Heh, ok, this post is way too long, but...
How do we prove/disprove cube as an allele at the motley locus, or, prove that cube is a selectively-bred motley and/or stripe?
IMO the first necessary step is to come up with a clearly-defined distinction between cubed and striped and motley. In order to attempt to separate it, we have to be able to say what the parents/offspring are, phenotypically. (If not, then it really doesn't matter if there's a cubed allele.
)
For example, we can clearly identify motley from striped phenotypes, by following a definition that the striped corns have the "4-lined" and "wide-dorsal-stripe" characteristics where motleys do not. This gives us a basis for making the IDs on the parents and offspring, so we can say "stripe is an allele to motley, and it's recessive to motley, and here's how it will behave" and match up our predictions with the actual results. (So far, they are doing a pretty good job of matching up.)
As another example, we have ultra and amel. This is much easier in one way, as we can clearly tell an amelanistic from an ultra/ultramel. There may be some trickiness with picking out "uu" from "au," but we still make very useful predictions and understand what a clutch will look like.
I guess what it boils down to is "what is the point of calling it an allele, instead of just thinking of it as a stripe or motley corn?" IMO it's because we want to be able to predict what results will happen if a "cube" allele really does exist and causes distinctive results.
So, the biggest question to start off with is, "where do we draw the line between cubed and striped?" From there, we can make test crosses, and start gathering the test cross results. If the results seem screwy then we might need to adjust our ID methods.
(Like they started to with motley X stripe crosses, "hey, wait a minute, these 'striped motleys' are not het for stripe at all... stripe might be recessive to motley instead!")
So, it looks like Joe has already done some crosses and we can start with those and go from there. We can start out with his ID method:
I basically considered any Cubed that had some striping as a Striped/Cubed and the Cubed/Cubes to have no striping at all. Sounds good to me. We can adjust it if it seems to need adjustment in the future.
So we'll assume that:
Cube/Cube = 100% cubed pattern (maybe some connection on the first 2-3 saddles?)
Cube/Stripe = Striping and Cubing mixed together
Cube/Motley = Cubes and Circles, like the above with the oval "sunspot" marking
Joe's results:
Cross 1:
Cube/Cube X Cube/Cube =
· 100% Cube/Cube.
Cross 2:
Stripe/Cube X Stripe/Cube =
· Stripe/Stripe
· Cube/Cube
· Stripe/Cube
(Did you collect a count of each type, by chance?
)
Cross 3:
Stripe/Cube X Cube/Cube = "no perfect Stripers." Does this translate to:
· Stripe/Cube
· Cube/Cube
(If all of the above, it really starts to lean toward an allele hypothesis. Also, do you have numbers for each type in this cross, too?)
The biggest hurdle is eliminating "selective breeding" from the equation though. The above is still possible (seems kinda unlikely, but definitely an easy possibility IMO) under the idea that "cube is just a selectively bred stripe." So we could dilute the selective breeding out, by crossing to generic normals that wouldn't be likely to add "just the right genes" into the mix and drop some of the "cubing" influences out of the picture...
I'd think the one cross to do, in order to really prove it out well, would be:
A- Cube/Cube X Normal --> F1(A) Normals het cube.
B- Cube/Cube X Normal --> F1(B) Normals het cube.
Where all four parents are as unrelated as possible, and the two normals are not all funky patterned. (Like don't use zigzags or aztecs.)
From the F1, choose the hatchlings with the most
normal-looking patterns.
(Do not pick ones that might "look like they're het for cubed" otherwise you might be continuing to select for cubing!)
Then cross:
F1(A) X F1(B) and expect --> Normals and cubes, and no motleys or stripes.
If that happens, it would be a really good indication that cube is an allele.
If instead you get stripes and cubes, or any kind of "mixture" (motley patterns, or stripey patterns, or half-cubey/half-stripey patterns) going on, it's probably a much safer bet (at that point) to say that cubes are selectively-bred stripes or stripe/motleys.
