Grammostola rosea: A possible insight into its genetics

[drgonzo]

In fact, incomplete dominance (what you are mistakenly calling co-dominance) is the ONLY way that we can get three color forms. (Unless, of course, something really weird is happening here. Possible, especially with tarantulas, but not very probable.) The test in our case is the difference in the appearance of the offspring between the two possible outcomes. If pink occurs in the real world, then the inheritance has to be incomplete dominance.

OK I'm trying to understand the incomplete dominance.bear with me

If Rr was bred to Rr the out come would be 25%RR 50%Rr and 25%rr
The 25%rr would they then be carring a hetro version of the Red genotype but not expressing the phenotype?(so if two of the rr from the same sac were bred they could produce Reds???)
And the 50%Rr would be carring the homo version of the Red genotype expressng the phenotype?
It would be like a visiable het that isn't always visiable????

 

[Stan Schultz]

... NCF I'm refering to this/pink?
View attachment 97522 ...

I would call this one a Grey = NCF = rr.

To see what I would call a pink, got back to my initial post, scroll down towards the bottom and look at the listing with the associated thumbnails.

... The only thing that was diffrent on my chart was If it was a codom trait
Rr x Rr = 25% RR, 50% Rr, 25% ...

Substituting incomplete dominance (INCOM?) into your sentence makes this part substantially correct.

... rr would be Rr X Rr = 25% RR,75% Rf,0% rr(making it impossible to produce all three color forms in one sac) ...

But, this doesn't make any sense to me. If you cross Rr x Rr you'll get this Punnett square:


(Uploaded with ImageShack.us)

That reduces down to
RR = 25% = RCF (Red)
Rr = 50% = PCF (Pink color form, for want of a better label)
rr = 25% = NCF (Grey)
if we assume the INCOM form of inheritance.

And:
RR = 25% = RCF (Red)
Rr = 50% = RCF (Red)
rr = 25% = NCF (Grey)

Or, 75% RCF to 25% NCF
if we assume the Dominant/Recessive scheme.

DISAMBIGUATION
We need to make a clear distinction between the genetic makeup of an organism (genotype) and its appearance (phenotype).

The genotype is the underlying source code for a particular characteristic or group of characteristics. A genotype is composed of one or more genes. The genotype is usually considered to be an abstraction created to explain inheritance. Recently, many parts of this abstraction have been proven to be real world things, although they are so small that we can't see them. In thought experiments we sort and juggle these bits of source code, this genotype, according to a set of rules to determine how many and what combinations are possible. We intentionally do not label them directly with real world names because we're trying to divest them of any bias or interpretational meaning. We're trying to deal with them on a much more basic level. We use letter designations as a carryover from algebra, which attempts to do much the same thing with other real world issues. In my example, the Punnett square and the letters "R" and "r" are only applicable to the genotype since they label the different states of a gene and how it might hypothetically be sorted during reproduction.

On the other hand, we have the real world "appearance" or "expression" of the characteristic on the organism. This is the phenotype. The phenotype is derived from the genotype in that the displayed characteristics depend on the genes, but the phenotype does not directly control, interact with, or influence the genotype. It's a one way street. Genotype => Phenotype. However, if we understand the rules that the genotype follows to produce a phenotype, we can infer ("backwards engineer") what the genotype had to be to produce the result that we see. We label phenotypes with real world names or their acronyms partly because we can somehow see, detect, or document the characteristics, and because we seldom perform mathematical or logical tests or manipulations on the things we call phenotypes. Nowhere is it written that there needs to be a strict 1 to 1 correlation between genotypes and phenotypes, and the last set of equivalencies just above this section demonstrates that beautifully. Two different genotypes (RR and Rr) produce the same phenotype (RCF).

In our example at the top of this post:

Grey = NCF = rr

Grey is the real world description for this characteristic. NCF is an acronym used to identify the phenotype. rr is the abstract code that we used to describe the genotype. And I listed all three this way to nail down their relationship in my discussions.

