by Richard Cryberg

February 20, 2009


Abstract:  The objective of the current study was to get some kind of minimum count on the number of bronzing genes required to produce a quality Modena bronze phenotype.  Modena bronze is shown to consist of at least three different bronzing genes.  Two of these genes bronze the coarse spread pattern area on the wings while the third acts like a very muted kite and shows mainly on the webbing of the primary flight feathers and coverts.  Two of these genes are fairly weak codominants, or possibly even very nearly recessive, when by themselves.  The third, the kite like gene that I am naming Modena flight bronze, is dominant.  The combination of all three as heterozygotes gives a bird with fair bronze coverage compared to good colored Modena bronzes.  However this bronze is very muddy colored versus when the three are homozygous.


Introduction:  There are many conflicting claims made about the inheritance of Modena bronze.  Most state it is recessive1,2,3 when crossed to wild type or at best gives only slight bronzing.  However Quinn states3 that on occasion fair bronze is seen in the F1s.  Sell states that mating two F1s will not produce the expected one in four well bronzed individuals indicating the trait is more then a single gene.  Quinn states4 that he spent five years trying to move Modena bronze into Rollers and finally gave up as the bronzes produced were so poor.  Gibson claims5 to have isolated a gene from Oriental Frills, that is part of the frill stencil package, which is dominant and produces Modena bronze phenotypes.  Unfortunately, none of these sources have provided publicly available data from so much as a single mating.  It seems well past the reasonable time to actually generate and present some breeding data for all to consider.


Discussion:  Modena bronze is actually a misnomer.  When all the needed genes are present as homozygotes the color is not bronze.  Rather the pattern areas are red.  Just as red as a recessive red that is also T pattern, spread and dirty.  It has none of the muddiness to the color seen in so many bronzes that include substantial amounts of black pigment.  Very few of the Modenas seen today in the US show this good red color, as few  Modenas are homozygous for all needed genes.  Modena bronze affects mainly the pattern areas in the wing.  On a check or T pattern bird it should extend clear to the shoulder.  On some birds it also extends to some minor extent into the head and tail feathers.  These regions outside the pattern areas are never a good red like seen in the pattern.  Rather they are a true bronze consisting of a mix of red and black pigment.


Based on hints from several people including Larry Long, Robert Mangile and James Gratz I mated a hen Gazzi T pattern Modena that showed excellent bronze to a dark check pattern Homer that showed no visible bronzing.  A few words about both these birds are in order before getting into details.


First of all why did I use T pattern?  Gratz suggested I use T pattern as in his words “the quality of the bronze shows better on T pattern then on bar.”  You can go to any show and confirm this by looking at the Modenas shown.  If you look at bar pattern birds you will see very decent coverage of the bar area with bronze almost universally.  This is even true of birds with a very poor bronze that has so much black pigment it is hardly even a bronze.  But when you look at checks and T patterns you will see that in the majority of birds shown the bronze does not push clear up to the front of the shield.  In some birds the bronze hardly pushes past the bar area leaving the whole front of the shield area looking like a normal check.  It is obvious that if I had just used bar pattern birds I would have wasted a lot of time and effort, as I would only be able to observe less then half of the total impact various genotypes have in terms of overall bronzing power.  In my results I have seen no real difference to date between using a dark check or T pattern and in fact many of the birds I have raised are dark checks.


The Modena I picked for this project was a careful choice.  Again when you go to shows and look at the Modenas you see great variation in the bronzing both in terms of extent towards the front of the shield and in terms of color. I also wanted a T pattern bird that had solid bronze shields showing as little black tips on the feathers as possible.  In fact the only bird I could find that had this quality of bronze turned out to be a very old style Modena, much like they looked 30 or more years ago.



Picture 1.  Hen Modena


As can be seen this hen is gazzi and the bronze pushes into the stray mismarked colored feathers on the hocks and chest.  She also showed some bronzing on her head and even a trace in her tail.  This hen was slightly under Homer size and weight.  She also was dilute.  I would rather have used a non-dilute and non-gazzi bird but my first concern was the quality of the bronze and she was the best overall I could find at the time I started this project.


