I'm going to take a guess, and suggest that if you intentionally crossed these with a variety that never shows variegation (say, a normal Samsun), only the female variegated parents (made by removing its anthers, providing pollen from the male parent, then sealing the blossom) would provide you with the 60% variegated seedlings, and that female normal variety parents would yield 0% variegated seedlings. That is, the pollen from the variegated plant would not yield any variegated seedlings when pollinating a normal Samsun female.
In keeping with Jessica's comments, the defect may be passed only in the cytoplasm (which always comes from the female parent), rather from the nucleus. So it doesn't follow basic Mendelian inheritance patterns.
I think it would be a bit tedious to actually do the crossing experiment on enough specimens to give meaningful numbers.
Bob