Australian Shepherd Health & Genetics Institute

Australian Shepherd Health & Genetics Institute

Generic selectors
Exact matches only
Search in title
Search in content

Breeding FAQs


What’s the difference between inbreeding and linebreeding?

Linebreeding is an animal breeder’s term that means the mating of animals that are related, often through one or more common ancestors, but not within one or two steps of relationship.  Inbreeding is the mating of those very close relatives.  In this context, inbred matings are so close they would be considered incestuous among people (parent/offspring, siblings, etc.)  To a geneticist, however, both are “inbreeding” because you are in both cases breeding related individuals with each other.

The advantage of inbreeding and linebreeding is that they increase your odds of matching up genes that occur in the common ancestry.  That is also their downfall.  When you inbreed or linebreed you do so to produce dogs with the positive traits of their progenitors.  But every dog, even the very best of them, will also carry genes for things you do not want.  You are just as likely to double up on the genes you don’t want as those that you do, because nearly all of the unwanted ones are either recessive or contribute to a polygenic trait and can be difficult to track through a pedigree.  The more inbred or “tightly linebred” the individual is the greater the chance of matching up genes both good and bad.  It is possible to calculate the degree if inbreeding in an individual by using a formula called Wright’s Coefficient of Inbreeding (COI).

How close a breeding is “too close?” 

There isn’t a simple answer to that.   There may be compelling reasons to do a fairly close breeding.  However, if it’s done repeatedly over multiple generations there can be an accumulation of genes you don’t want along with those you’ve been intentionally collecting.  Another hazard of close breeding is the popular sire.  That dog, however good he is, will have some bad genes.  The more often he and his close kin show up on both sides of pedigree, the more likely you are to double up on some of those unwanted genes.

Is inbreeding good or bad?

Advances in genetics are calling a number of long-held breeder beliefs into question, including our devotion to inbreeding as the best way to produce good dogs.  Inbreeding, which for this discussion also includes what we call “linebreeding,” certainly has its place.  None of our domestic animals or plants would exist without it.  Inbreeding must be used to develop a strain that breeds true for whatever traits are considered important for the breed.  But true-breeding strains of most purebred dog breeds were established decades to centuries ago.

Inbreeding concentrates the “good” traits that we want, but it also inadvertently concentrates things we do not want some of which are very bad.  Any dog, no matter how high in quality, will have some “bad” genes.  Beyond that there will be sub-optimal genes that in and of themselves only cause the dog to be a little less efficient at some bodily process or function, but if these genes accumulate a significant health problem may result.

Any dog or group of dogs that are used extensively and are then subject to extensive linebreeding will have both positive and negative effects on a breed.  The negative effects may be insidious and not easy to breed away from once most the breed is to some degree related to that dog or line.

There is definite advantage to using a proven stud from a well-known line—a dog that has offspring (preferably adult offspring) available for you to evaluate.  You can probably find out a great deal about the conformation, temperament, and performance qualities of the offspring.  But beyond standard health screening (hips and eyes in my breed) you may be hard pressed to get any detail on health traits.  Even with hips and eyes, affected offspring may not be reported openly.  For many health issues there is no screening test.

Some feel that a good sire that does not carry a particular problem can dilute the effects of the unwanted gene.  If the popular sire does not carry a particular recessive disease gene, his offspring cannot get that disease.  But if the gene is frequent in the population, it will surely re-surface in the subsequent generations.  If one of the clear dog’s carrier sons also becomes a popular sire you are back where you started.

Most diseases are not single gene.  Diseases with complex inheritance are very difficult to eliminate unless you have very good information on the status of many members of a dog’s family, collateral relatives as well as his descendants and progenitors.

A great deal has been learned about how inbreeding affects genetic health and longevity over the past few decades, though few of the studies have been on dogs.  However, what goes for other mammals is very likely to apply to dogs. One key example is the Major Histocompatibility Complex (MHC).  This group of genes controls important functions of the immune system.  It is a group of genes located close together on a single chromosome.  Therefore, they are almost always are passed on as a group.  Each dog can have only two sets of these genes, or haplotypes.  However each MHC gene has multiple versions, sometimes as much as a hundred.  This is necessary so the species has sufficient genetic diversity to deal with ever-evolving pathogens.  Some individuals will die from a nasty new bug, but others will survive.  But maybe not if there has been too frequent use of popular sires or lines and inbreeding.  The necessary MHC alleles may have been lost.  Loss of MHC diversity also contributes to autoimmune disease frequency.

Another thing to keep in mind about inbreeding:  It is not the rule in nature.  Most species have behaviors that will prevent them from breeding to near relatives unless circumstances give them no other choice (stuck on an island, for example.)

From a practical standpoint, you cannot avoid some level of inbreeding if you are going to breed dogs.  But breeders would be wise to know coefficients of inbreeding on their dogs and take them into consideration when making breeding decisions.  If COI in your kennel or the breed has been climbing generation after generation, especially in recent years (recent inbreeding seems to be more problematic than old or “historic” inbreeding), sooner or later there will be a price to be paid.

How do I know if my dog is too inbred?

First you need to know how inbred your dog actually is and a printed pedigree usually doesn’t contain enough information to tell you.  Find out what your dog’s 10-generation coefficient of inbreeding (COI) is.

