Three things every dog fancier has: a private formula for a sure cure for distemper; a pet method of feeding and conditioning; and a system of breeding. These three vary almost as infinitely as the total number of those riding the doggy hobby. Nevertheless, each personal variation can usually be placed under a general classification.
There have been, so I have read, something over three thousand distinct and infallible systems for breaking the bank at Monte Carlo; but all of these are founded upon four basic principles. It is much the same with all the different systems of breeding a champion. All these systems can be reduced either to a single one, or at best a combination of two or three, of six different basic systems. Strange to relate, three of these six are no systems at all.
The use of the word system applied to a plan or method of dog breeding is unfortunate. The word implies scientific order, a definite rule. It connotes infallibility, a sort of guaranteed success. No breeder should ever become so infatuated with a particular scheme of mating as to follow it blindly, and any breeding system that sets itself up to be a set of fixed and unchanging rules can at the very outset be regarded with suspicion. Such advice as "Always breed a bitch to her sire's sire" is wickedly foolish. The breeder who attempts to follow it is courting failure. The principles of genetics cannot be expressed in mathematical formula that will invariably work out a correct solution for any breeding problem. Other breeding systems are guesses founded upon practical observation, but misinterpreted by lack of experience or insufficient knowledge. Others are merely excuses for ignorance, or more often laziness. Neither a guess nor an excuse can rightly be called a system.
The best of the so-called systems are those that neither attempt to lay down fixed rules, nor to bolster up weakness, nor to offer short cuts to success. At best, they are merely guide posts pointing out the way to sound breeding practice, and even to call them systems is twisting the meaning of that word.
Because some breeding systems are good, others are bad, and still others are indifferent, it is well worth the serious breeder's while to consider them somewhat in detail. After so doing he will be better able to map out for himself a course that he is to follow in his breeding operations. Let us first consider those systems that are no systems at all.
The greatest of these is the trust-to-luck system. It is to be feared that a majority of fanciers breed their dogs on this basis. "Put two good ones together, and trust to luck." That sums up the idea of these men. It is this famous system that, as H. E. Packwood puts it, "floods the whole world with well bred bad dogs," for a dog can have a pedigree full of champions and yet be not only a poor specimen but also be downright badly bred. Some strains of blood mix about as well as oil and water, and to attempt to cross them results in positive retrogression. Selection is the only direct force in breeding that is at the breeder's command, and when he fails to employ it he ceases to be a breeder at all.
Next to the trust-to-luck breeding, and just about as popular and just about as pernicious, is the fashionable breeding system. By this method a bitch is invariably mated to the latest sensational winner. Pedigree, and all that it means in heredity, individual points, and all that they mean in variation,—both are blissfully ignored. Only the number of first and special prizes won count, and it is very well known that, given a certain amount of quality, the number of prizes won is almost always in direct ratio to the opportunities given a dog. A reasonably typical dog in the hands of a professional visiting every show in the circuit, or a reasonably good performer run through all the field trials from Connecticut to Mississippi, either will have a bigger list of wins to his credit at the end of the season, than a truly superior dog, the property of a fancier who only supports his own local fixtures. Trust-to-luck makes no pretense of serious selection. Fashionable breeding deliberately bases its selection on a gauge as stable as a weather vane.
It is not to be supposed that there have been no valuable sires who have also been popular winners, but a breeder who sends his brood bitches to the latest sensation is very apt to produce dogs like the one a fair fancier showed under George Raper at a show in the Midlands several years ago. The dean of the English judges gave the entry the gate and his owner later came up and complained bitterly that such things should not be because, forsooth her entry was sired by Champion So-and-so and his dam was Champion Such-a-one. Raper's answer was, "Next time, Madam, show the pedigree and leave the dog at home."
A third so-called breeding system is the egotistical system. A fancier breeds to his own dogs because it is cheaper, because it gives them greater opportunity as sires, or because it supplies the puppies with pedigrees that look as if he had established a strain of his own. Often he is absolutely honest in his belief that his stud dog is the greatest living sire, but this blind prejudice, even though it be innocent, is none the less at fault because it is blind and innocent. This egotistical breeding is the breeding vice of little minds.
