Thanks Micheal: I think about 95% of the talk about the differences in tone woods is simply to generate sales. I have a sneaking suspicion that the other 5% has more to do with density than anything else. Basically, it seems that you want a low density softwood for the top, and a higher density wood for the B&S.
With top woods it's easy to see why you'd want to go that way: it's about stiffness and weight. There is not much horsepower in a plucked string, so you can't drive anything very heavy with one and get any sort of treble response or sound power out. The thing that limits how thin and light you can make the top is stiffness. Any wood top that is stiff enough to resist folding up for a reasonable length of time will be strong enough to hold up under the string tension. Stiffness in a piece of wood is determined by the thickness, and the Young's modulus along the grain. As it happens, all softwoods have very similar structure on a microscopic scale, so the Young's modulus along the grain is primarily determined by the density of the wood. If you measure the Young's modulus along the grain and density of a lot samples of softwood and plot the results out you'll find that about 2/3 of them will fall within 10% of the same line. Thus density is not a totally reliable predictor of Young's modulus in softwoods, but it's pretty good.
For two pieces of material of the size and Young's modulus the stiffness will be a function of the cube of the thickness. We all know this intuitively: a 2x6 up on it's edge deflects a lot less under a given load than a than a 2x4 on it's edge. A 2x8 deflects about 1/8 as much as a 2x4, all else equal.
What this means in terms of making guitar tops is that you'll usually end up with a lighter top if you start with a low density piece with a low Young's modulus, and leave it enough thicker to get the stiffness you want. This is one of the big advantage WRC: it's generally lower in density than most of the spruces on average. Of course, there's a lot of variation within any species, and some pieces of wood are just stiffer than they 'should' be for their density. One key there seems to be the proportion in width of soft early wood lines and the harder late wood. Late wood adds more mass than it does stiffness, so dark heavy late wood lines tend to make a piece that's denser than the Young's modulus would predict.
The back has a different job then the front in general. The top is the part that is driven by the strings, and produces most of the sound, either directly, acting as a loudspeaker, or indirectly by 'pumping' air through the sound hole. The back 'steals' all of the energy it gets to vibrate with from the top. It's also facing the wrong direction for the audience to hear, and it's often up against somebody's pudgy aviordupois, so it doesn't have much of a chance to make usable sound. What is does seem to do is contribute 'tone color', but in doing so it also reduces the overall output in general. There are always exceptions, of course; very light backs, such as on Flamenco guitars, do seem to help in producing a 'punchy' sound, particularly in the low frequency range. In fact, if the back does contribute to overall power it's in that low range. Ther is always a debate among guitar makers about whether the back should be 'reactive' or 'reflective'. I'm squarely in the camp that says "Yes!": they should be reactive in the low range and reflective in the higher range. This argues that the back should be fairly heavy, and 'tuned' in some way to work with the top in the lowest range.
All of this ignores or glosses over a lot of subtle nuance. Many makers (most!) prize wood with low 'damping', particularly for the back. Wood like BRW, with low damping, tends to 'ring' for along time when it's tapped, and have a 'musical' sound because it vibrates within a very narrow frequency range. This may help in limiting the amount of energy the back 'steals' as it vibrates, but it's hard to say. There is so much going on n that range of tone that the number of variables you'd have to control to be able to say anything definite is just too great.
In terms of this conversation, and trying to answer a specific question, what that means is that denser 'rock' maple will tend to sound a bit like soft maple, but with more of a tendency toward the 'rosewood' end of things. In my experience, Black Locust, which tends to have similar stiffness to Indian rosewood, but lower density, tends to sound a bit more 'Flamenco' than a rosewood guitar, but how much of that is due to the fact that the color of locust is similar to that of cypress is hard to say. Lots of things to think about...