Central New York Betta Club

Do you find it confusing to read or hear talk about this gene or that in the fish you try to breed? Are you ever frustrated when you pay good money (maybe even ridiculous amounts) for a pair of fish having some outstanding trait and then cannot rear anything like them from your spawnings? If this sounds familiar, you are, as the saying goes, not alone. It happens to a lot of us. Probably the main reason is that we expect something to be true that isn't ... we thought we bought something ... a gene for a trait ... that perhaps is a bit more complicated. I haven't discussed the topic for a while but because of the recent expansion of interest in certain Betta types I thought it might be time to do so again.

The explanation is basically pretty simple. We, too, frequently tend to look upon an overall phenotype as representative of a single functional genetic factor when most often it is the result of some combination of factors. This is an extremely important distinction and I want to explain why, along with presenting some relevant examples.

1. A phenotype that does not represent a single gene. Yellow Bettas do not result from the action of a single gene. There is no such thing as a yellow gene that produces a yellow phenotype in Bettas. There are, however, yellow Betta phenotypes. The yellow color itself (in the earliest studied case) was one I labeled "non-red" back in the 1960's after our research determined that this recessive gene caused Bettas that possessed it to be yellow where they would normally be red. If you looked at a wild type Betta and noted where red showed on it, you would recognize that red is not all over and there are other colors present as well. If the others are unaltered, then the Betta with the non-red trait would still have dark and green metallic (iridocyte) color and all would be visible simultaneously in their respective areas on the fish. There were two reasons why I did not call the gene yellow (which I could have). One was that the term non-red had been used previously to describe similar abnormalities in other organisms. The other was that I wanted to try to avoid having people make the assumption that there was a single gene that would generate the yellow phenotype.

The IBC standards categorize yellow as a light-bodied, solid color type. To obtain a solid yellow, several changes must exist. The black and iridocyte colors must be minimized (eliminated if possible), the yellow must replace the red and the red/yellow must be extended to cover the entire fish. This requires four different alterations of three different pigment components! Two of these seem regulated by single locus recessive genes, the non-red already mentioned and the recessive cambodian gene which nearly eliminates dark pigment. The other two (the extension of red to cover the entire fish) and the reduction of iridocyte color to minimize green (or blue, if the green has been altered ... by still another gene) do not operate as though controlled by single genes.

2. Genetic variation that may be generated by polygenic action. The distribution of red or the iridocyte color in Bettas' (and possibly other species as well) colors seem to be of a type called continuous variation. This kind of variation is generally dealt with as though controlled by several genes working on the same trait or by a gene with "modifiers" regulating the distribution and/or density which is essentially the same thing. Again, when I studied these traits back in the 1960s I gave them names which I thought best described what they were doing. "Extended red" referred to whatever was happening to cause the fish to have red in places it was not present in the wild type. "Spread iridocytes" served the same purpose for the distribution of iridocyte color. I used symbols (Er and Si) to refer to them in a genetic sense although when I use them I always try to distinguish between these and proven single gene traits. The symbols are headed by lower or upper case letters to show that when crosses involving them occur they seem to act in a recessive or dominant (or at least partially dominant) way.

Those who have followed my writings about Betta genetics are aware that I adopted and recommend the "Drosophila (Fruit fly) method" of use of symbols where a + is used to designate a wild type allele. Symbol names are selected to try to reveal some genetic information and lower or upper case letters to indicate dominance or recessiveness to wild type or normal. For example, the cambodian trait (the prevention of most dark pigment formation) by this method is symbolized with a lower case 'c'. A Betta with normal dark color would be symbolized '+//+' at the locus representing cambodian. The cambodian would be symbolized 'c//c'. Crosses would generate '+//c' genotypes. Those fish would be dark since the cambodian gene is recessive.

The traditional way to look at multigenic traits is to assume that each gene of several contributes some increment of effect to the phenotype. If we called that 'A' and there were four sets of alleles we could array them from AAAAAAAA through AAAAAAA+, AAAAAA++, AAAAA+++ and so on down to ++++++++. This is one way to try to account for variation that seems continuous rather than either-or. This has been a useful and generally satisfactory way to try to explain such variation although I always thought it looked best on paper. It doesn't logically seem easy to explain how such a system could have developed through random mutation. There are, of course, ways to try to account for it but that's another topic. This interpretation is more the way variation of red and the iridocyte colors operates. For general purposes the Er and Si designations can be used as though they were single gene designations but one needs to remember that the results are not going to work in strictly that way.

