Yes I feel this point is worth highlighting. I may be wrong(!) but it appears we are confusing gene pairs at a single location with those that occupy different locations. If my understanding is correct, the breast colours purple, lilac, and white are affected by a gene pair at a single location in the genome and each of the two genes in that gene pair at that location is tri-state, i.e. the single gene can take on either purple, or lilac, or white, and thus, as a pair, the gene pair can take on P-P, P-L, P-W, L-L, L-W, W-W. Because this is a gene pair it cannot be "triple split", i.e. it cannot be P-L-W (that would require a gene triple, not a gene pair).
I think the reference to fawn and cinnamon in Star finches is a different situation (in my humble opinion, I may be wrong) because my understanding is that those two mutations occupy different locations in the genome. Although they affect the same area of the bird (the overall body colour) they are different locations and therefore act independently. Thus it is possible to have a Star that is normal split fawn split cinnamon. They must be different locations in the genome because one is autosomal recessive and the other is sex-linked recessive.
Another good example is the Bengalese finch and its many mutations like pearl, grey, ino, etc. These are mutations located at different positions in the genome, and thus a single Bengal can be split to all these mutations. This is not the case for Gouldian breast colour, because there we are talking about a single location in the genome.
Tiaris, did David specifically say that the Gouldian breast gene can be "triple-split"? Could it be that David's comment regarding Gouldian orange head working the same way as breast colour was that they are both autosomal? It would be interesting to hear David's reply when asked whether Gouldian breast colour can be triple split.So it is possible for a purple breasted bird (or 2 of them who are mated together) to carry both the white-breasted & lilac-breasted mutations (be split for both).
There is a huge amount of information on Gouldian breast colour on the web and reference texts, but from what I have read no one has ever suggested that Gouldians can be triple-split to the three colours simultaneously (a genetic impossibility because it is a gene pair).
Marek Buranský knows more than most about Gouldian genetics, and his forecaster does not list any triple-splits for breast colour.
As I have stated before I believe the most likely reason for the two breast colours of Alvin's offspring is that the hen as mated with another cock. Another possibility, perhaps less likely, is that a different hen laid an egg in the same nest. I have recently witnessed both these situations in a community cage; I've seen a hen paired to cock "A" mate to cock "B", and I have another hen that lays eggs randomly in various nests, including a nest that was occupied by another brooding pair. Alvin, do you discount these two possibilities, and if so,can you give a reason why? I understand that you believe the cock is the parent of those two offspring, but it must be possible that he is not the biological parent, surely? And wouldn't either of the above two explanations be far more likely than the possibly that your cock has a very rare purple breast that looks lilac?
Now if "triple splitting" is possible for Gouldian breast colour then please ignore all the above
By the way Craig, this site is in agreement with you when you said you believe "dirty white" and lilac are not related:
The above is all in my humble opinion and mentioned in the interests of robust discussion. I am more than happy to be corrected as that is a necessary part of learning.
Cheers.
Ross