The Morris water maze represents the de-facto standard for testing hippocampal function in laboratory rodents. In the field of adult hippocampal neurogenesis, however, using this paradigm to assess the functional relevance of the new neurons yielded surprisingly inconsistent results. While some authors found aspects of water maze performance to be linked to adult neurogenesis, others obtained different results or could not demonstrate any effect of manipulating adult neurogenesis. In this review we discuss evidence that the large diversity of protocols and setups used is an important aspect in interpreting the differences in the results that have been obtained. Even simple parameters such as pool size, number, and configuration of visual landmarks, or number of trials can become highly relevant for getting the new neurons involved at all. Sets of parameters are often chosen with implicit or explicit concepts in mind and these might lead to different views on the function of adult-generated neurons. We propose that the classical parameters usually used to measure spatial learning performance in the water maze might not be particularly well-suited to sensitively and specifically detect the supposedly highly specific functional changes elicited by the experimental modulation of adult hippocampal neurogenesis. As adult neurogenesis is supposed to affect specific aspects of information processing only in the hippocampus, any claim for a functional relevance of the new neurons has to be based on hippocampus-specific parameters. We also placed a special emphasis on the fact that the dentate gyrus (DG) facilitates the differentiation between contexts as opposed to just differentiating places. In conclusion, while the Morris water maze has proven to be one of the most effective testing paradigms to assess hippocampus-dependent spatial learning, new and more specific questions ask for new parameters. Therefore, the full potential of the water maze task remains to be tapped.