Genetics of Buff and Related Color Patterns in the Fowl1

J. A. BRUMBAUGH2 AND W. F. HOLLANDER

The Department of Genetics, Iowa State University of Science and Technology, Ames

(Received for publication September 4, 1965)

O NEpoultry
of the landmarks in the history of
breeding was the importation
been advocated by Jaap and Hollander
(1954), so that other types can be ana-
lyzed in terms of mutant factors.
of huge Chinese fowl into Europe about
1845. Tegetmeier's (1873) account indi- This procedure was almost followed by
cates that the names "Cochin-China" and Davenport (1909). He crossed Buff Cochin
"Brahma" were less correct than "Shan- bantam with a White Silkie, known by pre-
ghai." Fanciers crossed these into many vious breeding test to conceal the black-
other stocks, to give rise to American dual- breasted-red game or wild-type pattern.
purpose breeds and new color varieties of The F-t were described as "washed-out buff
others, especially the buff, cinnamon, ging- color . . . the Jungle pattern shows itself in
er-red, and light or Columbian. the black tail and slightly redder buff of
Genetic analysis of these color types has the wing-bar and hackles in the male."
been sporadic and incomplete. A good re- Only chick-down descriptions were given
view of such studies was presented by Jull for F 2 ; excluding the recessive whites, these
(1932), indicating considerable complexity. were 34 buff or "buff and black," and 7
Nevertheless he applied the symbol e, for game (wild-type). Davenport's interpreta-
restriction (non-extension) of black, fol- tion was that a dominant or incompletely
lowing Dunn (1923). This practice was dominant "xanthic" factor, X, was neces-
continued by Hutt (1949), as well as by a sary for buff coloration, and also a reces-
number of subsequent writers, e.g., Kim- sive factor ;' for absence of jungle-fowl pat-
ball (1956). tern.
In recent years evidence has been ac- Cock and Pease (1951) reported the
cumulating that a rather large number of cross of Light Sussex with autosexing
multiple alleles of E exist (Brumbaugh and Brussbar, essentially a barred black-red:
Hollander, 1965). These affect primarily "The F], birds resemble Columbians, al-
the chick-down pattern. Most of the adult though in some the black pigment is much
male plumage phenotypes are alike (wild- more extensive; but in backcrosses to the
type). The relation of buff to these pattern Brussbar the proportion of black-reds (or
types is the subject of the present paper. their silver counterparts) varies widely in
Probably the chief reason for confusion different families, and in most is too low to
and delay in analyzing the color patterns be explained by a single gene."
genetically has been failure to define the Kimball (1956) states that Columbian
goal of analysis. The use of a standard ref- Wyandotte crossed with Brown Leghorn
erence type, preferably the wild type, has (essentially wild-type color pattern) gave
typical Columbian F L In successive
'Journal Paper No. J-5210 of the Iowa State backcrosses of the Columbian-pattern birds
University Agricultural and Home Economics Ex- to Brown Leghorn, he obtained 1:1 segre-
periment Station, Ames, Iowa. Project No. 1380.
2
Present address: Department of Zoology and
gation, recognizable both in chick down
Physiology, University of Nebraska, Lincoln, Ne- and adult plumage. However, he continued
braska. to use the symbol e.
451
452 J. A. BRUMBAUGH AND W. F. HOLLANDER