The Homer I used came from a family that has seldom showed any hint of bronzing of any type.  In only one known relative of the Homers I used have I seen bronze and in that case it was a very slight bronzing in the primary webbing.  All blue birds seem to have a bronzing component in them based on Haase, et al6.  So I doubt there is such a thing as a truly bronze free pigeon.  I have often, but not always, seen a slight red cast in blue feathers under the microscope at 100X in birds that to the naked eye looked like clean blue.


The hen Modena mated to the Homer produced three young F1s.  Two proved to be cocks and one a hen.  All three are phenotypically as alike as peas in a pod in terms of the bronze.  Nearly all birds raised were rated as to bronze extent and color both in nest feather and after the first molt.  No significant changes were seen upon molting.  All three F1s showed the Modena flight bronze about equal to the same bronzing shown by the Modena.  This bronze in the flight webbing will be much more clearly shown in a later picture.  In picture 2 of one of the F1s the flight bronze shows best on the underside of the flight feathers on the far side of the bird.  The bronze coverage of the bar area was close to normal for a Modena bronze, although the line between the black area and bronze area is not nearly as well defined as in a good colored bird.  The bronze pushed fairly well up towards the front of the wing shield as seen in the picture 2, although the bronze coverage of each feather was poor by the front of the shield.  The color quality of the bronze is very poor.


Picture 2.  F1 from Modena X Homer, Band 8985-06


Gene Count:  My first objective was to try and get some count on the minimum number of bronzing genes involved.  Hollander gave us directions7 on how to tackle a problem like this.  What Hollander said is when an F1 is intermediate the proper next step is to mate this bird to either pure wild type or pure mutant type.  This is just common sense.  I chose to mate the three F1s to wild types.  I reasoned that there were likely at least two bronzing genes in the mix if not more.  The reason for this idea is quite simple.  Examination of Modenas in shows had revealed that a considerable number did not have the flight bronzing at all, but still had a fair amount of bronze in the pattern area.  Also I was surprised at the amount of bronze in my F1s.  When I first showed them to Quinn for instance he was amazed at the bronze in the F1s.  He stated they were far bronzer than he had ever come close to getting in F1s and as good as he had gotten after several years of matings trying to move it into Rollers.  This hinted to me that there were perhaps more than two bronzing genes in my starting bird.


The important data from this mating of F1s back to wild type is the birds at the extremes.  That is the birds that look like the F1s and the birds that show no bronzing at all.  From  these two numbers either together or individually I can get a count on the minimum number of bronzing genes.  During production of these B1 birds I raised a total of 26 classifiable young.  A few were so pied I could not classify them.  Several from one of the F1s were recessive opal that had come in from the original Homer cross.  No attempt was made to classify the recessive opals.  Of the 26 classifiable birds 2 were wild type with no visible bronze at all and 4 were as bronzed as the F1s with the rest clearly  intermediate between wild type and the F1s in bronzing.  I also identified 8 birds that seem to just have the flight bronze and no shield bronze.  However this classification is error prone as some of them may have had so little shield bronze it was masked by the flight bronze.


The rest of the birds with intermediate bronzing were impossible to classify into groups that I could really be confident were correctly grouped.  The shield bronzes were a near continuum from very little bronze at all to considerably more bronze but clearly less than the F1s.  Further the color of the bronzes also varied.  Perhaps with a lot more experience I will be able to see things I cannot see today.  At any rate the range of bronze phenotypes sure hints strongly that more than simply two genes are involved.  These shield bronzed birds consisted of birds that did and did not show the flight bronze.  Table 1 summarizes the expected results from 26 B1s if two, three or four codominant genes are present as well as the actual observed results.


Table 1.

