In the last few weeks of a course I've been taking (mathematical methods for physicists) we've learned about how to solve PDEs using the method of separation of variables. I think I pretty much know how to do it, technically. The one thing which I can't seem to grasp is why the most general solution is the linear combination of all of the separated forms. How do I know that there aren't more solutions which agree with some specific boundary conditions, while the separation method does not?
Thanks