Animal Breeding And Mendel's Law
( Originally Published 1912 )
The best farm animals of today are much better suited to present needs than the most popular types of the middle of the nineteenth century. For the most part the present types serve existing demands with greater satisfaction than was experienced by stockmen working with the best animals obtainable for the conditions of some decades ago. It is also not improbable that should conditions of use and rearing of fifty years ago again become operative that some of our prized animals might readily be discarded for the types of their early progenitors. The breeders of each age and each area of country retain those animals best able to do and give what is then and there demanded of them. In some cases breeders have perpetuated and intensified tendencies and characters that seemed to be of advantage in rearing or to give added value when selling, but sometimes selling value has been obtained at the expense of true economy of production as is evidenced in numerous discussions of size and bone, particularly in swine. Low cost of production has also been offset by reduced value as evidenced by market discriminations against animals very large and growthy but coarse and rough.
What changes in animal types the market demands and farming practices of the future may necessitate can not be foretold. The limit of improvement, or more properly of adaptation to artificial requirements, lies only in the effects of selection for characters that are opposed to growth, health, or the natural exercise of powers of reproduction.
In discussions up to this point, selection with its essential accompaniments has been prescribed as the basis of progress toward any desired stand ard. This however assumes the existence somewhere of the component characteristics of the animal it is desired to produce and to multiply. If we have a starting point of even a small variation toward what is desired, the cumulative effects of selection of the fit and the rejection of the unfit will render possible the practical development of everything having an existing basis. Artificial selection accomplishes for artificial needs what natural selection adapts to natural requirements. In each case the law of the "survival of the fittest" obtains. But how should it be if there developed a need for animals with characters not found in any of their kind? "Natural selection may explain the survival of the fit-test but it cannot explain the arrival of fittest."
The first hornless pure Short-horn of which we have knowledge was not the outcome of generations of gradually declining horn growth. It appeared suddenly without apparent reason and with strength to impress its offspring with its own peculiarity. Such an unusual character appearing without intermediate stages between itself and the usual form is called a mutation and the individual exhibiting the mutation a mutant. True mutants are also referred to as sports or freaks. A horse with a very long and distinctly curly coat would be a mutant if it were certain that it was not a case of reversion. Castle has bred a strain of guinea pigs that uniformly shows four toes on each hind foot, one more than the usual number. The start consisted of one with an imperfectly developed fourth toe. None of this animal's ancestors had been known to show even any rudimentary resemblance to an extra toe. Illustrations of animal mutations of practical utility are not easy to suggest. The question of mutations has been studied quite thoroughly in plants where it seems to have greater practical possibilities. The application of the principle has been so sanguinely commended to animal breeders by persons conversant with its use in plants that a review of the matter in that field is of more than passing interest.
There had existed for some time a growing dissatisfaction with the necessity of explaining the origin of all species and varieties of plants by the very slow and gradual operation of natural selection among variations of minor degree. It was felt that some varieties had come into existence more quickly than was probable by this method. De Vries, professor of botany in the University of Amsterdam, had observed distinct changes in plants occurring spontaneously, or at least with no apparent previous tendency in the same direction. This investigator undertook to secure the double flowered character in the cultivated variety of the corn marigold (Chrysanthemum grandiflorum). This plant averaged twenty-one ray florets to a flower, while the wild form averaged only thirteen. The seeds of six plants were planted separately and in five of the six groups there was a lack of constancy to the twenty-one floret type and consequently only seeds from the sixth group were retained. In 1896 De Vries found in the progeny of one of this group a plant with two secondary heads, having twenty-two florets, the terminal heads still showing twenty-one. Succeeding generations from this plant showed great tendency toward an in-creased number of florets, forty-eight being reached in 1898 and sixty-six in 1899. Late in the same season three secondary heads were found with florets on the central' part of the flower. This was accepted as the arrival of the hoped for mutation. Plants grown from seeds from those heads showed 100 ray florets and 200 in the next generation, a completely double-headed flower. Although the increase in the number of marginal florets was gradual, it may be said that the true double-flowering character appeared spontaneously though it would seem to have been connected with the previous selection.