Because of these conflicting reports, we type are clear buff, though more often trace
decided to make an extensive analysis of smudges of darkness are present in the dor-
buff, using Gallus gallus, wild-type red jun- sal region where the wild type has black
gle fowl, for the initial test. stripes. In any case, unlike true buff chicks,
wheatens develop dark feathering, and the
MATERIALS AND METHODS
males mature as essentially wild type.
Our breeding stocks and methods have At hatching, chicks were all carefully ex-
been described elsewhere (Brumbaugh, amined and recorded. Eggs failing to hatch
1963; Brumbaugh and Hollander, 1965), were opened, and if the embryos were
and are too extensive to repeat here in de- about full term they were also described.
tail. Most of the breeding, hatching, and Coloration and pattern could be best exam-
rearing was carried out on the Poultry ined after washing, drying, and fluffing the
Farm of the Iowa State University Poultry down. Representative chicks' skins were
Science Department, to which we are in- preserved flat for reference. The inner sur-
debted for technical assistance. face was dried with borax.
Our stock of Red Jungle Fowls was de- At the age of two months or later, the
veloped from a blend of two lines: one surviving birds were again recorded for
from Dr. Walter Landauer of the Uni- plumage and foot colorations. Unless
versity of Connecticut, and the other from needed for further breeding they were then
the Lincoln Park Zoo in Chicago. Our killed. Representative plumage samples
only other source of wild-type color pat- were also preserved for record.
tern was several Brown Leghorns of both
Light and Dark varieties. The latter type RESULTS
was found to be heterozygous for another A Buff Minorca cock was crossed with
allele in the E series, which we have desig- two jungle fowl hens, and 63 F t chicks re-
nated ep (partridge), and which Smyth sulted. All were considerably lighter than
(1965) designated eb (brown). standard wild type, the down color being
One of our principal testers was a stock described as tawny-buff with pale narrow
of recessive wheaten type, symbolized ey, striping (Fig. 1). However, none of the
derived in part from the Jungle Fowls of stripes was absent or interrupted. The beak
Lincoln Park and in part from mongrel was brown like that of buff breeds, or dark-
bantam sources. er, in contrast to the flesh-colored beak of
Another recessive allele in the E series the jungle-fowl chicks. In mature plumage,
we had extracted from an exceptional the Fi cocks were rich coppery-buff with
White Giant cock. This allele was called some black in the tail. The Fi hens (Fig.
"blurred" and considered probably the 2) were buffy-yellow with a mossy sprin-
same as that previously reported by More- kling of black over the tail and back, and
john (1955) and termed "speckled head," in the flights. In both sexes the feet were
with symbol es. This Giant cock, as well blue, lighter than those of the wild type.
as most of our other pure stock birds ex- Matings of F x inter se produced 464 F2
cept Rhode Island Reds, came from the chicks, most of which were buffy. Only 8
Murray McMurray Hatchery of Webster were classified as strictly wild type. The
City, Iowa. Our Rhode Island Reds were 381 birds surviving to plumage description
Parmenter strain, maintained by the Poul- were of numerous intergrading shades, in-
try Science Department. cluding dark red, buff, wheaten, and the F±
Some chicks of the recessive wheaten phenotype. Five of the wild-type chicks
 GENETICS OF BUFF COLOR 453

FIG. 1. Comparison of chick-down patterns. Left, standard wild type; center, female F, from Rhode
Island Red hen X standard cock; right, Fi from Buff Minorca X standard.

survived and remained standard. This com- with close study we found it possible to
plex segregation indicated that buff cannot recognize qualitative groupings. First was
be considered a single mutant. the group indistinguishable from the Fj.
Backcrosses to wild type were next Second were "reds," much redder (darker)
made, to help in the analysis. Using the F 2 than the F-L Third were "gingers." These
hens, we obtained 238 chicks; the recipro- as chicks were well striped but had tawny
cal matings gave 116 chicks. In both faces and ventral regions, and brown beak.
groups, chick-down phenotypes ranged be- As they matured, the plumage showed more
tween that of the Ft and wild type, with buffy-red than in wild type, with dark fluff,
the latter greatly in the minority. Of the the breast feathers of the cock having a
277 survivors, 28 were described as stan- coppery hour-glass pattern. Fourth were
dard wild type, while the remaining birds "mahogany-infused"; these as chicks were
showed various grades of restriction of indistinguishable from wild type but in ma-
black. These were confusing at first, but ture plumage were redder, the breast feath-
ers of the cock being spangled (red base,
black tips) to varying degrees. Fifth were
"dilutes," which in chick down were pale-
striped, but when mature, only the female
was lighter than wild type.
Second backcrosses of representatives of
each of the above groups were made with
jungle fowls. Twenty or more chicks were
desired from each mating for reliability of
ratios, and in most cases, this number was
obtained. Matings of the first group (re-
sembling Fi) tended to repeat the complex
segregation obtained in the first backcross.
From two females however, single male
progeny of a still different coloration
showed up: though wild type as chicks,
FIG. 2. Fi hen from Buff Minorca X jungle
fowl. Most of the plumage is buff, black being they showed diluted phaeomelanin in the
greatly restricted. mature plumage, somewhat similar to
454 J. A. BRUMBAUGH AND W. F. HOLLANDER

TABLE 1.—Segregation data from presumptive monohybrid derivations of Buff X jungle fowl (mature plumage)

