BlackSails wrote:I would guess that it went RNA->DNA->Coiled DNA, but I cant see any sort of useful intermediate between coiled and uncoiled.
Well, the transition from RNA to DNA most likely overlapped with the decrease in RNA acting as ribozymes and increase in them acting as code templates (or genes or cistrons, if you prefer) for proteins and enzymes which helped maintain their living(nearly living?) systems. Since DNA is more stable than RNA, it made a better information storage molecule. The replicators could stop doing their own work and just hold the code for things to do their work for them.
While making this transition, it's not hard to imagine selection favoring more compact, stable structures for the DNA. Aside from holding the genetic code, DNA doesn't have much of a function (though selfish gene theory and natural selection dictates that any DNA, even those bits without apparent function, which can manage to get replicated into following generations will grow and propagate in a population). It just sits there holding the instructions for everything else. A tightly bundled structure is smaller, easier to store, and easier to protect from would-be harmful chemical attackers. An uncoiled plain bit of "naked" DNA runs the risk of being broken down, deformed, co-opted, and so on. Of course, you have to pay for all of this with the cost of working with genes stored this way, but apparently the benefit outweighed the cost.
Stepwise, even a small amount of coiling around even an inefficient histone precursor would provide some advantage, simply by giving the DNA that much more protection. Little by little over the generations, the mechanisms and pieces for coiling and uncoiling grow more efficient by natural selection.
This is one possibility (and to me, the simplest).