THE HARLEQUIN FAMILY OF DOGS: Harls, Merles, Blacks, Whites & More

There always seems to be alot of confusion & "mythos" about harlequins, harlequin breedings & the merles that are born to harls. This little article is intended to answer the basic questions that most newcomers to the breed have, & to hopefully help clear up the confusion that comes from conflicting reports about harls, merles & their family. There is a little more technical information at the end of the article for those interested, as well as a list of references for those wanting to study the topic. If you don't want to wade thru all this, then remember, you cannot go wrong if you stick to the standard & breed ONLY show-marked & breed-quality Harlequins & Mantles to each other. That's why the standard was written in the first place; to act as a guide to breeders. So you don't need to understand all these details & theories of the genetics of coat color, but you do then need to stick to the standard when breeding. If there is anything you feel that needs to be added, etc. to this article, please email me & I'll do my best to see your question gets addressed.

First off, the basics of what you get when you breed. A note prior to presenting the data & links. Statistically the data are conclusive. This does not mean you will see these exact ratios of pups in litters, as such ratios only appear when large numbers are studied. This has been done. The data are extensive & it's well known what to expect from certain breedings. There is no real mystery to breeding harlequins; it's not "a crap shoot," & it doesn't require the use of mismarked dogs or weird combinations to be successful. There is NO secret recipe involving merles or other mismarked dogs. All this has been done time & again, so PLEASE don't think there is ANY need to try out odd combinations to "see what you get" or prove some elusive point. This is simply not an ethical approach to breeding: you do not care for the breed by "reinventing the wheel" and having litters to learn what has already been discovered, seen & written about. At the end of this article is presented an argument about theories of the harlequin variant. That does NOT mean that the basics are not understood, nor do these arguments have much to do with the practicalities of breeding harlequins. Read before you breed, please, and follow the standard when you do breed. A long list of references is offered at the end of this article & live links are offered thru-out to help you find the relevant articles & information you need. If you want to see the ratios of expected pups from various harl-family breedings, read Jane Chopson's article which gives all the offspring expected from various pairings or read one of the articles I've written, either the basic article or the more technical one. Louise Feddema also has one that outlines what to expect from certain breedings. I am NOT presenting myself here as "the" expert: I am presenting expert information gathered up in one place for the public to see that (hopefully) properly summarizes the available literature. Like I said, email me additions, corrections, etc. The best advice always is: follow the standard for the breed, in color, as in all other things. You cannot go far wrong as a breeder if you use the standard as your guide.


First off, regardless of what combos you try, large numbers of mismarked puppies are expected in ALL breedings involving harlequin (& merle) dogs. The best you can ever hope for is about 75% correct color & pattern; 50% is the average if you count both Harlequin & Mantle offspring. You can reduce the expected number of correctly marked pups by at least 1/4 if you want to talk about harlequins specifically. Merles are routinely born to harls, as are various other mismarks, including white merles (MM) also known as (deaf) white danes. Therefore the problems of euthanasia of defective pups and the careful placement of dogs who fail to meet the standard is a routine part of the responsibility of any ethical harlequin breeder. You can avoid all of this by breeding Danes of any other family but Harlequin. Harls have problems & challenges the other colors of danes do not face. Mismarks in every litter is just one of them.

HARLEQUIN to HARLEQUIN: You can expect, in a litter of 7, to see 2 (likely mismarked) black or Mantle pups, 2 merle pups, 2 harlequin pups & one deaf white pup (two are statitically conceived, but one of two whites conceived generally dies prior to birth). Naturally people have 7 males in some litters, so some people will have 7 harls..or even 7 deaf whites, but on average, about 1/4 of the four basic colors is what is expected. Of course harls that carry blue or fawn or brindle can produce these "porcelain" dogs, sometimes called bluekin, fawnikin & brindlikin respectively. To read about why fawnikins occur, click here. Flashy harls (with lots of white) could also produce piebalds: white dogs with "mantle heads & harlequin looking bodies" as they have been described. Harl-, merle- or boston-heads; dogs with white bodies & "caps" of color (i.e. color-headed dogs) also can result from the use of piebalds & flashy harls & mantles. These color-headed dogs, also called piebalds, are essentially mismarked Mantles. To read more on piebalds, click here. Any time two dogs with the merle gene (incl. harls) are bred to each other, white merle (i.e. defective, dominant white) offspring are to be expected. The dominant whites that do survive to birth are normally defective & commonly deaf. Ethical breeders generally euthanize all whites at birth for these reasons. To read more about white (MM) danes, click here.

HARLEQUIN to MANTLE: This is the preferred breeding, all things considered, as it does not produce the Dominant (deaf) White dane. Expect the litter to be about 1/2 mantle, 1/4 merle & 1/4 harl. So in a litter of 8, expect 2 harls, 2 merles & 4 mantles. If these black pigmented dogs carry for other pigments, you may get fawns with blazes, or blue instead of a black basecoat. (If that is confusing, read about why fawnikins are here to stay.) Piebald & color-headed dogs can occur, especially when flashy dogs are used.