Mating Number
of Progeny x2 P>
Cock Hen Matings
Gr/+ +/+ 4 62Gr/+:S7+/+
+/+ Gr/+ 0
Total 4 62 Gr/+:S7 + / + .22 .5
Mh/+ +/+ 5 113 Mh/+:88 + / +
+/+ Mh/+ 2 28Mh/+:28 + / +
Total 7 141 Mh/+:115 + / + 2.41 .1
Di/+ +/+ 1 8 Di/+*: 5 + / + *
+/+ Di/+ 3 12 Di/+*:15 + / + *
Total 4 20Di/+*:20+/+* 0
Cb/+ +/+
Gr/+
1 13Cb/+:25 + / + 3.78 .05
Gr/+ 4 42 GrGr:S8 Gr/+:41 + / + 4.46 .1
Mh/+ Mh/+ 5 27 MhMh:36 Mh/+:20 + / + 2.64 .25
:
Only female progeny classified.

"golden duckwing." We have called the fluff was dark gray. Most of these homozy-
coloration "champagne blond." One of gotes were low in vigor and died before two
these males was again backcrossed to stan- years old.
dard. All the progeny were standard in Four second-backcross matings were
chick-down color, but the survivors were made using dilutes. In each case, only the
easily classified in mature plumage (13 two classes of chicks were produced, indi-
champagne blond:25 standard). The cham- cating monohybrid constitution. Clas-
pagne blonds all died, mostly of anemia, sification in later plumage was possible
before two years old. No further analysis only with females. There were not many,
has been attempted. but the totals make a 1:1 ratio. Attempts
Second backcrosses involving the "reds" to mate dilutes inter se were quite discour-
all produced four or more types of progeny, aging because of poor egg-laying. Three
including all the above types except that classes of chicks were again obtained, how-
resembling the Fi. We therefore infer that ever. The new phenotype here was pale and
the "reds" are at least dihybrid. lacked the lateral black back stripes (Fig.
In the case of second backcrosses from 3). Unlike wheaten chicks, these showed
ginger, four matings were made, and mono- fairly well developed head stripe and eye
hybrid segregation was evident in each. stripe. In mature plumage these birds be-
The next step was to mate ginger heterozy- came reddish, with very pale nearly white
gotes inter se (four matings). Three classes feet.
of chicks were produced by each mating, Mahogany birds in second backcrosses
with fair approximation to a 1:2:1 ratio (seven matings) produced chicks all
(Table 1). The presumed homozygotes classed as standard. Of the 257 survivors,
were more tawny than the heterozygotes, 141 were classed mahogany and 116 stan-
with less distinct striping (Fig. 3), and in dard, a reasonable approximation to a 1:1
mature plumage both sexes resembled the ratio. Five matings of heterozygotes inter
Fi from buff X standard. The plumage se again produced chicks entirely standard
 GENETICS OF BUFF COLOR 45 5

FIG. 3. Comparison of chick-down patterns of standard wild type (center),

homozygous ginger (left), and homozygous dilute (right).

in phenotype. Survivors were classified into red breeds to the E locus, will be briefly
three groups with fair 1:2:1 ratio (Table summarized:
1). The presumably homozygous mahoga- Buff Leghorn hen X blurred (eses)
ny in both sexes is much redder than the cock produced 30 F x chicks, all of which
heterozygote, and approaches the Rhode were buffy with faint striping on the back
Island Red coloration. and blurred-speckled head markings. In
The buff X standard in F 2 had pro- mature plumage the F t were similar to the
duced some birds which appeared to be Buff Minorca X jungle fowl, but with
wheaten. To test this interpretation, we green feet. Fj X Fi gave 149 F 2 chicks,
made individual crosses of 15 first-back- of which none were wild-type—all were
cross birds to wheaten and made chick buff or blurred or something intermediate.
classifications. Three kinds of results were These results agree with the interpretation
obtained: (1) in five tests, all chicks were that at the E locus the Buff Leghorn is
standard (i.e., not wheaten); (2) in three homozygous for ey.
tests, half the chicks were standard and Speckled Sussex hen X blurred (eses)
half wheaten; and (3) in seven tests, half cock produced 51 F± chicks, all with
the chicks were standard and half were blurred head pattern and variable back-
quite light with some disruption of the striping. Speckled Sussex hen X jungle
striping, similar to that of the Buttercup fowl cock produced 33 Fi chicks, all of
breed. There was no relation between the standard down-pattern phenotype. In ma-
outcome of these tests and the phenotype ture plumage, these birds were mahogany.
of the tested parent. We interpret the re- F 2 chicks totaled 173, composed of 129
sults of these tests to mean that, in addi- standard:44 wheaten. These results indi-
tion to the semi-dominant factors indicated cate that the Speckled Sussex is also ho-
earlier, the Buff Minorca had ey (wheaten) mozygous for ey.
and ehc (buttercup). These recessives at Rhode Island Red hen X jungle fowl
the E locus would preclude the possibility cock produced 35 Fx chicks, all with pale
of any of the other factors being alleles narrow striping (Fig. 1) and brown beak.
of E. As adults, these birds were mostly red.
Supplementary breeding tests, made to Two Fj females were tested with wheaten
learn more about the relation of buff and males; 102 chicks were obtained, of which
456 J. A. BRUMBAUGH AND W. F. HOLLANDER