So, whenever you are speaking about inheritance you have to sit back a moment and try to figure out whether you're talking about the genes themselves (aka, the genotype) or merely the appearance that results from those genes (the phenotype). And when we speak about how a specific characteristic is inherited, we need to understand that the gene is inherited, not the characteristic, although the characteristic may be resulting from the gene in question. (But, note: The characteristic might also be the result of two or more genes acting independently or in concert. And from here it gets really deep and confusing!)

In this way at least the text editing features on our Internet browsers should allow us to review and edit our postings before we send them so we don't get those two terms confused again. Especially me, because I can never get anything right until the third try!

If at first you don't succeed, skydiving is not for you.
​

I hope this helps a little.

---------- Post added 01-03-2012 at 02:06 PM ----------

OK I'm trying to understand the incomplete dominance.bear with me ...

Of course.

... If Rr was bred to Rr the out come would be 25%RR 50%Rr and 25%rr
The 25%rr would they then be carring a hetro version of the Red genotype but not expressing the phenotype?(so if two of the rr from the same sac were bred they could produce Reds???) ...

No. Go back to the Punnett square and the definitions I edited into the very first posting.

"rr" is equivalent to "un-red, un-red." THERE IS NO RED GENE HERE! And, such individuals could not be responsible for any red coloring in their offspring. Any red would have to be due to the genotype of their mate.

... And the 50%Rr would be carring the homo version of the Red genotype expressng the phenotype?
It would be like a visiable het that isn't always visiable????

In the common parlance, I presume "hetro," "hetero," and "het" to be shortened forms of "heterozygous." Which means that the individual's genotype contains both versions of the same gene: "Rr" in our example.

"Homo," I presume, is the vernacular for "homozygous." Which means that both copies of the gene in question of the individual in question are identical: "RR" or "rr."

You have your definitions switched.

And, homozygous and heterozygous are not versions. They describe two contrasting CLASSES of genotypes (genetic makeups):

RR is homozygous for the "R" gene.
Rr is heterozygous for both.
rr is homozygous for the "r" gene.

Is this beginning to make any sense to you now?

---------- Post added 01-03-2012 at 02:13 PM ----------

OK I'm trying to understand the incomplete dominance.bear with me ...

It just occurred to me that you may not understand what the letters "R" and "r" mean. And, that would explain a lot of your confusion.

"R" stands for the fully functioning gene. When present it causes the production of the red pigment.

"r" stands for the altered, nonfunctional, or mutated gene. While this gene still exists in the creature's genotype it serves only as a placeholder or "filler" in the genetic sequence. Effectively, it's a blank.

Does this help?

 

[Fins]

This is very interesting stuff even though I only understand it on a basic level. So sorry if this was this was explained already. Just curious.

Does the difference between "appearance" & "expression" mean that an individual with Rr could "appear" to be a NCF or will they always "appear" pink?

 

[Stan Schultz]

This is very interesting stuff even though I only understand it on a basic level. ...

Actually, this is all pretty basic stuff that's taught in some high school biology classes. Oddly enough, more than 50 years ago I was taught this stuff in 10th grade biology in a school system that was at that time considered to be near the bottom of the list in the State of Michigan! Mr. Gerow was a fantastic biology teacher!

... Does the difference between "appearance" & "expression" mean that an individual with Rr could "appear" to be a NCF or will they always "appear" pink?

"Appearance" and "expression" are often used as synonyms for each other. I'm sorry if I gave you the impression that the geneticist might have different definitions for them. (In reality, it's possible that geneticists do consider them different terms with different meanings, but I've never run into that situation.)

In the hypothesis that I presented Rr will always appear or be expressed as PCF (Pink color form unless someone suggests a better term), not NCF (Normal color form, I think). By my definition, NCF is the common grey color with a genotype of "rr." Someone may present a different story, redefining the terminology to suite their hypothesis or personal inclination. This points out the necessity for getting precise definitions of terms sorted out at the beginning to avoid confusion down the road.

