Chapter 7 - The Implications Of Mendelism

Even if one does not discredit the Mendelian findings of heredity, even if one accepts them and assumes their truth, how may the knowledge be applied, how may it guide a practical breeder to the production of better dogs?

The theory of the chromosomes and genes, the Mendelian laws, the results of the twentieth century work that has been based upon those laws; none of these require further proof. They are not the imaginary figments of a cult, not mere hypotheses to explain the mysteries of life, not esoteric superstitions. They are scientific facts, fully accepted by every biologist worthy of the name everywhere, Soviet geneticists to the contrary notwithstanding.

They fit together almost too neatly, these facts, like the pieces of a jigsaw puzzle. They are so pat and so orderly that they are difficult to credit.

We have been accustomed to think of heredity as something fortuitous in its behavior. "Like begets like," we believed. "Like father, like son," we said, except that no son is exactly like any father—or mother either—for that matter. When like failed to beget like, it seemed that something was wrong. But what was wrong we did not know. Empirically, we bred together male and female most nearly approaching as individuals the ideal we desired in the progeny, and then trusted to luck.

In the face of a thousand observations of progeny which is much unlike its parents, we are still prone to parrot "like begets like." We see an agouti rabbit from two white parents; we see a smooth Fox Terrier in a litter of wire Fox Terriers of which we know the pedigree and that the parents and grandparents and great grandparents have been rough for generations. We are disgusted to find a "splash" in a litter of Boston Terriers of well marked ancestry; and yet we insist that "like begets like." We shrug our shoulders and mutter, "reversion, atavism —a throw back," without knowing what we mean.

And what is this reversion, this atavism? What is this "throw back"? Mendelism gives us our answer, as we shall see.

That early breeder who first uttered the dictum that "like begets like" was simply talking because he thought he had to say something. He observed that animals sometimes looked much like one of the parents, often looked somewhat like both the parents, and blurted out the false law. Perhaps because it sounded well, because the words had a nice epigrammatic finality, or, perhaps, because if it were only true it would solve all of our breeding problems, we accepted as law this glib aphorism of wishful thinking as if it had been engraved on tablets of stone and brought down to us from Sinai. Generations of breeders of animals have uncritically parroted "like begets like." "Like father, like son."

But does like beget like?

As the statement has been made and acted upon through the years since its acceptance by breeders, it is one of those half truths which are more harmful than good, outright, honest lies which can be nailed and stricken from the record. That like organisms always produce like progeny is simply not true. To modify the statement and say that like organisms tend to produce similar progeny takes from the maxim all of its finality and certitude, leaving us nothing for practical use.

The early promulgators of the "like begets like" idea knew nothing of Mendelism and less of the genes. The "like begets like" idea is much older than the modern science of genetics. What was meant is that animals look like their parents and grandparents. They do, sometimes, resemble them.

When the "like begets like" theory, however, is applied to the genes, it becomes true. Like organisms do not of necessity beget like organisms. Like dogs do not beget like dogs. Like genes do produce like genes. And like complexes of genes find like expression in the organism, be that organism a garden pea, a vinegar fly, or a Homo sapiens—a. Homo sapiens insipient enough, in the face of ocular evidence that like often begets difference, to continue to cry that like begets like.

Mendelism proves that the breeder is matching and balancing genes, not merely mating dogs and bitches. It reduces the operation to its smallest factors. Using those gene factors as stones, the breeder builds the mosaic of the new organism.

Fortunately, or unfortunately, in any single mating he must choose and utilize two animals only as the carriers of the genes he wants. He cannot choose genes for one kind of head from one dog, genes for a given kind of tail from another, genes for correct front from a third, and genes for a particular kind of coat from a fourth. He must, perforce, select two dogs for mates whose germ plasma s he believes to include the genes he chooses to perpetuate. If he wishes to introduce genes from another dog, he is compelled to wait for another generation.