51 were wheaten and 51 had complete tified (Brumbaugh and Hollander, 1965)
striping (standard or varyingly pale and do not occur in adult male plumage.
narrow). We consider this result good evi- It may be asked, if ey for example is
dence that the Rhode Island Red, like the ineffectual in adult males, why do buff and
Buff Leghorn and the Speckled Sussex, is red breeds have this mutant? Of course, the
homozygous ey. wheaten phenotype of the adult female
In 1961 Dr. Lewis T. Smith of the Iowa would seem a major step toward buff or
State University Poultry Science Depart- red, and might have been selected for.
ment gave us a red cock, superficially al- However, another explanation is selection
most identical in coloration with the Rhode in the chick stage, where this mutant is a
Island Red, which had been black as a potent restrictor. Fanciers probably would
chick. It was F 2 from crossing Rhode Is- prefer chicks of buff or red breeds to be
land Red with White Leghorn. We crossed buff rather than black, or striped, or other
this cock with jungle-fowl hens and ob- dark phenotype. Apparently it is possible
tained a range of chick colorations in- with E to have a pure breed which would
cluding .E-type black, rusty black of fairly be black in chick down and red in the adult.
uniform distribution, and wild-type. The By contrast, with homozygous wheaten and
striped: non-striped ratio was approxi- recessive black we have produced a pure
mately 1:1, suggesting that E was segre- type which is buff in the chick and black in
gating with a modifier, probably mahog- the adult (Brumbaugh and Hollander,
any. Here is evidence that e7 is at least 1965).
unnecessary in the red phenotype. Our data indicate that the buff adult
One further note on red is pertinent here. phenotype depends on four autosomal mu-
We crossed monohybrid ginger X monohy- tant factors, which we have named ginger,
brid mahogany birds, and obtained four mahogany, dilute, and champagne-blond.
phenotypes in the ten mature progeny: two We propose for these the symbols Gr, Mh,
standard, four ginger, two mahogany, and Di, and Cb, respectively. None of these in-
two red, which is about a 1:1:1:1 ratio. This dividually can be considered a "buff gene";
result confirms the previous analysis of reds the buff phenotype is an interaction effect.
from the first backcross. For this reason we have not used the sym-
bol X of Davenport (1909) or Knox'
DISCUSSION (1927) symbols Bu and Bu'.
Although our study is incomplete, the Tegetmeier (1873) described a color va-
results are sufficient to show that the sym- riety of Game fowls called ginger-red. The
bol e as used by Jull (1932) and later term ginger is still current among cock-
workers is invalid. The "restricted black" fighters, and an old-timer applied it to our
condition is not a simple one. Kimball's mature homozygous ginger birds.
(1956) claim that monohybrid segregation While the Rhode Island Red and the
resulted from Columbian Wyandotte F± Speckled Sussex have the sex-linked Id fac-
backcrossed to Brown Leghorn is the out- tor for light-colored feet, this gene is not
standing case of disagreement with our involved in the Buff Minorca or Buff
conclusion. Even if the Columbian Wyan- Leghorn. The light feet of such birds seem
dotte should be confirmed as an exception, attributable to the dilute factor, Di. None
which we consider most unlikely, there has of the other factors seems to affect the feet.
been no evidence that the restriction factor Champagne-blond, Cb, is the least-well-
concerned is allelic with E. The restriction established factor of the four we propose in
effects of E alleles which we have iden- buff, yet it seems essential. It appeared in
 GENETICS OF BUFF COLOR 4S7

