I've measured the run-out at the tool-tip and I'm getting 0.2mm(

Here's the thing, though - the spindle is quite rigid in the bearings - if I apply force in any axis there's no real play. As I revolve the spindle by hand, the "high" point on the DTI is coincident with the rotational angle of the spindle, and this translates to the offset at the tool-tip.
[Edit]... and for completeness, dialing the spindle above the bearing, the high point is 180 degrees opposed, so it is rotating around the bearing mount. [/Edit]
Now, my brain isn't geared for analysing three dimensions, but it would appear to me that rather than being necessarily a bearing issue, this is more likely an issue with how the spindle is sitting in the bearings (e.g. conceptually - a bit of swarf between two machined surfaces). Is this likely?, or am I relieving myself into the wind? (

Any suggestions of where to start? (I'm minded to remove the lock-nut on top of the spindle pulley and see where that takes me - is that a silly idea?)
Any advice gratefully received.
Mike