Over and Under Biasing as means for tone control

Well in consumer-land, tone control refers to bass and treble balance controls. I'm not sure bias will affect  this in that sense.

Under-biasing risks excessive distortion, over bias risks excessive heat and shortened (possibly VERY short) tube life.

Harmonics control may I say?
if you are talking about power tubes I highly discourage overbiasing to warm up
the sound.  It just gives you supernova and runaway tubes easily. 
Gain and driver tubes are a different story.  You can get pretty good returns from say biasing a 6SN7 from say 2mA to 4mA depending on B+ and plate resistor.  However since a lot of modern designs are direct coupled.  Increasing bias at gain stage by tweaking grid voltage or cathode resistor usually means decreasing bias at driver stage.  Not for the faint of heart.  

I read somewhere that under biasing tbe Joule 100 takes a bit of sparkle off, which might be desiarable with some speakers or source players. 

I use this to make sure bias is exactly what mfg. recommends

http://www.vhtamp.com/products/vht-accessories/tube-tester-2

Worth every penny, helps also rule out a bad tube or cathode resistor.  Not a tube tester but very useful .

That’s only a for power tube biasing.  Also won’t work for the likes of 6C33B. 

Using bias as a tone control is a fool's errand.  What people are calling bias is actually idle current measured across a 1 ohm (typically) resistor.  Bias is an negative voltage on the grid and varies by tube type and circuit. It runs between negative 10 and negative 80 volts. 

If you want more bass get and 30 band analog Equalizer.  Much better tool than bias. 

@dumbeat

I read somewhere that under biasing tbe Joule 100 takes a bit of sparkle off, which might be desiarable with some speakers or source players.

I have measured many amps of that design. I will say that the output impedance does rise as idle current is reduced (underbiasing being the popular but sadly incorrect term). When the output impedance rises bass will rise. Also the low impedance parts of the speaker's response curve will diminish in level which may explain the loss of highs. 

I have recently been working on a similar OTL from another maker and have noticed that idle current can affect the output impedance by 2x, and its already high. 

The problem with low damping (high output impedance) is that the speaker's response will follow the impedance curve in a way the speaker maker did not intend.

For more on this join the discussion please see:  Does anyone care to ask an amplifier designer a technical question? My door is open.

I replaced tubes in my push-pull monoblocs with a different type, and in an elder moment forgot to re-bias. Everything still sounded great. It was only a week later, when I upgraded a tonearm, that I noticed a slight but very discernible upper bass/lower midrange suckout of probably 3dB. Increasing bias until the tell-tale blue ionisation appeared in the power tubes fixed that instantly.