Richard Cryberg


Biosynthesis of eumelanins is widespread in the plant and animal kingdoms.  In plants commonly recognized colorations that are due to eumelanin like substances are such things as browning of apples or bananas or blackening fresh cuts on potatoes.  Much of the brown and black coloration seen in various fungi is due to eumelanin like substances.   However, not all dark plant-produced pigments are melanins.  Many are due to various other polymerized quinone structures such as napthaquinones or anthraquinones.  These are often formed as a result of the action of oxygen and tyrosinase, the key enzyme at the start of melanin synthesis, acting on the appropriate non-protein substrate.  Thus from an evolutionary standpoint formation of these types of pigment preceded the development of the melanin pathways 1(ff1).

In the mammals the main coloring agents are melanins in addition to pinks from hemoglobin in blood close to the surface.  Fish, reptiles and birds have two other common color forming mechanisms, lipochromes and structural colors.  Lipochrome simply means fat-soluble.  Brilliant reds and yellows formed from carotene or its derivatives are commonly deposited in birds fat, feathers and skin or scales.  The vivid yellows and reds such as seen in canaries for yellow or a cardinal for red are examples.  The red on a pigeon’s leg is lipochrome based2.  More muted yellows and reds seen in mammals or birds are often due to pheomelanins.  The so-called structural colors result from diffraction and refraction of light by the microstructure of feathers or scales to create a variety of colors ranging from blues to iridescence.  A great many birds show some effects due to structural colors on at least part of the feathers.  In general any blue seen in animal life is a structural color.  In both mammals and birds there are also some other minor pigment based colors that in some cases are very important.  As it is not obvious that these exist in pigeons I am going to skip them.

As a point of interest, the black ink from squid is among the most concentrated suspensions of melanin available.  This ink is largely uncontaminated with proteins and other cellular products and therefore has been of great value in learning about the chemical makeup of melanin.

A few words about how this is organized might help the reader.  The material presented in the main body is out of the published scientific literature and is well documented.  Most of it has been learned from species other then pigeons.  As biological processes are well preserved across species what we learn about pigment synthesis in animals other then pigeons is in general directly applicable to pigeons.  I have added a number of footnotes which contain unpublished observations I have made as well as what seem straightforward interpretations as applied to pigeons.  Without doubt some of my interpretations will prove modestly incorrect in the details as science moves forward.

Also I need to be very clear that nothing I say in this paper has anything at all to do with formation of the white areas of any of the various pied varieties.  The white areas in pied are the result of things like differential and selective migration of melanocytes in the embryo or with programmed cell death of melanocytes and not with turning pigment synthesis on in some regions of the birds body and off in others.  In pied variants the feathers in an area are either white or colored.  The colored areas of pied are governed by the rules given herein including those where the colored feathers show both color and white.

Plant Kingdom Melanins

The chemistry of formation of eumelanins in the plant kingdom seems to have been fairly well worked out 3.  There seem to be only three essential ingredients.  They are the enzyme tyrosinase, the amino acid tyrosine and oxygen.  Pheomelanins are unknown in the plant kingdom.   Tyrosine is oxidized to dopa that is then oxidized again to dopaquinone.  It is clear that both the enzyme tyrosinase and oxygen are essential for the second reaction, conversion of dopa to dopaquinone.  The exact chemistry of the first reaction is less clear.  During in vitro studies mixing tyrosinase, tyrosine and oxygen only leads to formation of dopa after a significant induction period.  A considerable amount of effort has been expended to understand the basis of this induction period 4.  It seems to mainly be a result of needing time to build up a small amount of dopaquinone.  Adding preformed dopaquinone or various metal ions to the reaction mix can considerably shorten the induction period.  This preformed dopaquinone then acts as the true oxidizer in conversion of tyrosine to dopa.  It is unclear what role the metals play although effective metals have two readily available oxidation states and likely simply act as co-oxidants together with the tyrosinase.  It is not clear if metals play a role in biological systems.

Dopaquinone, together with atmospheric oxygen is capable of slowly forming a variety of brown and black pigments in in-vitro experiments.  The rapidity of color formation and the color formed varies with conditions such as pH or presence of various free metal ions over and above the bound copper found in tyrosinase.  This plant based or in vitro produced pigment does not have the stability nor molecular complexity of animal produced melanins.  They are much lower in molecular weights also.

