What exactly is line breeding, I have been asked this question today.  Am I right in thinking it is when a dog has been used more than once in the pedigree of a dog i.e it could be a pups grandad and gt, gt, grandad??

It's rather more complicated than that - this article will help explain it.

Hi thanks for that I am afraid I am still none the wiser is line breeding the same as inbreeding?? it was a little too technical for me perhaps you could simplify?? please

Linebreeding is a less severe form of inbreeding. Inbreeding is normally explained as father to daughter, mother to son, brother to sister, but line breeding could for instance be cousin to cousin or half brother to half sister.

Is this common practice then to do line breeding

> Is this common practice then to do line breeding

How long is a piece of string?

Depends on the breeder and what they are trying to achieve.

It can be easier to think of it as inbreeding being what wouldn't be allowed with humans, and linebreeding being what would, within the same family. For example, it's legal for cousins to marry, but not siblings.

I'd say only top breeder's who have been in the breed for years and know what they want to achieve do this as they know their own lines. Is it common place within my breed you will see some kennels do it more than others, some of the boys up for stud have the one dog that pops up about 4/5 times over the two sides in one ped (a bit too much for my liking).

Line breeding should never be done unless you know what you are doing or have a mentor that knows.  

>Line breeding should never be done unless you know what you are doing or have a mentor that knows.

To be honest I'd say that outcrossing should never be done unless you know what you're doing or have a mentor that knows. There are far more things that can go wrong in that scenario; line breeding is the fail-safe of the three options.

Thanks guys think I understand it a bit more and it wouldnt be something that I would be looking to do, I think viewing it as you would with humans and cousins being able to marry is a much easier way to understand it...

>it wouldnt be something that I would be looking to do

Remember that using totally unrelated animals there is so much more research to be done before you do the mating; each individual only has to be researched once - the more there are, the longer and more difficult the task.

Agree with Triona

My mentor of one of my breeds, likes to put two dogs together that have the same champion in each dogs pedigree.  I know of some people that have the same champion both sides, several times.  So how close inbreeding goes is another factor.  Do you think it's ethical?

(Another can of worms!)

I have used half uncle to niece matings most successfully and my next two planned litters are the same.

Now in my own family my brother is living with our half sisters daughter (same thing half uncle and niece) and this is a relationship not allowed to marry, even though the relationship is actually less close than first cousins, because of the half blood relationship.

Obviously their choice caused a lot of upset in the family (they are the same age).

They have had a baby and went for genetic counselling and the doctors predicted no problems due to the relationship.

As for animal breeding there are not ethical problems based purely on relationship, the ethics come into what your breeding.

I'd sooner have a close mating on well researched known healthy and good temperament than out cross on total unknowns, or with animals that carried the same faults, even though unrelated.

Hi everyone,

I'd sooner have a close mating on well researched known healthy and good temperament than out cross on total unknowns, or with animals that carried the same faults, even though unrelated

Using your comments Barbara but not directed at you as hopefully you will agree with my additional comments.

People do describe above often BUT we have to always remember , well for me these described are the two opposite ends.

Today more than ever I believe there is a demand for close co efficent monitoring and healthy type breeding not necessary line breeding.   Not claiming to be an expert at it but I am a novice genetic enthusiast .. lol :-)  

Type breeding along has proven to be as successful as line breeding I believe. A few sucessful kennels in my breed rarely closely breed of course there are some who do line breed successfully also although I think you can only line breed so long I believe before it bites you in the bottom.

Some people like to line breed and are happy to take the risks with what the double up on good or bad, I know of a couple instances were the only problem litters have been line breeding but I have also heard of ones where they think the outcross was the problem.

I am not saying I am against it at all, well not completely but I personally would like to see far more great co-efficent control and far great control over stud dog use as we see the matador traits in many of our breeds..

I personally wont venture above 12.5% now .. I think below this can easily achieve what most need and keep the gene pool wide enough.    

It is quite scary when you run the co-efficents on some dogs even in he numerically high breeds.

