Home Color & Pattern Theories
PDF Print E-mail

Theories of Color and Pattern Dominance in Shelties

Notes on Blue Merle, Dilute and White Factors

by Cheryl Anderson, Bio

Copyrighted '03-'09


If one studies the history of our breed it becomes readily apparent that we have some unusual color issues from our past which carry forward to today's stock. In the very first stud book, called Stud Book A, for example, it lists some as whites with colored heads, black & tans, solid sables, solid blacks, brindled sables, chocolates and livers. Nowhere does it list blue merle. The "blue" dog seen in those early days was actually an anopthic color, which is commonly referred to as a dilute or Maltese, and this form of "blue" is now called slate, and is the color much like the blue Great Dane or Doberman. One of the first champions, Lerwick Peat, was a "blue" of the slate variety. The dilution gene, a simple recessive which takes "two to tango" and can be carried hidden for many generations, can cause shades of charcoal, slate, gray, chocolate, liver and other variations in sable, tricolor, or blue dogs. The shade and darkness of the dilute color depends entirely upon the basic color and depth of pigment the dog carries as a whole. The dilution affects the whole dog, not just parts, so that in the sunlight even the deepest of charcoals becomes apparent to actually be a dilute. The dilution affects also the eye rims, nose leather, lips etc. which is why our standard authors put in that line, "nose must be black", to make it understood these factors are undesirable. We also saw a lethal silver gene in days gone by, same as is found in the Collies, but I have not heard reports of any for some years now. These whole dog dilutions are called "C Locus effects" from where it originates on the dog genome.


 This anopthic dilution factor, much like the also simple recessive rusting factors in our breed, is carried completely separate from the other coat patterns and colors, and works independently as such. The smutting factor appears to be an autosomal recessive in some families and a dominant in others, likely due to the particular origin within each line. Some families, however, seem to have a higher incidence (gene frequency) of these anopthic, or dilution, recessives so they show up more often than in other lines, hence people tend to think they are linked, and in some cases they may be as some patterns, such as cryptics, are crossovers in themselves and may have picked up the pattern linkage from this. In other words, the genes for both may exist on the same chromosome as a result of two chromosomes crossing over each other, and since chromosomes tend to replicate intact, this would explain its concurrence. But this is pure conjecture and thorough records would have to be kept of the occurrences in various lines to conclusively validate or repudiate this idea.

The merling gene as we know it seems to have been introduced into our breed in about 1917 with a documented cross to a blue merle Collie. There were several later crosses to merle Collies also. So this would be the most logical source as none of the other Sheltie ancestors appeared to carry merling. We do know, for instance, that Border Collies are ancestors and that they bring in the Irish (Si) white gene along with the cryptic pattern. The term "cryptic", meaning hidden, was coined by Dana Quinney in the late 1970's in order for us discussing dogs around the Banchory kitchen roundtable to distinguish between the different "self colors" since both anopthic and cryptic, up to that time, were known singularly and simply as "self-colors".

Since tan points are a separate dominant gene, the bi-color dog simply lacks that pattern gene for the points. We also know that the Blenheim Spaniel who was bred into our breed gave the white background with deep red pinto spots so that Shelties carry both of what would be called the Tobiano and Overo white patterns as found in horses and other species. As a result, Shelties appear to have 2 white patterns, one, the Tobiano, or pinto pattern, which gives the normal color-headed whites, while the Overo, or piebald pattern, can produce a combination of effects, some of which are defective, in much the same manner as the double merle since it also acts upon the very same cell layer in the zygote. They are all known as "E Locus effects" from the genome. These defects can be loss of hearing, loss of normal sight from abnormal pupil placement along with loss of pigment in the iris. This ranges, in eyes, from very light China or walleyes to what the old breeders called "starburst" eyes with uncentered or small oval pupils. In double merles, dogs can also have more severe eye defects to the point of eyes being absent entirely or only partially formed, also of internal organs being immature or incomplete which may cause sterility or early death. This is why the issue of overly white dogs needs to be taken seriously as we can all too easily slide down this slippery path toward more and more defects as we build up on the Si genes involved in a dog.