Amimal Kingdom Melanins

In animals there is a regulatory scheme that controls the production of tyrosinase as shown in Chart 1 5.  A key step in this chart is that MC1R modulates the activity of the tyrosinase produced.  This modulation is caused by phosphorylation of the tyrosinase produced by the gene and increases the enzymatic activity.   Different activities of tyrosinase lead to very different phenotypes as low activities cause the chemistry to proceed mainly down the pheomelanin path while intermediate or high activities force the chemistry down the eumelanin path 6. Dominant MC1R mutants are not at all uncommon which lock the receptor site in a full on state 7, 8,9,10.  These full on states are called constitutive mutations.  With these mutations the MC1R site is in a full on state even without the help of melanocortin stimulating hormone.  In these full on states production of eumelanin is very high leading to black or near black phenotypes.  There are also some other kinease enzymes involved in the overall modulation scheme.  Their exact role is not yet understood in detail.  Nonetheless, it is clear that the things that happen before actual pigment synthesis even starts govern much of the final color produced.

Chart 1
Chart 1.  Formation of Tyrosinase

Eumelanin Formation
Chart 2 summarizes the well-proven chemistry of eumelanin formation 11,12.  Many of the steps have been proven by techniques such as adding model compounds to tissue samples where the model compound would be expected to only be able to move one or two steps down the synthesis pathway before stopping due to not fitting the needs of further steps.  All intermediates shown have been either isolated or proven to exist as transient species by such instrumental techniques as nuclear magnetic resonance or mass spectral examination of reaction mixtures during the course of the reaction.

Clearly all the details of what happens in the processes in this chart are not yet known.  For instance there must be enzymes that release free tyrosine as raw material for melanin synthesis involved.  Tyrosine deficient diets are known to reduce pigment in animals.  Other mutants may well produce products that promote or antagonize each step in this early part of the pathway resulting in increased or decreased pigment levels.  The point is that by considering the steps shown and such promotion or antagonism affects a great many of the known color mutants can be explained even before pigment synthesis starts.  This illustrates why much of the future progress in understanding exactly what causes different mutant affects is going to be discovered in lab studies and not in the breeding loft.  As breeders our main job is to find and characterize new mutants these days.  That cannot be done in the lab, at least not just yet.

Chart 2
Chart 2.  Biochemical Synthesis of Emuelanin

Tyrosinase seems essential only for the first two chemical steps that result in dopaquinone.  While tyrosinase is not essential it may promote some of the subsequent reactions in vivo.  The wild type gene responsible for production of tyrosinase is universally recognized as the locus responsible for albinism when the gene is defective.  Tyrosinase becomes more complex in higher life forms, but essentially always consists of a fairly large protein which complexes copper.  In plants and fungi only one copper is present per tyrosinase.  In vertebrates tyrosinase is a considerably larger molecule and complexes four coppers.  It is this copper which is the actual oxidant in conversion of dopa to dopaquinone.  The oxygen required is simply to reoxidize the reduced copper.  Any genetic defect which makes the enzyme incapable of binding copper will render the enzyme 100% ineffective with the end result being the organism is melanin free and a pure albino.  But, as often happens in biology we know of examples of near albinos where the enzyme apparently is rendered almost inactive but not quite 100%.  In these cases the copper binding sites are intact but the enzyme does not have the other structural characteristics needed to be an effective catalyst.  In humans we know today of over 60 alleles of this gene that lead to albinism 13 (ff2).

