Australian Shepherd Health & Genetics Institute

Australian Shepherd Health & Genetics Institute

Generic selectors
Exact matches only
Search in title
Search in content

Heresy Revisited

Is crossbreeding ever justified?

By C.A. Sharp

First Published in Double Helix Network News, Fall 2011

A dozen years ago I wrote an article called “Speaking Heresy.”  It was about crossbreeding, the intentional mixing of different breeds of dog.  Since then the designer dog fad and it’s plethora of whatever-doodles has, if anything, made purebred dog enthusiasts even more critical of the practice.  In spite of this, the Dalmatian Club of America recently voted to allow the “Backcross Dals,” descendants of a long-ago crossbreeding with a Pointer, into their AKC studbook.  In view of that and other developments I felt it was time for another look at the practice of crossbreeding, our almost universal knee-jerk condemnation of the practice, and whether it might have a place in purebred breeding.


For over a century cross-breeding has been looked on with scorn as an adulteration of pure canine blood that took generations of dogs and humans alike to develop.  “Designer” dogs are disdained for their mixed ancestry, as are those who seek to create new breeds by combining the genes of established breeds.  Within recognized pure breeds it is, with extremely rare exceptions, forbidden.  Whatever the reasons for crossbreeding today – and the motivations can be highly varied as to purpose and ethical practice – practitioners are generally sanctioned, socially and sometimes formally, by purebred dog enthusiasts and organizations.

Despite the overwhelming pejorative view of the practice in dogs, crossbreeding is a legitimate and useful breeding technique.  Early in the history of pure breeds it was commonly practiced for a variety of reasons.  Understanding this history, why those crosses were done, and how the practice can be effectively employed today is something every breeder should be aware of even though most will never employ it.

Early history

The modern concept of pure breeds is just that:  Modern.  Prior to the 19th century there were no breeds as we know them today.  There were particular types of dog found in certain geographical regions, developed over generations to meet needs varying from hunting to farm work, or vermin control.  Members of the aristocracy might selectively breed a unique strain but many types arose because dogs who would do a certain job were important to people’s livelihood.  None of these strains were purebred in the sense of studbooks and formal closed registries.  Breeding records might be kept by the literate, but in those days literacy was the exception.  Even so, though some non-literate cultures, like the nomadic Bedouin, had a strong oral tradition and might commit detailed pedigrees of valued dogs to memory.

Unhindered by closed studbooks and registry restrictions, if a nobleman of times past thought a different sort of dog had something to offer his bloodline or a farmer felt it might throw better work dogs with his bitch,  that other sort of dog would be used.  If the offspring proved worthy, they would be kept and bred from with no one thinking any the less of them for their mixed ancestry.

Crossbreeding was a feature in the development of many of our modern breeds, with perhaps the most well known being the Doberman, created by tax Collector and pound-master Louis Dobermann, who selectively bred a variety of dogs to get the personal protection animal his tax duties required.  However, the Doberman is far from the only breed that sprang from purposeful crossbreeding.  The Bull Terrier, originally called the Bull and Terrier was, as that early name suggests, derived from a cross between bulldogs, terriers, and possibly other breeds.  In 1859 the Victorian dog authority “Stonehenge,” John Henry Walsh, condemned the early Bull Terrier for its mixed heritage, but little more than a decade later Walsh included it as one of eight recognized terrier breeds.

In the 19th century a whippet was crossed with a Black and Tan Terrier (today called the Manchester) to produce a more elegant animal that appealed to city gentlemen who wanted a vermin dog but also wanted something stylish, forever changing the look of that breed.  For similar reasons Borzoi blood was introduced to the Collie, making the one-time farm dog more appealing not only to the aristocracy but to the up-and-coming middle class.

The Irish Wolfhound is the product of a purposeful attempt to recreate an historic breed, few if any of which had survived to modern times.  Dogs which had the look of the ancient breed, as depicted in artwork, were selected.  These dogs had little or no known pedigree and no clear connection to the original breed.  The related Scottish Deerhound was used to help fix the type.  Borzoi and Great Dane were introduced to increase size.

