DSJR,

When used passively, typically two notch filters are employed to flatten the midrange response, and often a third to suppress the upper resonance. The upper resonance is dangerously close to the pass band IME at 4.5khz. Without suppression this will be audible, even when using fourth order slopes. Ideally the unit should be linear at least an octave above the pass band.

I wouldn't have though directionality has anything to do with the crossover point, the dome only becoming directional at circa 4khz, the 15" bass driver at around 800hz. The 9" cone in the 50's will be directional earlier of course, yet the same crossover topology is used. It will have no affect at all on the midrange dispersion from what I can see. I suspect that the bass unit throws up spurious 2nd order harmonics much above 400hz, and that is the reason for adopting the low crossover point. Certainly when crossed over at 500hz and properly EQ'ed the dome has none of the charaterisitcs detractors find off putting in the ATC monitors.