BREEDING STRATEGIES & GENETIC CONSEQUENCES IN GUPPIES
© Alan S. Bias
Permission granted for non profit reproduction or duplication of photos and text with proper credit for learning purposes only.
January 6, 2012
 |
| Grey Asian Blau Vienna Emerald Lower with Z-Bar and Blue Peduncle |
PREFACE
The intent of this article is to point out how considerations beyond color, pattern and caudal shape play an important role in creating healthy & viable strains. As such specific references to color study are not made. They are better covered in publication of study geared to these criteria.
In the following article references to population(s) and strain(s) are interchangeable. Though population(s) will normally refer to individuals in a wild setting and strain(s) refer to individuals in a domestic population(s). Therefore, each is more or less one in the same. Island populations are defined as those in geographic isolation from another. Such conditions may results from water falls, mountainous separation or on an island.
Non random mating of genetic relatives will increase the frequency of homozygotes, and decrease the frequency of heterozygotes. Breeders utilizing linebreeding cause genotype frequencies to change thus are altering the proportions of phenotypes. We therefore increase the speed of natural selection on a captive population. What may take Mother Nature hundreds of years can be accomplished in several. By increasing the occurrence of homozygotes, linebreeding can expose any deleterious recessive alleles for culling.
In both wild and domestic populations, any decline in population size can result in genetic drift, which can lead to inbreeding depression. Inbred offspring can be substandard to outbred offspring in survival & reproduction. As a result, individuals homozygous for deleterious recessive alleles increase, and deleterious mutations become fixed due to such drift. In the wild such individuals face extinction. In domestic populations they should face ruthless culling by the breeder.
Genotype is usually inferred from physical appearance referred to as phenotype. In simple terms phenotype in domestic strains is no more than the product of coded genes resulting from a combination of genotype. It is influenced over time by environment and breeder selection. Individuals carry two alleles, which may be identical (homozygous) or different (heterozygous) for each gene. Strains with multiple allele types at a gene are polymorphic; those with only one type of allele are monomorphic.
 |
| Sibling Asian Blau |
DEFINITIONS
Dominance - one gene contributes more, or less, to the final phenotype.
Alleles - copies of genes inherited from both parents.
Effective Population Size - number of individuals required to have the same amount of random genetic drift as the actual population; this is often much smaller than the true population.
Founder Effect – founded by a small core group of individuals representing a smaller proportion of the source genetics.
Genetic Bottleneck – the reduction of a strain to very low numbers and subsequent loss of variation from rarer alleles.
Genetic Drift - gene frequencies within strains change over time due to random events by chance alone; resulting in a possible loss of certain traits. In small populations selection is often skewed and beneficial alleles may be lost.
Gene Flow – the movement of genes into or out of a population, or between populations. This may result in a change in the gene frequencies within a strain or population over time.
Genetic Potential - the greater the genetic diversity, the greater potential to respond to changing environmental conditions.
Gene Pool – a strains genetic variation consisting of alleles present and their frequencies across members.
Heterozygosity - percentage of gene loci at which the average individual contains alternative forms of alleles.
Homozygosity -percentage of gene loci at which the average individual contains the same forms of alleles.
Inbreeding Depression – a reduction in fitness due to the mating of closely related and often genetically similar individuals; creating a loss of robustness due to expression of deleterious recessive alleles. This may result in an inability to respond to selection from lack of residual heterozygosity.
Locus - position occupied by a particular gene or one of its alleles.
Monomorphic - both copies of the allele are always the same, indicating no variation at a particular locus.
Outbreeding Depression - a loss of robustness due to reproduction among very dissimilar individuals.
Polymorphism - the percentage of the genes with multiple types of alleles at a given locus; characterized as a measurement of the genetic capacity of a strain to adapt to change.
Additive Gene Interaction – All genes contribute equally to the final phenotype. In the absence of dominance and epistasis, the final phenotype is estimated as the sum of the individual effects of the two alleles present at each locus.
Nonadditive Gene Interaction - Nonadditive gene interactions include the effects of dominance of one allele over another and epistatic effects of one gene (or gene pair) on another, non-allelic gene (or gene pair).
POPULATION SIZE
A strain’s genetic diversity is not determined by Overall Population Size (OPS), but by the size of the genetically effective population. The latter is referred to as Effective Population Size (EPS). Genetically EPS is much smaller than OPS because of multiple factors which include: varied reproductive success of females, unbalanced sex ratios, or severe population crashes.
