Popular Science Monthly/Volume 84/January 1914/The Mechanism of Heredity as Indicated by the Inheritance of Linked Characters
|THE MECHANISM OF HEREDITY AS INDICATED BY THE INHERITANCE OF LINKED CHARACTERS|
By Professor T. H. MORGAN,
IT is generally recognized to-day that the central idea of Mendel's discovery in regard to heredity is that when two contrasting elements enter a hybrid, one from each parent, they separate in the germ cells of the hybrid, so that the germ-cells are pure like those of the original parents in regard to each element. Chance meetings of the germ cells give the ratios that are characteristic of Mendelian heredity. This is illustrated by the example that Mendel gave.
When a pea having green seeds is crossed with a pea having yellow seeds a hybrid pea is produced. When the germ-cells of the hybrid are ripened, each ovule carries either the element for green or that for yellow, but never both. Yellow and green have separated. The same separation occurs in the formation of the pollen. If self-fertilization now takes place chance combinations of the yellow-or green-bearing ovules with the yellow or green-bearing pollen give one pure yellow pea, two hybrid (yellow) peas, one pure green pea, as shown in Fig. 1. Mendel discovered that the same principle holds when two pairs of characters are involved. If one of the original parents had round seeds and the other wrinkled seeds, these characters separate independently of the green-yellow separation. We say technically that the pair yellow and green, and the pair round and wrinkled segregate independently of each other. Chance combinations of the germ-cells in a double hybrid of this sort give nine yellow round peas, three yellow wrinkled peas, three green round peas and one green wrinkled pea (Fig. 2).
The same rule applies to three or more pairs of characters. Mendel assumed in fact that independent assortment always takes place no matter how many characters are involved.
In more recent times evidence has been accumulating which shows that the chromosomes are the bearers of most of the elements (factors) that produce those characters that we study in heredity. I can not take up the work that seems to me to place this hypothesis on a very probable basis, but I shall simply assume that it is a reasonable conclusion from the evidence at hand.
Fig. 1. Diagram to illustrate Mendel's law of segregation. Individuals (zygotes) are represented by superimposed circles, whose colors stand for the factors involved. Gametes (germ-cells) are represented by single circles.
Fig. 2. Diagram to illustrate inheritance of two pairs of Mendelian characters, viz., yellow-green and round-wrinkled peas. The 16 small squares represent the composition of the 16 F2 peas; viz., 9 YR, 3 YW, 3 GR, 1 GW.
Fig. 3. Diagram to show the pairs of chromosomes (a, c) and their behavior at the time of maturation of the egg. Three pairs of chromosomes are represented; three from one parent, three from the other. The six possible modes of separation of these three are shown in the lowest line.
gets one of each kind of chromosome from one parent and one of each kind from the other parent. At the maturation of the germ-cells the maternal chromosomes and the paternal chromosomes come together in pairs and appear to fuse (Fig. 3, d, e). They then separate, and after two peculiar divisions one chromosome of each pair goes into the egg and one into the polar body. Similarly for the sperm. There is no evidence that all the maternal chromosomes go to one germ-cell, and all the paternal to the other, more frequently than chance assortment calls for, and we are free to suppose that a random assortment of chromosomes takes place, so that each egg and each sperm gets one of each kind regardless of its parental origin (Fig. 3, lower line).
With the acceptation of this view it appears at first sight that in a given race there can not be more independent pairs of characters that show assortment than there are pairs of chromosomes. Since the number of chromosomes is fairly limited it might appear that we could reasonably expect to make still more probable the chromosome hypothesis by finding that the number of independent pairs of Mendelian characters is not greater than the known number of chromosomes in a given race; or else we might expect to disprove the chromosome hypothesis by showing that the number of independent pairs of characters does transcend the number of the chromosomes. As a matter of fact, several strains are known in which the number of Mendelian characters is greater than the number of chromosomes; but just here a remarkable phenomenon has come to light that shows in certain cases that many of the characters do not segregate independently, but are linked to each other as though they belonged to some common system. As a result, peculiar ratios appear, that differ from the expected Mendelian ratios in so 'far as that expectation rests on independent assortment. The question here raised has, therefore, taken on a new aspect, and it has become essential to discover whether there are as many, or more, or fewer groups of linked characters than there are kinds of chromosomes.
Correns appears to have been the first to call attention to a case in which peculiar ratios appeared which he attributed to coupling. Bateson and his coworkers have described several instances of the same kind. They have found that certain characters in sweet peas do not fulfill the expectation for independent assortment of different pairs of characters, although they do show Mendelian segregation when each pair is taken separately. This phenomenon has been described as coupling or repulsion. I shall refer to it as linkage.
