Originally posted by virtuallyhorses
I would correct the person who said that dorsal stripes are not a dun thing - indeed they are! when you see them in other colours they are the evidence (usually) of a recessive dun gene for instance EeAaCCDnndDnnd would be a bay horse and may or may not show any traits of the recessive Dun gene. The exception is a foal which starts out with a dorsal stripe which then fades.
But then we are faced with all the duns who are genetically dun, but show no dorsal stripe, along with the 'layman's' dorsal stripe on non-duns, which is in fact a form of counter-shading caused by the sooty gene (probably among others).
A clear, sharp, true dorsal stripe is certainly a dun characteristic - but there are also 'pseudo' dorsal stripes that really just confuse the whole issue.
One of the things that really fascinates me is the dun gene. In spite of the fact that there is, almost without doubt, a simple dominant dun gene, it's also been shown that horses left to breed for several generations without human intervention 'develop' dun colouring. That would support the theory of a reccessive form of dun. Dun as described here is described in the simple dominant form though - mainly to try and keep it simple. In order to register a horse as dun it must show the characteristics of the dominant form of dun - including the need for at least one dun parent.
ShinyMcShine - the answer to that one really depends on which theory you subscribe to. It has been shown in research in France that seal brown actually carry Agouti, but are not considered bay because two Seal browns will never produce a bay foal. Current thinking is that there are in fact four alleles at the Agouti locus instead of simply A or a for bay or not bay.
Thsi would mean that at the Agouti locus we now have - A (bay), a (black), A+ (Wild-type bay - where the black on the legs is mixed with red, so points appear paler - may also be linked with the lightening of flanks and muzzle which may also be caused by or in part by Pangare or Mealy gene) and a^t - Seal brown, where Agouti partially restricts black.
Agouti, because it affects black pigment, doesn't influence chestnut. Chestnut can, however, be influenced by other genes - Pangare, Mealy and Sooty among them. It's possible that liver chestnuts are red base with sooty modifier - again, it doesn't satisfactorily explain all the variations. I have 3 liver mares. One is a dark reddish liver, with mane and tail the same liver colour. One has darkened over the years and now has mane and legs so dark they can easily be mistaken for black at times. The third is a gingery liver with a grey mane and tail. It would seem there is more than one gene at work to cause the variations in shade.