MANTE TO MANTLE: Produces only Mantle. That is, it cannot produce harl, merle or white danes. Now piebald danes that are mistaken for whites and for harlequins can be born to Mantle parents. This is more likely to happen when breeding flashy Mantles with big collars & broken blankets than with dogs who have more pigment. Mismarked blacks can also be born to this breeding. It is possible (if unlikely!) that very flashy Mantles bred to each other could produce a puppy so white s/he could be deaf. Possible--not probable--& AFAIK it's never happened in Danes. BUT it does happen in Boxers & Dalmatians.

MANTLE TO (harl-bred)BLACK. Would likely produce mismarked blacks. Mantles & (solid) Blacks also are possible.

BLACK to BLACK: Would likely produce mismarked blacks when harl-bred blacks are used. Mantles & Blacks possible.

BLACK TO HARLEQUIN: Produces as does Mantle to Harlequin, except, since a Black lacks the white collar of the Mantle, expect more mismarks under the AKC standard to occur, both in Harlequin & especially in potential Mantle offspring. Blacks will likely not be solid Black, but mismarked black, sometimes even called "harlequin black" it is so common to black to harl breedings. These "harlequin blacks" are disqualifying under the American standard.

These above breedings are authorized by the Great Dane Club of America (GDCA) & by various other breed clubs. Note that the Great Dane Club of Germany has banned Harl to Harl breedings in an effort to avoid the production of white danes. To read more about the ban, click here. Some European countries have gone so far as to ban ANY dog who carries the merle gene from being bred to another merle-factored dog & some have banned merles entirely, all because merles produce deaf & defective white puppies & pups who die in embryo. The merle gene is "semi" or "sub" lethal--and kills about 50% of those pups who have two doses, while is leaves most of the living pups deaf or with other defects. Read more about whites here. The breedings below are NOT authorized by the GDCA. All involve dogs DISQUALIFIED worldwide under the standard for the breed. Disqualification is done specifically to insure that a dog with such a defect will not be bred.

BLACK TO WHITE: (or Mantle to White). This breeding is done by some people because it sometimes offers the possibility of producing at least 50% harls or even a whole litter of harls. Of course a whole litter of merles is just as likely as getting all harls & there are lots of problems with using whites in a breeding program. Whites may all look alike, but they don't all breed alike! And of course most of them are deaf. To read more about whites, click here. White to black breedings that produce black, mantle or piebald pups are not true black to white breedings; the "white" used is actually a mismarked harlequin or a misidentified merlikin. Using deaf dogs in a breeding program is reported to increase the number of deaf pups thru several generations.

WHITE TO HARL: This would produce a litter with 50% deaf whites, the rest split between harl & merle, depending on what kind of white was bred. Again, if this breeding produces black, mantle or piebald pups the "white" used is actually a mismarked harlequin or a misidentified merlikin.

HARL TO PIEBALD: This would increase the number of mismarked pups, as you'd expect to get harl-heads instead of harlequins & piebalds in the place of mantles, especially if the harl used is flashy or even just has that clean white neck. To read more about the piebald, click here. Piebalds are mismarked Mantles, not a kind of Harlequin.

MERLE TO WHITE: 50% deaf (or dead) white is expected; 50% merle, with reduced litter size due to merle-related deaths.

MERLE (or merlikin) TO BLACK (or to Mantle): This is a breeding using TWO mismarked dogs. (Most of the others use at least one parent with correct markings!) Since some harl-bred blacks & mantles can carry harl genes (H-factor) unseen & merles obviously carry the Merle gene, recombining the two can occassionally produce harls. But mostly, again, this breeding produces mismarks: black with white markings & merle dogs. And the dogs used in these sort of breedings are commonly sub-standard in other ways than color.

MERLE TO HARL: This is like breeding harl to harl, except you reduce your chances of getting harls & increase your chances or getting mismarks and various defects by all reports.

MERLE TO MERLE: This is not deliberately done in Danes, although it is occassionally done in other breeds. It is universally discouraged & is against the law in some countries. Vague reports of such breedings in Danes make the rounds from time to time. From what is known, at least 25% or the litter is expected to be defective & there are anecdotal reports (in Danes & other breeds) of merle x merle litters ending in "disaster" (all lethals, all defects, absorbed litters, etc.) More on breeding merles below:


1) Merles are a disqualification under all standards for the Great Dane--and have always been. That is enough, in many registries, for a dog to be denied "papers," so this is obviously always been thought a very poor breeding choice. (That the AKC will register a dog does not imply approval of it as a breeding choice--the AKC will register *ANY* dog with two registered parents of the same breed. Any dog.) One must assume, historically, such animals were used & the results generally not good, hence the fact they have always been disqualified. There *are* anecdotal reports that using merles in a harl breeding program not only increases the number of mismarks in a litter, but also results in a much higher number of defective puppies than in a harl to harl breeding. In other breeds which allow the merle (very few BTW), merles are not supposed to be bred to each other, to avoid the production of white merles: those defective, predominately white pups. Merles LACK the necessary genetics on their own to produce harls: noone has ever documented a merle to merle breeding that produced harls. So by all reliable reports, merles as a group do not productively contribute to the harlequin-mantle gene pool.