Number of Codominant

Bronzing Genes



Expected Number of Each Phenotype


Wild Types                Like F1s








  Experimentally observed 















Clearly, by inspection, the experimental data does not fit with either a two gene or a four gene model particularly well.  Both experimentally classifiable groups are smaller than expected for a two gene model and larger than expected for a four gene model.  The total observed in the two experimental classes was 6.  A binomial statistical analysis8 says for the two gene model the odds of getting 6 or less of these two phenotypes is 0.5% probable.  Likewise for the four gene model the odds of getting 6 or more of the experimentally observed phenotypes is 10%.  The normal statistical rules for discarding a hypothesis are to discard when it becomes less then 5% probable.  The two gene model is far below this probability and should be discarded from further consideration.  The four gene model just barely manages to avoid failing.  But it is so close to failing it seems unlikely.  It is possible that there are full recessive genes in the mix that do not show in either the F1s or the back cross birds.  Such a gene would not have been detected in these matings.  If such a gene exists a four, or even more, gene model would be correct.


On the other hand the three gene model fits very well.  Binomial analysis says the odds of getting at least 6 is 48% while getting not more then 6 is 52%.    For the moment the data demands  that there are at least three genes required to get full expression of a quality Modena bronze.


Table 2 documents the results of all the F1s and B1s produced, including phenotype observations for the B1s.



Table 2.  Summary of Matings

            Parents        F1s

           Cock Homer
           Split band number 40

           Hen Modena
           band number NPA 144-02

Split number 9



          Parents         B1s

         Cock F1
         band number 8961-06

         Hen Homer
         band number  8973-06


    656-07 Wild Type

   651-07 Wild Type

     671-07 Pattern Bronze*    

     701-07 Pattern bronze
     713-07 Pattern bronze

     691-07 Modena flight bronze**
     692-07 Modena flight bronze
     694-07 Modena flight bronze

   642-07 Modena flight bronze

     714-07 Modena flight bronze

       Cock F1
       band number  8985-06 


        Hen Homer
        band number  9010-06

    676-07 Like F1

     696-07 Like F1
     697-07 Like F1
     677-07 Pattern bronze
     682-07 Pattern bronze
     654-07 Pattern bronze
     707-07 Pattern bronze
     708-07 Pattern bronze
     679-07 Modena flight bronze


      Cock Homer
      band number  8992-06


      Hen F1
      Split band number  9

     675-07 Like F1
     679-07 Pattern bronze
     700-07 Pattern bronze
     703-07 Pattern bronze
     714-07 Pattern bronze
     649-07 Pattern bronze

      686-07 Modena flight bronze



*I have included every bird that showed obvious bronzing in the pattern all in one group.  The coverage of this bronze as well as the color varied greatly from bird to bird.  Some only showed a very small amount in part of the bar while in others it extended into the shield to a greater or lesser extent.  In general the color of bronze was even poorer than seen in the F1s.  Several of these birds also had the Modena flight bronze while others showed no trace of it.


** Some of these birds may have had small amounts of pattern bronze that were missed in the exam to classify them.



Work Currently in Progress:


The obvious future goals are to isolate each of the three bronzing genes both as heterozygotes as well as homozygotes.  The way to accomplish this isolation is to take select birds thought probable to have only one of the bronzing genes and mating back to wild type for the third consecutive generation.  If the bird under test has only one codominant bronzing gene 50% of the young, within normally expected random variation, should show the gene and 50% should be bronze free wild types.  It should then be easy enough to mate two heterozygotes to make a homozygous bird.


This last goal of making homozygous birds containing only a single bronze is important in a final determination of dominance, codominance or recessiveness.  The rules of genetics call for determination of dominance by comparison with wild type in animals that have no other interfering mutants.  It is obvious that in the case of Modena bronze the three bronzes cannot be expected to be independent of each other.  It is quite likely that they are more than simply additive.  Consider the well-known case of combining slipper with grouse to make muffs.  The muffs you get are far bigger than you would expect from simply adding the effect of grouse and slipper.  In fact when two independent mutants impact the same trait a combination of the two is often greater than additive.


