Summary. This paper was motivated by a controversy concerning the role of basic sciences in medical education. A problem underlying this issue is that it is unknown how basic science is used in clinical reasoning. The experiment was designed to address this issue. Three texts were constructed dealing with basic science knowledge relevant to a clinical problem. Subjects were asked to read and recall the texts. Next, the subjects were required to read and recall the clinical text describing a patient problem. Finally, they were asked to provide a diagnosis and an explanation of the underlying pathophysiology. Subjects were first-, second and fourth-year medical students. Detailed analysis of subjects' protocols arc presented. In general, the results show that when basic science information is given before the clinical problem, the basic science knowledge is used either incorrectly or inconsistently in explaining the clinical problem by all subjects. The authors interpret these results to indicate that the basic sciences and the more practical clinical knowledge form two separate domains with their own individual structures and the clinical information cannot be embedded into the basic science knowledge structure.