Gene function in Yellowface Budgerigars

By: Peter Bergman (Sydney, Australia)

Part 1 - When is a Yellowface not a Yellowface?

Yellowface budgerigars have probably had more written about them than any other variety. Opinions differ about what Yellowfaces really are and debates about Yellowface genetics have at times become quite heated. The prevailing view today is that the three Yellowface varieties are brought about by alleles of the genes for Blue and Green. It makes sense that this should be so because it allows the same genetic infrastructure which produces the yellow pigment in the Green bird to produce the yellow pigment in the Yellowface varieties as well.

If the Yellowface genes were not alleles of Blue and Green then it would require a restructuring of the yellow producing apparatus which would be more complicated to evolve than the simple mutation of a pre-existing gene. Not all breeders agree that the Yellowface genes are alleles of Blue (bl). Some breeders claim that they have paired Green birds to Blues and produced Greens, Yellowfaces, and Blues from the one pair. If the Yellowface genes are alleles of Blue (bl) this result should not be possible. However,it should be pointed out that in the instances I am aware of where it had been claimed that a Green to Blue pairing yielded Greens, Blues, and Yellowfaces in the nest, the Yellowfaces involved were Goldenfaces. That in itself suggests the Green birds in the pairings may not have been true Greens at all but heavily pigmented Goldenfaces.

Patterns of inheritance

Most articles about Yellowfaces have focused on the genetics. Genetics is fine as far as it goes, but genetics is only about patterns of inheritance. To understand why the Yellowfaces look the way they do one has to step back and take a broader perspective and attempt to unravel what is going on in terms of the way genes are known to function.

The Yellowface varieties are the budgerigar equivalent of the Parblue varieties in other species of parrot. However, uniquely in budgerigars, the double factor birds contain less yellow than single factor birds. This paradox has caused much puzzlement over the years and is the source of the debate about whether the Yellowface alleles are yellow producing genes or yellow reducing genes.

Blue budgerigars first appeared in the early 1880s. These Blues were established and their descendants distributed around the world. It took several decades for Blues to become widely available, but within a decade of their becoming a common variety all three Yellowface varieties appeared in quick succession in the mid 1930s. The nearly simultaneous appearance of all three Yellowface genes suggests that they lay hidden as splits in the Green population for a long time.

The most curious of the Yellowface varieties is the Mutant 1 Yellowface in which the double factor bird looks exactly like an ordinary Blue. This simple fact begs the question;. how can we be sure that the common ordinary Blues aren`t really the double factor Mutant 1 Yellowfaces and that those Blues we think of as double factor Mutant 1 Yellowfaces aren`t in reality the true Blues?. The way I see it there is an important detail the hobby has overlooked, and that is:There is no rule that says mutant genes have to appear in any particular order.

We only assume that the common Blues are true Blues because they happened to have appeared and become established as the .true Blues. first. The Mutant 1 Yellowface gene appeared later but it could just as easily have happened the other way round. How can we be sure the hobby hasn`t got it backwards and that the double factor Mutant 1 Yellowfaces are really the true Blues? If that sounds farfetched then have a look at the situation in Peach-faced Lovebirds. Breeders have established two types of Parblue but true Blues have yet to make an appearance.

Imagine a different World

Imagine a world in which our double-factor Mutant 1 Yellowfaces had become established as the true Blues first and those birds we currently think of as Blues appeared later and were dubbed double-factor Mutant 1 Yellowfaces...... Our greenish single-factor Goldenfaces would be thought of as Goldenface / Mutant 1 Yellowface composites while those birds we currently think of as Goldenface / Mutant 1 Yellowface composites would be regarded as single-factor Goldenfaces. Single-factors would have the same distribution of yellow as the double-factors but would be somewhat paler in their depth of yellow.

The Goldenface and Mutant 2 Yellowface alleles would be seen to function in a straight forward manner like the Parblue varieties in other species of parrot. A single Yellowface gene would put yellow pigment into the bird and a second dose would simply produce a deeper yellow. No longer would fanciers have reason to regard the Yellowface alleles as suppressor genes.

