Your saying texas dont fade but when breeders cross texas to midas they get faders so what your saying is wrong. Texas obviously dont need the fade gene if breeder cross red/ midas, blood parrot, and king kong parrots to texas and they get faders the first generation so the statement about texas needing the fade gene is wrong and texas may infact already cause fading on their own.
Firstly, I never said that the fading gene(s) was recessive. It could be dominant, and hence easily show in F1. It could also be controlled by multiple genes, hence showing in some and not in others.
Secondly, a recessive gene can actually show in a hybrid. The reason is because when crossing different species, there can be locations on the chromosome where a gene is present in one species and not in the other. So in a hybrid, even though the gene may be recessive on one chromosome, the fact that there's no equivalent in the corresponding chromosome means that the gene would not be masked anyways, and hence expressed.
The thing is with the 'pink/white/albino/leucistic/blah' gene, it would seem to be related to the presence/absence of body color as a whole, so the texan body color will probably mask the 'absence' gene.
Lemme explain Convict genetics a bit:
The reason why the leucistic gene is recessive is because it's actually just producing a mutant version of the pigment, which does not actually show up as any sort of color. That's why when you have a Convict with the stripey allele and the leucistic allele, stripes are still produced, because whilst the leucistic allele produces mutated pigments that doesn't show up, the stripey allele still produces the normal amount of pigments, and hence the heterozygous Convict is still stripey. But when it has two version of the leucistic allele, all it is producing is the mutated pigment, which does not show up as stripes.
When you cross a texas and a pink convict, well, to be honest it doesn't really matter if the leucistic gene is located on the same locus as whatever is producing the texan coloration - even if it is on a separate locus, the texan will still have one allele producing its normal body color, and so that's why I would assume that the texan x pink convict will still have a normal body color. Simply because stripes, pearlings, body color and all that are one sort of trait - coloration. I'm not gonna say that they're definitely on the same loci or not (though I'm quite sure they're not).
In the case of the fading gene(s), it is a separate type of trait, it's the control of coloration. So when it's passed down, chances are there's nothing on the corresponding's chromosome's loci to act against it, so recessive or dominant or otherwise, it'll get expressed. Why there's different level of expressions though, and this is what I'm pretty sure about, is that it's either a trait controlled by multiple genes, or a gene but with multiple regulatory elements. When a parent passes down their half of their genome, recombination occurs before that, and any combination could be passed on to the child, so some may have er, 'more fading' in them, whilst other not.
Basically imagine ABCDEFGHIJ as genes and/or regulatory elements affecting fading. In say, a red devil, it may have ABCDEFGHIJ on one chromosome and ABCdefgHij on the other, and let say the more 'capitals' it has, the 'better' it fades. So because of recombination, some of the offsprings would receive say, ABCDEFGHij, whilst some may receive ABCdeFgHiJ for example, and therefore the latter will have 'worse' fading than the former.
Etc. etc.
[EDIT]
P.S. It's not THAT hard for recessive genes to arise from dominant genes. A lot of times it may take just a single point mutation to mess up a gene (and its product) completely, and therefore produce a recessive version of the gene.