Research should always be regarded as one of the activities of teaching clinics. Such clinics are usually relatively well equipped in the way of laboratories and endowment; and the very fact that the laboratories are closely associated with the wards gives an exceptional opportunity to the staff and thus places a responsibility on its members for the investigation of disease. It is fair to assume that a large part of the progress that will be made in our knowledge of the diagnosis and treatment of disease will come through the medical clinics, even if much of the fundamental work on which this practical advance is based is the outcome of investigations which have been carried on in the laboratories of so-called pure science. (As a matter of fact, it has interested me recently to see how frequently the clinical investigator, studying a problem in disease in man, is forced to go back and tackle the most fundamental aspects of it--anatomic, physical, chemical--because the necessary facts have not been made available by workers who specialize in these various fields.) Research should also play a part in the clinic because it is a type of training which develops critical judgment; and even a limited experience in a research problem, undertaken under skilled guidance, is a valuable discipline for every one, including the man who subsequently goes into practice, for among other things it teaches him to estimate the worth of the publications of other men. Finally, it is generally true that the investigator is a more stimulating teacher than the man who is not actively laboring at the forefront of scientific advance.