2) It is certainly and always the case that anyone who has not established an *sterling* reputation & can explain to the general dane fancy's satisfaction the reason to break such a general rule of breeding better be prepared to be considered unethical by most if not all others concerned about ethical breeding. This is a case where making an exception to the rule is better NOT done but for the rarest of cases where the person in question has a long track record of good practices, the production of Champions under ethical constraints, AND who has an exceptional & rare enough situation on their hands to warrant a probably once in a lifetime event. People who sell & breed merles (including so-called merlikins) as breeding stock on a regular basis can simply be generally assumed to be commercial (i.e. for-profit) breeders whose goal is cash cropping harl pups, not breed protection. These dogs are also bred in ignorance by people only casually involved with the breed. Needless to say, however well meaning, these breedings do not contribute to breed betterment either.

3) A boston merle, the most likely choice as a breeding partner to a harl, is essentially a "true" boston (i.e., Mantle), with the added problem of producing deaf white & other defective puppies, which does not occur in a Mantle to Harlequin breeding. So a Mantle is always preferable, on this account, to a merle.

4) "Merlikins," a white base-coated dog with black and grey markings, have been bred on occassion to Mantles to produce harls. But this is a risky breeding for several reasons. Merlikins may be deaf and have eye defects (and without a CERF no one can say a dog's eyes "are fine"!), most (Mm) merlikins will NOT produce properly marked harls under any circumstances, & the harls they DO produce are likely the result of their Mantle (boston) mate carrying unseen the factors needed to produce harls. There is significant evidence that some (many?) merlikins are actually (MM) double merles, i.e. genetic whites. There is also a well-established theory that merlikins are simply merles which carry the Tweed gene and the Tweed gene produces that swirly-looking "merli-harl" that is a harlequin mismark. And they are disqualified under the standard. All this means merlikins are established as a poor breeding choice as a rule. So, again, a properly marked animal, in this case a show-marked Harlequin, would be preferable to the merlikin.

5) The long and the short of breeding merles (and other mismarks, for that matter) is, if you have to ask about the whys and wherefores, you probably "can't afford it;" can't afford the disaster and dishonor likely awaiting you. Unless you are dealing with a publicly esteemed breed expert really caught in a corner, you can pretty much assume someone using mismarks (e.g. merles, piebalds, and other mismarks) routinely in their breeding program is less than knowledgeable and less than ethical. This *is* a case of guilty until proven innocent.


I published an article in the Great Dane Reporter in 1997 on the action of the S Locus itself (when alone) as it applies to the Mantle (Manteltiger) entitled: "Mantle Dane Genetics: How to Get & Keep Boston Patterned Dogs".  There are also somewhat harder to find, but very useful, articles published over the past two decades by such as Paul Hardiman and Laura Kialenaus on the subject of the Mantledane. I tried to summarize their message in the article I wrote in 1997. This article discussed some of the implication of attempting to stabilize a "tux-n-tails" bicolour Great Dane as a breeding partner to the Harlequin. This also is a complex issue, but the basic take home message is:

1) Breed only properly marked Harlequins & Mantles to avoid not only mismarks but an increase (thru loss of pigment) of pups with sensory defects. The breeding of piebald or flashy boston-type dogs (those with extensive white, i.e., more than 1/3 white) will result in a loss of harl offspring (to such as "harl-heads"), and could cause an increase in defective pups due to loss of pigment from these recessive white genes combining with the dominant white gene(s) that make up the merle/harl complex. Therefore, avoid the breeding of mismarks, especially predominantly white mismarks, such as harl/merle and boston heads, whites, mostly white, white headed, piebald and porcelain (merlikin/fawnikin etc.) dogs whenever possible, as the effects are multi-generational and generally deleterious to both health and color. They simply carry too much recessive white genetically to be safely combined with harldane (merle) dogs. White to black (including Mantle) breedings are not unusual these days in North America. But recall "whites" are often an unknown genetic quality. Also Mantles are now accepted. More reasons to select other choices. Plus, using deaf dogs in a breeding program is reported to increase the number of deaf pups thru several generations.

