The Lightning Project

The ongoing saga of the PNG Lightning Maroon Clownfish Breeding Project

Browsing Posts tagged genetics

Filials, What the F?

No comments

It seems that at least one irate aquarist thinks I’ve made up the concept of “F1″, “F0″ and so forth. Maybe as a marketing ploy even?! Nope, sorry, they’ve been around forever.

Originally, this content was intended for inclusion in the Conservation Breeding Chapter of my half of the book, Banggai Cardinalfish, which was published last year. Most of that last chapter was cut…we were 17 pages over-length and the most esoteric parts of the book were the first to go. But, I found the text in my drafts, and I’m publishing it here, very slightly edited from what I found in my drafts.

Filials Denote Generation and/or Inbreeding

You will routinely see the use of tags like F1, F2, and F3 attached to aquarium-produced specimens of wild-type forms in the freshwater trade, most often attached to fish from private versus commercial breeders. These notations are called filials, and are meant to convey the generational distance from the last “outcrossing,” that is, a mating between two completely unrelated fish.

Officially, F0, or generation zero, represents a pairing of two unrelated fish regardless of the source, and the resultant offspring are F1, the first generation. Mate those F1 fish together and you have F2, and so forth. However, in some cases, including in the general aquarium hobby, F0 more often used exclusively to mean a wild-sourced broodstock. This slight difference in use can cause a significant level of confusion.

From an inbreeding standpoint, and for tracing the extent of inbreeding, the use of F0 to denote the parents of an unrelated mating is both practical and proper. From a conservation standpoint, breeders are often concerned with “distance from the wild.” Given the possible conservation implications, and that the general accepted practice among freshwater hobbyists is to use filials to show distance from wild genetics, Banggai breeders might consider implementing the methodology outlined here.

F+ the number of generations away from wild-sourced genetics. Thus:

  • F0 denotes a wild fish.
  • F1 is generally the progeny of a wild pairing.
  • F2 is the offspring of an F1 X F1 pair, related or not.

When fish of different generations are mated, the resultant offspring are F+1 to whichever mate is already further away. For example, F3 X F5 = F6, not F4.

Problems arise when a captive line is outcrossed back to a wild fish. This fundamentally resets the inbreeding that has occurred, so F0 x F6 should be called F1. Many aquarists and breeders may disagree, considering such a “reset” as deceptive because only half of the parentage is wild, and thus insist that F0 X F6 should be denoted F7. Some breeders sidestep all this debate, using F0 for wild fish, F1 for progeny of wild fish, and “aquarium strain” to denote everything else.

The important final message is that filials are used by different people in different ways. From a genetics standpoint, the filial is hardly the final word, because it can mean two fundamentally different things: one breeder’s F6 fish may be far more genetically solid than another breeder’s highly inbred F3 specimens, and a third breeder’s F1 fish might be 10 generations removed from the wild. If there is any doubt, a conversation with the breeder should help clarify the meaning of the filial label he or she has applied.

As promised, I finally got time to look at the offspring that Mike Doty had managed to rear from my F1 Lightning X Lighting Maroon Clownfish pair.  While Mike had initially stated he felt the split was 50/50, today’s visit showed something different.

All told, I only managed to count 3 white stripe offspring in the BRT. Meanwhile, there appeared to be 6 distinctive Lightning Maroon-type offspring. What I didn’t see were any fish that, at this point, looked atypical from either known phenotype.




With only 9 (approximate) juveniles, the sample is far from conclusive. The 3/6 split could be representative of anything from a 50/50 split to a 25/75 split. A new phenotype could be missing because none simply survived in this first successful run.

That said, the 3/6 split, if a valid sample, would represent something we perhaps don’t want to  see.  It would imply that Lightning is straight dominant, but it would also imply that a double-dose of Lightning is in fact fatal to the offspring and they fail to develop.  This is one of the current working hypotheses for the Snowflake gene in Ocellaris, and anecdotal reports continue to bolster that line of thinking (reports are anecdotal in so much as a breeder saying “It’s about a 60-70% snowflake result” is not the same as a breeder saying “I got 140 Snowflakes and 65 wild types in this clutch”).

It is fair to say that there is still hope for these fish – they are QUITE YOUNG and all we can truly discern at this point is that they outwardly either are white stripes or aberrant.  My tune could change as these babies develop, or if subsequent results are different. It is not surprising to me though, to see similarities emerge in various mutation types.  We already have Picasso/Platinum and the sister mutation of DaVinci/Wyoming White. Much as I wonder if these twin mutations could represent the same gene in different species (or simply a similar genetic mutation in sibling species), would we come to find out that Lightning is in fact not unlike Snowflake? After all, genetic analysis has revealed that the Maroon Clownfish are in fact very closely related to the Percula/Ocellaris complex, so it would not be surprising to find similar or the same genes present in all these species based on their common shared ancestor.

It is also an interesting footnote to observe that these tiny juvenile clownfishes DO represent a the first F2 generation of Lightning Maroons.

Here’s a quick rundown.

The 5th spawn of the F1 Lightning X Lightning Maroon Clownfish pairing was collected and hatched by Mike Doty while I was away.  In short, he scraped off the eggs, hatched them in a 1 gallon jar in a water bath with simple aeration, 75% clean new water.  Come November 28th, Mike relayed that settlement had started.  The moment we’ve been waiting for was here – is there something new?

Well, Mike’s first words were “About 90% sure we have some normal striped fish.”

This, of course, does the following:

  • likely rules out Lightning as a simple recessive gene. If it WAS recessive, then both parents would be “double dose” aka. homozygous, represented as l/l, which means that each parent could only contribute a recessive lightning gene, and thus, each offspring would also get one copy each, one from mom, one from dad, and thus, could only be l/l as well. For the moment, while another couple test matings will bolster the data, the fact that there are white-stripe offspring pretty much precludes this being a standard single allele, single locus, recessive trait.
  • does not rule out straight dominance. If it was straight dominance, and each parent is “single dose” aka. heterozygous, represented as L/+, then 25% of the offspring would not get a gene from either parent, and thus, 25% would be white stripe maroon clownfish.
  • nor does it rule out partial dominance. This of course, would work the same way as dominance, except that 25% of the offspring would get a lightning gene from EACH parent, and would be homozygous for Lightning, represented as L/L. This is the scenario that most people are hoping for, because with the new homozygous state, there comes the potential for a new phenotype that could be different from the Lightning that we know.

As of today, 12-8-2014, I spoke with Mike briefly and have to relay this news – while he doesn’t have many babies left, he believes that the phenotype split is roughly 50/50.  That is to say, half white stripes, half lightnings.  So far, he also has not seen anything unique or new in this F2 generation.  I have yet to see the babies for myself, and have yet to take pictures or do a headcount, but these cursory, informal results, mirror another clownfish mutation that seems to not fit the mold as we’d expect – SNOWFLAKE in Ocellaris.  It’s my hope to get over to Mike’s today yet to see for myself.

In other news, the 6th spawn of the F1 Lightning X Lightning pair was put down on 11/29/2014. It appears I finally won the battle of the tiles:

LxL Spawn #6

LxL Spawn #6

Meanwhile, I brought some new clownfish into the fishroom earlier in November (the 19th and 22nd) and was trying out the new Ick-Shield food from New Life Spectrum.  This is basically a Chloroquin-laced pellet food that is meant primarily to prevent disease such as Crytopcaryon, Amyloodium, Brooklynella etc…pretty much the things which are sensitive to the active ingredient. I decided to not simply feed this fish to the new arrivals, but also to feed it to one of my holding systems AND the wild Lightning Maroon and her mate as preventative medication, just in case.

Well, it turns out that there is an unfortunate side effect to this feed; it seems to shut down breeding activity.  All my routine pairs stopped spawning. The Lightning and her mate did finally put down a spawn on December 1st, 2014, #46.




Unfortunately, it appears as though the spawn was not fertilized…the eggs didn’t develop, and after 48 hours they were gone. A few days after that, I read, anecdotally, that Chloroquin can cause male sterility??? Not permanent according to the rumor, but certainly a potential setback. As far as the efficacy of the food, my jury is out. The larger fish which were feeding well on it by and large remained disease free, but not all did.  I still had a Brooklynella outbreak, although not in the fish I would have necessarily expected. Once that outbreak started, it then affected other fish as well despite their feeding on the pellets.  I’m also seeing either Cryptocaryon or Amyloodium on fish which were visually “clean” upon arrival, which were in dedicated QT systems, feeding on this food from day one.  So the question here is were they simply getting ENOUGH feed as they were small fish which cannot readily eat the small pellet size.

So of course, one is left with questions, not answers.  There is no way to say the food didn’t work, nor is there any way to prove that it does work. Absence of disease is not proof of prevention, that much I know for certain. Lack of a cure, or lack of prevention, which IS documented, only raises questions about why it didn’t work as suggested and certainly requires investigation (eg. would a smaller pellet size be better accepted…could these failures stem from simply lack of feeding, or lack of sufficient feeding, thus insufficient dose to the fish?).