The cause of the origin of such a mutation is not suggested by the botanist. Manifestly it was not the product of environment. The nearest approach to a satisfactory understanding lies in the reference of such occurrences to the maze of possibilities in the combination of the chromatic elements of the reproductive cells as discussed in Chap-ter VI. In more recent years we have received accounts of the production of a new color in plants by the injection of solutions of mineral substances into developing ovules.Subsequent attempts to secure such results have proved failures. Even were such procedure practical in plants it would not be so in animals. The introduction of any new transmissible element into the hereditary material is inconceivable in view of what is known of that substance. We are again reminded of the limitations of our knowledge of the ultimate source and nature of the chromatin bodies. Procurement of mutations is wholly dependent upon chance and the capacity of the breeder to detect them. The distinction between mutations and variations is really one of degree. It will doubtless be more helpful to think of future modifications of animal form as having their beginning in minor variations occurring without design and offering opportunities to those best fitted to recognize and utilize them. Any new feature promising value, no matter how little developed, when favored by an encouraging environment and the most careful selection, may in course of time be brought up to a useful degree. However, should mutations of pronounced utility present themselves they may be utilized even if found in inferior stock, as has been done in forming our single standard breeds of polled cattle.
Each of the breeds of live stock, even of those kept for the same purpose, is characterized by special features of excellence. Of course breeders of each of the competing breeds, while retaining the admitted superiorities of their own stock, try also to secure as much as possible of the pronounced good features of their rivals. Not infrequently the crossing of established breeds is resorted to in the hope of combining the good features on both sides. Bearing in mind that every quality, desirable or undesirable, is represented in the hereditary material, regard must be had for the fact that in a cross there is no virtue that can obscure the weaknesses. These must be expected with the rest and are just as likely to be contributed from both sides as are the good traits. The crossing of breeds is sometimes favored as a means of securing new variations and new forms. The mixture of hereditary material from two dissimilar sources, each of which has been rendered pure to its usual properties and at no point allowed to receive any vestige of a taint from the other, may yield unusual results. Such procedure is quite practical among plants where many random trials are practicable even if only one in thousands yields anything of promise, and it is by crossing that Mr. Burbank has secured some of his more useful plant creations.
The making of a cross cannot be expected to originate any new character; yet the breaking up of forces among old tendencies may so balance and engage one another as to give opportunity for previously dormant and restrained possibilities to evidence themselves.
Crosses sometimes show likenesses to very remote parents, which likenesses really constitute reversions as exemplified in the case of a crossbred (Hereford-Shorthorn) bull bred to a pure Angus cow. The offspring was creamy white in color with black muzzle and black hair in the ears, features of the very early native British cattle.* In the main, however, the most that can be looked for in crossing is a fortuitous combination of existing characters such as is evidenced in the offspring of a Hereford and Angus cross which exhibits the polled head and black body together with the white face. The Oxford breed of sheep was made by selections from Hampshire and Cotswold crosses, and is the only breed of consequence that has resulted from crossing old established breeds.
The demonstration of the actual occurrence of mutations as sudden and considerable departures from the usual order of things encouraged much hopeful speculation as to changes of magnitude the breeders of the coming years might effect. It may be repeated that the hopes of accomplishment among animals was based on a somewhat over-drawn analogy between the plant and animal kingdoms, especially in economic aspects. Even more stirring than the mutation theory was the announcement at about the same time of the discovery of Mendel's law. The discovery of the operation of this law or principle was announced almost simultaneously in 1900 by De Vries, Correns and Tschermak. It was soon learned that similar work had been done and similar conclusions published in 1865 by Gregor Mendel, a monk in an Austrian monastery at Brünn. In view of the priority of Mendel's work and announcements his name is always used to designate this interesting phenomenon. Mendel's law is of special interest in conjunction with mutations as suggesting practical procedures in their perpetuation. It also throws some light on the behavior in transmission of existing characters, and is therefore worthy of careful study.
