Any student is expected to know that correct prediction is a necessary, but not a sufficient, condition any theory must meet to be acceptable. However, an eventual weakening of reflective and critical attitudes when dealing with "whats" and "whys" in science learning can easily lead to an identification of the predictive success of a given theoretical model, albeit useful as an intermediate state of reasoning, with the fundamental explanatory power of such a model. This is not always taken into proper consideration by teachers and textbooks. Examples from chemistry teaching are discussed in this paper. (Contains 1 figure.)