A half-brother/sister cross would have a COI of 12.5%.  A parent/offspring cross would be 25%.  In both cases this assumes there are no other common ancestors in the pedigree.  10-generation COIs in Aussies average about 14%.  Breeders should try not to let the COIs on their dogs continue to creep upward.  There may be excellent reasons for doing a very close breeding, but it should not be done generation after generation.

In general you should try not to exceed an average of the two parents’ COIs.  If at all possible, shoot for something below the breed average (12.5%).  In a perfect world, they should all be below 10%, but this is not a perfect world.

The COI of any individual is less important than the breed average and trends within particular lines.  You might have a very good reason to do one high-COI breeding, but doing it generation after generation can lead to inbreeding depression and can unintentionally increase the risk of inherited disease.

If you have a really good litter, isn’t it best to keep repeating the cross?

If you had exceptional results once you will probably do so in repeat breedings.  The down side is that you will then have that many more dogs of exactly the same breeding in the gene pool.  If the sire of this cross is popular, there will be even fewer less-related dogs available as mates.

Males have the potential to sire offspring with a variety of mates.  Even a fairly limited breeding career can produce many different genetic combinations with his genes.  A bitch’s potential is limited by the number of litters she can produce.  If you breed her only to one dog there will be only a limited number of genetic combinations from her. You maximize a good bitch’s potential contribution to the breed by not breeding her multiple times to a single sire.

 There is no dog or cross that is totally free of unwanted genes.  It is far better to have offspring with different combinations of genes available through crosses with different mates than to have many full siblings.  If you like what you got from a cross, seek other studs of equal merit to the first one, whose qualities complement those of your bitch, and who are not close kin to the first one.

 I want to import a dog, what tests should I want done first?

At an absolute minimum the dog’s hips and elbows should be screened (assuming it is an adult), it should have a clear eye exam within the past year, and be tested clear or cleared by parentage (both parents tested clear) for MDR1, cataract, CEA and PRA.  Ideally, it should have a PHA test with a negative result as well as all the DNA tests available for Aussies except where cleared by parentage to protect you from importing anything carrying those things for which we have tests.  If you are importing semen the stud dog should have all available tests done and screenings.

Which genetic problems can you breed away from and which should you not breed relatives?

When a genetic disease or defect is identified, there are a number of issues that come into play in the decision about whether to breed that individual or not and whether relatives should be withdrawn from breeding.  There are four major areas of consideration:

  1.  How seriously does it impact the dog’s health and ability to function?  Example:  Hemangiosarcoma is lethal; a missing pre-molar has minimal effect on dental function.
  2. How common is it in the breed?  Example:  The MDR1 mutation is extremely common; Von Willebrand’s Disease is extremely rare.
  3. How large is the available breeding population?  Example:  In the United States there are many Aussies, in South Africa there are relatively few.
  4. What is known about the mode of inheritance?  Example:  Collie eye anomaly is recessive; hip dysplasia is polygenic.

If a dog’s problem is so minor that it will never require medication or surgery and has at most very minor effect on the dog’s ability to function, one might breed it.  However, selecting an unaffected littermate might be the better choice.  The classic example would be a dog with only one or two missing pre-molars.  If you breed such a dog, you must take care to mate it to animals that have full dentition and have no recent family history of missing teeth to reduce the risk of producing it in the offspring.

If the problem is extremely common, it may not be possible to avoid breeding an affected animal if the breed itself is to be preserved.  Fortunately, we do not at present have anything that is this extremely common in Aussies other than the MDR1 mutation, which is only an issue if we give the wrong medications.  However, a more serious health issue could become common in smaller, geographically isolated populations as the breed makes its way around the world.

If the local population is very small, as is the case for Aussies in some countries outside North America, one may have to weigh the risks of various problems in the available lines.  If CEA were very rare and hip dysplasia very common, it might make sense to breed a dog that had CEA but was from a line clear of hip dysplasia.  It shouldn’t be bred extensively and the offspring would have to be followed and only the ones free of both defects used for further breeding.

Another example might be a working dog which exhibits extremely valuable working traits but has a hereditary disease or defect.  Working dogs are less numerous than the non-working population, so breeders have fewer options.  In such a case, the breeder might use that dog with one that has no family history for the unwanted trait, and then proceed with the clear offspring.

If you know how something is passed, and it is simple in inheritance, this can be a guide for which dogs to keep and which to pass by.  PRCD/PRA is a recessive. If it is produced, both parents carry it.  Hemophilia is X-linked.  If a dog has it, he got the gene from his dam not his sire.  If you don’t want to produce these defects, you either don’t use dogs which are proven carriers or, where there is a DNA test – as with PRCD – you breed them only to dogs which you know do not carry the trait.

If the mode is complex (polygenic or genetically predisposed) it is difficult to know how much each parent contributed but both will have contributed something.  It may be 50/50, but it might be 95/5—with no telling who gave which.  In this case, a close examination of family history–not just the ancestors on the pedigree but as many collateral relatives as possible—may give an indication of how common the problem is behind one parent or the other.  Certainly, if a particular individual produces the unwanted trait in multiple offspring, especially with different and unrelated mates, odds are high that individual is “loaded” with the genes and it would be best to cease breeding it.

