# Popular Science Monthly/Volume 67/July 1905/Recent Discoveries in Heredity and their Bearing on Animal Breeding

THE

POPULAR SCIENCE

MONTHLY

JULY, 1905.

 RECENT DISCOVERIES IN HEREDITY AND THEIR BEARING ON ANIMAL BREEDING.[1]
By Professor W. E. CASTLE,

HARVARD UNIVERSITY, CAMBRIDGE, MASS.

EVERY breeder of animals is familiar with the great complexity of hereditary processes. He knows that characters of the most varied sorts are inherited. These relate not only to general size and proportions, but also to the structure of individual parts; and not merely to structural, but to functional peculiarities as well. Thus, in certain races or strains of animals, we find inherited great fecundity, or early maturity, or ability to put on fat, or to produce abundant milk; in other cases, speed, keen scent, fierce or gentle disposition, and numberless other characteristics are plainly inherited. Very rarely are any two heritable traits necessarily associated. The cow with a good flow of milk may or may not be gentle; the keen-scented dog may or may not be speedy. Accordingly, we must conclude that different hereditary characters are inherited independently of one another, and are probably represented by different structural elements in the sexual element or germ. We know, further, that the laws of transmission of different characters are different, so that we can not estimate the force of heredity in the lump, but must fix our attention on one character at a time if we wish to analyze the complex processes in operation.

Francis Galton (1889) was the first to recognize that in the case of certain characters the result of inheritance is a blend of the conditions found in the two parents, while in other characters inheritance is alternative between the conditions found in the parents.

Fig. 1. A Brown-coated, Lop-eared Rabbit.

Fig. 2. An Albino, Short-eared Angora Rabbit.

A good illustration of blending inheritance is found among rabbits which differ in size of ear. Lop-eared rabbits have ears two or three times as long and as wide as those of ordinary rabbits. (Compare Figs. 1 and 2.) A cross between lop-eared rabbits and ordinary ones produces offspring with ears of intermediate size, which sometimes stand erect and sometimes lop. (See Fig. 3.) The ear-characters which were so distinct in the parents have in this case lost their identity in the offspring, and apparently can not be recovered again in their original condition, for the offspring transmit to their young the blended character, rather than the extreme conditions found in their respective parents.

It has been thought until quite recently that hereditary processes in general were of this sort and that any result other than a blend was exceptional. But recent investigations do not bear out this idea.

Fig. 3. A Black-coated, Half-lop Rabbit, son of the two rabbits shown in Figs. 1 and 2, respectively.

Alternative inheritance is illustrated in a cross between the so-called Belgian hare and an albino rabbit. The Belgian hare is simply a gray-coated variety of the European rabbit, while albino rabbits are pink-eyed animals of the same species and have white hair; the Belgian is pigmented like the wild European rabbit, the albino is essentially unpigmented. A cross between the two produces offspring all of which have the pigmented or Belgian coat, none being albinos. Compare Fig. 3; in this case the parents were an albino and a brown pigmented animal, respectively. The young, nine in number, were all black pigmented, like the one shown in Fig. 3.

The effect of crossing a pigmented rabbit with an albino is similar to that produced when two pieces of glass, one transparent, the other opaque, are held up together. We see only the opaque one. Nevertheless, the two conditions have not blended; each retains its original distinctness, and the two can be separated again at will. So it is in the Belgian produced by cross-breeding with an albino. The albino character is there, though unseen, and will appear as a distinct entity when the cross-bred reproduces, for it will be represented in approximately half of the sex-cells formed by the cross-bred animal, the alternative or Belgian character being represented in the other half. It is as if the two pieces of glass, combined originally to illustrate the formation of a cross-bred animal, were separated again to illustrate the formation of the reproductive elements by the cross-bred. For every element formed having the opaque character, there will be another

Fig. 4. A Guinea-pig with Long, Rough, Albino Coat.

having the transparent character, but there will be no elements of an intermediate nature.