Our whole perception of Yellowfaces has been moulded by a random event in history. Simply changing the order in which the Blue and Mutant 1 Yellowface alleles appeared would cause a 180 degree reversal of how Yellowface budgerigars are perceived. Unfortunately for Yellowface breeders, the wrong allele appeared first.

All of this might be good enough to correct the situation in Mutant 2 Yellowfaces and Goldenfaces but we are still left with the question of why does the double factor Mutant 1 Yellowface look blue? This is a very difficult question to answer in terms of the way genes are known to work. Perhaps we have been asking the wrong question. This brings me to the second detail the hobby has overlooked: There is no rule that says there cannot be more than one allele for blue.
As it now stands we are expected to believe there is one Blue allele (bl) and three Yellowface alleles (bl yf1, bl yf2, bl Ay) . What is wrong with there being two different Blue alleles (bl 1 and bl 2 ) and only two Yellowface alleles? What defines a variety?
In the double factor bird we see the action of a gene in its purest form. The double factor bird defines the variety. Therefore if the double factor Mutant 1 Yellowface looks Blue then a Blue is what it is. It is a Blue (bl ) allele not a Yellowface allele.
This is how I have come to regard the Yellowface and Blue series. For my own purposes I have designated the common allele as Blue Mutant 1 (bl 1), and I have redesignated the traditional Mutant 2 Yellowface allele as Blue Mutant 2 (bl 2 ). The birds are still all the same, I have merely shifted my perspective a bit. Instead of having to answer the difficult question why does the double factor Mutant 1 Yellowface look Blue?,¯ the question has become why is it when you combine two different kinds of Blue you get a Yellowface budgerigar?. With a bit of knowledge of biochemistry this problem is n`t too difficult to solve. To understand how this can happen we have to understand how genes work in the first place.

Genes are like blueprints

Genes are sets of instructions for making proteins. Genes are like the blue-prints, the proteins actually carry out all their functions. If genes appear to behave in a peculiar manner such as two different kinds of Blue producing a Yellowface, then we have to see if it is explainable in terms of the kinds of proteins the genes could be making. Most of the proteins genes code for are enzymes. Enzymes are the proteins which perform the chemical reactions in the body. In a Green bird the gene for Green (designated bl +) codes for an enzyme which is responsible for the production of yellow pigment. The bl 1 allele is a mutant form of the gene therefore it produces a slightly different version of the enzyme. This bl 1 enzyme is defective so no yellow pigment is produced and we have a Blue bird, but it is important to note that the enzyme is still there. Mutant 2 Blues have no yellow pigment because the bl 2 allele likewise produces a defective enzyme. Because the bl 1 and bl 2 alleles are different each version of the enzyme has a different defect.
1. Mutant 1 Blues have two bl 1 alleles so can only make bl 1 enzyme; therefore no yellow.
2. Mutant 2 Blues have two bl 2 alleles so can only make bl 2 enzyme; therefore no yellow.
3. Mutant 1 Yellowfaces have one bl 1 allele and one bl 2 allele; and so can make both types of defective enzyme.
The question now becomes , how can two kinds of defective enzyme work together to make yellow pigment?

Protein structure

To find the answer to this question we have to look at protein structure. Proteins are made as chains of amino acids and are reconfigured into the final product afterwards. Different kinds of enzymes have different structures. Some kinds of enzymes are made of only a single protein chain. Others are made of two copies of the protein chain. Still others are made of four copies of the protein chain.
If the enzymes produced by this series of alleles consisted of only a single protein chain then there would be no Mutant 1 Yellowfaces. Birds with the genetic constitution bl 1/bl 2 would look like ordinary (white-faced) Blues. What the Mutant 1 Yellowface tells us is that the enzyme which produces yellow pigment is made of more than one protein chain. It might be two protein chains it might be four. For the purposes of this discussion it is reasonable to assume the enzyme is made of only two protein chains until proven otherwise.
Based on the evidence at hand:
4. Pure Greens (bl + /bl +) produce only bl + enzyme.
5. Mutant 1 Blues (bl 1 / bl 1 ) produce only bl 1 enzyme.
6. Mutant 2 Blues (bl 2 /bl 2) produce only bl 2 enzyme.
7. Mutant 1 Yellowfaces (bl 1/bl 2 ) produce both bl 1 enzyme and bl 2 enzyme.