And yet, utilizing the genes from a single pair of dogs, he is considering those genes as derived from all their line of ancestry. Mendelism throws another spotlight on the significance of the pedigree. The pedigree becomes to us a guide as to what genes may lurk hidden and unexpressed in these dogs whose genes we are about to unite.

When an attribute not present in either parent appears in the progeny we call it a "throw back," or atavism or reversion, recognizing perhaps that it is a "throw back" to some ancestor identified in the pedigree. But have we stopped to consider what atavism is? Except for Mendelism, it has no explanation. Have we not been given to the fatalistic acceptance of the phenomenon as some trick of a joke-loving Providence? It is no such thing. It is rather the working of immutable law. The genes which determine its expression lay hidden in the parents and, when those genes are brought together in the right combination, the atavistic attribute finds its expression in the progeny.

It may be that it is a mere recessive trait which has lurked hidden in hybrid dominance in the parents. We know that, except for sex-linked traits, it requires a pair of recessive genes for the expression of a recessive factor. Each of the hybrid dominant parents contributes one recessive gene of the pair, and the recessive trait in the phenotype results: a wall eye from dark-eyed parents, or a smooth coat from wire-haired parents. Such an attribute may lie buried for fifty generations and finally come to the surface of expression in the phenotype. Mendelism shows us how and why. The longer the recessive gene is buried, the less the likelihood of its appearance. We know that this is true because it is impossible for a recessive trait to manifest itself in the phenotype of an organism of which even one parent is a pure dominant for the set of genes of which the recessive trait under consideration is one of the alleles. If both parents are hybrid dominants for the particular set of genes, then there is a three-to-one chance for the appearance of the trait if it be a dominant, a 25-75 chance if it be a recessive.

We shall see in Chapter X, "Inbreeding, Line Breeding, and Out breeding," how consanguineous breeding serves as a sieve in which to catch such recessives. They may slip through our sieve in combination with dominants, but when two genes for a recessive factor come together they will not go through the mesh and the breeder can nab them and do with them as he sees fit. If the genes are undesirable, he can then gradually cast them out of his stock until he has animals which are pure dominants for those factors, in which case their recessive expression will never occur again. Or, if the traits are desirable, he can hold onto them and perpetuate them with almost absolute certainty.

These so-called atavisms, on the other hand, may not be recessives. They may be hybrid dominants between two sets of factors, of which one dominant gene from each member is requisite for the expression of the atavistic trait. Recall the origin of the agouti rabbit from the crossing of two distinct and different varieties, the white Polish albino and the white Beveran with dark eyes.

Such a trait, however desirable it may be, will fail to breed entirely true when mated to its like, for it will tend to split out the parental types in approximately 50% of its immediate offspring. The parental types, though, when crossed will continue to produce the trait under consideration.

And even if undesirable in itself, such an atavistic animal need not be a complete loss for breeding. If it has some other traits which one wishes to conserve, a cross with the desired parental type should in 50% of the progeny get rid of the undesirable atavism forever and may add the other traits which it is wished to conserve.

But in any event, Mendelism takes the fatalism out of breeding. "As ye sow, so also shall ye reap," remains as true as ever and even more apparent. But we must remember that we are sowing genes, not whole dogs. It is, however, very easy to sow unawares some genes we do not want to reap, hidden and buried underneath others which we do wish to reproduce. An intelligent analysis of the pedigrees of our animals reveals the source of such buried genes and we can make up our minds whether we shall utilize an animal for his genetic virtues or discard him for the faults that we know to be a part of his heritage.

Mendelism enables us to evaluate an animal not for what he is but for what he will produce. It explains why of two litter brothers, equally good looking, one may be a potent force for the betterment of his breed and the other may be utterly worthless to improve his line.

This is not, of course, usually the case, for the desirable animal is usually the one which produces desirable progeny. The genotype expresses itself very largely in the phenotype, and the experienced breeder can hazard an at least partially valid opinion about what kind of progeny an animal should produce merely from an examination of the animal.