This simple principle, that in alternative inheritance sex-cells of two sorts are formed by cross-bred individuals, constitutes one of the most important discoveries ever made in the study of heredity. The discovery was made about forty years ago by an Austrian monk, Gregor Mendel, who was engaged in the study of cross-bred garden peas. It, however, attracted little attention at the time and was soon forgotten. Meanwhile, a great body of workers was studying with great minuteness the material basis of heredity, the sexual elements. Their investigations disclosed in the cell a complete basis for just this kind of alternative inheritance and led up to the rediscovery of Mendel's law simultaneously by several different experimental breeders, foremost among whom was the Dutch botanist, de Vries. Mendel found that in cross-breeding between alternative characters, one uniformly dominates in the offspring from its very nature, while the other disappears. Just as, when the two pieces of glass are held up together we see only the opaque one, the transparent one being invisible. Mendel called the character seen in the offspring dominant, the unseen one he called recessive. In rabbits, the gray pigmented or Belgian hare coat is dominant over the albino coat, the latter being recessive (unseen) in cross-bred animals. Similarly, in mice, guinea pigs, and even in man, mating of an albino with a pure, pigmented individual produces only pigmented offspring. In guinea-pigs the rosetted or rough coat is dominant over smooth (normal) coat (see

Fig. 5. A Guinea-pig with Short, Smooth, Pigmented Coat.

Fig. 6); and short coat is dominant over long (or angora) coat (Fig. 8). In rabbits, also, the normal or short coat dominates over the angora coat (Fig. 3), and the same is probably true in cats and goats as well.

Among guinea-pigs there occurs a series of alternative pigment types which show Mendelian relations one to another. If we write them in this order, (1) agouti (i. e., black ticked with yellow, the ancestral or wild type of coat), (2) black, (3) yellow, (4) albino, we may say that each is dominant over all which follow it, and recessive in relation to all which precede it. Thus agouti mated with black, yellow or albino gives only agouti offspring; black mated with yellow or albino gives either black or agouti, but never yellow or albino, while yellow dominates over the albino only. In man, a condition of hypophylangia (two-jointed instead of three-jointed digits) is dominant over the normal condition. In mice, the peculiar waltzing habit of so-

Fig. 6. A Guinea-pig with Short, Pigmented Coat. Its parents were the animal shown in Fig. 4 and one like that shown in Fig. 5.

Fig. 7. A Guinea-pig with Long, Smooth, Albino Coat. These three characters are all recessive in nature.

called Japanese mice is a recessive character in heredity. In man, a peculiar dark-colored condition of the urine, known as alkaptonuria, is inherited as a Mendelian recessive character. Many other illustrations might be given, but these will, perhaps, suffice to show that Mendelian

Fig. 8. A. Short-haired, Smooth, Pigmented Guinea-pig, illustrating the result of mating two animals like those shown in Figs. 5 and 7 respectively.

Fig. 9. A Short-haired, Smooth, Albino Guinea-pig.

or alternative inheritance is neither a rare nor an exceptional phenomenon, and that it applies to the inheritance not only of characters purely structural, but also to those which are physiological.

From the facts that cross-bred animals form sexual elements (or gametes) of two sorts, and that the two sorts are equally numerous, it follows that among their offspring dominant and recessive individuals will occur in definite proportions. It has been found by experiment that when two cross-bred (or hybrid) dominant animals are mated together, the offspring consist of a mixture of dominant with recessive individuals in approximately the proportions, three dominant to one recessive. Further, when a hybrid dominant is mated with a recessive animal, half the offspring are hybrid dominants, half recessives. These proportions of necessity result, provided neither sort of gamete has greater affinity for one kind than for the other. For consider all the possible unions between two sets of gametes each ${\displaystyle D}$ and ${\displaystyle R}$ (the case in which two hybrid dominants are bred together):

 ${\displaystyle {\begin{array}{rcl}D&{\text{and}}&R\end{array}}}$ ⋮⋱⁠⋰⋮ ⋮⁠⋱⋰⁠⋮ ⋮⁠⋰⋱⁠⋮ ⋮⋰⁠⋱⋮ ${\displaystyle {\begin{array}{rcl}D&{\text{and}}&R\end{array}}}$ ${\displaystyle \underbrace {DD+2D(R)} _{3}+{\underset {1}{RR}}}$

They are 1 ${\displaystyle DD}$, 2 ${\displaystyle DR}$ and 1 ${\displaystyle RR}$, or three unions involving the ${\displaystyle D}$ character to one involving the ${\displaystyle R}$ character only; hence three dominant individuals will be produced to one recessive. Further, one of the three dominants (${\displaystyle DD}$) will be pure, while the other two are hybrid in character. Recessive individuals are, however, necessarily pure and breed true inter se. Thus, smooth-coated guinea-pigs produce only smooth-coated offspring, albinos only albino offspring, and long-haired ones only long-haired offspring, when mated to animals like themselves. The reason for this will be clear if we return to the illustration with the glass plates. Pairs of transparent plates can be separated only into transparent pieces, and these can be recombined only into transparent pairs.