The most commonly seen (2% or more) inherited issues in Aussies are (in rough order of frequency in the breed):

  • MDR1
  • Dental faults
  • Allergies
  • Hip Dysplasia
  • Autoimmune Disease
  • Hereditary cancers
  • Hernia
  • Epilepsy
  • Cataracts
  • Retained testicles
  • Elbow dysplasia
  • Persistent Pupilary Membrane (in dogs over 6 months old)
  • Distichiasis
  • Laryngeal Paralysis

Here is same list re-ordered to indicate health impact:

  • Hereditary cancers (generally lethal)
  • Epilepsy (potentially fatal)
  • Autoimmune Diseases (potentially fatal)
  • Hip Dysplasia (painful, potentially crippling, and can require extensive surgery)
  • Elbow Dysplasia (painful and potentially crippling)
  • Laryngeal paralysis (late onset with significant quality-of-life impacts, ultimately
  • fatal)
  • Cataracts (potentially blinding)
  • Allergies (can cause severe discomfort)
  • Distichaisis ( occasionally causes discomfort or requires surgery to avoid corneal
  •    scaring)
  • Persistent Pupilary Membrane (occasionally blinding)
  • Retained Testicles (can result in sterility, surgery generally done to reduce cancer risk)
  • Hernia (occasionally requires surgical correction)
  • Dental Faults  (when severe will interfere with jaw function)
  • MDR1 (not a problem, so long as the dog isn’t given the wrong drugs)

While one might argue exact placement within a position or two, most people familiar with these conditions would agree that those at the top of the list are by far more serious than those at the bottom.

The only thing in the above list with a proven simple inheritance is MDR1.  Others are all either polygenic or the mode is undetermined.  For some, including hip dysplasia, autoimmune disease, and allergies, the environment plays a role.

Unfortunately, there are no simple rules to apply to the management of genetic disease.  But if a breeder chooses to breed an animal that is affected with something, or one that has produced it, she should do so only with the greatest care to avoid producing affected offspring and with complete openness about what the disease is and where she has produced it, keeping in mind that there is no such thing as a 100% genetically “clean” dog or line.  They all carry genes for something you don’t want.  The trick is to minimize risk for producing the bad stuff while you maximize the probability of getting the things you want.

ASHGI advocates making health a breeding priority. How can I do that and keep producing quality dogs?

People are justifiably concerned that they might take a step backwards on their breeding goals, either for competitive venues or real work, if they make health risk reduction a major factor in breeding selections.

Whatever your goals are, if you shift exclusively to a different set of priorities you will probably not make as much progress, or may even backslide a bit, on your goals.  A breeder should never make any single criterion, whether winning in the show ring or reducing epilepsy risk, the only reason for doing a breeding.

Forget health for a minute and think about conformation or working ability.  If you’ve consistently produced a trait you really like but have a problem with another that you want to address, you select a mate that you hope will improve on that weak point but is also of good quality as regards other important traits, though you may need to tolerate somewhat lesser quality on some item where your dog is strong in favor of the overall package.

There are many factors you need to consider when planning any breeding, but too often we have a tendency to consider everything else first (“Oh, but he’s such a nice dog!) and let health be an afterthought.

How can a new person learn more about health and other qualities in the breed if they want to start breeding?

First you need to learn the questions to ask.  Do everything you can to educate yourself on the traits necessary to the type of dog you want (conformation, stockdog, agility dog, etc.) as well as genetic issues in the breed.

Remember that winning isn’t everything.  If someone is very good at promoting his/her kennel and has a lot of winners out there, it does not automatically follow that those are the best dogs.  The best promoted and most skillfully trained and campaigned, perhaps, but not necessarily the best quality or the soundest in health.   They may be, but you need more information than what you get from listings of show and trial wins or promotional copy.

For health traits, this website is an excellent resource.  Follow the links to other sites and read what you find there.  Use social media and e-mail lists to read and participate in breeding discussions.  You should include breed lists, multi-breed, and health or genetics related forums to tap a broad knowledge of dogs.

Talk to people.  Keep your ears open at shows, club meetings and other places breeders gather.  If you approach someone with a question, don’t be abrupt or accusatory if you want answers.  Keep the questions general.  (“I’m new and want to learn about breed health issues.  Can you tell me something about what I need to study up on?”)  This may lead to more fruitful and specific discussions.

If you have an Aussie and you are considering breeding, you should discuss pros and cons of the dog and its pedigree with the breeder of your dog.  When you know the weak points, you should discuss those with owners of mates you are considering.  While not everyone is totally forthcoming, most will not want to double up on a fault or health issue when an affected pup could later be associated with their kennel.

Becoming a student of human behavior can help.  Sometimes non-answers, evasions, and other behavior can tell you as much as a more constructive conversation.  It won’t tell you exactly what that person wants to hide, but it does indicate that maybe you need to find other sources for information.  If you feel uncomfortable, there is a reason even if you can’t put a finger on it.

Can you have a line that is “clean” for something and “suddenly” get it when you do an outcross?

You are less likely to double up on unwanted genes with an outbred pedigree than by inbreeding (which includes linebreeding).  However, it is possible to wind up with something you didn’t think you could get when you do an outcross, especially if the inheritance is complex.  This is sometimes called “outcrossing surprise.”

Outcrossing surprise occurs because the “clean” line no longer has all of the genetic puzzle pieces to make the whole picture.  If the outcross line happens to have the missing pieces – Bingo!

