Studying Unknown Traits: A Look At How It Is Done

By Gene A. Lucas g.a.lucas@worldnet.att.com

Originally Published:

FAMA Magazine

June 2002

Vol. 25 - No. 6 - Pg. 102

What would you do if you discovered a new, previously unknown trait in one of your Bettas (Guppies, Goldfish, etc.)? Of course you would have to recognize it if it happened but assuming you did would you know what to do? Not to put anyone down but I'm fairly convinced that not many would, so I thought I might go over some procedures that would be most helpful to follow.

First let me provide two scenarios.

1. A light bodied mutant turns up in a spawn in a Malaysian jungle swamp. Along with its spawn mates it survives a few weeks. It is conspicuous in its surroundings, however, and easy to see against the dark background of the debris on the bottom below. After numerous narrow escapes and before it has a chance to mature it is discovered and devoured by a predator bird. It has no chance to leave descendants and the genetic fault responsible is gone.
2. A fish with the same genetic fault appears in the tank of a Betta breeder. It stands out like the proverbial sore thumb. Its owner notices it and excitedly nurtures it to see what it is going to be like. It receives special care and after maturity is specifically selected to parent a brood. It passes the genetic fault to many of its progeny and eventually establishes a type. This mutation survives in the many offspring it is passed to.

These scenarios reflect what happens in "real life." The mutations occur randomly and would be expected with the same frequency in either wild or domesticated stocks. The significant difference is that in the wild most deviant forms are at a naturally selective disadvantage compared to the normal or wild type of each species, one of the reasons wild types are so much alike. Under domestication deviant forms are observed and saved if there is anything interesting or desirable about them, thus the tendency under controlled or artificial selection for steady increase in variation over time.

It is rather obvious that in popular aquarium fishes that have been kept for many years and are easy to breed, there are numerous unnatural types generated either by the preservation of new genetic mutations or by guided selection to enhance minor variations in the natural forms. This has been done throughout the history of human propagation of all domestic plants and animals, and long before any knowledge or application of the principles of genetics. But what is known today about the mechanisms of reproduction and genetics prepares us to note useful variations and systematically develop ways to exploit them. This is what I will try to explain.

I think the best way to do it is to go through a chronological sequence. At each step I will explain why the step is where it is and comment at that point on whatever I think might be worthy of note. Some of it will be obvious and some of it is practiced already, even without formal preparation. Even some of the genetic terminology is widely used, though not always correctly. So let us proceed:

1. First, one needs to observe. I always spend a good deal of time carefully examining my growing spawns. Partly this is because the matings I make are to try to reveal something which I hope to find because of the mating I made. But aside from this, close scrutiny will help in identifying the most (and least) promising individuals. The best of these may turn out to be even better if they are given extra attention as they grow. The poor ones may be discarded assuring more space, resources and attention for the better ones.
   
   
   Close scrutiny will also reveal those individuals which may show something beyond what was expected which to me is the most exciting prospect. The broader tail with more rays, the extra fins, the brighter color, the odd pattern ... the something new and different. These things are the material from which new strains or breeds can be developed. They may require additional work to perfect a type but it would be a shame to have the foundation for something new turn up in a spawn then be overlooked.
2. Preserve any new trait as soon as possible. I would give special attention to the new type from the moment it was discovered. This means isolating it, jarring it, feeding it better, giving it the best water changing routine, watching for parasites or disease and treating promptly for anything needing attention ... full VIP treatment. Remember that anything can happen and if it's bad it probably will. At the first opportunity I would try to get a spawn from it. If there is anything genetic about this new feature, and I would go on the assumption that there is, then by all means we want to try to preserve it before something happens to it. I don't know how many times over the years I have been frustrated by not being able to get a spawn I really needed from some fish and the cause was that I put it off until the fish got too old, or got sick, or jumped out, or the cat ate it, or I may have inadvertently given it away.
3. Follow the pattern for studying a genetic trait. Be aware of why mating sequences are made the way they are. Analysis of a genetic trait occurs in a systematic way. The first mating involves the individual which showed the new trait and, if possible, another which has a high probability of also carrying it. This would most likely be a sibling or to the parent of the opposite sex. The latter possibility may be eliminated in species with short life cycles, like Bettas. The parent may be too old to breed before the offspring reaches sexual maturity. In horses, for example, which can live and reproduce until they are at least twenty, the young are sexually capable at three or even two so there are long spans of time in which offspring to parent matings are possible.

