Let's assume that the colour of South Seas is based on the red that it inherited from 'Chicago Apache' and the cream or light yellow that it inherited from tet 'Lullaby Baby' producing an orangey colour in 'South Seas'.
Selfing 'South Seas' can then theoretically produce reddish, orangey and light yellowy/creamy flowered seedlings. The colours possible include all the in-between colours from reddish through orangey to light yellow/creamy in potentially a wide range of shades. Some might be eyed and some not obviously eyed.
On the other extreme, there is a genetic effect that can occur in certain plants, called 'fixed heterozygosity'. If a plant shows this effect when a red is crossed with a yellow and produces orange seedlings then when those seedlings are self-pollinated all the resulting plants will be orange more or less like their parent.
Genetically the red tetraploid parent could be R1/R1/R1/R1 where R1 means the gene produces red pigment. The light yellow or cream parent could be R2/R2/R2/R2 where R2 means the gene is not able to produce the red pigment. The plant from the cross (the equivalent of 'South Seas') would be R1/R1/R2/R2 and orangey. When that plant produces pollen (or ovules) the expectation is that it can produce R1R1, R1R2 and R2R2 pollen but when fixed heterozygosity is present it only produces R1/R2 pollen/ovules (or produces R1/R2 in far far greater abundance than it should). It is not known whether tetraploid daylilies suffer from fixed heterozygosity or whether some tetraploid cultivars do and others don't.
Secondly, tetraploid ratios, even when the plant obeys Mendel's laws perfectly for dominance and recessiveness are not as simple as diploid ratios.
In a diploid with perfect incomplete dominance (additivity) if one self-pollinates a plant from a cross of a pure-breeding red with a pure-breeding cream/yellow one expects 1/4 red, 1/2 orange and 1/4 yellow. Fully one half of the seedlings would be the extreme range of possibilities. However, in tetraploids only 1/36 would be red and 1/36 would be yellow; the other 34/36 would be intermediate type flower colours. Only 1/18 seedlings would be the extreme range of possibilities.
Those fractions are the theoretical expected fractions. To be
certain that in any group of selfed seedlings being grown one actually found a red or yellow one would probably have to grow more than one hundred seedlings.