You are far more likely to have problems crop up by breeding close on dogs from a family that already has the problem–particularly when information about affected dogs is not freely exchanged.  Doing some homework on the outcross line may give you a better idea whether surprises might be lurking there.

A final caveat–very often when people say they have an “outcross” it isn’t really all that much of an outcross.  There may not be any common names on the printed pedigrees of sire and dam, but if you go a generation or three farther back you will find them.  A coefficient of Inbreeding (COI) on your dog and the potential cross is a better indication of how close the breeding would be.

What is a coefficient of inbreeding (COI)?

COI is a measure of the common relationship between the two sides of a pedigree. Specifically, it is the measure of how likely any particular gene is to be homozygous (two like copies of the gene) by descent.  In other words, the odds that a gene will have two copies coming down from the same ancestral source.  As an example, let’s say an Aussie is red and he has several red or red carrier ancestors but Old Granddad, a red carrier, is the only one who appears on both sides of his pedigree.  COI will give us the odds that our Aussie’s two red genes are both down from Old Granddad, not one or two of the other ancestors in the pedigree.  But COI applies to any gene and any common ancestor.  If a COI is 12.5% your dog has that probability that any of his thousands of genes could be two of the same from an ancestor who shows up on both sides and most purebred dogs have many common ancestors on both sides of the pedigree.

 Isn’t COI just a theory? 

Coefficient of inbreeding (COI) isn’t a theory, it’s a mathematical calculation.  It isn’t about “good” genes and “bad” genes, It’s a measure of how likely any given gene is to be identical by descent based on a particular pedigree. “Identical by descent” means that a dog has inherited two copies of the same gene version from the same ancestor.

Of the dog’s thousands of genes, breeders actively select for a relative few, mostly to do with behavior, structure and appearance.  That leaves a great many genes not subject to conscious selection.  Other genes that are located closely on the same chromosome as a gene subject to selection will tend to travel along with that gene.  This is probably why some unwanted traits, including diseases, may be very difficult to get away from – because the genes for them may be linked to genes for highly desirable traits.

During the formation of sperm and eggs, the dog’s DNA gets shuffled.  The chromosomes pair up, swap segments, uncouple, and split up into sets containing half the parent’s chromosomes.  A full set is reformed at fertilization.

The probability that any particular copy of a gene the dog has will be passed to offspring can be calculated mathematically.  A COI calculation does this over all the dogs in a set number of generations.  The use of the word “likely” in this instance means probability.  Simply put, the more ancestors appear on both sides of a pedigree the more probable – or likely – it is that a gene will be identical by descent.  Higher rates of homozygosity (pairs of like gene versions) are associated with health and reproductive issues.

Why is COI important?

High COIs are associated with inbreeding depression, decreased longevity, and increased risk of some diseases including cancer and autoimmune disease.   There is also indication that recent inbreeding, that which has taken place within the last dozen generations or so, is of more concern than historic inbreeding, or that which took place 20 or more generations ago.  The Aussie is a recently developed breed, so the bulk of inbreeding that has taken place is recent.

Can we run COIs just on what we have on a printed pedigree?

No.  Using only the three to five generations that appear on written pedigrees is not sufficient.   It is very possible to find two Aussies that have few or no common names in these pedigrees.  This might look like an outcross, but they may be quite related if you go back a few additional generations.

 What is the average COI for Aussies?

About 13.5%

 How can we get breed average COIs lower?

It may be possible to lower the average if there is some portion of the breed population that is not so related and is not heavily represented.  In Aussies one can draw from the working lines.   In other breeds one might look at country of origin dogs or dogs that are in other countries and relatively unrelated to what is found where you are.

Even so, every breed will have a certain level of background inbreeding due to whatever founder events or bottlenecks have occurred in its history.  There is nothing you can do about that, so you must determine what portion is background and realize that you cannot go below that or run your calculation so that it does not go back to the foundation/bottleneck to get better picture of what has taken place since.

Should I make low COIs my primary breeding goal?

No sensible breeder makes the COI a prime criterion for breeding.  It has to be balanced along with all the other important factors   Very low COIs should be considered a plus, but you still have to look at conformation, working or performance traits, temperament and health concerns.  For example, making a low-COI cross from two individuals whose respective families are riddled with hip dysplasia would obviously be unwise when there are families around where HD isn’t such a problem.  The best possible cross for a particular dog may sometimes increase COI.  The point is to avoid doing it on every breeding and not to do it generation after generation.

What’s the best way to use COIs when making breeding decisions?

Strive not to exceed the average of the two parent’s COIs and below breed average if possible.  If a dog has a COI that is above breed average (for Aussies around 13.5%) give priority to matings that would produce puppies  that are below average or at least 10% lower than the COI of the over average parent.  If both parents are over average, 10% lower than the parent with the lower COI.   (Example:  Parent’s COIs are 25% and 20%, litter not to exceed 18%).

What is the most generations you can use with COI?

You can use any number you want but every additional line of pedigree doubles the number of dogs to search (10 generations has 2048).  At some point you might bog down your computer’s processing speed.  In Aussies it is possible to follow a line of ancestors back 30 or more generations, but most lines will peter out long before that.  In any event, going that far back will either not make any significant difference in the result or may introduce some level of “background noise” if your breed suffered a bottleneck at some point (like during the World Wars) or you reach the founders in a breed which relatively few.

How common is zero COI?

