Brian: I haven't explored the low-disk-loading regime methodically. All I have is anecdotes.
My 'Bee has 24.5-foot Rotordynes and an ultra-stripped configuration. I weigh 180 or less. The result is a disk loading of about 1.0. The machine flew fine, and climbed reasonably well with its 447 and wood prop. It would do over 70 mph flat out, but was happier ghosting along at 40.
By "climbed reasonably well," I mean in comparison to my 1980's 447 Air Command, which had a rate of climb down in the 200 FPM range in the summer. I found that rather scary (the saving grace was that we don't have much summer here in VT). The Air Command was heavier than the 'Bee and used 23-foot McCutchens.
I'm not certain what people believe is the danger of moderately low disk loading in gyros. Obviously, the secret of a gyro's "stall-proofness" is the fact that the wings (blades) travel at a tip speed of 3, 4, 5 or more times the aircraft's forward speed. As you slow the rotor down with lower disk loading, this margin gets smaller... but not radically smaller. The ratio of tip speed to forward speed is still on the order of 3:1.
As this ratio (known as the mu ratio) gets smaller thanks to low disk loading, you encounter more and more stalling of the inner portion of the rotor at high cruise speeds. I SPECULATE that this does not become catastrophic, however, with disk loadings in the 1 lb./sq.ft. level. Rather, the rotor flaps back more and becomes increasingly draggy at higher forward speeds until you run out of power. Again, this is speculation based on limited experience in just two gyros with low disk loading -- my 'Bee and my tandem Dominator.
The latter gyro actually had a rotor tach. Solo, it had a disk loading of about 1.15 and flew very nicely at 100 mph, at mu= 2.9 or so. Interestingly, though, it, too, used least power cruising at about 45, solo.
NACA gathered some data on these topics and published them in their 1930's reports about autogyro rotors. IIR, they found that mu= 0.4 was the most efficient ratio.
Chuck Beaty reported experimenting with a very slow rotor (like 220 RPM) on a light Bensen-style gyro. He found that the gyro would barely move forward, though it did leave the ground. Most likely, the large stalled region of the rotor acted as a drag brake.