Circling back to LxL Spawn #6, as the week progressed an interesting change in behavior occurred starting around December 4th, 5 days post spawn.  The larger female F1 Lightning became belligerent towards the male, and over the day drove him from nest tending duties.  December 5th, a Friday, would have been 6 days post spawn, and the night of the first hatching.  I was simply swamped with preparations for sending our dog to live with my brother, and failed to pull the tile.  By morning, Saturday, December 6th, 2014, it appeared that I had not missed much…most if not all the eggs were still there. The pair remained at odds.  We left for the weekend to ship our family dog, and upon Sunday, December 7th, there were still a few dozen eggs remaining, although they appeared potentially dead and disappeared throughout the day.  By nightfall, the pair was starting to be less antagonistic, but I am still keeping a close eye on them.  Hopefully, we’ll get another spawn soon – this was the only mature pair in the house that didn’t receive Chloroquin-laced foods (as they don’t reside in the fishroom with the rest of the fish).

So yeah…October 2nd, I’m just walking by the tank and I see this…

F1 (Lightning X Lightning) first spawn for the Lightning Maroon Clownfish

We are about to embark on the next exciting chapter of the Lightning Maroon Clownfish and genes.  My hypothesis is that the Lightning gene is partially dominant or dominant. The offspring from this spawn, if successfully raised, will yield important answers that may shed new light on the genetics at play. Will we see something NEW come out of this? Well..I certainly hope so and it looks like I’ll get to name it!

Let’s give that a closer look:










I have photos of this spawn stashed away in the camera, but I figured I better get this on the books before I completely forget. Spawn #19 was put down on 12-14.  With tons of stuff going on with the holidays, I did a boneheaded thing and pulled the nest on the night of the 19th going into the 20th (because I looked at my phone after Midnight, didn’t realize it was that late, saw the 20th, and said “shoot I gotta pull this tonight!”).

Spawn #19 during H2O2 sanitizing dip.

Spawn #19 during H2O2 sanitizing dip.

So the eggs went into a clean tank with all new water after a 15 minute bath on hydrogen peroxide.  It was at least 24 hours early, and yet despite that, I had a handful of offspring the next morning…that’s technically 5 days after spawning!  A few more offspring hatched out the subsequent night (which would have been 6 days later), and yet after that, the nest died.  There has been plenty of RotiGreen Omega and rotifers in the 10 gallon blacked out tank, but by the time I departed for our holidays away, there were maybe only 3 babies I could see alive.  Upon returning the day after Xmas (12-26), I could find only one baby in the tank, and by the 27th I could no longer find any larvae.  Pulling my hair out.

There is one thing I’m looking at right now – I haven’t had access to wooden airstones for incubation in ages, and Joe Lichtenbert swore by them for direct egg aeration during hatching.  I’ve been having all sorts of nest deaths in the Lightning Maroons when putting coarse air directly on them, so I’ve generally tried to get good ambient flow, which works better but not consistently.  Well, if it’s my damn airstones I’m going to be really ticked, but my plans are to have wooden airstones on hand very soon, going out of my way to obtain them since most online suppliers have long since dropped them from their offerings.

And yet, I’ll get another shot at it still.  I discovered the latest nest, Spawn #20, when checking out my tanks when I returned home from the holidays.  I cannot say for certain, but given that the Lightning Maroon’s ovipositor was still down, I believe that Spawn #20 was laid on 12-26-2013.  Based on that timing, the nest was pulled on the evening of 1-1-2014, going into the morning of 1/2/2014.  Once again the 10 gallon tank was drained, cleaned, rinsed with freshwater, and set up. This time I used 50% new water, 40% broodstock water, and 5% RODI water to bring the specific gravity down slightly.  I did not sanitize the eggs prior to pulling.  There was very little orange left in the larvae, which I believe bodes well for my pull occuring at the right time to maximize whatever hatch I get.  A small LED flashlight was set up at the far end of the tank to draw hatched larvae away from the heavy aeration – wooden airstones have yet to arrive, so once again I’m pummeling the eggs with coarser bubbles from a glass airstone.

You may recall the “other” pair of wild (F0) White Stripe Maroons I’ve had set up from PNG.  I’ve had the pair a long time..they were paired before the Lightning Maroon ever got around to it.  By my records, they put down their first spawn on 12-20-2013.  Upon returning home today, I checked their nest; just a few unhatched eggs remained, which means that they hatched even FASTER than the Lightning Maroon.  I would have had to pull them the late evening of 12-25….just over 5 days post spawn.  The other possibility is that the parents ate most of the eggs prematurely, so I’ll have to watch the pair another few times through before I see what their pattern is.  But seriously…as little as 120 hours from spawn to hatch?!

I will absolutely raise a batch from this other pair this year, and the reasoning is simple.  First, we’ll want to see if by some odd chance they throw out Lightning ( I don’t believe for a second they will).  Next, we’ll want to see how many show up with horns and spots and other aberrant, but not “Lightning”, markings. Ultimately, regardless of what we get, this new and completely unrelated line of F1 PNG maroons will be the IDEAL outcross for the F1 Lightning offspring.  Instead of the F1 generation of PNG Maroon Clownfish stemming from only 2 foundation fish (which is really not enough), we’ll double the foundation population to 4 fish.  According to FAO documentation on conservation breeding of fishes in captivity, this is still too small of a genetic bottleneck, but since I am aware of a couple other breeders who have Lightning Maroons and claim to have unrelated F0 White Stripes to mate them with, we could realistically be on very solid footing for the long term captive viability of the PNG provenance lineage of Maroon Clownfish even if we have no further access to PNG fish for months, years, decades….

F1 PNG White Stripe Maroon Clownfish, WS12

WS12 – does this fish carry critical Lightning genetics, or is it just a plain white stripe Maroon?  We simply don’t know!

More than any other question I’ve received in the past year, is one that once again, came up in a private communication from Carl W. Phillips Jr. tonight.  He asked,  “Mr Pedeersen whats you opinion of what the stripe maroon will throw some lightning ??”

My response: “I’ve written about this rather extensively on and elsewhere. My gut feeling is that they are not carrying any special genetics; their value is in maintaining the PNG provenance, as well as the unknown results of Lightning X Lightning…what if that cross is fatal? I think the results of the parental cross do suggest a genetic inheritance, and it’s reasonable to assume that the mating of the Lightning Phenotype to the regular White Stripe Phenotype ought to result in the same phenotype spread in the F2 generation. Here’s some details for you to review ->

Obviously, we still do not have answers to the genetic questions surrounding the Lightning Maroon…so far all we can say, based on observations, is that the pattern seems heritable (of the 2 very small that hatched and survived in the spring 2013, one was White Stripe, one was Lightning).  However, the 50/50 spread in the F1 generation is a possible phenotpye ratio for all three common forms of genetic expression.  It could be that the white stripe siblings, with their slightly scalloped edges on their stripes and increased horns and spots might be representative of a very subtle “singe dose” form of a partially dominant lightning gene.  Then again, it’s just as possible that very subtle patterning is irrelevant / separate from Lightning altogether.  And that could mean that they could carry a hidden single-dose recessive lightning gene, or, in my opinion, they might simply not carry any lightning genetic material themselves.  As I relayed to Carl, it’s best to review the full rundown of the Genetic Possibilities behind the Lightning Maroon Clownfish – in drawing on what we know from other fish and other clownfish specifically, I put less than a 1/3 chance that Lightning is recessive, which means I also think there is less than 1/3 chance that the regularly patterned offspring carry a hidden lighting gene.

So to purchase a white striped sibling on the speculation that two of them mated together will produce lightnings is just that – speculation.  Then again, if Lightning was revealed to be the result of 2 recessive alleles, or if “Lightning” is the double dose of a partially dominant allele, then these white stripe siblings would be very important to the creation of MORE Lightnings.

In the end, it’s fair to say that we absolutely do not know, and therefore couldn’t make any promises of any kind regarding the genetics.  It will take many more crosses with different pairings, and counting of all the offspring that result, to help determine with some certainty exactly how the Lightning gene works.  I’m going to guess that it will take another 3-4 years before we have that answer.

- Followup – not even 12 hours later, I’ve answered this exact question at least a half dozen more times today, and seen at least one internet post from someone who stated that the Lightning trait is “recessive” (which while possible, is in my opinion highly unlikely and absolutely something we cannot know at this point…the data we have doesn’t exclude or point to any one option).

A special thanks goes out to Shane & Len at Advanced Aquarist for co-publishing this blog entry there - it seemed like the perfect contribution given the style of material they like to cover.

The Lightning Project

Just to bring you up to speed, it’s been over 2 years since the “Lightning Maroon” from PNG made it’s way to the US and ultimately into my home tank.  It was a long time coming but in the late spring and early summer of 2012, we finally got a glimmer of success with the first and second spawning between the “Lightning” Maroon, and a normally patterned “wild-type” Maroon clownfish.  The pairing below are the two parents responsible for everything we’re about to cover.

