Journal of Genetics 16: 33-43 (1926)
"Somatic Segregation" in Domestic Fowl
By A. S. Serebrovsky

(From the Anikovo Genetic Station and Institute of Experimental Biology, Moscow)

FROM most other domestic birds and from mammals domestic fowl differ in the absence of a kind of coloration called "piebald." When we say "piebald" we mean the presence of more or less irregular spots on the skin with no pigment; while the derivatives of the skin—feathers and hair—are upon these spots also deprived of pigment. Among mammals we know piebald horses, asses, cows, deer, goats, sheep, swine, dogs, cats, rabbits, guinea-pigs, rats, mice, wild bisons etc. Among birds we find piebald geese, ducks, pigeons, canary-birds, sometimes sparrows, etc.

The study of fowl-hybrids of the race Plymouth Rock with other races showed, however, the presence of coloration not only resembling the "piebald" condition, but also serving to explain its origin.

As is known, the race Plymouth Rock has barred feathering, each feather being traversely striped black and white (Pl. II, fig. 1). The number of such stripes is usually five black stripes and five white ones, but this number may vary more or less, depending upon the size and character of the feather, and upon other reasons. In the "hackle" of the cock, and especially in the long tail feathers, the number of stripes can greatly increase.

When carefully examining the feathering of Plymouth Rock hens, one notices on most an occasional feather, differing in colour from other feathers. Generally such feathers are quite black. Sometimes such occasional feathers are grouped. Once we happened to observe in the hackle of a hen six black feathers tightly grouped together.

Still more often it has been noticed, that not the whole feather but only a part of it is of exceptional coloration. In most cases the distal end is normally coloured, the other half being quite black; sometimes the proximal end is striped on one side.

Such exceptional black feathers occur almost always on hens only. In homozygous Plymouth Rock cocks these abnormalities are not known to us, but only a few cocks have yet been studied.

We are inclined to consider the occurrence of these exceptionally coloured feathers as a manifestation of an odd "piebald" coloration, for the reason that in other animals the piebald coloration is also not limited by the distribution of coloured and uncoloured spots. For instance in mammals another kind of piebald coloration often occurs in which red spots are spread on a black ground. These spots are irregular in the same degree, and often not symmetrical; their variability is also very marked as is that of black and white piebald coloration, and there is no fundamental difference between the two varieties of piebald coloration. There are black and red rabbits, guinea-pigs, mice, dogs, cats, gold-fishes, etc.

In dogs we have also a third kind of piebald coloration in which, on a uniformly coloured ground, are spread piebald spots, where coloured hairs are mixed with white ones, as in "silver" rabbits.

Here we have different kinds of piebald coloration. However, the following case of piebald cats leads us to another distinct kind of "piebaldness." It is well known that white blue-eyed cats are deaf. Genetical study established that this deafness is in some way correlated with the blue coloration of the eye, and this last is itself part of the piebald coloration. For if a white piebald spot happens to fall upon the region of the eye, the corresponding eye turns out to be blue, instead of the usual yellow. Such piebald eyes also occur in men, dogs, rabbits, etc. But with cats there is an additional complication, and the ear on the side of the blue eye is deaf.

It is therefore possible to point out the essential character of the piebald condition, viz. the mosaic distribution of some features, not only of the skin coloration, but also of the hair, or of iris coloration, or of the composite elements of hearing.

Such an explanation of the piebald condition might be adapted to facts we have observed in Plymouth Rocks. Here we have to deal with the irregularities in the distribution on the body of some elements of coloration, forming the irregular regions, the size of which varies from part of a feather to a group of many neighbouring feathers. These spots are distributed at random and asymmetrically.

Let us follow the analysis of this character. For this purpose it is necessary to notice the genetical elements of the Plymouth Rock coloration.

