This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.

In the study of patients with disease of the lungs, a roentgenogram of the chest is nearly always obtained. This is because the chest roentgenogram generally provides valuable information which helps the doctor in making a diagnosis. Chest roentgenograms are generally available even in the smallest hospitals and in fact in the offices of many physicians. In spite of its great utility, the chest roentgenogram does not provide an exact etiologic diagnosis. For example, the characteristic picture of tuberculosis can be mimicked by other diseases, such as upper-lobe bronchiectasis. Then, too, it should be emphasized that if the chest roentgenogram is normal, lung disease is not excluded. The roentgenogram of the chest does not even tell us whether the patient was dead or alive when it was taken. It is a favorite trick of Dr. Julius Comroe to start a lecture with an essentially normal-appearing chest roentgenogram and to announce