Despite what people say:
bamboo does stretch in tension on the back of a bow... Just test it.
The bow index of the wood database only indicates that the "good woods" have a low MOE and a high MOR for its density. And preferrably both in tension and compression.
In an ideal world, we want bow woods with high MOE and high MOR: stiff woods that still bend very far. Bamboo species are in that category (have data for Moso and Guadua bamboo to back this statement), at least in tension. They are crap in compression (unless heavily toasted, and with the dense outer layer on the belly too, like in split cane fishing rods).
For real wood species, there's a trade-off between stiffness (MOE) and elasticity (MOR), the combination of which determines how much energy a wood can store in bending. Basically, there's pretty little variation among energy storage capacity among all wood species, because of this trade-off. That brings us back to the findings of Comstock and Baker in their archery books: any wood allows you to make a bow, you just have to adapt the design to the wood properties and your own (MC) conditions.
Low-stiffness woods (relative to their density!) are just easier to turn into bows (more tolerant to tiny mistakes) and are likely also more stable in their tiller towards MC changes.
So that's why we use bamboo as a backing. Ipe is very dense and stiff wood, with just above average compression properties for its density. This makes it a good partner for bamboo. Osage and yew are less stiff, but better in compression. Net energy storage capacity will be similar, but requiring a different design.
J