However since the protein chains are assembled into the enzymes randomly Mutant 1 Yellowfaces also produce bl 1/bl 2 hybrid enzyme. It is the bl 1/bl 2 hybrid enzyme that produces the yellow pigment. Each blue allele makes an equal contribution to the hybrid enzyme. The Mutant 1 Yellowface is really only a by-product of the interaction of two different kinds of Blue. There is no Mutant 1 Yellowface allele or more precisely it was incorrectly named in the first place. Neither of the two blue alleles is dominant or recessive to the other. They stand at the bottom of the allelic dominance hierarchy on an equal footing; bl + > bl Ay > bl yf2 > bl 1, bl 2. This type of inheritance pattern has sometimes been called allelic complementation.
The Mutant 1 Yellowface is a composite variety along much the same lines as the full body coloured Greywing. The full body coloured Greywing has one greywing allele and one clearwing allele, but there is no full body coloured greywing allele.

Why not Creamface?

Here in Australia the Mutant 1 Yellowface is commonly referred to as the Creamface. The term is both descriptive and does not mislead fanciers into thinking there is an actual Yellowface gene involved. I will adopt the term Creamface for much of the rest of this discussion. By knowing how to read the clues, the Creamface has already given us much information:
1. There are two different alleles for Blue.
2. The enzyme responsible for the manufacture of yellow pigment is made of more than one protein chain.
3. The two Blue mutations occurred in different locations on the original gene for Green and would therefore be classed as heteroalleles with respect to each other.
The reasoning behind the third point is as follows:

In order for the hybrid bl 1/bl 2 enzyme to function at all, the bl 1 and bl 2 protein chains have to be able to at least partly compensate for each others defects. This can happen if the defect in the bl 1 protein chain is in a different location to the defect in the bl 2 protein chain. The bl 1/bl 2 hybrid enzyme works, but doesn`t work particularly well, which is the main reason the yellow is so pale. After all, each hybrid enzyme molecule still contains two defects. Since the defects evidently occur in different locations along the protein chain it automatically follows that the mutations occurred in different locations along the original gene for Green.

Part 2 - Two kinds of yellow pigment!

The true Yellowfaces

Part 1 of this article focused on the dubious status of the Mutant 1 Yellowface. The Goldenface and Yellowface Mutant 2 by contrast are both true Yellowface alleles in which the double factor birds are visual Yellowfaces. The breeding pattern of the two varieties is identical, the difference being that the Goldenface has a deeper yellow than the Mutant 2.
In Australia Goldenfaces have erroneously been referred to as Mutant 2 Yellowfaces in various publications, the authors having been largely influenced by the rather superficial treatment of the two varieties in Genetics for Budgerigar Breeders. Consequently both indigenous Mutant 2 Yellowfaces and Goldenfaces have been unwittingly lumped together under the banner Australian Yellowface, leading Yellowface fanciers to believe they are dealing with a single highly variable variety. Evidently in Australia the Goldenface allele appeared on the scene first and when the Mutant 2 allele came along it was not recognised as a separate variety because the two varieties have the same breeding pattern. One doesn`t have to look long or hard in Sydney pet shops to find Yellowfaces which fit the description of the proper Mutant 2 among even the smallest of pet quality birds.
There has been much debate over whether to classify Yellowfaces as blue birds with yellow pigment added or as green birds with some of the yellow removed. This is really a hollow argument and is more of a reflection of how the human mind works than the natural order of things. Some scientists have described the human mind as a great dichotomising machine. We like to pigeon-hole things in either one category or another such as, black or white, right or wrong, good or evil.
In the case of budgerigars it seems we like to think of Yellowfaces as either green series or blue series. Have a look at the genetic formula for the double factor Goldenface bl Ay / bl Ay . It can hardly be called a green series bird because it doesn`t actually have the green allele in it, nor can it be called a blue series bird because it doesn`t have a blue allele in it either.
The double factor Goldenface belongs in a category of its own as an intermediate form being neither fully blue nor fully green. Parblue would be the operative word. The same principle applies equally to the double factor Mutant 2 Yellowface bl yf2 / bl yf2.