It seems sure that some genetic influence impacts some steps in this sequence.  However at this date no genes have been proven to influence the chemistry.  The eumelanin pathway leads to brown and black products and in very rare cases to red products. Dopachrome itself is red and has been isolated as an end product from the sea worm Halla parthenopea 14.  Also in hair some of the very reddest hair examples lead to degradation products consistent with eumelanin. Elemental analysis of this red pigment is inconsistent with pheomelanin 15.  Little seems to be definitely known about what exactly causes eumelanin to range in color from red on rare occasions to the normal browns and blacks.  The reds seem to involve some oxidative degradation of the 5,6 dihydroxyindole units resulting in bleaching much as happens in hair treated with hydrogen peroxide.  The difference between black and brown seems to be in part due to the size and shapes of the pigment particles as well as total pigment concentration.  Size and shape can affect the way light is reflected and refracted causing different perceived colors.  A dark brown pigment will appear black when at high concentrations.  It is also widely accepted that melanins can consist of heteropolymers made of monomers from both eumelanin and pheomelanin 16.  Such mixed polymer molecules could appear brown in color.  However, it is also clear that some well explored species of domestic birds are known that produce both browns and blacks but at least to date have never been known to produce reds via pheomelanins.  Canaries and geese are good examples.  So such mixed polymers are not essential to production of brown pigments.  It is clear, without doubt, that whatever causes phenotypic color to shift between black and brown happens during or after formation of the polymeric products.  There is absolutely nothing in the small molecule chemistry that impacts this color shift.

Pheomelanin Formation
At the dopaquinone step the synthesis pathway forks.  Chart 3 shows the chemistry after this fork. The amino acid cysteine combines with dopaquinone followed by chemistry of this adduct, eventually producing the well-recognized red pheomelanin pigments.  Which fork the chemistry follows is largely a matter of tyrosinase activity.  At low activity of tyrosinase the fork to pheomelanin is followed 17,18 while at high activity the fork to eumelanin is followed.  As the tyrosinase activity is modulated by MC1R, as shown in chart 1, MC1R turns out to be the primary determinant in red pheomelanin versus brown/black eumelanin pigment production.   There are many known recessive alleles of the MC1R gene that in effect break the function of the site resulting in red hair or feather color.  In mice four recessive alleles are well-characterized 19.  Recessives are also known in many other birds and animals.  All lead to red/yellow phenotype feathers or hair.  This modulation of tyrosinase activity can also be turned on and off by organisms.  Many animals are known where part of the hair is colored with pheomelanin and another part with eumelanin.  We clearly see this effect in many of the pigeon mutants where both colors are visible on different areas of the same feather.
Chart 3
Chart 3.  Biochemical Synthesis of Phoemelanin

As is the case in eumelanin little is known about the actual pigment formation process as soon as the small molecule sequence ends.  Again much of the small molecule chemistry after formation of the dopaquinone-cysteine adduct is formed seems to be close to spontaneous other then needing peroxidases at a couple of stages.  Just before the final polymerization trichromes are formed which are poorly characterized mixes of compounds that seem to be dimers and trimers.  These materials are red in color and have been isolated in small amounts from biological organisms as seemingly final products.  It is believed that trichromes are generally involved in the subsequent polymerization to form pheomelanin.  There are no known genes involved in the chemical steps subsequent to formation of dopaquinone.  This will ultimately prove to be untrue as undoubtedly genes are involved that control pigment polymerization, particle size and shape of pigment produced and pigment transport processes in both the eumelanin and pheomelanin pathways.  But most such genes may well prove to have little impact on phenotypic appearance (ff3). 

In some species of birds the ability to make pheomelanin seems to be absent.  Perhaps this is because they do not have sufficient free cysteine present to allow production of the cysteine adduct with dopaquinone.  In such cases broken versions when homozygous simply result in a white bird.  An example would be snow geese 20.  Interestingly this is the only example of a co dominant MC1R allele that I have run into in my searches.  It would be reasonable to expect that some versions of melanocyte stimulating hormone might have intermediate activity and thus lead to co dominance but in the snow goose case the specific defect was identified in the MC1R gene itself.