These three systems, the trust-to-luck, the fashionable, and the egotistical are all actively bad. By a lucky fluke some few good dogs may have been bred according to them, but any one of the three followed consistently is sure to result in great waste of time, effort, and good breeding material, without having made the least progress. They are the ones that are no systems at all, and together they are undoubtedly responsible for that vast army of thoroughbred dogs whelped every year that are never good enough to be anything more than a yard dog or a companion.
The three systems that remain are much more serious attempts at careful selection based upon definite principles. If they were no more, they would be great improvements over those we have discussed. Properly employed, each is a valuable aid. Their abuse has lessened their usefulness. Not one of them can be considered an infallible rule of sound breeding, for any one persistently employed will bring a breeder upon the rocks. These systems are in-breeding, line breeding, and out-breeding.
There are, we must remember, but three forms of true in-breeding. In-breeding necessitates the direct crossing of one individual's blood. This is only possible when a sire is mated to his own daughter; when a dam is bred to her own son; or when full brother and sister are crossed. Probably nine fanciers out of ten consider the mating of half brother and sister (dogs with either the same sire or else the same dam) as in-breeding. This is the primary cross of line-breeding. In fact, as we discovered in our pedigree studies, most of the ideas about the prevalency of in-breeding results from a confusion of these two terms and what each really means.
The mating of a sire to his own daughter obviously results in a tremendous intensification of his own blood. Using the figures of the numerical expression of the law of ancestral heredity, there will be 31 per cent. of his blood in the puppies. Plainly, such a cross is the most effective means of perpetuating the qualities of a certain dog. The same is true in the case of a bitch bred to her son. These two are the closest forms of breeding, though, of course, if the individuals employed be themselves in-bred, or even line bred, the intensification of blood will be even greater.
The breeding together of full brother and sister is the most effective means of preserving and magnifying a certain cross. A bitch bred to a certain dog has produced a litter of very exceptional quality. To repeat the cross will be an uncertain experiment at best, but to breed together the best dog and the best bitch of the exceptional litter will fix the nick of blood firmly.
In-breeding for generation after generation will result in degeneration of physical and mental powers, and, if persisted in, will eventually mean sterility and the production of monstrosities. Blind puppies, and ones with malformations, especially of the legs, feet, ears, and tail, are the goal of continued in- and in-breeding. How soon the deterioration will appear depends upon the ruggedness of the variety experimented with and the conditions of kenneling, exercise, and feeding under which the breeding stock lives. It is, however, perfectly safe to say that to-day no breed of dogs is in-bred to anywhere near an injurious point. The total number of in-bred dogs certainly does not exceed 10 per cent., and in most breeds probably not 5 per cent. These figures show that there is much exaggeration in the belief in the commonness of in-breeding.
The danger of in-breeding lies not, as so many suppose, in deterioration, but in the magnification of the heredity of a single individual. This is, of course, the very object of all inbreeding, but since the good and bad points are both intensified, in-breeding is a double edged sword. It is just as difficult. If not more so, to remove faults as to improve excellencies. The removal of in-bred faults is a "bad job." For this reason, in-breeding can only be judiciously used. However, the fact remains that in-breeding, properly understood and wisely applied, is a most directly effective weapon in the hands of the breeder. Much of the improvement in our blooded stock—this is particularly true of cattle and poultry—has been the direct result of skilful and prudent in-breeding.
Judged by the results produced, however, line breeding, although its results have been slower and are less sensational, has been even more effective. In our pedigree studies we saw that line breeding has, in the case of two typical Terriers, produced more than five times as many champions as in-breeding, and the breeding of a champion is the object of the dog fancier's breeding operations. Line breeding may be considered as a slower, safer, surer method of close breeding.
Line breeding can be defined as the combining of the blood of a certain individual without the direct use of that same individual. It is fairly represented by the marriage of cousins in whose children the blood of the grand-parents is again combined. It is possible—in fact, fanciers often use the term in this way—to have a dog line bred to a certain great individual without that individual's name actually appearing in the pedigree at all. Such a line bred dog offers all sorts of opportunities for the use of certain blood in various combinations, since he can be mated directly to the exceptional individual in question without the attendant dangers, both active and passive, of direct in-breeding.