In genetic parlance the initial mating in a sequence is called the parental (P 1 for short) generation. Subsequent generations are called filial generations, the first filial is called the F 1, the next the F 2 and so on. In terms of the analysis of a new trait the P 1, F 1, and F 2 are of greatest importance since additional generations may only be of value in maintaining a stock bearing the trait or for use in outcrosses for other purposes. The F 1 is also the P 2 in this sequence or P 1 if starting some other sequence. Obviously the correct term for use is based on the individual that is the focal point of the analysis ... known as the propositus. I hear people talk about F 4 or other later generations as though they had significance but it is the first two that provide the most and are most important.

What we want to find out is if the trait will reappear in later generations which would prove that it is genetic, rather than an accidental occurrence generated by some factor other than a genetic mutation. If we see F 1 progeny developing the new trait we assume that it is genetic and was passed from the affected individual to its offspring. If that happens we also know that the trait is probably dominant although there is the possibility that it could be recessive and the other parent carried only one allele for it (an allele is one member of a pair of genes). In that case the other parent "carried" the trait but did not show it because it had a normal dominant allele which prevented it from developing. If two individuals came together, both possessing one or more recessive mutant genes, then the mutation would have to have occurred at some previous time.

It is quite likely that many genes that turn up in domestic organisms are recessive and invisibly being spread through a population until at some time two individuals that carry it mate and produce some progeny which have only the mutant recessive genes. If our new trait is recessive and the selected mate does not carry it, it will not appear in progeny of the F 1 generation. This is why it is imperative that an F 2 be secured. We know that if the trait is recessive, the original parent who showed it had to have both genes for the new trait and none for its normal alternative. Its mate must then have had only normal genes and all the F 1 progeny had one of each.

Mating two of the F 1 progeny together will ensure that approximately one fourth of their F 2 progeny will have two of the recessive genes and no dominant normal ones, thus the trait would re-emerge. Not only that but two-thirds of their normal siblings would carry the trait, although their mutant gene would be suppressed by their normal counterparts. Finally, one-third of those normal siblings would not carry the mutant gene and could not pass it on. This is a bit frustrating because it isn't possible to distinguish between the normal siblings that carry the mutant gene and those that do not. This makes it more difficult to use any of them to try to perpetuate the new trait.

One solution is to test cross some of them. A test cross involves mating one of the normal appearing fish with one that shows the trait. If the tested individual produces affected offspring then it must have carried the mutant gene. If it doesn't then it must not have. The first type could be used in matings to carry on the trait while the second would need to be excluded. There are some drawbacks. One is that test crossing requires an additional generation and because of the short life span of Bettas it may not always be practical. Another is that the test individuals are randomly selected and there is a one in three chance that the one chosen will not carry the desired gene. This makes it necessary to test several at a time to increase the probability of finding one that does.

An example of a recessive gene in Bettas is Doubletail. In order to have the doubletail trait the Betta must have two mutant genes (and no normal ones). If the fish is a Singletail it has either one or two normal versions of the gene. These can be symbolized as follows: (The + symbol indicates the normal version of the gene and the letters indicate the abnormal or mutant version. Upper case letters indicate alleles that are dominant or partially dominant to the normal, and lower case letters indicate alleles that are recessive to the normal.):

+//+=Singletail, +//dt=Singletail, dt//dt=Doubletail

Another possibility is that the mutation is first appearing in the affected individual. If it shows up when present with a normal allele it is said to be dominant. Some dominant genes are not completely dominant and an intermediate type can be produced. In Bettas, dark body color is completely dominant over the albinistic cambodian which has a light body, thus dark is completely dominant over light, but the iridocyte (metallic) colors come in three shades. Green is the wild type, blue is produced with one mutant allele and Steel Blue when both genes are abnormal. These are properly symbolized as follows:

+//+=Dark, +//c=Dark, c//c=Light (Cambodian).