It should be noted that in domestic stock breeding often robustness = fitness; or is in direct response to it. However, in the world of Mother Nature fitness = viability; an overall ability to survive and reproduce. Breeders often forego this in an attempt to impose phenotypic traits on a strain.
An accepted rule of thumb states: deleterious mutations will lead to a reduction in robustness of about 1% each generation without additional influence either positive or negative. When trying to counteract the effects of such mutations, the first concern at a strain level is determining EPS. That is, at what number would a group of individuals contribute equal genetics vs. a theoretical infinite population, with further numbers failing to increase genetic contributions?
In the Guppy world this number is rather hard to define for numerous reasons encompassing multiple traits, colors and patterns. In a typical homozygous livestock breed of limited color and variation such as Barbados Blackbelly Sheep, EPS is in the neighborhood of several hundred individuals in OPS of 1000 potential breeders.
Population diversity can be defined by 3 major components:
• Intra-individual (heterozygosity);
• Among (= between) individuals within a population;
• Among (= between) populations.
A strain can lose much genetic diversity during population crashes resulting from many factors not limited to disease or downtrends from loss of breeder interest. If numbers are recovered quickly, so will EPS. In a small population inbreeding depression and genetic drift are potential problems resulting in the loss of often important genetic diversity contained by rare alleles. Sustained low EPS can drastically reduce genetic diversity with each passing generation. Through genetic drift it is thought such lows can expose in homozygous form the already existing recessive mutations needed by isolated populations for survival without genetic migration from others. Otherwise, extinction is a possible result of extreme reduction in EPS. In a small population, the effect of genetic drift is stronger than almost all levels of selection.
Ironically, successful breeding programs can actually reduce EPS under certain circumstances. For example: If OPS in a strain is small and a high percentage of young from these strains are retained, they can contribute a large percentage of replacements to the total breeding population. In such a case EPS may decrease.
A Minimum Viable Population (MVP) is one that permits long term survival of a population or strain in current environmental conditions. MVP must be sufficient to protect against extinction resulting from genetic, demographic or environmental factors. Studies indicate most mammal (breeds or species) have little genetic variation vs. plants resulting from small or static inbred populations. Only infusions of genes from migration or exchange in combination with mutation avoid serious inbreeding. For example: The entire population of African Cheetah is virtually identical in nuclear DNA, indicating the former population crashed severely. So identical in fact that skin grafts from one individual to another are not rejected.
Inbreeding depression is normally characterized by lowered robustness or vigor. Heterosis or hybrid vigor tends to manifest by increases in size, growth rate, and such in offspring of two inbred lines. Yet, the Cheetah passed through a serious bottleneck without extinction. Since that time many generations ago natural selection has eliminated the most deleterious genes arising from inbreeding depression by infant &/or adult mortality. Otherwise, they simply would not have survived into modern times. They are still a highly endangered species, in part due to their high degree of homozygosity. Remember that evolution through natural selection only increases frequency of an allele in proportion to its benefit on survival. Breeders should select and cull on the same principles for strain fitness.
GENETIC ISOLATION
A domestic population usually exists as a group of strains that may or may not experience gene flow in the form of breeder exchange. In established strains exchange on a regional level comes from both male and female individuals through limited numbers of individuals. The genetic variation or heterozygosity created in such cases can be grouped by local diversity within regional strains. Breeders may create variability in the form of local genetic adaptation to conditions, which result from co-adapted gene complexes.
These combinations of alleles at different loci after several generations can confer optimal levels of robustness on individuals in static strains. Introduction of outside genotypes to any strain, while beneficial on many levels can cause a loss of local adaptation and should be a concern to breeders. Such loss of vigor has come to be known as outbreeding depression. While its effects on domestic strains are unknown to many breeders, biologists have long understood the process in rare & endangered wild species.
Reduced genetic diversity within small populations may result from:
1. Founder Effect;
2. Genetic Drift;
3. Demographic & Genetic Bottlenecks;
4. Inbreeding Depression.
1. Founder Effect
Founder Effect is a term used to describe negative impacts within a strain descended from a single or a several common ancestors. They appear many times in the background of the strain, typically possessing many traits breeders wish passed to offspring. Along with good genes, negative recessives associated with the desired genetics my manifest themselves in later generations in a homozygous state. In the wild Mother Nature will quickly remove such flaws. As breeders we must do the same with little emotion. Founder Effect can be easily studied in populations that occur on
islands or in geographic isolation, such as breeder fishrooms.