As many animals are known to have upwards of 50 different genetic loci that impact the final color and pattern of pigment deposition it is obvious that the above chemical steps and known associated genes are only a fragment of the whole color forming process.  However it needs to be pointed out that many of the pigment deposition patterns found in feathers have been mathematically modeled and found to follow fairly simple differential equations 21.  This paper shows pictures of computer-generated solutions to the equations used while substituting various values into the constants.  Clearly the simulations generated essentially all the color distribution patterns I have ever seen on real feathers as well as the patterns I have seen under microscopic examination of feather parts.  This illustrates that great variation in phenotype can be achieved by very simple means with surprisingly simple genetic control.  The math used is much the same as the math that has been used for many years to model oscillation in electronic as well as chemical systems. All that is required to see real life feathers showing exactly the same affects is a relatively simple feedback mechanism in the chemistry of the pigment forming steps.   For example a gene that produced a product that both increased the production of tyrosine while also dropping in the production of this gene product in proportion to any one of the chemical products of pigment synthesis would fill the need.  What the above paper shows is by adjusting the intensity of feedback essentially any pattern of pigment deposition can be achieved.  Such simple processes can easily account for bars and checks for instance.  Or the many beautiful black stripes you see on the blue area of a pigeon feather at magnifications of only 50 or 100X.

We also know of bacterial studies that show that simple random noise can have phenotypic outcomes 22,23.  Random noise in the pigment production process could easily explain such genetic effects as seen in the grizzle or almond series of alleles where pigment production is turned on and off and even back and forth between two different pigments.  Just as stable oscillations can be governed by relatively simple feedback systems random noise can be generated by the very same systems.  In some cases all that is needed is to choose different constants for the same mathematical equations (ff4).


This paper would not have been possible without the help of several others who deserve recognition.  My wife, Dawn, did a significant portion of the internet searches and also cajoled our local library to get copies of several papers and also books on interlibrary loans.  Much of this information was critical in writing the paper.

Dr. Daniel Smith read and gave very helpful suggestions during the writing process.  Most of what is presented on the mode of MC1R action comes directly from his study of the technical literature.  Without his help this critical part of the pigment forming process would not have been covered in anything close to an adequate fashion.

David Rinehart also deserves recognition.  David was responsible for prodding me into starting down this study path and has cajoled and tormented me regularly for the past six months to get me to continue.  He is also the person who made me aware that genetic noise was a factor that needed to be considered when we talk about genetic mechanisms.  His very challenging questions have kept me thinking through the whole process of trying to learn about the known technology of pigment formation.  It was at his insistence that I got out my optical microscope and learned to see pigment particles and take pictures of what I saw.

I also need to recognize those who laid the groundwork in pigeon genetics for all the fine background information and thought provoking questions that they raised.  In particular Hollander, Sell and Quinn wrote books and papers containing actual data that are relevant and timely even if dated in some cases.  We need more actual published data to further our understanding.


ff1.  Enough is know today about melanin chemistry that we can make a fair reconstruction of the total evolutionary pathway that has occurred.  At some early time our single cell ancestors had a problem with UV light.  At that early point in earth’s history there was no oxygen in the atmosphere and thus no ozone layer to shield the surface from UV.  Thus while all life was then confined to the oceans the UV meant that organisms could not even venture too close to the surface.  Some primitive organism accidentally made a gene that produced a protein that bound copper and had enough oxidation activity to convert naphthalene adducts to napthaquinones.  Quinones in general are black and excellent UV absorbers, thus the organism derived some reproduction advantage and passed on its new gene.  When we look at the tyrosinase in today’s primitive organisms we find that they are small proteins and only bind one copper per molecule.

At some point fairly early on the organism had another genetic accident.  This accident produced enough free tyrosine in the cell to allow interaction with the already existing tyrosinase and oxidation of the tyrosine to dopaquinone.  The dopaquinone spontaneously underwent the needed reactions, aided by peroxidases where oxidation was required, to produce the first primitive melanin like products.  Again we find exactly this mechanism at work in some fungi and plants today.

With time the organism had more molecular accidents that resulted in duplication of the tyrosinase gene.  Perhaps a deletion or duplication in a noncoding region near the gene prevented exact lineup during sexual cell division and the nonaligned cross over resulted in one cell winding up with two primitive tyrosinase genes end to end.  The new gene had two copper binding sites and was a more efficient enzyme then the original version.  In fact, in today’s higher animals the tyrosinase enzyme has doubled again and binds four coppers and is several times the size of the tyrosinase found in today’s primitive organisms.

While the evolution of the tyrosinase enzyme was happening the rest of the cellular processes we find in modern animals was also evolving such that eventually pigment was only produced in special cells at in areas where it was actually needed.  Further the polymerization of the small molecules was better controlled to produce stable pigment in of the best shape and particle size to provide UV protection.