+//+=Green, +//B1=Blue, B1//B1=Steel Blue

A helpful feature of these kinds of possibilities is that each type is readily identifiable. The numbers and types can easily be tabulated for analysis. These are examples of "Mendelian" (term derived from the name of Gregor Mendel, the nineteenth century scientist who recognized and publicized the mechanics and mathematical relationships associated with this "discontinuous" variation).

Still another possibility is that the trait represents the extreme of a continuum which in its other forms was not obvious. If this is true, careful examinations of any available original sibling should show others grading into it. When a trait varies "continuously" through a population it is referred to as "non-Mendelian." Two examples are commonly seen in Bettas. One I call "Spread iridocytes" and the other "Extended red." The terms are derived from the density and distribution gradients when compared to the normal or wild type distribution of these colors. They could very simply be described as follows:

In the case of Green or Blue or Steel Blue,

+//+=limited distribution, +//Si=intermediate spread, Si//Si=extensive spread

In the case of Red,

+//+=limited distribution, +//Er=moderately extended red, Er//Er=very red

Unfortunately, these two traits do not sort neatly into three categories but actually form broader ranges of less distinct grouping. They fit better into classifications thought to be generated by multiple genes or the types are just so variable within the classes that they overlap and blur the distinctions. Were we to attempt to classify them as we did the others, they would both need to be considered as incompletely dominant.

We can conclude, then, that there are limited possibilities about the genetics of a new trait that suddenly appears and that we can make certain assumptions about them. To summarize:

1. The trait is a non genetic "accident" (like a birth defect) and does not reappear in progeny in further generations.
2. The trait is the first individual with a dominant mutation and more like it will appear in its first generation progeny. It may be only partially dominant but it will still be identifiable. 
3. The trait is recessive and the affected individual can pass on its mutant genes but they will only appear in an F 1 generation if the other parent also carries the mutant gene. If it is recessive, all F 1 progeny will carry it from the affected parent, no matter what the other parent is like. If this is the case, sibling matings will generate mutant types in the F 2 generation. 
4. The new type is not the result of a single gene mutation but something more complex. It may be the extreme of some continuously variable condition and may require more careful study and additional types of matings to analyze.

Discussion

I suspect most people who find a new trait in one of their fish would want to see if they could produce more like it so most of the time such fish are probably set up to spawn. If they aren't I think they should be. A major reason is that new traits may not be what I would call finished products. They could provide the genetic basis for selection into some new desirable type even if they didn't look like much themselves. Also, every new gene adds to the accumulation needed to determine possible genetic linkages or have some other genetic or biological value. They may be combined with existing traits to add to the many varieties we already have. For example, a new ruffled fin edge might be added to Doubletails of Half-moons or a different color or pattern might be established which could then be supplemented to the prevailing types.

I find that many breeders will make the first mating in an exploratory series and if they see no results in the first generation progeny, abandon the effort. Most of our best known genetic types in Bettas, and the easiest to work with, are the result of recessive mutations (Cambodian, Melano, Doubletail, Non-red, etc.). There is a greater possibility of finding recessive traits distributed throughout the Betta gene pool which means there is a greater chance that one of these might turn up than a dominant one. If the F 2 generation is not obtained, the trait may disappear again and not emerge in a place where it could be exploited for a long time, if ever. There is always the possibility that it could be lost forever ... before it is ever "found."

Breeders should make an effort to understand the meaning of terms like P 1, F 1, F 2, and so on and use them properly both as they work with their fish and as they talk or write about them. I assume the relationship of terms like progeny and offspring, genes and alleles, mutations (changes to a gene) and mutants (individuals exhibiting the effects of a mutation) are clear but if not they should be understood. It would also be extremely helpful if people would learn and use a common symbolic system (which I have been "preaching" for many years!) which in turn would be a big help in communication.

I hope this has made a case for following through on the two generation sequence when exploring something new and possibly speeds up the process of building more variety into popular domestic species. I also hope it might lead to better understanding of some routine genetic vocabulary and processes and improved communication between breeders of domestic fishes, especially Bettas (Siamese Fighting Fish).