A founder effect can arise when a strain is perpetuated by only a few foundation individuals. In the extreme case, a single viable female might be utilized to resurrect or establish a strain in a new location, and may result in a type of genetic bottleneck. For example: After the foundation strain of Castle Milk Moorit Sheep was dispersed, they were nearly lost and revived from a single sire and six females. Today the population continues to increase with additional interest by breeders in the United Kingdom
2. Genetic Drift
Isolated strains and island populations are also affected by genetic drift. This random change in genes can affect frequency of type. Example: A strain utilizing only blond (gold – US) sires bb in genotype and five grey females BB in genotype. Too keep things simple we assume each female is replaced by her daughter in any given year. 1st year offspring will be grey and Bb in genotype. From 2nd year on offspring will be either Bb or bb. Each year the number of Bb will decrease while bb will increase. At some point in the near future all individuals will be bb and the gene for grey will have been lost to the population via genetic drift. Extreme loss in genetic variability often found in island populations or isolated strains can create adaptation to a specific environment. Yet it also can reduce the ability to adapt to changes in either environment or preference. This is commonly seen with strains geared towards a limited phenotype.
Genetic variation in guppies is jointly determined by breeder selection, initial genetic variability, naturally occurring genetic drift, and mutation. Genetic drift has a high relative importance in small populations. The negative effects of recessive deleterious alleles may become much more prevalent, as the frequency of the recessive allele increases. When the recessive allele replaces the dominant allele (or visa versa) a fixed trait results. The loss of any positive or negative allele is most likely to occur in a small gene pool. The reduction in the number of forms of an allele in the extreme case leads to a monomorphic state where only one form exists.
Consequences resulting from a loss of genetic variation include inbreeding depression &/or the inability of a strain to adapt and evolve to changing conditions of its environment. While inbreeding depression has been accepted within domestic strains, scientists are just beginning to substantiate it in selected wild or feral populations. The occurrence of inbreeding depression indicates that the underlying causes are yet to be fully understood by the breeder or are breeder enhanced by poor mating selection and retention.
Phenotypic characteristics (appearance) are often used to divide individuals into populations & strains. While there is great diversity across strains, there may be limited genetic variation within any given strain. Domestication of wild guppies has led to many strains and increased “between strain variations.” Guppy strains are bred to many standards, resulting in selection for different characteristics in different strains. We have selected individuals for specific appearance. These selections over time can lead to indirect selection for behavioral traits and disease resistance.
3. Demographic & Genetic Bottlenecks
Small island populations of wild guppies in native habits, and feral introductions around the world present opportunities for examining the bottleneck effects of both recent and prolonged inbreeding. From these populations isolation has created an impressive diversification of type with the aid of genetic drift. On a smaller scale isolation by distance has had a similar effect on domestic strains over the last hundred years in breeder tanks. As a rule small populations will not persist without added genotypes via migration (breeder exchange) or homozygous expression of new combinations of rare mutations that are exposed by genetic drift. Breeder exchange allows strains to function as a collective group.
Many guppy phenotypes are unique in the animal world as they are sex-linked &/or autosomally enhanced. Males that consistently pass homozygous or fixed traits to offspring are prepotent. In essence, he doesn’t have two different copies of desired alleles to pass on. Breeding "like to like" is a time-honored system in the pedigree world of livestock. The number of different alleles available in the gene pool decreases creating consistency. Genetic bottlenecks can occur for a number of reasons. The repeated use of a sire in backcrossing or single pair matings are only two factors reducing a gene pool. Another cause of bottlenecking is a reduction in numbers via disease or loss of breeder interest. Remaining members are bred together in an effort to continue the strain.
4. Inbreeding Depression
Inbreeding depression occurs when deleterious recessive alleles become homozygous. The condition can manifest as reduced fertility, growth, age of maturity, loss of maternal traits such as birth or litter size, or structural deformities to name but a few. In a wild setting inbreeding reduces recessive heterozygous lethal alleles by elimination. As they are expressed in a homozygous state mortality increases. In a domestic setting it is up to the breeder to perform the task by culling of undesired genetics &/or breeding around it with the aid of genetic test breeding.
Problems associated with inbreeding depression are found in many domestic strains and captive wild populations. Yet, the outstanding reproductive successes of many breeding programs indicate poor choices are potentially responsible. In Example: The Austrian Kärntner Brillenschaf is an endangered domestic sheep breed. They descended from local sheep crossbred with two Italian native breeds. As common with many native breeds it was replaced by more market driven varieties during the 20th century. From a low of 17 females and six sires the total pedigree population has grown to over 1500. Current genetic diversity within the breed determined by DNA micro satellite markers compares favorably to diversity with other strains showing a high number of alleles per
locus.