This leads us to the second major part of this discussion which is also very important and even more so at this time since there are some Betta types attracting wide attention which are being discussed as though they were being produced by known genetic processes.

3. Traits which may be enhanced through selective breeding but which represent only the extreme of variation in a single characteristic. There is a natural range of variation in all traits, even if the genetic types are the same. Identical twins may be a lot more like each other than they are like non-identical siblings (or people with different genetic makeup) but they still are not exactly alike. I doubt if anyone would deny that their interactions with the world around them, especially such things as kinds and amounts of food they eat, the kinds and amounts of exercise they indulge in, the things they read or watch or listen to, or the disease organisms they might be exposed to, cannot possibly be identical. Their development and progress through life is unavoidably affected by these and many other factors.

I would superimpose upon genetic potential the nutritional, physiological, and behavioral opportunities organisms have to achieve optional development. Bettas raised in fishrooms in jars with constant water changes and temperature, being regularly and bountifully fed with excellent food and lacking the stresses of searching for food or escaping predation should be able to reach that level. No one would expect genetically similar Bettas developing under less fortunate conditions to do the same. These external factors contribute significantly to the natural variability of the fish and their assorted characteristics.

I bring this up because of the current popularity of such Betta types as "Half Moons" and "Crown Tails." These types are visually distinctive and represent a noticeable deviation from the ordinary or "normal" Betta tail types (whatever they are). The best way to define normal in traits such as the tail fin types discussed here is to look at the wild type. The wild type has 12 soft rays, each branched into two "secondary" rays reaching to the edge of the fin. These rays support a membrane or webbing which normally fills out the space between them. The spread of the wild type tail fin is such that it forms a full rounded posterior margin and the outer rays approach the vertical, in other words, something approaching a half circle. The outer rays are a bit shorter so the vertical profile is more oval shaped. The tips of the rays may protrude just enough to make it obvious where they are, but usually not much. Some species of Bettas and some populations of B. splendens may have slightly more noticeable protrusions of the rays in the middle.

This description is important because what we see in the Half Moon and Crown Tail types is on the one hand variation in the size and shape of the fin; and on the other, a difference in the way the fins rays and membrane appear. In ideal Half Moons the outer rays are proportionally longer ... about the same length as the central rays and there is enough webbing to allow the fin to spread to an approximation of a half circle. Breeders refer to this as a 180 degree spread contrasted to individuals having only 170 degree or 160 degree angles formed by the outer rays. Individuals without 180 degree spread would then not be true Half Moon phenotypes.

In Crown Tails, or what I prefer to call Fringe Fins since other fins beside the caudal (tail) also have ray extensions, the rays extend beyond the edge of the membrane portion of the fin. I think that what happens is that the membranous portion of the fin does not grow as far out as the rays themselves, thus it appears that the rays protrude. Another problem I have with the term "Crown" is that it doesn't work for more than a few of the affected individuals. The membrane seems to extend slightly further between the branches of the rays than it does between the main rays themselves. If one rotates the tail so the rays stick up and the extension is right, the contour does resemble a King's (or Queen's, Princess's, etc.) crown but the trouble is that only certain ones meet that criterion.

Actually, the rays seem to merely grow out to normal length so the abnormality really is inadequate membrane development. This deficiency can be characterized by referring to the percentage of inadequacy similar to the degree of spread in Half Moons. The ray extension might range from say 2-3 percent to 5 percent, all the way to 35-40 percent or more. The more extreme ones have what I like to call a "Rooster" tail because of their resemblance to the tail feathers of a fighting cock. At the time of this writing I have not seen rulings passed by the IBC judging board or statements from breeders as to which, if any, of these variations constitutes the ideal Crown Tail. I trust this subject is being addressed and there will be guidelines of some sort in the near future.

Are There Half Moon Or Crown Tail Genes?

In my opinion the answer is no! Both traits exhibit continuous variation. Both also turn up in relatively small numbers in spawns. Both seem to nearly disappear in F1 hybridization with strains that do not "carry" the traits. True breeding strains of either do not, to my knowledge, exist. They certainly are not traits that are inherited through the single gene, dominant-partial dominant-recessive (Mendelian) mode. They also don't really conform to the multiple factor or multigenic mode although that might come closer. So how can they be explained?