Well down the evolutionary pathway, well after multicell animals had evolved, an organism had a genetic accident that resulted in enough free cysteine being present in the same cells making tyrosinase that pheomelanins could be produced.  And today’s animals live with the benefits of all these genetic accidents.

So when the anti evolution crowd say complex processes are too complex to have evolved by accident they simply show a lack of understanding of how simple accidents far back in time can have been modified and improved in many small accidental steps and result in the complex biochemical processes we see today.  And it has only been in the last few years that sciences like biochemistry, biology, chemistry, engineering and genetics have been forced to start cooperative efforts that digging out the details of how such complex processes happened in small steps could be identified.

ff2.  If having 60 alleles at one gene locus seems excessive I suggest it is far past time for the pigeon community to get used to the idea of multiple alleles.  When examined at the molecular level multiple alleles are proving the norm rather then the exception.  For the human disease cystic fibrosis we now know of over 1000 alleles of the CF gene.  We are adding new alleles at a rate of over ten per month these days.  There is even a commercial company that has a screen in place to determine if an individual has one of more of over 1000 known defects in his genome.  The present cost is roughly $3 per defect screened for by this company.

ff3.  Joe Quinn with remarkable insight anticipated what we would learn about the chemistry of pigment formation when he wrote “The Pigeon Breeders Notebook.”  In that book he suggested considering all the bronzing genes as possible alleles of each other suggesting that he believed that on a chemical level the same chemical pathway caused all bronzes 24.  Well, Quinn came very close to getting it right.  He just stopped a little too soon.  In fact we know today that while all the bronzes are not alleles (but so did Quinn) exactly the same chemistry produces the pigment in every single one of them.  Further it is obvious that a variety of other genes, such as ash red, indigo, recessive red and recessive opal, also exert their influence at this same chemical step.  In fact, we now know that all pheomelanin pigment produced results from some action that causes dopaquinone to react with cysteine rather then simply ring close to an indole.  And we also know without doubt that many of the reds are really pheomelanin and not some unusual eumelanin thanks to the quantitative analyses of feather pigments published by Sell et al 25.

 Now, many of these bronzed colors also are perfectly capable of making eumelanin as well as pheomelanin.  For example I have microscopically examined the tail feathers of a hen ash red bird and clearly shown the presence of black eumelanin.  It has made little sense for quite a few years to think that ash red and brown were actually alleles rather then simply closely linked genes.  Understanding the chemical pathways that lead to pheomelanin vs eumelanin and understanding that the choice between black and brown eumelanin takes place at a very different point in the chemical pathway is simply another proof that ash red is not a brown allele.  As soon as you realize they are not alleles it is easy to understand how an ash red hen can make black eumelanin.

We have been sitting on phenotypic clues for years that hinted that many of the bronzing effects were caused by the same or very similar mechanisms.  Consider how much some homozygous recessive opals and homozygous indigos look like ash red for example.  I think it is very likely that the feedback system that turns red pigment production on and nearly turns black pigment production off acts on exactly the same one or two steps in the synthesis pathway for all the bronzing genes.  In fact it would be far more unlikely for this to not be true.  While we know that it takes low activities of tyrosinase to produce red pigment this is not the whole story.  What we do not know is what this low activity must be relative to.  For instance is it a low activity of tyrosinase relative to tyrosine or relative to dopa that is important?  Or perhaps relative to dopaquinone or tyrosine?  And are there other enzymes that also get in the act?  Until we know answers to these kinds of questions we are not going to know the whole story.