Not all crossbreeding took place early in purebred history.  The Red Setter, a field version of the Irish, was banned from the Irish studbook after breeders introduced English Setter to improve working traits.  More recently breeders who desired a coated sighthound of moderate size crossed a Shetland Sheepdog to a whippet then back-crossed to whippets selecting for the long coat.  The result was the Long-Haired Whippet, which has a modest and dedicated following but remains an object of scorn to Whippet enthusiasts and lacks AKC recognition.

The Rise of Registries

Dog registries were originally developed for record-keeping purposes but ultimately morphed into arbiters of canine purity.  That aspect of their function is the child of the class-conscious European social thinking of the 19th century, when “pure blood” was tantamount to good breeding for humans, not only among the nobility but also for the rising merchant class and gentleman farmers.  Marrying “beneath one’s station” was thought to introduce “inferior blood” to a family and would necessarily result in offspring of poor character and limited ability.  This thinking was transferred to the breeding of animals, including dogs.

Other species have pure breeds with formal studbook registries, too, but only in dogs is the use of crossbreeding almost uniformly condemned.  Pure breeds of livestock are maintained, but for commercial production growers don’t hesitate to crossbreed if the result is a more marketable carcass or improved production of food or fiber.  Various horse registries will admit crossbreds under certain circumstances:  Racing lines of Quarter Horse can have some Thoroughbred background and part-Arabians are allowed registration.

Why crossbreed?

We tend to focus on negative applications of this very old breeding practice:  The blatant commercialization of the “designer dogs” or the sub rosa use of one breed to improve ones chances of competing successfully, as with the use of Pointers with some field dog breeds, Border Collies with working line Australian Shepherds, or the alleged use of an Afghan Hound a couple decades ago to improve show coat in Irish Setters.  Turning dogs into consumer fad items or cheating in competitive events are rightly condemned, but was the fault here crossbreeding or some combination of hubris and greed?

Crossbreeding has some well-established benefits, which explains its continued use in livestock.  The first generation offspring of a cross between two pure strains exhibits hybrid vigor, an improvement of health and other biological qualities derived from increased heterozygosity of the genes in the offspring.  Designer dog merchants often tout improved health conferred by hybrid vigor when marketing their product.  However, hybrid vigor only occurs in the first generation.  If you backcross those dogs to each other or either parent breed, you reduce the level of heterozygosity and no longer have the benefit of hybrid vigor.

Crossbreeding can be used to develop an improved strain suited to a specific purpose, to reintroduce a lost trait, or improve on existing ones in an established breed.  We saw examples of this in the formation of our pure breeds as well as in livestock and horse breeding.  There are legitimate examples of crossbreeding of dogs in recent history.  Some guide dogs associations have crossbred Labrador and Golden Retrievers because they found the crossbred offspring were less prone to hip dysplasia and successfully completed their training more frequently than purebreds of either breed.  Before Labradoodles became fashion statements, their early breeders’ goal was a well-tempered, readily trainable, low-maintenance pet for the average family.

Crossbreeding can be effective in introducing a trait that is lacking, correcting a problem, or expanding a tight gene pool.  Its use for purposes like these should be considered legitimate, provided the crossbreeding program is open as to its practices and goals and, if focused on a particular breed, it works within the established breed organization and registry.

Documented Crossbreeding Efforts

The Backcross Project – Prior to this project, all Dalmatians had two copies of a mutation that causes high uric acid, which frequently leads to bladder stones.  In the mid-1980s Robert Schible PhD, a medical geneticist, crossed a Dalmatian bitch to a Pointer in order to introduce the normal version of the gene causing high uric acid.  For the subsequent four generations the progeny were bred to Dalmatians, resulting in dogs that looked like purebred Dals, though they did have a tendency to have patches in addition to spots, a significant color fault in the breed.  This effort was done with the knowledge and support of the board of the Dalmatian Club of America (DCA).  At the request of that board, the American Kennel Club agreed to register two of the 5th generation pups.  However, club membership had been unaware of the project and when they found out they were outraged and overturned the sitting board.  The new board requested that AKC no longer accept the backcross dogs and the registry complied.  In spite of the set-back, a dedicated group of Dalmatian breeders maintained the Backcross bloodline which today is indistinguishable from other Dalmatians.