The Lightning Maroon Clownfish and her wild-type mate.

The Lightning Maroon Clownfish and her wild-type mate.

The Results Are In

So not only did we get a spawn that made it, but we got roughly 50 juveniles post-settlement in the very first rearing attempt.  Now, more than 2 weeks in, it seems ever more likely that we have a roughly 50% rate of “Lightning” babies in our group of offspring – note I have yet to do an actual headcount, this is just a ballpark guesstimate on the numbers.  Initially the babies showed up with blue “caps”; thicker headstripes that were readily discernible.  As they progressed, they looked more and more like Picasso Percula babies.  While still possibly premature to say conclusively we have “Lightnings”, we’re definitely starting to see signs that the Lightning trait will come through with defining characteristics that will clearly match up with those that the two original wild Lightning Maroons shared.

17 day old Offspring from the Lightning Maroon & a wild type Maroon from the same island in Papua New Guinae

17 day old Offspring from the Lightning Maroon & a wild type Maroon from the same island in Papua New Guinae

Dismissing the Hybrid Hypothesis

Before going into the genetics discussion, I’m going to address one “possibility” that some creative thinkers might propose, either though just being “creative”, or through having read, misread, or misunderstood what someone has posted on some forum somewhere.  The hypothesis is this; the Lightning Maroon is a different species than a normal white stripe maroon.  And thus, are these offspring “hybrids”?

Categorically I firmly believe no, the Lightning Maroon does not in any way represent a species other than Premnas biaculeatus.  In most hybrid scenarios between two species, the initial primary hybridization generally yields a predictable intermediate form between the two parental species – I am sure there are examples of a primary hybrid where the offspring “range” from one parent to the other, but that is far more common in the second generation if it’s going to happen.  Since we have no intermediate forms in the offspring of this pairing, I believe we can safely rule out the “hybrid” hypothesis without further delay (the same cannot necessarily be said if we look at the “White Stripe” vs. “Gold Stripe” Maroon…the more I read and learn and see…leads me to believe these may in fact be two distinct species in the wild).

Let’s talk Genetics, Breeder Style

I’ll state up front that I’m no geneticist, and that I’ve been known to get my terms confused. So I’ve taken the opportunity to run this by Adeljean Ho (a good friend of Dr. Matthew L. Wittenrich, and the scientist who published work in CORAL that suggests a unique genetic basis for the “Red” form of the Green Mandarin).  Hopefully he caught any errors I may have made in attempting to distill and disseminate these ideas.

Remember, I really downplay the “designer” aspects breeding of marine fish with mutations, but taking on the preservation of this wild trait has forced me to learn it.  Understanding the genetics allows a breeder who is working with “designer” fish to quite literally “create” what he or she envisions; the upshot of this knowledge is that it also levels the playing field for breeders, forcing them to turn back to producing QUALITY fish in order to differentiate themselves.  For me, the emphasis on quality, as driven by “open sourcing” the genetics of a fish, is the best route we can go if we must pursue “designer” variants going forward.

In this discussion of possible Lightning Maroon genetics, here are the important terms. We will try not use the term “gene”, because it kind of gets used interchangeably and thus will probably only confuse. The important terms here are “locus” and “allele”. “Locus” being a specific point in the genetic code where a particular pairing of alleles resides; the alleles being the pieces of genetic information, one from the father, and one from the mother, that come together at the locus to form the genetic makeup of the offspring.

We also cannot neglect the terms genotype and phenotype.  Genotype refers to the genetic “code” specifically, which is important because alleles can be present yet not “expressed” in the phenotype.  Yes, the phenotype is the outward appearance as driven by the genetics.  And this is the conundrum; due to the way certain alleles interact with other alleles, there are traits that can be masked, surpressed, or unexpressed, that is to say you won’t know a fish carries a hidden albino gene in its genotype just by looking at it (and seeing it’s phenotype).

The other important terms to remember are “homozygous” and “heterozygous”; all that really means is whether the two alleles in the loci pair are the SAME genetic code (homozygous, such as A/A or B/B), or different (heterozygous, often abbreviated as “het” for short, such as A/B). Considering the entirety of our genetic makeup, it all boils down to loci (plural of locus) and what pairing of “alleles” is inherited at each locus. Obviously, the outward result of these traits is the result of all these separate loci together, and certainly some observable traits may be governed by multiple loci, which makes it difficult to ascertain the genetics and inheritance behind them. By the same token, the possible individual alleles that can be present at a loci are perhaps infantasimal in their variation (for example, ABCDEFGHIJKLMNOPQRUSTUVWXYZ), but only two individual alleles (eg. A/Z, C/C, or B/Q) will be present in any particular locus.

That said, most all of the genetic variations that we’ve come to openly understand in fish seem to be the result of the genetic makeup of an individual loci, and from there, the combination of multiple traits at different loci  is what gives us a well-understood, massive diversity of ornamental fish varieties (as some would call them, “Designer” fish).  Freshwater Angelfish (Pterophyllum scarlare) make the perfect example as they are well understood genetically (see The Angelfish Society’s Phenotype Library). Combining multiple traits from individual loci is how we get a Pearlscale Lace Clown Veil Angelfish. Those names refer to phenotypes; outwardly discernible traits, in this case those names refer to scale structure + dark gene + stripe genetics + fin length. In the Angelfish breeding community, this would get denoted roughly as (p/p) – (D/+) – (Z/S) – (V/+).  Because we know which alleles at which loci contribute to each end result, in theory any breeder can “make” a Pearlscale Lace Clown Veil Angelfish; the breeder just needs to have the proper parents with the proper genotypes.  The breeder also knows that mating two Pearlscale Lace Clown Veil Angelfish together will result in a plethora of unique genetic combinations, 27 to be exact, all of which have their own name.  One example?  Pearlscale Blushing Superveil Angelfish, (p/p) – (+/+) – (S/S) – (V/V).

I’ll borrow notation from the freshwater Angelfish world to try to lay out the options for the Lightning Maroons, and I’ll propose that “L” will stand for the Lighting allele. “+” will stand for the wild-type, default state allele (aka. a normally striped fish). Thus, a wild fish, without any “Lightning” genetics, would be represented as (+/+). Note that in this notation, capital letters are normally used for dominant or partially dominant traits, whereas recessive traits are generally denoted using lowercase letters.  I’m going to assume right up front that “Lightning” is a trait that directly controls the “striping” of the fish.  We are going to assume here that there are only two possible alleles involved in what we are seeing, and that the Lightning Phenotype is driven by one specific allele (L in our examples) and is not in fact the result of two unique alleles coming together (eg. Lightning = L/X, wild fish being +/+).  We are also going to assume that the Lightning trait is the result of genetics at one locus only. A brazen assumption, but it seems likely at the moment.

To explain the multiple locus issue another way, we are assuming it does not take the genetics of two (or more) loci to result in the Lightning phenotype.  In the Angelfish world, there are  phenotypes like Platinum that are the resultant combination of two independent loci, and the presence of specific recessive alleles in homozygous pairings, that result in the all white Platinum Angelfish – in this case the recessive gold trait on the “dark” locus, and the recessive Philippine blue trait on the “philippine blue” locus. Independently, you’d have a Gold Angelfish, or a Philippine Blue Angelfish, but “activate” both of those recessive traits through breeding choices, and you wind up with the possibility of all white platinum offspring.  Yes, you can “make” a Platinum out of parents that are not outwardly “Platinums” themselves, and that is the beauty of understanding the genetics.  What you cannot do is use only Platinums to breed back to the wild form of an angelfish – and that is the curse of “designer” breeding (which is one reason why designer-focused breeding can get in the way of conservation minded breeding – the “ornamental” genetics can function as actual genetic contaminants…but that’s for another day).

Angelfish Phenotype Examples

Genetics in action - the large fish in the foreground is a wild Angelfish, generally presumed (+/+) unless it carries hidden recessive alleles, and called a "Silver", which is the default striping pattern in Pterophyllum scarlare. The Blue one at right is a "Blue Ghost", representing 2 doses of the Philippine Blue allele, and a single dose of the partialy dominant stripeless allele, so (pb/pb) - (S/+). The white angelfish in the back is a "Platinum", the result of a fish being both homozygous for the recessive Philippine Blue allele, as well as homozygous for the recessive Gold allele, thus (pb/pb) - (g/g).

Based on the quantity of Lightning Maroons in the very first batch of offspring, there are three possibilities for how the Lightning Maroon trait genetically functions.  It could be a recessive trait, whereby there must be two alleles for Lightning present in order for the Lightning pattern to be observed.  Lightning could in fact be a dominant trait, whereby it only takes a single dose of the Lightning allele to mask the normal stripping pattern.  And the Lightning trait could be the result of a partial (incomplete) or even codominant allele, where a double dose fish will look different than a single-dose fish, which is still different from the wild type, normally-barred fish.
The Case and Implications for “Lightning” being a Recessive Trait

So let’s look at inheritance and expression of the genetics in play. We’ll start with the easiest to understand, a recessive trait like albinism (I think we all understand how albinism works on some basic level). Another good example – the recently discovered Philippine Blue gene in Angelfish is thought to be recessive.  A fish with a single dose of this allele (pb/+) shows no real difference with the wild form.  But put on a second dose, and *Bam*, you have a Philippine Blue Angelfish.

