Carcinoma of the bladder represents a spectrum of disease ranging from benign papillary carcinoma to a highly invasive and aggressive anaplastic tumor. The variability of its course and lack of a generally acceptable classification has obscured rather than clarified the problem. This is not to imply that a dearth of literature exists on this subject, but the absence of a generally accepted clinical grading which can be utilized by both the urologic surgeon and radiotherapist is reflected by the variety of systems proposed. Patterson suggests the division of bladder carcinoma simply into noninfiltrating and infiltrating varieties. In contrast, Friedman (7) has designated twelve clinicopathologic types for convenience and suggests that twice this number actually exist and would aid in determining treatment and predicting prognosis. This leads to a dilution of numbers and makes it difficult to report results. A pure clinical grouping, as suggested by Rusche and Jaffe (13) fails to consider the histology, which is as important in classification as anatomical extent. Recently, accurate classification systems have been evolved by leading urologists. These systems are reduced to two essentials known to influence prognosis: (a) the physical extent or depth of infiltration of the tumor; (b) its inherent aggressiveness or histologic grade. Accordingly, Marshall's dual classification (12), which permits evaluation of a given tumor as to its inherent vigor (grade) and present anatomic extent (stage) seems most worthwhile. This is, however, a surgical and not a clinical classification, which in many instances does not allow for grouping of all patients treated by radiation therapy. Marshall (12) states that “no one knows what proportion of vesical cancers fall into what stage at the time the patients present themselves for treatment.” This is borne out by the fact that 40 per cent of his patients with deeply invasive lesions (B2 and C groups) had histologically positive nodes at a time when routine preoperative studies did not indicate such carcinomatosis, implying stage D lesions. He rightly concludes that specimens removed by radical cystectomy or pelvic extenteration should provide firm knowledge in this respect. Such a “biopsy procedure” for accurate classification, however, eliminates the need for additional therapy. In attempting to evolve a meaningful classification, the factors to be regarded can be divided into primary or essential considerations as to the prognosis or evolution of the disease in an average patient and secondary factors or considerations as to the success of treatment in a specific individual. That is, a slow growing, superficial carcinoma, classified according to primary factors, should show a favorable response to a simple form of therapy, but may terminate fatally in the presence of an associated advanced cardiac disease.