It was found, for instance, in sweet peas that when plants with blue flowers and long pollen grains were crossed to plants having red flowers and short pollen grains in the grandchildren, the blues had for the most part long pollen and the reds short pollen. Again when blue and red flowers with erect and hooded standards were used all the red grandchildren had erect standards.
We have met with these same phenomena at nearly every step in our studies of heredity in the fruit fly, Drosophila. Over one hundred mutants have appeared from many of which pure races have been formed. At present we have studied fifty-nine of these sufficiently to show that they fall into three great groups.
The characters in the first group show sex-linked inheritance. They follow the sex chromosomes. The second group is less extensive. Since the characters in this group are linked to each other we say that they lie in a second chromosome. The characters of the third group have not as yet been so fully studied, except to show that they are linked. We place them in the third chromosome without any pretensions as to which of the pairs of chromosomes are numbered II. and III.
The arrangement of these characters in groups is based on a general fact in regard to their behavior in heredity, viz., A member of any group shows linkage with all other members of that group, but shows independent assortment with any member of any other group. In Drosophila ampelophila there are five pairs of chromosomes. According to Stevens, the sex chromosomes are attached to one of the other pairs. Three of the five pairs are occupied according to our view by the three groups of linked factors that we have studied. There are as yet two more pairs of chromosomes than there are groups of linked factors.
On the chromosome hypothesis we can readily see that if the factors that stand for characters lie in the chromosomes, those that lie in the different chromosomes should give independent assortment, and the ratios obtained in breeding experiments should be the expected Mendelian ratios. On the other hand, it may appear that the factors that lie in the same chromosomes should always march together through successive generations. If this were true the linked characters would be absolutely linked to each other. Experience shows, however, that the linked characters are not absolutely linked, but that to a greater or a less degree, according to the factors involved, interchanges must in some way take place. Here fortunately there is a cytological relation that may be utilized to explain how interchanges between like chromosomes may take place.
It has been observed when the homologous pairs of chromosomes unite before maturation of the egg and sperm that they twist around each other. In consequence, parts of each chromosome may come to lie on one side of the twist and other parts on the other side. If at times the chromosomes break at the crossing point, and each then unites with that part of the other chromosome that lies on the same side, the new chromosomes that emerge later from the pair will be made up of parts of each chromosome to the extent to which breaking has taken place at some of the crossed levels.
In consequence the two new chromosomes are no longer made up of the same parts as the original chromosomes, but of pieces of both. If we think of all the factors that lie in one chromosome as linked because ordinarily they go together, in the sense that they are likely to remain in that chromosome, this linkage will be disturbed, or broken, at one time only in the history of the chromosomes, viz., at the time of conjugation of the pairs, when an interchange between the members of an homologous pair becomes possible.
Let me illustrate by means of two concrete cases, and by preference cases that belong to the sex chromosomes, because the conditions here are simpler and more convincing, and because we have more definite information concerning the mode of distribution of these chromosomes than of any other.
When a male fruit fly with yellow body color and white eyes is mated
Fig. 4. Diagram illustrating the results of crossing a yellow (stippled), white eyed male to a gray, red-eyed female of Drosophila. To the right the sex chromosomes are represented, colored in the same way as the flies.
to a gray (wild) female with red eyes, the daughters are gray with red eyes, and the sons also (Fig. 4). The explanation of this result is shown in the diagram in which the sex chromosomes are represented by X's which are marked like the characters they stand for. The daughter gets a "red-gray" chromosome from the mother and a "white-yellow" chromosome from the father. Since red and gray dominate the color is determined by these factors. The son gets his single sex chromosome from his mother which carries the factors for gray body color and red eyes.
When the hybrid (F1) flies are inbred they produce one kind of female and four kinds of males as shown in the next figure. In order to understand how these classes arise let us follow the history of the sex chromosomes.
The F1 female had two kinds of sex chromosomes, that we may call briefly WX and RX. Either may pass out into the polar body leaving the other chromosome in the egg. Consequently there are two kinds of eggs. The F x male has only one sex chromosome GRX which goes into the female-producing spermatozoon. The other, the male-producing spermatozoon, does not carry an X chromosome. When the female producing sperm fertilizes either kind of egg it brings in the two dominant factors GR; hence all the females are gray in body color and have red eyes. Since the male-producing sperm does not bring into the egg
Fig. 5. The two upper lines represent the two sex chromosomes of the female, showing their original composition, viz., GR and YW. The lower line shows "crossing-over" between the two pairs. The numerals to the left give the F2 males, whose composition is the same as that of the chromosome that stands next to them. The males receive one or the other of these chromosomes from their mother.
any X chromosome, the sex-linked factors in that egg are those of the egg itself. Since without interchange between the sex chromosomes there are two kinds of eggs there will be two kinds of males, namely, gray-white and yellow-red.
