PROMOTE THE HOBBY THROUGH OPEN MINDED EXCHANGE OF KNOWLEDGE AND IDEAS - Share your experiences as a breeder or novice both good and bad. Pass on your experiences and share results with the next generation. A successful breeder will be remembered for such efforts...

Monday, November 26, 2012

A Discussion of the Pink Moscow (Panda) Phenotype …creation, modification and when needed reconstruction

© Alan S. Bias
Permission granted for nonprofit reproduction or duplication of photos and text with proper credit for learning purposes only.

November 26, 2012

Panda Male (XYMw PkPk

Introduction:  As I don’t breed purely by scientific method, nor solely for show and consider myself a stock breeder first and foremost, I’ve been searching for a term or phase to encompass this amalgamation of three distinct schools of thought.  One that describes what guppy breeding means to me and my approach towards it.  Recently, long time Betta breeder and judge, now fellow Swordtail  Guppy breeder, Prof. Leo Buss offered up the following in correspondence:  “Pursuing aesthetic goals in a scientifically informed fashion.”  As this is much more eloquently stated than my abilities allow for, that’s what it shall be.

It is generally accepted the basic Pink Moscow (Pk + Mw) or Panda phenotype arose in several locations to include Europe and Asia.  The late Japanese breeder Yoshiki Tsutsui is generally credited for the first documentation and stabilization into a recognizable strain, which he named “Panda.”  The following is not intended to be all encompassing, and will focus on color / pattern variants and not those involving finnage:  long fin variants.  It is based for the most part on personal experience and observations of other breeder’s results.  To a lesser degree incorporates prior publications, of which I found to be minimal.  My breeding experience and fascination with Panda’s goes back more than a decade.  In the world of delta guppies this hardy, fertile and long-lived phenotype often gets little more than a second glance.  Yet, for these same reasons it maintains a dedicated international following with guppy breeders who also enjoy active strains…

Breedings:  The first Panda strain I bred in MT was rumored to have been descended from Tsutsui stocks.  It had potential for very long dorsal extension, at least in the world of Panda’s, and periodically produced homozygous blau pink individuals.  In this mutation the dorsal was always clear.  The caudal expressed as clear narrow flag tails with bits of red and translucent blue bodies devoid of other color pigment.  For this reason I assumed it to be what is now commonly referred to as European Blau (Eb), which in my mind put some doubt on the strains origins.  This also suggests the basic Panda phenotype, at least in this strain, was epistatic to red.  Barring was only present in a portion of the males and always partially masked.  I did not attempt to verify if autosomal Z-bar (Ze) or sex-linked Tigrinus (Ti).  One of the few outcrosses I did with this strain involved a Purple Moscow Delta * Panda breeding.  The F1 were sib bred to produce a nice line of Panda with Purple Body Mutation (Pb) with a noticeable reduction of coloration in pectoral fins.

Two years ago I intended to obtain Panda’s and again use the homozygous pink (Pk) females joined to select Purple Moscow (Mw) males as foundation for a new strain.  By doing so would produce a Panda strain for both blue and purple peduncles.  But this would have to wait.  After obtaining, my current Panda strain initially demonstrated poor viability in females which necessitated an outcross.  Subsequent outcross with Panda males as sires seemed to confirm weakness in females.  With no available classes for North American show entry, interest and availability of Panda’s in North America has seriously decreased over the last decade. 

Strain Reconstruction
Generation1:  For initial outcross I set up a breeding group consisting of a grey Panda male with three existing sons to be bred with several Golden (gg), IFGA Bronze Delta, females obtained from breeder Rick Grigsby.  These females were utilized only for initial outcross to produce 1st generation F1 and discarded shortly thereafter.  All heterozygous Pk males in the F1 were discarded.  I’ll note that all F1 males were traditional dark shouldered Moscow Delta phenotype with a weak snakeskin pattern in the peduncle and dark variegated dorsal / caudal.

R. Grigsby Bronze Delta Females

Generation2:  Resulted from a BC1 backcross, again involving all four original P sires and a select group of 6 F1 females.  In this manner I could reinforce any retained X-link Moscow color traits and also produce 50% Panda.  It had to be considered such a route could reinforce any remaining detrimental X-linked fertility traits I was trying to eliminate.  Heterozygous Pk males and females in the BC1 were discarded.  Only those males and females which expressed homozygous Pk in conjunction with Mw were retained.  Staring with this generation females were selected for increased green iridescence on the topline.

Generation3:  Production of the 3rd generation left me with a bit of a dilemma.  The 2nd generation BC1 in theory had recombined the original X’s in 50% of females and 100% of males also possessed the same X.  However, with continued backcross it was possible to lose the autosomal recessive gg.  Not only did I wish to retain this trait into the 3rd generation, but wished it to express.  For this reason I sib bred the 2nd generation BC1 offspring, and hoped 1st generation simple outcrossed had resolved issues with females.

Several 2nd generation BC1 males were individually set up with 2-3 sibling females each.  Drops were collected from each breeding group to identify which, if any, males and females were heterozygous for gg by production of Golden offspring.  After one male was identified, all other breeding groups and their offspring were discarded.  This decision based on his superior phenotype and as a means to conserve space

2nd Generation Heterozygous Golden Panda Male (XYMw PkPk gg)

From this sole breeding group it has been possible to produce strong, hardy and viable Panda with the addition of an autosomal Golden Pink Moscow Variant.  Collection of fry from this group is ongoing to identify any sons which exhibit superior phenotype to their sire.  In effect, an intentional genetic bottleneck for specific intent.

