Border Collie Colours


**Please Note - All text and images on this website are copyright to Bryning Border Collies unless otherwise stated and may not be distributed or reproduced without permission**

While everybody usually thinks of the border collie as being black and white they do come in quite a wide variety of coat colours (as you will see from our own dogs). With all colours the traditional ‘classic’ markings are a white blaze up the face, white chest and forelegs, a white collar around the neck, white hind feet and a white tip to the tail, these markings are often referred to as 'Irish Spotting' although they can vary greatly, for showing purposes in the UK any colour and markings are permissible but the white should not dominate.

Coat colour is controlled by a number of dominant and recessive genes inherited from the parents. For any given characteristic, including colour, a puppy will inherit one gene from the father and another from the mother and the distribution of these genes may vary between puppies in a litter. I have tried to explain this briefly below followed by some examples of the variety of colours present in border collies with a very simplistic overview of their genetic makeup.

Colour genetics (the basics!)

I've added this bit because so many people ask me about this and I thought it would be easier if I had somewhere to refer them rather than keep repeating myself.

I will try and describe this in the simplest way as I understand it but it is worth remembering that while the understanding of colour genetics is very useful when wishing to breed puppies of a particular colour, the colour should be the last consideration when planning a mating...the breeders requirements for health, temperament and type must all be fulfilled before colour is taken into account and when you find all the things you want, the colour is the icing on the cake!

Ok...here we go...

The hair follicles are made up of cells containing two colour pigments – eumelanin (black) and phaeomelanin (true red); so basically eumelanin is the pigment that produces the darker black and brown shades, phaeomelanin produces the true red (golden red). There are a vast number of genes that control the production of these pigments within the cells and therefore control coat colour.

Firstly remember that there are always two of every controlling gene, one comes from the sire and one from the dam. Where a gene is expressed as a capital letter it is dominant, where it is lower case it is recessive.


Where an animal has two identical genes this is called homozygous and where they are different it is heterozygous.


Where a gene is dominant the animal only needs a single copy from either parent to express that colour. Where a gene is recessive an animal MUST have two copies of that gene in order to express that colour; so they must get one copy from either parent. In this way recessive genes can be ‘carried’ i.e. the animal only has one copy so does not express that colour but has the gene to pass on to it’s offspring.

Where an animal is heterozygous for a particular gene (i.e has one of each) it will statistically pass the dominant copy to 50% of it’s offspring and the recessive copy to the other 50%. Where an animal is homozygous for a particular gene (i.e. has two copies the same) it will always pass this copy on to ALL it’s offspring.

Black/Brown - 'B'

So lets say a ‘normal’ animal is black…in a chocolate/brown animal the eumelanin pigment granules are a slightly different shape which makes the coat appear brown. The gene said to be controlling this is expressed as B for black and b for brown. Brown is a recessive gene so both parents must have the gene to produce it in a litter, Black (B) is dominant and where present it will always be expressed.

So… a brown/chocolate dog will be bb (homozygous brown)
a black dog carrying brown/chocolate will be Bb (heterozygous black)
and a black dog not carrying brown/chocolate will be BB (homozygous black)

Merle - 'M'

Merle is a dominant gene so is expressed as M – merle; m – non merle
So only one copy of the gene is required for a dog to appear as a merle and therefore only one parent need be a merle in order to produce the colour in a litter. Merle CANNOT therefore be carried (it is a common misconception that animals with one merle parent will carry the merle gene…they cannot unless they themselves are merles!!)

The merle gene acts on the eumelanin pigment causing various grades of colour dilution and therefore the patchy colour appearance…merle can therefore act on both black and brown.

The vast majority of merles are therefore heterozygous (Mm) i.e. only have one copy of the merle gene. When two merles are mated together you will get a proportion of the litter (~25%) that are homozygous merles (MM). There are quite a large number of health issues caused by this genetic combination…deafness, poor eyesight, malformation of the eyes, infertility etc. These animals are usually fairly easy to spot, due to the double merle influence they are often excessively white with very little colouring on them and have unusually small eyes. Because of these associated problems two merles should not be mated together and no responsible breeder would even contemplate a deliberate merle to merle mating. There are no known health issues related to the normal heterozygous merle.

So… a normal healthy merle - Mm (heterozygous merle)
Non merle - mm (homozygous non merle)
Double merle - MM (homozygous merle)

Dilution - 'D'

The dilution gene D is again recessive and acts by slightly altering the shape of the pigment containing cells in the hair follicle and therefore affecting the pigment granules. The dilution gene acts on the Eumelanin pigment so dilutes Black to Blue and Brown/Chocolate to lilac.

