Skip to content

Design of betta greenhouses for selective breeding

Design of betta greenhouses for selective breeding

Dimensioning ornamental fish greenhouses, especially for bettas, requires medium and long-term planning and vision. This is because the number of fish tends to increase a lot, considering the number of fry that are born and survive each litter – which depends on the breeder’s management.

Often the breeder starts with a couple of bettas, has an aquarium for the mating and growth of the fry, and other accommodations that are used for the couple separately. Or, it already has aquariums or plastic boxes, which can be of different sizes, which makes standardization and better use of available space difficult.

However, an important observation is necessary here: there are two basic paths or purposes that will characterize the dimensioning of the greenhouse:

  • Without genetic control of the strains;
  • With certified genetic control of these strains;

The method here called without genetic control of the strains is characterized that juveniles of different litters will be accommodated together, causing mixing and loss of ballast when identifying which are their matrices or ancestors, except when compatible to avoid mixing (we will see more forward). As a rule, it will reduce the number of aquariums needed to accommodate fish.

On the other hand, the design of betta greenhouses with certified genetic control requires more resources, both in space, as in investment and efforts for control. In this article we will detail the design of betta greenhouses with certified genetic control.

This type of work is based on premises. Let’s see:

Premise 1 – regarding management:

  • a litter of a betta takes two to three months to develop, from birth to placement on individual tanks, therefore, a tolerance of three months was given in the occupation of the tanks due to the stragglers;
  • a litter will always be separated into two growth tanks (or plastic boxes) – called tanks;


Premise 2 – regarding blood lines within the same lineage:

  • there will be, whenever possible, three different blood lines;
  • in each of them, it will only be possible to advance to F3;


Premise 3 – other important points to note:

  • the number of lines that will be worked simultaneously;
  • there are strains that can coexist growing in the same tank, as long as their development is compatible. In other words, there is a lineage that develops faster than another; therefore, placing them in the same tank can be detrimental to the one with the slowest development. Another point that can be commented on is having strains with completely different phenotypes. Eg, a lineage of reds, and a lineage of blues (because, if the lineages of blues turn red, or vice versa, these “lineages” are very bad, and it is not even worth continuing with any type of work improvement or selection).
  • similar phenotypes, however, from different lineages, cannot be together in the same tank.

As an example, we can mention copper and black copper. If we want more rigorous control over genetic work with these strains, it will not be possible to accommodate such litters together, as we know that from black copper bettas copper can always arise.

Breeding bettas with certified genetic control: start of work

a) after obtaining the desired pair, we perform the crossing, placing it inside tank 1 (TQ1). So we started our bloodline 1 (L1).

b) this litter (F1) will be together for fifteen days in TQ1 and, after that, a part of it (more or less half) will be placed in another tank (TQ2), where they will remain for three months for its growth and development, in the same way as those in TQ1, of course;

c) concrete mixers must be made available to accommodate the males that stand out. Many floating plants (water ferns, eg) in TQ1 and TQ2 tanks are recommended to avoid the stress process between bettas. At this point, we have mixers with the father, the mother and the males.

d) the following should be chosen:

– a female to continue L1,

– a brother to cross with her, and continue L1 (in F2), and

– another to go back to the father and create blood line 2 (L2).

I advise, at this point, to:

– to undo all latecomers who are in TQ1 and TQ2;

– separate another couple of brothers as security and

– to undo all other siblings (males and females).

This procedure must be followed in any blood line, and in any “F” generation within that line.

At this point, we will have at least seven tanks occupied, as follows:

  • dad;
  • mother (although it may no longer be used – or only in urgent cases);
  • male (who will continue L1);
  • female (who will continue L1);
  • male (reserve);
  • female (reserve) and
  • female (who will start L2).


  • Note that the same female was not caught to start L2 and continue L1, thus avoiding an increase in the blood rate, because even among siblings, there will always be some differences in genetic loads;
  • The female matriarch mother was not used to breed with her child, because in most cases, the females grow a lot, gain weight, and the children are unable to mate. But, nothing prevents that, if such a mating is possible. It will be another line that will emerge. This possibility will not be considered in this work, but after understanding the work philosophy presented here, feel free to dimension the necessary structure with such variable.


Premise 4 – continuing

a) at that point we already get F2 at L1, and F1 at L2. See the diagram below:

a) in L1, F2 was obtained by crossing two F1 brothers;

b) in L2, F1 was obtained by crossing the father with the daughter (from L1 in F1).

c) at this point, we take a male from L1 in F1 (preferably the one that was used as a reserve and that was not used) and crossed with a female from L2 in F1, generating the blood line L3.