1Our genetical nomenclature is as follows. Either gene is symbolised with a number similar to the principles of decimal bibliographical classification. The genes for coloration belong to the 8th group, the ground genes for colour belong to the 81st group, the genes for albinism to the 811th group, and so on. By aid of a special table these numerical symbols are transferred in short words. For a recessive allelomorph to this word is added "a": tedu is the gene for colouring, atedu is its recessive allelomorph, the gene for albinism, and so on. In this paper the following genes are mentioned;
tedu-gene (8114)—the ground-gene for colouring; atedu—albinism.
tife-gene (8215)—the "melanic"-gene, see p. 34; atifa—the normal condition.
tode-gene (8311)-the "inhibitor of black pigment in Blue Andalusian"; atode—the normal condition.
todi-gene (8312)—the inhibitor of most colouring, the gene for epistatic white in White Leghorn; atodi—the normal condition.
tuge-gene (8721)—the gene for "silver," see p. 35; atuge—the gene for "gold."
trage-gene (8521)—the gene for barred-condition; atrage—its absence.
trakla-gene (8535)—the gene for light-coloured marking and light shaft; atrakla—its absence.

The Plymouth Rock coloration differs from the wild fowl coloration by the following principal genes1:

  1. Tifa-gene, determining the increased development of the black pigment, and in some combinations capable of giving the self-black coloration of Minorca, Langshan, etc.
  2. Trage-gene, determining the transversely striped distribution of pigments in the feather; in the presence of the tifa-gene the transverse stripes can be black and white, in the absence of same-red and white, or brown and white, etc.
  3. Tuge-gene, i.e. the gene for the silver coloration. Normally this gene is not to be observed in the Plymouth Rock coloration, and the typical coloration of the Plymouth Rock can exist without it. That it cannot be observed is chiefly owing to the tifa-gene. Nevertheless, in most Plymouth Rocks there is a tuge-gene. It suppresses the development of golden elements in the coloration. We can usually notice that fowls with tifa and trage without tuge (i.e. in the presence of its recessive allelomorph atuge), have some golden edges or spots in the hackle, and probably owing to these golden spots, spoiling the standard of the race, the fowl-breeders have conserved the tuge-gene in the composition of the race.

These last two genes tuge and trage are localised in the sex chromosome, and tifa in an autosome. Therefore on crossing a Plymouth Rock hen with a "common" black-breasted red cock all the sons show the Plymouth Rock coloration, and all the daughters a pure black one, inheriting from the mother the tifa-gene, which is localised in the autosome, while they cannot get the trage-gene, localised in the sex-chromosome. In such a way the hens lose, so to speak, the trage-gene and turn out pure black.

We suggest that it is possible to consider the "exceptional black" feathers in a fowl as having lost for some reason the transversely striped design, thereby revealing the latent black coloration.

This point of view is decisively confirmed by considering the hybrids of the first generation from Plymouth Rocks with other races. We have had some hybrids between Plymouth Rocks and Orloff fowls, and also English-game race: the last two races are alike in coloration, the abovementioned genes, tifa, trage and tuge being absent. The cocks have a black breast, and a more or less bright red back, hackles and saddle.

When we designate the genes tuge, trage and tifa by corresponding symbols Tu, Tg, Tf, and their recessive allelomorphs by tu, tg, tf, we get for the hybrids of first generation Plymouth Rock-Orloff the following formula:

Chromos. I Tu Tg Chromos. II Tf

tu tg tf

Careful observation has shown the presence of exceptional feathers in such F1 cocks. Most of these exceptional feathers proved to be black with red borders (cf. Pl. II, fig. 2). We can get the same exceptional feather, if we add to the Orloff's coloration the sole gene tifa: this gene would darken the middle part of the feather to an intense black, leaving its borders red (by simultaneous action of other genes which we do not mention just now).

Therefore, if the exceptional feathers were potentially black and, as it were, removed from the influence of the trage-gene, it is obvious, that these feathers are at the same time removed from the influence of both of the genes, which are localised in the sex-chromosome: they are neither transversely striped nor silvery.

In the same cock (N 2344) we have observed in small numbers also other exceptional feathers: white, black and so on, but we will speak of them later on. We would emphasise here that almost all exceptional feathers in Plymouth Rock-Orloffs are black with red borders.