So… dilute (blue/lilac) - dd (homozygous dilute)
Carrying dilute - Dd (heterozygous non-dilute)
Not carrying dilute - DD (homozygous non-dilute)

The dilution gene can act with the merle gene to give slate and lilac merles.

Extension - 'E'

It is the extension gene that is responsible for producing the true red (Australian, ee, golden red) colour.

The extension gene controls the extension of the two pigments in the hair follicle. E is dominant and allows the extension of eumelanin, so the dog is black or brown. e is the recessive form and allows the extension of phaeomelanin but prevents the extension of eumelanin, so there is no eumelanin in the pigment and therefore the animal is red. So the red colour is produced from a recessive gene so the same principles apply, a dog must be homozygous to exhibit the colour and must get one gene from each parent.

So… true red (golden, Australian red) - ee (homozygous red)
carrying true red - Ee (heterozygous non-red)
not carrying true red - EE (homozygous non-red)

What is interesting about this gene is that it will mask the dogs ‘real’ colour. As such you can get black, chocolate, blue and lilac dogs that all appear as red because of the lack of eumelanin in the hair follicles. As it only affects the hair follicles and not other pigmentation the dogs’ nose and eye rims will express its hidden colour. There is also an added danger here…the merle gene cannot be expressed on a true red dog (remember merle acts on eumelanin to) so it is possible that when red puppies are born from a merle parent they may well be merles but you simply cannot tell. Don't be fooled by those people that claim they can see merling on ee red puppies...this is simply not chemically/physically possible.

The danger is that these puppies could then be inadvertently mated to a merle down the line…a merle to merle mating which, as we aready know, may result in numerous birth defects. As such most responsible breeders would NEVER mate a true red to a merle.

Agouti - 'A' and Dominant Black - 'K'

This is the gene responsible for producing sable, seal, saddleback pattern and tricolour in border collies...there are numerous alleles of this gene and it is still not fully understood but I am concentrating on those that concern border collies.

ay - sable
at - tanpoint - recessive to ay (as such a tricoloured dog cannot carry sable)

As we already know each individual has two copies of each gene but this isn't quite as straightforward as some of the other genes we've mentioned...these genes exhibit variable incomplete dominance...

Genetically pure for sable (ay-ay) is generally clear sable in colour, while heterozygous (ay-at) is usually sable with some extent of darker fur in the coat (shaded sable). This is incomplete dominance (one copy of the recessive allele produces some effect).

ayay - clear sable
ayat - shaded sable
atat - tanpointed

But...it doesn't end there...there is another gene that effectively switches the agouti genes on and off...this is referred to as dominant black and is written as 'K'

It is thought that ALL border collies have the genes for sable and/or tanpoint but their ability to express and produce these colours is not only determined by their presence and combinations but also by those of the Dominant black 'K' genes.

KK - Homozygous dominant black, cannot be or produce sable/tanpoint regardless of the A alleles
Kk - Heterozygous dominant black, cannot be sable or tanpointed but can produce the patterns it carries on the A alleles
kk - Homozygous, will exhibit sable or tanpoint depending on it's A alleles

In conclusion...
KK atat - neither sable nor tanpointed, cannot produce sable or tanpoint but can pass the tanpoint genes on to offspring
KK ayat - neither sable nor tanpointed, cannot produce sable or tanpoint but can pass the tanpoint or sable genes on to offspring
KK ayay - neither sable nor tanpointed, cannot produce sable or tanpoint but can pass the sable genes on to offspring

Kk atat - carries tanpoint
Kk ayay - carries sable
Kk ayat - carries sable and tanpoint

kk ayat - shaded sable coat colour (also carries tanpoint)
kk ayay - clear sable coat colour
kk atat - tanpointed coat colour

Any solid colour can exhibit sable or tanpoint pattern (black, chocolate, blue, lilac); the pigmentation and eye colour of the dog will usually reveal it's base colour in the case of sables. Because the sable gene affects the extension of eumelanin (similar in effect to the extension gene) in it's extreme (clear sable) merling will not be visible, as such merle to sable matings are generally not a good idea.

**update 30.03.07**
The assumption by breeders, that the differences in the degree of shading in sable dogs is determined by whether they are homozygous or heterozygous for ay, is now known to be not entirely accurate; it is believed that some to all of this variation is caused by another (as yet unidentified) gene. DNA tests to distinguish this allele "a" in Shelties, Tervuren/Groenendael, and a few other breeds are now offered by Healthgene based on recent findings.