Assuming you are running the calculation over 10 generations, extremely uncommon if not absent in an established breed, including the Aussie.  For a calculation over a low number of generations you can easily come up with zero; every one-generation pedigree (dog and parents) has zero COI because dad and mom can’t be the same dog!

This applies to established breeds.  If you are breeding Labradoodles and only doing it by mating a Lab to a Poodle, all the pups will be zero COI.  If you are trying to establish a new breed or have a recently created breed you may have dogs with zero COI depending on how the breed was established and how many dogs of differing breeds were used.

If you are looking at historic pedigrees you may find zero COIs because your pedigree data runs out shortly beyond them.  In a breed where country-of-origin stock that is unregistered or from unrecognized registries is introduced, puppies of those dogs will have zero COI either because nothing is known about their pedigrees (Basenjis) or their pedigrees may not contain any dogs that are in the recognized registry (desert bred Salukis.)

Won’t COIs start going down when some of the historic “bottlenecks” like the Flintridge dogs behind the show lines “fall off” the back end of our pedigrees?

Probably not since COIs continue to climb because most breeders emphasize line breeding and popular sires remain, well, popular.

Does low COI mean that risk for disease is low? 

No. The coefficient of inbreeding (COI) is only a measure of the level of inbreeding a particular dog or cross might have.  It is not, all by itself, a predictor for any specific trait.  However, the closer the breeding, and therefore the higher the COI, the more likely you are to produce any trait, good or bad, known to occur in that line.  That is the reason people linebreed, the hope of concentrating genes for the traits they want to perpetuate.

However, if a line has health problem or any other unwanted trait, tight breeding increases  the risk you will produce that disease.  On the other hand, you can do a breeding with 0% COI, say a Labrador to a German Shepherd Dog who both had hip dysplasia prominent in their pedigrees, you would increase risk for HD even though the COI was zero.

Do working line Aussies have better COIs than show lines?

Not necessarily.  The working lines vary.  Some have high COIs and other are rather low.  One difficulty the working lines face that the show lines do not are the overall population size.  There are far fewer stockdogs than showdogs so the risk of increased inbreeding in the stockdog population is higher because there are fewer places to go.

If you intend to confine your breeding within a population (i.e. stockdogs only), whether a COI is high or low has to be considered with the average for the population in mind.  Since we do not at present have an average COI calculation for stockdogs only, breeders would do best to assume it is the same as that for the breed as a whole.

Do high COIs cause autoimmune disease? 

No, but a dog with a high COI is more likely to have immune system deficiencies that can lead to autoimmune disease.  A number of the genes for the immune system are inherited in a block called the major histocompatibility complex (MHC) because they are all close together on the same chromosome.  As a result, there is much greater chance of a dog having two copies of inadequate defective of those genes the more inbred it is.  Research in dogs and many other species has shown that being heterozygous (having two different versions of these genes) results in a stronger immune system.

How long do Aussies live?  

Results of the 2009-10 ASGHI health survey indicated that the average lifespan for an Aussie is about 11 years but individual dogs have been reported to live as long as 18 years, though that is highly unusual.  Average lifespan can change over time.  In the late 1990s John Armstrong of the University of Ottawa conducted a canine longevity study which included Aussies.  The found that early dogs in the breed averaged 14 years while more recent (at the time of the study) dogs averaged only 12 years, largely due to an increase in cancer deaths in dogs that were not old.  The health survey indicates that the trend toward reduced lifespans is continuing.

How can I tell how long my Aussie will live?

There is no way to make a precise prediction for any individual dog, but if your dog is healthy and has been healthy all his life, he could easily live a longer than average lifespan.  The longevity of his deceased relatives who died of non-accidental causes would be another indicator.

Does inbreeding impact longevity?

Yes.  The canine longevity study conducted by the late Dr. John Armstrong of the University of Ottawa showed that dogs with higher levels of inbreeding, measured by a 10-generation coefficient of inbreeding, tended to live a couple years less than dogs of the same weight group which were not very inbred.

What else affects longevity?

Smaller/lighter dogs tend to live longer than bigger/heavier ones.  Life-style factors like diet and exercise can impact longevity in dogs just as they do in humans.  If you feed your dog a balanced, healthy diet and keep her lean and well-exercised she is more likely to live her full span.  Diseases and injury, even if not directly fatal, may sometimes impact the body sufficiently that lifespan is reduced.

With all the DNA tests we have now, why should anyone analyze pedigrees?

There are over two dozen health issues, dental faults, and disqualifying colors that need to be given due attention by breeders. The great majority of these do not have DNA tests.    In addition, not every dog will have had every DNA test.  Some that still appear in current pedigrees lived and died before some or all of these tests were available.

Pedigree analysis can take a lot of time and effort.  What’s the point?

If a breeder does not study pedigrees—with reference to both desirable and unwanted traits—all he is doing is producing litters and trusting to luck for good results.   This is not an effective way to breed healthy, high-quality dogs on a consistent basis.

 How far back should you look in a pedigree to determine whether you might be doubling up on a problem? 

When looking for specific health traits, five generations is probably sufficient.  Look for affected dogs and the parents or grandparents of affected dogs.  If there is nothing in five generations, odds are the risk of that particular problem is negligible.  Also run a coefficient of inbreeding (COI) over 10 generations to determine how closely related the two sides of the pedigree are.