2) Attempting to stabilize a full and complete collar and the approximately 1/3 white called for in the ideal of the Mantledane requires one to 'straddle' a gap between two alleles (si=irish and sp=piebald)--SO--many dogs with this same phenotype are going to have VERY different genotypes & most of those genotypes are NOT going to be homozygous (e.g., Ssp/se = pseudo irish occurs). The long and the short of that is, that many Mantledanes who look just alike are going to produce very differently, & the more mismarks used in the breeding program (especially predominately (>50%) white dogs), the more mismarks that can be expected, especially "skip generation" recessives. Consider the problem with whites and checks in Boxers, as this is the exact same phenomena-less our problems with the merle gene. What is called "white factoring" occurs in these breeds--i.e. the presence of a piebald recessive white allele--which results in predominately white dogs from predominantly pigmented parent. This phenomenon is present in a variety of breeds from Collies to the Akita. Broken collared dogs and other dogs who are "safely" (at least 2/3) pigmented are a better breeding bet for overall health than rather flashy marked animals; particularly "extreme" bostons and out-right piebalds (Plattenhunde) and most particularly when bred to flashy marked harls.

Checks & whites? If not familiar with this phenomenon, see how "Plain," "Flashy," & "White" phenotypes in the Boxer (roughly) corresponds to Solid Black, Mantle & the Piebald in Danes. Click here for an illustrated tour of this two gene, three phenotype phenomenon typical of incomplete dominants. See also "Spots Before Your Eyes" for an illustrated guide to the complete & complex phenomenon of the recessive spotting (S) locus with its four alleles, incomplete dominance, 10 genotypes, modifiers, & multiple, overlapping phenotypes, which is likely a much closer analogy to the situation in Danes.

3) It would seem inescapable that some & not all Mantledanes carry some sort of allleles, be they modifiers or a more "direct" & seperate H-factor (the "harl" locus), that contribute to the production of harls (over merles) in harlequin litters. This is implicit in much of what has been written, but is rarely explicitly stated. It certainly should be addressed at part of the family inheritance, as should the implication that follows directly from this observation; namely that Mantledanes, as individuals, contribute "unevenly" to the production of harls, with some unable to offer any "help" to increase in the percentage of harls in a litter, while others might, at least under certain theories, carry strongly enough for whatever constitutes "harlequin genes" to not only increase the percentage of harls in a litter (over merles) from a harl x mantle breeding, but actually produce harlequins from a merle x mantle breeding.


There are a variety of theories publically offered to account for the variation of harl to merle phenotype. These I have summerized and discussed in detail in the 1996 publication entitled: "A Summary of Theories concerning the Harlequin Variant in the Great Dane." (published in the GDR (Great Dane Reporter) May/June 1996 issue). Generally all these theories fall into the following categories:

1) Harl is a variation on merle, with the M allele of merle being somehow modified to produce this different phenotype. Various ad hoc caveats are added in to explain observed phenomena that do not fit the specific theory. None of them (e.g. Burns & Fraser) are at all satisfactory & most cannot account for the (regularly observed) production of merles from harldane breedings. If you read in such as Willis, you can see some reports on such theories. As they don't match with observation & data, they have generally been discarded & are included here simply for thoroughness.

2) Harl is a variation on merle, with a seperate Locus being necessarily present to produce the harl variant (the W/H theories of harl). This is the Bagalla (W) or Sponnenberg (H) theory of Harlequins, that Neil O'Sullivan published on (with comprehensive statistical data) & Jane Chopson so thoroughly diagrammed.(To read this explanation, click here.) The basic notion here is merles lack a gene that harls have. This "H gene" makes harlequins by "clearing" the grey background that the merle gene leaves behind. So merles are grey dogs with black torn patches, whereas harlequins are white dogs with black torn patches. Harls are HhMm; Merles are hhMm. Harls are therefore "H-factored" merles if you will. Merles, therefore, cannot produce harls as they lack this H-factor. There are no living homozygous (HH) harls, as homozygosity is uniformly lethal. There are homozygous or "white" merles (MM) however, although about 50% of these dogs are also lost prior to birth. (To read more on the white or white merle dog, click here.) So, the merle allele is sub-lethal as is the harl allele, but both seem to be generally deleterious only to the homozygote, and the link for both deaf & defect is not direct, but rather thru the loss of pigment & pigment cells. (See below for more information on Dominant White Lethals.) MM loss produces a 12.5% (1/2 of 4/16) reduction in expected litter size. HH loss produces an additional 18.75% (3/16) loss in litter size. Together under this theory a 31.25% loss is expected. Merle x merle (hhMm dogs) cannot produce the Harlequin variant (a HhMm dog). Merlikins under this theory are simply heterozygous (Mm) merles who are homozygotic for the recessive
Tweed (Tw) gene. This Tweed gene produces that "swirly" or "mixed" (messy black-n-grey patching, as opposed to harls with the required solid black) patches on a harlequin. So adding this recessive gene means increasing the number of mismarks, & for this reason merlikins should not be used in harlequin breeding programs. Merlikins are (recessive) affected (tw-tw) dogs (hhMmtwtw) who produce both carriers (Tw tw) from clean stock (Tw Tw) & affecteds (tw tw) from carrier (Tw tw) stock. This "H-factor Theory" not only accounts for all observed variation, but is fairly easy to diagram & use. For shorthand, when discussing harlequin genetics, Harl or H-factor can be used to describe the variation from harl to merle, regardless of the "true" explanation or any lack of certainty that still exists about these details of harlequin inheritance. And this theory can be used to demonstrate the expected results from various parings in the harlequin family of dogs. It is a very practical and very popular explanation for the Harlequin variant.