We are missing important genes that we should be seeing in pigeons.  For instance either recessive red or spread in pigeons is the same as the MC1R gene.  But it cannot be both as recessive red and spread are not even on the same chromosome.  In other species such as mice where large numbers of animals have been raised it is common to have both recessive red and a gene analogous to our spread gene as alleles.  So in pigeons we are either missing the dominant black phenotype that is an allele of recessive red or we are missing the recessive red phenotype that is an allele of spread.  It seems more likely to me that when the gene sequencer guys have the answer we will find that our recessive red is equal to MC1R.  My reason for leaning this way is easy.  In other species there are generally more alleles for the recessive red phenotype then for the spread phenotype.  Yet in pigeons we know of only two alleles for recessive red.   And ember is clearly less broken then recessive red itself.  But we have hints of a whole variety of alleles at the spread locus.  Some of the later do not even darken the albescent strip.  And others need all kinds of help to produce a decent solid black.  Some are able to turn a blue bar into a very nice black with no other darkeners.  There are a lot more ways in general to break a gene then lock it full on.  I think we simply have too many spread alleles to make it the best choice for MC1R.  It will be great when we know for sure.  The possibility exists that we do have the constitutive gene for MC1R and simply have not recognized it.  We know that both ember and recessive red do not cover the albescent strip on the tail feathers without help form other darkening genes.  We also know of apparent spread genes that leave this strip unchanged from wild type.  No one, as far as I know, has ever taken one of these poor spread phenotypes and checked to see if it was actually a spread allele or if it might instead be an allele of recessive red.  There is no real reason at this point to rush to the breeding loft and make such tests as we will probably know from lab studies shortly if recessive red is in fact the MC1R locus and at that point it will be far easier to check various spread variants in the lab and see if any are also at the MC1R locus.

The color brown in pigeons remains a bit of a puzzle.  As I have pointed out above brown coloration is not well understood.  Sell, et al showed clearly that a brown bar pigeon had less eumelanin and more pheomelanin in the bar areas then the blue counterpart.  In the smooth spread areas this same bird had considerably less eumelanin and almost no pheomelanin compared to wild type.  In spread brown both eumelanin and pheomelanin were considerably lower then in spread black.  Brown is well known to bleach easily when exposed to sunlight.  While the lower concentrations of eumelanin could tend to look brown rather then black the bleaching is a strong hint that the pigment in browns is already partially degraded upon deposition in the feather.  Work that shows that eumelanin based reds are actually the result of oxidative damage to the eumelanin would fit with our brown pigeons simply being due to lightly oxidized eumelanin.  Once some oxidation has happened it is generally easier chemically to get further oxidation.  Pigments exposed to sunlight are going to absorb UV and get kicked into either singlet or triplet states that can easily combine with atmospheric oxygen.  Such damage would be expected to show up to naked eye exam faster on browns then any other color due to the relatively low total pigment concentrations.

Just to make life interesting there are also some bird species known that are brown because of porphyrin based pigments 26.  These pigments are reported to be the main coloring agents in the feathers of bustards, owls and goatsuckers.  I know of no evidence that these pigments are either light sensitive or in pigeons.

ff4.  I do not want to get bogged down in a lot of math.  In the first place I would have to go back and relearn all kinds of stuff I have not thought about for over 40 years except in nightmares.  For anyone who is interested a good EE text on the math of electrical feedback systems would be a good place to start.  Then follow up with one of the recent books on chaos theory.  Between the two it should be rapidly obvious that simple linear and nonlinear feedback systems that tied into the front end of pigment synthesis can easily generate any of the pigment patterns we see, either naked eye or microscopically.  When I say pigment patterns I am referring to bars, checks and various bicolor types.  I am not talking about pied types.  These patterns include the effects seen in all the grizzles and almonds.  It is mainly a matter of picking the correct constants to feed into the equations.  By including the effect of low activities of tyrosinase leading mainly to pheomelanin and high leading to eumelanin you could easily understand why grizzles often show bronzing, particularly in the transition between eumelanin areas and white areas.  All you need to do is drop the tyrosinase concentration a bit slowly while in the eumelanin production mode and pigment production will switch to bronze and then white at a very low activity of tyrosinase.  No gene other then the grizzle gene is needed to explain such bronzing.  Exactly the same sequence would explain what we see in almond.