What I will now list is a hierarchy of blue colors ranging from the most dominant to the most recessive recognizable in our breed. After that I will list the merling patterns which layer on top of the color bases that often hand-in-hand with certain patterns. All these colors and patterns are determined by modifiers which exist in our breed. These modifiers determine the depth of color (intensity of pigment) and how much and where the merle gene disperses itself. It is these modifiers that determine a pattern, and the loss of which occurs in a specific order of dominance. Therefore, once the most dominant or numerous modifiers are gone, they are gone forever, and cannot be retrieved unless one consciously breeds out to stock carrying those modifiers. Tricolor and sable dogs can carry these modifiers, but they are just not phenotypically apparent because their background color is relatively solid. Depth of pigment is the only visible outward gauge along with what is produced within the family line. These modifiers become readily obvious when combined with a merle gene. This is why we see so many variations in our breed, and all these colors and patterns I will list are not cut & dried, but gradually slide from one into the next. What I will describe is the relative category in order from most to least dominant, sliding down a "totem pole" each of color base and pattern type.



The deepest pigmented blues, called "Pigeon" blues in the old days, are no longer seen in Shelties in the USA. This deep, deep color, comparable to a steel wool pad with a very distinct deep blue hue, same as a park pigeon whence it gets its name, can still be found in Australia and the UK. We are talking about the color base seen on the "blue" areas of the body. Most did not like this color as it was quite dark, and in the deepest, one had to look again to see that it really was a blue and not a black dog. However, this is NOT the same as the very recessive cryptic pattern. This is a blue basic color here.

The next step down from Pigeon blue is deep "Pewter" blue. Some of these still exist in the USA, but are becoming quite rare. It has a deep pigment base, on the blue areas of the coat, less than Pigeon, and with a bit more silver or brownish cast, just the way pewter itself looks. Most don't care for it, considering it too dark, but not realizing the value to be found in depth of pigment. I don't know if it holds true in Shelties, but in many other breeds such as Collies and Australian Shepherds, a lack of pigment can often go hand in hand with the autoimmune disorders. It is also one of the reasons many other breeds insist on deep pigment, make it highly desired in their standards. It's something to observe for yourself and consider anyway.

The next color down on the totem pole is "Powder" blue. A good, rich Powder blue was considered by the old time breeders to be the perfect blue color. It has a distinctly "blue" cast to it, unlike the more popular lighter silver dogs. In its clear form (no rust factors) it is a very clean and truly beautiful color of intensity without darkness, almost glacial blue.

Next down the pole is the "Silver" color, which is most popular currently, but lacks any blue cast to it. We are entering a more dilute pigment realm here, and while very striking in its contrast to the dark spots, denotes a lack of pigment base starting to happen.

Further down the line is the "Dilute" white. Under a microscope (and please send samples to Dana Quinney who is investigating this) the hair shafts appear to have some pigment, but very, very little in comparison to the higher colors. In whelps, it may be hard to determine where white markings leave off and blue pigment begins, it's so pale. These dogs, as all blues do, do darken with age, but it never has the fully pigmented appearance of the other more dominant colors.

Last, but not least, and operating under different genetic mechanisms than the above colors are the double merle and pattern white (bodied) dogs which lack brown or black body coat pigment. The double merle can be defective as the skin cell layer affected is deepest and more organs are derived from this layer than many of the more normal pattern whites (Tobiano type, anyway as Overos can exhibit the same defects as double merles). Some sources feel the difference is in 6 factors comprising the Si makeup -- if you have less than 6, you get Tobianos, if you have all 6 or more you get Overos. My feeling is this needs more research in our breed due to the different ancestral sources of white patterns. Albinos take this one step farther by being a dilute form which has no pigment at all. These are very rare.

Next, layered on these pigment bases are the patterns. This is what really makes our dogs truly distinct.