Some recessive genes are lethal in homozygous form. However, sometimes the heterozygote may be selectively superior to even the homozygous normal genotype. In example let us consider the partially dominant autosomal Asian Blau (Ab) gene which removes red from the body. Non Asian Blau abab males are normal with red spots which confer an advantage in breeding as based on female preference. However, it also makes them conspicuous to predation in certain lighting. Heterozygous Abab, while suffering reduced fecundity in breeding from a lack of red spots are less visible in low light, so are selectively superior. On the other hand homozygous AbAb not only lack fecundity conferred by red spots, they are highly reflective allowing for increased predation in conjunction with reduced vitality. Many breeders cull them because of their reduced colorfulness.
STRAIN MAINTENANCE
In a wild setting inbreeding and genetic drift are countered by natural movement (migration) often inhibited by island demographics. In domestic populations this is accomplished by breeder exchange. In small populations it takes as little a one individual per generation to maintain genetic diversity. Natural selection (evolution) causes changes in wild populations over many generations. This often results in many sub species originating from the founder population. Domestication enhances and speeds up the process via artificial selection dictated by rapid change. The primary limiting factors are numbers of interested breeders and fishroom sizes.
Guppies have been sub-divided into many different phenotypes when breeders select against certain traits (alleles), while retaining other desirable alleles. Example: In both my Line 1 and Line 2 Vienna Lower Sword strains the allele for red fins is selected against, in favor of X-linked yellow, by phenotype. In Line 1, a basic Vienna Emerald Green (VEG) pattern, the red genes (alleles) have been selectively eliminated and yellow enhanced with additional Y-linked gene(s) for yellow fins. In Line 2, also VEG with additional autosomal Z-bar and blue peduncle, red fins have only been suppressed (epistasis) as an outcross will allow for their reappearance. A conscious effort has been made to suppress this particular allele in both gene pools.
Can breeder selection accumulate bad genes with long term repercussions for a strain? Yes, in the fishroom we can use the following example: strain longevity can be quickly influenced by the killing off of breeders at an early age over successive generations. Ever wonder why most show strains lack the longevity found in wild-type or feral populations? Breeders rarely keep breeding age females around extended periods of time vs. non showing breeders, resulting in a loss of alleles for longevity. This would seem to be a relevant concern to show breeders since many males outside of swordtails cannot be exhibited in more than several shows in a season. Yet they continue to select for growth rate & size over slower maturity & longevity, and associated hardiness, resulting in a shorter life. If this one trait is so easily manipulated, what others are we unknowingly influencing to hinder a strain’s ability to thrive under stress-free conditions?
Can good environments accumulate bad genes with long term beneficial results to a breed or species? Again yes, recent studies indicate inbred Soay Sheep are more susceptible to internal parasitism during peaks in population cycles. Naturally occurring periodic winter die offs of individuals highly parasitized by gastrointestinal nematodes guaranties survival of those with genetic diversity. Thus those with the least resistance to parasites are sacrificed to keep genetic variation intact by Mother Nature. In the wild inbreeding is not all bad if it occurs over multiple generations permitting selection to purge any recessive deleterious alleles before they become fixed.
Breeders of guppies must continue the role of Mother Nature in the fishroom. Combined genetic diversity results from both local (intra) and regional (inter) population differences. Gene flow can increase intrapopulation variability, thus reduce potential inbreeding in a strain. Yet it has the potential to also reduce interpopulation differences. Additionally, any short-term benefit derived from outside infusion into a strain could be negated in future generations if it destroys co-adapted gene complexes geared towards survival.
Breeders interested in conservation of genetic diversity tend to focus on OPS. A problem arises if we choose OPS as the chief unit of conservation, since wild and domestic populations rarely exist as one interbreeding group. Social hierarchy in the wild and breeder exchange in domestics should create loosely-connected genetic “unrelatedness” with gene flow to other groups.
Recessive and mildly deleterious mutations are common. When located close to a desired allele they may inadvertently be selected. Crossing over may separate the selected gene from the harmful one and it can then be eliminated through selection. Selection cannot remove the effects of inbreeding if it is due to over dominance.