But to answer your question in a generic sense, providing that there isn't something strange going on with some other gene (but see below), yes, with incomplete dominance the heterozygous condition (i.e., one member of the pair of genes is normal, and the other is "different") will "always" produce an intermediate phenotype (appearance). ("Always" is a long, long time, so we must take it with a proverbial "grain if salt," because sooner or later we're going to find an exception. Promise.)

Re: "providing that there isn't something strange going on with some other gene" - In the real world, very few characteristics are effected by only one gene pair. The majority of characteristics may be primarily controlled by one gene pair, but are also secondarily modulated by other gene pairs. As an example, in humans one gene pair enables the production of melanin, but as many as 7 other gene pairs control the magnitude of that melanin production to give us the almost continuous range of skin colors that we can see just by taking a ride on a city bus! Further, there is a reaction to ultraviolet light that is further modulated by other genes to vary the intensity of melanin in our skins. I'm a blue eyed blond, and my scalp becomes sunburnt under my hair even though I'm not bald. And, with great difficulty and buckets of sunblock I can develop a tan (without burning) that lasts for about 2 days. However, while black Africans and those from the Indian subcontinent who are black may sunburn, they generally don't do so to anywhere near the extent that I can if I'm not careful. They just get darker when exposed to raw sunlight. These varying reactions to environmental stimuli are also mediated by other genes, and their activity differs depending on racial or family backgrounds.

Several decades ago some research scientist was researching some question about human skin color, but his research was thwarted until he found a patch of skin that was almost never exposed to sunlight: the skin on our right buttock! So, some number of college kids had to drop drawers to get photometry done on their @$$. All in the name of science!

Fortunately, we don't have to do that with our tarantulas.

 

[somethingbig]

So...
IF
RR always produces RCF, Rr always produces PCF, and rr always produces NCF

THEN
RR x RR will always produce RCF
rr x rr will always produce NCF
and
Rr x Rr will theoretically always produce all 3

At least that much should be easy enough to either strengthen or disprove the hypothesis in just one generation. Of course this doesn't account for multiple genes, alleles, and whatnot..

 

[drgonzo]

We need to better define the three seperate color phases
rr=grey if this isn't what was being sold years ago as Grammastola sp.north with the white setea,and is the Normal color phase.
Can someone post pics of the three color phases in similar lighting?
I mainly wanna see what were calling "pink"

 

[Stan Schultz]

So...
IF
RR always produces RCF, Rr always produces PCF, and rr always produces NCF

THEN
RR x RR will always produce RCF
rr x rr will always produce NCF
and
Rr x Rr will theoretically always produce all 3 ...

[size=+1]BINGO![/size]

At least that's true if the incomplete dominance model is appropriate. And, the fact that we get three colors is pretty strong evidence for that.

... Of course this doesn't account for multiple genes, alleles, and whatnot..

I refer you to posting #13. The one about "Quacks like duck", "Occam's Razor," and "Simple is Better." Don't make it any more complicated than you have to.

 

[drgonzo]

I would call this one a Grey = NCF = rr.

To see what I would call a pink, got back to my initial post, scroll down towards the bottom and look at the listing with the associated thumbnails.



Substituting incomplete dominance (INCOM?) into your sentence makes this part substantially correct.



But, this doesn't make any sense to me. If you cross Rr x Rr you'll get this Punnett square:


(Uploaded with ImageShack.us)

That reduces down to
RR = 25% = RCF (Red)
Rr = 50% = PCF (Pink color form, for want of a better label)
rr = 25% = NCF (Grey)
if we assume the INCOM form of inheritance.

And:
RR = 25% = RCF (Red)
Rr = 50% = RCF (Red)
rr = 25% = NCF (Grey)

Or, 75% RCF to 25% NCF
if we assume the Dominant/Recessive scheme.

DISAMBIGUATION
We need to make a clear distinction between the genetic makeup of an organism (genotype) and its appearance (phenotype).