But the experiment shows that there are four kinds of males. Their origin can be explained if we assume that in some of the eggs the sex chromosomes prior to the extrusion of the polar bodies have crossed, and the parts have reunited at some point between the factors in question. This is shown in the next diagram (Fig. 5) where the crossing over is represented. The number of times that this occurs will be measured by the number of the two other classes of males, the gray-red and the yellow-white. The actual results are:
Expressed in percentages the crossing over takes place in 1.3 per cent, of the total number of males produced.
The hypothesis of independent assortment for two pairs of characters calls for equal numbers in each of the four classes of males in the cross just given. The numbers show how far the actual results depart from this expectation.
There is one further point here that demands consideration. If the factors lie in a linear order in the chromosome as the hypothesis requires it is evident that the nearer together two factors lie the smaller will be the chance that a twist occurs between them. Consequently the frequency of crossing over can be taken as a measure of the distance of the factors from each other in the chromosome. On this basis the position of these factors in the chromosomes has been calculated. I shall return to this point later.
When three pairs of sex-linked characters are involved the result is essentially the same, but the possibility of another class of individuals, viz., those produced by double crossing over offers certain relations of peculiar interest. If a female fly with the characters yellow body color, white eyes and miniature wings is mated to a wild male with gray body color, red eyes, and long wings the daughters are like the father and the
Fig. 6. Diagram illustrating the, two sex chromosomes in the F x female derived from a YWM female and a GRL male. The first (upper) pair of lines show the chromosomes without crossing over. To the left the number of the F2 flies of the composition of the two chromosomes respectively are represented. The second pair of lines illustrate the results of "single crossing-over" between YW and GR. The third pair of lines represent the "single crossing-over" between WM and RL. The fourth pair of lines represent the results of "double crossing-over."
sons are like the mother. The explanation is here the same as in the last case. These inbred give the following significant classes:
The diagram, Fig. 6, shows the imagined relation of the chromosomes to each other. There are two classes, YWM, GRL, representing "non-crossing-over." There are four classes, YRL, GWM, YWL, GRM, where "single crossing-over" is expected. There are also two classes, YRM, GWL, representing "double crossing-over."
Other experiments show that the factor for miniature gives a high percentage (about 34) of cross-overs with yellow or with white. We place it at a correspondingly distant point which means that the likelihood of a twist occurring between the loci for miniature and that for yellow (or for white) is large. Other experiments show that the factor for yellow body color and that for white eyes rarely cross over. We place them therefore near together. The chance that a twist should occur between Y and W and at the same time another twist between W and M is very small. It is to be anticipated that the double crossingover would be rare, as it is, in fact.
I have given the argument on which we base our scheme of the linear order of the factors in the chromosomes. I need hardly say that there is no pretension that the distances (calculated in per cent.) correspond to real distances, for we know nothing of the actual space occupied. But the numbers give the relative positions of the loci in the chromosomes. The procedure justifies itself in one all-important respect. By its means we can calculate results before they are tried, and experience shows that the prediction comes true. For example, if we know the location of a factor Y and of another factor W, then when a new factor, M, appears we need only determine its position in regard to Y and we can predict what will happen when a cross is made between M and W. In a word, we can by determining the position of a new factor in regard to any other known factor calculate the results for all other known factors in the same chromosome. When we recall the wide departure, due to linkage, from the accepted Mendelian ratios based on random assortment it is no small gain to be able to calculate the results of all possible combinations by determining two known points. I make this statement even though we may at any time find that linkage is influenced by the environment, or by characteristics peculiar to individuals or to pure strains. Nothing would be more harmful at this stage than that the situation be prejudiced by absolute statements.
If any one objects to locating these points in chromosomes and prefers to treat biological problems in terms of mathematics he can make the same predictions from the data that can be treated without regard to the mechanism of the chromosomes. But since we find in the chromosomes all the machinery actually at hand for carrying out this procedure, it seems to me reasonable to base our conceptions on this mechanism until another is forthcoming. And if it should prove true that we have found the actual mechanism in the organism that accounts for segregation, assortment and linkage of hereditary factors we have made a distinct advance in our study of the constitution of the germ plasm.