OUTBREEDING ENHANCEMENT
Heterosis or outbreeding enhancement is the opposite of inbreeding depression and is referred to as hybrid vigor. Hybrid vigor is the opposite of inbreeding depression. It has been described as “the masking of recessives through crossing of unrelated genotypes.” In domestics regional populations have dissimilar recessive deleterious alleles in their genetic composition. Matings from two populations may create offspring that are heterozygous for them. The hybrids benefit as any deleterious alleles are masked by heterosis in a recessive state. In successive generations a higher measure of robustness is evident in the nucleus population, but Mendelian recombination will again turn out homozygous deleterious alleles in certain individuals. This principle will hold true in most situations when limited to alleles on a single locus.
OUTBREEDING DEPRESSION
Many breeders claim genetic diversity is a good thing, touting how the benefits of hybrid
vigor will negate inbreeding depression. However, co-adaption in the form of gene complexes of multiple loci evolves to create environmental strengths in isolated strains. When such sub-populations are crossed the overall robustness of offspring can be greatly reduced. Such a situation is referred to as outbreeding depression. It has been proven outbreeding depression could have a negative impact if individuals from two different populations that are each well adapted to their local environments are bred to each other.
Outbreeding depression can result in two ways. In the first and most complex form locally co-adapted gene complexes can be disrupted by breedings to distant or unrelated strains. This is due to non-additive gene interaction of the same genes from different genetic backgrounds caused by a breakdown of biochemical or physiological compatibilities. The details of the epistatic interactions are often different in these cases. Co-adapted gene complexes are the least understood in most cases.
In the second form offspring between parents from isolated or unrelated strains may have phenotypes that are not suited for either location. This is common when a breeder brings in individuals from a long time breeder located in a dissimilar environment or adhering to a drastically different routine. Individuals bred multi generation in an automated system with aged alkaline water and high protein feed are tailored to this setting. On the other hand individuals bred with limited water exchange acidic in nature, with sparse feedings of minimal value are adapted to their own location. Crossing of the two types may create individuals with transitional characteristics not suited to either place.
As if things were not intricate enough both outbreeding depression and outbreeding enhancement can occur concurrently in a strain (or individuals) when utilizing an outsourced genotype. If strains or populations have not split long enough to acquire divergent gene complexes it is unlikely any outbreeding depression will occur. If European Schimmelpennig Platinum Double Swords are bred to Asian stock differences may be visible, to North or South American stocks even more probable. The degree of potential outbreeding enhancement cannot be predicted and you must test strain individuals to evaluate benefits.
When I bring in a new male he is rarely mated to the majority of females. Only after constructive test breedings show positive result will his genetics be utilized on a larger level. The same goes for homebred males utilized in linebreeding. They must first pass on beneficial results via limited test breeding. Tank space is too valuable to waste on unproven results on a large scale.
As a breeder you may be well served by showing restraint in obtaining unrelated stocks from distant sources in a quest to diversify a strain. The effects of outbreeding depression may not manifest for several generations. Initial offspring may indicate a beneficial increase in overall strengths. After several generations when the genetics have had a chance to “recombine” negative effects may become perceptible. In these cases only time and breeder choice can eradicate problems by means of concentrated culling.
TESTING
DNA analysis has quickly become one of the techniques most often used by large animal stock breeders to scrutinize strain genetics, and by associations to substantiate purity of strain via fingerprinting of sires. Two available options are: nuclear DNA and mitochondrial DNA (mtDNA). As expected current testing of many breeds indicates a greater diversity between the breeds than within the breeds.
All species obtain nuclear DNA from both parents. A highly useful nuclear DNA marker can assess genetic variability through micro satellite repeats. These repeats are currently used for interpopulation comparisons to detect variation in genetically distant populations, and assess relatedness between individuals.
Extracted mtDNA allows us to map the genes on the mtDNA molecule. Individuals having the same map were related in the past. Mitochondrial DNA does not recombine with other DNA as does nuclear DNA, thus allowing for a static means of identification. mtDNA only comes from your mother and passes via a maternal line. While male offspring possess mtDNA it is not passed to offspring. Even though passed mtDNA is indistinguishable from generation to generation it is modified by mutation over extreme periods of time. Therefore, it is possible for guppies deriving from the same source to have different mitochondrial alleles when separated for any length of time.