This fact is to be explained as follows: since the exceptional feathers are simultaneously exempted from the influence of two genes, which possess a single common feature, namely being localised in the same chromosome, the whole phenomenon can be explained by the supposition that these feathers are exempted from the influence of this chromosome.

In the descendants of these hybrids, which are heterozygous in relation to tuge and trage, a segregation is of course to be observed. Omitting the possibility of crossing-over, which we shall examine further on, we can say that some of the offspring get both genes, included in a sex-chromosome, while others do not acquire this particular chromosome, and are at once deprived of both genes.

Thus the simplest suggestion would be, that, for some reason or other during development, the exceptional feathers lose one of their sex-chromosomes, and if this chromosome turns out to be a chromosome bearing tuge and trage, the feather becomes the same as in descendants completely deprived of this chromosome.

Thus in the development of exceptional feathers there would appear to be some form of somatic segregation whereby some feathers get one set of genes and other feathers another set; and as in the normal segregation the genes tuge and trage, localised in the same chromosome, show linkage, this same linkage is also to be found in the "somatic segregation."

I offered such an interpretation of this appearance of exceptional feathers at the colloquium of the Institute of Experimental Biology in March, 1923. At the same time I pointed out the method of verifying that suggestion. If the character of exceptional feathers depends upon the chromosomal linkage of the genes tuge and trage, it ought to be altered by the change of the hybrid structure of the bird; as, for example, in the case when the genes tuge and trage are localised not in the same chromosome but in two homologous ones. The normal feathers containing the two chromosomes would have in both cases the same appearance, but the exceptional feathers should look quite different. If the chromosome bearing tuge vanishes, the feather becomes tu Tg, i.e. transversely striped with golden tint, but the trage-bearing chromosome having vanished, the feather will not be striped, but only bordered, and this border will be of a silvery, not of a golden tint.

Now we have succeeded in getting two samples of such cocks. From the crossing of the trage-atuge cock and atrage-tuge hen we reared two adult cocks. The prediction of the character of their exceptional feathers was quite justified. Thus golden unstriped feathers have not been found either in the hackles or in the saddle, but we have noticed the silvery unstriped feathers, quite similar to the feathers of true atrage-tuge, in great quantity (cf. Pl. II, fig. 6). The atuge-trage feathers were much rarer, but at some places-in the wings and tail-were found some feathers, one side of which was gold and barred and the other silvery but nonbarred (cf. Pl. II, fig. 5).

This prediction being justified, confirms the supposition that the exceptional feathers are determined by the chromosome structure. This circumstance leads us to the following questions:

Haldane's (1921) investigations as well as ours (1922) had established that the genes tuge and trage show rather loose crossing-over, in such a way that their connection breaks off in approximately 30 per cent. This number probably varies with the age of the bird and with other circumstances, but of course it is easy enough to observe the breaking apart of these two genes. Now one ought to inquire what happens to the same genes by the somatic segregation (or by somatic non-disjunction). In other words, do the exceptional feathers always get both genes together, or does separation between them also occur?

The solution of this question is essential to the theory of the crossing-over. Morgan's point of view allows of crossing-over only at the moment of the "conjugation of chromosomes" in reduction division. On this interpretation the homologous chromosomes may exchange their parts and genes in no other cell division, and therefore in every cock's soma the sex-chromosomes ought to keep their initial composition, in the present Tu Tg and tu tg (F1, Plymouth Rocks x Orloffs).

If crossing-over could also occur in somatic divisions or in the stage of the resting nucleus, there could be formed in the cocks' soma chromosomes of a new structure, namely tuge-atrage (Tu tg) and atuge-trage (tu Tg). Lack of the former of these two chromosomes would lead to the formation of golden-transversely-barred feathers; lack of the second would mean silvery-unbarred feathers.

The thorough examination of the feathers of the hybrids Plymouth Rock x Orloff seems to confirm the possibility of such cases. It is true that it is not possible to determine the number of these crossings-over being 30 per cent. and so on. But in some cases we succeeded in finding some feathers, which, in the complete absence of transversely striped design, were distinctly silvery, i.e. atrage-tuge.

