by Tina Barber
June 1999

For centuries breeders had to rely on the minimal amount of data available through pedigrees and their personal knowledge of the dogs themselves. During the past four decades I have worked with many dedicated breeders that were successful in creating their own type through inbreeding. By examining their "formula’ and observing the results it produced, I was able to learn more then anyone could possibly "get" from just reading a book on genetics. In the late 70’s I wrote an article for the GSQ titled "PRACTICAL GENETICS." In those few pages I tried to share some of the basic knowledge I had accumulated up to that point. In order to prove these theories correct-or not, I set out on a major breeding campaign producing dozens of litters per year, and documenting as much of the littermate information as possible, in order to "find" the recessives that were running in my genepools. The information I gathered was astronomical! I managed to accumulate boxes full of notes about various combinations, many of which are not even in the present Shiloh Registry because they were unsuitable for entry. I have a large filling cabinet full of pedigrees & data on "non-Shiloh-GSDs" that were prevented from entry into the "separated-breed" genepool for various reasons. Now I am thrilled that I kept all of those records because they are immensely helpful in my RC research!

Some of the dogs in those files are very closely related to our foundation stock, and many of them were developed by inbreeding on GSDs that now show up with a high % on the RC reports! In past decades I spent a lot of time holding seminars in many states in order to help breeders better understand how to apply some of the basic principles I had discovered through my own experimentation with the basic 4;

Inbreeding - helps you to establish a particular "line-type" in the shortest amount of time. In the first few generations you will increase size, and "pull" specific qualities (good as well as bad) from the ancestors in the line you are using. If you are not getting the desired results, try a different line to work with!

Linebreeding - Is just a bit safer to use, and you will probably be able to "pull out" more of the specific qualities you are looking for, without taking as great a risk.

Type breeding - Can be utilized in conjunction with distant linebreeding, by mating two dogs that "look" the same, but do not have the same ancestors for several generations. This works much better then a total outcross, yet accomplishes the same desired results.

Out-crossing - In most cases this would consist of a breeding to a totally unrelated dog, that will flood your genepool with an entirely fresh group of genetic variations that must be discovered & documented!

During my seminars I used my "3-leg" pedigrees to explain how CHD can hide for several generations, and then "pop" up when least expected even out of OFA parents! Furthermore, I always emphasized the importance of the "Shadow Dog." When you plan a breeding, you take great pains to look at the dogs you are using. You evaluate both the sire & dam impeccably! Yet most people never bother to investigate their "Shadows" and that is what your puppies will turn out to look like! Example;

The stud you chose has very strong ears, yet he was only one of 10 puppies, and seven of them had soft ears! Without this knowledge, you may choose to breed him to your female - whose dam had four soft eared littermates, and her sire had two littermates with soft ears. Her "shadow" (the dog you do not see) has soft ears, even though hers appear to be sound. After this litter was produced and a large percentage of their puppies had soft ears, you would be shocked. Yet, if you had taken the time to research the "shadow" of both the sire & dam that you bred, you would have found that probability easily! Other recessives can be a lot harder to determine! Since each dog has 78 chromosomes, and 39 are passed from each parent, the probable combinations can be very difficult to calculate! It is for this reason that most long-time breeders are so careful about choosing an outcross!

The new RC reports help us in locating those "shadow" dogs, and even give us an idea of how strongly they can effect the progeny we produce! There is a lot more to breeding then just putting two dogs together, and producing a litter of pups! Sometimes nature does an even better job of selection then humans! I am not referring to two mutts running loose, such foolishness is caused by "humans’ NOT nature! When you have the time, you may want to do some intense research into POPULATION GENETICS.

In order to attain the desired results, old-time breeders would begin by doing an intense 7 - 9 generation pedigree study on both the sire & dam. First they would look for the heaviest in/line breeding of the same dogs on both sides. This would help them in starting to determine the probable "shadow dogs" that would be lurking in the genepool. Then they would investigate progeny produced by the dogs that showed up several times in the pedigree. Example;

Let’s say in the first 7 generations "cloud" shows up four times. They would not only look at his progeny, but concentrate on progeny he produced with other dogs from the "lady" line, etc. Then they would "back-up" even further, since "cloud" is a ‘Sword" g-grandson, they would look for other "sword" lines in the pedigree! If several showed up, they would weight his probable contribution to the planned breeding. If he produced similar faults to those produced by other dogs in the pedigree, they would realize that this fact would increase the probability of those same problems showing up in their litter, and may choose to evaluate the benefits of using another stud. If a particular virtue was being sought for, then the risk would be worth it, and the progeny produced would be carefully evaluated so that any specific puppies (with that quality) could be retained. Such puppies would then be "type’ bred to a dog with a compatible pedigree that also carries the sought after virtues, without the same faults! Most breeders drew up a plan that would carry them for many generations, until they achieved the goals they had set! Then they would just set new goals for the next few generations, etc. etc. :-)

All this involved a lot of dedication, since the research needed was achieved through a very painstaking process! Those without the foresight and determination to see such a program through to fruition, usually failed to develop anything worthwhile! Such people are often referred to as "backyard" breeders, because their only interest is in producing some puppies to sell, instead of dedicating themselves to an ultimate goal!

