Mendel's rules are the based on simplifying assumptions--the genes being considered are in unrelated places in the genome, that they live on 'generic' nuclear chromosomes, and that gene products under study have no interaction with each other. What happens when one or more of these assumptions are invalid? Using a hypothetical organism (hey--the truth is out there, but it isn't very tractable!) you'll learn how to recognize and distinguish between these sorts of misbehaviors.
NDSU site with some stuff that you may find useful
Some information about statistical testing: chi square