigment - it is almost entirely pink. This can also be seen on breeds such as the Pharaoh Hound and Cirneco dell'Etna. In fact, most recessive red dogs with liver pigment will appear to have pink pigment. It is possible that the recessive red gene affects liver eumelanin to a greater extent than black eumelanin.

The Nova Scotia Duck Tolling Retriever comes in both black and liver pigment. The dog on the left shows black (but faded) pigment, and the dog on the right is a liver (note the bright pink again). Tollers can be phenotypically either sable or recessive red. Without genetic testing it is very difficult to tell, but for our purposes we will assume both dogs above are recessive reds as they have clear coats (no shading). Remember that liver shading can be very difficult to distinguish on a deep red coat though, adding an extra layer of difficulty to establishing the genotype of a dog such as a Toller. The faded pigment on the dog on the left is a very strong indication that it is a recessive red as this would be much more unusual in a sable.

Grizzle and Domino

Recently two new alleles have been discovered on the E locus - Eg and Eh. The Eg gene is specific to the Saluki, where it is known as grizzle, the Afghan Hound, where it is known as domino, and the Borzoi, where it is usually called sable (not to be confused with the normal sable allele Ay). It has not yet been identified in any other breeds, although it is possible it also occurs in other related sighthounds.
Grizzle looks very much like a shaded sable or creeping tan (see A series), and follows a specific pattern with a widow's peak on the head. The shading colour varies from black/grey to brown. Sables in breeds such as the Rough Collie look very similar to grizzle Salukis but the gene is not thought to be the same.

Grizzle is dominant over all E locus alleles other than Em, which means a grizzle dog can never have a mask (an EmEg dog would have a mask but not be grizzle). Grizzle can also only occur on dogs that are atat on the A locus. In other words, the dog must have the tan point allele as well as the grizzle allele. A dog with the grizzle allele (Eg) but not the tan point allele will not be grizzle and will just express its A locus normally. This means that breeds such as the Greyhound may carry grizzle but never express it as they do not have the tan point allele.
In this way, Eg can be seen as a modifier of at, in the same way that harlequin (H) is a modifier of merle. Is it possible that other hypothesised modifiers of at, such as creeping tan and the saddle pattern, are also located on the E locus? This is something we don't know yet, but it is certainly a possibility.

Dark "sable" (Eg) Borzoi, submitted by Emily Leamon, and medium-shaded grizzle Saluki, submitted by Lauren Johnson. This shade in Salukis is known as "deer grizzle".

Kryslaur's High Flyer (Flyer), submitted by Shawn Snead, is a "domino" Afghan Hound, caused by the same Eg allele as the Borzoi and Saluki above. His red areas have been "diluted" to white (see I locus page). However, Flyer is also a brindle. The effect of brindle on grizzle is fascinating, as it seems to produce the opposite effect to brindle on sable or tan points. Usually brindle is confined to red (phaeomelanin) areas and does not appear on black (eumelanin) areas, however on a grizzle it appears to do the impossible. As you can see, Flyer's black areas are brindled but his red areas (points) are not. Some grizzles will show brindling on their tan areas, but in almost all cases it fades towards the feet.
This is interesting because it suggests that the red/tan areas on a grizzle are essentially the same as recessive red - i.e. the cells are unable to produce eumelanin (black) pigment. It may be that the grizzle pattern is caused because the ability to produce eumelanin is reduced towards the outer "points" of the dog (muzzle, chest, legs etc) due to a fault in the allele. If this is true, grizzle could be considered "partial recessive red".

The photos above, taken by Tara Meikle, show the same phenomenon in Salukis.

The Eh Allele

Eh is an allele recently discovered in English Cocker Spaniels. Cockers with this allele are known as "sables", although again this is not the usual Ay sable found in most breeds. Eh is almost identical to Eg, and just like Eg it is a modifier of tan points. All Cocker Spaniels are "fixed" for tan points (meaning they are atat), however only some carry Eh.

There is one important way in which Eh (Cocker sable) differs from Eg (sighthound grizzle/domino). While all grizzle Salukis tested so far have been kk on the K locus (non-black, able to express A locus), all sable Cockers tested so far have been Kk or KK (dominant black). kk Cockers with the Eh allele have been reported as "dirty red", not sable. This suggests that Eh works differently to Eg as it appears to "override" dominant black.

These photos show the sable English Cocker Spaniel "Andara vom Luthersbrunnen", taken by Elke Gildemeister. Note how the pattern is extremely similar to the Eg sighthounds in the section above.

Quick Summary!
No time to read the whole thing? Here's the quick version!

There are five known alleles on the E (Extension) locus - Em (eumelanin mask), Eg (grizzle/domino), Eh (Cocker sable), E (normal extension), and e (recessive red).

*- Em produces a eumelanin mask on the muzzle and/or ears (only visible on sables, tan points and agoutis - see A locus page). Sometimes this may also spread to the tail and the front of the chest.
*- Eg is known as grizzle or domino and occurs only in certain sighthound breeds. It is a modifier of tan points, so can only be expressed when the dog is also atat on the A locus.
*- Eh is very similar to Eg but occurs only in Cocker Spaniels and also affects KK and Kk (dominant black) dogs.
*- E (normal extension) has no effect on the phenotype.
*- e is recessive red, which turns all eumelanin in the coat to phaeomelanin. Recessive red dogs will always be solid red (with or without white), regardless of which alleles are present on the other locii, meaning recessive red can even mask merle. It is genetically impossible for a recessive red to have any black/liver/blue/isabella in its coat, although eumelanin in the eyes and nose is generally unaffected.

Further Info and Links

The E locus in dogs has been mapped to MC1R (Melanocortin Receptor 1), which is implicated in coat colour/pattern in a large number of species. MC1R tells melanocytes to make eumelanin, so in the absence of a signal from MC1R the melanocytes will produce phaeomelanin instead. This means that the E locus controls which cells produce eumelanin and which do not (in the case of recessive red, no melanocytes receive a signal to produce eumelanin, hence no eumelanin can be produced in the coat).

MC1R is responsible for ginger hair in humans, recessive red in mice, and black vs red in both horses and cattle. Only one E locus mutation affecting coat colour is known in cats however, and this occurs only in one breed (Norwegian Forest Cats).

Links to studies:
A new mutation in MC1R explains a coat color phenotype in 2 "old" breeds: Saluki and Afghan hound:
MC1R studies in dogs with melanistic mask or brindle patterns:

** Please note that I am not a research scientist, and the information on this page comes from my own knowledge and observation of dogs, observational and testing data provided via e-mail by site visitors, any research papers linked on the page, and the information provided by Dr Sheila M. Schmutz on her excellent website

For further genetics resources, see the Links page