The genotype is the underlying source code for a particular characteristic or group of characteristics. A genotype is composed of one or more genes. The genotype is usually considered to be an abstraction created to explain inheritance. Recently, many parts of this abstraction have been proven to be real world things, although they are so small that we can't see them. In thought experiments we sort and juggle these bits of source code, this genotype, according to a set of rules to determine how many and what combinations are possible. We intentionally do not label them directly with real world names because we're trying to divest them of any bias or interpretational meaning. We're trying to deal with them on a much more basic level. We use letter designations as a carryover from algebra, which attempts to do much the same thing with other real world issues. In my example, the Punnett square and the letters "R" and "r" are only applicable to the genotype since they label the different states of a gene and how it might hypothetically be sorted during reproduction.

On the other hand, we have the real world "appearance" or "expression" of the characteristic on the organism. This is the phenotype. The phenotype is derived from the genotype in that the displayed characteristics depend on the genes, but the phenotype does not directly control, interact with, or influence the genotype. It's a one way street. Genotype => Phenotype. However, if we understand the rules that the genotype follows to produce a phenotype, we can infer ("backwards engineer") what the genotype had to be to produce the result that we see. We label phenotypes with real world names or their acronyms partly because we can somehow see, detect, or document the characteristics, and because we seldom perform mathematical or logical tests or manipulations on the things we call phenotypes. Nowhere is it written that there needs to be a strict 1 to 1 correlation between genotypes and phenotypes, and the last set of equivalencies just above this section demonstrates that beautifully. Two different genotypes (RR and Rr) produce the same phenotype (RCF).

In our example at the top of this post:

Grey = NCF = rr

Grey is the real world description for this characteristic. NCF is an acronym used to identify the phenotype. rr is the abstract code that we used to describe the genotype. And I listed all three this way to nail down their relationship in my discussions.

So, whenever you are speaking about inheritance you have to sit back a moment and try to figure out whether you're talking about the genes themselves (aka, the genotype) or merely the appearance that results from those genes (the phenotype). And when we speak about how a specific characteristic is inherited, we need to understand that the gene is inherited, not the characteristic, although the characteristic may be resulting from the gene in question. (But, note: The characteristic might also be the result of two or more genes acting independently or in concert. And from here it gets really deep and confusing!)

In this way at least the text editing features on our Internet browsers should allow us to review and edit our postings before we send them so we don't get those two terms confused again. Especially me, because I can never get anything right until the third try!

If at first you don't succeed, skydiving is not for you.
​

I hope this helps a little.

---------- Post added 01-03-2012 at 02:06 PM ----------



Of course.



No. Go back to the Punnett square and the definitions I edited into the very first posting.

"rr" is equivalent to "un-red, un-red." THERE IS NO RED GENE HERE! And, such individuals could not be responsible for any red coloring in their offspring. Any red would have to be due to the genotype of their mate.



In the common parlance, I presume "hetro," "hetero," and "het" to be shortened forms of "heterozygous." Which means that the individual's genotype contains both versions of the same gene: "Rr" in our example.

"Homo," I presume, is the vernacular for "homozygous." Which means that both copies of the gene in question of the individual in question are identical: "RR" or "rr."

You have your definitions switched.

And, homozygous and heterozygous are not versions. They describe two contrasting CLASSES of genotypes (genetic makeups):

RR is homozygous for the "R" gene.
Rr is heterozygous for both.
rr is homozygous for the "r" gene.

Is this beginning to make any sense to you now?

---------- Post added 01-03-2012 at 02:13 PM ----------



It just occurred to me that you may not understand what the letters "R" and "r" mean. And, that would explain a lot of your confusion.

"R" stands for the fully functioning gene. When present it causes the production of the red pigment.

"r" stands for the altered, nonfunctional, or mutated gene. While this gene still exists in the creature's genotype it serves only as a placeholder or "filler" in the genetic sequence. Effectively, it's a blank.

Does this help?

100%
Incomplete dominance is they same as co dominance
I see were I was making a mistake on my out come with RrXRr,I don't know how I was comming up with that but it was 100% wrong.All three color phase would be present.
Now the only thing I'm not 100% on is the "pink" color phase.