If this were not true, the whole scheme of artificial selection would be invalid and our fine breeds of domestic livestock could never have been developed and improved by that empiric method. Merely by weeding out the worst and breeding from the best, nineteenth century breeders, and earlier ones, endowed our century with all of our true breeding varieties. There is no gainsaying that the method, however crude and empiric it may be, accomplishes the end in view. Indeed, the general run of present day breeders continues to employ that method, and if all could be induced to employ it consistently, if all could be deterred from breeding from the undesirable, the improvement of the breeds would gather an increased momentum. But Men-delism gives us the clue to a better method.

More artificial selection of phenotypes was all the technique which the breeders had, all that most contemporary breeders employ. Mendelism enables us to select the genotype. Why should we rub two sticks together to obtain fire (although there is no doubt that such friction will produce fire) when we have matches and cigarette lighters? Why waste the time and effort which might be devoted to getting warm?

These earlier breeders did not, and most current breeders do not, recognize the difference between genotype and pheno-type. That the genotype to a large degree determines the pheno-type is all that has made possible the progress that has been made. But how much of the best germ plasma has been lost because its carriers were not typical specimens of their breeds, and how much bad germ plasma has been perpetuated because its carriers have satisfied a breeder's eye, we shall never be able to estimate. It is only because good animals carry more good germ plasma than do bad animals that progress has been possible.

That one can no more determine from looking at it what kind of progeny an organism will produce than one can determine by looking at a frog how far he will jump, is illustrated by the well known work of East and Jones with Indian corn. This work has gone far and will go further to revolutionize and increase the acreage yield and the quality of that crop.

East and Jones chose two fine strains of corn. For twelve generations they forced these strains to self-fertilization, keeping the two strains separate. This will be recognized as the closest sort of inbreeding. All sorts or undesirable recessive traits appeared and were eliminated. The strains were permitted to degenerate as they would, except for the elimination of these undesirable recessives. At the end of the twelve generations, when degeneration seemed to have ceased and no further recessives appeared, the two inbred lines were hybridized. Immediately, in the first hybrid generation arose corn which was surprisingly superior, not only to the degenerated, inbred parents but to either of the two excellent strains with which the experiment was begun.

This superior hybrid corn is not yet a true breeding variety, being but hybrid dominant as to most of its sets of genes. It is only the first cross of the two degenerate strains which results in the superior corn and generations of self-fertilization and careful artificial selection will be required to so eliminate the reversion to the degenerate ancestry and produce a true breeding variety.

The experiment with corn is cited here only to show that the genetic possibilities of the organism cannot be accurately surmised from the phenotype. No intelligent, practical farmer would have planted the seed from those degenerate ears of corn in expectation of a bumper crop. The fine corn to be derived from the degenerated strains could no more have been guessed from their appearance, without consideration of their ancestry, than could agouti coloration have been expected in the progeny of the two varieties of white rabbits. So with a dog—the phenotype may not reveal the genotype.

It should be needless to say that the reader is not to accept the mention of that experiment as an encouragement to dispense with artificial selection or to, include degenerate or atypical dogs in his breeding program. However, a good dog is not to be discarded because of a single fault without consideration of whether it is one that is likely to reproduce itself. And no matter how excellent a dog may be, if it fails consistently to produce desirable progeny when discriminately mated, it should be discarded for breeding and in its place substituted even a poor dog provided that it produces better puppies.

One can be guided by three considerations in choosing a dog or bitch for breeding. First, is what is the dog like itself, how nearly it approaches the breeder's ideal of that breed and how far and in what manner it falls short of that ideal. Second, its ancestry must be considered, its pedigree analyzed to ascertain what excellencies and what faults it may have derived from which of its parents and grandparents. Third, and perhaps most important of all, one should ascertain what kind of puppies it produces when adequately mated. This last is not always possible with a young dog, and especially with a young bitch, and it is often necessary to make one's own tests of the animal's ability to transmit the typical virtues of its variety.