The agouti genes are fairly complex and are also responsible for a number of other patterns in border collies and other dog breeds, such as saddle pattern, wolf pattern, chinchilla and seal.

Since producing a suprise Seal/White pup in March 2007 the research I have done has discovered that the most likely explaination for this is that an additional modifier (possibly linked to the dilute gene) allows the sable pattern to show through faintly when a dog is heterozygous for dominant black (Kk ayat + modifier) producing a sort of 'ghost' sable pattern on the coat. This phenomenon has also been reported with ghost tan point and saddleback markings too. The fact that seal is usually produced with sable and/or tanpoint in the same litter, and the anecdotal evidence about the way it breeds on, would support this theory.

Seal is still a relatively rare and not widely recognised colour in the breed. See the pics below of Wren (Bryning Goldfinch) to see how the colour develops as the pup grows.

Border Collie Colours...

Black & White

Black is a dominant gene and the black and white colouration will always be produced unless other modifying genes are present.

pictured below...our very own Lloyd

Image: Locheil True Scot

Black Tricolour

The same as black and white but with tan ‘points’ on the cheeks, ‘eyebrows’, legs and underneath the tail. Tricolour is a recessive gene so two copies of the gene must be present in a dogs genetic makeup in order for it to be a tricolour. Both parents must carry the gene in order for it to be produced in a litter. If you mate two tricolours you will only produce tricolours

pictured below...our own little Faith, Bryning True Believer


Blue Merle

Merle is a modifying gene that causes ‘patchy’ pigmentation over the body, this can also extend over the nose and eyes producing pink noses and blue or odd coloured eyes. Merle is a dominant gene so a blue merle will only have one copy of the gene. When a merle is mated to any other ‘colour’ merles can always be produced. Blue merle is when the merle gene is acting on a black/white base.

pictured below...Roger Brown's 'Josie' (Bryning Bolt from the Blue)


A ‘double merle’ is a merle that has two copies of the merle gene, this is to be avoided at all costs as there are numerous health issues associated with this genetic combination, mainly deafness, poor eyesight and general poor health and development. A double merle will often appear excessively white coloured and may have abnormally small eyes; these are produced when two merles are mated so merle to merle matings are to be avoided. Problems can occur when you get a ‘phantom’ merle, a dog that genetically is a merle but does not appear to be one; this dog could then inadvertently be mated with another merle.

Blue Tricolour Merle

Same as black tricolour but with the merle gene also present. A blue tricolour merle dog has two copies of the tricolour gene and one of merle.

pictured below...Julie Kondor's Blue (Bryning Blue Jay)


Chocolate (Red/brown) and White

There is some confusion internationally over the correct name for this colour, the shades can vary greatly and it may be referred to as red, chocolate, brown, liver etc. throughout this site we will be calling it chocolate! Chocolate is also a recessive gene so two copies of the gene must be present in a dogs genetic makeup in order for it to be choc. Both parents must carry the gene in order for it to be produced in a litter. If you mate two chocs you will only produce chocs.

pictured below...our very own Niamh


Chocolate Tricolour

Same markings as black tricolour but with the choc ‘base’ colour. A choc tricolour dog must have two copies of both the choc and tricolour genes to exhibit the colour.

pictured below...Echo (Bryning Leap in the Dark) owned and loved by Ann Eiserman, Doggone Borders, Belgium


Chocolate Merle

Same as blue merle but the merle gene is acting on a choc/white base. Two copies of the choc gene and one copy of the merle gene are present.

pictured below...Sadghyl Pip me to Comebye CDX UDX


Chocolate Tricolour Merle

Same as choc tricolour but with the merle gene also present. A choc tricolour merle dog has two copies of both the choc and tricolour genes and one copy of the merle gene.

Pictured below...our very own Mac!


Blue and White

Blue is the colour produced when black is diluted. The dilute gene is also recessive so two copies of the gene must be present in a dogs genetic makeup in order for it’s base colour to be diluted. Both parents must carry the dilution gene in order for it to be produced in a litter. If you mate two dilutes you will only produce dilutes.

Incidentally there is a relatively rare genetic condition found in both blues and lilacs (in other breeds not just BC's) known as dilution alopecia which can cause bald patches over the body and particularly on the ears. It is commonly thought that the breeding of dilute to dilute will increase the incidence of dilution alopecia but there has been no evidence to support this and a blue puppy from two black parents is just as likely to suffer from this disease as one from two blue parents (in fact I've never seen this from two blue parents).

pictured below...Bond, Locheil Time Moves On


Blue tricolour

Same as black tricolour but with two copies of the dilution gene also present. A blue tricolour dog has two copies of both the dilute and tricolour genes.

