I find this thread interesting and also confusing. Apparently Bill's "problem" is that there is a rather small but steady drift between each image, not enough to make the stars elongated, so it is not really a problem. Still it is interesting and I see the same thing from time to time but rarely worry about it.
I assume the argument for differential flexure is that gravitation and the movement of the mount slowly pulls the guidescope in one and the same direction during the session (since all movement between subs is in the same direction).
Could not also poor polar alignment be the cause, or will that cause field rotation rather than a steady movement?
What about a poorly leveled mount?
In any case, Olly has a good point about guide scope rings. I never had any and always bolt my guidescope as steady as possible to the main scope (or use OAG when possible, which is on everything except my RASA). Why do they sell those rings?