Background
Pachypanchax is considered ancestral to the killifish and comes in two major types: the fully marine P. plafairi that has adapted to fresh water and the nine nominal species on Madagascar that because of the restricted and stable environment of the island places few forcing functions in the evolutionary path of Pachypanchax ans they are largely unchanged very primitive dino killies that tell us a lot about the origin of the killifish. For example where the backbone meets the tail in killies is complicated and consists of many boned. In Pachypanchax it's a big crude plate, found only in three other killies, all in the Americas indicating they are the least evolved. North America split off 100 my before south America thus are even more primitive. While this was going on fish in Africa were speciaing and developing the bright colors and patterns we know now. South American fish have been doing this for 100MY less and north american fish the least. Indian fish had 100my to develop while in isolation as an island as India "swam" from where it used to be, closer to Gabon, until it came rest in the Asian subcontinent. There is little difference in the color patterns of Aplocheilus and Pahypanchax playfairi. The red dots went away bu the spangled sparkley scales did not. The color pattern in the unpaired fin did not change much which siggests the red dots change the fastest.

Nothobranchius have either plain clear tails of a speficic color or a pattern. Did the pattern evolve from a clear unpattenred fin or vice versa? One species, VRG, has what looks like the remntant of a pattern seen in other nothos but complete. Other nothos still hve taile od clear solid clor i the same eact shade. This pattenr moves outward not inwanrd pver time, thus "pushing" the caudldl bar off the fin leaving a solid color behind.

Rows of red dots are common to nearly all killies and rows of dots of any color are common to even more and these dots follow certain mathematical patterns.

(article about genes unfilding into rows of dots)

So the gene can be seen ti encode for, say "dot space space space space" and "repeat six times so scales come out, from bottom to top just like that. This gives six rows of colored dots.

But look what happened in killies over time: those six rows of red dots morphed into all sorts of different things.

First we have the fish that still look like this - OES, whose color pattern and large size make these at least on paper the least ecolved of any Aphyosemion or Fundulopanchax and could pass for a skinny greenish playfairi. Related fish, Endlers, turn green when alge is abundant with the same metallic green so we can guess what conditions were like for OES, trapped on the island of Fernando Poo and just like Pachypanchx on Madagascar, to real reason to need to adapt of evolve to such unchanging conditions.

There are not that many killies with six rows of red dots still. STR comes to mind ut we see evidence of other variations on this theme: one and two rows of dots. Did Pachypanchax evolve from one to six rows of dots the exace same was as South American Rivulus and African Aphyosemion did? Hardly and it seems much more likely those six rows of red dots remain in a very small number of fish while others have variatios rangfing grom none to one to two an dmoe. Ormage austreale in less than 100 years has thrown spotless and striped varieties, the ofrmr einf stale the latter not.

In Cynadonictys we see two tstripes, and note thaey are not limited to the body. When they eave the body tey keep right on going in the fin and if you look at this riv you can see how CEL or CIN could have eaisy go their tail from a process just like this.

Ths clincher is Roloffia monrovia however as this fish is a rosetta stone for the red dots. From this we can just make out six rows of red dots onan orange background. THe metallic spangle of Pachypanchas is missing and the form (and function) of the body has altered but other than that it's not different from Pachpanchax; bth occur in blue and red forms.

But look at where those six rows od red dots are, ony in the front of the fish. In fact you can look at the worm like markings around the face and you may notice these are contunuations of the red dots that were in rows. Jsut as they don't stop in the body and go into the tail in CYAN then tooo tey go tintot the face of the fish and look like squigkly lines in many spcies of killifish but here they are just forming and were can still see continuety between the rows of red dots and whre they and and the throat markings and where they begin - at the same place.

So that's no mystery, any fish with throat markings used to habe rows of dots that colow and in fish nown for it (SJO) note tht the pattern s consistant in cameroon and nigerian forms bt the oclot snpt,the latter scrlet and gor foemet crmine.

More interestingly the orgin of ghe bars towar th rear o the fish is also easoil explaied here, and these bars show up in BUA, SJO, ELE, KUZ, and even some rivulus.

Look carefully at the zig zag lines in MOR. There much be a differnt mechanism fo the front of a killifish and the back as they are often different from the front, SJO beinf the most notable example, rows of dots at the front, bars at theback. Some gobies have the same color pattern right down oto the vermiform markings on the fase.

But look at this Roloffia and note the zig zag bars, what has happpened here is the gene that "prints" out the pattern "dot space space space" for coloring scales has changed. Now it's off a little nit and each row starts in a different place, one dot hiher each row and instad of going striaght they go sideways. What owuld expain this?

A mutaton would. Whetther they're on two different genes or on two spots on the same genes doens't matter, the point is the color pattern on the back half underwrnt a mutation while the front is unchaged.

What a mutation would have to do is rotate the gene sideways alittle. These are 3d molecules and a small shift inthe phase angle of the chromosomes - and this cangle can probbaly be caldulcated - wold be enojgh to transform the horioxntal rows of dots into diagonlay lines and as har angle changes the lines, whcih are acually rose of dots, what look like bars on this ROloffia are actualy bits of roes of dots and you can still see the dots - and keep in mind this angle canno be arbitrary, bond angles of dna were amongt this first bits of dtaa we gleaned from nascebt milecular bioligy.

This bars in the rear half of the fish can be made like this, and it must have been the same in WAL, BUA and others.

Now, when as is the case of the rivulus the bars extend into the tail fin then when they undergo the same mutation the same tjig happens but as the tail expands in lenth as the fin rays elongate the pattern is stretched. This the odd tails in ROlofia are the earlust morphingof rows of dots that underwent a single mutation in colot pattern combined with the tentandy of a post-packpacans caudal fins to elongate.