STRIKING A BALANCE
Dr John Armstrong
©1997
Lessons from the wolves
Too many breeders follow the "breeding for extinction" paradigm. Start with too few founders, close the registry to new imports, inbreed and breed your preeminent males as many times as possible. It is a guaranteed recipe for degrading the gene pool. Breeders will visit the Diversity web site, read some of the information, and come away still believing that what I propose will lead to deterioration of their line, often insisting that the best and most successful breeders don't breed this way. They don't see the relevance of the strategies for saving endangered species or populations, and maintain that the factors taken into consideration for the Mexican or Ethiopian wolf don't apply to Canis familiaris.
To be sure, the domestic dog, as a species, has considerable genetic diversity, and is not in any danger of extinction. However, I see no fundamental difference between breeds of domestic dog and subspecies of wolf. If the Mexican wolf is distinctly different from the Ethiopian wolf, and if there is merit in trying to save both, is there not also merit in trying to preserve the Poodle, the KyiApso, the Malamute, etc. The difference is only that the evolution of the different wolves has been determined less by man and more by nature, and that man has generally done a worse job.
It is amazing how many breeders accept that "fading" puppies, stillbirths, small litters, inability to conceive, and even the occasional disaster litter (where most or all are lost) are "normal". They will dismiss the statistic that, for example, a Standard Poodle lives, on the average, two years less than a Miniature Poodle with the statement that "big dogs don't live as long as small dogs". They never question the logic. Do small animals live longer in the wild? (Don't tell me that such a comparison is not relevant unless you can come up with a logical reason for why a smaller animal should live longer.)
I have been asked, several times, what to tell such a person. I'm afraid that my reaction is largely one that you make someone aware of alternative possibilities, but if they insist on clinging to their old beliefs it is pointless to persist. Altering someone's fundamental beliefs is not easy. The best you can do is to show them that there are alternatives to consider, and hope that they will wake up and recognize the reality of the situation.
How Many Founders?
This question has come up again in relation to the attempts of a few breeders to rescue the KyiApso, which I will describe, probably unfairly, as an overgrown Lhasa Apso (~ 26", 80 lb.). Six have been imported to North America and bred. There are now about two dozen breedable animals including four of the original imports. Is this sufficient to establish (and possibly save) the breed, which may be threatened with extinction in Tibet?
In "The Millenium Ark", Soule et al. conclude that 6 is the absolute minimum number of founders, and that with fewer, the group will lose more than 10% of its genetic diversity as soon as it reproduces. Twenty is a much safer number. Furthermore, the population size should be increased to 200-300 as quickly as possible to prevent significant loss of diversity. The necessity for increasing the population size is to prevent random loss of alleles that may be important.
In genetics we deal in probabilities. Each fertilization is an independent event, and one does not influence another. Sometimes probabilities seem counterintuitive. Surely if I keep tossing this coin, and it keeps coming up heads, the probability of getting a tail must be increasing. The answer is "no, it is 1/2 each time". What does change with an increased number of trials is the probability of getting all heads (or all tails). If we toss that coin twice, the probability of getting a tail is 1/2 each time, but the question should be "what is the probability of getting two tails when you toss it twice." In effect, you are asking "what is the probability of getting a tail on the first toss AND ALSO getting a tail on the second toss." If you ask the probability of two independent events, A and B, both happening, you multiply the probabilities: 1/2 x 1/2 = 1/4.
From the practical perspective of planning what matings need to be done to retain founder alleles, the numbers game is much the same. We want to know how many progeny are necessary to be reasonably certain of retaining a desirable allele for some gene. If you have only one son or daughter that carries on the line, the probability is 1/2 that the allele will be "dropped" (not passed on). If you retain six progeny, the probability is reduced to (1/2 x 1/2 x 1/2 x 1/2 x 1/2 x 1/2) = 1/64 (about 1.6%). This should assure that the allele is retained better than 98% of the time.