However, the observation of a very few litters from a dog or bitch, with a consideration of the mate and of the mate's pedigree, will reveal with what consistency it tends to produce good progeny. And dogs do tend to produce progeny with an approximation to a given excellence of type with much greater consistency than is usually believed.

Mendelism explains that consistency. If, for a given set of genes, a dog is a pure dominant, it will transmit a dominant gene of that pair to every gamete it produces. If the dog is a hybrid dominant male, then for a given set of genes it casts the dominant member of the set to half its sperm and the recessive member of the set to the other half. If a bitch, there is an equal chance of which member remains in any ovum. If the dog is recessive for any set of genes, every haploid set of chromosomes will contain one of those recessive genes, although it cannot find expression in the immediate progeny unless it find its gametic mate in the other haploid set of the zygote.

Much time, effort, and labor of love is lost in trifling along with dogs which somehow ought to produce good puppies but equally, somehow, fail to do so. While sometimes such a one will at long last "kick through" with a paragon of excellence, the chances are that if after being given adequate opportunities it fails to give one what one wants, it will be rather consistently a producer of indifferent stock, and it is wise to replace it in the breeding program with another known or believed to produce consistently well.

All of which brings us to the very heart of the mystery—the proponent individual.

The individual dog or bitch proponent to produce high excellence in its progeny is the greatest asset the breeder of dogs can possess. From such a proponent individual the strain can be developed and maintained. If it be a bitch, so much the better, for the services of a dog of recognized prepotency can be obtained to mate with it and it is worth sending to the ends of the earth to find for it the right partner.

A gem of purest ray, such a one! Flora Berkemeyer, the German Shepherd dog bitch, Annie Laurie, the Scottish Terrier, Etfa von der Saalsburg, the Great Dane. What do not their breeds owe to those bitches? Those three are outstandingly famous, but every breed has many of them.

These propontent bitches can each produce but a limited number of progeny, it is true, whereas a normal male dog can beget literally thousands of puppies if given the opportunity. But one can usually find a good producing male to mate with a good producing bitch, whereas there are not enough of these exceptional bitches to enable a much used male to maintain so high an average of excellence in its progeny.

That these proponent dogs and bitches exist is generally recognized among breeders and is much discussed, but it is only the astute or the fortunate who take advantage of their existence. They are usually thought to be something of a mystery, to arise without cause, to spring like Athene full-panoplied from the head of Zeus.

They are no such things! They are not mysteries. They are but animals whose gametes are pure dominant or recessive for the various factors which produce the typical characters of their breed. They are purebred animals in the Mendelian sense of the word "purebred." From the breeder's viewpoint, that is the only sense of the word. What does it matter if all of a dog's ancestors have been members of his variety and stud-book-registerites for however many generations if he does not register that variety's characteristics upon his progeny? Even though one of his grandparents were a mongrel, if the genes for such mon-grelism did not enter his germ plasma , he is just as much a purebred example of his variety as if all of his ancestors, clear to the dog-Adam, had been of that variety.

It is alleged that there was a white Bull Terrier not very high up in the branches of the family tree of the Airedale Terrier, Elrudge Monarch. If it be true, then breeders of Airedale Terriers will wish that there were more Bull Terrier germ plasma behind some of their other dogs, for Elrudge Monarch was one of the proponent dogs which made Airedale Terrier history. Any reasonably good bitch bred to him could be expected to have excellent progeny.

This is no plea for mongrelization of our varieties of dogs. On the other hand, it is a plea for purity of race; not merely for stud book purity, but for genetic purity, for the sorting of the genes and the retention of those which determine racial excellence and the elimination of those others which make for degeneration of type.