1. Giuseppe Prota, Melanins and Melanogenesis, Academic Press Inc (London) Ltd, 1992, pp. 10-11.
2. W. F. Hollander, Origins and Excursions in Pigeon Genetics, pp. 31, 141.
3. Giuseppe Prota, Melanins and Melanogenesis, Academic Press Inc (London) Ltd, 1992, pp. 1-7.
4. Giuseppe Prota, Melanins and Melanogenesis, Academic Press Inc (London) Ltd, 1992, pp. 42-57.
5. Professor Daniel A. Smith, Chemistry Department, Goshen College, private communication.
6. S. Kerie, J. Lind, K. Schutz, P. Jensen and L. Andersson, Melanocortin 1-receptor (MC1R) mutations are associated with plumage colour in chicken, Animal Genetics, 34, pp.  241, August 2003.  Also web address:
7. E.Theron, K. Hawkins, E. Bermingham, RE. Ricklefs, NI. Mundy, The molecular basis of an avian plumage polymorphism in the wild: a Melanocortin-1-receptor point mutations is perfectly associated with the melanic plumage morph of the bananaquit, Coereba flaveola. Web address:
8. Maria K. Ling, Malin C. Lagerstrom, Robert Fredriksson, Ronald Okimoto, Nicholas I. Mundy, Sakea Takeuchi and Helgi B. Schioth, Association of feather colour with constitutively active melanocortin 1 receptors in chicken, Eur. J. Biochem. 270, pp. 1441-1449(2003).  Web address:
9. Dongsi Lu, Dag Inge Vage and Roger D. Cone, A Ligand-Mimetic Model for Constitutive Activation of the Melanocortin-1 Receptor, Molecular Endocrinology 12(4): pp. 592-604, web address:
10. S.M. Doucet, M.D. Shawkey, M. K. Rathburn, H. L. Mays Jr, and R. Montgomerie, Concordant evolution or plumage colour and feather microstructure and a Melanocortin receptor gene between mainland and island populations of a fairy-wren, Proc. R. Soc. Lond. B, pp. 1-8, 2004, web address:
11. Aaron Bunsen Lerner and Thomas B Fitzpatrick, Biochemistry of Melanin Formation, Physio. Rev., 30: pp. 91-126(1950)
12. Giuseppe Prota, Melanins and Melanogenesis, Academic Press Inc (London) Ltd, 1992, pp. 6.
13. Richard A. Spritz, Molecular genetics of oculotaneous albinism, Human Molecular Genetics, 1994, 3, pp. 1469-1475.
14. Giuseppe Prota, Melanins and Melanogenesis, Academic Press Inc (London) Ltd, 1992, pp. 64.
15. Giuseppe Prota, Melanins and Melanogenesis, Academic Press Inc (London) Ltd, 1992, pp. 68.
16. Giuseppe Prota, Melanins and Melanogenesis, Academic Press Inc (London) Ltd, 1992, pp. 69.
18. Johnathan L Rees, The Melanocortin 1 Receptor (MC1R): More Then Just Red Hair. Pigment Cell Res. 13: pp. 135-140(2000)
20. Nicholis I. Mundy, Nichola S. Badcock, Tom Hart, Kim Scribner, Kirstin  Janssen, and Nicola J. Nadeau, Conserved Genetic Basis of a Quantitative Plumage Trait Involved in Mare Choice, Science, 203, pp. 1870-1873(2004)
21. Richard O. Prum and Scott Williamson, Reaction-diffusion models of within-feather pigmentation patterning, Proc. R. Soc. Lond. B (2002) 269, pp. 781-792.
22. Yina Kuang, Isreal Biran and David R. Walt, Simultaneously Monitoring Gene Expression Kinetics and Genetic Noise in Single Cells by Optical Well Arrays, Anal. Chem., 2004, 76, pp. 6282-6286.
23. Jonathan M. Raser and Erin K. O'Shea, Noise in Gene _Expression: Origins, Consequences, and Control, Science, 23 September 2005, Vol; 309, page 2010-2013
24. Joseph W. Quinn, A Pigeon Breeder’s Notebook: An Introduction to Pigeon Science, 1971, pp. 76.
25. E. Haase, S. Ito, A. Sell, and K. Wakamatsu, Melanin Concentrations in Feathers from Wild and Domestic Pigeons, Journal of Heredity, 1992, pp. 67-67.
26. Geoffrey E. Hill and Kevin J. McGraw, Bird Coloration, Vol 1, Harvard University Press, Cambridge, Massachusetts, 2006, pp. 245.

Copyright 2006 by Richard Cryberg.
# # #

Article Index

Home Page