 

[Tarac]

Why are we focusing on this particular genetic system? My impression is that we have virtually no genetic data available for virtually all tarantulas. So is this just conjecture for the sake of demonstration or are you really trying to use this to set up some type of breeding parameters? Did I miss something? I read the whole post, but I apologize if this is redundant or not relevant. Why do we think the T in question has a simple allele system, rather than something with more than just two?... more than two isn't something weird or funny that is rare, quite the opposite- usually only two can occupy a single loci, this is true. But there could be more than two options overall. In general we use punnett squares because it's a very very simplified way to predict phenotypic makeup in a pairing of parents with known genotypes but it's largely demonstrative. It's a concept model really. You've got only a part of the story here. There are only two alleles at each loci because each parent contributes one. But it doesn't mean that all parents in that population have either A or B, there could be individuals out there with C,D,E, or Z as well. So it's not really accurate to think about inheritance in terms of only two possible alleles until you have some reason to suspect that a particular gene is strictly bimodal.

The most obvious example is blood type- there are only four phenotypes for blood type in humans (note: six genotypes) however tons of alleles at that loci possible, more than 50 (I don't remember the exact number off the top of my head, but I'm sure someone will check that lol. For simplicity we think of three, A-B-O but that's also the watered down version of blood type inheritance as there are many different polymorphisms of A and B and O so the gene A can be built with a vast number of different nucleotide arrangements with substitutions at various different residues that ultimately still produce "A" even though it's not exactly the same as the next A and so on).

I think in order to hazard these kinds of guesses you really need to know for a fact how many alleles you are dealing with. There could be another allele or two or 50 that are possible at that same loci. Obviously each loci should probably only have two each, but you might have three total represented in your cross if, for example, that white one pictured was actually W so you have RW and RR or Rr or something along those lines. You can only consider dominance once the number alleles and hierarchy of heritability are known, otherwise no sure way to predict phenotype because there could be lurking alleles you aren't factoring in. If you think of the multitude possible for blood type in humans that reduce down to a set of only four phenotypes you can see why assuming that small number of phenotypes= simple inheritance pattern is not necessarily accurate.

To avoid making the little figures if you want, the punnett squares can be calculated by simply multiplying the sum of each parents' genotype together, so parent 1= AB, parent 2= AB and your offspring are (A+B)(A+B) = A^2+AB+AB+B^2 = AA, 2AB, BB in 1:2:1 or 25%:50%:25%

---------- Post added 01-04-2012 at 01:07 PM ----------

What the heck is that little monkey? lol

 

[jayefbe]

Why are we focusing on this particular genetic system? My impression is that we have virtually no genetic data available for virtually all tarantulas. So is this just conjecture for the sake of demonstration or are you really trying to use this to set up some type of breeding parameters? Did I miss something? I read the whole post, but I apologize if this is redundant or not relevant. Why do we think the T in question has a simple allele system, rather than something with more than just two?... more than two isn't something weird or funny that is rare, quite the opposite- usually only two can occupy a single loci, this is true. But there could be more than two options overall. In general we use punnett squares because it's a very very simplified way to predict phenotypic makeup in a pairing of parents with known genotypes but it's largely demonstrative. It's a concept model really. You've got only a part of the story here. There are only two alleles at each loci because each parent contributes one. But it doesn't mean that all parents in that population have either A or B, there could be individuals out there with C,D,E, or Z as well. So it's not really accurate to think about inheritance in terms of only two possible alleles until you have some reason to suspect that a particular gene is strictly bimodal.

The most obvious example is blood type- there are only four phenotypes for blood type in humans (note: six genotypes) however tons of alleles at that loci possible, more than 50 (I don't remember the exact number off the top of my head, but I'm sure someone will check that lol. For simplicity we think of three, A-B-O but that's also the watered down version of blood type inheritance as there are many different polymorphisms of A and B and O so the gene A can be built with a vast number of different nucleotide arrangements with substitutions at various different residues that ultimately still produce "A" even though it's not exactly the same as the next A and so on).