But, why not use an L1 male in F2 instead of the L1 male in F1?

Here there is a range of discussions and will always depend on the quality of the genetic material that is emerging (phenotypes). If the male of F2 of L1 is better than that of F1 (of L1, of course), you can use it and get rid of the other. However, if the difference is not very significant, the “genetic distance” (loss of genetic load, or heredity) of the L1 F1 will be greater than that of the L1 F2, for the generation of the female L2 F1.

This fact is not linked to age, but to a greater distance from the sets of genes (genetic loads) of both generations and different blood lines, giving chance to other different combinations of genes, with less action by “harmful” genes that can arise from these matings between very close relatives, due to high inbreeding. At least, I see it that way, without any scientific basis to support me in this statement. My experience of years, at least, has seen such a trend.

Viewing the discussion so far:

From the above, we see that the use of an L1 male in F1 will present an L3 with a less closed genetic load than that using an L1 male in F2.

  1. a) whatever the choice to generate L3 using a female from L2 in F1 with a male from L1 – L1 in F1 or L1 in F2, we have as a result:

Premise 5 – we arrived at F3 at blood line 1 (L1)

a) according to premise 2, when reaching F3 on any line, there is a need to do an “out crossing” – a total opening of the blood line – to start that line again.

b) in this case, L1 – as expected – first reached that mark. You can open that line with an individual (a male or a female, it doesn’t matter) from another breeder. This matrix must be as far away as possible genetically from those with whom one is working from the point of view of kinship. In other words, no relationship between the couple is the desired situation.

Then, we can show the evolution of the crossing schedule:

Note that the same discussion that took place in (5d) could be reopened here, regarding the “new cycles” of blood lines L2 and L3. But why not use individuals who are genetically distant at the expense of those closest to them?

The new L2 cycle could be developed with an individual from L3, but in F1 – instead of F2 -, for example.

The same reasoning could be applied to the new L3 cycle, using F2 as an individual, instead of F3 (or even F1).

This implies a greater amount of concrete available to accommodate these individuals, in order to have them available at the right time.

With this strategy, one can continue the work safely and continuously.

Premise 6 – final considerations

a) we must not forget that at any time one can make use of openings of blood, or even, crossings between different lines of blood (line breeding), without the obligation to reach F3. It will depend on the results obtained in each case.

b) the same applies to advancing into each line (inbreeding) and passing F3 (F4, F5 etc.). The care I recommend to take for this last topic is the careful observation of the structural results of the litters. Regardless of what grade “F” is that blood line. Possible increases in the number of defective individuals: weak, sick, with little development, “draggers”, with the appearance of bumps or cavities in the body – especially in the head – are some of the typical problems. In addition, bettas with a protruding back (dolphin head and very “pointed”) and / or crooked fins (or their absence) are also defects. And yet, disproportionate bodies comparing the set of fins (lack of aesthetic balance, where the betta has a small body and immense fins, or a huge body and small fins).

c) preferably, photograph all bettas with which you have worked with the same quality and image / photo characteristics: same camera and settings, same lighting and aquarium, with glass and water transparency. In addition, if the line is made of metallic bettas, or with a high iridescence index, these conditions become even more necessary.

d) lastly, it is worth remembering that all this work was developed for a single lineage! If you want to work with more than one strain, multiply all the quantities involved by the total number of strains.

e) as to the number of concrete mixes required, I will leave it to anyone who wishes to apply the information and recommendations contained in this work to dimension them, because depending on the philosophical option that I use, this number can vary a lot.

It is clear that, if its management is not appropriate to enable an adequate development for its bettas, in such a way that the times of 3 months are not sufficient for their relocation in individualized concrete, all the total times will change, but not the positions relative to the occupations of the aquariums and tanks in the schedules. In other words, the schedule will stretch.

f) A separate strategy is that, when you get to F3, in any job, with 3 blood lines, eg, instead of “out crossing”, with bettas from other breeders that we often don’t have the information regarding their genetic histories and, for that reason, we may put all our work to waste. In these cases, we can do “line breeding” (exchange of copies between blood lines) with the three lines, as follows:

What will guide you in choosing whether to be the male in line 1 with the female in line 2, instead of the male in line 2 with the female in line 1, eg, will be the quality shown in the respective phenotypes (shapes, colors, and color distributions), and also, its objective to be achieved with that specific work.

In this case, we will have the new F1: L1 with L2, L1 with L3, and L2 with L3.

There may also be a case where the results presented on a given line are not in accordance with its objectives, which will lead to abandoning it, with total disposal, and working with one line less.

We wish you success with your Bettas!

(c) 2014 - 2022 Betta Project - A DNA Experience. All rights reserved