3) Little's original (1955) theory is that harl is simply a phenotype variant of merle, with the action of alleles at the S locus (in combination with the M allele) producing the variation from merle to harl. This has been thought to be an incomplete, if not wholly inadequate theory for a variety of reasons which revolve around explaining all the multitudinous variations of harl to merle & the observation that MmSs or even Mmss does not seem to produce harl in any other breed but the Great Dane. If expanded, this could be called the Modifier Theory of the harlequin variant. A suggestion I made in Yousha, 1996 was that M modifiers (functioning quite like the recognised modifiers at the S Locus) might account for the declaration made that merle to merle breedings can produce harls (which not all other theories cannot account for). And a "Modifier Theory" of the Harlequin variant does not require the 31.25% expected reduction in litter size that the H-factor Theory would need to function, so might better account for the large litter size not uncommonly reported in harl-family dogs. It also fully accounts for black to white breedings which do not produce harls, as well as those "in-between" "harlie-merle" animals; the wide & seeminly continuous variation in phenotype seen in both harls & merle, and the commonly reported claims of merle to black dane breedings producing harls. In short the idea of such a Modifier Theory is that merle is a phenotypic continuum with two axes: S for pattern distinct from M for coloration(hue). Modifiers genes at the M Locus "add" & "subtract" pigment, particularly from (merle) areas of reduced ("dilute") pigment that appear grey-brown. Many positive modifiers allow for a merle phenotype and as these modifiers "move" to be predominately negative, harlequin phenotype begins to appear. Modifier genes from the S locus account for the incremental variations from solid (mouse) merle through boston & piebald merle, merlikin, color-headed dogs, as well as "clean white necks" & other overall pattern phenomena seen in harls. It is a complex & somewhat unwieldy idea; and without clear categories (since it constructs a continuum).

We know recesive white (S Locus) spotting genes interact with Dominant White (M/H Loci) genes when it comes to the percentage of white overall (with more defects resulting in dogs who carry recessives at the S Locus). This is something observed in all breeds that carry the merle gene, not something particular to danes. Little (1955) postulated a synergism occurs via positive & negative modifiers of the S Locus acting on the M allele, which act to increase or decrease the amount of pigment the main S gene. The S locus is documented as a late acting pattern locus involved in a very specific action at a late (embrionic) stage & acts naturally on the extremities inward, (in a well documented and somewhat predicable pattern), and is affected by modifiers which allow for a variation in range around the axis allowed by the main gene (e.g. S=<10%white, si=<30%, sp=10-90% white & se=>90% white). The more recessive the allele & the more negative the modifier series, the early the action of the gene begins to interfer with A/E Loci (which direct for a fully pigmented body). The M Locus begins to act very early in fetal development (i.e. neural crest stage), with full expression occuring later in development & independently in (molecular leve) action from the S allele(s). The dominant (M) mutation affects the melanocytes prior to/during pigment migration, disrupting the resulting banks of cells' ability to produce pigment, thus causing the mottled coloration. The dominant (H) mutation (or such modifiers that otherwise exist) is not expresssed except in the presence of the Merle allele: when H & M alleles are combined, a Harlequin results.

The HH homozygote is 100% lethal in embryo. The MM homozygote is ~50% lethal in embryo. The MM (the double merle, white merle, or defective white) danes that survive to birth generally suffer from sensory defects (e.g. deaf &/or blind) as this gene's effect on melanocytes affects much more than simply external coloration (i.e it is pleiotrophic), although these effects are neither uniform or universal. Their combined (M/H & S Loci, plus modifiers) effect is to visually increase the amount of white on the dog. But these genes disrupt the proper formation of pigment. The S (spotting) Locus, however, simply "clears" areas where no pigment will normally be found, leaving smooth "edges," white skin and a distinct bicolour dog. S=spotting alleles are inherited recessively & the S Locus demonstrates incomplete dominance. (To read more on the S Locus as a distinct phenomenon, click here.) The M (merle) Locus apparently affects (fetal) pigment cells individually, completely disabling some cells (white areas), partially allowing for pigment expression (merle areas) & having no affect on some cells at all (black areas). Mottling or merling, with distinct "ragged" edges & a random blotching pattern of varying hues results. The M allele can only act, however, in the areas unaffected by the restriction of pigment due to carrying recessives at the S Locus. The H allele (or modifiers) cannot act except in the presence of the Merle allele, as detailed above. So merles are without the H-factor (if they had it they would be harls). But blacks (including Mantles) with the H-factor are simply black in appearance.