I think in order to hazard these kinds of guesses you really need to know for a fact how many alleles you are dealing with. There could be another allele or two or 50 that are possible at that same loci. Obviously each loci should probably only have two each, but you might have three total represented in your cross if, for example, that white one pictured was actually W so you have RW and RR or Rr or something along those lines. You can only consider dominance once the number alleles and hierarchy of heritability are known, otherwise no sure way to predict phenotype because there could be lurking alleles you aren't factoring in. If you think of the multitude possible for blood type in humans that reduce down to a set of only four phenotypes you can see why assuming that small number of phenotypes= simple inheritance pattern is not necessarily accurate.

To avoid making the little figures if you want, the punnett squares can be calculated by simply multiplying the sum of each parents' genotype together, so parent 1= AB, parent 2= AB and your offspring are (A+B)(A+B) = A^2+AB+AB+B^2 = AA, 2AB, BB in 1:2:1 or 25%:50%:25%

---------- Post added 01-04-2012 at 01:07 PM ----------

What the heck is that little monkey? lol

You make a lot of good points, but I do think that there is enough information present to believe that the most likely scenario is a single locus with two alleles. 3 discrete phenotype classes makes this the simplest model. While it is possible that there may be some form of epistasis that can lead to the same distinct coloration classes (coat in labrador retrievers), a single locus with 2 alleles is the simplest. Going beyond this thread to the literature, most examples of adaptive coat coloration shift are caused by a single locus with two alleles (one functional, one not). I'm not extremely familiar with the molecular pathways responsible for pigmentation in tarantulas, but I imagine they are similar, at least in a broad manner, to those found in vertebrates. Additionally, most known examples of adaptation in which the genetic mechanism underlying the trait is known have involved a single locus or very few loci. While it is certainly true that most traits are polygenic, and there is a bias for simpler genetic traits due to their ease in determination (comparatively), simple Mendelian traits playing an important role in adaptation isn't rare.

On another note, while it may be different in the medical research world, thinking of adaptation as occurring through 2 alleles per locus is a common and effective viewpoint from an evolutionary perspective, and is nearly always implicated in cases of adaptation. Unless there is some form of balancing selection, multiple alleles will not be maintained over time. Assuming no selection, drift will fix one allele. Off the top of my head, the only cases in which multiple alleles for a single locus plays an important role involves some sort of extreme frequency-dependent selection (MHC loci for example).

 

[Stan Schultz]

Why are we focusing on this particular genetic system? ...

1> Because we have to start someplace.

2> Because we have to start with a simple enough model that without too much work the average-Joe (or Jo)-tarantula keeper has a chance of understanding.

3> Because most of the people who are reading this have to learn basic genetics from scratch to even begin to understand what we're saying, and this is an important learning exercise for them.

4> Because once we are fairly certain that we've nailed the heredity of this one facet of tarantula biology for this one, very popular and interesting species, we will have the confidence and the background to move onto other, more challenging, genetic projects.

5> Because of

[size=+1]SCHULTZ' PARADIGMS[/size]​

A> If it looks like a duck and quacks like a duck it's probably a duck. (I.e., trust the data.)

B> Occam's Razor. (Survival of the simplest, least troublesome, etc.)

C> Simple is better. (The fewer things there are to go wrong, the better.)

6> My simple, little hypothesis fits oh-so-well!

7> Because I'm (bored) as oop: with all the "Help! My Chilean rose won't eat!" postings. (Read the fricking care sheet, d*** it!)

8> Because I think a bunch of people on these forums are actually interested in exploring and learning some REAL, NEW information about both the world they live in and the spiders they keep.

9> And, because it's for the pure fun of it!

And, I use the little graphics because people are graphics orientated. Looking at, much less understanding, a seemingly endless written list of letters and numbers is absolutely guaranteed to lose your audience, unless they make 6 figure+ salaries as accountants or computer programmers by doing so. Walt Disney proved the point many decades ago, "We love cartoons!"

:biggrin: : :biggrin: :cute: :giggle: :laugh: :love: :: :roflmao: :smile: :tongue: :wink: ​

Now, all your concerns are certainly valid, if you're a geneticist with a hefty research grant and decades at your disposal for research. But, they are of little practical importance unless you can show that my hypothesis is incorrect. Then it's up to you to propose an hypothesis that fits the data better.