With the dominant offspring of hybrids, however, the case is different. Only pure dominants (${\displaystyle DD}$) will breed true when mated inter se; hybrid dominants (${\displaystyle DR}$) will continue to give a mixed progeny. For pairs of opaque plates, when separated, can be recombined only into opaque pairs; but pairs composed each of an opaque and a transparent plate, when separated, may be recombined either into opaque pairs or into transparent pairs, the chance frequencies of the two sorts of combination being as three to one.

Accordingly, any pair of recessive individuals may form the beginning of a pure race of recessives, but in starting a pure race of dominants we must test each animal used, to make sure that it does not contain, unseen, the recessive character. Otherwise we may keep getting mixed lots of offspring. The simplest and surest way of making such a test is to mate the dominant animal with a recessive. For in that case, if the dominant is pure, all the offspring will be dominants; but if the dominant parent is a hybrid, half the offspring will be recessives, half dominants. When the purity of two dominant animals of opposite sex has once been established by breeding-test, we may use them as the starting point of a race of dominants which we may be sure will breed true.

We must not, however, fall into the error of supposing that any pair of dominants which produces only dominant offspring is, therefore, pure. For progeny of this sort will be obtained if only one of the parents is pure, the other being hybrid. In starting a race of dominants which will breed true we must test each animal individually, by mating it, preferably with a recessive, or else with a dominant known to be hybrid in character. A test of the former sort should, as stated, give 50 per cent, of recessive individuals if the dominant is impure; a test of the latter sort should give 25 per cent, of recessives, if the dominant is impure. Either sort of test should give only dominant offspring if the dominant tested is pure.

The statement has already been made that many characters are independent of one another-in heredity; T hope now to demonstrate the correctness of this idea m cases of alternative inheritance, even when the independent characters relate to the same bodily parts. For this purpose the coat-characters of guinea-pigs and rabbits are well adapted, since they are exterior structures easily studied in the living animal. I hope to show first that pigmentation of the hair is inherited quite independently of its length, and secondly that hair-arrangement (in smooth or rough coat) is inherited quite independently of both pigmentation and length of hair.

When an ordinary short-haired guinea-pig (Fig. 5) is mated with a long-haired albino guinea-pig (Fig. 7), all the young produced are short-haired and pigmented, these being the dominant characters. (See Fig. 8.) Xow if the cross-bred young are bred together, offspring of four different sorts are produced. Two of the four sorts are identical with the grandparents in character; they are short-haired pigmented animals (Fig. 5) and long-haired albinos (Fig. 7), respectively. But the other two sorts represent new combinations of characters; they are short-haired albinos (Fig. 9), and long-haired pigmented animals (Fig. 10). Further, these four sorts of individuals occur on the average in definite numerical proportions, viz:

 9 short-haired pigmented animals, 3 short-haired albino animals, 3 long-haired pigmented animals, and 1 long-haired albino.

Considering pigmentation and hair-length separately, we see, first, that there are 12 pigmented animals to 4 albinos, or 3 to 1, as expected; and, secondly, that there are 12 short-haired to 4 long-haired animals, again 3 dominants to 1 recessive. But if we consider the relation of each pair of characters to the other, we find absolutely no correlation between them. Albinism may or may not be associated with long hair, and pigmented coat may or may not be associated with short coat in the offspring, though they were so associated in the grandparents. As a matter of fact, when the animals are tested one by one, to determine the presence of recessive characters, we find that albinism, visibly present in 4 out of 16 offspring, is present recessive in 1 others, and that in half

Fig. 10. A Long-haired, Smooth, Pigmented, Guinea-pig.

of these cases it is associated with short coat, while in the other half it is associated with long coat.

In another experiment which I have performed with guinea-pigs, a cross was made involving three pairs of alternative coat-characters, length, pigmentation and roughness of coat. A long-haired rough albino (Fig. 4) was mated with short-haired smooth pigmented animals (Fig. 5). The young were all short-haired, rough and pigmented (Fig. 6). The coat-characters seen in these offspring are the three dominant characters, two of which were received from one parent, one from the other; the three alternative recessive characters are present but unseen.