True breeders may attain many glorious victories, but along the way, they will also feel the agony of defeat! Those that learn from their mistakes, and forge on nevertheless, pave the way for their chosen breeds future. Every litter born will display some new "trait" or quality that was passed down from one of their ancestors! Just as in the human world, where no two individuals are exactly the same, so it is with all of nature! God tells us that even the snowflakes are all different! If you truly look at all the puppies in a litter, you will notice that each one is a "bit" different from the others. I have repeated the same combinations many times, producing several dozen offspring with the same parents, and never found two puppies to be identical in all respects! Although such repeat breedings are very important toward proper data accumulation. The more progeny you can evaluate from the same parents (and from breeding the littermates of those parents) the faster you can determine the strong (prepotent) traits that obviously are running in that particular line. I used to call this method ‘stacking" my deck. If you can remove most of the number cards (and replace them with 10’s, queens, kings, jacks and ace’s) you stand a much better chance at winning the game! This will not guarantee that you will win every hand, but the odds in your favor are much better! The same principle can be applied to your breeding program! There is no such thing as a "perfect" dog! Each breeder must work toward eliminating as many faults as possible, while trying to "concentrate" on the virtues that can be "pulled" in. This can only be accomplished through intense pedigree research, and in depth knowledge of the individual dogs (and their littermates if possible) that will affect your progeny!

The RC program provides us with a helpful tool because it actually calculates the Relationship Coefficient on our planned breeding, for up to 30 generations! Furthermore it even provides us with the Homozygous possible % effect from various individuals. Since our entire breed is based on four very inbred lines, it is important to "draw" a lot more information from the foundation stock, then it would be with any other breed! Without a full understanding of the dogs used in the creation of those lines, and the progeny they produced through heavy inbreeding, it is nearly impossible to avoid the inevitable pitfalls just waiting to pop-up in future litters!

Over the past 8 months I have spent countless hours, and days, researching our genepools. Even though I personally knew all of the dogs used in the "development" of the Shiloh, and can easily write pedigrees on nearly any registered dog for up to nine generations, I am constantly amazed at some of the discoveries I keep running into with nearly each report! Since most pedigree studies are done on 5 - 7 generations, (not a good thing when you are using heavily inbred lines) even the most dedicated breeders may not see the names of dogs that they will be inbreeding on! For Example;

Take out a pedigree on a current 2 year old, you will probably see the name of Super Sweet Sabrina Selah somewhere in the 4th or 5th generation, maybe even twice, but this fact would not alarm most breeders ! After all she is the mother of great dogs like GV Ch. Shiloh’s Captain Caliber Baker, and Ch. Morris Good Morning Mona, the dam of Shiloh’s Wolfin Sasquach, etc. The fact that her name shows up several times in your pedigree only proves that you will be producing large puppies, right? Wrong! One of the things you will find from researching her pedigree is that she is very heavily inbred on Mia Martini Rall and JD’s Dark Victory Czar. How much information do you have about their "shadow" dog? You probably didn’t even know that these dogs were even IN your pedigree, besides - if they are that far back, what difference does it make anyway? A lot!

Unlike a pedigree that can only go "so far" listing your dogs ancestors, this computerized tool is capable of actually calculating genetic information for up to 30 generations! I have tried doing 10 generation reports, and then running a 20 & 30 for comparison. The further back you go the more things will changes, because as you are "picking up" dogs from those distant generations, they are being added to the calculations. Such an endeavor would be humanly impossible to duplicate, much less so quickly! Does this mean that every breeder should run out and purchase this program? No, because you would not have sufficient data to enter to begin with, and even if you did (break into the TCCP and steal all of the ISSR records), you would not know enough about the dogs that you are inbreeding on anyway. I knew most of these dogs personally, many of which I whelped, raised & trained. Even more so, I was able to collect data on them through their progeny! This is one reason that the Homecoming is so important! It gives breeders a chance not only to see the dogs they produced, but also progeny from other litters as well! Books & pictures can never surpass personal experience! The more time you devote to educating yourself about this breed (the dogs from the past) and by meeting real dogs in person as often (and as many) as possible, the more successful your breeding program will become! There are no "short cuts" provided for anyone, time has always been the best teacher! Just because we are now capable of utilizing a computerized program to help us understand our genetic probabilities better, does not mean that we have found some "magic potion" that will map out all of our breedings for us. It is still only a tool, one very different from those used in the past, but nothing more! It should never supersede all other methods, no more then it should be ignored due to a difficulty in understanding it.

