I appologize in advance if this is a simple/stupid question. I am playing around a lot with jME (and love it) and with so much to learn I thought it might be fastest to just throw this question up here.
What is the difference between ModelBounds and WorldBounds? Why would I use one vs. the other, or should I set both on the objects that I add to the scene graph.
Thanks for any help!
Keep up the great work!
Never mind, I think I answered my own question… 
From what I can tell ModelBounds are used for Picking operations and WorldBounds for clipping. I understood WorldBounds, just not what I needed ModelBounds for. Now I guess I know.
Thanks anyway.
As irrisor said, this should be covered in the wiki, particularly this line:
"After the model
Your assumptions are not entirely correct.
Maybe have a look at the wiki: bounding volumes
And then feel free to ask any unanswered question here 
For even more insight into concepts used in jME you can take a look at a game engine book mentioned here (important chapter can even be downloaded).
Thanks for the help. I read the Wiki per your suggestion. I will start there next time (very nice work, btw).
I guess I just struggle to understand why, not just how. So from what I have put together, in order for the world bounds on a geometry spatial to be accurate it needs model bounds which know how to figure stuff out from the geometry. Is it any more efficient during rendering if I were to just set the world bounds? The scene that will be rendered will be created by a home grown application, so I have the opportunity to perform any shortcuts at creation time as opposed to rendering time.
While I am asking about it, my goal is to build a quad tree that contains all the static geometry in the scene off of one node in the scene graph, while all the Spatial content that can change/move is in another node off the root of the tree. Is this a good idea, or is there a better way to go about it in jME?
Once again, keep up the good work! It is much appreciated.