Wouldn’t it be better to do both of the above for the same number of generations?

No.  5 generations are too few for an accurate COI.  Looking back ten generations for specific traits gets too cumbersome – there are 2048 ancestors in a ten-generation pedigree.  Some information on dogs beyond the 5th generation may be available, but if nothing has been reported in recent generations odds are it wasn’t passed down.

 What is the value of vertical pedigrees? 

Vertical pedigrees include the siblings of dogs that appear on a standard (horizontal) pedigree.  They allow the user to review breadth of pedigree as well as depth, the only way to deal effectively with traits with unknown or complex inheritance.  Lacking a DNA test, it can also be useful for single-gene traits particularly when you are unable to determine status on all progenitors in recent generations.

It can be harder to keep tabs on pet pups for doing a vertical pedigree, particularly if they were bred by someone else.  Given that, how useful is it, actually?     

Vertical pedigrees are extremely valuable for predicting what traits you are likely to get in a given litter or produce from a given dog.  However, tracking information on dogs that go into pet homes can present some difficulties. The average pet owner isn’t going to pay for eye exams or hip and elbow x-rays unless there’s something wrong that needs to be diagnosed.  It would be prohibitively expensive for the breeder to pay for it all and, no matter how hard you try to stay in touch, you will lose track of some of your pet people.

A breeder should do the best she can to keep tabs on puppy buyers.   Make a friendly check-in call at least once a year just to see how things are going.  Most people will appreciate the attention and concern.  If something is wrong they will let you know.  Try to see the dog again at least once after it is grown to evaluate conformation and behavioral traits.  Set up a chat list to keep in touch long-term.  Even though you will lose track of some pups, the data you gather on the others will give you a good idea of what is happening, both good and bad, with each cross you make and your breeding program as a whole.

You are even more likely to have difficulty in getting information on collateral relatives of the sire or other ancestors of the litter if you don’t own them.  If you are considering a dog for purchase or as a stud, try to see as many relatives as possible.  Conformation, teeth, the basics of temperament, and performance qualities can be evaluated at shows, trials, and other get-togethers.  Utilize ASHGI’s IDASH Open Health Database.  Search the OFA database and other on-line health registries to check status of relatives.  In the US and Canada reporting is voluntary, but many European clubs have some degree of mandatory health result reporting.  This information will be printed in club magazines or, in some cases, available in searchable on-line databases.  The English, Swedish and Finnish kennel clubs are among those who have health data on their websites.

Another thing you can do is talk to people; not only owners of dogs but other people active in the breed.  Some of what you hear will be drivel and you need to get a feel for what is and what isn’t reliable information.  Question sources.  If you hear exactly the same thing from two independent sources who are in a position to know and don’t have any obvious motivation to either promote or disparage a dog, the information is probably true.  Make note of whatever you learn in a pedigree database.  To simplify your searches when working with vertical pedigrees, note not only dogs that had a trait but whether they produced it or had offspring that did.   can be cumbersome because so many dogs can are involved, especially if there are one or more popular sires in the pedigree.  I use a system that can be incorporated into a standard pedigree.  If I know a dog had a trait, I mark its parents as having produced the trait and the grandparents as being suspect-producers.  If it has happened with different mates, or a particular cross produced multiple affected offspring, I note that as well.

There are so many things a breeder has to think about – dozens of different aspects of structure, behavior, health – how can I possibly make ALL of the improvements I want to in one cross? 

Breeding is very much an art of setting priorities combined with the ability to know when and when not to compromise.  You cannot get everything you want in one litter, never mind all of them.  To gain something here you may have to give up a little there.  The more important things need to be given higher priority.   If the risk for something unwanted is high, seek mates with as little family history of the unwanted trait as possible.  Realize this won’t guarantee you are safe, but it will be a step in the direction you want to go.

If I get pedigree analysis done on each of my dogs when I first get them, will I be able to eliminate health risk in my breeding program?

Until we have DNA tests for all the inherited diseases we’ll never be able to avoid health problems entirely and even then it will take generations to achieve.  However, doing pedigree analysis on your breeding stock, whether through a service or your own research, will be much more effective than just trusting to luck or a hunch.  We know much more about a dog’s health and that of its relatives by the end of its life than we knew when it was born.  Because of this, pedigrees of breeding animals should be periodically reevaluated.  AHHGI recommends every three years while the dog is being bred or semen is stored.

All ASHGI does with pedigree analysis is talk about bad stuff.  What about selecting for GOOD things in pedigree analysis? 

The method ASHGI uses for pedigree analysis, developed by C.A. Sharp, is more useful for unwanted traits.   There is a method handy for wanted traits called the “stick dog” method.  It and was developed by Dr. Carmen Battaglia that is useful for traits you want.  He recommends that you make up three-generation pedigrees depicting each dog as a stick figure – lines for the body, legs, neck, head, ears and tail.  The last item would be optional for most Aussies, depending on where you live.  Color code each part of each dog indicating excellent, good, fair, or poor.   If you don’t know something, leave it blank.  You can make notes, either on the pedigree or elsewhere, to explain why you used a particular color (“super topline,” “pasterns a little weak,” “straight stifles,” etc.)  You can tell at a glance where the conformational strengths and weaknesses in a pedigree are.  A similar method could be developed for performance and behavioral traits.