Once the population size reaches several hundred, we no longer need to keep 6 progeny - as long as the number of animals carrying the desired allele is reasonably high and everyone has a fairly equal chance of contributing to the next generation. However, once you impose selection, the rules change. Even the decision not to breed an affected individual, while desirable from the viewpoint of reducing the incidence of that particular problem, also eliminates all the other genes carried by that individual. In effect, you are hoping that other, unaffected individuals carry the good alleles you are discarding. A program to eliminate all the carriers of a "bad" allele for some gene carries, in my opinion, an unacceptably high risk.
The Preeminent Male
I repeat -- The decision NOT to breed an individual also eliminates all the other genes carried by that individual. Such a decision may be largely taken out of the hands of the owner, as everyone flocks to the current star. This is a form of selection. Intentionally or otherwise, when you select for one thing you are selecting against others. If one line gains a significant advantage and maintains it, the others will be denied the opportunity to pass on their genes and diversity will be lost.
Some might argue that if this line is really the outstanding one, we are improving the quality of the breed. To take a real example, let's argue that Annsown Sir Gay (b. 1949) was the best black poodle ever bred and that anyone who wants a good black poodle would be overjoyed at having a Sir Gay. However, even if he truly represents perfection in the breed, favoring his line is not producing clones of Sir Gay.
Recently, a visitor to the Diversity Website put forward the argument that surely the outstanding males from his line produced enough offspring that most of his genes should be preserved in the gene pool of today -- and that it would not be just his bad genes that were passed on. Though that may be true, she has missed the point. Even if all his genes have been preserved, when you sample the gene pool to create a new puppy you are not going to get the exact "Sir Gay collection." If the gene pool is only Sir Gay's, and if he carried only one bad gene, the laws of genetics predict that 25% of our new creations are going to get 2 copies. Too bad! (Pun intended)
Something for the poodle-people to think about: the top 10 (and very likely the top 20 or more) top-producing Standard males of the past 40 years - ALL colors included - are direct descendants of Sir Gay.
My own attitude toward breeding is that one should try to breed like to like and best to best while at the same time minimizing common ancestry. Practiced consistently, it does work. I know of a number of breeders who breed this way and will tell you that they are pleased with their results.
Too many breeders follow the "breeding for extinction" paradigm. Start with too few founders, close the registry to new imports, inbreed and breed your preeminent males as many times as possible. It is a guaranteed recipe for degrading the gene pool. Breeders will visit the Diversity web site, read some of the information, and come away still believing that what I propose will lead to deterioration of their line, often insisting that the best and most successful breeders don't breed this way. They don't see the relevance of the strategies for saving endangered species or populations, and maintain that the factors taken into consideration for the Mexican or Ethiopian wolf don't apply to Canis familiaris.
To be sure, the domestic dog, as a species, has considerable genetic diversity, and is not in any danger of extinction. However, I see no fundamental difference between breeds of domestic dog and subspecies of wolf. If the Mexican wolf is distinctly different from the Ethiopian wolf, and if there is merit in trying to save both, is there not also merit in trying to preserve the Poodle, the KyiApso, the Malamute, etc. The difference is only that the evolution of the different wolves has been determined less by man and more by nature, and that man has generally done a worse job.
It is amazing how many breeders accept that "fading" puppies, stillbirths, small litters, inability to conceive, and even the occasional disaster litter (where most or all are lost) are "normal". They will dismiss the statistic that, for example, a Standard Poodle lives, on the average, two years less than a Miniature Poodle with the statement that "big dogs don't live as long as small dogs". They never question the logic. Do small animals live longer in the wild? (Don't tell me that such a comparison is not relevant unless you can come up with a logical reason for why a smaller animal should live longer.)
I have been asked, several times, what to tell such a person. I'm afraid that my reaction is largely one that you make someone aware of alternative possibilities, but if they insist on clinging to their old beliefs it is pointless to persist. Altering someone's fundamental beliefs is not easy. The best you can do is to show them that there are alternatives to consider, and hope that they will wake up and recognize the reality of the situation.
How Many Founders?