For more info on neural crest development see: or Their respective biochemical action of these recessive & dominant white (spotting) genes seem distinct & discrete, but the combined result is a dramatic loss of pigment. This fact is worth taking account of as lack of pigment is not simply an aesthetic factor. Loss of pigment & pigment cells affects the structure and functions of many organs, particularly eyes, ears & skin. So "pretty" predominately white dogs are distinctly at a functional (and reproductive?) disadvantage & (even inadvertently) breeding to increase recessive white in merle/harl dogs is a practice that can be predicted to increase the number of defective heterozygous (Mm) (lightly marked harl, white bodied & merlikin) dogs.


Naturally the same data are all explained by all these theories outlined above. There is no basic dissagreement about that. There is no dissagreement about what is expected or seen in harlequin litters. Also there is no disagreement about the data produced or the information given above the "Technical Section" above. It's simply a question of more than one theory that fits the data at hand, which is not uncommon in science, & really has little practical significance to breeders anyway. The details of harlequin genetics that these theories differ over are fairly insignificant & mostly are just an intellectual puzzle. It cannot be said that there is any real mystery or controversy about harlequins & the merles they produce ( at least for educated breeders), and quibbles over theory details should not be used to exploit the breed by mating all sorts of dogs, or claiming breeding harls is uncontrollable or wildy unpredictable. This is simply not so. Color genetics is the "crayola crayon" level of genetics: easy to manage and easy to comprehend compared with more complex (& more important) genetic issues that affect dogs & breeding decisions. And it cannot be said too often that the general rule of breeding harlequin family dogs is to use show-marked & breed-quality animals only.

In summary, the Merle mutation is a documented incomplete dominant with 3 distinct phenotypes (e.g MM Mm & mm) possible. The phenotype harlequin, as a distinct "fourth" type, is somewhat controversial, but is obviously present at least as a phenotype, even if it is only (apparently) found in the Great Dane. Since Little's original notion of modifiers was not developed enough to satisfy Harlequin breeders & his data were apparently limited, other theories have been postulated to better agree with current data & observations. The most widely accepted theory for the Harlequin variant was postulated by Sponnenberg & by Bagalla, is supported by the work of O'Sullivan, & explained in precise detail for laymen by Jane Chopson. If it were ever to be documented that a true merle to merle breeding produced a true harlequin, however, this theory would be discredited, as it does not cover than contingency. This theory also requires fairly large (~1/3) litter losses due to lethal gene combinations. The notion I suggested of M as well as S modifiers allows for harl offspring (if rarely) from merle dogs, and does not require such massive litter losses. It offers more "moving parts" to explain the wide variation in pattern & pigment seen in both merles & harlequins. (It also makes the whole process of figuring out "who has what" rather unwieldy!) But of course it does not offer a molecular or actual physical explanantion; it is just a theoretical explanation that remains "undefeated" in the face of the recurring (but never documented!) reports of merles producing harls, and does not require such a large reduction in litter size from fetal death. But whatever the "physical truth," the easiest way to practically predict offspring is to use the H-factor theory, even if it is used only as a sort of shorthand for details still not in evidence. I suspect we will all have to wait for the molecular data for a detailed answer as tothe question of what exactly makes harlequins genetically different from merles.

Please note that, regardless of the theory to which you subscribe, it is inescapable that the current trend in American harldanes to produce ever "whiter" harldanes has negative implications for the overall health of the resulting puppies as well as producing a whole series of recessive mismarks uncommon (in not unknown) in the parent stock. The original selection was for a strongly pigmented dogs & ONLY the AKC standard calls for a "white neck preferred." FCI and all other Great Dane standards (including the parent country) simply ask for a black and white dog with "torn patches." They generally allow solid blacks & even our "mismarked" blacks to be used as part of the harlequin breeding program, but frown upon black dogs with extensive white markings; presumably to avoid the production of recessive white dogs in a color which already carries for dominant white. So even when breeding for this "preferred" (AKC) white neck in harl & mantle danes, one should not lose sight of the following facts:

1) It is never possible to control for a predicably fully white neck, even when two such animals are bred consistently together; the genes (as described above) do not admit of such control. So it is not a trait worth giving an enormous amount of attention to, and the standard for the Mantle doesn't require a fully white neck anyway.