The primary or closest cross of line-breeding is the mating of half brother and half sister. This, like in-breeding, offers a means of concentration of one individual's blood. The outer limits of line-breeding shade off into out breeding. It is not always possible to determine just when the one begins and the other ends, but for practical purposes a rough and ready rule can be based on the law of ancestral heredity by disregarding as line-breeding anything beyond the fourth generation.
Two subordinate modifications of line-breeding principles are the figure system and the alternate generation system. The former was originally worked out for race horses and was first presented to dog fanciers by C. J. Davies in his book on breeding. In his monograph on Scottish Terriers he traces out the system in the pedigrees of this breed. The distinguishing feature of the figure system is the transfer of the emphasis from the sire to the dam. Instead of studying the sires in a pedigree, the dams are analysed, and all dogs are placed in families traced back through their dams to the original females of the breed. The system has one very obvious advantage. The fact that a sire usually has many more progeny than a dam makes the determination of his true worth as a breeding individual a much more complicated and difficult task. With his many opportunities, a dog at public stud may sire several winners and yet not be a truly exceptional sire. On the other hand, if a bitch produces the same number of good offspring she is almost surely an exceptional breeder.
The figure system traces back the different families through the bitch line, and by this means discovers the strains that have been specially prepotent. As an example, Davies, in tracing out the Scottish Terrier blood lines, discovered that the vast majority of the champions of the breed have come through their dams from the bitch Splinter II. Bitches who trace in tail female to this bitch and their sons are members of the Splinter II family, and, as such, are peculiarly desirable breeding stock. The figure system has never gained any general popularity among dog fanciers, probably because of the labor necessary to trace out the female lines of the different breeds. Few dog fanciers have the patience and the industry of the chief supporter of this system, which, moreover, has no very apparent advantage over the method adapted from Davenport in discovering the prepotent strains through the better known and more easily traced sires.
The alternate generation system of breeding is based upon the belief that exceptional quality is accustomed to skip a generation. A famous dog is noted for his bitch puppies, or a dog never gets a son as good as himself, are two very common and very dogmatic statements. Their support rests upon the fact that several great dogs have failed at stud, and that their sons have often been markedly successful sires. We have seen, however, that this does not hold as generally true. The shining exceptions have led fanciers astray. The principles of genetics would certainly not bear out any belief in certain characters skipping a generation, except in the case of reversions which are, however, very irregular in their action and quite uncommon. The drag of the race would be at work to make the get of exceptional individuals less exceptional, i.e., closer to the average of the race, but there is no biological evidence to support the idea that it is better to breed to the progeny of an exceptional dog or bitch, if the parent is available.
Straight out-breeding, the scrupulous avoidance of all close breeding of any type, is a child of the super-fear of the noxious effects of continued close breeding. From what has been said, it is plain that this is foolish and unprofitable. Moreover, from a practical point of view, it is almost impossible. The line of descent of most breeds has come to be concentrated in certain families which have proved to be overwhelmingly prepotent and from which most champions and exceptional breeding individuals have been bred. Quite naturally, sometimes consciously and often unconsciously, this has resulted in more or less consanguinity throughout the variety. Any breeder, whose breeding operations were extensive and extended over any considerable period of time, would find it difficult to live up religiously to a system of complete out-breeding. If he succeeded in so doing, his dogs would be such a jumble of mixed strains and contrasting heredities that any success that he might achieve would surely be sporadic and fleeting.
Destructive criticism is not nearly so valuable as constructive. It is a fruitless and thankless task to pull apart the various breeding systems without piecing together a substitute to take their place. "Piecing together" is the correct verb, for the breeding system I am going to recommend is not an original production, though I am sure that in the form I offer it, it has never been presented to dog fanciers. I have borrowed freely from the work of other dog owners, from the practice of other breeders, and from practical application of the principles of genetics, always working upon the foundation of my own experience.
First the dog breeder who seriously intends to make every effort to guarantee his own success will learn all he can from the history of his breed, and next, after thoroughly understanding their points, he will draw up for himself a very definite ideal. His actual breeding operations will be directed towards the establishment of a strain that will as closely as possible approximate this ideal.