Stud book purity is desirable, too. But it is desirable only because it helps to establish the Mendelian purity which results in prepotence. Mendelian purity can exist without stud book purity, but it usually does not. Practically, the stud book is of the greatest value to the breeder for it is the conservator and custodian of the race's purity. Elsewhere we have seen how it would be possible, although the phenomenon has never perhaps actually occurred, that a dog could be of no kinship whatever to its own grandsire. The white genes in Elrudge Monarch are further back in the pedigree than grandfather. Furthermore, it is generally recognized that Bull Terrier blood (sic!) was used in the making of the Airedale Terrier breed, for the breeds are in many respects similar. The right Bull Terrier might be expected not to prove harmful to Airedale Terrier type as it appeared a few generations later.

Credit is lent to the allegation of Elrudge Monarch and his Bull Terrier great grandfather by the fact that a son of Monarch out of Monarch's own daughter, Champion Clonmel Rough and Ready, habitually produced puppies with white feet.

How does this Mendelian purity which results in prepotency arise? It comes from a confluence of good germ plasma , from the combination of desirable genes. Usually it springs from parents and grandparents of recognized excellence and prepotency. Some few times it simply crops up from the right combination of ordinarily good type and germ plasma , even from the fortunate mixture of the good genes of parents only hybrid dominant for many of the allelic sets. In such a case, the propontent individual is a freak, and its litter siblings may be of but mediocre merit and little breeding value. However, it is just as pure and just as potent for all time as if it were of more spectacular lineage and all its brothers were champions.

While such breeding animals may crop out from mediocre ancestry, such freaks are rare, indeed. Even when the pedigree fails to blazon the family excellence, a close analysis will usually reveal the source of the potency. There are very likely to be great, even if obscure, dogs somewhere close behind it in the pedigree.

The great, proponent dogs and bitches usually derive from a great line of proponent ancestors. The converse, however, is not true, for a dog may have a spectacular pedigree and never produce a puppy worth the water to bucket him. The explanation lies in the genes.

When a proponent individual arises in a breed, is it possible to perpetuate that prepotency in a strain? Must the breeder sit idly by and watch this prepotency dissipate itself in a generation or two? The answer is that such prepotency can be conserved and maintained from generation to generation but it seldom is.

It must be noted that a recessive gene must have a recessive allelomorph in the zygote for its expression. Therefore, there can be no such matter as prepotency for recessive characters unless the other party to the mating be recessive or hybrid dominant for such characters. And if that second party be hybrid dominant, the recessive trait can find expression in but approximately half the progeny. However propontent a dog or bitch may be, its progeny can manifest recessive characters only if the recessive gene be derived from both sides of the house.

That is not true, however, of the genes for the desirable dominant traits. They will manifest themselves in whole or in part if a single parent is pure dominant. The individual proponent for dominant characters will stamp them upon its progeny regardless of the other parent.

And that is how prepotency is lost. The breeder, seeing an animal's consistency in transmitting excellence, depends upon that animal to such an extent that he grows careless in the choice of mates for him. A dog may stamp the excellence of its germ plasma upon its immediate progeny with great consistency and without very much consideration of the kind of mates it has, but if such progeny has received from only one of its parents the excellence it manifests, it has little prospect of transmitting that excellence to subsequent generations.

And in the mates, it is to be remembered, recessive genes may lurk in hybrid dominance.

If an equally proponent mate is found for a propontent dog, his prepotency can be perpetuated and all of the immediate offspring of the mating can be expected to manifest the prepotency. Prepotency is but the purity of dominance or the recessive ness of the various sets of genes. And one gene of each set is derived from each parent.

The breeder's task is to sort and re-sort the genes until he has the desirable genes in pure sets within one organism, and then not to permit their purity to disintegrate. It is simply stated and sounds easy to do but it is a task to try the ingenuity and pertinacity of the most avid breeder.