The genetic mutation that causes high uric acid was identified a couple years ago and there is now a test available.  A proposal was made to the DCA membership that the backcross dogs, whose only non-Dalmatian heritage was that one Pointer a quarter century ago, be allowed back into the AKC registry.  Initially, the membership voted the proposal down.  Proponents launched an education campaign and earlier this year the DCA members voted, by a narrow majority, to allow the dogs in.  These dogs and the DNA test will allow Dal breeders to gradually reduce the frequency of the high uric acid mutation in the breed.

The Bob-tail Boxer Project – With the probability of a docking ban in the United Kingdom, Bruce Cattanach PhD, a medical geneticist and Boxer breeder, developed a plan to introduce the bob-tail gene into his long-tailed but traditionally docked breed.  He bred one of his bitches to a bob-tail Pembroke Welsh Corgi then backcrossed to Boxers for four generations, selecting the most Boxer-like offspring to create each subsequent generation.  Despite the significant morphological differences between the breeds, the 4th generation dogs looked like purebred Boxers.

Docking is now banned in the United Kingdom.  The Kennel Club agreed to accept the dogs into its registry, but with the bob-tail gene present in the breed, Boxer breeders can produce short-tail dogs if they wish.

The Swedish Clumber Spaniel Project – The Clumber Spaniel has a restricted gene pool.  Members of the Swedish Clumber Spaniel Club have been concerned about this for years and discussed crossbreeding as a way of widening their gene pool.  Drs. Lennart Svennson and Per Erik Sundgren agreed to assist.  In 2003 a Clumber bitch was bred to an English Cocker Spaniel of working type.  The ten resulting puppies were examined by show judges and the two bitches of the most consistent Clumber type were admitted into the Swedish Kennel Club’s registry, which includes an “X” in the registration numbers of project dogs so they can readily be identified.  At this time those bitches have had three litters between them.  Some of those dogs are competing successfully in conformation across Europe.  Since most of the dogs are still relatively young the health benefits and working quality can’t yet be assessed but it is hoped the new genetic material will result in improved health and longevity.

How-To Manual

If crossbreeding is to be employed it must be for a well-defined purpose and with a detailed plan of action.  It must also be open, with everyone involved aware of the heritage of all dogs produced in the effort.  If the project is intended to introduce something lacking to an established breed, the national club and the registry must approve the effort beforehand, be fully aware of the details, and club membership needs to be informed and educated about the purpose and progress of the effort.  Failure to do so can doom the project, as exemplified by the long, rocky history of the Dalmatian Backcross Project.

The DCA board of the time failed to inform membership about what was being done.  It was presented as a fait acompli after AKC registered the two Backcross dogs.  Not surprisingly, the membership voted the responsible board members out of office and the new board stopped the registration of any further Backcross dogs.  It took a quarter century to undo the damage.

The Bob-Tail Boxer and Clumber projects have had much better initial success.  Both were done with the knowledge and approval of all parties.  Dr. Cattanach at the time was a member of the KC’s health committee.  The greater majority of the Swedish Clumber club’s membership were the guiding force behind that effort and allowed the SKC close oversight of everything they did.  However, the Boxer project received a setback when the German breed club (Germany is the breed’s country of origin) petitioned the Federation Cynologique Internationale (FCI), which governs shows and kennel clubs throughout most of the world, to ban the showing of any Boxer with a “naturally stumpy tail.”  The bob-tail Boxers can still compete in the UK, Australia and the United States, whose kennel clubs are not FCI members.

Where to go?

The choice of the other breed for a breed improvement crossbreeding project requires careful consideration.  The more similar in appearance and behavior the other breed is, the easier it will be to get consistent quality in the backcross generations.  A Pointer was chosen for the Dalmatian project because the breeds are similar in body structure.  In addition, Pointers are typically mostly white with patches and small ticking spots where properly colored Dals are entirely white with large round ticking of relatively uniform size and distribution.  The field-bred English Cocker was used for the Clumber project because the breeds are related and have similar hunting styles and coat patterns.

The use of a Corgi for the Bobtail Boxer project may seem a stretch, but there is no bob-tail breed that is morphologically similar to a Boxer.  Dr. Cattanach demonstrated that even with such a drastic difference one can produce quality within a very few generations.  It helped that two of the most un-Boxerlike Corgi traits, short legs and a longer coat, proved to be recessive and easy to breed out.