First, and most dominant, is the intense roaning pattern, or "Pigeon Roan" as it was most often found in conjunction with the Pigeon blue base color. Roan is a mixture of black and color hairs so that there are almost no spots, only blue hairs evenly mixed with lots of black hairs. You can still see this pattern, although more and more rarely, in some Australian Shepherds. The blue of the Australian Cattle dog is a lighter version of a roaning pattern.

The next most dominant pattern is the light, or "Steel" roaning pattern. It often was found on Pewter color base dogs, or a steel color, halfway between a Pewter and a deep silver color. Steely Dan was an excellent example of a Pewter steel roan while Deep Purple had the pattern on the steel colored base. Again, spots were few. The roaning predominated the pattern.

Next on the totem pole of dominance hierarchy is what we called the "Snowflake" merle. This is where it is a very light roaning pattern with very small and well spread out spots of black. When it grows out in an adult dog, it almost appears as though the dog were completely gray even though this is just an illusion. Tiger Rag is an example of this pattern. The blue base was usually light pewter, powder, or even silver.

The next, more recessive and more common, pattern is what we call the "Silver Dollar". Again, this was the pattern the old time breeders most desired and considered ideal. They even preferred a dog with a blue or merled eye along with this pattern and with the deep Powder blue color base to be their epitome of the classic blue merle. As the name describes, the merling pattern was light, with very little roaning apparent, and the black was well broken up all over the body in small "silver dollar" sized spots. The true "polka dot" dog.

Lower down on the pattern totem pole recessively is the very popular pattern known as "Domino". It is extremely flashy with big splotches of blue, white and black all mixed up. One of the hallmarks of this pattern is that somewhere on the body will always be an anopthic gray patch, big or small. It doesn't mean the individual will produce dilute dogs, it is just part of the makeup of the Domino merling dispersal. This pattern is most often overlaid upon a Powder blue, Silver blue, or Dilute color base. It has been the most popular, flashy, splashy pattern in the last decade or more. The rage started with the Banchory "Strike Me Silver" get and hasn't slowed yet. However, this pattern, being the recessive bottom of the totem pole, carries several genetic caveats along with the dilute background counterparts. When laid on a Dilute background, the color and pattern are known as Harlequins, just the same as a Harlequin Great Dane. Pups are often born white and only turn blue later. It is in the same family, at this point, as the Dalmatian pattern, with a white background and black spots.

At the lowest end of the pattern pole are the "Cryptic" patterns. This is where the dog is predominately black -- note: the old time breeders considered predominant quite literally, that any dog over 50% black was more a black dog than a blue dog, and this was a form of self-color. The ideal desired blue pattern was well broken up black spots on a blue background, not solid areas of black predominating as it does with the Cryptic pattern. This was not considered a true blue merle pattern to them so they would pet them out when pups of this pattern or dilute specimens appeared in litters. That is why they were virtually unseen before the last 20 years.

When they appeared, they were just culled. As I understand it, this is still the practice with most Canadian AOAC breeders, and I must commend them for this practice as it helps retain the genetic diversity of modifiers so needed in the breed. Cryptic is a color fault, just like a pale sable or rusty tri or dilute blue. It is minor in the whole picture of the standard, but still needs to be taken into account as a fault, and is not the ideal desired. You can see in the standard the two parts of self color described by the older breeders, along with the dilute description, before we had the cryptic and anopthic distinctions:

"Faults…..Washed out or degenerate colors such as pale sable and faded blue [Dilutes]. Self-color in the case of blue merle, that is, without any merling or mottling [Cryptic] and [ed. note, perhaps should have been more correctly "or"] generally appearing as a faded or dilute tricolor" [Anopthe]. The two self-color type distinctions meant within the standard were what was taught to me by Betty Whelen and Libby Babin as I did not fully understand that passage before their clarification to me, and it makes sense in light of what we actually see and now know to be genetically accurate. This, of course, is only my own evaluation of the passage. It, and the entire standard, is always open to individual interpretation and much discussion. This, to my mind, is the crucible of refinement, and in the end, what eventually changes any standard.