The Mendelian doctrine simplifies that task. The realization that he is shuffling genes rather than mingling bloods enables the breeder to eliminate many of his false starts from his program before he has made them.

How often we see a breeder of dogs flash like a comet across the prize lists of the dog shows and after only a few years of success to fizzle into mediocrity! Proponent strains built up either by design or good fortune are permitted to peter out, not because of the breeder's indifference but because he fails to distinguish between the phenotype and the genotype of his dogs. He permits the pure dominance of various sets of genes to become hybrid dominant, and then cannot understand why a given dog fails to produce excellence in his progeny with the consistency its sire exhibited.

The degeneration of good strains is no more a mystery than is the establishment of them and the building up of prepotence. Acquaintance with Mendelism, even the simplified exposition of it contained in this volume, gives one to know how prepotency works, how strains are built up, and how they are usually permitted to lose their individuality.

The adjective "pure" is not a comparative term. A thing is either pure or it is not pure. A dog is purebred for any set of genes if both members of the set are recessive or both dominant. One or the other may be the desired expression in the phenotype. Hybrid dominance, though it may find an expression in the phenotype similar to that of pure dominance, is not purity.

Many breeders of dogs say and believe that the more intensely a trait is bred into an animal the more sure it is to manifest itself in the immediate progeny and through the generations. This is a fallacy. If the genes necessary for the expression of that trait be pure, one from each parent (no matter whether the parents each received one from each of their parents or from only one), they cannot be more intensely carried. And a pure set of genes may be changed to hybrid dominance in one generation, and into allelic purity in two generations. Purity has to be built up. Being reached, it may not be intensified. But having been established, it requires to be maintained.

Mendelism shows us what that purity is, how attained, how perpetuated. It gives a new meaning to purity of blood. "Blood," of course, has nothing at all to do with it. It is purity of gene pairs. If we could bring ourselves to discard the term "blood" from our consideration of heredity and substitute for it "germ plasma ," "chromosomes," and "genes," we could shed much of the fuzz from our thinking about the subject and reorient ourselves along the proper genetic lines.

But "blood" has been so long employed that the term is difficult to eliminate. Even when our use of it is awardly figurative, it is confusing. The term "blood" implies blending. Until Mendelism came into being, blood was a literal term. We thought of heredity as a current of confluent bloods. Now we know that it is rather a current of sands, any white grain of which may be removed and replaced by its black allelomorph.

Excellence in breeding cannot be achieved by the effort to blend two faulty extremes to arrive at the happy mean. If a fault is to be corrected in stock it must be balanced with excellence and not with its antithetic fault. The late Jack Holgate, successful breeder of so many varieties of dogs, whose South-boro prefix was one to conjure with, pointed that out in a discussion of Fox Terrier size some forty years ago. "Don't put a big one to a little one to get the right size, because you will get big ones and little ones, not correct ones. Use the right size for one of the mates," he explained in essence. Whether Holgate knew why or whether his advice was only empiric is impossible to say. But Mendelism furnishes us, as it would have furnished him, with an adequate explanation.

If the good of Mendelism were only in the isolation of attributes which we know and recognize as due to Mendelian unit factors, it would explain little and be of little practical value. Sufficient experiments have not been made, adequate intelligent observation has not been brought into play in our dogs to enable us to recognize a great many of these unit factor traits. In each breed there are traits which the breeder should be able to label definitely as dominant or recessive.

Once found, the breeder should surely take advantage of his knowledge, for he can control those traits with absolute certainty. We know that wall eyes are recessive to pigmented eyes, parti-color recessive to self-color, Dudley noses recessive to black noses, congenital deafness recessive to normal hearing, red recessive to black in many breeds, and several other reces-sives and dominants are recognized. The value of this specific knowledge is great and data should be accumulated.

But it is not in such cut-and-dried facts that the real value of Mendelism lies. There are not yet enough of those facts to make them more than incidentally useful. One should not permit these little sapling trees of unit factors to obscure the vision of the vast forest of Mendelian phenomena.