When the young were bred together, they produced offspring of eight different sorts, including all possible combinations of the three pairs of alternative characters.

One large class was like the parents, short-haired, rough and pigmented. Two other classes were like the grand-parents, viz, shorthaired smooth pigmented, and long-haired rough albino. In addition, there were five other new classes not represented among the parents or grandparents. These were: short-haired rough albino (Fig. 11), short-haired smooth albino (Fig. 9), long-haired smooth albino (Fig. 7), long-haired smooth pigmented (Fig. 10), and long-haired rough pigmented (Fig. 12).

The eight classes of young produced in this experiment are not all equally numerous. The largest class is that which contains the three dominant characters (Fig. 6), the smallest that which contains the

Fig. 11. A Short-haired, Rough, Albino Guinea-pig.

three recessive characters (Fig. 7). Theoretically, they should number 27 individuals and 1 individual, respectively, in a total of 6-1 young. These proportions are roughly approximated in the observed result.

This experiment illustrates two important principles in heredity: First, if as regards the hair alone there exists such a variety of characters separately heritable, how great must be the number of such characters in the body as a whole, and how remote the probability that any animal will in all characters resemble any individual ancestor, provided that in a considerable number of heritable characters a choice is offered between 1 alternative conditions. Secondly, the experiment shows how a variety of new organic forms may quickly be produced by cross-breeding, leading to the combination in one race of characters previously found separately in different races. Thus, in guinea-pigs, one can obtain within two generations any desired combination of the three pairs of alternative coat-characters, if one produces

F:g. 12. A Long-haired, Rough, Pigmented Guinea-pig.

Fig 13. An Imperfectly Long-haired Guinea-pig. Compare Figs. 4, 7 and 12.

a sufficiently large number of individuals; but to obtain the desired combination in individuals which will breed true is not so simple a matter. If the desired combination consists wholly of recessive characters, any pair of individuals manifesting that combination will breed true. But if the desired combination contains one or more dominant characters, then each animal selected must be tested for the presence of undesirable recessives before one can be sure that the new race will breed true. In practise it is found best by the breeder not to work with too many characters at a time, but to eliminate the undesirable recessives one by one. Otherwise the search for the one individual in a large number which will breed true may prove a long and tedious process. If we deal with one character at a time, the chances are that one in four of the second generation of animals reared will meet our ideal; if we deal with two characters at a time, the chances are one in sixteen; while if we deal with three characters at a time the chances are only one in sixty-four; and so on, with the chances of success diminishing in a geometrical series.

From what has thus far been said it would appear that in alternative inheritance characters behave as units, and, more than that, as wholly independent units, so that to forecast the outcome of matings is merely a matter of mathematics. While this is in a measure true, it is, fortunately or unfortunately, not the whole truth. In alternative inheritance characters do behave as units independent of one another, but the union of dominant character with recessive in a cross-bred animal is not so simple a process as putting together two pieces of glass, nor is their segregation at the formation of gametes so complete in many cases ae: the separation of the two glass plates. The union of maternal and paternal substance in the germ-cells of the cross-bred animal is evidently a fairly intimate one, and the segregation which they undergo when the sexual elements are formed is more like cutting apart two kinds of differently colored wax fused in adjacent layers of a common lump. Work carefully as we will, traces of one layer are almost certain to be included in the other, so that while the two strata retain their identity, each is slightly modified b} r their previous union in a common lump.

Thus, when we cross short-haired with long-haired guinea-pigs, we get among the second-generation offspring a certain number of long-haired animals with hair less long than that of the long-haired grandparent, or with long hair on part of the body only (Fig. 13). Further, certain of the short-haired animals have hair a little longer and a little softer than that of the short-haired grandparent. Again, rough-coated guinea-pigs produced by cross-breds often have coats less fully rough than that of their rough ancestor, lacking certain of the typical rosettes (Fig. 14). Finally, when an albino is crossed with a fully pigmented animal, the result may be not a wholly pigmented animal, but one spotted with white. While such a cross-bred animal forms a full quota (one-half) of albino gametes, the pigment-bearing gametes formed by it frequently bear this spotted or modified pigmented condition.