While such testing is not readily available to the average guppy breeder, this too may change in the future. Mapping of the guppy genome is no longer a fantasy. Genome studies across breeds are showing that the amount of similarity across not only breeds, but species, brings validity to using each as models for others.
 |
| 2 month old Grey Vienna Emerald Lower * Bronze Delta |
SUMMARY
Guppies are unique in the vertebrate world. They have been bred primarily for color and to a lesser degree shape over the last one hundred years. In many instances with limited consideration for other valuable attributes commonly considered in the livestock world. For this reason many strains lack survivability.
Contributions for many selected phenotypes (color & pattern) are predominantly sex-linked on the Y chromosome and not limited to autosomal genes. Or so it was thought. Many recombine between X & Y chromosomes. Current scientific and breeder theory indicates this may not be totally accurate. We are finding that certain traits thought to be Y-linked are often enhanced or expressed in combination with autosomal components determining expression in independent zones of the body.
As a breeder we can manipulate gene frequencies by non random mating selection in conjunction with: genetic drift, mutation, or the domestic version of population migration in the form of individual acquisitions. It must be remembered that genetic variation can only come about via mutation in existing genes.
Through mutation deleterious alleles constantly arise, and are always present at low frequencies. At a given locus, some alleles may confer higher levels of survivability on an individual than other alleles. These other alleles may be rare deleterious recessives, which in homozygous form reduce robustness.
In the future with the aid of modern technology it may be possible for guppy breeders to characterize strains for phenotype and genotype. In large animal breeds genetic distance is tracked by gene mapping to determine the degree of relatedness between established strains. While phenotypic performance of various traits could be evaluated for adaptation to individual breeder environment. It is done regularly by non-guppy breeders in other venues.
Hopefully this article illustrates the importance of understanding guppy population genetics in planning effective management of strains consisting of limited numbers. Color selection in guppies, albeit primary, is but a single consideration in breeding a strain if one seeks viability. Inbreeding, loss of variation through drift or overuse of sires, genetic differentiation among regional populations, etc. are worthy of contemplation in the overall management scheme of guppies at the strain level.
After all, how well your strain succeeds will determine how well it meshes into the overall fabric of guppies as a whole.
REFERENCES
Cooke, D. A. (????). “Heterosis and Inbreeding Depression”. http://members.ozemail.com.au/~davcooke/heter.htm (Last viewed: 1.5.12) davcooke@ozemail.com.au
Cooke, D. A. (????). “Hybrid Breakdown”. http://members.ozemail.com.au/~davcooke/break.htm (Last viewed: 1.5.12) davcooke@ozemail.com.au
Michael Lynch, Dept. of Biology, University of Oregon U.S.
Wolosz, Tom,
Schwend, Katharina (Oct. 2001). UNTERSUCHUNGEN ZUR GENETISCHEN VARIABILITAT DER KARNTNER BRILLENSCHAFE IN OSTERREICH. Doctoral Thesis. University of Vienna
C. Dalvit, E. Saccà, M. Cassandro, M. Gervaso, E. Pastore, and E. Piasentier (Nov 2008). “Genetic diversity and variability in Alpine sheep breeds.” Small Ruminant Research - November 2008 (Vol. 80, Issue 1, Pages 45-51, DOI: 10.1016/j.smallrumres.2008.09.005)
O’Brien, Stephen J., Chief, Laboratory of Genomic Diversity, National Cancer Institute (Apr 2001). “Species Conservation and the New Genomic Technologies”.
http://symposia.cbc.amnh.org/conservation-genetics/transcripts/o_brien.pdf (Last viewed: 1.5.12)
Menotti-Raymond, M., O’Brien, S. J. (1993). “Dating the genetic bottleneck of the African cheetah.” PNAS 1993 90 (8) 3172-3176. http://www.pnas.org/content/90/8/3172.full.pdf (Last viewed: 1.5.12)
Kennedy, Delma, Sheep Specialist, Genetics, Reproduction and Performance Program, Ontario Ministry of Agriculture, Food and Rural Affairs (OMAFRA) (Oct. 2000). “Glossary Of Sheep Breeding Terminology”. http://www.omafra.gov.on.ca/english/livestock/sheep/facts/00-079.htm (Last viewed: 1.5.12)
Petrescu-Mag I. V., 2009, “Winge’s sex-linked color patterns and SDL in the guppy: genes or gene complexes?”, AACL Bioflux 2(1):71-80., Printed version: ISSN 1844-8143. Online version: ISSN 1844-9166 available at: http://www.bioflux.com.ro/docs/2009.2.71-80.pdf © 2009 Bioflux. (Last viewed: 1.5.12)