2) Even the AKC standards say NOTHING about white "fronts" & chests and forelegs being even desirable, let alone preferred on harls or mantles. Note also the illustrated standard says markings on the neck & legs, etc. of a harldane are NOT to be faulted. So faulting those equally correct dogs is short sighted, even if certain patterns within the accepted range are fashionable or even more aesthetically pleasing.

3) Since harl & mantle dogs of less than ideal pattern are inevitably necessary to a breeding program, it is best to stay within the acceptable range of pattern as documented by the illustrated standard. All sorts of romantic legends about the value of mismarks in breeding programs abound, but it's a simple fact that no truly dedicated breeder would ever prefer a mismark to an animal correct to the standard. Apart from there being absolutely no documentation that mismarks "help" produce harls (in fact the evidence is quite the contrary), it is also obvious far too many mismarks are included in breeding programs simply as a cheap way to make harls.

4) Mismarks make more mismarks. Even if harls cannot breed "true," they & their mantle mates produce more correctly marked animals than mismarks do. And the problem with breeding & getting harls is not liscence to sell and breed every imaginable mismark that comes out of a harl litter. Many of these myths about the value of breeding mismarks are so obviously self-serving. Many who breed them are simply uneducated about harlequin genetics it would seem. The mismark who is of actual breed value is a rare creature from an unusual breeding. Predominately white dogs from harl breedings, if kept, should be BAER & CERF tested, according to DDC (German parent club) recommendations. Deaf/defective dogs used in breeding programs appear to have an increased incidence of deaf/defective offspring, even several generations later. Something to think about, as losses of normal & useable animals are to be expected when deaf dogs are used.

5) When using dogs either under or over marked, it is better judgment (if less the fashion or the common taste) to use dogs with more, rather than less pigment. This was the original intent of the standard of the country of origin & is still so designated in most standards worldwide. The AKC standard for the Great Dane has altered its color description during various revisions for a dog with more white than called for or originally prefered. It is obvious for better health that more white is less prefered.

For us to have the CORRECT harl & mantle range of color put up in the ring, we have to show the judges the correct range of acceptable markings; not some individual (or group's) aesthetic choice, or a fashion of the times--but TO THE STANDARD for the purpose of breed preservation. For us to safely breed dogs carry "lethal" & "semi-lethal" genes we need to consider the implications of simultaneously carrying 2-3 white genes well documented to cause serious health defects when "used to excess" to produce predominately white dogs. We should not encourage an increase in defects purely for the sake of a (currently) prefered aesthetic, sale of breeding stock, or ease of breeding choices.

And as not to forget the art in the science (of breeding), I offer this short poem by Gerald Manley Hopkins, who surely would have included in his beloved "dappled things" the glory that is the Harlequin Great Dane.

Glory be to God for dappled things--For skies of couple-colour as a brindle cow; For rose-moles all in stipple upon trout that swim; Fresh-firecoal chestnut-falls; finches' wings; Landscapes plotted and pieced--fold, fallow, and plough; And all trades, their gear and tackle and trim.

All things counter, original, spare, strange; Whatever is fickle, flecked (who knows how?): With swift, slow; sweet, sour; adazzle, dim; He fathers-forth whose beauty is past change; Praise him.

(Gerald Manley Hopkins)