When you look at your first RC report, it may totally confuse you! Let’s start by looking at the RC report on the Sabrina/Shane litter. This is an old 10 generation inbred-outcross. To start with you may want to note that the RC factor is only 0.44772% the lowest you will ever see in the "Shiloh" world. Since practically every Shiloh carries these two dogs in their pedigree, I felt it would be interesting for most breeders to look at this one first. Please note, on the left side of the page you will see a list of names, unlike a pedigree that follows the parents, g-parents, etc. These dogs are listed as per possible effect of genetic % of contribution to the progeny of that specific combination. The page continues to list columns of data. The last 2 columns show the number of times that individual dog appears in the pedigree, and his/her maximum genetic % of influence. This does not mean that the parents each contribute exactly 50% to each puppy! If that were the case then why even bother looking at any of the other numbers, or even at grandparents in a pedigree! It does mean that they cannot contribute more then 50 % of their (genetic makeup) into any one puppy, some of which most certainly enter in the form of hidden recessives. Just think of the gene game as being similar to the Lotto. When you look at those balls, and think about the fact that there can’t be that many of them, it may lull you into thinking that you have a chance of winning. Millions of people buy tickets regularly, but their odds of ever winning are still very slim! The only thing this report can help you with is to tell you how many of which number are "in the pot." It still doesn’t guarantee that you will win, but it sure helps you beat some of the odds ;-) By knowing exactly how much influence a particular dog (or combination of dogs) can have on your litter, it will help you make better breeding decisions, if you know what that dog and his "shadow" are like! The RC report can only tell you his/her % of influence, the rest is up to you! If you look at most normal reports, the numbers in the far right column will decrease by 50% at each generation. Parents - 50%, Grandparents - 25%, great-grandparents - 12.50% etc. When a report is run on any inbred dog, these numbers will change drastically! For example, in this report things start out normal, but then you have 5 dogs listed that have a higher % then even the g-grandparents! The heavier the inbreeding, the more these numbers will fluctuate! In this short example please note that Mia & Czar (although way back in the pedigree) have nearly as much of a % of influence as the g-grandparents!

The second report (Indy/Rambo) that I would like you to look at is a good example of the type of breedings done (during the early 80’s) to establish the foundation lines. This particular one was done to draw in more Ursa influence. Please note that she shows up 7 times, and provides a higher (31.250%) then the grandparents! Even Lance, who appears 41 times in this pedigree has a higher influence then the grandparents! Now it is time to look in the Homozygous column! Since the first report we looked at was that of an outcross (even though both parents came from inbred lines) that column reflects only 0’s! In this report both parents came from the same inbred lines, thereby creating a "Homo" (same) % of affectability!

The third report (Goldie/King) is that of a litter of pups recently born at Zion kennels, and typical of the type of combinations used now. In this report you may note that the grandparents have more of a consistent influence, then they did in the past, but the desired influence from the dogs in the past is still being drawn! Note that in this pedigree Lance shows up 356 times with a 15.060 % of possible genetic contribution. Furthermore Gunsmoke carries a higher % (18.469) on the progeny of this breeding then he did on his g-grandkids 2 decades ago! Also, you may want to note that the homozygous effect, although not as strong, is still quite evident in this breeding. Finally, the total RC of 19.94763% is still within the safety limits!

The final report that you should look at consists of a breeding that will not happen, but could have! It is also a good example of what could happen in the event such a heavy Inbreeding did take place. To start with the RC factor on this "proposed" litter is an outrageous 66.68329%! The HOMOZYGOUS percentages are astronomical! There can be no doubt at all that some of those genes must carry some very "bad" copies! For a more in-depth research into this subject please go to GENETICS, Multiple Gene Traits, and also be sure to read all of the wonderful articles written by John Armstrong!

As I looked at 47 randomly selected RC reports on litters produced in the late ‘90’s, only 3 did not have Gunsmoke listed on the first page! Most showed that he was contributing a possible 16%, and many were much higher! Some showed a homozygous effect of over 2%! Lance of Fran-Jo also shows very high percentages, as do so many of the dogs used over 20 years ago! Each one is positioned a little differently, but the same names come up time and again! What does this all mean? Extreme caution must be taken with every combination planned! Continued (irresponsible) inbreeding on these dogs will lead to unimaginable horrors!