monohybrid condition only in the second blond (Cb). Cb is thought to be closely

backcross to jungle fowl, and then only in linked with Gr. Additional tests of £-locus
two individuals. As an explanation, we sug- constitution in Buff Leghorn, Rhode Island
gest rather close linkage with Gr. It is Red, and Speckled Sussex revealed only ey,
quite possible that a high proportion of our which cannot account for the restriction
apparently monohybrid ginger birds had phenotype of these adult males.
Cb as well, since the latter is not evident in
the chick. The observation that most of the REFERENCES
Cb birds died of anemia is inexplicable at Brumbaugh, J. A., 1963. A study of the genetic
present. Also it is curious that Di birds and control of black-red pigment patterns in the
homozygous Gr were low in vigor. The fowl. Thesis (Ph.D.), Iowa State University.
Brumbaugh, J. A., and W. F. Hollander, 1965. A
Buff Minorca and Buff Leghorn and the Fi further study of the E pattern locus in the fowl.
did not show such troubles. Iowa State J. Sci. 40: 51-64.
The differences between adult Buff Or- Cock, A., and M. Pease, 1951. The genetics of the
pington, Rhode Island Red, and New white pile pattern in the domestic fowl. World's
Poultry Congress, 9th, Paris, 1951. Official Re-
Hampshire (light red) have been investi-
port 1: 49-53.
gated by Somes and Smyth (1965) by all Davenport, C. B., 1909. Inherited characteristics in
possible reciprocal crosses and F 2 , as well the domestic fowl. Carnegie Institution of
as some backcrosses. A multiple-factor in- Washington Publication 121.
terpretation was proposed, with no evi- Dunn, L. C , 1923. Color inheritance in fowls: The
dence for sex linkage. The authors were un- genetic relationship of the black, buff and Co-
lumbia colorations in the domestic fowl. J.
able to identify the effects of any individ- Heredity, 14: 23-32.
ual genes. According to our interpretation, Hutt, F. .B, 1949. Genetics of the Fowl. McGraw-
all these breeds presumably have Gr and Hill Book Company, New York, New York.
Mh, as well as ev, so that the differences Jaap, R. G., and W. F. Hollander, 1954. Wild type
among them would rest primarily with Cb as standard in poultry genetics. Poultry Sci.
33 : 94-100.
and Di. Whether this is an oversim-
Jull, M. A., 1932. Poultry Breeding. John Wiley
plification or not may best be answered by and Sons, New York, New York.
individual analysis of the breeds as we Kimball, E., 1956. Genetic relation of wild type
have done with the Buff Minorca. and Columbian plumage patterns in the fowl.
Poultry Sci. 35: 1274-1284.
SUMMARY Knox, C. W., 1927. The genetics of plumage color
A reinvestigation of the genetical basis in poultry. Iowa Agricultural Experiment Sta-
tion Research Bull. 105: 106-131.
for "restriction of black" plumage patterns
Morejohn, G. V., 1955. Plumage color allelism in
indicates that the previously postulated Red Jungle Fowl (Gallus gallus) and related
gene e is invalid. Distinction between domestic forms. Genetics, 40: 519-530.
chick-down and adult plumage effects is Smyth, J. R., Jr., 1965. Allelic relationship of
necessary. The Buff Minorca crossed with genes determining extended black, wild type,
and brown plumage patterns in the fowl. Poul-
standard Gallus gallus produced Fi striped
try Sci. 44: 89-98.
as chicks but more resembling buff as Somes, R. G., Jr., and J. R. Smyth, Jr., 1965.
adults. Backcrosses to jungle fowl, as well Feather phaeomelanin intensity in Buff Orping-
as F 2 , indicated segregation at five loci, the ton, New Hampshire, and Rhode Island Red
only recessive factor from the Buff being at breeds of fowl. 2. Inheritance studies from
whole feather extracts. Poultry Sci. 44: 47-52.
the E locus. The other four factors from
Tegetmeier, W. B., 1873. The Poultry Book (sec-
the Buff were named ginger (Gr), mahoga- ond edition). George Routledge and Sons, Lon-
ny (Mh), dilute (Di), and champagne don.