A Blue Silver Angelfish

A Blue Silver Angelfish , (pb/pb). The angelfish breeding community is thoroughly convinced that pb is a recessive trait on its own locus.

This angelfish is a "Silver", and happens to be a sibling to the Blue Silver angelfish shown above. There is a 2/3 chance that this fish has a hidden Philippine Blue allele, denoted as (pb/+), otherwise it is wild-type in every known sense, written as (+/+). If these two fish were mated, and none of the offspring developed into blues, that would prove the 1/3 chance of this fish having no hidden blue allele.

If “Lightning” is a recessive trait (one that requires two “doses” of the Lightning allele), then the Lightning parent could only be homozygous (l/l).  A fish that is heterozygous (l/+) would appear “normal”. Thus, if our (l/l) fish is mated to a wild type (normally barred) fish with no Lightning genetics (+/+), all the offspring would be (l/+). Such a pairing would result in 0% discernible Lightning Maroons, as all offspring are (l/+) (Figure 1).

Recessive Lightning to Wild homozygous Mate = all hets = no Lightnings.

Figure 1. Recessive Lightning to Wild homozygous Mate = all hets = no Lightnings.

Thus, if “Lighting” is recessive, we know that the Lightning Maroon must be (l/l). If recessive, to have found Lightning offspring in the first generation mating, that implies that the standard-barred mate must carry a “hidden” Lightning allele, and thus be (l/+) itself. Mating (l/l) to (l/+) would give you a 50% expression rate IF (and that’s a big if) the Lightning trait is recessive. Mathematically, the door is open for this trait to be recessive (Figure 2).

Recessive Lightning to Wild heterozygous Mate = 50% Lightnings.

Figure 2. Recessive Lightning to Wild heterozygous Mate = 50% Lightnings.

Now, there is an upside if this trait is recessive; it means we got lucky. Primarily, it means I got lucky on the selection of the non-Lightning mate, because there would be no way of knowing it carried a single-dose, non-expressed Lightning gene. It would mean that the game plan of using a mate from the same island paid off. If you find a wild albino fish, you are most likely to find more albinos in the same geographic region because they’d probably be siblings. Not to mention that many of the non-albino siblings in the area could potentially carry a single albino gene as well.

The other way we will be lucky is that IF Lightning is recessive, and if the initial percentage is in fact roughly 50%, it would mean that all the siblings would then have to carry a single, non-expressed Lightning allele (because their only option from the Lightning parent is to receive a Lightning allele). This would mean that every fish in the group if mated would produce 25%, 50%, or 100% Lightning Maroons. To put it in a commercial context; if we definitively knew that this was a recessive trait, then even the normally striped offspring would be tremendously valuable to breeders, because simply mating two of those together yeilds 25% Lightning.  In an interesting twist, it seems most people expected the Lighting trait to be recessive if genetic, and assumed that we would get the results shown in Figure 1, and only in the 2nd generation would we get more Lightnings, as shown below (Figure 3).

Recessive Heterozygous F1 Offspring, Mated together, produce 25% Lightnings.

Figure 3.Hypothetically recessive heterozygous F1 Offspring, mated together, produce 25% Lightnings.

Still, I’d love to hope that this trait is recessive because it means all the siblings would then carry a hidden Lightning allele. In looking at the number of wild Lightning Maroons presumably observed (and thus caught), we know of only 2. This rarity could suggest a recessive trait, as two wild fish with hidden Lighting Genes, mating together, would produce 25% Lightnings. Given that a clownfish pair’s minimal reproductive goal is to produce two replacements, you can quickly see how a single pair of clowns, constantly churning out babies that are 25% “Lightnings”, might only yield a handful at best (remember, marine fish have been shown to suffer massive mortality in the earliest hours and days of their lives – most never even make it to settlement, and most of those, not past their first year).  Lightning Maroon babies truly stand out in the rearing tank while their normally patterned siblings are difficult to see; you can’t help but assume Lightning offspring be much easier for predators to locate. So the rare Lightning making it in the wild would fit well with a recessive trait hypothesis.

But what are the odds that I got “lucky” with the mate I selected? Impossible to say, but Occam ’s Razor suggests that the following scenarios could be more likely.

The Dominant Scenarios for “Lightning”

Let’s deal with straight up dominance. If this is a dominant trait, then you only need one “dose of the gene” to express the trait. To simplify, breeders tend to view dominant traits as being pretty uniform in their expression, and there’s no difference whether you have one dose or two. In other words, a Lightning Maroon Clownfish could either be (L/L) or (L/+) and would look the same. A good example of this, to borrow from the Angelfish community, is a trait called “Zebra”, which adds extra bars and patterning in the fins. There’s no visible difference between a homozygous Zebra (Z/Z) or a heterozygous Zebra (Z/+).

A Zebra Angelfish

A young Zebra Angelfish, straight up dominance means this fish could be (Z/+) or (Z/Z) - the only way to know is through planned and controlled matings and observing the results.

Let’s again weigh the options. If Lightning is dominant, then the non-lightning mate can only be (+/+). Why? Because any fish that is (L/+) is going to be Lightning. So in this scenario, the normal mate can only be (+/+). That leaves the Lightning Maroon to be either (L/+), or (L/L). Now, here’s where it gets interesting. If the Lightning Maroon was (L/L), we would have 100% Lightning Maroons in the offspring, because every fish could only get a ( L ) allele from the Lighting Maroon, and all (L/+) offspring would then be Lightning (Figure 4). Since we don’t have 100% Lightnings in the offspring, we can rule out the Lightning Maroon being (L/L) if this is a dominant trait.

Homozygous Lightning Father X Homozygous Wild-type Mother = 100% Lightning offspring

Figure 4. Dominant Homozygous Lightning Mother X Homozygous Wild-type Father = 100% Lightning offspring

That would leave (L/+) as our only genetic option for the Lightning Maroon, which would thus result in a roughly 50% expression rate in the F1 generation. The inheritance of the ( L ) allele from the Lightning parent is a just a coin toss, 50% of the time they get a +, and 50% a L.   Once again, the rules of genetic expression and inheritance suggest that this is a possible genetic explanation given the initial results we’re seeing (Figure 5).

Dominant Heterozygous Lightning Father X Homozygous Wild-type Mother = 50% Lightning offspring

Figure 5. Dominant Heterozygous Lightning Mother X Homozygous Wild-type Father = 50% Lightning offspring

Now, my problem with this trait being dominant starts immediately from the fact that it requires at least one outwardly visible Lightning Maroon Clownfish to be breeding in the wild in the first place (unless there is a wild-type pair that is predisposed to throwing off the odd “Lightning” mutation once in a blue moon – afterall, these traits can appear spontaneously). If this trait is dominant, then it might also suggest that this mutation ought not to be as rare as we currently are led to believe it is. And to make matters worse, it does seem that we haven’t seen much straight up “dominant” variations in ANY of our designer clownfish to date; it seems all are either recessive or the result various doses of partially dominant traits. And surprise again; looking back at the Angelfish (which happen to Cichlids, which are a closely related family to the Damselfish, and thus to the Clownfish), we see this: 0nly 1 truly straight-up dominant trait. Meanwhile, there are currently 5 known recessive traits, and 7 traits that are either partially dominant or codominant. Dominant traits just don’t seem that common in clownfishes.

What if “Lightning” represents Partial Dominance?

So what if this is a partially dominant (aka. incomplete dominant) or co-dominant trait. The difference is nuanced, but in the angelfish world co-dominance can cause “blending” of traits in certain mixes, dominant expression in other mixes, whereas partially dominant traits present more of an A/B/C result. To draw a parallel, some might say that if the Lightning trait were codominant, then a fish with a single Lightning allele should still show the white stripes “underneath” the lacy pattern of the Lightning.  I’ll dispense with codominance for the time being and just refer to this option as the partial dominance possibility.

Partial (incomplete) dominance is perhaps the most plausible and most exciting of the three options.  As the scenarios are about to play out, they suggest that the Lightning, in a partial dominance scenario, would only be the “first step”.  Partial dominance is well documented in angelfish, and the stripeless allele is a great example.  A normally striped angelfish is Silver (+/+), a single dose is a Ghost (S/+), and a double dose is a Blushing (S/S).  Take a look at a Ghost and compare it to a Blushing that happens to be showing a second partially dominant trait, the “veil” fin trait (impossible to say at this young size whether our example fish is simply veil (V/+) or super veil (V/V)).

A single dose stripeless angelfish, (S/+), aka. a "Ghost".  You can see a "Silver" (wild type, standard barred) Angelfish in the backround at right for comparision.