Remembering always that until the drag of the race can be transformed into an ally, it is the breeder's worst enemy, he should strive to accomplish this, rather than to attempt any chance success in the production of a flyer or two. How can this be done, and how can it be done most quickly and with the least expense?
Even a casual study of any breed will reveal the fact that certain points "come good" in the majority of the dogs. Other points are commonly bad. If in the selection of the brood bitches of his kennels, a breeder get two or three of sound average type, but excelling particularly in those points in which their breed, as a breed, is weak, he will have made the best possible start toward the establishment of that ideal strain. Naturally, these bitches should not only excel in these weak points of their breed, but should, so much as possible, be bred from stock strong in these same characters. Bred to dogs excelling in these same points, and better in others, the foundation of the strain is well laid.
In every subsequent mating that takes place a breeder should always strive to hold every good point possessed by his bitch, and to add to them something extra from the stud dog. The dangerous pitfall that trips hundreds of thoughtful breeders is an attempt to balance points, good and bad, against each other. A bitch excelling in eyes, skull, and ears but lamentably bad in foreface will be bred to a dog with a capital foreface, but shocking in eyes, ears, and skull. Or a bitch with speed and hunting sense, but lacking in bottom, will be mated to a solid dog with substance to spare, but little else to recommend him. As an example of glorious optimism such matings are splendid, but as breeding operations they are pathetic. The result is more than apt to be a spoiling of whatever good points were possessed by both parents, for these points were probably above the average of the race, and the principle of regression would tend to pull the average of the puppies back closer to the breed mean. Always hold then whatever good points we have, endeavoring in each successive mating to add to these other good points.
The advantage of starting at the weakest points of a breed and working up toward those points that are commonly good in the variety with which one is dealing is almost self-evident. It tremendously simplifies the problems of selection. The more generally a certain desirable point is possessed by a breed, the easier it will be to find a suitable mate possessing it. Sooner or later every breeder is forced to compromise. He must sacrifice one point for another, but the longer he can postpone this compromise, the firmer will be the foundation of his strain. Moreover, a dog excelling in points notably weak in his variety will never fail to catch a judge's eye. This very practical, almost mercenary, consideration cannot be overlooked in these days of keen competition.
Working to found a strain from weakness through to strength, and always scrupulously holding all the good points and trying to add other excellencies to them is a broad policy that will be found to be eminently practical and valuable. Working on this basis, there are two things useful in narrowing the selection of the individual dogs in any particular mating. Always judge a dog as a breeding unit not by its own points, but on its ancestry and progeny.
To form a right estimate of the ancestry of a dog a knowledge of the points, both good and bad, of the dogs of the past is absolutely necessary, and a wide acquaintance with the dogs of the present is imperative if one is to judge a dog by its progeny. There are great practical difficulties in the way of doing this. It is, however, the best and surest means of arriving at a correct conclusion of the true worth of any individual as a breeding unit. Every scrap of definite knowledge of this kind can be regarded as a valuable asset, but breeders are fortunate in having another alternative that gives valuable assistance in selection. If an accurate tracing out of the winning strain cannot supplant knowledge of the points of the dogs of a breed, at least it can ably supplement it. In the Airedale and Scottish Terriers we found, and the same is probably true of every breed, that a few exceptional sires had been responsible for the vast majority of the improvement. It is very much worth while for a fancier to trace out the winning strain in his own variety. An indifferent dog bred in this strain will almost surely be a better sire than an exceptional winner in whose veins this desirable blood does not flow.
The elimination of all guess work and the willingness to accept considerable length of time before success comes should be the first resolution made by a breeder. This means study, first, of the principles of genetics; next, of the breed with which one is dealing; and lastly, of the individuals employed in every mating. Working upon a foundation of excellence in the weak points of his breed and always retaining good points gained and adding others to these, the breeder, provided he judges his breeding stock by their puppies rather than by their own points, is sure to establish a strain upon which he can count for results. This, however, cannot be done in a season. The breeder must possess those qualities we all admire in our dogs, patience, gameness, and faithfulness.