One thing to think about is the reason why humans not allowed to closely mate !! 

I think we in my breed tend to avoid going too close as we always have a limited gene pool anyway so if you go close it means you have nowhere to go next time.

We seem to need to import every other generation, so do go on phenotype breeding.

I do believe that if you bring in out cross blood you do need to check what you have got by doing a closer breeding on the import blood, like half siblings or uncle niece, it will show up any problems brought in with the out cross (hopefully not health ones as research and testing would limit this likelihood).

I think generally that most people will line-breed after an out cross so as not to loose the type they want to keep.

Even in the country of origin of our breed I think it is hard to find any dog that doesn't share a common ancestor in a five- seven generation pedigree.

I wish the KC would give us a test mating facility to get the coefficients, though they would be of limited use as we import so often.

Half siblings are as close as I would go, but most often the closest I have gone is half sibling of a bitches parent.  do you have an easy way of calculating the percentages without having a program to do it?

I think the stud I used in Finland had under 3% inbreeding index, but my bitch had ancestors in common with him about 8 generations back.

My half sibling mating had two imported grand sires one from US and one Norway with no relations in common to each other for at least 10 generations.  So even that may be less inbred that some seemingly less line-bred animals with  a lot of inbreeding in the background.

My mentor of one of my breeds, likes to put two dogs together that have the same champion in each dogs pedigree

What it has to have a champion!!!.  For fear of criticising someone will hold me tongue .... hopefully it is not actually meant literally.

Hopefully most good knowledgeable people with an cup of common sense would use the best dogs regardless of their show title status.

I just used a dog that I think is stunning. Owner can't show him just now due to personal circumstances but was shown a couple times a sa puppy, when I saw him as a puppy and he caught my eye from the end of the hall.  I went to have a look at him at 18 months and thought he was a nice. 

My 2 previous litters were by dogs without titles but now have their titles.

> I just used a dog that I think is stunning

I'm a 'born again' phenotypical dog breeder.
After many years of line breeding in one breed ( boxers), a breed which is now very tight and close on one Australian import, I moved towards phenotype matings...ie picked the dog I though looked the best :-)

In Cavaliers it is impossible not to breed fairly close - so much effort is being given to finding outcrosses, as far as possible and to that effect I'm attempting to get my COI to below 5 or ideally 3%.

I'm also thinking about letting friends abroad have pups to mate to outcrosses ( as such) in their countries, thinking long term plans.

In Cavaliers it is impossible not to breed fairly close - so much effort is being given to finding outcrosses, as far as possible and to that effect I'm attempting to get my COI to below 5 or ideally 3%.

I am sure we are related in the 3rd of 4th generation :-D :-D     The Sweds now aim for a max of above 6.25%  that is really my aim to stay below this it makes sense I think in the long run... 

People jump up and down and say , "but when you outcross you get mixed types in litters"    BUT if your type breeding you will always pick the type that you are drawn to..hence generations of type without the inbreeding.

The highest I have in my house is I think  9% but she was bought in.     

Forgetting things like Hips elbows all things you can readily test for now, recessive/polygenetic genes are responsible for the couple occuring things in my breed, so when people go out and get an affected they are often quick to point the finger BUT it had to be there in the bitch or dogs line in the first place.

I find it really hard to believe in breeds of huge numbers just how small the gene pool actually is.   It has a double whammy as in the future people have to do the opposite of what they are trying to do now, IE they have to go out to lesser quality of dogs to get an outcross because the dogs they know are already all too close.. or too sensibly close.

I know it is a subject that can be close to peoples hearts but the more knowledge we are armed with the more it is showing us just what "we" are doing in some breeds..

I'm also thinking about letting friends abroad have pups to mate to outcrosses ( as such) in their countries, thinking long term plans. Ditto, I have done this with two dog, to two good dog friends in the hope it will help in the future.  Even simple things like Size is something I think can be lost without blinking....

I should have said there that I won't venture about 12.5% in my stud dog selection. In my own house it is a lot less.