The great good of Mendelism is in its implication, not in its hard facts. It is a key which unlocks the profound secret of heredity. It explains prepotency and atavism. It reveals why brothers may be unlike and why remote cousins may bear strong resemblance. It offers a fundamental, material basis for all the unexpected traits which crop out in our dogs and which we had hitherto deemed merely fortuitous. It codifies the immutable laws whose workings we had thought but the antics of Puckish Providence.

Practical breeders have been accustomed to scoff at Mendelism—not at the truth of it but at the possibility of its practical application. They have accepted its truth but blinked its implications. Few of them, indeed, understood what Mendelism really is. They have read brief and sketchy expositions of it but have failed to comprehend its workings. "Mendelism" has remained a mere word to them, something having to do with growing garden peas. It has seemed to them to have no more to do with better dogs than has relativity.

They have gone on the old way, their father's way and their grandfather's way, the way which has seemed to them the easy way, but is really the hard way. But the time has come when the breeders who refuse to be guided by Mendelism will be unable to compete with breeders who accept, understand, and use it.

The old fashioned breeders will continue to make old fashioned progress with their old fashioned selection, breeding from the apparent best and eliminating the apparent worst. But the progress they make will be in no wise comparable to that of the man who accepts and utilizes the newer knowledge. It is not that one stands still and the other moves ahead, but rather that one moves ahead so much faster than the other and at such increasing momentum that competition is impossible for him who refuses to draw on the seven league boots of modern biologic science.

The late Luther Burbank, even before the more recent work with gene complexes had been accomplished, employed the Mendelian principles and techniques in his breeding of plants. Strangely enough, Burbank did not apparently realize that he was Mendelizing. There are, no doubt, many breeders of dogs who employ the Mendelian knowledge without themselves being aware that they are doing so. It is to be questioned that Jack Holgate had a clear idea of Mendelism when he gave sound Mendelian advice about the breeding of Fox Terriers.

The world is in the debt of Burbank for billions of dollars in the economic worth of the new varieties of plants he produced; and it is not to be gainsaid that Holgate was one of the foremost breeders of fine dogs the world has known. Using the Mendelian principles unconsciously, they made vast progress. Today they would use those principles consciously and persistently, else they would lag in the arts in which they had led.

By means of Mendelism, a virtue which characterizes one strain may be added to a strain which lacks it without sacrifice of virtues already possessed.

In the plant breeding world this is a recognized procedure. Sir Rowland Biffen* using Mendelian principles deliberately, bred the two now most grown varieties of wheat to order. The Dutch government in 1921 undertook to produce a new disease-resistant sugar cane for growing in Java. By 1924 the new variety was ready and now two-thirds of the cane grown in that island is of the improved variety. This, too, was deliberate, conscious Mendelization.

But it will not succeed with dogs? As early as 1828, thirty-eight years before Mendel published his experiments, Good-lake's Courser's Manual contained an account of Lord Orford's introduction of Bulldog blood into the Greyhound to improve the courage of that breed, after which he bred the hybrid back and back and back to purebred Greyhounds, retaining the Greyhound type but including Bulldog tenacity. This was a thoroughly Mendelian procedure, even though it anticipated the recognition of the Mendelian laws. Stonehenge on the Dog, published in 1859, included illustrations of dogs used in another, similar breeding experiment by one Sergeant-Major Hanley.

It is well recognized that in the nineteenth century Russian Wolfhound infusion was used to refine the skull of the Collie, that the English Foxhound was used to improve the stamina and olfactories of the Pointer, that the long-haired Dachshund was the result of crosses with Cocker Spaniels.

While the breeders who used such drastic means as the crossing of two unlike breeds to obtain a single quality for the improvement of one of them did not possess the label of Mendelism to attach to their technique, yet the genes behaved just as the same genes would have done after the new laws had been discovered. Had those breeders possessed the knowledge we now have, they might have undertaken their work with greater surety and accomplished it in less time and with better results.