I needed some book keeping system so I have assigned tentative symbols to these three genes subject to experimental confirmation.  Hollander1 was the first to specify a symbol for Modena bronze.  He considered it a single gene he named mahogany and symbolized ma.  Present breeding data indicates the muted kite like gene that bronzes the flight webbing is a dominant.  Thus I am calling it Mfb for Modena flight bronze.  The other two bronzing genes impact only the pattern areas and both may turn out to be weak codominants.  I am calling these two mahogany a and mahogany b after the Hollander precedent.  The symbols are Maa and Mab9.  It is likely that Mab will prove to be so close to recessive when by itself that the symbol may have to be changed to lower case.


The present state of research is, I believe, I have isolated Mfb as a heterozygote with no other bronzes present.  Picture 3 shows what I believe is simply Mfb with no other bronzes.  This bronze is exhibited mainly in the webbing of the flight feathers but also a bit in the webbing of the coverts. This bronze looks much like a muted kite to me.  Others, who I consider knowledgeable, assure me this is not the normal kite.



Picture 3.  Band number 642-07 Hetero Mfb


It does not show the intensity of bronze shown in a normal kite.  It also shows bronzing clearly on both sides of the rachis while kite tends to be much more bronzed on the proximal side of the rachis.  Mfb is easy to recognize on any pattern of blue Modena based on looking at many examples in Modenas at shows.  It shows up just as clearly on a bar pattern as on a T pattern.


Mfb does not seem to have been recognized as an individual trait by US Modena breeders.  Close to all blue Modenas exhibited in the US show this trait. European breeders have recognized this bronzing11 and consider it a flaw.  Thus, in Europe there is selection pressure to eliminate this bronze.  Curiously, it still appears in  birds that I have seen of European origin.  This implies, that it does enhance some aspect of the bronzing or it should have been easily eliminated long ago.




Picture 4.  Band number 682-07 Hetero maa


Picture 4 shows what I believe to be Maa.  The kite like bronze in the flight webbing is totally absent in this bird.  Compared to the F1 shown in picture 2 the bronzing in the pattern area is practically absent.  Even in the bar area where the bronze coverage is always best it is poor in bird 682-07 as only the upper half of the bar has a poor bronze.  And there is next to no bronze showing in the upper shield area.


From birds produced in the mating of F1s back to wild type a very few birds were produced with even less bronze coverage than shown in Picture 4.  These birds had no Mfb and only showed pattern bronze in the bars on two or three of the feathers closest to the birds back when the wing is folded.  I believe these may be hetero for only Mab.


Short Tail Feather:  Modenas have a distinctly shorter tail than wild type.  As I am getting data on the short tail trait for free I have monitored this aspect.  I measured the third tail feather from the outer edge of the tail after the molt to adult feathers.  I did this by pulling the feather and measuring the total length of the rachis including the part normally below skin level.  Wild type birds of roughly Homer size measured about 130mm.  Pure Modenas measured about 105mm.  All the F1s had tail feathers within 5mm of wild type.  Out of five young produced to date from a mating of F1 back to pure Modena two showed tail feather lengths of 110mm.  While the data is too limited to draw a final conclusion it seems short tail feather is a single gene recessive.  I propose the symbol stf for this trait.


Other Observations:  In both my own birds and also in birds I have seen in shows I have often seen black feathers on the upper leg area.  I know there has been speculation for years about sooty being a needed part of a well-bronzed T pattern Modena.  Yet in my crosses I have seen no hint of sooty as part of the mix.  I have seen these same black feathers on an occasional Homer in my loft.  Yet I do not have the sooty gene in those Homers.  I have no real idea what gene is involved in producing black feathers in this area.