The choice of the other breed is also dependant on the purpose for the effort.  For the Dalmatian and Boxer projects, the goal was to introduce the desirable version of one gene.  An individual from any breed in which each dog has two copies of the desired gene version would pass a copy on to each of its crossbred offspring.  However, using a breed physically and/or behaviorally similar – if available – would make the job easier.

If the problem to be addressed involves many genes – as with augmenting a tight gene pool – the problem becomes far more complex.  The cross could influence many important traits, not all of them easily observed or measured.  Insofar as this improves the health, viability, and longevity of the breed, this is all to the good.  But behavioral and physical traits that don’t conform to the breed standard will need to be selected against and more generations may be needed to fix breed type.  At this point there is no easy way to track all of the newly introduced gene versions from one generation to the next, so we can’t know how many are retained as we backcross to the original breed.  We will lose those that are linked (close on the same chromosome) to genes that we don’t want – those that contribute to incorrect appearance or behavior – because of strong selection against that neighboring gene.

The Clumber project is still in the early stages so traits more typical of the English Cocker still occur.  In time that will be sorted out, but in the process some Cocker genes not causative for the Cocker traits will go with them in the effort to get proper Clumber type back.   Thus far the Clumber project has used only a single outcross and the long-term benefits remain to be seen.  With projects of this sort it may prove necessary to repeat the process more than once to maximize the inflow of beneficial gene variants.  The different outcrossed lines would need to be kept distinct from each other not only until acceptable breed type was established but for a few generations thereafter to make sure the type remained consistent.  During this time, the strains could be crossed with any suitable pure line.  The point would be to avoid having more than one dog of the other breed in the recent pedigree.

MHC profiling  

An approach toward improving health in breeds with extremely tight gene pools that no one, to my knowledge, has tried yet would be to focus on introducing increased diversity at the Major Histocompatibility Complex (MHC) a group of genes that govern the functioning of the immune system.  Those genes are all grouped on canine chromosome 12 and tend to be inherited together in sets called a haplotypes.  MHC genes are highly variable and many different haplotypes exist in the dog as a species.  However, individual breeds each have only a subset of the total.

Popular sire breeding can reduce the number of haplotypes in a breed by unintentionally selecting against those which the popular sires do not possess.  A dog can have, at most, two haplotypes – assuming the one inherited from his sire was different that the one he got from his dam.  Breeds with small founder bases or small populations also tend to have fewer haplotypes.  This can impact immunity to infections and the frequency of immune mediated diseases.

At this point researchers have identified MHC haplotypes for around 80 breeds, though the findings thus far may not be complete.  Also, there is no commercial diagnostic test available to tell you what MHC haplotypes a particular dog has, so launching a MHC diversity crossbreeding project at this point isn’t practical.  In time it may be.

If MHC genotyping becomes available and the haplotypes present in a breed are known, a breed with different haplotypes could be identified and individual dogs tested to verify that they carry two “new” haplotypes before they are used for a crossbreeding.  Such dogs could be bred to bitches of the first breed and the progeny backcrossed to the original breed then selected for breed traits and the novel haplotypes.  In time one could introduce greater MHC diversity and better health to the original breed.  More than one outcross should be done so multiple new haplotypes could be introduced.

Measuring Success

Crossbreeding is not something to be approached lightly.  For it to be successful it must be well planned and executed with the support of all stakeholders.  Provisions must be made to reintroduce descendants of the outcross to full registration as members of the breed.  England’s KC will do so after four backcross generations and the AKC requires seven. Perhaps the most important way to ensure success is to engage the breed club’s membership.  Without them the project will fail or, like the Dalmatian project, may take decades to gain acceptance.

Beyond the political considerations, success must be measured in how well the descendants are integrated into the breed as a whole.  Even with full registration, if they remain shunned by the majority of breeders, the benefit derived from the crossbreeding will be limited.

Crossbreeding is not a panacea for pure breeds with substantial health issues or which have lost once-valued traits.  However, with due consideration and planning, it is a legitimate and effective practice which could benefit some breeds. To dismiss it out-of-hand is short-sighted and of no benefit to our dogs.