It has been pointed out to me, more than once, that the views here presented concerning the "architecture" of the chromosome are similar to the views (assumed to be discredited) that Weismann advanced several years ago. But it should not be overlooked that Weismann's purpose in locating his determinants in the chromosomes was only that he might separate them again during development. He tried, in fact, to explain development in this way without, however, explaining what determines during development the orderly disintegration of the chromosomes. Nothing of the sort is postulated, or implied, on my view. Weismann's hypothesis was purely speculative. My own conception of the constitution of the chromosomes rests on numerical data obtained from hereditary characters. All of the chromosomes are supposed to go intact to every cell of the body as observation, so far as it goes, shows to be the case. How differentiation takes place is a question quite remote from the idea of the architecture of the chromosomes in their relation to hereditary characters.
There is but one fundamental similarity between my own view and that of Weismann. The chromosomes, looked upon as the vehicles of heredity, are assumed by both of us to have definite structures and not to be simply bags filled with a homogeneous fluid. The discrete parts (factors) of these structures are supposed to influence the course of differentiation, but there the resemblance ends. A factor, as I conceive it, is some minute particle of the chromosome whose presence in the cell influences the physiological processes that go on in the cell. Such a factor is supposed to be one element only in producing characters of the body. All the rest of the cell or much of it (including the inherited cytoplasm) may take part in producing the characters. So far as such things as unit characters exist I look upon them merely as the most conspicuous result of the activity of some part of the chromosome. A single factor may affect all parts of the body visibly, or a factor may preponderantly influence only a limited section of the body. As a matter of fact, if we look carefully, we can generally find farreaching effects of single factors. On the other hand, Weismann's idea of development emphasizes the intimate relation between his determinant and a specific character of the body. His writings often leave the impression that he supposes the determinants of the chromosomes to pass out into the cell, multiply there, and become the differentiated part. Perhaps this is only due to his attempt to visualize his conception, and he might grant that the differentiation of the determinants may depend on the interaction of many parts of the cell. But if we take his view literally to mean that the determinants are the materials out of which specific structures are directly built up, then his conception of the nature of a determinant is widely different from my own concerning the relation of "factors" and body characters.
- Mendel speaks of characters as forming pairs. To-day we speak of factors or genes as the paired, elements (allelomorphs) in the germ-cells, and these are supposed to act as differentials in producing the characters in the adult animal or plant. The English school considers the presence of a factor as one allelomorph and its absence as the contrasting factor. For instance, if yellow color is due to a present factor then if it were lost the color that results is green. But since we know nothing about the material in the germ-plasm that by interacting with other parts gives yellow in one case and green in the other, it seems to me gratuitous to postulate the nature of the change in the germ plasma. It is only necessary to assume that the original factor and a new factor form a pair without in any way committing one's self as to how these two allelomorphic factors are related to each other.
- The term segregation applies strictly to the process of separation of the contrasted factors (allelomorphs). When more than one pair is involved, the distribution that follows the segregation of each pair is called assortment in the text, and assortment is a different process from segregation; although it is the resultant of segregation so far as each pair is concerned.
- Except in those eases where the male has one less chromosome than the female. In such cases the single sex chromosome of the male has no mate.
- Bateson and Punnett formerly defined coupling as the association of two factors and repulsion ("spurious allelomorphism") as the condition where two factors are usually not associated in the same gamete. They point out that the same idea is expressed by saying that if two dominants come from one parent and two recessives from the other coupling is observed; but if one dominant and one recessive come from each parent, repulsion will be found. For the fly, Drosophila, we have pointed out (Jour. Expt. Zool., 1911) how both these results can be accounted for on the hypothesis that the factors concerned are carried by the chromosomes. Bateson has more recently changed his conception of coupling and repulsion.
- The twisting of the chromosomes has been described by a number of writers. Janssens has observed that at the time when the pairs are about to separate, cross-bridges between the pairs (more strictly between the halves of the pairs) can be seen. Whether these cross-bridges are the result of the kind of crossing referred to in the text can not be discussed here. Janssens points out that the mechanism of interchange between homologous chromosomes, by means of the cross-bridges furnishes an interesting explanation of those cases where the number of distinct allelomorphic pairs of characters is greater than the pairs of distinct chromosomes. The evidence seems to me to indicate furthermore that independent assortment occurs when factors lie in different chromosomes, while the interchange between homologous chromosomes accounts only for the relatively small proportion of crossing-over. Only when the factors lie very far apart is there a numerical approach to the independent assortment of factors lying in different chromosomes.
- The first and best description of double crossing over is to be found in Sturtevant's paper (Jour. Expt. Zool., 1913).