Generation4:   To date most members of this generation result from ((Golden Pink Moscow) * (Golden Pink Moscow)) breedings.  As expected all are gg and will likely be maintained as a distinct line.  Golden guppies are known for less density of coloration in the body and finnage.  Pk still has the same effect on melanophores on scale edge, only now more exaggerated not only in body, but especially finnage combined with gg.  Many grey Panda express partial barring that is near completely masked with maturity.  So far this barring is present in 100% of gg males, but was not expressed by their 2nd generation sire in the photo above .  This would seem to suggest either a X-linked Tigrinus (Ti) or an autosomal dominant version of Zebrinus (Ze).  As males of this phenotype near maturity the barring is still completely visible.  The peduncle has darkened heavily, but retains very iridescent features.

Golden Panda Females (XX PkPk gggg)

Immature Golden Panda Males (XYMw PkPk gggg)
Additional non-Pk females containing genetics for Pb, blond (bb) and albino (aa) have been introduced for production of other Panda variants.  Sib breedings of F1 offspring have commenced.  In each case of outcross utilizing Panda males the expressed Moscow Phenotype in F1 hybrids remains surprisingly consistent:  1. Blue shoulders, 2. Silver snake-skin like peduncle and 3. VAR finnage.  This being nearly identical to the earliest reported Moscow half body snakeskin strains.  Is this indicative of the initial Moscow inputs of the strain or does it include those from introduced females?  I believe it does reveal the three traits to be in “complex” as little if any deviation is visible.  The most visible stems from introduction of yellow color pigment via my Vienna LS females to produce blue-green shoulders, yellow snake pattern and green VAR finnage.   

F1 Hybrid (Panda * Albino Vienna LS)

Discussion:  Pink Moscow as they better known in Europe, are a composite phenotype created by the blending of predominantly Y-linked Moscow (Mw) and homozygous autosomal Pink (Pk) traits.  In a basic grey body form are notated as:  XYMw PkPk  and commonly referred to as Panda’s.  A quick search of images and breeder webpages on the internet will reveal subtle variations between individuals, both male and female in grey form. It will also show variants with additional autosomes.  Are these examples all reflective of the original phenotype documented in Tsutsui’s Panda strain, and for that matter what constitutes a strain?

Generally, a named strain is a very line bred or long-term bred to produce recognizable phenotype.  Individual breeder(s) do so in hopes of a strain which breeds true to itself with a high degree of replication.  It takes line breeding to reinforce many of the subtle traits a strain is known for.  Those that go beyond simple sex linkage or autosomal inheritance resulting in phenotypes from combination and complex.  A hybrid phenotype is just that, it is not a strain.  The IFGA and other show guppy organizations tend to get around this quandary in a rather simple fashion.  They view things almost exclusively based on categorization by visible phenotype as expressed at the moment, and not strain names.  This works rather well for judging purposes, in most situations, and avoids a lot of the pitfalls they may result from differences of opinion on genetics.

Some strains are highly fixed to a single phenotype with little variation and many breeders adhere to this in both practice and philosophy.  Others may be recognized for autosomal variations such as grey, blond (IFGA gold), golden (IFGA bronze), or albino.  They may also reflect variations such as solid or variegated finnage as in snakeskins. Or even short fin or long fin variants.  The list goes on.  Again, many breeders adhere to this more “fluid” definition of a strain in both practice and philosophy, while others do not.  I fall into this latter group of breeders preferring very complex, yet fluid strains that express a host of variation.  The only true precursor is you are breeding for intent.

When breeder intent is to duplicate the original definition and you continue breeding along criteria that will further refine the results to meet that original definition, you are breeding towards a named strain.  To further argue that source genetics must be documented as descending from the original breeder is counterproductive.  Simply do so if you feel a need.  Most guppy phenotypes are readily reconstructed from unrelated sources.  What is important is your understanding of them.  Each of the variants presented either in discussion and/or photographs in this article are genetically Pink Moscow’s.  So to answer the initial question:  Are these examples all reflective of the original phenotype documented in Tsutsui’s Panda strain?  No…  

As a breeder you have the discretion to put your own individual signature on a strain through modification.   I find it best to acknowledge any deviation and resulting variants.  In example, my primary interest in guppy breeding has always been and will likely always be Swordtails.  In specific Vienna Emerald Lower Swordtails.  My strain is comprised of several lines composing many phenotypical traits that result from specific combinations of X, Y or even X & Y-link traits further influenced by several autosomal traits.  These traits may be standalone or they may be in linked complex.  I acknowledge they do not meet the original documented European definition of a Vienna Emerald, for this reason I refer to my swordtails as of “Vienna Type” and not “Vienna Emerald”.  In documenting I normally refer to specific variants by genetic name and notation.  Example:  Asian Blau Vienna Lower Sword YLs SmIr X Ab.

3 month old Golden Pink Moscow Males (XYMw PkPk gggg)

As demonstrated in my recent breedings, with a little knowledge of guppy genetics and patience a breeder can often easily reconstruct phenotypes, including the Panda from a single male in as little as two generations.  The inheritance of the Pink trait is no different from many autosomal recessive alleles.  Either sex is capable of passing Pk in unexpressed heterozygous formStill it is best to qualify use of the term “unexpressed”, as some visible modifications to melanophores in heterozygotes are visible in both Pk and gg.  Visually Pk shares several similarities with Golden (gg), also known as IFGA Bronze, in relation to melanophores. 