A single dose stripeless angelfish, (S/+), aka. a "Ghost". You can see a "Silver" (wild type, standard barred, aka. (+/+)) Angelfish in the background at right for comparison.

Blushing Angelfish

A Blushing Angelfish with two doses of the "Stripeless" allele, (S/S).

If “Lightning” is a partially dominant trait, the results in the offspring push us to only one genetic possibility. Let me step back to explain why. There are currently only 2 forms of observable pattern in the offspring; “Lightnings” and “normal”. Simply put, the Lightning cannot be (L/L) in the partial dominance scenario. If a partially dominant allele is present in a homozygous state (L/L) and mated to a wild type fish (+/+), we should get all (L/+) – something intermediate between the Lightning and the Wild form, and they should all be the same (emphasis again on the fact that there would be no Lightnings, and no normally barred fish either) (Figure 6). We don’t have that result, so (L/L) is ruled out if Lightning is a partially dominant trait.  Or is it?

Partially Dominant Homozygous Lightning X Homozygous Wild Type = 100% Intermediate Offspring

Figure 6. Partially Dominant Homozygous Lightning X Homozygous Wild Type = 100% Hypothetical Intermediate Offspring

The second consideration for parental genetics would be (L/L) x (L/+), but once again here, the (L/+) cannot look like the wild form, as (L/+) represents an “intermediate form”. Someone out there is going to say “but what if (L/+) does in fact look like the wild form?  If it did, then by definition Lightning would be a recessive trait as I described earlier (Figure 2)!  So this scenario is ruled out.

The third consideration would be (L/+) x (L/+), but then again that would mean both mates should be “intermediary” forms and roughly look the same (which they obviously don’t in our pairing). This alone is enough to scrap this mating as a possibility.   But if you’re not convinced, this hypothetical mating would also mean that 25% of the offspring would be (L/L), 50% (L/+), and 25% (+/+) – if the fact that the parents would have to look the same didn’t throw this out for you, consider that there would still have to be THREE (3) phenotypes in this batch of offspring for that proposed genetic combination in the parents to make any sense (which it can’t, because the parents are not the same).

The only way that “Lightning” works as a partially dominant trait is if the Lightning Maroon is (L/+), and the mate is (+/+). This produces a nice occurrence of 50% like the Lightning, 50% like the male parent (Figure 7). This also takes a lot of the “luck” out of the equation; we didn’t have to stumble upon a mate with a hidden allele like we would have in the recessive scenario.

Partially Dominant Heterozygous Lightning Maroon X Wild Type Male = 50% Lightning Maroons

Figure 7. Partially Dominant Heterozygous Lightning Maroon X Homozygous Wild Type Male = 50% Lightning Maroons

This also seems to be how some currently known traits may work (Picasso in Percs, maybe Snowflake in Ocellaris). If you believe that “Black Ocellaris” are a melanistic variation within Ocellaris, then “black” in ocellaris could also potentially be partial dominance…with “Blacks” having “two doses”, and when you mix Black with Ocellaris, you get “Mochas” which in all photos I’ve seen, are muddy intermediaries. The real question to be asked is what happens when you breed 2 Mochas together – do you get 25% Blacks, 50% Mochas, and 25% normal Orange Ocellaris? I don’t know that anyone has done that and tested the results yet (but I also know that I don’t believe they are the same species of fish at the moment either…you have to throw the genetics out the window when you start hybridizing)

But getting back to the Lightning; if this trait is “partially dominant”, then the most exciting part is yet to come, because it would mean that all the fish we’ve seen so far only have ONE dose of the Lightning allele (L/+), and thus, the designer breeders out there will be clamoring to mate two Lightnings together so they can discover what a (L/L) fish is going to look like.

And that’s the crazy part, because there should only be one of two things that could when we mate Lightnings together – either we’ll get 25% being something new, or we’ll just get more Lightnings. If we get 100% Lightnings, we are either looking at a recessive trait or a straight up dominant trait (or, in a less likely case the difference between a partially dominant (L/+) and (L/L) is simply too minimal to discern, and you’d then just treat it as dominant anyways).

The “Lightning Precursor” Hypothesis – Dealing with Horned and Flaked Maroons

I suppose at this point we have to step back and objectively define what we *think* a Lightning Maroon “is”.  What is the phenotype?  We have to consider the two fish that have been given that label to-date (the less familiar one being the first wild-collected Lightning Maroon from 2008).  Well, the best term I’ve heard used lately was to describe the Lightning Maroons as filigreed. Other’s commonly call the patterning “lacey” or “net-like”.  Whatever it is, the most notable place for this Lightning variation is in the headstripe.  The headstripe is dramatically wider in the Lightnings, and it is “pitted” with normal coloration.  “Horned” and “Flaked” maroons fundamentally lack this very distinct patterning and the wider headstripe it takes to make it.

The other part of the Lightning phenotype is the breaking up of the mid-stripe and tail-stripe into the lacey, interconnecting patterns that split apart and at times, reconnect.  None of the “horned” maroons show this patterning that I’ve seen, while many “horned” maroons simply exhibit broken bars or “extensions” trailing off.  Admittedly, only the most recent “Lighting Precursor” was really suggestive of the body stripping seen in the two wild Lightning Maroons, but the stripes showed a more “smooth” outline and did not reconnect (I’ve been told the other side of this fish was unimpressive) – I think this fish is better considered a more extreme form of these “Horned” Maroons being found in PNG waters.

In drafting this genetics rundown, I realized I had one other genetic possibility on the table; the notion of the “Horned” Maroons being collected in PNG potentially represented the “intermediate” form in a partial dominance scenario (eg. the hypothetical heterozygous offspring shown in Figure 6).  One such Maroon recently made the rounds in the internet being called by some a “Lightning Precursor”.

After examining the data provided publicly by EcoAquarims PNG, it seems these aberrant Maroon clowns  appear to be quite common in the waters of PNG, with various atypical Maroon clownfish being caught approximately every 11 days.  We also had other fish like Mike Hoang’s Goldflake Maroons which indeed, as young fish, had me wondering if we’d see Lightning-like traits as they grew up (sadly the best marked offspring were lost, and those that remain look no different than the “Goldflake” Maroons output by Sustainable Aquatics).

Let’s deal with the “Goldflakes” of the world first.  It turns out that abnormally spotted and overbarred Maroon Clownfish (what I’m calling “Flaked” here) are indeed commonplace in captive culture.  So far, these fish have seemed to elude genetic categorization, apparently really behaving fundamentally more like “misbarring” in other species of clownfish.  Most recently German breeder Sylvio Heydenreich shared some videos depicting some highly overbarred Maroons on the MBI website; when asked about these fish, he stated quite directly that, “Die Fehlzeichnungen lassen sich ganz leicht über die Wasser Qualität steuern.”  Or as Google likes to translate it, “The failure drawings can be controlled easily through the water quality.”  Failure drawings of course, being what is probably a literal translation for “misbarring”.  And to that end, we already are aware that misbarring in clownfish has environmental causes, not genetic causes. So as much as we like these “Goldflakes”, all observations to date suggest we think of this type of patterning as a likely non-genetic occurance.

Meanwhile, those “Horned” Maroons coming out of PNG had all of us, even me, convinced that the Lightning Maroon could be a homozygous (double-dose) example for a partially dominant trait. Simply put, the breeding results don’t really suggest this possibility because we lack the intermediaries (I would’ve expected the 100% “horned” batch to show up, like Figure 6).  Still, I do have two normally barred fish that show spots.

Note the extra spot on the back of this normally barred juvenile.

Note the extra spot on the back of this normally barred juvenile.

Is this baby a “Horned” Maroon?  Well, here’s the kicker.  There are only two ways you get hypothetical intermediates (intermediates being the proposed placement of the “horned” Maroons).  You either get 100% in the F1 batch, or the male parent has to be an intermediate itself, in this case, a “Horned” Maroon. And this is where there’s still an outside chance – the male has a single broken tailbar.  But…if this was in fact an “intermediate”, what genetics must we get in the offspring?  75% Lightnings, and 25% intermediates – NO wild-types.  Again, let me be explicitly clear – for “Horned” Maroons to be “Lightning intermediates” or “Lightningprecursors”, I would have had to encounter “Horned” Maroons  in the offspring and at a rate of 25% -or- 100%.  So…if the babies all wind up showing extra horned bars and spots as they grow up over the next few months, and the ratio of Lightnings to non-lightnings is 3:1, there could still be “hope”.  Otherwise, we have probably closed the book on the “Lightning Precursor” hypothesis that tried to link the Horned Maroons to the Lightning, at least for now.

All of that said, what I really think we’re seeing here is something much more fundamental in the Horned Maroons.  We are seeing this “flake” overbarring, a commonplace occurring in captive-bred maroons, showing up on a few random offspring.  You wouldn’t notice it in the Lightning offspring because it’s just “painted over”, but you can see it in the normally barred fish.  Years ago, breeders would have destroyed these types of fish as “culls”…that’s when the 3-bar wild-type fish was considered something to aspire to as a breeder, and not “common” and “boring” as many hobbyists may consider a wild-type clownfish today.  Given that we know of a possible causal relationship between “overbarring” and “environment”, perhaps there is something environmentally going on in the waters of PNG to show us more “environmentally overbarred”, aka. “Horned” Maroons, than perhaps we might expect in other parts of the ocean.