Mendel was also a botanist, and the experiments which led up to his discovery were conducted with the common sweet pea. The stockman's interest in Mendelism may be better discussed after the law has been explained in its applications to the plants with which its discoverer worked. The law as set forth by Mendel does not lend itself to terse statements and it will therefore be more satisfactory to outline the experiments in their natural order. In 1857 Mendel planned and inaugurated the experiments which at the end of eight years justified such important conclusions. Two of the varieties of peas selected represented extremes in regard to length of stems, the plants of one uniformly having stems measuring from 6 to 7 feet in length, while in the other the range was limited with equal uniformity to between 9 and 18 inches. These two varieties were crossed, and when the resulting seeds were planted the following season it was found that all of the plants had stems fully equal in length to those of the longer stemmed parents. These cross-bred plants while not hybrids in the hue sense are erroneously so designated for convenience. The chromatin representing the short stem was of course present in the hybrid plants but had not asserted itself, and the short stem character in this case is therefore termed recessive and the long-stem character dominant. Those hybrid long-stemmed plants were pollinated exclusively by pollen from plants of their own group. In the third season there were reared as the progeny of hybrid parents on both sides 1,o64 plants. Of this number about one-fourth, or 277, had the short-stem character of their grandparent which had been recessive in the parent, the remaining three-fourths, or 787, all had long stems. This was a very suggestive fact and Mendel proceeded to investigate the breeding qualities of these two groups. The flowers of each plant were fertilized exclusively by pollen from plants of their own kind. Six other similar experiments were carried on simultaneously with such characters as position of flowers, form of pods, and form of seeds.
As it was impracticable to breed from every plant reared, Too plants were selected from each of the two groups preserved from the long and short-stemmed hybrid lot and their seeds were planted the following season. The progeny of the short-stem plants exhibited short stems exclusively. Of the Too representative long-stemmed plants it was found that 28 produced only long stems, while 72 produced some of each kind. This pro-portion is approximately three to one, but applying the actual figures to the whole number it shows that the 1,o64 plants really comprised 277 capable of producing only short stems; 567 capable of producing either, and 220 capable of producing only long stems. Each of the two smaller groups continued to reproduce their own kind exclusively through succeeding generations as long as bred within their own groups. Further work with the larger and unstable group, however, showed that it behaved exactly as did the original hybrids giving off one-fourth of its number to produce only long stems, an-other fourth for short stems, and the half of hybrid character to again break up into the three kinds.
First year.—Long and short stems crossed.
Second year.—Hybrid plants raised from seed produced the first year; allowed to fertilize each other.
Third year.—Seeds produced the second year planted and 1,064 plants reared ; of these, 277 had short stems and 787 long stems ; plants of each group fertilized by their own kind.
Fourth year.—Two groups of the third year were tested; short-stem group found to produce its own kind exclusively; of the larger group 220 were found to be pure to the long type, while the seeds of the other 567 produced both long and short stems.
Fifth year.—Seeds planted from offspring of each of three groups revealed in previous season. The off-spring of the pure long and pure short groups bred true. The offspring of the 567 plants producing mixed progeny, breaking up into 142 short-stem plants and 425 with long stems. On further test those two groups prove to behave exactly like the two groups of the third year.
Knowing the breeding records of the various groups, the divisions of the original number may be arranged so as to show their relationship.
It must be borne in mind that while the plants pure to the recessive short-stem character in generation two were separable as soon as they appeared, the 220 with no tendency to short stems could be separated from the larger groups remaining mixed only by examination of the plants grown from their seed. In all instances plants were fertilized by others of their kind.
The discovery thus made was that hybrid parents produce offspring of which one-half are again hybrid while one-quarter are pure to each of the original parent forms. The figures 277, 567 and 220 are only approximately in the proportion of I, 2 and I, but the entire numbers of offspring could not be tested and the figures used represent the rather unfair proportions derived from the actual test groups. It is, not claimed that the proportions will occur with exactness except in very large numbers, although the summary of all Mendel's tests shows a very close adherence to the set proportions. From a single individual possessing a desired character or mutation that obeys Mendel's law it would be possible in time to procure a large number of others equally strong in the same character. As will readily be seen, however, progress would be much more rapid in dealing with characters that are recessive since they may be selected as soon as found in the second generation from the cross without awaiting the breeding test that is necessary to segregate pure dominants from hybrids with the dominant character. Also in practice the hybrids can be rebred to the parent and thus one-half the offspring will possess the parental character. Many characters are not Mendelian and so do not remain distinct but mix with their opposites. Mendelian or non-Mendelian characters can only be deter-mined as such by test.