Homozygous Pink Moscow fry, regardless of additional autosomes, are easily distinguished at birth by a highly visible float and eyes which seems to fade with age.  Panda are found to have either black or silver eyes at birth.  With the appearance of secondary coloration from Moscow traits over the eye socket dominant or flashing males take on a nearly pure black eye.  Careful examination of fry under handheld magnification reveals a dark region over the topline extending to and increasing in coverage over the cranial region and above the eyes.  This stems from enhanced manipulation of melanophores in these areas and to a lesser degree in the flanks above the abdominal cavity.  This continues to increase with age and will lesson perceived brightness in regions found at birth.

Grey Panda Females (XX PkPk)

Just the opposite seems to occur in the lower peduncle, with an increasing reduction from below the dorsal base towards the caudal base.   Further revealing an “opaqueness” of flesh that is found over the entire body in Pk homozygotes.   Again, a similar phenotype found in homozygotes for both Pk and gg phenotypes.  

Without the addition of the Half Black genes Nigrocaudatus (Ni or NiII)  to Pk it is hard to see the gene pink as “color pink”, as it more resembles “white”.  Pink guppies are considered capable of expressing all colors typically found in guppies except red, as a result of the Pk gene.  It should be noted the existence of several Pingu strains w/visible red Mw in the shoulders.  This and the ability of Panda to express Pb in the peduncle  and Homozygous Eb Pink Moscow to express limited red in the peduncle casts some doubt on the theory of homozygous pink completely suppressing expression of red pigment via epistasis.  When Pink is found in combination with the Ni or NiII red color pigment in finnage is modified to a pale orange.  Fixed strains in this combination are often referred to as Pingu.  In some instances it is difficult to tell the difference between this modified orange and unmodified yellow pigment.  Red pigment may result in clear or light dorsals, while yellow can enhance.   

There are nearly as many phenotypical variations in Pk females as males.  This stems from their individual genetic background and how much effort has been put into re-enforcing original X-link Moscow traits of foundation sires.  After outcross, initial homozygous Pk females are often lighter than those from line bred strains.  A “pink” peduncle is still visible in either case, although the body may appear little different from a heavily reticulated non-Pk grey bodied female.  Even in these instances they will change with age as lighter “fleshy opaqueness” is still present, albeit reduced in expression.  Heterozygous gg females express more intense reticulation compared to regular siblings. Homozygous Pk females with a single dose of gg often express a visible difference in scale pattern from simple homozygous Pk siblings.

Grey Bodied Heterozygous Golden Pink Panda Female (XX PkPk gg)

Panda are not a round tail guppy in a traditional sense as a result of genotype.  Caudal shape is a result of homozygous Pink modification through suppression of Pigmentierte Caudalis (Cp), but may be further enhanced  if X &/or Y-link round tail (Rndt) genetics are present.  Most breeders find it wise to select for the darkest caudal and dorsal with clean round shape to avoid either exaggerated clear trailing edges, clear patches, or irregular shape.   

Homozygous Pk Vienna Swordtails -  Photo Courtesy of Björn Lundmark

It is routinely assumed by many breeders that homozygous Pk in conjunction with Mw modifies parental strain finnage (dorsal and caudal).  This assumption is twofold in error; 1.  Extension genetics are not removed, just suppressed to varying degrees by epistasis. 2.  Homozygous Pk in itself is capable of suppressing extension genetics in several strains.  In example above are Homozygous Pk Vienna Swordtails bred by Björn Lundmark.

In recent years I have noted several autosomal homozygotes routinely result in body modifications.   In well-bred Panda strains both males and females exhibit a type of “chunkiness”.  Being defined as; more tightly coupled in both the abdomen and peduncle than either parental stock.  Overall structure of the head and gill plates are also more “rounded” than either parental strain.  Though expression may be enhanced from combination of Mw + PkPk, there is a similar effect found in homozygous AbAb fish.  An autosomal recessive Zebrinus (Ze) in homozygous form has had just the opposite effect in my Vienna LS;  1.  Delaying maturity to produce a “long lanky” body.  2.  Diminishing dominant epistatic effect of Y-link LS over X-link DS to reveal “topspikes.”

Panda Male with X-link yellow pigment (XYYMw PkPk) showing clear edge on caudal

One of the more noticeable traits in Panda is prevalence for clear trailing edge on caudals.  As dorsals commonly exhibit less of this modification it would be easy to believe the effects of pink are limited to the body and caudal coloration.  Or are they?  Panda caudal and dorsal color is biased towards blue iridophore and melanophore modification.  While the dorsal in a Panda often looks very dark and solid, outcross often reveals Variegation (VAR) to be present.  This is a noticeable deviation from expressions held in common with gg.  Homozygous pink is likely suppressing expression of VAR, while it is commonly expressed in gg.   Depending upon mood dorsals can be very lightly colored in Pink Moscow.  This may result if red pigment is suppressed by Pink or if color is predominantly based on blue iridophores.  Incorporating an X-link for yellow fin coloration can override some of these issues.   

Without any additional color pigment and reduced melanophores Blond Panda often express clear dorsals and are rather flat in body color.  Blond Glass Belly Panda with reduced iridophores even further exacerbate this expression.  While Albino Panda can be similar to blonds in expression, there are marked differences in both iridophore and melanophore expression.  Not so “washed out” comes to mind.  Male and female grey Glass Belly Panda, with higher retention of melanophores, clearly show reduction of iridophores in both the body and finnage.  

Glass Belly Panda (XYMw PkPk GbGb ) courtesy of P Shaddock

The overall body color on Panda is a result of blue iridophores in combination with homozygous pink.  It is easily modified to purple with the addition of Pb or green with the addition of yellow pigment.  While a Panda may seem black, closer inspection reveals standalone blue is still visible in select areas on the skull, shoulders, peduncle and caudal base.   Most breeders select for a solid black coverage with attention to degree of iridescence.  A flatter black Panda has been achieved by European breeders, and appears periodically in my stocks, similar to a Black Moscow Delta counterpart.  From available evidence it appears this was accomplished with increased silver iridophores at the expense of blue.