The Goldenface

The double factor Goldenface is the bird which really defines this variety. The golden yellow pigment is found mainly in the cap, face, and mask. The yellow pigment is also found in the short lateral tail feathers to varying degrees. In adult feather traces of yellow pigment can also be found scattered throughout the rest of the body but the body colour remains predominantly blue.
Double factor Goldenfaces produce only bl Ay / bl Ay enzyme. Unlike the enzymes in the two kinds of blue, the Goldenface enzyme is not as severely damaged so the ability to produce moderate quantities of yellow pigment is retained.
In addition to the double factor there are two types of single factored Goldenfaces. The more familiar traditional type (bl Ay/bl 1 genetically) tends to become quite green and is the type which actually comes to mind when one thinks of a single factor Goldenface. The other kind of single factor Goldenface has the genetic formula bl Ay/bl 2. They tend to resemble double factor Goldenfaces in their distribution of yellow but are noticeably paler. The latter are traditionally thought of as Goldenface - Mutant 1 Yellowface composites. However as was pointed out in Part 1 there is no Mutant 1 Yellowface allele. Therefore there can be no Goldenface - Mutant 1 Yellowface composite. These birds are in reality an alternate form of single factor Goldenface based on the Mutant 2 Blue allele bl 2 instead of the Mutant 1 Blue allele bl 1.
Single factor Goldenfaces produce three versions of the enzyme but the total quantity of all three enzymes is no greater than the amount produced by the double factor Goldenface.

Blue 2 Single-factor Goldenfaces
Blue 2 single factor Goldenfaces (bl Ay /bl 2 ) produce Goldenface bl Ay enzyme which is known to be functional, Mutant 2 Blue (bl 2 ) enzyme which is known to be defective, and bl Ay/bl 2 hybrid enzyme. It is difficult to judge whether this particular hybrid enzyme is functional or not. If it is functional then it doesn`t work particularly well and adds little yellow to the final appearance of the bird. The bulk of the yellow pigment would have to be produced by the Goldenface fraction (bl Ay/bl Ay). Since the Blue 2 single factor Goldenface produces less functional enzyme than the double factor Goldenface, the Blue 2 single factor bird is a paler shade of yellow than the double factor bird.

The fact that the bl Ay /bl 2 hybrid enzyme seems to contribute little if any yellow to the bird tells us that the bl Ay and bl 2 protein chains do not compensate for each others defects. This in turn suggests that their respective defects occurred on or about the same location along the protein chain. It therefore follows that the mutations which produced the bl Ay and bl 2 alleles occurred on or about the same position in the original gene for green. When two different mutations occur in the same site within a gene producing two distinct alleles they are termed homoalleles. It would require DNA sequencing to determine whether we have a pair of true homoalleles or whether we simply have two alleles differing in two different sites which are in approximately the same position. However since the bl Ay and bl 2 alleles seem to behave like homoalleles I shall refer to them as such in order to simplify this discussion.

Blue 1 Single-factor Goldenfaces

The traditional single factor Goldenface is the Blue 1 single factor Goldenface bl Ay /bl 1. These are the birds which cause fanciers so much frustration because of the spillage of yellow pigment into the body colour which turns the bird green. Blue 1 single factor Goldenface bl Ay /bl 1 produce Goldenface bl Ay enzyme, Mutant 1 Blue (bl 1 ) enzyme which is known to be defective, and bl Ay/bl 1 hybrid enzyme. The Blue 1 single factor Goldenface differs from the Blue 2 single factor mainly in the kind of hybrid enzyme it produces.
In Part 1 it was determined that bl 1 and bl 2 are heteroalleles and that the protein chains they produce compensate for each others defects well enough to manufacture the modest quantities of yellow pigment seen in the Creamface. It was determined above that bl 2 and bl Ay are homoalleles which means that bl 1 and bl Ay are likewise heteroallelic with respect to each other. The bl 1 and bl Ay protein chains compensate for each others defects very well and produce an extremely active hybrid enzyme. It is the bl Ay/bl 1 hybrid enzyme which is responsible for the over-production of yellow pigment turning the traditional single factor Goldenface green.

