You can either store the wav data in a constant like and array of clusters, each cluster would have the wave data (8 or 16 bit mono or stereo: which are 1-d or 2-d arrays of intergers) plus the config info rate resolution or you can store an array of references to wave file paths. I would go with the second version since you dont bind the data at compile time, allowing for the application to change "sound tracks" without recompiling. When you build the application you can specify these wave file as support files and lock their paths (probably ./application/data) you can even store the path in the registry if needed. Good luck.
Paul