Grey Panda with pronounced silver iridophores

All finnage on a basic Grey Pink Moscow expresses nearly solid black, regardless of additional pigment, though iridophores can bleed heavily into the caudal base.  This includes; dorsal, caudal, pectoral, pelvic, and anal.  Homozygous Pk results in intensified collection of black melanophores in certain regions of the body and fins.  At first glance the addition of homozygous gg to homozygous Pk appears to reduce this effect; Magnification reveals it to be further amplification of collected melanophores in limited regions.

5 month old Golden Pink Moscow Males (XYMw PkPk gggg)

One of the more genetically intriguing variants combines both Pingu and Panda phenotypes:  (XNi YMw PkPk).  The anterior still resembles Panda.  There is deviation from Panda phenotype in all finnage.  Not just limited to the typical lighter colored Pingu caudal and dorsal.  It includes the pectoral, anal and ventral fins, which have lost all or most black coloration as the result of Ni.  (Note:  Reasons will be expanded upon further down.)

Pingu Panda (XNi YMw PkPk) courtesy of D Czajkowski

The posterior is classic Pingu peduncle and finnage, though pink stops just short of early versions which extended to dorsal base.  Many Pingu today are similar and exhibit black on the topline in front of the dorsal with a Mw shoulder, likely from Moscow Delta foundation sires.  Original Pingu had two distinct shoulder types.  One was darker in shoulder and topline, but did not possess Mw genotype.  The second was an “Old Fashioned” shoulder pattern in which a red stripe was suppressed via epistatic effect of Pk.  Leaving for the most part only white leucophores or silver iridophores visible (see photo below.) In both cases Pk often extended into the abdomen.

Ca. 1985 Pingu (XNi Y PkPk).  Descendants of David Liebman stocks
 maintained by late Don Sauers and myself

Pink Moscow Variant coloration at the juncture of peduncle  / caudal (not inclusive of all potential)

Conclusions:  Pink Moscow are routinely bred and shown in both Asia and Europe by a dedicated following.     A lack of defined class in North America has restricted both interest and development. In all locations a loose definition of what actually comprises a Panda Guppy has allowed for a diversity of type to arise from a singular phenotype.

Based on the ease of creation it would be safe to ascertain that few if any Pink Moscow found today actually descend from either initial European or Asian (Y. Tsutsui) stocks.  The diversity of phenotype in today’s Panda strains not only is indicative of a variation in foundation Moscow genetics utilized for Y-link traits, but also female genotype.

Many guppy breeders view the Panda phenotype as a simple combination of Moscow blue and homozygous Pink.  This fails to encompass:

1.      Additional autosomes breeders can incorporate to create complex phenotypes very diverse from original descriptions,
2.     Neglects the line breeding needed for refinement into individual Strains,
3.      The diversity of genetic sources used in foundation lines,
4.      Degrees of intensity based on mood of the individual fish,
5.      A non-demanding physical structure.

Basic Panda are fairly similar in genotype.  After initial crossing of parental stock, differences can be readily visible for several generations depending upon breeder selection, compounded by loss or retention of any X-link Moscow traits.  There is clear evidence of manipulation of black melanophores at birth in Pink Moscow’s. 

As with most guppy strains a clear progression of secondary color / pattern development commences at onset of sexual maturity.  Increased iridophores to reveal silver / blue and expression of barring; followed by appearance of color pigment; and finally, in contrast to cellular layering, further enhancement of black melanophores to partially overshadow iridophores.  Suggesting some black in Panda’s is ectopic in nature and not limited to lower cellular levels.

Panda Moscow Lower Sword (LS + Mw + Pk + Ssb),
Photo courtesy of breeder Olof “Ollie” Boberg

In all but a few Pink Moscow strains coloration is highly motile in nature as a result of Mw gene(s) interaction between black melanophores in conjunction with structural iridophores & color pigment.  When possible it is wise to delay selection of your breeders to identify those with best density / intensity of coverage in body and fins.  As a result of foundation Moscow Blue (blue iridophore) genotype, descendants of Tsutsui Panda stocks were very dark colored in the peduncle.  There may have been a blau component (Eb or Ab) which was not documented.  Today, depending upon breeding Panda are also found in both green and purple body mutation (Pb).  This could stem from use of Pb females or Purple Moscow males as parental stock instead of Blue Moscow.  

Observation reveals to breeders how zones of regulation are allowing for the varied appearance in today’s Panda.  Easily evident are increased melanophores in topline and finnage, iridophore / pigment modifications anterior and posterior.  Less noticeable at first glance are zonal effects on patterns such as Zebrinus / Tigrinus and Snakeskin.  Normally found over the entire peduncle, like many other strains, they can be restricted to either upper or lower quadrants in Pink Moscow.  In the males below barring is present in both upper and lower, while snakeskin is restricted to the upper.  I have not witnessed the latter in inverse, snake in the lower, as is possible with other traits such as Saddleback (Ht).

(Left photo) Panda Moscow w/barring in upper
 & lower peduncle. Snakeskin in the upper peduncle. 

(Right photo) Blond Vienna LS w/barring
 in upper peduncle quadrant 

Homozygous Pink has often been characterized as having its greatest effect on iridophores.  However, this is not necessarily the case.  Pk exerts a far reaching impact on all levels of cellular structure in regards to color and pattern.  Most breeders acknowledge guppy color cell layering to consist of:  A lower layer of melanophores (black); a middle layer of leucophores/iridophores (minimally being blue / silver / white); a top layer of zantho/erythrophores (yellow / red).