Pictured below...Jace (Bryning Ray of Sunshine); owned and loved by Julie Kondor, Trevellis Border Collies, UK

Image: Jace

Slate Merle

Same as blue merle but the merle gene is acting on a diluted black/white base. Two copies of the dilute gene and one copy of the merle gene are present.

pictured below...our very own Canen Spangle

Image: Canen Spangle

Slate Tricolour Merle

Same as blue tricolour but with the merle gene also present. A slate tricolour merle dog has two copies of both the dilute and tricolour genes and one of merle.

Pictured below is Ghost (Fast Forward da Casa d'Arinia), Trust Your Heart Border Collies, Germany

Image: 'Ghost'

Lilac and White

Lilac is the colour produced when chocolate is diluted. Both the dilute and choc genes are recessive so two copies of each of the genes must be present in a dogs genetic makeup in order for a choc base colour to be diluted. A lilac and white will have two copies of both the choc and dilute genes.

pictured below...Pip aka Packhams Ever Ready Pipin owned and loved by Lissa Searle, Hampshire Harriers Flyball Team

Image: Pip

Lilac Tricolour

Same as chocolate tricolour but with two copies of the dilution gene also present. A lilac tricolour dog has two copies of the choc, dilute and tricolour genes. This makes them rather rare as it requires 3 lots of recessive genes!

pictured below...Astra Mist; owned, worked and loved by Laura Cunningham, Corrie Dhu Sheepdogs


Lilac Merle

Same as choc merle but the merle gene is acting on a 'diluted' choc/white base. Two copies of both the dilute and choc genes and one copy of the merle gene are present.

Pictured below is Lizz (Qualified to Comebye of Pinewood Country) owned and loved by Danielle Boushouwers of Nice of you to Comebye kennels in Holland.

Image: Qualified to Comebye of Pinewood Country

Lilac Tricolour merle

Same as lilac tricolour but with the merle gene also present. A lilac tricolour merle dog has two copies of the dilute, choc and tricolour genes and one of merle. These are also rare as it requires 3 lots of recessive genes and the merle gene!
(so unusual I don’t have a picture! If you have a picture you would like to send me please feel free!!)


Red (ee red, Golden, Australian red) and White

This colour is very rare in the UK and there is therefore confusion over the correct name; we usually refer to this as golden/yellow or Australian red although again, the shades can vary greatly. Obviously in Australia/NZ they just refer to this colour as red. Red is also a recessive gene so two copies of the gene must be present in a dogs genetic makeup in order for it to be red. Both parents must carry the gene in order for it to be produced in a litter. If you mate two reds you will only produce reds.

Neither the tricolour or merle gene can be expressed on this coat colour so it is unwise to ever mate a golden red with a merle unless you knew the lines very well; if the merle carried the golden red gene you would almost certainly produce phantom merles.

pictured below...our very own Sunny!


Sable and White

Each hair is more than one colour/shade, usually black at the root and a light sandy brown at the tip. Sables are relatively rare in the breed and again, the tricolour and merle gene cannot be fully expressed on the clear sable coat colour so as with the ee reds it is not really wise to mate a sable with a merle as you may produce phantom merles.

The sable pattern can act on any base colour i.e. black, chocolate, blue or lilac. See the pics below for an example of sables with different base colours

1) 'Rico' aka Bryning Gold Blend - black shaded sable

2) 'Monkee' aka Bryning I'm A Believer - blue shaded sable

3) 'Falck' aka Steady Lines Falck - chocolate shaded sable

4) 'VJ' aka Fire n Ice van 't Hof Diabolo - lilac shaded sable

Image: 'Rico' - Bryning Gold Blend (black sable)

Image: 'Monkee' - Bryning I'm A Believer (blue sable)

Image: 'Falck' - Steady Lines Falck (chocolate sable)

Image: 'VJ' - Fire n Ice van 't Hof Diabolo (lilac sable)

Seal and White

You can think of Seal as being a sort of 'ghost' sable. The dog is carrying the sable gene on the agouti allele but does not have the double recessive k on the Dominant Black alelle so is unable to fully express the sable pattern. This can also sometimes be seen with the tan point and saddleback pattern genes, where dogs will exhibit 'ghost' markings.

This is Wren, Bryning Goldfinch (Rico's litter sister); here are some pics of her as she was growing up showing how the seal colour develops (Thanks to Joy and Becca for the slideshow)

This is by no means an exhaustive list of all the possible colours and coat patterns found in this breed, there are quite a number more including brindle, brindle point, saddleback tri etc but I have tried to cover those most commonly found and those of particular interest to me!

