http://www.exoticfinches.com

Warning: All rights reserved. This article appeared in the issue of the National Finch and Softbill Society Bulletin. Vol. 21. No. 5. Sept/Oct. 2004. (p. 6-9). Anyone wishing to reproduce this article for another bulletin, newsletter, article, journal, CD, or any other public forum needs express written consent of the NFSS and of the author michael@exoticfinches.com

If you are going to improve your birds through selective breeding it is essential that you understand the difference between a phenotypic trait controlled by a single gene and a trait that is quantitatively controlled (i.e. a polygenic trait). The breeding strategies for simply inherited traits can be quite different than for polygenic traits and, therefore, it is crucial that one is certain about what controls the trait before launching on a multigenerational breeding scheme.

What makes breeding for “standard” so difficult is that most of the genes that regulate traits such as shape, size, and feather quality are not simply inherited. More commonly, simply inherited traits are the color mutations where a single gene mutation alters a biochemical pathway for a specific pigment. Examples include albinism in society finches, blue body in Gouldian finches, and black cheek in zebra finches.

With simply inherited traits there tends to be an “either/or” category. For example, a Gouldian finch either has a white chest or it does not. A shafttail is either fawn or it is not. In polygenic traits there tends to be a continuum from one extreme to the other. For example, head shape in finches is a continuum between flat headed and round headed with everything in between existing. The adult weight of a clutch of six finches tends to range between x and y and not be either x or y.

Because of the way they can be categorized, simply inherited traits are often called categorical or qualitative traits whereas those that are controlled by many genes are quantitative traits and are not easy to put into defined categories.

One of the more distinguishing characteristics of simply inherited traits is that they tend to be unaffected or only slightly affected by environment. A blue body Gouldian or an albino society finch will never look wild-type (“normal”) no matter what light level, diet, or temperature scheme it is subjected to.

Polygenic traits on the other hand are affected by interactions among many genes with no single gene having a dominating influence. Environment can impact the final product since it is the delicate balance between many different genes that results in the expression of the final characteristic. Obtaining uniformity year after year is a challenge when quantitative traits are being manipulated. Variation due to environment has been demonstrated in humans where identical twins that have been separated at birth (e.g. by independent adoptions) and raised in very different environments do not look “identical.” While they are still unmistakably “identical” twins, they are not as “identical” as twins that are raised in the same environment (e.g. same diet, same exercise regime, same level of stress, substance abuse, etc.). As you will see in future articles, it is this environmental x polygenic interaction that can influence the show potential of a show quality bird.

In this article I will concentrate on the manipulation of simply-inherited traits, using color mutation as the standard example. One of the most rewarding things about finch breeding is trying to create the best line of a specific color morph, for example, having the best penguin zebra finches in the country or the nicest blue Gouldians at the show. To achieve such goals one often begins with an average representative of “your favorite mutation” and works toward making mutation birds equal in quality to any great line of the normal colored bird of that species. It is wonderful to go to a bird show and see a blue body Gouldian of higher quality than a normal or a fawn gray society finch larger than the winning chocolate (i.e. normal). To achieve such results one must know how to move single genes within a population. The goal is to move the mutant gene into a superior line. Ideally, after many generations you essentially have all the genes of the superior line with the exception of the one gene that changes the color.

I will not discuss the dominant single gene mutations for two reasons. First, they are rare. Most of the single gene mutations of interest are recessive. Secondly, dominant traits show up in the first generation of crosses between normal and mutant birds so it does not take a rocket scientist to move them around to different lines in a flock. In addition, I will also assume that you know simple Mendelian inheritance (e.g. Aa x aa yields 50% Aa, 50% aa). If you do not, there are numerous web pages listed at the end of this article that can help. My article in the last issue of the journal gave you definitions you might need.

Recessive single-genes can be hidden in birds that are split (heterozygous). We often do not have extensive pedigree data since most of us obtain stock at bird shows or from breeders that do not keep extensive pedigrees. Therefore, we are uncertain if a bird is split for a recessive trait, be it one that is desired or one that is not. In species, like zebra finches, there are so many mutations it can be a nightmare trying to find stock that has the genetic makeup that you desire. The only way to be certain if a bird is split for a recessive mutation is to perform a test cross, mating it to a bird showing the mutation (i.e. homozygous).

For example, suppose that you want to develop the best line of fawn zebra finches and know that if the bird also carries the recessive penguin mutation it might shrink their final size since penguin has a tendency to produce smaller birds. If you suspect that your zebra finch is carrying penguin you would need to cross it to a penguin zebra to make a “test cross”. On average half of the offspring would be expressing penguin if the suspected bird was split. A clutch of two, both non-penguin, is not a high enough number to be sure that the bird in question is “penguin free” since there is a 50% chance that a pairing of a split x a fully mutant bird could yield splits which, of course, look normal. A course in probability is beyond the scope of the NFSS journal but statistics show that in order to be 99% sure your bird is not split for the recessive trait you need to cross the bird in question to the mutant bird enough times to obtain at least seven non-mutant offspring and no mutant offspring. Remember, the lower the number of offspring the higher the probability that penguin can still be there hidden as a split.