In Goodrich, et al., (1944) regulation of phenotypic expression of melanophores by autosomals Blond and Golden are studied.  He describes Wild-type as “in some respects intermediate between the Golden and Blond.”  His  research divulges several types of melanophores:  1.  Dendritic (on the scales), 2.  Corolla and Punctate (on the body).  For the most part they are present prebirth, at birth or shortly after birth.  While Wild-type and gg guppies have the largest dendritic melanophores, bb has the smallest. Yet wild-type and bb have nearly the same amount (concentration) per square mm of tissue sample.  

Goodrich, et al., (1944)

In Nayudu, et al., (1979 No. 2) melanophores in Wild-type and three other sex-linked traits are studied.  Of particular interest in relation to Pk are her results for Half Black (NiII).  She defines them as “mutant”, being ectopic and larger than normal.  The mutant melanophores of NiII, Flavus (Fla) and Cp are present in at least the posterior peduncle region, caudal / peduncle juncture and dorsal base.

Nayudu, et al., (1979 No. 2)

Autosomals blond and golden are both considered epistatic to black in homozygous form.  Not by directly affecting melanin synthesis, but by regulating melanophore size.  Blond has little or no effect on Half Black (NiII), which has been shown to consist of larger sex-linked mutant black melanophores.  In contrast autosomal Pink aggregates concentration of regular melanophores and suppresses mutant melanophores, based on Panda and Pingu phenotypes.  Thus, being in divergence from gg and bb, Pk appears to regulate in both instances…  

Pink Moscow Male – top  (XYMw PkPk)
 Golden Pink Moscow Male - center (XYMw PkPk gggg)
 Het. Golden Pink Moscow female  - bottom (XX PkPk gg)

I used to worry that this wonderful little phenotype would disappear from the scene like so many others have in the past.  Yet, the simplicity of genotype allows for quick reconstruction from many sources.  While the more obvious phenotypes have already been produced in multiple locals, it’s hard to tell what may be presented by breeders in the future.

On a last note, it should be mentioned that Panda Moscow are a notoriously slow growing, but long lived strain.  After a quick growth spurt the first month it takes considerable time to reach mature size and coloration.  If you have an exceptional bodied male in with a group of siblings that just never seems to color up to expectations, remove him to another setting.  Often this is all it takes to reveal the full potential as a result of the motile nature of color found in Pink Moscow.

The final words in this article belong not to myself, but the individual who bred, named and documented the Panda strain.  In his words:  My answer may be abstract Japanese way... I first draw the exact image in my mind the guppy I want to create, then print it in my mind. Then keep thinking of the image everyday when I feed my fish. After a while, fish start to lead me on what to do and my eyes become more keen, so often find the guppies I need from my fishroom without adopting guppies from somewhere else.  This is what I call "Guppy Alchemy." ~ Yoshiki Tsutsui ca. 2005

1.      Robert Gall, (2005), Guppy Labs e-Bulletin, Base Body Color Pink, Volume 7, (July 2005), (no longer active site - wayback machine capture 2012)
2.      Telephone Interview by Tomoko Young, (2005), A convsersation with Yoskiki Tsutsui, Guppy Labs e-Bulletin, Volume 5, (January 2005), (no longer active - wayback machine capture 2012)
3.      Shaddock, Phillip (2010).  Guppy Color Strains.  Printed in Canada, Pocket Cine Books. [General Reference; Moscows pgs.:  15-36 / Pinks and Pingus pgs.:  113-120]
4.      Heike Savelsbergh, Guppyheike (webpage),  Moskauer Rundschwanz,  (10.31.12 active)
5.      Winge, O., Genetic Laboratory of the Royal Veterinary and Agricultural College, Copenhaen  (1927).  "THE LOCATION OF EIGTHEEN GENES IN LEBISTES RETICULATUS" [gene tables and descriptions]
6.      Petrescu-Mag I. V., et al., (2007(, Interallelic interaction between the autosomal Blond and the sex-linked Nigrocaudatus gene in the guppy (Poecilia reticulata). AIR 2(1).
7.      Kirpichnikov, V. translated by G.G. Gause., 1981, Genetic Bases of Fish Selection.  Berlin / NY, Springer Verlag (1981) [gene tables and descriptions]
8.      Shaddock P.,(2008), Blond: a regulatory gene in the guppy (Poecilia reticulata Peters 1859). AACL Bioflux.  1(2):161-164.  Printed version: ISSN 1844-8143
9.      Nayudu P.L., et al., (1979 No. 1), Genetic Studies of Melanic Color Patterns and Atypical Sex Determination in the Guppy, Poecilia reticulata, Copeia 1979, pp. 230-231,
10.  Nayudu P.L., et al., (1979 No. 2), Cytological Aspects and Differential Response to Melatonin of Melanophore Based Color Mutants in the Guppy, Poecilia reticulata, Copeia 1979, pp. 240-242.
11.  Goodrich H. B., et al., (1944), The cellular expression and genetics of two new genes in Lebistes reticulatus. Genetics 29(6):584–592.


Breeders of many fish species are intent on maintenance and preservation of what already exists.  Guppies present a unique set of challenges if only by diversity of genotype.  As breeders it is easy to believe we can add to the equation while forcing them into a specific mold.     An old adage in the pedigree livestock world states, “You start with a 100% package and you end with a 100% package.”

 All we can do as breeders is re-shuffle the deck for gains in one area and corresponding losses in another.  This is a constant when breeding any species.  There is no way around it.  Seek a balance in your efforts in conjunction with your goals to reveal what Mother Nature will allow within your particular environment…


Monday, August 13, 2012

Ginga Sulphureus... ...Evolution of a new strain or variant upon existent?