Or, and this is still a possibility; the “Horned” Maroons of PNG could yet represent another, distinct genetic variation.  It’s certainly possible – breeding them could give us the answers, although it may be difficult in the face of commonly-occurring “flaked overbarring” potentially giving you a fish with the same basic phenotype.

The Odds on the Lightning Pair’s Genetics

Let’s get back to the Lightnings.  If we give equal weight to all three possibilities for the interaction of the “Lightning” allele, we are left with three scenarios for the genetics of the parents.  Once again, notation here…(female first) X (male second).

Recessive, where we have (l/l) X (l/+)
Dominant, where it can only be (L/+) X (+/+)
Partial Dominance, where it must be (L/+) X (+/+)

By this alone, each has a 1/3 chance of being right.  There is a 2/3 chance, or 67% roughly, that the Lightning is (L/+).

However, for the sake of doing something interesting, what if I used the genetic ‘spread’ in Angelfish to derive an alternate baseline for the odds of a trait being dominant, recessive, or partial/codominant within the clownfish family?

Recessive = 5/13, or roughly 38%
Dominant = 1/13, or roughly  8%
Partial/Codominant = 7/13, or roughly  54%

If this was at all representative of the odds for trait expression in clownfish (and it’s really probably not, it’s just a fun way to think about it), then we have a 62% chance that the Lightning Maroon is (L/+), and within that 62%,  it would then represent a 87% chance  that the trait would be partially dominant (again, roughly 54% overall).

Overall, whether we weight the system or not, the odds remain in the rough territory of 2:1 that the Lightning Maroon is (L/+), and the mate I used is (+/+), vs the only possible alternatives of (l/l) and (l/+).  The kicker for me is when you move beyond “probability” alone, and put in the observations and the way mother nature seems to work.  I’ll get to my prognostication in a minute, but first, I must point out that this puzzle can be solved.

How are breeders going to help figure it out?

In a nutshell, this project will soon turn to the massive “cloud computing”, or in this case “crowd breeding” effort of marine aquarists who get these offspring.  It has always been my intention to get the F1 fish out to other breeders to both diversify the risk, but also to leverage the collective efforts of breeders to provide for rapid, definitive answers.  In a nutshell, anyone breeding with my offspring, you have my formal request to track your project at the MBI, and to do so openly.  You also have my request that you must track your offspring numbers and take photos of each one on both sides, because it is the headcounts and photos that will help determine the genetics in the end.  Here’s how we’ll do it (again, assuming that “Lightning” is the result of a single locus and a single allele).

We can determine (or rule out) a recessive trait by mating the non-lightning siblings together; if recessive, 2/3 of the F1 babies will carry a hidden Lightning gene. This means that picking any random 2 fish, the odds are roughly 40% that both are (L/+), so four out of 10 random pairings would yield Lightning offspring to the tune of 25%, if this is a recessive trait. The only way you get Lightnings out of pairing 2 normally-barred siblings is if this trait is recessive.

We can also determine this trait to be recessive by matings of Lightning Maroons to their non-lightning siblings. In this scenario, 2/3 of the pairings would produce 50% Lightning offspring, while the remaining 1/3 would produce nothing but normally striped fish.

We don’t need to use the siblings to specifically test for a recessive trait, but non-sibling fish present a conundrum – you have less insights onto what their genetics could possibly be.  Still, you can simply mate Lightings to unrelated white stripe maroons (and breeders out there, I will work as hard as I can to produce offspring from the other PNG White Stripe pair in the house so we have a clean PNG bloodline which we can outcross to, and Dale Prichard in the UK hopefully can contribute more, or you can look to the other PNG maroons being exported from EcoAquariums PNG now).  If the trait is recessive, then you have to consider the unknown odds that any randomly-selected, unrelated fish, could be carrying a single hidden copy of the recessive allele.

However, if the trait is partially dominant, any Lightning paired with a wild-type sibling, or any outcross (mating of a Lightning to unrelated normal fish) should yield a percentage (50%) of Lightnings in the offspring. Conversely, again, if the trait is recessive, these outcrossed matings will produce nothing but normally barred fish UNLESS, once again, you get “lucky” to stumble upon a fish with a hidden allele. But that’s the rub – you’re far more likely to find that hidden gene in the normally barred siblings.

If we get something “new” out of the Lightning X Lightning mating, it should be 25% of the “new” variety, and that would convincingly clinch the genetics as partial dominance. Sounds far-fetched? Well, in Percula, Picasso X Picasso is where we get Platinums from.  If mating Lightning X Lightning simply makes 100% Lightnings, then the trait easily falls into the category of a straight up dominant trait.

I’m a betting man if the wager is bragging rights…so my guess is…

…partial dominance.  Ultimately, my gut call is for partial dominance because it seems to be the most commonplace type of genetic trait we’ve seen in our designer clownfish, and it’s the most prevalent in a widely cultivated and well-documented group of related fish (the freshwater Angelfish). The odds also do slightly favor partial dominance.  Partial dominance may also be one of the easiest to prove – just mate two Lightnings together and see what you get. Partial dominance (and in this case, straight dominance) also requires less luck to have had the outcome I seem to have had with my initial pairing. If ever there was a project that had just about everything except “luck” on its side, it is The Lightning Project.

One last wonderful caveat – every possibility laid out above could wind up being 100% wrong.  Until we get those second generation fish produced, and aquarist start gathering the data and sharing it, we simply won’t know.

A special thanks to Adeljean Ho for acting as a sounding board and editor on this piece.  I am sure Adeljean, with his strong interest in genetics, was probably as excited about this as I am!  Thanks!

"Oh snap, was that thunder??" - image courtesy EcoAquariums PNG, Ltd.
“Oh snap, was that thunder??” – image courtesy EcoAquariums PNG, Ltd.

EcoAquariums PNG, Ltd, the successor of the spot formerly filled by SEASMART, has continued to turn up abberantly patterned Maroon Clownfish collected in the waters of Papua New Guinea.  Between SEASMART’s own collections, and with no less than 3 unique Maroons shown off on the EcoAquariums PNG Facebook page this year, we are looking at the very least, well over a half-dozen PNG-sourced Maroon clowns that are highly “atypical”.  SEASMART referred to many fish like these as “Horned” Maroons, owing to the common barring pattern of “prongs” leading off the headstripes in either direction.  It’s a unique fish, but digging deeper, there’s even more information behind these unique maroon clownfish.

With EcoAquarium’s label tracking system, we actually are given a window into pretty much every fish that the company collects, and that’s where things get interesting.  You see, EcoAquariums records every fish collected along with a slew of other data, and makes all this information publicly available in easy to use and search PDF files!  You can download them here:

Breakdown of Maroon Clownfish Captured per grouping of 1000

0001 to 1000 = 30 maroons
1001 to 2000 = 25 maroons
2001 to 3000 = 25 maroons
3001 to 4000 = 24 maroons
4001 to 5000 = 18 maroons
5001 to 6000 = 2 maroons
6001 to 7000 = 12 maroons
7001 to 8000 = 10 maroons
8001 to 9000 = 16 maroons
9001 to 9723 = 3 maroons

Total Maroon Clownfish harvested to date = 165 = roughly 1.7% of total exports

Breakdown of Percula Clownfish (Amphiprion percula) captured, per grouping of 1000

0001 to 1000 = 359 percs
1001 to 2000 = 388 percs
2001 to 3000 = 412 percs
3001 to 4000 = 316 percs
4001 to 5000 = 200 percs
5001 to 6000 = 286 percs
6001 to 7000 = 371 percs
7001 to 8000 = 370 percs
8001 to 9000 = 275 percs
9001 to 9723 = 497 percs

Total Percula Clownfish harvested by EcoAquariums PNG to date = 3474 = roughly 35.7% of total exports.

I share the Percula figures because a) they surprised me and b) it kind of speaks towards the general overall demand for Amphiprion percula, vs. Premnas biaculeatus, in the trade.  The other interesting part – the data as currently provided by EcoAquariums PNG Ltd. is an unprecedented look at what is presumed to be the entire marine aquarium life trade in Papua New Guinae, and could someday form the basis for a lot of interesting research by academics.  It’s an amazing data set, assuming the accuracy is there (which, in theory, it should be).  The transparency provided gives us an unparalleled opportunity to question our supplier, and at the same time, investigate some really interesting questions on our own.