The advantage of these two approaches is that you can do them yourself if you are motivated to do your homework and, in the case of Sharp’s method, a little basic math.

How important to breeding is breed type?

Breed type is those traits that distinguish one breed from another.  However, the word “type” has come to mean many things which are not truly “type.”

Type is what makes you say when you see a dog, “It’s an Aussie.”  If the dog looks a lot like a Great Dane or a Chihuahua, it clearly is lacking in type.  For that matter, if it looks more like a Bernese Mountain Dog or a Sheltie than an Aussie, it would be lacking in type.  Among closely related breeds (Aussies, English Shepherds and Border Collies, for instance) there can be some overlap in type that could make the breed of some individuals difficult to discern if seen out of context, or with a tail in the case of Aussies.

Over time, “type” has come to mean other things–show type vs. working type, for example.  Some people take it another step and define type as the look of the dogs in a particular line or out of a specific kennel, i.e. “BigBucks Kennel’s dogs are the correct type and Average Joe Aussie’s aren’t.” (Probably because BigBucks wins more often than Average Joe.)  This is more properly a “style” than breed type.  Finally we come to the arguable assumption that winning a lot = correct type.

Whatever you are breeding for, you want your dogs to be clearly recognizable as Aussies.  Not only because they are tailless and merle but because to the average person they look like Aussies.  You want to keep the standard in mind and not stray from it.  Never dismiss a quality dog as a breeding prospect because the style isn’t exactly what you prefer.  And never equate a lot of wins in any venue as “proof” of quality in type.

 How do I go about developing signature type for my line-to-be.

You are referring to a style, not a type.  Style is attractive window-dressing.  Having a style that people can point to and identify your kennel is nice for the ego but has nothing to do with producing quality dogs.  The top breeders didn’t start by asking, “What shall I make my style?”

You need to have a clear vision of what your ideal dog is, not just in appearance but in every other aspect.  If you work consistently toward producing dogs that meet that vision, your style will develop along the way.

What are steps of relationship?

In the case of breeding dogs it refers to how closely related one dog is to another.  First step relatives are parents, offspring and full or half siblings.  Second step would be grandparents, grandpups, parents’ siblings and siblings’ offspring. Each succeeding step would be made up of the first-step kin of the prior step.

Beyond the first couple steps things can get murky if there has been inbreeding or linebreeding in the pedigree.

Knowing something about the virtues and faults of first and second-step kin can be helpful in breeding.  If wanted traits are consistent across those individuals, odds are good that your dog or the one you are interested in will produce those desirable traits.  On the other hand, the more individuals within a couple steps of relationship who themselves have connections to specific health concerns within a couple steps of themselves, the greater the risk that your dog or the one you are looking at might pass that health issue along.

Why is it that the bitches tend to be better quality than the dogs?

More than one dog person of long experience say that it’s to the benefit of a breed when the quality of its bitches is high.

Good breeders pay a lot of attention to the quality of their bitches because bitches are in-house while studs may belong to someone else.  The bitch carries and raises her litters.  She has to get along with your other dogs.  There are lots of things that are very important to kennel management that surround bitches, so their health, mothering ability, and behavior are just as important as their conformation or performance traits.  Bitches are the foundation of any breed or breeding program.  A breed needs quality dogs, but it needs even more quality bitches.

A bitch’s production capacity is limited, where a dog’s is not.  A stud owner can find out relatively early whether a young dog’s promise will play out in the litter box.  Every breed has hot young studs, usually with very successful early show careers, which don’t live up to their potential.  They disappear from the scene and even their best get may not receive much attention.  Through all of this the owner’s investment in the dud stud is minimal (the show career still goes to the breeder’s credit.)

Not so with a bitch; the breeder needs positive results from the first litter on because a bitch can only have a few.  The best way to do that is by using the best bitches, breeding for, and keeping the best bitches.

What is prepotency?

Prepotency is the tendency of a sire to consistently produce his qualities in his offspring.  For example, in the early 1970s breed type was still extremely variable.  A prominent sire of that time, Ch. Wildhagen’s Dutchman of Flintridge, CDX, PC (Dusty) was bread to a number of bitches of widely varied pedigrees – much more varied than you find today – and he consistently improved type and conformation in the get.  Dusty was a product of the breeding program of Dr. Weldon Heard and was homozygous dominant for numerous genes for desired traits which he passed to his offspring who also exhibited those traits.

Today, overall breed quality is much improved from where we were in the 1960s, but you can still see the stamp of a prepotent dog on his offspring, whether in the way they look and move or how well they perform.

What’s the best way to choose a mate for a dog?

Look for potential mates that complement your dog.  They shouldn’t share its faults and should have the traits you feel are most important.  A potential mate’s family should not be known for having faults you are trying to improve on.   Make sure you include health issues in your list of important traits.  Select the mates that are least related to your dog.  If you have a bitch, avoid over-used sires, you will have fewer options for where to go when it comes time to breed any puppies you keep.

Know your dog’s 10-generation coefficient of inbreeding (COI), that of the mate and of the potential litter.  The litter shouldn’t exceed an average of the two parents’ COIs.  Breed average is about 12.5%, the equivalent of a half-sibling mating, so if you can find a mate that has the qualities you want and will contribute to a lower-than-average COI, give it strong consideration.  Keeping COIs low helps in maintaining genetic health – risk of doubling up on disease genes is increased the higher the COI goes.

