Some of the problems of applying theory, so successful in the physical sciences, to complex problems in neurobiology will bee discussed. As an example, a theory for the development of important properties of cortical cells will be presented. The emphasis will be on logical structure: postulates and conclusions as well as comparison with experiment.