I believe these variants represent an extreme of some part of the normal range of variation of whichever genetic type they are. Another way to put it is that all the fish from a spawn are basically the same genetically and they only vary within the scope of possibility for their type. There doesn't need to be mutant genes for Half Moon or Crown Tail to provide for the types. I believe breeders have seen something desirable about these respective tail shapes and tried to, by selection, breed to enhance them. To some degree they have succeeded.

Selection can and does guide the development of types but the range of possibility is quite different from what it becomes if there is a new genetic mutation. Attempts to "fix" a type that only emerges after persistent selection are likely to be frustrated by the number of generations required and the slow progress toward the goal. In the case of these two Betta types it is clear that the types have been refined and there are strains that produce the ideal, but we still don't have strains that breed true, meaning that matings can be made that will produce all of that type.

Although I haven't worked extensively with either Half Moon or Fringe Fin (Crown Tail) Bettas, I have worked with them and I have a considerable amount of anecdotal information about them from others who have. I believe that anyone who knows will agree that good specimens of Half Moon occur in low percentages in even the best strains. I am talking about numbers like five percent. The rest of the fish in the spawns do not have full half circle caudal fin anatomy. Even if you breed from the best you can get, you are likely to get a low percentage of good ones. If you breed from lesser quality individuals, even though they come from the same spawn, you can expect to get proportionately lower numbers of good ones ... if any at all.

While there may be an ideal type for Crown Tails in Indonesia, where most of the best developed ones seem to be coming from, it isn't clear to breeders in this part of the world what that may be. I think strains are a bit farther along than Half Moons in the sense that a higher percentage of individuals with some form of ray projection appear in spawns but this might be a function of how strictly the type is defined. It could well be that the number of fringed finned individuals might only contain a small percentage which would be considered true or ideal Crown Tail and the rest inferior. A lot like those 170 degree or 160 degree less-than-Half Moons. In the spawns I have, that I got when I crossed Crown Tails with Bettas that have smooth edged fins, none of the progeny show anything more than a very faint suggestion of ray extension. If you weren't specifically looking for them you wouldn't recognize anything different about them at all.

The main concern I have is that we are dispensing information that is incorrect or at least that has thus far not been proven. I hear people routinely refer to fish that are "Half Moon geno," for example, since what they refer to are fish that come from lines that have produced Half Moon phenotypes. Use of the term "geno" implies that these fish have a genotype (from which "geno" has been contracted) or carry a gene that will produce more of them. Since there has been no demonstration of the existence of a mutated gene that is responsible for this trait, and breeding patterns cannot be carried out that could be if there were, it is inappropriate to use that kind of language when discussing them.

From what I have been able to determine so far, I believe the same should be said for the Crown Tail trait, although I'll feel more confident after I am able to analyze more than F1 generation progeny. In conclusion, I'd like to provide a summary statement here to cover the points I have been trying to make:

A distinction should be recognized between traits that have developed by selective breeding and those that occur because of genetic mutations. Oral and written references to either of them should honor that distinction. Traits generated by selective breeding cannot be presumed to be inherited in the more precise manner of those resulting from single gene variations. Specifically, as related in this discussion, the traits in Bettas known as Half Moon and Crown Tail are not the result of single gene mutations and until proved otherwise should not be considered as such.

I would consider it fraudulent to have someone offer for sale either of these fish types (and others that are generated this way) as being represented by any "gene" for the traits. Fish that come from spawns that produced good Half Moons or Crown Tails, without themselves possessing those traits, should not be so designated and cannot be expected to produce more than a very few that are better than themselves.

I might suggest a comparison with race horse breeding. I can't imagine any horse breeder taking two mediocre runners, even those descended from the best, and mating them with the anticipation that they will get superior runners as a result. I realize it could happen but who would waste the time and effort after considering the probability? In the case of a race horse, like these Betta types, there are a number of small factors which determine how one of them will turn out. True, there are genetic factors but there is also nutritional, health care, conditioning and training, and somewhat intangible characteristic of attitude which also contribute to the final product.

The reader may think this is much ado about nothing important but I disagree. I see and hear regularly the inappropriate use of terms genetic with regard to the traits I have been discussing. I have also heard and seen the disappointment, frustration, and anger of people who have been misled and I can't see how this can do any good for this hobby. I think it is the responsibility of breeders who sell fish, especially to newcomers, to understand what they are dealing with and to be as accurate and forthcoming as possible while they do it.

Food for thought?