There! The gauntlet has been thrown! The challenge has been made! The ball is in your court! Do or die! (Oops! I've run out of idioms!)

:biggrin:

(And, thank you very much for the opportunity to vent. Now I must go commit laundry. )

 

[somethingbig]

I think the only thing left to benefit this conversation is some input from people that have bred:
RCF x RCF
PCF x PCF
and
NCF x NCF

I guess RCF x NCF would help too seeing as 100% of the offspring should be PCF according to the hypothesis.

 

[Necromion]

Something else that could help in this case is if someone had access to the materials to do run the complete genetics for a G. rosea and compare the different color forms genetic squencing, being someone who is interested in taxonomy I know that this in the cases of some beetles, has been done to clarify certain species (sorry more word of mouth from beetle taxonomists I know) to the point that two species are geneticly different enough that they become classified as to seperate species. however this could also show us that the only difference is in the color form, which in that case would prove that it is quite possible (although seemingly rare), to get all there color forms from one single egg sac. Although if it would be proven that the Rcf is in fact a different species, and depending on how the test goes for the pcf (whether or not third species), It could show that the Pcf is a possible/naturally occuring hybrid of the two.

Just my two cents, and a test I would love to run (as I have all three color forms) if I had access to the materials.

 

[drgonzo]

I think the only thing left to benefit this conversation is some input from people that have bred:
RCF x RCF
PCF x PCF
and
NCF x NCF

I guess RCF x NCF would help too seeing as 100% of the offspring should be PCF according to the hypothesis.

A picture of a WC PCF would be nice also.
In referance to the photo's on post #1
Picture three an picture two are under very diffrent lighting conditions and NCF color varries a lot durring a molt cycle.
Picture three looks like a freshly molted NCF/"gray"
If you put picture #2 in photshop and bring the brightness down to the level of picture #3 it looks like a NCF/"gray" too...
Examples
For NCF
fresh molt 12/24/11




Pre molt



Examples for RCF
Fresh Molt 12/4/11




Premolt RCF

 

[jim777]

View attachment 97521

Wow, that is a cool looking spider! It looks almost entirely white.

Stan, great thread - thanks so much!

 

[Tarac]

But, they are of little practical importance unless you can show that my hypothesis is incorrect. Then it's up to you to propose an hypothesis that fits the data better.

Here's the first problem (after the part where you refer to data that doesn't exist anywhere)- I can demonstrate that there are other ways to get these forms, so not proving the negative doesn't validate your hypothesis, that's not how conclusions are drawn. It is in fact pertinent as it shows that there are more than one way to get the forms we see, so the hypothesis is not really anything until you either A. prove your hypothesis correct (which will be hard with the genetic information I mentioned) or B. prove all of the other possible hypothesis incorrect, the way a theory would be adopted in science (also hard without the aforementioned data). As far as I can see neither has been investigated, so it's really mute other than no, you cannot go around saying you've figured out the inheritance pattern and allele set of the G. rosea/porteri- I believe the entire post is predicated with that disclaimer in fact. It's not a right or wrong scenario, it's about figuring out what's going on so having a vested interest in either approach is just clouding the matter with personal bias anyway. I was asking simply to find out what the application of this information was intended to be- if you're afraid you are hybridizing? Are you trying to decide whether or not these are distinct species? Are you trying to figure out exactly how many forms there are and if there are intermediates? Because the application will change how you approach the problem and to what extent it is worth investigating. If the point is to consistently re-create a certain "look"- breed line mentioned previously- for this animal then who cares what alleles are involved as long you figure out a way to reproduce that look. But if you want to know about the genetic distinction between the two than using genetic data would obviously be the best, probably the only, way to tackle it.