Ackan, A. and W. Wegner. 1983. Veranderungen an Sehbahn und Sehzentren beim Merle-Syndrom des Hundes. Zeitschrift Fur Veruchstierkunde. 25(2): 91-9.
Ackerman, Lowell, DVM. 1996. Dr. Ackerman's Book of Great Danes.
Neptune City, NJ: T.F.H. Publications, Inc.
Burns, M. and Fraser, M.N. 1966. Genetics of the Dog: The basis of successful breeding.
Edinburgh: Oliver & Boyd.
Carroll-Draper, Nancy. 1981. The Great Dane: Dogdom's Apollo.
New York: Howell Book House.
Cattanach, B. (1999). The 'dalmatian dilemma': white coat colour and deafness. J. of Small Animal Practice 40: 193-+.
Clark, Ross D., DVM, and Joan R. Stainer. 1994. Medical and Genetic Aspects of Purebred Dogs.
St. Simon Island, GA: Forum Publishing.
Chopson, J. 1992. Inheritance of Great Dane Coat Color. GDCA Color Committee.
Daush, D., Wegner, W., Michaelis, M. and
I. Reetz. 1977. Opthalmologische Befunde in einer Merlezucht. DTW (Deustche Teirarztliche Wochenschrift). 84(12):468-75.
Daush, D., Wegner W., Michaelis, W. and
I. Reetz. 1978. Augenveranderungen beim Merlesyndrom des Hundes. Albrecht v Graff Archiv fur Klin. u. Exp. Opthal. 206(2):135-50.
Flach, M., Dausch, D., and W. Wegner. 1980. Floureszenzangiographie bei Teckeln. Weitere Befunde zum Merlesyndrom des Hundes. 8(3):375-83.
Gelatt, K.N., Powell, N.G., and K. Huston. 1981. Inheritance of micropthalmia with colomboma in the Australian shepherd dog. Am. J. Vet. Res. 42(10): 1686-90.
Green, B.K. 1974. The Color of Horses.
Flagstaff, AZ: Northland Press.
Greibrokk, T. 1994. Hereditary Deafness in the Dalmatian-Relationship to Eye and Coat Color. JAAHA 30: 170-176. Hoskins, J.D. 1990. Veterinary Pediatrics: Dogs and Cats from Birth to Six Months.
Philadelphia, PA: W. B. Saunders Company.
Johnson, Di. 1994. Great Danes Today.
New York: McMillian.
Klinckmann, G., Koniszewski, G. and Wegner, W. 1986. Light-microscope investigations on the retinae of dogs carrying the Merle factor. J. Vet. Med. A. 33:674-88.
Klinckmann G., Koniszewski, G., and W. Wegner. 1987. Lichtmikroskopische Untersuchungen an den Corneae von Merle-Dachshunden. DTW (Deutsche Tierarztliche Wochenschrifte). 94(6): 338-41.
Klein, E.,
Steinberg, S.A., Weiss, S.R.B., Matthews, D.M., and T.W. Uhde. 1988. The relationship between genetic deafness and fear- related behaviors in nervous pointer dogs. Physiology and Behavior 43: 307-312.
Klinckmann G., and W. Wegner. 1987. Tonometrien bei Merlehunden. DTW (Deutsche Tierarztliche Wochenschrifte). 94(6): 337-8.
1935. Dominant dilution and other color factors in Collie dogs. J. Hered. 26: 424-30. Krautwurst, Friedmar. 1997. Harlequin Breeding Forbidden in
Germany (translated by J.P. Yousha & Monika Olbrisch,
with assistance from Katja Hasslebach). Great Dane Reporter. M/A.
1991. Die Deutsche Dogge.
Jena: Salle-Verlag. Little, C.C. 1957. The Inheritance of Coat Color in Dogs. New York: Howell Book House.
The MERCK Veterinary Manual, 6th Edition. (1986)
Rahway, NJ: Merkc and Co., Inc.
Nouc, Winfired. Deutche Dogge: Das Rasse-Portrait. 1990. Muerlenbach: Kynos Verlag. O'Sullivan, Neil. (1988-89) "Harlequin colour in the Great Dane dog." Genetica. 78(3):215-8.
Padgett, George A., DVM. (1998) "Control of Canine Genetic Diseases."
: Howell Publishing.
I., Stecker M., and W. Wegner. 1977. Audiometrische befunde in einer Merlezucht. DTW (Deustche Teirarztliche Wochenschrift). 84(7):273-7.
Robinson, R. 1982. Genetics for dog breeders.
Oxford: Pergamon Press.
Schaible, R.H. and Brumbaugh, J.A. 1976. Electron microscopy of pigment cells in variegated and nonvariegated piebald spotted dogs. Pigment Cell. 3: 191-220.
Siegal, Mordecai, ed. (1995) UCDAVIS Book of Dogs. New York:
Sorsby, A. 1970. Ophthalmic Genetics. London: Butterworths.
Sorsby, A. and Davey, J.B. 1954. Ocular associations of dappling (or merling) in the coat color of dogs. 1. Clinical and genetical data. J. Gene. 52: 425-40.
Sponenberg, D.P. 1984. Germinal reversion of the merle allele in Australian shepherd dogs. J. Hered. 75:78.
Sponenberg, D.P. 1985. Inheritance of the harlequin color in Great Dane dogs. J. Hered. 76:224-5.
Sponenberg, D. P. and A. T. Bowling 1985. Heritable syndrome of skeletal defects in a family of Australian shepherd dogs. J. Hered. 76(5): 393-4.
Sponenberg, D.P. and Lamoreux, M.L. 1985. Inheritance of tweed, a modification of merle, in Australian shepherd dogs. J. Hered. 76(4):303-4. Steel, K.P., and C. Barkway.1989. Another role for melanocytes: their importance for normal stria vascularis development in the inner ear. Development 107: 453-463.
Wegner, W., and A. Akcan. 1980. Auswirkungen der Merlefactors auf die Area optica beim Hund. DTW (Deutsche Teirarztliche Wochenschrift). 87(9):342.
Willis, Malcolm B. 1989. Genetics of the Dog. New York: Howell Publishing. Yousha, JP. 1996. A Summary of Theories Concerning the Harlequin Variant in the Great Dane. Great Dane Reporter: M/A.


  Return Home


Copyright 1998 (revised 2001) J P Yousha, CHROMADANE.  All rights reserved.  Our thanks to the willingness to share this article for educational purposes.