This webpage is part of a series on Dog Coat Color Genetics and was last updated on March 24, 2010 by Sheila Schmutz, Ph.D.


Merle is a pattern with patches of intermingled colored and white hairs and other patches of solid color as is shown in the Australian Shepherd at the left. Patches of white may or may not be present. Flair is a merle with one non-merle parent and is therefore a heterozygote for merle. She also has tan points.

The Great Dane, Shetland Sheepdog, Collie, Border Collie, and Cathoula are other breeds in which a pattern called merle occurs. When merle occurs in Dachshunds, such as Sydney, it is called dapple. They bear certain similarity but not complete similarity in phenotype and side effects. The "merle" Great Dane pup pictured at the right is merle, a pattern that is disqualified from the show ring in that breed. This pup's pattern changed quite dramatically as it grew older and this is not typical in merle Australian Shepherds.

The merle pattern is inherited in dogs with a heterozygous genotype. One of the homozygotes is solid colored on the torso and the other is white or mostly white. Frequently these white dogs are deaf, and less commonly blind. This is an example where genes involved in pigmentation of hair or skin can also be associated with development of nerves. Most breeders have learned to avoid this problem by never breeding two merle dogs together.

The merle pattern was recently found to be caused by a mutation in the gene called SILV or PMEL17 by Keith Murphy's group at Texas A & M. The mutation is not a single base pair change but an insertion of a variable number of base pairs that appears to cause problems in processing the gene. This same gene causes white in chickens and silver in mice and silver dapple in horses and dun in some breeds of cattle. Although their study concentrated on Shetland Sheepdogs, they also studied dogs of other breeds with the merle pattern including Collie, Border Collie, Australian Shepherd, Cardigan Welsh Corgi, Dachshund, Great Dane. They also detected merle in some additional breeds such as American Pit Bull Terrier, Catahoula Leopard Dog, Chihuahua, Miniature Poodle, and Pyrenean Shepherd.

  • Clark, L.C., J. M. Wahl, C. A. Rees, and K. E. Murphy 2006. Retrotransposon insertion in SILV is responsible for merle patterning of the domestic dog. PNAS
  • (2004-

    An epistatic interaction of the MC1R e/e genotype prevents the merle phenotype. This is illustrated by the clear red Chihuhua, Mellow. Mellow is heterozygous for the the SINE & TTTTT..... insertion which causes merle. Her gentoype by band size was 200/500. Although most conscientious breeder try to abide by the rule not to breed merle to merle to avoid the mostly white dogs that are often deaf, no one could "see" that Mellow is a merle. Therefore in such breeds where both merle and clear red occur, one should not breed merle to merle or merle to clear red. This would apply to Chihuahua, Dachshund, Pomeranian and others.

    Hector (left) is a dapple Dachshund. His merle is difficult to see on his reddish coat but once one knows what to look for, it's visible.

    Cindy (right) is also a dapple Dachshund. Her merle pattern is almost more difficult to detect, but visible if one looks closely enough.

    Recall that dogs that are e/e do not have a single black hair, not even an eyelash. Merle doesn't show on e/e clear red dogs. Dogs that are reddish because of the ay allele often have a few black hairs or black tipped hairs. Merle can be difficult to see, but is detectable on them.

    If you are not sure which type of red your Dachshund is, then you could have it DNA tested at HealthGene for this. If the dog is e/e then it's best not to risk breeding it to a merle dog. HealthGene is not licensed to test for merle. The test was patented and a sole license was sold. The test has not been widely available therefore, and may continue to be unavailable.

    The dog on the left is an interesting example of how this coat color pattern appeared to affect eye color since the right eye is blue and accompanied by a large white patch on that side of her face and her left eye is amber and accompanied by a merle patch. She sees normally but is bilaterally deaf. A few homozygous merle dogs have microphthalmia or small eyes, which typically are malformed and have no visual function.