At first thought the occurrence of the Mendelian proportions seems to be entirely out of line with all ordinary procedures of nature; however, a very plausible explanation is at hand. In the experiment referred to for the purpose of explaining the law it was seen that short-stemmed plants bred to long-stemmed ones produced hybrids, all with long stems. The hereditary element for the short stem was restrained from showing itself but must necessarily have been present in the germ cell material of the hybrid. It manifests itself in the production of 25 per cent of the offspring of the hybrids that contain no long-stem material, as is shown by the fact that they and all descendants remain true to the short-stem character. The hereditary material contained in the reproductive organs of a hybrid individual must contain elements for both characters;
whether each chromosome received from the long-stemmed parent contains the element for that character or whether it resides only in a single chromosome, our knowledge of those bodies will not permit us to conjecture. However that may be, the occurrences strongly suggest that each germ cell produced by a hybrid parent represents only one of the characters. If this be true, and inasmuch as the stock of hereditary material is equally supplied with both kinds, then the number of germ cells of one kind will be the same as for the other. Speaking only in regard to this single character, it will now be seen that only two kinds of ova can be produced, those with the element for short stems and those for long stems. The same is true of the spermatozoa. Any ovum that is fertilized is as likely to be of one kind as of the other. Any particular spermatozoon sharing in fertilization has also equal probabilities of having the long or the short-stem element.
In mating a female and a male hybrid four cases may arise, as shown in the diagram, Fig. 12 : Case I.—The ovum S is the one presented and is fertilized by the spermatozoon S; the progeny SS, can produce only short stems so long as bred to others like itself. Case 2.—It is equally likely that the ovum S would be fertilized by the spermatozoon L, giving LS, a hybrid progeny which like his hybrid parents might produce either kind of germ cells. Case 3.-It is equally likely that the ovum presented would be of, the L type; if so and fertilized by spermatozoon S another hybrid offspring would result. Case 4.-An L ovum joined by an L spermatozoon would produce an offspring pure to the long type. The chances for SS, purity of short stems, are the same as for the op posite, while the probability of LS, the hybrid form, is twice as great as for either of the pure forms. It is therefore not surprising that in a considerable number of cases the proportions of I, 2 and I should appear. On this basis the cause of the Mendelian proportions is apparent.
Botanists usually speak of germ cells as gametes.
This explanation of Mendelism assumes the purity of the gametes to a single character. Whether the gametes are actually pure in one character or whether they contain a predominating amount from one of the parents cannot be stated. The facts would suggest the former and that perhaps the representation of one parental character is conveyed only in a single chromosome.
Before discussing Mendelism among animals it will be of value to gain a clear idea of unit characters. It is fully established that some animal characters follow Mendel's law in transmission.Long-haired and short-haired guinea pigs mated in the experiments of Professor Castle gave progeny with the short-haired character dominant. The progeny of these hybrids consisted of twelve short-haired guinea pigs to four with long hair, the recessive character. This is the expected Mendelian proportions, 3 to i, for the first generation from hybrid stock. The twelve when tested were found to contain four that bred true to shortness of hair and eight still producing mixed progeny. The same results were apparent in crossing albinos and colored guinea pigs in the same experiment, the colored coat character being dominant over the albino condition. The two characters were studied in the same animals. Starting with long-haired albinos and short-haired pigmented or colored animals, sixteen guinea pigs were reared from the hybrid stock.
Each of the groups of twelve shown on left and right, on further test was found to be made up of four pures and eight hybrids.
The results of this experiment illustrate the meaning of the term unit characters. It is apparent that there was no relation between the length of hair and its color; each was transmitted entirely independent of the other. The length of hair is therefore one unit character, while the color of hair is another.
