When the effectiveness of a form of treatment for a certain disease is to be evaluated, there is really no other way of doing this than by comparing the results achieved with this treatment in the cases in one group with the course of the disease in another group in which the treatment in question has not been used. Before the evidence from such a comparison can be considered conclusive, however, it is essential that the two groups be homogeneous in all other respects except the factor under examination—or at least as nearly alike as it is possible to achieve with biologic material—and also that the number of individuals available for study should be sufficiently large and approximately the same in the two groups. In order to be able to eliminate discrepancies in the two groups to be compared, it is necessary for the examiner to have a thorough knowledge