Breeding strategy

For decades breeders have been advised to pair single factor Goldenfaces back to common blues and select from those young with the least amount of spillage. The irony of it all is that it is the common allele for blue itself which is instrumental in the overproduction of yellow pigment. Consider the birds with the following genetic formulae bl Ay/bl Ay, bl Ay /bl 2, and bl Ay /bl 1. The first two have the desired distribution of yellow and the last exhibits substantial quantities of yellow spillage. What is instrumental in producing the desired distribution of yellow is not the doubling of the bl Ay allele but the absence of the bl 1 allele, i.e. the inability to produce the active hybrid enzyme bl Ay/bl 1. The bl 1 allele is not the innocent bystander it has been generally assumed to be.
Since bl Ay/bl 1 hybrid enzyme is part and parcel of the traditional single factor Goldenface it cannot be bred out of the bird. The enzyme and the associated yellow spillage will be a perpetual problem. History has born that out. Goldenfaces look the same as they did sixty years ago. The best thing to do is to avoid the bl Ay/bl 1 hybrid enzyme in the first place by breeding the alternate single factor Goldenfaces bl Ay /bl 2. However breeders have traditionally been advised against breeding Blue 2 single factor Goldenfaces bl Ay /bl 2 because of the erroneous belief that they are breeding a bird which is not only a double factor Yellowface but a composite between two varieties bl Ay /bl yf1.
In a world in which the two blue alleles had been discovered in the reverse order breeding our bl Ay /bl 2 Goldenfaces would be standard practise while our traditional greenish single factor Goldenfaces bl Ay /bl 1 would be regarded as the bane of Yellowface breeding. What has made the Goldenface a problem variety is that the wrong blue allele was discovered first coupled with the subsequent failure by the hobby to realise that if the double factor Mutant 1 Yellowface looks like a blue then it most probably is a blue.

Yellowface Mutant 2

The Mutant 2 Yellowface represents another allele in the series which like the Goldenface produces its own unique version of yellow-producing enzyme. It has the same breeding pattern as the Goldenface and the same foibles including spillage of yellow into the body colour due to the overactive enzyme bl yf2/bl 1. The only real difference between the Mutant 2 and the Goldenface is that the Mutant 2 enzyme is less efficient at producing yellow pigment than the Goldenface enzyme.
Composites between Mutant 2 Yellowface and Goldenface do exist and likewise produce three versions of the enzyme namely bl Ay/bl Ay enzyme, bl yf2/bl yf2 enzyme, and bl Ay/bl yf2 hybrid enzyme. All three versions of the enzyme are active but the hybrid enzyme is not overactive. These birds closely resemble the double factor Goldenface. The Mutant 2 Yellowface allele can therefore be regarded as another homoallele of bl Ay and bl 2.

Two yellow pigments

There is a whole other dimension to budgerigars to which the hobby appears to be oblivious since I haven`t seen anything on this topic discussed in the budgerigar press. The budgerigar doesn`t contain one kind of yellow pigment in its feathers but two kinds.
A German scientist, Otto Völker, discovered in 1936 and 1937 that budgerigars and many other Australian parrots glow in the dark. Herr Völker found that parrots with yellow feathers have spots, stripes, neckrings, and other markings that glow when placed under ultra-violet light. Those of us old enough to remember the blacklight posters of the late 1960s can construct a mental image of what the phenomenon looks like.
The crests and yellow underwing feathers of the Sulphur-Crested Cockatoos glow brightly as do the yellow areas of the Cockatiel; to name two examples. This phenomenon has not been found in Lorikeets, African Lovebirds, or most South American species. It is thought that parrots see a wider range of colours than humans do, ranging into the ultra-violet spectrum.
I have seen photographs of museum specimens of wild green budgerigars taken under UV light. The part of the bird that glows corresponds precisely to the distribution of yellow in a well-patterned double factor Goldenface; the cap being the brightest area. This includes the stray traces of yellow seen scattered throughout the body on otherwise well-patterned double factor birds. To my mind this is no mere coincidence and it allows us to construct a more precise hypothesis to explain how the Yellowface alleles function.