This experiment also illustrates the practical bearing of another point. If long coats and albinism were infrequently found and were preferred it would be possible to increase the number of individuals pure to either character' to one-fourth the whole number that the second generation might: comprise. These two characters being recessive could easily be recognized in the individuals pure to them. Should it be the dominants that were sought for, a breeding test would be necessary to isolate them from the hybrid forms. Among plants the rearing and preservation of such numbers of individuals may be at-tempted even in a commercial enterprise. Among larger animals however such practice would be difficult even in large experimental work; though should it be possible, accomplishments of considerable value might reasonably be hoped for. Animals with. characters known to be Mendelian may perhaps be handled to some satisfaction so long as it is possible to base selections completely upon the one specific character. Conditions may be imagined under which such a course would be practicable but are not likely to arise.
If it were sought to combine in the same animal two unit characters that were also Mendelian, one in sixteen of the second generation could be expected with that combination as is seen in the case of the long-haired albino guinea pig representing the fourth group shown on page 203. If a combination of dominant unit characters were desired the detection of the individual of that class would of course be more difficult.
Many features of comb and plumage of poultry have been shown to be inherited in accordance with Mendel's law. The waltzing habit in a variety of fancy mice has been shown to be a Mendelian recessive. This was found to be associated with a peculiar lack of development of the semi-circular canal of the ear which supposedly accounts for the waltzing movement. It has been suggested that the pacing gait in horses may be a Mendelian unit character. So far however this gait has not been found to be uniformly associated with any physical peculiarity and the manner of its appearance does not indicate that it can be regarded as Mendelian. Professor Spillman has presented figures which seem to show that the hornless character when appearing in horned breeds of cattle is a Mendelian recessive. Breeders of blue Andalusian fowls experience great difficulty in breeding the desired color. Birds of the desired colors when mated produce numbers of black, blue and white in proportions of I, 2 and I. The blue is regarded as the hybrid though how it splits up into the two colors while they appear to blend in development is not clear. Data has been presented strongly supporting the idea that red and white colors in Short-horn cattle are Mendelian characters.' If such be the case, however, neither can be regarded as dominant since the roan hybrid exhibits both colors.
All the animal characteristics that have been mentioned as illustrating unit characters following Mendel's law are external ones and of only secondary importance in breeding. Where any of them are desired and can be selected for without interfering with more essential features some advantage may be gained. Selection based solely on a single external character is entirely impracticable in regular breeding. Valuable mutations have been perpetuated and added to existing types in a comparatively short time by breeders who had no acquaintance with Mendel's work, as is shown in our double-standard Polled Durhams. It does not 'seem that the time required for perfection of the Polled Durhams was any greater than it would have been if the breeders had sought to utilize a knowledge of Mendelism while regarding a polled head as only one of many essentials. Some advantage would have been afforded by familiarity with the law in the certainty regarding what the results would be.
While length and color of hair are separate unit characters in guinea pigs, it would seem the length and fineness of wool fibres may be unit and Mendelian characters. This, however, is not true of the Cheviot-Leicester cross and, as was stated, the Oxford Down represents and transmits a blend of two distinct types. Probably one of the most interesting things about Mendelism is the evidence that the animal body represents a vast number of unit characters which may or may not be transmitted in definite proportions. Investigators have not attempted to determine unit characters of the body, though it has been suggested that the width of hindquarters of beef cattle is a unit character.* It has occurred to the author that the size of the Percheron foot may be transmitted as a dominant unit character and also that peculiar "typey" head of the Berkshire may be a Mendelian unit character. No data have been collected on these points. Additions to our knowledge of what are unit characters of the body are not likely to come through planned experiments. They are more likely to come through retrospective study of records of regular breeders who record and preserve full data regarding the individual peculiarities of each animal reared or used in breeding. Mendel-ism also suggests the value of careful inquiry regarding individuality of ancestors, to guard against the existence of undesirable characters that cannot be recognized be-cause recessive but which might reappear.
Although the results of Mendel's work have been freely spoken of as promising to revolutionize practical breeding it now seems that the matter is still chiefly of scientific interest. The breeder may aid the scientist by preservation of records that will facilitate this study and in turn the scientist may then be able to make suggestions of possible application in breeding.