The ultimate performance which can theoretically be obtained with a fluoroscopic imaging device is determined by the quantum nature of the x-rays themselves. In medical fluoroscopy, the available x-ray intensity is set by the permissible patient dose, so that the information carried by the x-rays is quite definitely limited. This fact was first pointed out by R. C. Mason (1) in 1944. Since then, experimental work and calculations have been carried out by several others. The limitation has usually been expressed in terms of the minimum brightness gain of an ideal intensifier in which the perception is limited by the scintillation effect rather than by the observer's eye. Mason estimated that the scintillations would be noticeable at a brightness gain of about 10, and expressed serious doubt as to whether very large brightness gains (i.e. 10,000) would be of interest. An experimental test of the effect was carried out for a particular case by Coltman and Ebbighausen (2, 3). Under conditions equivalent to a gain of 5,000, they found that the contrast discrimination for small objects was not seriously impaired by the quantum noise. Morton (4), in discussing the photon limitations of the eye, ventured the opinion that in x-ray image amplification there was little point in going beyond a gain of 10, since no further information could be obtained from the x-rays. Sturm and Morgan (5), in a more elaborate analysis, arrived at a figure of 50 to 100. With the advent of the first operable image amplifier tubes (6), it became possible to prove directly that at brightness gain of 50 times, no measurable loss in visual acuity could be ascribed to the scintillation effect. At this brightness gain, the visual acuity was still limited by the eye rather than by the instrument. Thus it became certain that some of the early estimates of the scintillation limit were unduly pessimistic. Though Sturm and Morgan's work can be revised to yield a very much higher gain figure when certain previously neglected considerations are taken into account, the subject still remained in a very unsatisfactory state, first because of the wide variance of conclusions, and second because all of the methods were more or less indirect and employed extrapolations and assumptions which had not been adequately tested. The research reported here was undertaken to make a direct determination of the capabilities of an ideal fluoroscope limited only by quantum fluctuations. The problem resolves itself into two parts: first, a determination of the number of significant x-quanta available at the screen; second, a determination of the ability of the eye to integrate a scintillating picture into a meaningful pattern. With these data the performance of an ideal image intensifier over the entire range of fluoroscopic conditions can be predicted. Determination of Available Quanta