This question has come up again in relation to the attempts of a few breeders to rescue the KyiApso, which I will describe, probably unfairly, as an overgrown Lhasa Apso (~ 26", 80 lb.). Six have been imported to North America and bred. There are now about two dozen breedable animals including four of the original imports. Is this sufficient to establish (and possibly save) the breed, which may be threatened with extinction in Tibet?
In "The Millenium Ark", Soule et al. conclude that 6 is the absolute minimum number of founders, and that with fewer, the group will lose more than 10% of its genetic diversity as soon as it reproduces. Twenty is a much safer number. Furthermore, the population size should be increased to 200-300 as quickly as possible to prevent significant loss of diversity. The necessity for increasing the population size is to prevent random loss of alleles that may be important.
In genetics we deal in probabilities. Each fertilization is an independent event, and one does not influence another. Sometimes probabilities seem counterintuitive. Surely if I keep tossing this coin, and it keeps coming up heads, the probability of getting a tail must be increasing. The answer is "no, it is 1/2 each time". What does change with an increased number of trials is the probability of getting all heads (or all tails). If we toss that coin twice, the probability of getting a tail is 1/2 each time, but the question should be "what is the probability of getting two tails when you toss it twice." In effect, you are asking "what is the probability of getting a tail on the first toss AND ALSO getting a tail on the second toss." If you ask the probability of two independent events, A and B, both happening, you multiply the probabilities: 1/2 x 1/2 = 1/4.
From the practical perspective of planning what matings need to be done to retain founder alleles, the numbers game is much the same. We want to know how many progeny are necessary to be reasonably certain of retaining a desirable allele for some gene. If you have only one son or daughter that carries on the line, the probability is 1/2 that the allele will be "dropped" (not passed on). If you retain six progeny, the probability is reduced to (1/2 x 1/2 x 1/2 x 1/2 x 1/2 x 1/2) = 1/64 (about 1.6%). This should assure that the allele is retained better than 98% of the time.
Once the population size reaches several hundred, we no longer need to keep 6 progeny - as long as the number of animals carrying the desired allele is reasonably high and everyone has a fairly equal chance of contributing to the next generation. However, once you impose selection, the rules change. Even the decision not to breed an affected individual, while desirable from the viewpoint of reducing the incidence of that particular problem, also eliminates all the other genes carried by that individual. In effect, you are hoping that other, unaffected individuals carry the good alleles you are discarding. A program to eliminate all the carriers of a "bad" allele for some gene carries, in my opinion, an unacceptably high risk.
The Preeminent Male
I repeat -- The decision NOT to breed an individual also eliminates all the other genes carried by that individual. Such a decision may be largely taken out of the hands of the owner, as everyone flocks to the current star. This is a form of selection. Intentionally or otherwise, when you select for one thing you are selecting against others. If one line gains a significant advantage and maintains it, the others will be denied the opportunity to pass on their genes and diversity will be lost.
Some might argue that if this line is really the outstanding one, we are improving the quality of the breed. To take a real example, let's argue that Annsown Sir Gay (b. 1949) was the best black poodle ever bred and that anyone who wants a good black poodle would be overjoyed at having a Sir Gay. However, even if he truly represents perfection in the breed, favoring his line is not producing clones of Sir Gay.
Recently, a visitor to the Diversity Website put forward the argument that surely the outstanding males from his line produced enough offspring that most of his genes should be preserved in the gene pool of today -- and that it would not be just his bad genes that were passed on. Though that may be true, she has missed the point. Even if all his genes have been preserved, when you sample the gene pool to create a new puppy you are not going to get the exact "Sir Gay collection." If the gene pool is only Sir Gay's, and if he carried only one bad gene, the laws of genetics predict that 25% of our new creations are going to get 2 copies. Too bad! (Pun intended)
Something for the poodle-people to think about: the top 10 (and very likely the top 20 or more) top-producing Standard males of the past 40 years - ALL colors included - are direct descendants of Sir Gay.
My own attitude toward breeding is that one should try to breed like to like and best to best while at the same time minimizing common ancestry. Practiced consistently, it does work. I know of a number of breeders who breed this way and will tell you that they are pleased with their results.