I have just a pedigree program and I do trials on it when considering matings. In fact I do them for half my freinds.  A freind of mine in the US has what I believe our biggest pedigree database , I bought the program and downloaded the contents of her database :-)    I do loads of trials 2 and 3 generations down when I am bored..

Does someone in your breed keep the pedigree database?? most programs do have the Co efficent calculation facility..

If would be nice if the KC would give people access though..A read only but enough so it lets you do trials .

This is a suggestion for someone to put to them :-)

> Does someone in your breed keep the pedigree database?? most programs do have the Co efficent calculation facility..
>
> If would be nice if the KC would give people access though..A read only but enough so it lets you do trials .
>
> This is a suggestion for someone to put to them

I know one or two who have bought the entire CKCS database and a
have entered every Cav in the BRS from the year dot to complete it. Funnily enough there is a couple of cavalier forum threads about the pre 1960-70 discrepancies where all Kennel Club registrations were hand typed and logged. I've still got my old hand typed ones ( active register ones)
The database costs though and I feel it should be open to all.

Problem is as I use overseas lines and expect to do so again it is not particularly useful to only have the KC database for our breed.

What I really want is a simple tool so I can enter any of my pedigrees and calculate it on a case by case basis.

I don't want to get involved with a whole database program.

> some of the boys up for stud have the one dog that pops up about 4/5 times over the two sides in one ped (a bit too much for my liking).
>

Now if I saw a pedigree like that and it was the same prolific stud dog who sired lots of Ch./Sh.Ch. progeny, I would relish that pedigree.   All my LM's go back to the very influential stud dog - Ch. Axel von Esterfeld of Raycris and if you look at my current 2 in the LM database he is the most significant dog in their pedigree.

My sister's ESS had Sh.Ch. Hawkhill Connaught (A VERY influential stud dog) as his gt.grandsire, gt.gt. grandsire & gt.gt.gt. grandsire on the paternal side of his pedigree alas the dam's side was all working bred so outdid any chance of a show type ESS.

Barbara I swear the pedigree program is easier to use than champdogs. Lol.    You need a database to work it out as it is calculated generally over 10 generations. 

We add the dogs from all over.  I send my friend the BRS she tops it up.   A trial takes 10 seconds.    I would ask some of your breed collegirs.  It is very to add dogs to. 

> What it has to have a champion!!!.  For fear of criticising someone will hold me tongue .... hopefully it is not actually meant literally.
>
>

Yes it was.

> Now if I saw a pedigree like that and it was the same prolific stud dog who sired lots of Ch./Sh.Ch. progeny, I would relish that pedigree

I believe that is the sort of thing this particular judge does.

Was their any bitches that had a really good impact do you think Christine out of curiousity?

Some bitches produce good of spring wherever they are mated I think and I often think some stud dogs are fortunate enough to have wonderful bitches brought to them.   In fact I would probably say that I now believe bitches have the greatest input in their offspring having mated different bitches to the same dogs with such different results.

It is funny how males take so much credit, mind you the poor sods get the blame for everything too :-)  :-)

Such a shame really ..

Like one or two earlier posters I think breeding to phenotype is generally safer - not saying that line breeding should never be done but it shouldn't be the rule and I'd steer well clear of anyone practicing inbreeding.  It isn't just about specific genetic diseases but modern science suggests that inbreeding as a whole (including line breeding) tends to reduce vigour - stuff like immune system, reproduction and life expectancy in general.   I can see line breeding becoming less and less popular in future which imo would be a good thing.

Peoples might find this article interesting/useful (by Dr Bruce Cattanach-geneticist and Boxer breeder)
http://www.steynmere.com/ARTICLES8.html

Plus this one written by him re inbreeding and published in Dog World Feb 2009, it does reference back to the link above.
http://www.steynmere.com/INBREEDING.html

> Was their any bitches that had a really good impact do you think Christine out of curiousity?
>

It depends on what you determine as impact?   Consistency in type or the amount of titleholders they produced?  If you go with the latter then maybe these bitches I mention are noteworthy.