Here’s a rundown of ALL the special / abberant maroons recorded to date:

0164 Benard Ora maroon clown, unique Premas biaculeatus lg 22-Nov-11 S 9.5046, E 147.0954 FD, HN, BN
0182 Aila Kila maroon clown, unique Premas biaculeatus md 22-Nov-11 S 9.5046, E 147.0954 FD, HN, BN
0470 Olema Kila Maroon, unique Premas biaculeatus lg 8-Dec-11 S 9.4988, E 147.0062 FD, HN, BN
0680 Nou Karawa Maroon, unique * Premas biaculeatus md 9-Dec-11 S 9.4988, E 147.0062 FD, HN, BN
0681 Nou Karawa Maroon, unique * Premas biaculeatus md 9-Dec-11 S 9.4988, E 147.0062 FD, HN, BN
1632 Geno Au maroon clown, unique Premas biaculeatus sm 15-Feb-12 S 9.5384 ,E 147.1021 FD, HN, BN
2221 Gia Laka Maroon clown, Highly unique Premas biaculeatus lg 25-Feb-12 S 9.4900, E 147.0348 FD, HN, BN
2262 Geno Au Maroon clown, Highly unique Premas biaculeatus lg 25-Feb-12 S 9.5046, E 147.0954 FD, HN, BN
2979 Kunini Sam maroon clown, unique Premas biaculeatus md 2-Mar-12 S 9.4900, E 147.0348 FD, HN, BN
3199 Ralai Kila Maroon Clownfish Premas biaculeatus Unique, sm 19 April 2012 S 9.5046, E 147.0954 FD, HN, BN
4109 Voi Karawa Clown Maroon, spots Premas biaculeatus Md S 9.4900, E 147.0348 29 April 2012 FD, HN, BN
4594 Kunini Sam Clown fish Maroon Premas biaculeatus Highly uniqueS 9.5384 ,E 147.1021 2 May 2012 FD, HN, BN
4611 Nou Karawa Clown fish Maroon, horned Premas biaculeatus Lg S 9.4988, E 147.0062 2 May 2012 FD, HN, BN
4612 Nou Karawa Clown fish Maroon, horned Premas biaculeatus Sm S 9.4988, E 147.0062 2 May 2012 FD, HN, BN
4841 Ralai Kila Clown Maroon spotted Premas biaculeatus Lg S 9.5046, E 147.0954 4 May 2012 FD, HN, BN
5108 Kala Kila Clown Maroon, one horn Premas biaculeatus Md S 9.4900, E 147.0348 11 May 2012 FD, HN, BN
6215 Samuel Kila Clown Maroon, thick horn Premas biaculeatus Lg S 9.4988, E 147.0062 25 May 2012 FD, HN, BN
7063 Voi Karawa Clown Maroon unique Premas biaculeatus Lg S 9.4900, E 147.0348 5 June 2012 FD, HN, BN
7247 Gia Laka Clown Maroon spot,horn Premas biaculeatus Lg S 9.4900, E 147.0348 5 June 2012 FD, HN, BN
8131 Pauline Paul Clown Maroon,horned Premas biaculeatus Md FI zone A 20 June 2012 FD, HN, BN
8393 Pauline Paul Clown Maroon, misbar Premas biaculeatus Lg FI zone A 27 June 2012 FD, HN, BN
9004 Olema Kila Clown Maroon unique Premas biaculeatus Sm FI zone A 1 July 2012 FD, HN, BN

Currently, the data runs from November 22nd ,2011, through July 5th,2012, and covers 9,723 fish and inverts.  Out of those fish, there were 165 Maroon Clowns collected.  Out of those Maroon Clowns clowns, 22 were flagged as ‘unique’ in some fashion, with two three of those twenty-two being further classified as “highly” unique.  There are so many interesting ways to look at this – we cover 226 days in this sampling, which means at current catch efforts, 7 out of every 10 days, a maroon clownfish is caught.   Slightly over 14% of the collected maroons are classified as unique in some capacity, and  odds are, roughly every 11  days, a “unique” maroon is collected by the folks diving in PNG (just under 3 per month).

Obviously, we cannot extrapolate this to necessarily say that 14% of the maroons found in PNG waters are “abnormal”…without a doubt there is possibly, if not probably, a mandate and emphasis placed on unusual maroons, that is to say “even if we don’t need maroons right now, if you see something atypical, you should collect it”.

When I originally drafted this article, I had a lot of “genetics” on my mind.  While I think we will have better answers, here’s where my thinking was last month.

Clearly, deviations from the normal striping seem to be prevalent in PNG waters where EcoAquariums operates.  And looking back at all this, and how we’ve come to learn that Picasso Percs are not necessarily as exceedingly rare as we may have initially thought, this does all start to make you wonder – in these aberrant wild maroon clowns, are we seeing a low level occurance of the equivalent of “picasso” type forms in a wild population in PNG?  Could it in fact be that these “close but no cigar”, highly unique maroons, may in fact be the picasso equivalent or as one blogger put it, a “Lightning Precursor“?  And, if the genetics of Lightning were to work like we think the genetics work in Picassos , could it be that the two fish we’ve called “Lightning” to date, may in fact be the equivalent of the Plantinum Perc?

And here’s the kicker…the fish above does show traits that certainly speak to it being “lightning-esque”.  But when I look at the two fish we’ve called “Lightning” to date, here’s what I see – more of a netting effect, particularly in the headstripe but also in the midstripe and tail stripe.  Let’s ignore my Lightning Maroon (#2) and go back and look at #1 - - and here’s where my thinking goes.   If you double up, and mate these “close but no cigar” unique or aberrant Maroons, do you get a redoubling of the gene that causes the stripe abnormality, taking the phenotype from stray prongs, spots, and splits, and amplifying it into the “Lightning” form we all know and love?

It may sound insane, but there are definitely examples of this genetic story  in other fish, including clownfish such as Picasso Perculas which appears to be a “single dose” of a dominant gene, and Platinums being a “double dose” of that same gene.  It may or may not be that way, but further offspring counts should nail it down, and some breeders may already know the answer and just aren’t sharing / thinking it’s worthwhile to mention.  Of course, as I recently learned, there is not shortage of genetic understanding in other fish where different genetic loci and the alleles at those loci are known to drive a plethora of diverse phenotypes – amazing levels of information exist for freshwater Angelfish (Pterophyllum scalare) via the Angelfish Society’s genetic documentation and others (such as a release about the Philippine Blue locus made by a collaboration of parties independent of the Angelfish Society).  We can only hope that those breeders working on designer breeding start paying greater attention, and can realize the value presented here by sharing information.  The level of breeding CONTROL one is afforded has elevated the discourse and pursuit of Angelfish breeding in my opinion.

Turning back to the Lightning Maroon genetic mystery, my original hypothesis about the wild-caught “horned” type Maroons from PNG was all speculation at that point, and when the ideas came to me and I first wrote them down, I had yet to see any baby Maroons from our Lighting breeding efforts.  Despite that data deficiency, we’re certainly seeing a continuum of stripe aberrations in these fish that were perhaps suggestive of a genetic basis (given the geographic restriction and frequency of occurrence).

Granted, now that we have babies, the story is about to get a heck of a lot more complex….and yet, possibly much clearer.


So I think many of the regular readers of the Lightning Project know I have an outspoken stance on breeding. Specifically, that preservation of natural biodiversity trumps the creation of captive-originated ornamental forms and the intentional and unintentional hybridization of wild forms. Not going to get on that soapbox today (I’m busy doing that on the MBI), but in doing my “homework” I came across an interesting publication. More on that in a moment.

The concept I was trying to find some documentation on relates to what we’d call “foundation stock”. Foundation stock being the individuals used to create a new population of a species. In the case of species survival programs, foundation stock would be the wild collected specimens used in zoo breeding programs to maintain the captive population with the goal of arking the species and hopefully one day reestablishing wild populations. In the case of aquarium breeding, it’s no different..i.e. the 4 breeding pairs of Mccullochi Clownfish (Amphiprion mccullochi) owned by Ryan Dwyer are the 8 specimens that will be the ancestors of all future Mcc’s (unless the unlikely happens and more wild fish are brought into captivity).

There’s another way of looking at this question, a darker viewpoint. Functional extinction. This is a term I came across first when reviewing IUCN Red Lists (Banggai Cardinalfish anyone…still on that list as endangered). What “Functionally Extinct” means is that while the species still exists, the population is so small that from a breeding and genetics standpoint, it fails to have sufficient genetic diversity to prevent a total genetic meltdown (and ultimately, extinction). For example, you probably could not bring back a species of bird if there were only a single pair left – it would be termed functionally extinct. In such dire circumstances, actual intentional hybridization might be thrown out as a last-resort mechanism in the hopes that you could one day breed back to some approximation of the ancestral species, although in truth, you’ll never get back to the actual species.