It is not here advocated that crosses outside the various breeds of dogs be made for their improvement. Such crossbreeding is not now necessary. These instances are cited merely to show that Mendelian principles can be applied to the breeding of dogs, even to the extent of crossing of varieties and eliminating all of the foreign genes excepting those which produce the expression of traits for which the cross is made.

If breeds may be improved by crossing, so can strains within the breeds and the virtues of one strain annexed to another.

Indeed, every improvement in any breed that has ever been made has been due to the favorable Mendelian behavior of the genes. Every degeneration has been due to unfavorable behavior of the genes. With a sound knowledge of Mendelism, it is possible so to sort the genes that improvement is constant. Much trial and error can be eliminated. The improvement can go always forward.

One objection which may be offered to the extensive employment of Mendelism is that its utilization requires a vast kennel of dogs and many generations to obtain favorable results. Such an objection is not a valid one.

It is true that in the breeding of plants selection can be made from large numbers of experimental organisms. This is essential in the radical out-crosses that are made. In the crossing of breeds of dogs, such as we have here discussed, larger numbers might have hastened the results.

However, in working with purebred stock of a single breed, when desirable qualities have been obtained and fixed in their purity, a single pair of animals only is required for the maintenance of that genetic purity. The extensive breeder can continue the old haphazard breeding, selecting the good stock when it occurs, ignoring the wasters that arise. The small breeder, by utilizing modern knowledge, can eliminate entirely from his breeding stock the producers of merely mediocre progeny, and can, by concentrating his efforts upon stock of tested breeding worth, produce a very high percentage of dogs of great excellence.

There are and have always been many small kennels which turn out excellent puppies with a consistency seldom known in the larger establishments. A conscious application of Mendelism should not destroy that consistency, but rather should increase it. It will enable the small breeder as well as the large breeder to make his mating with an assurance and confidence, with a definiteness of the end in view, with an absence of guess-work which were hitherto impossible.

It makes possible the formulation of a program of breeding for several generations in advance. Such a program may include the utilization of dogs whose grandparents are yet unborn. It facilitates a planned genetic economy.

To the objection that the employment of the knowledge gained from science changes the breeding of dogs from a creative art to an exact science, the rejoinder is that the more is the pity it does not do such a thing. In that far-off millennium when we shall have a complete map of the genes of the thirty-nine hap-loid chromosomes of the dog, we many consider dog breeding an exact science.

Meanwhile, it must remain an art to which we may apply what scientific knowledge we may possess or be able to obtain. Is the Sistine Madonna a better picture because Raphael did not know enough of chemistry to use pigments which would not alter in four hundred years? Windsor and Newton make better pigments than Raphael used, but we have no Raphael to put them on the canvas with his genius. The chemistry of paint has not removed the artistry from painting. On the other hand, it has facilitated what artistry exists.

And so the knowledge of the genes and chromosomes would no more destroy the breeder's art, as an art, than the knowledge of atoms and molecules would destroy the painter's art.

By the ideals he formulates, and by the approximation of the dogs he produces to those ideals, the breeder of dogs expresses his personality. This self-expression, like any other, is art. The breeding of dogs has not been and cannot be reduced to a formula. The Mendelian laws and the later knowledge based on them do not make for any such formalization of procedure. They but clear the way for the realization of such ideals of their respective varieties toward which the breeder chooses to work.

The implications of Mendelism in dog breeding can only be partially stated. Every breeder who recognizes its value and seeks to comprehend it and to utilize it will read into it implications of his own. It will lift veil after veil, open up horizon after horizon, until the mirage of perfection seems within his grasp.

It is an inescapable law. Used, it will lead to better dogs. Abused, it will wreck any strain.

Are You Ready To Move Onto The Next Lesson? Click Here...