I must thank everyone that has allowed me to use pictures of their dogs to illustrate the different colours above.

Please be aware that these pictures must not be copied or distributed in any way.

Colour to colour matings….the myths

I thought I’d add this to the website because I get pretty tired of hearing people spouting these fallacies as fact.

1. You shouldn’t mate chocolate to chocolate or red to red because it ‘dilutes’ the colour

There is absolutely no genetic or biological basis to this statement and to be perfectly honest I have often found the reverse to be true in practice. Some of the darkest red and white dogs I’ve seen (my own Sunny included) are from red to red matings. That said, I’m not sure why people think that darker = better?? In my experience the paler coloured reds tend to have the better nose and eye rim pigmentation. The same goes for chocolate, there is no right or wrong shade of colour, only different ones and these are controlled by various modifying genes, it makes no difference whether these genes come from two parents that are the same colour or two different coloured parents.

2. You shouldn’t mate blue to blue because you get weak and sickly puppies

Again, a common myth that stems from the diagnosis of a condition referred to as ‘Gray Collie Syndrome’

This is a condition, similar to TNS that affects Rough and Smooth Collies; the grey colour seen in the coats of affected animals is actually a product of the condition and is not in any way related to the dilute gene we have in border collies.

3. You shouldn’t mate blue to blue because you get Dilution Alopecia (hairloss) in the puppies

This is something else I hear a lot that has its origins in some truth but has been misinterpreted. Colour Dilution Alopecia is a real condition that does affect dilute (blue and lilac) coloured dogs in a variety of breeds. However, the fact an animal is double recessive for the dilution gene and is therefore either blue or lilac is what makes it a possible candidate for the condition, coming from two dilute parents has no bearing on this. Bear in mind that dilute breeds true in some dog breeds, Weimeraners being a classic example, ALL weimeraners are dilute so all matings are dilute to dilute!

It’s worth noting that dilution alopecia is very rare in all breeds and especially so in border collies. I have bred and met hundreds of blue and lilac border collies and have only ever seen one affected with this condition, while visiting a dog show out on the continent. Any animal found to be affected should not be bred from.

Colour to colour matings…the facts

There is only really one ‘no, no’ regards colour to colour matings and that is merle to merle; this combination results in a proportion of the litter being homozygous for the merle gene (two copies) which usually results in excessive white colouration and leads to eye and ear abnormalities; deafness and poor eyesight.

Because a couple of other colours, ee red and clear sable have the ability to mask the merle gene; it’s not advisable to mate either of these colours to a merle either.

There is a sound argument for taking care when breeding for recessive traits (such as colour) on the basis that other, currently unknown recessive traits in the lines can be more likely to appear. This is on the assumption that the recessive gene is relatively rare (as most are) and will stem from a small number of animals so that those animals carrying or exhibiting the same recessive traits will share common ancestry and may therefore share other, less desirable recessive traits too. We have seen these sorts of genes come to the fore with TNS, CL and CEA in recent years and are extremely fortunate that we have DNA tests to deal with each of these.

A classic example of this in the UK bloodlines is that of Wiston Cap, he and his sons were so widely used that he is behind almost all UK bloodlines around today, he carried the chocolate gene so is likely the source of most of the chocolate we see. Another more recent dog is Bobby Dalziel’s Int Sup CH Wisp, a proven dilute (blue) carrier and largely responsible for the vast increase in the number of working bred blue dogs we see today. Sh CH Clan Abby Blue Aberdoone is also behind most of the blue dogs we see in the showring today, this NZ import was used extensively at stud and his genes have spread all over the UK and Europe as a result, he is the source of much of the dilute genetics we see in show lines. It is reasonable to expect that other recessive traits from these dogs will have trickled out down the generations too, some good and some not so good.

I’m obviously not saying that breeding colour to colour (recessive to recessive) shouldn’t be done, just that it should be done with a degree of care and consideration, as should apply to all breeding decisions in my opinion. I don’t think many people really stop and think about where certain traits have come from or where they might be going or consider the risks they may be taking. All breeding involves risk taking, part of being a breeder is accepting that and working with it, not choosing ‘rules’ and arguments to justify what you want to do.

Back to the top


©2010 Bryning Border Collies

Other pages:

This is the text-only version of this page. Click here to see this page with graphics.
Edit this page | Manage website
Make Your Own Website: 2-Minute-Website.com