@sooby if you divide a clone of a self-incompatible daylily such as H. citrina then crossing between the two will presumably not work, at least initially. But if they were separated, say for example 50+ years ago, would you still expect the same result? I have a recollection from a plant propagation course years ago that over time even a vegetatively propagated clone can develop changes.
Yes, Sue, if you were to separate two individual ramets (one fan of leaves, its portion of the crown and its associated roots) from a clump of
H. citrina and grow them separately and repeatedly divide the two clumps down to single fans/ramets then over time somatic mutations would accumulate in each fan and the two fans would no longer be exactly identical genetically. However, there are probably about 25,000 to 40,000 different genes (plus an unknown number of other important sections of DNA) in each diploid daylily. We could not expect to have much more than one new mutation per generation or perhaps two generations. In 100 years we might expect 50 to 100 genetic differences between the two. Unfortunately most of those differences would probably be effectively "silent". That is, if we sequenced the DNA we could see the differences but if we sequenced the proteins made by the mutant DNA we would not see any differences at all quite often and at least some of the rest of the time we would see a difference but it would not be of any consequence. A proportion of the genetic differences would have visible effects. It is possible that some of those genetic differences might affect the self-incompatibility. However there are other aspects of the self-incompatibility of daylilies that would probably have stronger effects.
Stout indicated that some of the daylilies that he examined had what might be considered a typical genetic self-incompatibility system. But others had what might be considered a primitive type of self-incompatibility. In that form, an incompatible pollination appears initially to have been successful. Only after a few days is the apparently developing pod aborted (late-acting self-incompatibility system). That is similar to what happens sometimes when daylilies of two different ploidies are crossed. The exact mechanism(s) that might be involved in such primitive self-incompatibility is not yet known. It may be that such mechanisms are strongly affected by environmental factors or that they may be intrinsically unstable. In any case it is possible that incompatible pollinations may sometimes/rarely "escape" or "avoid" the incompatibility and produce one or more seeds. Perhaps daylilies may sometimes have a form of "pseudo self-compatibility".
Stout also found that incompatibility between two daylilies may only be one way. That is A x B fails but B x A works. Incompatibility between two particular different daylily individuals is a potential consequence of self-incompatibility.