Cross-breeding, accordingly, is a two-edged sword which must be handled carefully. It can be used by the breeder to combine in one race characters found separately in different races, but care must be exercised if it is desired to keep those characters unmodified. If modification of characters is desired at the same time as new combinations, then cross-breeding becomes doubly advantageous, for it is a means of inducing variability in characters, as, for example, in the intensity of pigmentation and in the length of hair, quite apart from the formation of new groupings of characters. Sometimes it causes a complex character to break up into simpler units, as the agouti coat of the wild guinea-pig into segregated black and yellow, or total pigmentation into

Fig. 14. An Imperfectly Rough-coated Guinea-Pig. Compare Fig. 6.

a definite series of pigmented spots. In other cases it operates by bringing into activity characters which have previously been latent in one or other of the parental forms. Compare Fig. 3 with Figs. 1 and 2. Black pigmentation was latent in the albino parent (Fig. 2) and was brought into full activity by a cross with a brown-pigmented animal.

Now, what bearing, we may ask, have these theoretical matters on the practical work of the breeder? They show (1) that a race of animals is for practical purposes a group of characters separately heritable, and (2) that the breeder who desires in any way to modify a character found in this group, or to add a new character to the group, should first consider carefully how the character in question is inherited.

If the character is alternative in heredity to some other character, cross-breeding between the two, followed by selection for pure individuals, will within two generations give the desired combination of characters in individuals which will breed true. This process of selection is simplest when the characters to be combined are recessive in nature, but individual breeding-tests become necessary when dominant characters are included in the combination desired.

If a character gives blending inheritance, it must be treated in a different way. Suppose, for example, that we desire to combine lop-ears in rabbits with albinism, but that our lop-eared stock consists wholly of pigmented animals. How shall we proceed? First, mate a pigmented lop-(Fig. 1) with a short-eared albino (Fig. 2). The offspring will be pigmented half-lops (Fig. 3). If two of these be bred together their young will all be half-lops, and about one in four of them will be albinos. Now these albino half-lops may be mated with pure pigmented lops. The young will again all be pigmented, but will this time be three-quarter lops, and by breeding these together albino three-quarter lops may be obtained in the next generation. By continuing this process of back-crossing with the lop-eared stock, and selecting the albino offspring obtained, the lop-eared character may be steadily improved in the albinos until it is practically as good as in the original lop-eared stock. The rate of improvement possible can be readily calculated. The albino young will be:

 After 2 generations, one half lops, After 4 generations, three fourths lops, After 6 generations, seven eighths lops, After 8 generations, fifteen sixteenths lops, After 10 generations, thirty-one thirty-seconds lops, etc.

This will be the result on the hypothesis that no secondary variation occurs in the lop-eared character. If, however, variation is induced by the cross-breeding, then it is possible that the desired end may be reached sooner, or that an even better lop may be obtained in the albino cross-breds, than that of the original pigmented stock.

Latent characters are an important element in practical breeding. Sometimes they greatly aid the breeder's work; sometimes they impede it. If a stock contains undesirable latent characters which are brought into activity by cross-breeding, these latent characters will have to be eliminated, or a new stock tried.

Since cross-breeding is likely to modify characters even when these conform to the laws of alternative inheritance, and is certain to modify them when they give blended inheritance, it should be practised with extreme caution, and only by the breeder who has a definite end in view and a fairly clear idea of how he is going to attain it.

The purity of standard breeds should be carefully guarded, and much attention should be given to pedigrees; for even when individual excellence is not apparent, it may be present in a recessive or else in a latent state, which suitable matings will bring into full realization, provided the ancestors were superior animals.

At the same time the breeder should be on the lookout for individual peculiarities of merit. And he should not be discouraged if these are not transmitted to the immediate offspring. Ordinarily a desirable character which disappears from the children, but reappears among the grandchildren, can at once be made a racial character, for it is recessive in heredity.

Inbreeding is not invariably an evil. It is often necessary to cause the reappearance of a vanished recessive character, and is indispensable in the formation of races which will breed true. Two or three generations of close inbreeding usually suffice to realize the practical benefits of the process, if intelligently carried on. The inbreeding should then be discontinued as soon as the desired end has been attained. Otherwise, loss of vigor or infertility may result.

1. Published by permission of the Carnegie Institution of Washington, to whose officers the author is deeply indebted for aid in the prosecution of his studies, and in particular for the loan of figures 4–14 of this article.