Wooly Mammoth build mistake sounds good - explanation?

[RandomGlitch]

Hi, I've just built a Wooly Mammoth clone from the following schematic and vero layout

http://analogguru.an.ohost.de/193/schematics/Zvex_WoollyMammoth.gif
http://tagboardeffects.blogspot.com.au/2011/02/zvex-woolly-mammoth-vero.html

When I first tested it, I had neglected to connect the "Wool" pot.  I still got a distorted sound, but there was an interesting "vocal" or "talking" type tone to the notes, a little like a very subtle auto-wah effect.

I'm going to add a switch to my clone to get at this sound. 

Can anyone offer me an explanation as to why this changed the sound so, and if there might be any other mods that I could do to further exploit it?

Many thanks!

[Mike Burgundy]

P2 sets the gain of Q2. It doesn't affect Q2's bias (as in a fuzzface) through R1, but it does max out AC gain when the pot is set to 0 Ohms, shorting AC to ground through C6. Removing this pot maxes out AC feedback from Q2's emitter to Q1's base.

[RandomGlitch]

Thanks Mike, for that explanation!

[Mike Burgundy]

...and for the sake of completeness:
emitter resistors (unbypassed by a cap) are also a way of having feedback for a transistor stage. If you bypass that with a cap large enough to pass all signal, you've eliminated that feedback (there's tone shaping possibilities here using caps that only pass *some* of the audio, providing negative feedback for lower frequencies). All of this is very much the same as cathode resistors for triodes.
So Q2's local feedback is also maxed, as well as the feedback from Q2 to Q1.

[RandomGlitch]

Thabks Mike awesome explanation!!