At any rate, the document published by FAO (The Food and Agriculture Organization of the United Nations) is part of the Fisheries and Aquaculture Department. Conservation of the genetic resources of fish: problems and recommendations provides general guidelines on minimum population sizes for short and long term captive maintenance, and actually explains, quite clearly, some very interesting talking points on inbreeding and genetic integrity.  While this is all generalizations, here’s some interesting talking points:

  • A minimal captive population size of 50, in equal sex ratios, randomly mating, will keep inbreeding at roughly 1%.   At this population size, the total population will lose 1/4 of it’s genetic variations in 20 to 30 generations.
  • An interesting aside to the first point, the number of individuals in a small population (let’s say 25) is roughly equivalent to the number of generations that population can exist while maintaining a reasonable level of “fitness” (so 25 generations if the total population is limited to 25 individuals in each generation, as well as the other stipulations of equal sex ratios and random matings).
  • It is argued that 500 individuals is the minimum necessary to maintain a long term captive population that loses variations to genetic drift at a rate compensated for by new variations gained through mutation.
  • The most interesting one I’ll quote verbatim – “the number of founders in a colony, so long as it is greater than about five individuals, is not nearly as important as the long-term maintenance size of the colony (Nei et al., 1975; Denniston, 1978). That is, a single bottleneck event followed by rapid growth to a large size, say 2Ne greater than 500, does relatively little damage, compared, that is, to a chronically small Ne.”.  Or to try to paraphrase – a captive population started with only 5 individuals can still be enough (in a fish) to potentially establish a stable captive population, so long as that population is brought up to at least 500 breeding individuals in short order.

So what does that all mean?  Well, for starters, it means that even with only 8 founding individuals, the captive population of Amphiprion mccullochi could be very stable and with us for a long time so long as the fish is actively bred by many people and it’s done in as few generations as possible.  Pile on intentional avoidance of any inbreeding in the first generation or two, and the future could be very bright.  And on the flipside, even if the population of breeding individuals was kept well below 500 (and it’s entirely possible that could happen), the generation time for that to be a problem, as measured in years, could be centuries or more (given that you can spread out the generation time to be 10-20 years in a clownfish if you plan for that).

How does that all relate to the Lightning Maroon and it’s PNG mate (and neighboring pair)?  Well, if nothing else, it means I really should get one more PNG white stripe maroon pair into the mix if I ever can, just to get above that foundation minimum of 5 ;)

So with this latest shipment of PNG Maroons, I now have the following inventory:

1 PNG Lightning Maroon Clownfish, sex indeterminant
1 PNG Morse Code Maroon Clown, sex indeterminant
4 PNG Maroon Clownfish, your typical White Stripe variety, all still at sizes where they should be male or “juvenile”.
1 “Labrador” Maroon Clownfish, a massive fish that could only be treated as a female at this point.

What’s a breeder to do?

Let’s circle back to the two most interesting fish, the “Lightning Maroon” and the “Morse Code” Maroon (I called it a “Dash Slash” or “Dash Hash” but then Jay pointed out the “dots”, so it was a “Dot Dash” and then Morse Code…you get the idea).  I do hold the assumption that there is a genetic basis to the Lightning patterning.  I’ve elaborated on that prior.  I also think that possibly, this “Morse Code” patterning, the random dot or dash, could be indicative of some genetic variation as well…not simply a “misbar”, but the starting point, the most basic example, of what might represent a continuum or range of pattern variation within the Maroon Clownfish.  In simple terms…if the Lightning is the equivalent of an “A-Grade Picasso” Percula, then this “Morse Code” Maroon might very well be the “B-Grade” version of the same genetic barring mutation.

The PNG Lightning Maroon - A-Grade by ANY standard!

Unofficially calling this the "Morse Code" Maroon. Is it just a B-Grade Lightning?

You might think I’m “reaching” here, until I ask you to consider that this is in no way the “first” “Morse Code” Maroon Clownfish I’ve seen.  No, I think I’ve actually seen at least a dozen of these fish, in multiple venues.  At least one commercial breeder has had fish looking just like this “Morse Code” Maroon and offered them for sale.  Maroons are known to throw some variation  on striping.  For example, check out the spotted Maroons from Reef Hot Spot or this other “one of a kind” Maroon Clownfish thread on Manhatten Reefs.  This type of “extra barring” and “spotting” apparently isn’t really all that rare of a thing in Maroon Clowns apparently (and corroborated by at least one author…but I can’t find that reference at the moment!)

Pile on the simple fact that this is another “extra barred” Maroon coming out of PNG, where we already know 2 “Lightning Maroons” have come from.   If this subtle, small barring variation is genetic, and is a less extreme version of what produced the “Lightning Maroons”, we could have just struck “gold”.  With ALL of that said, in the absence of having a second “Lightning” Maroon on hand, this “Morse Code” Maroon just became the most desirable mate I could utilize.

And yes, there’s a catch.  I have yet to put these two fish side by side, but I can tell you this.  The Lightning Maroon, and the “Morse Code” Maroon, are about the same size.  Both are in the “limbo” size where they could either be mature males, or easily become females.  Up until now, I’ve treated the Lightning Maroon as if it were male, in the hopes that if it IS in fact male, it could remain male.  Again, I’ve elaborated on the “why” before.

But now there’s this other ambiguous fish.  If it was notably SMALLER, it would’ve been a no-brainer….pair it with the Lightning, making the Lightning a female.  A great pair, and arguably the best I could hope for.  But CAN I do this?

It’s hard to say.  I have to fight the likihood that BOTH the Lightning Maroon and the “Morse Code” Maroon could be female.  Jake Adams already drew the conclusion that the Morse Code was female without even knowing its size…just looking at a picture.  So yeah, a fair amount of folks might already call the Morse Code female, and in most other settings, it would be paired up to be the female.

But what if one, or both fish in question, can still be male?  Well, as you know, the Lightning Maroon has been treated as male up until this point.  The XXLG “Labrador” Maroon from Frank and Mary has been prowling the main tank in the hope that its presence would keep the Lightning male (assuming it IS male).  They have not been allowed to interact…this has been a “show of force” more than anything else.

Ideally, while in Quarantine, it would be wise to attempt the same “social pressure application” on the Morse Code Maroon.  In other words, try to pair it, or at least house it, next to a much larger Maroon.  That way, if it’s male, it will remain male while in QT.  And then, when the time comes, it could be paired with the Lightning Maroon.

To accomplish such a pairing, the ideal mechanism would be to allow the Lightning Maroon to pair up with any one of the small PNG Maroons that was shipped in.  Of course, we’re looking at at-least a few weeks before the “all clear” siren is blown and that process could start.  Get the Morse Code Maroon paired up as well, but as a male.  Allow the Lightning to grow substantially in size, and then SWAP out the males, bringing the Morse Code in as the male, and the Lightning as the female.

It all works in theory.  But what if the Morse Code is already female?  Well, if that happens the best pairing scenario would be to form 2 pairs.  A Lighting + PNG pair, and a Morse Code + PNG pair.  The practical upshot is that we could attempt a “Lightning X Morse Code” (or if you prefer, an “A Grade Lightning X B-Grade Lightning”) pairing once the NEXT generation is born.  It’s a less direct route, but it’s by far the safest route to go.

But what if I’m not fully committed yet to allowing the Lightning Maroon become a female?  There is that odd chance that I could continue to apply social pressure to the Lightning Maroon, and could even move forward with highly controlled pairing to the Labrador Maroon.  I’d be foregoing the PNG geographic lineage in any offspring that develop.  In the meantime, I’d also have to let the Morse Code Maroon grow substantially, and that could take years.  And THEN try to pair the Lightning to the Morse Code, putting the Lightning at greater risk acting as the male.  As you can see, I’m clearly leaning towards letting the Lightning go female, but I’m probably going to by myself some time and try a few different things.

My ideal plan is to construct a far more permeable yet secure enclosure for the Lightning to allow better interaction with the Labrador Clown.  I’ve heard annecdotes of clownfish being able to fertilize spawns through physical barriers.  I figure, WHY NOT at least let there be a chance of some reproduction over the next month?

But if I do stay on the path with the Labrador, what do I do to push the Morse Code Maroon in the direction of being male?  Well, quite simply, I’ve found myself in need of yet another ridiculously large Maroon Clownfish!  I’d pretty much want to do the exact same thing to the Morse Code that I’m doing to the Lightning.  Being smaller than the Lightning, there is still a better chance that the Morse Code could remain male.

What if none of this works?  Well, there’s a reason I had a standing order for 4 PNG Juvies from Blue Zoo.  Quite simply, I could find myself with SEVERAL female Maroons, and I’d like to have mates for them all.  If I’m honest, the most likely and safest scenario is to just mate the Lightning to a PNG Juvie, do the same with the Morse Code.  If there’s any disasters, I have 2 backup males.  If everything works flawlessly, I could pair both “spare” Maroon clowns with little Juvies from PNG, have the Lightning X Morse pair, and 2 extras, just in case.  But most likely, I might end up with 4 pairs of Maroons in the long run, at which point I’d probably part with the BIG Maroons so I could focus on 2 PNG pairs.

Yes, that’s a lot of thinking, and I’m a Maroon shy of my plan at the moment.  Quarantine buys me time to think this through a bit more, and we’re not out of the woods yet.  More on the “Morse Code’s” issues in my next installment.

Social Widgets powered by