Two of these feathers are pictured on the coloured table (Pl.II, figs. 8,9). It is true that inasmuch as these feathers are exceptional, the probability of chance errors is here very great. But in atuge-birds such silvery feathers are never found; there is no doubt about it. It is more difficult to guarantee the absence in this case of the trage-phene, i.e. the manifestation of trage-gene. The gene trage can manifest itself in different degrees. In tifa forms it is clearly manifested; but in atifa forms it is greatly weakened, especially in hens, where it is often very difficult to decide the question, whether she bears trage or not.

The feathers of the opposite type (trage-atuge) being so difficult to establish, have not been found by us. Therefore the problem of crossing-over in this case remains open. We can only state, that by "somatic segregation" the frequency of crossing-over is small, but we cannot confirm the non-existence of it, because of the "suspicious" feathers.

Does the phenomenon of "exceptional" feathers represent a phenomenon concerning only the sex chromosome? We can assert, that genes localised in the autosomes are able to give rise to similar exceptions in their distribution among feathers. We noticed that exceptional feathers are to be found among thoroughbred Plymouth Rocks, though almost always in hens. This fact is easily explained if one remembers that with fowls the female sex is heterozygotus, and therefore hens have only one sex-chromosome in the cells of their body, whereas cocks have two. If the sex-chromosome with tuge and trage does not fall on some feather of a hen, this feather will be deprived of both these genes and this will be immediately revealed. But with cocks it is necessary in order that the same effect be obtained, that both sex-chromosomes fall out. But this can only occur much more rarely than the missing out of one sex-chromosome, and therefore "exceptional feathers" are to be found in cocks much less often than in hens. Heterozygous cocks, however, are in this respect equal to hens, and even surpass them, for one can find several of these exceptional feathers on each heterozygous cock. We can even venture to observe that the constitution of such cocks seems to promote a disposition towards such exceptional feathers. We will speak of it again further on.

Therefore we must look among heterozygotes for exceptional feathers of another type than tuge and trage. And actually we find with our Plymouth Rock x Orloff cock (N 2344), with which we began our investigation, a series of other exceptional feathers. We found black feathers, red and albinotic ones. This cock was really heterozygous for the tifa-gene (which calls forth black colouring in the middle of the feather), and for the tedu-gene (for colour as opposed to albinism). The presence of purely black feathers is explained by the fact that the cock is heterozygous for the trakla-gene, which is to be found in Orloffs but not in Plymouth Rocks. However, we have not yet investigated sufficiently the trakla-gene, to be able to make final conclusions.

We could discover through an examination of the "exceptional" feathers in this cock those parts of his hereditary formula in which he was heterozygous.

Are the exceptional feathers a peculiarity of Plymouth Rocks and their hybrids? We must point out the disposition of all birds having the trage-gene to have exceptional feathers. Birds without breed which had never been interbred with Plymouth Rocks, are inclined to have exceptional feathers as soon as they reveal trage-marking.

The trage-gene, or some other element closely related to it, brings into the mechanics of mitosis some kind of disorganisation. The exceptional feathers of the hybrids tuge and trage, which have the structure (Tu tg / tu Tg), are more often deprived of the trage-gene than of the tuge-gene (or at least of the phene).

However, we can observe the same occurrence in the case of other genes, first of all in the case of todi-gene, which calls forth the epistatic white colouring. The todi-heterozygotes have among white feathers a certain quantity of pure black ones, or of some other colouring (blue, etc.), hens more often than cocks. They are very intensely coloured in distinction from some feathers of the albinotics of a light grey or yellow colour, and their colouring is exactly that, which it would be were the todi-gene to be removed. We have here a perfect illusion of the todi-gene having never been for some reason in these feathers. We can also observe that in the presence of the tifa-gene in the geno-type of the bird the quantity of exceptional feathers is greatly increased.

We can observe the same occurrence with the presence of the tode-gene. Tode inhibits the black pigment in the heterozygous condition until it becomes blue, and in the homozygous condition, until it becomes white. The blue colouring is to be found in Andalusian hens (with the presence of tifa), and among the blue feathers we can always find parts of feathers and even whole feathers of an intense black colour. These feathers seem to be deprived of the tode-gene.