© Alan S. Bias
Permission granted for nonprofit reproduction or duplication of photos and text with proper credit for learning purposes only.

August 9, 2012

Ginga Sulphureus   ...Evolution of a new strain or 

                                                      variant upon existent?

F2 Ginga Sulphureus

It seems most new strains come about from one of four scenarios:  1. Random hybridization. 2. Mutation.  3. Crossover.  4. Planned modification.  The latter should stem from a predetermined breed plan with well set goals working with a recognizable parental strain(s) and any additional genotype you wish to infuse.  In such instance one of the first questions that comes to mind is just how much modification is needed for distinction as a new strain from it's parent?  When is it a new strain and when it is just a variant of the parental phenotype?

So what do we consider a variant?  In a simple sense when limited sex-link or autosomal traits are introduced.  Such as blond, albino, another body or fin color.  Many breeders will often limit their discussion of variants to autosomal effect. In a strict sense whenever a breeder acquires a set of genetics, maintains it for several generations in his/her own fishroom under their own devices it becomes your strain.  Even when bred towards original well defined standards of the creator or a breed standard.  This results from a change in both environmental and genetic inputs that, while similar, are no longer under the same controls.

Ginga Rubra - photo K. Tanaka with permission
A couple of years ago I first saw a picture of Kenjiro Tanaka's Ginga Rubra.  Ken has been well known in Killifish circles for many years.  His 30 plus years of manipulating guppy genetics is far less known to breeders outside of his native Japan.  In email correspondence and review of his website: it becomes clear Tanaka's interest in guppy breeding favors non delta strains that are small, highly active, with alot of coloration.  In variations of his Ginga you can find finnage that parallels caudal shape found in some killifish species.

Ginga Rubra - photo K. Tanaka with permission
For quite some time I pondered the Ginga Rubra phenotype and whether it would hold my interest.  A first step of any breeder when considering acquisition of new genetics should be an analysis of the potential makeup and it's value to your current breedings.   It was  apparent that a fair amount of the Ginga traits were autosomal and that the phenotype would always be fairly fluid in many respects.  Similar to how I maintain most of my Swordtails.  To those who appreciate non-delta type guppies Ginga are quite stunning to the eye, but in the end still a red dorsal guppy and they rarely hold my interest for long.  Which I have always found strange in that my primary focus in large animal livestock breeding has always been towards red:  Red Galloway cattle, Moorit Shetland sheep, Rhode Island Red chickens, and more.  So, after looking over limited photos of the strain for a number a months I felt I had a decent understanding of the genotype comprising this strain.  The honor of defining Ginga Rubra as a strain belongs to Tanaka as the breeder. In correspondence he indicates, "My Ginga strain is extremely variable and I'm working with many types to get (it) established."  I'll only list, describe and then discuss several visible components.

Ginga Rubra - photo K. Tanaka with permission
Today most informed breeders are inclined to view genotype based on gene complexes governed by regulation, rather than single genes resulting in specifics traits. The Overall Ginga phenotype is comprised in such a manner.  Where it differs from many modern strains is being a complex composition of recognizable traits described many years ago on a basic wild-type form.  Ginga genotype is not hidden like many modern strains if you know how to look.  You might describe it as a series of overlays and overlap.  Very similar to the Vienna type Swordtails I have worked with for nearly 30 years.

In total the Tanaka Ginga Rubra phenotype is bred for and contains at least the following visible traits:  Blond (IFGA gold), wild-type, possibly reduced Emerald Green Iridescence (EGI) from no visible peduncle spot, Iridescence, purple body mutation, barring overlayed in purple, roundtail (though some males exhibit sword-like extension), ocular spotting (in peduncle base into caudal), increased number of Wingean "orange spots" along the lateral line, red dorsal (sex linked), multiple genes identified by Winge and Kirpichnikov for caudal color patterns and shape.  

Winge, O. (1922b), male exhibiting Sulphureus and Rubra

O. Winge in his 1922b and 1923 papers describes both the sulphureus (su) and ruber (ru) genes and hints at the possibilities of crossover to explain phenotypes with only partial components of each.    In Winge's 1927 publication he goes one step further and not only illustrates this process of crossover, but reduces both su and ru from two complex genes into to four distinct genes.   The complex sulphureus becomes vitellinus (vi) and coccineus (co), while ruber becomes minutus (mi) and sanguineus (sa).  Why is this important?  Because it forms the basis of multiple components of Ginga Rubra and Ginga Sulphureus.  In 1922b he describes each:  Sulphureus;  1.  Sulphur yellow colour in the dorsal fin and a dark dot that only at times is visisble in the said fin.  2.  Sulphur yellow colour in the tial and in the caudal fins.  3.  Red colour in the lower edge of the caudal fin.  Rubra;  1.  Red colour proximally in the upper edge of the caudal fin.  2.  Large oblong red side spot, lying for the most part below and behind the dorsal fin.  3.  Dark side-dot in the tail at the base of the caudal fin.

Winge, O. (1922b)

Winge, O. (1927)

Winge, O. (1927)
Winge further describes in his 1927 publication two other genes that are potentials in the makeup of Ginga.  Those being Tigrinus (Ti); an X-link consisting of 1-2 black pigment stripes, and Zebrinus (Ze); a recessive autosome consisting of 2-5, generally 3 dark pigment stripes.  He indicates they resemble Ti, but are more pronounced.    As an autosome this trait can be both heterozygous or homozygous in form.  A heterozygous sire will statistically pass the trait to 1/2 of his sons, but not necessarily expression.  This is determined by the recessive or dominant nature of the autosome.