What if the dog hasn’t been bred before?

While you may have some ideas about a young dog’s potential, whether male or female, based on what you know of its line and near kin, it remains a cypher until there are offspring you can evaluate.  Therefore, it is best to take unproven breeding stock to proven mates – mature animals who have adult offspring that meet your criteria of quality.

You will know your young dog’s conformation, disposition, health, and something about its ability to achieve in your chosen areas of canine activity.   You won’t know its full potential for competition or work, nor will you know what health issues will arise later in life.   Because of this you want a mate that has a track record that meets your goals and exceeds them in areas where you know or think your dog may have shortcomings.  You can’t know this if you breed two young dogs together.

With an older dog that has been bred you have the opportunity to follow the progress of its littermates, any other full or half siblings, offspring, and possibly other descendants.  You will know (or should be able to find out) something not only about their health status but how they turned out in conformation, temperament, etc.  All that information will be useful in guiding you toward a suitable mate.

Finally, an experienced mate in the hands of an experienced breeder is less likely to contribute to a negative experience for our young dog which might unnecessary stress for your youngster as well as having potential for troubles with its breeding behavior down the line.

 If something goes wrong in a dog’s first litter, should it be bred again? 

That would depend very much on what went wrong.  With a very few exceptions (horrible diseases like epilepsy being an example), if the dog (male or female) is good quality, choose a different mate for the next litter from a family of dogs not known for whatever went wrong in the first litter.  That means the second one also can’t be kin to either your dog or the first one it was bred to.   If the next litter has the same problems then it may be time to throw in the towel.

If you have a dog that has a conformation fault – too much of something – and you breed him to a bitch that has too little, will it even out in the pups? 

Unfortunately, no.  Many aspects of conformation involve skeletal structure, which are not single-gene traits and inheritance is not simple or easy to predict.   Any time a dog is off the ideal for a conformation trait, but not so far off that it shouldn’t be bred at all, the best plan is to select mates that are near perfect in that respect and come from families where correctness of that trait is the norm.

Multiple genes, regulatory DNA, and sometimes environmental factors combine to produce a particular phenotype.  Breeding results can fall anywhere within a given range:  Long (or tall) to short, heavy to light, big to small, etc. relative to the breed norm.  Most individuals in a given breed cluster around the norm.  The breed as a whole will form a “bell curve” distribution because there are relatively few individuals at either extreme.  But if you breed dogs from the extremes to each other the pups may fall anywhere in the middle and are unlikely to be consistent for the trait.

Couldn’t we use frozen semen to get back to the type of Aussie we had before big hair and big bone became the rage?  And couldn’t we also get back to a place where health issues weren’t as bad as they are now?

Frozen semen is a mixed blessing.  It has great potential for positive benefit, but the dogs whose semen will most frequently be used or stored are those who have already been bred extensively.

Long-term semen storage could be a method of bringing the breed back to an older type or bloodlines that have been neglected.  However, the use of frozen semen can also extend the use of already over-used sires.   Wide use of popular sires reduces genetic diversity within the breed and has the potential to spread widely whatever undesirable genes those dogs have, with the possibility of making a formerly infrequent genetic issue common.

When it comes to using semen from long-ago dogs, the technology is relatively recent and I doubt there are any great number of Aussie studs who died before 1990 from whom semen is still available.  In any event, I’m not sure how well this would work in conformation breeding because the dogs of 20-30 years ago wouldn’t necessarily be considered suitable in the present time and therefore there wouldn’t be much interest.  In the case of performance animals, however, being able to go back to a top performer of decades past could be a benefit to a breeding program.

From a health standpoint, it wasn’t that the diseases we are dealing with today weren’t present, they just weren’t as frequent as some of them are now.  You might be able to bring in blood that didn’t have something very common, like cataracts or epilepsy, but every dog carries genes for something and overemphasis on the “new
blood from long-ago dogs could lead to a new frequent problem.

Before storing frozen semen or using it consider the big picture rather than a single breeding.  Look at the potential for both good and ill from every angle, rather than just health or appearance.

Is there any legitimate reason why someone would crossbreed?

Yes.  It’s done all the time in commercial livestock.

In purebred dogs it is generally shunned—not that it doesn’t happen behind the barn.  For example, the color merle has appeared in several breeds that had no previous history of merle.  Odds are that some creative individual either introduced the gene and then fixed breed type by backcrossing or here was an “oops” that someone decided to capitalize on.  This isn’t a gene that mutates readily.

Legitimate reasons for crossbreeding include:

  1. Increasing the diversity of a rare breed’s gene pool.
  2. Reducing the frequency of genes for a serious or lethal disease in a breed population where many individuals carry those genes.
  3. To produce dogs more suitable for a particular type of work or performance event.
  4. Developing an entirely new breed.

Many breeds have a history of crossbreeding in their past; the Doberman was created in this manner.  More recently crosses have been done to address important issues (Dalmatian and Pointer to get away from bladder stones, Boxer and Corgi to introduce a bobbed tail).  There is nothing wrong with crossbreeding if it is open and for a well-defined purpose.  (Yes, Labradoodles have a purpose.)  Crossbreeding has been used, legitimately, to produce dogs better for a purpose, like guide dogs or agility.

However, fraudulent crossbreeding within an established breed through use of falsified documentation is ethically wrong and against the rules of every legitimate registry.