    The variety of coat color amount and pattern seen in homozygous merle dogs is extensive as shown in the color plates below. All of these Australian Shepherds but one are homozygous M/M. All are reported to be deaf and/or tested as deaf with BAER testing. In fact 27 of 29 of the Aussies we genotyped that were M/M were deaf. The one dog shown here that is M/m, could hear and was only deaf in one ear by BAER testing. After you've tried to find that exception, scroll to the bottom of this page for the answer to see if you chose correctly. It is harder than one might think!

    Subsequent to the study described in Aussies above, a study was conducted by George Strain and others using BAER testing and genotyping of the SILV mutation. They studied dogs of several breeds. Their study included 40 dogs that were M/M. Twenty-nine of the 40 were Catahoula, a breed with no white markings. Only 2 Catahoula were deaf in both ears and 1 was deaf in one ear. Of the remaining 11 dogs, 4 were deaf in both ears and 3 in one ear. These 11 dogs included 5 Aussies, 3 Collies, 1 Sheltie, 1 Corgi, and 1 Great Dane. Only 1 of the 113 dogs with a M/m genotype was deaf, confirming again that heterozygous merle dogs are not prone to deafness.

    Previous studies to find the gene causing Merle

    We had tried to study this color pattern with the help of many breeders and owners. C.A. Sharp, a noted dog writer and speaker, has been instrumental in helping us obtain the much needed DNA samples from Australian Shepherds. We also have one litter of Dapple Dachshunds. We had hoped to find the gene causing merle and then to try to understand why this gene appears to be associated with deafness and/or blindness in certain cases. We ruled out KITLG (previously known as MGF) as this gene, although we did show that it does cause roan in cattle.

  • Schmutz, S. M., T. G. Berryere, and C. A. Sharp. 2003. KITLG mapping to CFA15 and exclusion as a candidate gene for merle. Animal Genetics 34: 75-76..
  • We also studied a gene called microphthalmia transcription factor or MITF through an Acorn Grant from the Canine Health Foundation. This gene has a form unique to the melanocytes and than other forms in other tissue types. The DNA coding sequence of the melanocyte form (MITF-M) was not different between merle and non-merle dogs. Although this gene made considerable sense as a candidate gene because it also affects eye development, since the melanocyte form is present in the retinal pigmented epithelium, our data suggested MITF was not the gene causing merle in Australian Shepherds (nor microphtalmia in Akitas as explained on another page). However MITF has recently been shown to cause white spotting in some breeds, if not all (see the page on spotting).

    We checked if the merle pattern co-segregated with markers within or near 4 more genes, using these litters. We also ruled out EDNRB which causes the overo spotting pattern in horses which has gut enervation problems in the homozygote. In mice a pattern called mottled is caused by a tyrosinase mutation and we have also ruled this gene out. Finally, we have ruled out 2 other spotting genes in mice: PAX3 and KIT.

    Benoit Hedan, a Ph.D. student in France presented a very comprehensive study of candidate genes for merle at the International Pigment Cell Conference in September, 2005. He used a very large family of Australian Shepherds. He excluded about 15 genes, including those we studied. Although his work did not detect the gene causing merle, since PMEL17 was not studied, it does provide a good collection of markers which could be used to study spotting, ticking and roan.

  • Hedan, B, S. Corre, S. Dreano, T. Vilboux, B. Denis, F. Galibert, M.D. Galibert, C. Andre. 2005. Coat color in dogs: a powerful model for mammalian pigmentation. Pigment Cell Research 18 ( Supp. 1):65.
  • Hedan B, Corre S, Hitte C, Dreano S, Vilboux T, Derrien T, Denis B, Galibert F, Galibert MD, Andre C. (2006) Coat colour in dogs: identification of the merle locus in the Australian shepherd breed. BMC Vet Res.

  • ANSWER to merle "quiz": Cayene, the dog doing an agility jump is M/m whereas all others tested M/M. Note that the others lead great lives and many do agiity too!

    back to Dog Coat Color Main Page