I'm going from memory here as I can't access the LM database with ease.   Sh.Ch. Datroy Ovation of Raene had 4 or 5 litters.   From one litter to the dog I mentioned she had a Champion son & daughter.   Another litter to a son of 'Axel', she had a litter of 5 of which 4 gained their Sh.Ch. titles in this country.   And to a different 'Axel' son, she had a Sh.Ch. daughter.      She was top DW brood bitch one year or 2nd, again I can't remember.

Sh.Ch. Ichbin Heavensent of Yeldoaks had just one litter of 7 and from that 3 males became Sh.Ch.'s also one daughter won her title and another daughter won a CC.

Ch. Raycris the Kqrac Dreamer had 2 litters and produced a Champion son, 2 Sh.Ch. daughters & a CC winning daughter.

The imported bitch, Konny von der Langen Weide had 4 or 5 litters (Again this is from memory), she has from one litter a Champion son (Top CC winning LM), a Ch. daughter & a Sh.Ch. daughter.    I think another litter produced a Sh.Ch. daughter but again my memory's not as good as it was.

For me, breeding to a phenotype as oppose to line breeding is the way forward, provided you have researched your lines well - in every aspect.
I hope I have worded that correctly because in my ignorance, this is the first time I have come across the word "phenotype" lol. I do hope I am correct in assuming it means breeding to other dogs that are of the same "type" as your own but without necessarily sharing any common ancestory ?

Phenotype is just a genetic term meaning how the genes ( genetic makeup) express themselves...in it's simplest terms- how the animal looks... and by looks it can encompass things we cant see...eg MRI scan show us the inside of the organism.

Definition given as:The observable physical or biochemical characteristics of an organism, as determined by both genetic makeup and environmental influences.
The expression of a specific trait, such as stature or blood type, based on genetic and environmental influences.

Breeding for type is fine as long as you have other breeders breeding to the same type and following a similar format!

In my breed there are maybe half a dozen breeders breeding to their type and are very consistant but it's not the type that I would want.  The vast majority use the latest winning dog, which often means that the ancestors are just a complete hotch potch.  A couple of reasonably large, successful kennels line breed within themselves but often include REAL pet quality dogs and bitches which mean that in each litter I stand a chance of producing pups that I really wouldn't be proud of! :(

So I tend to line breed for a couple of generations and then out cross to a similar type line - twice in and once out as the old breeders used to say.  But it's becoming increasingly difficult to find stud dogs who are bred to type these days as most are just an odd nice pup from a mixed quality litter and that's not what I look for.  I want a whole litter of quality pups so that my pet homes can also have quality, typical dogs, not just one for me!

"[  But it's becoming increasingly difficult to find stud dogs who are bred to type these days as most are just an odd nice pup from a mixed quality litter and that's not what I look for.  I want a whole litter of quality pups so that my pet homes can also have quality, typical dogs, not just one for me!"/b]

Indeed. This is why I said you have to know your lines, since using a stud that is the same type as your own is no good unless he is typical of his breeding (if you know what I mean). A randomly nice dog will just help throw a mixed bag.

Edited to say, sorry about the bold print - it was just supposed to be the quote in bold.

Thanks toolz, that is what I thought it meant, but you have put it better than I, of course !

I think I would agree with the above comment. As a relative newcomer to the breeding game, I have a bitch who has been quite tightly line-bred (it wasn't until after I had her and researched her pedigree that I found out her father and grandmother on her dams side were littermates). This meant that the three of the four lines of the pedigree were closed down.

I've been doing extensive research into a possible mate for her, and have followed several lines of enquiry (gone back 3 - 4 generations, looked at siblings/progeny - extended the line laterally (does that make sense?)). I've been having sleepless nights thinking about this. Luckily for my breed there is a central database which I can do test matings, research lines right back etc, but each time I think I'm onto something and do further research I either find out something I didn't want to know (or glad I found out before it was too late) or come to a dead end.