1. Since the blue alleles block the production of both pigments the two yellow pigments must be chemically related and have steps of their biochemical pathways in common.
2. Double factor Goldenfaces and Mutant 2 Yellowfaces produce yellow pigment in precisely those areas of the bird where the fluorescent yellow pigment is found. Therefore it is reasonable to conclude that what we see in the aforementioned birds is the fluorescent pigment only. Evidently these two alleles produce versions of the enzyme which function to produce the fluorescent yellow pigment but fail to produce the non-fluorescent yellow.
3. We may conclude from point 2 that the enzyme produced by a green bird acts on two different substrates. One substrate would be the precursor of the fluorescent yellow pigment and another chemically similar substrate would be the precursor of the non-fluorescent yellow. Furthermore the split in the biochemical pathway which leads to the formation of two discrete substrates, and ultimately two discrete yellow pigments, occurs at a point prior to that governed by the Yellowface and Blue allelic series.
4. The active hybrid enzymes bl Ay/bl 1 and bl yf2/bl 1 produced by the common single-factor Goldenface and Mutant 2 Yellowface respectively appear to act on both substrates. It is expected that the undesirable yellow spillage seen in these birds consists of the non-fluorescent yellow pigment.
5. The active hybrid enzyme bl 1/bl 2 produced by the Creamface is more difficult to judge. Either it produces reduced quantities of non-fluorescent pigment        only, or it produces reduced quantities of both types of yellow pigment.

To verify or refute the points above the various genetic types of birds would have to be examined under UV light of the appropriate wavelengths. It would be unethical to expose live birds to UV light. Plucking feathers from strategic areas should suffice or alternately a collection of preserved whole skins could be assembled over time.

One aspect of the two types of yellow pigment appears to be less clear. We can see the limits of the distribution of the fluorescent yellow in double factor Yellowfaces, but it is more difficult to visualise the distribution of the non-fluorescent yellow. Do the feathers of the face, cap, and mask of green series birds contain both types of yellow pigment or does the distribution of the non-fluorescent yellow finish where the fluorescent yellow begins? The answer to this question has a bearing on an interesting theoretical question asked by fellow fancier Andrew Wisniewski:  Is there a reasonable possibility that a Whitefaced Green variety might one day appear?¯

It would be prudent to say that in the long term anything may be possible, even Red budgerigars. However to be more specific, if the feathers of the face of a green bird contains both kinds of yellow pigment then the answer would be no. A hypothetical allele which blocks the formation of the fluorescent yellow whilst permitting the production of the non-fluorescent yellow, would still produce a visually green bird but with a paler than average yellow face. On the other hand if the two yellow pigments are segregated into discrete zones then the answer is yes. In theory at least, we may one day see an allele which does what is essentially the reverse of the Yellowface Blue alleles. In summary

The alleles of the Yellowface and Blue series can be divided into two groups. One group contains the homoalleles (or alleles with mutation sites in approximately the same position) bl 2, bl yf2, and bl Ay. The other group contains only a single member thus far, the common allele for blue, bl 1. In the latter, the mutation clearly occurred in a different area of the gene giving it the property of being able to interact in such a way that a series of very active hybrid enzymes is produced.

There is no denying that the Mutant 2 Blue (bl 2 /bl 2) is every bit as much a true blue as the common Mutant 1 Blue (bl 1/bl 1) and should be recognised as such. The Creamface may be thought of as the Yellowface formerly known as Mutant 1.

The budgerigar produces two types of yellow pigment. Each allele or combination of alleles in the series codes for enzymes which synthesize these yellow pigments to varying degrees.


Völker, Otto. (1936) Ueber den gelben Federfarbstoff des
Wellensittichs [Melopsittacus undulatus]
Journal für Ornithologie 84, p. 618-630.

Völker, Otto. (1937) Ueber fluoreszierende, gelbe Federpigmente bei Papageien,
eine neue Klasse von Federfarbstoffen
Journal für Ornithologie 85, p. 136-146.

Boles, Walter. (1991) Black-light Signature for the Birds?
Australian Natural History, Vol 23 No. 10 p. 752.

© 2003 - Inte Onsman - MUTAVI Research & Advice Group

WebDesign © 2007 - Martin Rasek
Valid XHTML 1.0! Made with valid CSS 2.0! Dogma 4W
parrot picture