It's much more complicated when ya start adding extra genes in there, even if they're all co-dom.
The key problem in your logic, there, is the "or." None of the co-dom genes listed here are allelic, so you have a 1 in 2 chance of getting a Pastel, a 1 in 2 chance of getting a Pin,
and a 1 in 2 chance of getting a Butter. So if you look at the list of possibles I put up there, you'll see that we do get 4 out of 8 with the Pastel gene, 4 out of 8 with the Pin gene, and 4 out of 8 with the Butter gene.
For the Punnett square itself, the Pastel Butter Pin (the way I do it, anyway) can look like this:
PpBbNn
Pp = Pastel
Bb = Butter
Nn = Pin
This snake can only pass on one of each pair of genes, so the possible combinations are as follows:
PBN, PBn, PbN, Pbn, pBN, pBn, pbN, pbn
The normal, to keep from getting too confused, would look like this:
ppbbnn (no Butter, Pastel, or Pin genes),
and so passes on only
pbn.
Now, technically, it passes on "pbn" eight times since that's the total possible combinations for three pairs of genes (2 genes * 2 genes * 2 genes = 8), but since they're all the same, we don't really have to worry about the others.
So the square will look like this:
____
PBN PBn PbN Pbn pBN pBn pbN pbn___
pbn |
If you fill in the chart, you get the following combinations:
PpBbNn = Pastel Butter Pin (1 out of 8)
PpBbnn = Pastel Butter (1 out of 8)
PpbbNn = Pastel Pin (1 out of 8)
Ppbbnn = Pastel (1 out of 8)
ppBbNn = Butter Pin (1 out of 8)
ppBbnn = Butter (1 out of 8)
ppbbNn = Pin (1 out of 8)
ppbbnn = Normal (1 out of 8)
I hope that helps!