The same occurrence is to be observed with the tode-homozygotes, and it is here even more striking. The exceptional feathers are not black among blue ones, but blue among dirty-white ones. As the dirty-white feathers have two tode-genes it is evident that the exceptional blue feathers are obtained by the falling out of one of the tode-genes. But while the falling out of the only tode-gene gives the heterozygotes at once a black colouring, here the falling out of one of two tode-genes gives a blue colouring.

We can sum up as follows: a series of colouring symptoms in fowls acts as if some feathers had not the genes which caused the given colouring. In the case of two genes localised in one chromosome (sex) they reveal a linkage or a repulsion according to the genetical construction, only with the difference, as compared with the gametes, that crossing-over cannot be either proved, or denied. But even if it exists, it exists in a much lower degree than in gametogenesis.

The above facts would have but a limited importance were we speaking of casual occurrences. But we already know of accidental elimination of chromosomes in certain regions of the body for Drosophila in connection with the sex-chromosome in cases of gynandromorphism, etc.; and comparable phenomena have been found by Emerson (1921) in maize. Of these we can offer an explanation because the corresponding chromosome had at the same time two genes, one for colouring and the other for the structure of starch.

We think that the most interesting problem arising out of these facts is the problem of the localised piebald coloration. Considering different cases of piebald coloration in different animals we are able to make out a series of transitions from casual piebald coloration to cases where the piebald colouring is often distributed with great precision of localisation. The piebald coloration of rabbits and Jaroslaff cattle shows us a piebaldness that repeats itself in on individual after another with comparatively few variations. If the piebald colorations arise in connection with irregularities in the distribution of genes (or whole chromosomes) we are obliged to admit in the case of localised piebaldness the existence of a special mechanism that directs these irregularities. The admission of that kind of mechanism must revive the Roux-Weismann mosaic theory, and therefore the experimental verification of this theory would be very interesting and important.

Examining fowls we find some occurrences that are very interesting from that point of view. We speak of the development of red feathers on the shoulders and on the back of cocks. These feathers by their character and the original distribution of black and red pigments remind us very much of our exceptional feathers. They are very often "segregated" in two halfs, one red, the other black, and very often there occurs a sharp transition from black feathers to red ones. We were not able to find any difference in principle between the colouring of the feathers of the back and the exceptional feathers. All the difference that is to be found consists in the fact that the appearance of these feathers in cocks is quite regular, and the place where they appear is exactly determined. Whether we have to do here with regularly appearing exceptional feathers, which appear owing to a special disposition on the part of cocks to give such exceptions, or whether it is merely an analogy, must be decided by further observations, and will probably require special methods of experiment.



Fig. 1. Normal feather from cock N 2344 with the constitution
tuge trage   tifa

atuge atrage   atifa
Fig. 2. Exceptional feathers from the same cock with "somatic segregation" in tuge-trage parts and atuge-atrage parts.
Fig. 3. Normal feather from cock N 5334 with the constitution
tuge trage   atifa

atuge atrage   atifa
Fig. 4. Exceptional feathers from the same cock with "somatic segregation" in tuge-trage parts and atuge-atrage parts.
Fig. 5. Exceptional feather from cock N 6942 with the constitution
tuge atrage   atifa


atuge trage   atifa
showing "somatic repulsion."
Compare with the right feather of Fig. 2. The normal feathers from this cock are indistinguishable from Fig. 3.
Fig. 6. Exceptional feather from the same cock, with tuge and without trage, as distinguished from Figs. 2 and 4, and according with the genetic constitution
tuge atrage

atuge trage
Fig. 7. Exceptional very rare feather from cock N 2344 (see Figs. 1 and 2), probably without tifa as also without tuge and trage in most parts.
Figs. 8 and 9. The rare exceptional feathers from the same cock, N 2344. Possible "somatic cross-overs" with tuge but without trage in most parts. Compare with Figs. 7 and 2.