Nick:
So is the Cof G of 59% the determining factor for the MMO or is it some other factor?
The top of the boundary is FL600, largely an artificial number - the airframe is good for rather higher than this, but I believe air supply and ramp scheduling could become an issue not so far above this level.
Mmo - ditto. As others have said, Mmo was originally going to be higher (M2.2) but was reduced to extend fatigue life as the aircraft design 'grew'.
The significance of the shaded 59% portion of the graph is that it shows the envelope at that CG - in this case the relevant line is the bottom of the shaded area - M1.56. This is the MINIMUM mach number that can be flown with the CG at 59% (normal for supersonic cruise). You will see it represented on the Machmeter (a few pages back) as the "AFT" bug. i.e. you can't fly slower than this without moving the CG forward.
So it can be seen that the decel must be done in concert with CG transfer - and as (mostly) always the designers had made it as straightforward as possible. Transferring forward from Tank 11 using the two electric pumps the rate of txfr pretty well matched the standard decel profile, leaving the FE to make the occasional tweak to keep the flight envelope in concert with the CG envelope through the decel/descent.
In the case of abnormal procedures depriving one of electrical power then some other way had to be found to enable a descent (which required a decel) and that is why there are also two hydraulically driven fuel transfer pumps in tank 11.
It's a bit confusing at first, but there are two overlapping flight envelopes - the speeds/alts drawn on the basic envelope and those determined by the CG postion at the time.
In practice - one had a takeoff CG, a landing CG, a subsonic crz CG, a supersonic cruise CG and the only area one had to keep a close eye on was the transition between the last two. There were several visual and aural warnings to back up the CG and Machmeter bugs.