Indeed, it's not really obvious what's going on, so I shall explain more.
To show the twist on the stave I've taped on some short lengths of thin cane (from an old roll down window blind, it's useful stuff).
With twist like that, you don't really know how it will react when the bow is being drawn back, so I need to get it on the tiller fairly soon.
I've previously had staves which looked perfectly straight but insisted on twisting when drawn.
On this one I decided to try to follow the shape of the surface of the log as much as possible. Some of the twist is due to my laying out the bow at an angle along the stave to avoid the two knots E.G. The bow sort of spirals around the stave,( but only by a matter of a few degrees).
The other factor is the asymmetry of the log. The central pith of the log was very close to one edge, where the grain is very dense, this can bee seen a few posts back, (There's a picture of a polished cross section of the log on the Yew Longbows page of my website). The effect of this is that on one edge of the bow the sapwood may be 6mm thick, but on the other edge it may be half that thickness.
The second pic shows the same limb tip viewed from either edge, you can see how although the limb is about the same thickness either side the heart/sap thickness is different. You can also see I've used a temporary tie on nock (this technique is used on some native American bows I believe) this is to avoid cutting grooves into the tips, as I want to keep as much wood as possible until I know exactly where the string will need to line up..
As I'm roughing out the limbs I'm trying to make the thickness along each edge about the same, whilst retaining a few mm of heart wood showing on the edge.
Now it wouldn't necessarily matter if the sapwood was breaking through onto the belly of the bow. I've seen several primitives with an edge of sapwood showing along either side of the belly, but these were from narrow saplings and were symmetrical, I've also seen some beautiful primitives with huge degrees of twist (but I suppose these would be seen as character bows).
Taping the spills of cane to the bow has done me a favour as it's shown that most of the twist is on one limb, (the 4 spills nearest the camera are virtually parallel).
Once it's on the tiller and being braced and draw the wood will settle down a bit and will find it's own axis to bend along. this may mean the grip rotates a bit from how it is at the moment. The important thing is that once it's finished it doesn't rotate during the draw. (Taping a spill to the grip and watching as it's winched back will test this)
This is a good reason for not actually clamping a bow during tillering, but allowing it to move freely so you can see what it's doing. Sometimes it's best to let the bow move how it wants and build it to blend in with that rather than trying to fight it.
We saw what happened on the last bow where I tried too hard to shift it to what I wanted.
I've just added that second pic, having done a bit more, it's up on the tiller now and is just flexing a bit at about 30 pounds with out any nasty tendency to twist.
I shall now check carefully over the limbs taking measurements and getting a nice even thickness taper, then I'll start serious tillering.
Hopefully it will all turn out fine, heat bending will be a last resort if there are major sting alignment issues.