The danger in these lower colors and patterns, i.e. Silver and Dilute, and Domino and Cryptic, is that they are the genetic recessive dregs of our blue color spectrum. If dogs with or producing these color/patterns are not bred out to other lines or dogs carrying the more dominant modifiers, if they are bred only within their color & pattern family, even a good Domino Powder blue can only produce itself, Harlequins and Cryptics. This is because they are the most recessive patterns lacking other modifiers. And technically, by the standard description itself, these are not correct blue merle patterns. So with the increasing frequency, the fad popularity, that these colors & patterns are being bred, and the less the older, correct colors/patterns are used both in breeding and in the ring, we are changing the gene frequencies rapidly and could eventually see the permanent loss of correct patterns & colors from the breed. It is imperative that breeders select the correct colors and patterns to prevent this loss, and for judges to not be dazzled by the flash by understanding what they may be doing, genetically, to this breed.

When these Domino patterns exhibit excessive white, meaning over 50% white on the whole dog the danger of defective piebald, overly heavy Si (buildup of numerous Irish Pattern genes) dogs emerge. Dogs with white on ears or up to or surrounding eyes should be suspect for hearing loss and recommend checking for such by Baer testing. Dogs with uncentered pupils in China eyes should be suspect for sight imperfections, and both deafness and blindness are disqualifications under AKC Rules.

The issue for most has been, what exactly compromises over 50% white? Well, any dog that has four white legs, white creeping up the sides of its body, perhaps with some white body spots, a big white shawl collared neck, big white tail tip, white blaze up its face, all white under its chin and under its belly is more than 50% white. The problem has been the standard currently being misinterpreted as applying only to any white that appears on the saddle area of the body. The standard says, "Specimens with more than 50 percent white shall be so severely penalized as to effectively eliminate them from competition". Note it says "with more than 50% white". It does NOT say "with more than 50% white on the body". So if you took the dog in question and skinned it out, would the dog, with the full pelt nailed to a wall, appear to be more than 50%? In many cases today, yes. Unfortunately, the way the standard is currently written, it carries only the weight of a very severe fault, but is not a disqualification. The same problem, for judges, applies to the blue eye in the sable merle, and while not at all desirable, is more a true cosmetic issue than one of genetic defects being brought along with the excess white.

One more note in this discussion, this on the Cryptic pattern which is appearing more and more frequently due to the major breeding popularity of the Domino pattern. Many true Cryptics are appearing which look for all the world like tricolors but are genetically blue merles. The bad part for the breed is that the AKC registers dogs only by phenotype (how they appear) rather than by genotype (what they actually are genetically). As a result, many of these true Cryptics are being shown in Open Black and used by their unsuspecting owners as tricolors in their breeding programs, and have had some bad surprises when they produce blue pups. Also, they are more likely to produce more Cryptics like themselves, unwittingly to many breeders as they appear, again, as if they are normal tricolors when in fact they are not. So……..what is the upshot? Well, this is a source of blue-eyed tricolors, for one. The other is, what happens when you breed an unsuspected Cryptic tricolor to a regular blue merle? You can get blue merles, regular tricolors, Cryptic tricolors (really blues), double merles and Cryptic double merles. Now, ask yourself this one……what does a Cryptic double merle look like? Why, a "normal" blue, of course. But usually these are dogs who may often have excess white on the ears, and up to or through the eye area, and maybe the staring or starburst eyes. These dogs should be Baer tested also. I've seen many of this type of dog in the rings today out of blue named parents from lines known to carry the Cryptic pattern. This will be an increasing problem, and this is the main reason the old time breeders stayed away from breeding the Domino pattern.


This article is copyrighted and can not be copied or distributed in any way without the written permission of its author, Cheryl Anderson.

Please feel free to contact me at This e-mail address is being protected from spambots. You need JavaScript enabled to view it   if anyone wants further information or advice on these patterns and breeding situations. My interest is the health and future of this breed I love.

More involved discussions of these factors will be forthcoming in my in-process book, Stamping The Look, A Breeder's Guide To The Sheltie. These and similar topics are similarly discussed in my Stamping The Look Seminars, a non profit program for the benefit of Sheltie and all breed clubs.Visit my website at: http://www.cherdensheltie.com