I have been told10 that slight bronzing in the flight webbing can indicate the presence of hetero recessive white.  As I started with a gazzi marked bird all of my F1s and half of my B1s should be hetero gazzi.  I had some concern that this might be causing me some problem as perhaps the Modena flight bronze I am seeing was due to hetero gazzi.  However I also have a hen Modena that is hetero gazzi by breeding tests but totally lacks this bronzing in the flight webbing.  I have also seen well marked gazzi phenotype Modenas that did not show any hint of this bronzing.  At this point I am comfortable that hetero gazzi is not creating a classification problem with respect to Mfb bronzing.


A limited number of backcrosses of my F1s to more or less pure Modena bronze has yielded birds intermediate between the F1s and good Modena bronze.  Most of these backcrosses have had more bronze coverage, particularly in the shield area, than the F1s.  Most of them also have poorer color than a good Modena bronze should have.  Only a small percentage have had what looked like full Modena bronze.  The difficulty in recovering good color in a backcross further supports the idea this trait is a multi gene complex.


Every few months a picture is posted of a bird on the yahoo geneticsforpigeons group of some bird with considerable bronzing in the bar area when the bird was either a dark check or even a T pattern.  A year ago I saw a T pattern African Owl in a show that was a very nice example of such a bird.  Many have suggested that this was simply part of the Modena bronze complex.  I do not believe such bronzes have anything at all to do with Modena bronze.  The area covered by the bronze is different than seen in a Modena bronze when only part of the Modena bronze genes are present.  I have produced birds that show this coloration from crosses between a white Indian Fantail and blue Homers.  One such bird I raised is shown in picture 5.  Some of these birds look much like a T pattern bird with a bronze bar when the wing is folded.


Picture 5.  Indian Fantail X Homer cross


Testing has shown this bird is hetero kite, hetero recessive red and dirty.  Such birds look to me much like some I have seen posted from time to time.  Notice the difference in the distribution of bronze in the primary flights of this particular bird vs Picture 3 of Mfb.


Experimental:  All breeding experiments were carried out in individual breeding cages.  Food, water and grit were provided ablibitum.  The food was a mix of whole shelled corn and a 20% protein game bird grower pellet manufactured by Blue Seal Feeds, Inc and brand named AVIPELLS.  The corn and pellets were an equal weight mixture.



  1. Origins and Excursions in Pigeon Genetics, W. F. Hollander,  Second Printing 1994, page 115.

  2. Breeding and Inheritance in Pigeons, Axel Sell, 1994, Schober Verlags Publisher, page 106.

  3. The Pigeon Breeder’s Notebook:  An Introduction to Pigeon Science, Joseph W. Quinn, 1971, page 77.

  4. Private discussion with Quinn on his unpublished results.

  5. Genetics of Pigeons, Lester Paul Gibson, 2005, page 45.

  6. E. Hasse, S. Ito, A. Sell, and K. Wakamatsu, Melanin Concentrations in Feathers from Wild and Domestic Pigeons, Journal of Heredity, 1992, pages 64-67.

  7. W. F. Hollander’s PhD Thesis, Hereditary Interrelationships of Certain Factors in Pigeons, University or Wisconsin, 1937, page 2.

  1. The on line calculator used was:

  2.  Some pigeon genes have been symbolized by some letter or letters followed by sequential numbers when multiple genes contribute towards some common phenotype.  For example Ts1, Ts2, etc. This symbolization is clearly not allowed under the rules of genetic nomenclature.  Symbols ending in a number are specifically reserved for cases of paralog genes proven by DNA sequencing, certain recessive lethals known to be on specific chromosomes, or other highly special cases.  The practice of assigning symbols ending with a number to pigeons should be stopped and those genes presently symbolized in this way should be resymbolized providing experimental data can be provided to justify any symbol at all.  See this link, sections 2.3 and 2.4 for nomenclature details:

  3. Larry Davis, private communication.

  4. Axel Sell, private communication.



Back To Start Page

Email the author,  Richard Cryberg:

Copyright 2009 by Richard Cryberg
Permission is granted to download or copy for non-commercial individual use only.
The author retains all other rights under copyright.

Ronald R. Huntley
Web Page Designer
Duncan SC, 29334
phone: (864) 249-0276