Winge, O. (1927)

Phang, V (1999)
There is a 3rd form of barring described by V. Phang, et al. (1999).  In conclusion she proposes:  "The designation of (bar) for an autosomal recessive gene, when in homozygous condition barbar and present with snakeskin body pattern (SSB) produces a vertical barred pattern".  The dominant and allele being bar+.  There still remains alot of speculation among breeders if bar and Ze are not one in the same.  Source fish listed were limited to a strain of Y-linked Singapore Yellow Snakeskins.  No where in Phang (1999) is there any reference to reciprocal outcross of bar+ Y to non snakeskin strains composed of barbar XX or barbar+ XX females.  Therefore, it is possible that bar and Ze are one in the same as her statistical analysis does not prove bar and SSB were found in complex.  Simply co-expressed on males as individual traits.

Iridescens (ir)  Winge 1922b, 1.  A characteristic mother of pearl sheen on the body.  2.  2-3 red smaller side spots.  3.  Black side-dot on the tial near the caudal fin.  4.  Black side-dot on the anterior body.

Winge, O. (1922b)
Ferrugineus (fe) Winge 1922b, 1.  Black rust-coloured part in the proximal part of the cuadal fin.  2.  Black side-dot in in the tail near the caudal find (viz. with ocular spotting in the caudal).

Winge, O. (1922b)
Purple Body Mutation - undescribed, is a very important Ginga trait that needs to me mentioned here.  Tanaka's Ginga Rubra are bred exclusively for Pb.  To the best of my knowledge there is no formal description of this trait in guppies, having found no published research or reliable documentation of it.  Discussion with other breeders and researchers has also provided minimal insight.  To my knowledge this trait is found in only two of three published members of the guppy family comprising the subgenus Acanthophacelus; Poecilia reticualata, Poecilia obscura, and Poecilia Wingei.  This trait does not seem to occur in any of the pure Poecilia Wingei (Enders) that I worked with, nor are there any photos online from reliably documented populations.  It is only expressed in hybrids.  In the past, having been able to breed for it reliably, it did not seem necessary to set up breeding groups, dedicate tanks and document results statistically.  My test crosses to decipher it, in several strains, to include Ginga, Vienna and Wild-type, are not yet conclusive.  

This trait I refer to as Purple Body Mutation (Pb).  It's counterpart is Green.  My test breedings to date with non Ginga strains point towards Pb being an autosomal recessive trait.  Green presumably could be a dominant wild-type allele or epistatic trait as evidenced by lack of Pb in Endlers.  Until it is proven whether the latter is a wild-type allele (Pb+) or a distinct trait (Gb) I see no use in defining it as either.  To do so would require a larger number of free tanks than I have available.  Two strains; one homzygous for purple and another for green.  Documented analysis of reciprocal F1 crosses followed by F2 sib breedings.   
Vienna LS in both Pb (upper) and green (lower)
Purple Body Mutation as an autosomal recessive trait is a precursor in determining visible color resulting from pigment overlay.   Pb is what allows for creation of a true purple phenotype (viz. IFGA Purple).  Effect is limited to certain regions of the body.  Always in the peduncle in strains with reduced iridophores in this area.  In conversion of Wingean spots from orange to pink.  The expression can also extend along the shoulders and into the topline.  In most males, the belly region does not appear to be altered by the effect of Pb.  Green males express Wingean spots in orange.  Pb males express Wingean spots in pink.   While Pb is most noticeable in the peduncle region of male guppies, it is also expressed in females.  Along the lateral line and in the shoulders towards the topline, appearing as an iridescent purple sheen on the scales.  

In late summer of 2011 I arranged to obtain a trio of pure F1 Ginga Rubra from an individual who had imported directly from Tanaka.  My only requests were a male having a round tail and ocular spotting with no visible extensions, roundtail being a recessive not easily identifiable in my Vienna.  This request was fulfilled with the above below as he appeared after 4-5 months growth.
Ginga Rubra (P sire partially colored)
Ginga Rubra (P sire mature)
 As some of you have surmised based on yellow based breeding program my primary focus, at least initially, would be to remove red  in the dorsal and caudal via selective breeding.  If not feasible to at least mask it via epistasis from traits found in my Vienna LS.  To do this I would initiate breedings with a pure Ginga male and several pure Vienna females, blond and grey, to include both green and purple body mutation.  Then utilizing F1, F2 and F3 hybrid males exclusively bred the first 2-3 generations in backcross to pure Vienna females.  All hybrid filial sibling females were culled prior to onset of maturity.   Such a breed plan would avoid the pitfalls of segregation when using siblings.  In result:  (((Ginga * Vienna LS) * Vienna LS) * Vienna LS).  All initial Vienna females came from my Line1 as this line does not posses partially dominant autosomal Z-bar.  It does contain both Pb and green in composition.

One of the first things I wished to know about about Ginga had to do with barring.  Was this trait traditional autosomal Z-bar or the elusive recessive sex linked Tigrinus as documented by O. Winge early in the last century.

