In this paper, the learning of qualitative two- compartment metabolic models is studied under the conditions of different types and numbers of hidden variables. For each condition, all the experiments, each of which takes one of the subsets of the complete quali- tative states as training data, are tested one by one. In or- der to conduct the experiments more efficiently, a back- tracking algorithm with forward checking is introduced to search out all the well-posed qualitative models as candidate solutions. Then these candidate solutions are verified by a fuzzy qualitative engine JMorven to find the target models. Finally the learning reliability and kernel set under different conditions is calculated and analyzed.