The present study of roentgenographic and isotopic determination of residual urine was initiated because of conflicting reports on their values. Brailsford et al. (1) claimed accuracy to within 30 c.c. in comparisons of anterior and lateral roentgenograms of the bladder with cystograms where the volume of contrast media was known. In selected patients known to be without retention of urine, Bretland (2) found poor correlation on excretory urograms between the area of the bladder and the volume of urine voluntarily voided. In fact, the variation in volume for a given area was 50 per cent or more about the mean. Hershman (3) compared the postvoiding cystogram of the frontal area to the catheterized volume taken immediately after roentgenography in 100 patients with prostatism and claimed an accuracy to within 30 c.c. Mulrow et al. (4) estimated residual urine by injecting radioiodinated Diodrast intravenously and counting over the bladder before and after voiding. By equating the difference in counts to the measured voided specimen the retained urine was calculated. The estimated values agreed more closely to the catheterized volume than those reported roentgenographically by other authors. This report deals with a comparative controlled study of both procedures. Part I is an investigation of the roentgenographic method of determining residual urine, and Part II is a study of its isotopic measurements. Part I—Roentgenographic Study Method and Materials: In the roentgen determination of residual urine, 16 patients (5 males, 11 females) were examined. Of these, 3 had known vesical disease (unspecified), 7 had pelvic carcinoma (of cervix, bowel, etc.), 6 had apparently normal bladders. All voluntarily emptied their bladders and immediately thereafter were catheterized on the x-ray table. The catheter was left in, but clamped. Via needle puncture of the catheter, successive known volumes of 4 per cent NaI were injected, and following each increment a roentgenogram was obtained (36-inch target-film distance, angled 15 degrees caudad, subject supine). At the end of the procedure the bladder was drained through the catheter into a measuring cylinder and another roentgenogram was made to insure almost complete removal of urine. The excess drained over that instilled was assumed to be due to urine accumulation at a constant rate during the time (about twenty minutes) required for the procedure. The excreted urine volume was calculated for each cystogram and added to the volume of NaI instilled. This total vesical volume was then plotted against its area as measured by planimeter from the roentgenogram. Each curve represents one patient and consists of about 7 points (Fig. 1). Results: A study of the set of curves obtained reveals only a qualitative relationship between area and volume. Any given area corresponds to a wide range of volumes. For example, 25 sq. cm. can represent a range of 15 to 65 ml.; 35 sq. cm. is equivalent to a 25 to 160 ml. range.