And, I use the little graphics because people are graphics orientated. Looking at, much less understanding, a seemingly endless written list of letters and numbers is absolutely guaranteed to lose your audience, unless they make 6 figure+ salaries as accountants or computer programmers by doing so

I don't think you give the folks on this board enough credit, there are lots of highly intelligent and motivated people that aren't necessarily working in some hi-brow facility. Maybe because they are not graphics-orientated but rather knowledge-oriented... *clears throat*... and so will parse through page after page of discussion because it's interesting and satisfying to have a better handle on some issue we are curious about- some times we crave more than an over-simplified book with glossy pictures. Since when is honest discussion and curiosity only for the six-figured and since when are the six-figured excluded? Knowledge isn't a lofty pursuit despite popular (mis)portrayal.

As far as the multiple allele issue goes, it's not just medicine- I get that the punnet system is demonstrative and used to roughly pick out frequency if you were breeding or know the genotypes of the parents (for inheritable genetic defects, for example) and need to know offspring potential, but for evolutionary biologists it's going to be the molecular data that is the holy grail, no cutting that any other way. So much of what we thought by reverse-engineering via morphology, ESPECIALLY with invertebrates, has been re-organized and corrected since the advent of better sequencing technologies that it's going to require molecular data to be convincing at all. That's precisely why I pointed it out- if you are really trying to figure this out you need to know more about the allele set in question. There are lots of examples of poly-allelomorphic loci, I just used blood type because almost everyone, even the graphics-orientated, is familiar with it. Full of complicated inheritance patterns even though the morphology would appear to come from a much simpler set- only four phenotypes, right?- or sometimes just the opposite where a seemingly complicated set of forms comes from the simple presence or absence of some tangent gene (epistasis) or by simple mutation, like retrotransposons which account for a very large amount of our own variation and have nothing to do with any kind of the canonical environmental pressures. Immune systems for most organisms have many alleles for a single loci. How about sexual dimorphism? The path of least resistance to our mind is not always the path of least resistance to biology over time. At any rate, saying that this one way is more common than this other still doesn't answer the question- which path does the rosea/porteri/whatever follow? Nor would it convince an evolutionary biologist at all. All they want these days is genetic material for analysis. Mind you there are pitfalls there too of course, but they're certainly not going to get out a pen and paper and begin drawing up 10th grade biology punnett squares to start cladistic analysis.

I do understand the importance and function of the Xsquare, I just don't really understand what the application of the discussion is which would change what to stress and what not to. If this is about avoiding accidental hybrids then probably breed strictly same with same and you avoid in all likelihood, but if it's being used to try to validate the claim of a new species porteri v. rosea then it's not really a reasonable way to tackle the question- it is a bad test if you are looking for those kinds of answers. You have to design a different kind of test to determine that beyond just counting offspring forms that returns meaningful data. That's where the molecular data is paramount.

It is really interesting though one way or the other, just throwing in some more information to chew on because what study of any quality has ever been done without at least considering as many possibilities as you can. Plus I have a new found love of rosies- where did that crazy white one come from? It's very impressive. Anyone know why there is so little work being done on the professional level in terms of taxonomy? Lack of funding? Seems to me there is ample interest so that must be a part of it.

 

[somethingbig]

I think the only thing left to benefit this conversation is some input from people that have bred:
RCF x RCF
PCF x PCF
and
NCF x NCF

I guess RCF x NCF would help too seeing as 100% of the offspring should be PCF according to the hypothesis.

So no one has ever bred one of the above combinations? I find that hard to believe...

 

[Stan Schultz]

So no one has ever bred one of the above combinations? I find that hard to believe...

We don't know! No one has yet stepped forward and volunteered the information.

[size=+1]COME ON PEOPLE!

WE NEED BREEDING REPORTS WITH INFORMATION ABOUT WHAT THE PARENTS, BABIES, AND SPIDERLINGS LOOKED LIKE AS THEY GREW!

WE NEED PHOTOS OF THE PARENTS, BABIES AND SPIDERLINGS TOO, IF POSSIBLE!

"ENSIGN: REPORT!" (Capt. Jean-Luc Pickard, "Star-Tek: The Next Generation")[/size]​

Send them to me directly. E-mail address is in my sig, below.