Here's a interesting way of looking at the chart. It lists the kick drum's range (fundamentals and harmonics) as 50hz-8khz. If you take the midpoint of that range and follow it down to the subjective descriptions near the bottom of the chart, you'll end up between the subjective "fullness/mud" and "whack" and "tinny" sounds. Many of the frequency ranges contain both positive and negative subjective narratives. Which description actually applies depends on how the instrument should actually sound (by design or tuning), or ones personal preference. The midpoint range of the kick drum actually falls in the subjective "honk" sound...but since the kick isn't a brass or a woodwind instrument this may not apply But that midpoint includes the "fullness/mud" subjective range, but does not include the "whack" or "tinny" ranges.
Hmm. This has got me thinking about stuff I haven't thought about in years. Prior to buying the Omegas I have now (the Super 3S model) I've used Fried speakers for as long as I can remember, and for a while was having regular conversations with him about this stuff. He always tended to cross the woofer over at about 100Hz and I've taken that for granted for so long that it's kinda cool to have this thread pop up and get me thinking about it again.
Bud had a few rules about this woofer-to-midrange crossover that are worth mentioning in light of this conversation and the diagram that's been posted:
1. Get it as far below 650Hz (where most musical energy resides) as practical;
2. Drivers should be capable of performing well two octaves from the crossover point; and
3. Gentle slopes (I believe around 6dB/octave) are inherently more phase-linear.
Those rules — in conjunction with his ears and the drivers he had access to) led him to choose ~100Hz as the crossover point, almost always. Anecdotally, Bud talked a lot about evaluating a speaker by ear, and using very familiar instruments like the piano or the human voice as a guide. Everyone knows what those things should sound like, he argued, and a listener is instantly aware when something is "off" about the reproduction even if we can't pinpoint what's wrong.
Now look at that chart, specifically the junction between fundamentals and harmonics for the various instruments. With just four exceptions (snare, tuba, contrabassoon & piccolo) a 100Hz crossover is at least two octaves removed from that junction point for the other instruments and also corresponds to the lower limit for human voice. I don't think that's a coincidence. So there might be an argument there for avoiding a higher crossover point. Mightn't there?