I have therefore decided to start the search for a 'phenotype' stud. My breed is numerically huge, but going back only 20 years or so you'd be amazed at the few numbers of prolific studs which appear time and time again. I have read that it is possible to change 'type' by putting your bitch to a dominant sire (ironically, 'dominance' in terms of genes only seems to occur in line-bred dogs).

My search is ongoing, but with advice from others in my breed, I think I'm finally getting there:).

>> I have therefore decided to start the search for a 'phenotype' stud.

I don't know if you've read the second link given by Boxacrazy but it warrants reading well.

Bruce ( Cattanach) did extensive research into all the CC winning boxers over a large period of time and deduced that, so-called 'linebred' dogs won no more CCs than outcrossed ( per say -as some breeds it is almost impossible to do) dogs and I believe Simon Parsons found the same when he tried.

I have a bitch who has been quite tightly line-bred (it wasn't until after I had her and researched her pedigree that I found out her father and grandmother on her dams side were littermates). This meant that the three of the four lines of the pedigree were closed down.

This is where studying comes in because although that is quite close you have to look at the co efficient % on the pedigree really to work out how close it is.

You can mate two dogs with high % then the off spring becomes quite low % because the two dogs are unrelated or you can mate two with low % but because they are similarly bred you end up with higher %..

The coefficient % on many of the dogs I'm researching seems to be around 20%. I've seen so many pedigrees which have the same dogs going back 5 generations.

I will read the research document again - it really is a difficult thing to grasp, but I feel more confident I know what I'm looking for all the time! One thing's for sure, putting the research time in does pay in the long run!!

im a little confuzzled by all of this too, is it ok to mate if they share a dad but have different mums? both have excellent temprements no history of genetic defects etc. will the kc allow it?

Yes they will allow it but too little information to say if this is an advisable mating.

> The Sweds now aim for a max of above 6.25%  that is really my aim to stay below this it makes sense I think in the long run...

I'm not sure about Sweden but in Finland we have the same 6.25% recommendation by KC but this is the first 5 generations they usually refer to, not 10 like it seems to be in UK? Naturally as a breeder you need to look much further than 5 generations!
I just thought to mention this as 5 or 10 generations and 6.25% makes a big difference...
It is actually quite difficult to find 6.25% in 10 generations in my breed... My girl for example is seen as pretty out crossed and her 10 generation COI is 6.95%...

I'm afraid that the health, quality and temperament of the dogs involved is more important than random percentage figures.

If the dogs are all rubbish in those departments, it doesn't matter how low the percentage is - the pups will be rubbish!  They may not share the same genes, but they can still share the same problems which will be amplified.

Novices think that as long as there are no shared relatives, then it's an OK mating.  To me the important thing is to actually KNOW the dogs on the pedigree and the more dogs that are involved, the less chance there is of the breeder knowing them all.

That is the closest I would ever go, and only if the other parents were totally unrelated.

But if you took her to an outcrossed line of simliar % it would decrease it greatly.

it is very difficult to find outcrossed lines!
My breed is relatively new and had 5 dogs left after the world wars so they ALL are closely related not that far back.
It is actually pretty scary to look at the pedigrees 50 years back as they are just a mingled mess of same names!
No surprise, we have some health issues popping up and it does seem closely bred litters seem to have a bigger risk(just my own observation before anybody shoots me down!).

The situation might be little different to many breeds where there are bigger populations from the start or hundreds of years of breeding behind them so current day dogs probably have more variation as a whole... but in my opinion in my breed, you do need to look at the percentage figures too...

> I'm afraid that the health, quality and temperament of the dogs involved is more important than random percentage figures.

IMO that's what phenotypical based breeding is....finding the animals with the physical attributes ( including the above) to mate together, rather than looking for relatives of a named individual.

If they share close family members ...then I consider the in breeding coefficient, but health, quality and temperament are phenotypes. 

I agree with you Toolz but most breeders in my breed either line breed or outcross to the latest winning dog and so the pedigrees have little consistency of type, temperament or quality! :(
You need other breeders to be consistantly looking for similar traits to be able to do it. :(

Amen to that W