Ginga Sulphureus (F1 sire)
The F1 surprisingly resulted in at least two distinct phenotypes from several litters.  One was very Ginga in appearance to the sire with the addition of yellow finnage as hoped.  From the available males the one above was selected to sire the next generation.  You will notice a residual amount of red in the upper caudal.  While I did not do test crosses, this likely was the result of Y-link mi as it does not manifest in my Line1 Vienna stocks.  Dorsal size and shape was near identical between father and son at maturity.  Moderate in length and tapering to a well defined point.  Retention of overall body pattern still hinted at Y-link &/or autosomal traits.  In both F1 phenotypes barring was present in 100% of males from litters of four (4) females to varying degress  This without a doubt says autosomal Z-bar and not sex-linked Tirgrinus.  It also suggests an autosomal dominant version of Ze and not a recessive version as my Line1 Vienna LS traditionally lack Ze.  In F2 backcross Ze is still present nearly 100%, albeit weakened in some males.  This continues to suggest that the Ginga are comprised of a autosomal dominant version of Ze and not a recessive.  You will notice the appearance of a green peduncle spot in this F1 male.  It would be hopeful to suggest an autosomal input from his Vienna dam.  It is more likely visible from a reduction in barring.  Z-bar is often epistatic to peduncle spots when bred for posterior coverage toward the caudal base.  

F1 Ginga * Vienna hybrid
I mentioned a second distinct F1 phenotype that appeared in both blond & grey siblings.  A single grey female likely was the dam of all.   I have gradually culled or given away the blonds, but kept back a lone grey male who I am now breeding to both pure Vienna and F2 Ginga Sulphureus females.  His finnage is very reminiscent to the types of guppies I bred as a youngster back in the late 1960's and early 1970's.  An interesting Lyretail with little extension and in appearance mimics many Aphyosemion killifish species.  It will be interesting to see how much of this caudal type was passed, from his pure Vienna dam.  I have always suspected that much of either Y or X-link DS trait is carried by females both in sex link and autsomal, and is masked by epistatic Y-link LS.

F1 Ginga * Vienna hybrid
Outside of his caudal fin, other body color / pattern components are still very Ginga.  Note that all F1 males of this phenotype lacked dorsal color pigment.  Either being  clear or lightly colored  by metallic (Mg) and this is probably in complex with the unique caudal shape.  If so, a possible Y-link as my females will pass X-link yellow dorsal nearly 100% when linebred.  These males also lack a green peduncle spot.  In both F1 phenotypes the amount of orange / pink spotting seemed to increase and intensify with age.  Both in anterior regions and over posterior barring.

Segregation in the F2 has come about in the expected ratios.  While Ze has remained strong, the degree of intensity (viz. total number of bars) has decreased in some males from several generations of backcross.  A mixture of blond / grey in both green and purple body mutation.  For F3 breeding groups of each have been set up with sibling females to further this new strain at least one more generation.  With emphasis on increasing what I am sure Kenjiro Tanaka would consider not only a flaw, but one of his greatest challenges in breeding this unique strain.  Removing or at least masking the peduncle spot while increasing both ocular caudal coverage, also known as "peacock spots", and the degree / intensity of Wingean orange / pink spotting.

F2 Ginga Sulphureus
F2 Ginga Sulphureus

F2 Ginga Sulphureus

F2 Ginga Sulphureus
F2 Ginga Sulphureus

F2 Ginga Sulphureus

I come across quite a few comments in correspondence or on the web hinting at Ginga's deriving much of their phenotype from possible Poecilia Wingei (Endlers Livebearer) influence.  Surprisingly, all of the individual traits I identified, less some finnage, were already existent in either the Line1 / Line2 or both of my combined Vienna Emerald Lowersword (LS) strain.  Those few that are not common today, such as ocular spotting, were often found in guppy strains of the past.  This has allowed me to utilize Vienna in conjunction with a new direction in breeding Ginga's.  That being a reduction of red finnage in favor of pure yellow.  The primary trait in found in Ginga Rubra lacking in my Vienna LS is the layering of purple striping over Ze.  I also found that one of the most striking traits on the Ginga to be a seeming ability to express both Pb and green simultaneously.  Again, this trait while rare is also found in P. Reticulata strains.   The male on the right has no Ginga genetics in his makeup, yet posses the same ability to express both orange anterior and pink posterior.  Note that I have never run across an inversion of this combination, pink in the anterior and orange in the posterior.

F1 Ginga Sulphureus (l) and Schimmelpennig Platinum DS crossover male (r).
 Both expressing Pb posterior in peduncle and orange anterior in shoulders.
Breedings continue to indicated Pb as an autosomal autosome.  Similar to Asian Blau (Ab).  Analysis of several prior non Ginga breeding test results, consisting of Vienna / Schim / JB albino, and lack of Pb in Endlers point to green being a wild-type allele Pb+.  As Tanaka line breeds for Pb in Ginga Rubra, it is assumed both males and females are homozygous for the Pb trait.  Yet, no green presented in F1 from a pure Ginga Rubra single sire mating to 2 females (Ginga * Vienna LS).  It is doubtful both of these females were homozygous for Pb as their sire had been green.

I'm looking forward to a series of reciprocal Vienna LS * Ginga Sulphureus crosses a couple generations from now.  It will be interesting to see what autosomal traits will be contained in the genotype of linebred  Ginga Sulphureus females.  In the long-run when I work with a new strain the end result is not only to see what traits I can stabilize, but identify those which I can feasibly incorporate into my Vienna LS.

Winge, O., Genetic Laboratory of the Royal Veterinary and Agricultural College, Copenhaen

Phang, V. et al., Dept. of Biological Sciences, Faculty of Science, Nat. Univ. of Singapore (1999). "Interaction between the Autosomal Recessive bar Gene and the Y-linked Snakeskin Body (Ssb) Pattern Gene in the Guppy, Poecilia reticulata"

Tanaka, K & Bias, A. S. (2012).  Email correspondences on Ginga Rubra

Which do you find of more challenge as a breeder, creation and stabilization of a new phenotype or continued propagation and improvement of and existing one?
  Each a unique challenge in their own right.   


Click on blog photos to enlarge

Click on blog photos to enlarge