It is impractical and actually biologically impossible to test most species of finches for all mutations since the life expectancy is too short to perform so many tests. Therefore, getting stock from breeders with extensive records is most helpful. In addition, in zebra finches, and other species that have many recessive mutations, it becomes critical to recognize undesirable birds (e.g. to know what a fawn penguin looks like) so that you can cull the bird and trace its pedigree to eliminate the mutation from your lines. Yet, if you have a very nice bird (e.g. fawn zebra) that you want to keep but you now know it is split for something you don’t want (e.g. penguin) you do not need to get rid of the bird if you use an intelligent breeding scheme. If you mate a bird split for an undesirable gene penguin (P= normal, p= penguin) (Pp) to one you know is not carrying the undesirable gene (PP), the offspring on average will be one split (Pp) to one not split (PP) even though they all look normal.

If you do not want to take the time to do many test crosses to penguin you can take another approach. You use these ‘unknown’ offspring and cross them to birds you know are not split. There is a 50% chance you have eliminated that split since half of the ‘unknown’ birds will not be split. The more generation you mate potential splits to known normals the higher the probability that the split is eliminated from your line. In the first cross it is 50%, if you repeat it in the next generation it is 25%, the next 12.5% and so on. This highlights the importance of reducing inbreeding since inbreeding will tend to fix the undesirable mutation in the population by allowing a higher proportion of splits to remain.

While I stated I would not discuss dominant mutations there are a few mutations that fall between recessive and dominant, the so-called semidominant or codominant gene. An example that comes to mind is the Florida Fancy in zebra finch where the splits are actually Florida Silvers (once called Isabel but it is no longer thought to be the same mutation as the European) (so, ss = normal, Ss = Florida Silver, SS =Florida Fancy).

These codominant genes are easier to transfer because the F1 generation shows the split phenotype. You can then cross splits to normal to achieve a 1:1 normal: splits and when you finally believe your splits are of the same quality as the normals you can cross split to split to recover the homozygous mutant bird (e.g. Florida Fancy zebra). Thus, there is no need to do extensive backcrossing and progress can be made in half the number of generations than with recessive mutations.

Unlike with codominant genes, moving desirable recessive genes into a better line will involve some backcrossing since the splits look normal and you risk losing the mutation all together if you do not sib cross or backcross. To avoid inbreeding or going backwards in quality it is not wise to backcross to the homozygous mutant parent since the parent would be the inferior bird. Probably the safest strategy is to make splits with two different normals and then cross the unrelated splits saving only the mutant birds (one of four should show the trait).

Do your best to avoid inbreeding but remember that you always want to cross the mutant or split bird to a superior normal bird. When I discuss quantitative genetics in future articles it will become clear that mating splits to splits may achieve the recovery of recessive mutants more quickly but the progeny will also segregate greatly for all of the polygenic traits of interest. More on that in future articles.

The above discussion gives many predicted ratios and makes the assumption that fitness and viability are equal for different mutations. This is unlikely to be true for all mutations and ratios can be distorted if certain mutations and combinations of mutations are weaker. It has been documented in some animals and in many plants that the fitness of sperm (pollen in the case of plants) is reduced with certain mutations thereby distorting expected ratios. For example, if mutant sperm is weaker swimming or the mutant pollen slower to germinate it is outcompeted by normal sperm or pollen distorting expected ratios. Not much can be done about this but it can explain difficulty in recovering the desired mutation if offspring numbers are low.

As you can see from the previous discussion with a little brushing up on Mendelian inheritance one can develop a plan to acquire the desired color combinations while at the same time improving the overall quality of the bird.

This article highlights the fact that single gene traits are easier to manipulate than polygenic traits. However, there is an issue called “linkage” that makes the development of birds with combined single gene mutations a bit more challenging than one might think, and it is this subject that will be discussed in my next breeding feature.

You can test your knowledge of single gene inheritance on these web pages:

http://curriculum.calstatela.edu/courses/builders/lessons/less/les4/casino/cas1ck.html

http://www.borg.com/~lubehawk/mendel.htm

http://www.exoticflock.com/autosomal_traits.htm

http://www.shadypines.com/punnet.htm

http://anthro.palomar.edu/mendel/mendel_2.htm

http://www.emc.maricopa.edu/faculty/farabee/BIOBK/BioBookgenintro.html

http://www.synapses.co.uk/genetics/wrkshp2q.html

http://www.ksu.edu/parasitology/biology198/mendel.html