Search

Population genetics has a strong theoretical basis. Indeed, a great deal of its theoretical foundation was laid in the 1920s and 1930s, by the famous triumvirate of Ronald A. Fisher, J.B.S. Haldane and Sewall Wright. Richard Lewontin wryly observed in 1963 that they "had said everything of truly fundamental importance about the theory of genetic change in populations and it is due mainly to man’s infinite capacity to make more and more out of less and less, that the rest of us are not currently among the unemployed."

This "death" pronouncement for population genetics proved to be premature, of course. Subsequent technical innovation enabled the collection of data at a molecular level, and this development reinvigorated population genetics. This sort of progression -- where theories precede the experimental data needed to prove or disprove them -- is hardly unusual in science. In physics, for example, some of the theories proposed by Einstein required decades of technical advancement before they could be demonstrated experimentally.

Prior to the molecular epoch in population genetics, for example, test breedings were generally required to infer genotypes -- a painstaking process. For reasons of practicality, a tremendous amount of such research was, and still is, conducted on a single species: drosophila melanogaster (fruit fly). Despite the molecular techniques which now provide a way to study genetic variation in populations outside the laboratory, the data only give us a snapshot of things as they are now. Much of what we know, or think we know, is still based on theory.

But as the famous Spanish philosopher Jose Ortega y Gasset once wrote: "Life cannot wait until the sciences have explained the universe. We cannot put off living until we are ready. The most salient characteristic of life is its coerciveness; it is always urgent, ‘here and now,’ without any possible postponement. Life is fired at us point blank."

A case in point is the branch of conservation biology concerned with the captive breeding of rare and endangered species, whose management requires that decisions be made now -- not next week, or whenever science gets around to providing definitive answers. Although the goals of captive breeding are different from purebred dog breeding, a brief historical overview of captive breeding should be of interest to dog breeders, because there are definite parallels between the two activities. For example, both deal with small (or relatively small), genetically isolated populations that persist through time.

First, it may be helpful to mention the perceived importance of population genetics within conservation biology. In the past, some conservation biologists may have overemphasized the importance of population genetics. It had been thought that the loss of genetic variation might be a primary factor in the extinction of species. More recent thinking suggests that demographic factors (habitat destruction, etc.) are primary, and that a reduction in genetic diversity is more of a symptom than a cause of species decline. In nature, it seems, bad demographics kill you off sooner than bad genes. Or so goes the current thinking. Nevertheless, population genetics continues to play a vital role in conservation biology.

Surprisingly, the widespread application of population genetic concepts to captive breeding is of fairly recent origin. In 1965, for example, only a handful of zoological parks were using these concepts to manage species; by 1996, over 250 were. The old idea, borrowed from agriculture, of selectively breeding only the "best" animals has largely been abandoned. What caused this sea change in thinking? The seminal event was a study in 1979 by Ralls et al., which was the first to quantitatively document a lower survival rate for inbred (versus noninbred) captive endangered species. Most zoos now avoid inbreeding as much as possible. Stated more positively, their goal is to maximize the retention of genetic variation in captive populations. To this end, the following strategy is typical: start with as many founder animals as possible, maximize the number of breeding animals per generation, grow the population as rapidly as possible to carrying capacity, equalize the number of male and female breeders, equalize the genetic contributions of founding animals and families in subsequent generations, and minimize fluctuations in population size once carrying capacity has been reached.

This is all very well, you say, but what has this to do with dogs? More than you might think. Although it is certainly true that the dog, as a species, is wildly successful in terms of demographics, due to its symbiotic relationship with mankind, it is also true that there are now more than 400 separate breeds of dogs. Furthermore, under the purebred registry system which has prevailed now for over a century, there is no provision for gene flow between breeds. In effect, we have artificially created a whole lot of endangered (sub)species of dogs. But unlike zoos, our goal is to maximize physical uniformity, as measured by a breed standard. Far from conserving genetic variation, we usually try as far as possible to eliminate it. To this end, the following strategy is typical: start with as few founder animals as possible (admittedly this is not usually done by design, but it is typical of many breeds nonetheless), minimize the number of breeding animals per generation (Ubbink documents that only 3 to 5 percent of registered purebred dogs in the Netherlands contribute to the next generation), keep the population as small as possible for as long as possible (common dog wisdom says popularity is always a bad thing), skew the sex ratio of breeding animals as much as possible (neuter all but the very best males, but be much more lenient with females), and skew the genetic contributions of founding animals and families by overusing some dogs and completely neglecting others. Dog breeders rate a pass on minimizing fluctuations in population size, since the dominant strategy is to keep populations small at all times.

Does this strike anyone else as a recipe for potential diaster? There is no doubt that inbreeding is what enabled dog breeders to fix breed type. Indeed, dog breeds probably could not have been created in the first place without inbreeding. But dog breeders need outbreeding too. And in most breeds, outbreeding is no longer possible due to the effects of remote inbreeding. I'd like to close with a quote from Stephen Budiansky, who in his recent book The Truth About Dogs, had this to say: "From a scientific point of view, inbreeding is merely one tool, to be used to achieve a desired end: it is not an end in itself. But the establishment of closed breeding populations and registries of breeds by their very nature sets inbreeding up as something to be cultivated for its own sake. It is, as far as modern genetics is concerned, a slight absurdity... It is very hard to dispute that in the single-minded pursuit of a fairly narrow set of mostly visual criteria, many dog breeders have tipped way too far toward uniformity on the uniformity-diversity continuum. There is no inherent reason one cannot breed for looks and healthiness or looks and good temperament simultaneously. But it requires striking a better balance between inbreeding and outbreeding than has been pursued by many to date."

Mike McIntyre
http://www.homestead.com/dogbreedersguild/files/popgen.htm
Tenpenny@compuserve.com