Ever since the discovery of X-rays in 1895, X-ray imaging has played a large role in the cognitive and practical organization of medicine. This article analyses the way X-ray images were introduced and made sense of in medical thinking and acting around the turn of the century. The implicit assumption in many histories of radiology is that the specific (diagnostic) message of the X-ray images resided inside them from the beginning, and that it is obscured either by technological or epistemological problems. These being solved, it would then be no problem to see directly what information the image contains. In this article this assumption is contested. It is argued that the specific content of the images was shaped by the activities of X-ray workers within the context of medical developments of the time. This shaping, as it is historically reconstructed here, consisted of four methods. X-ray workers (be they physicians, technicians or scientists) experimented with the technology, the images, the photographic materials and the objects that were X-rayed. They used X-ray images of dead bodies to compare them with radiographs of living patients. Radiologists tried to'translate'diagnostic information acquired with other methods into the shadows of the X-ray images. And finally they compared images with images. The process of shaping the content and use of X-ray images, of making them represent reality, took place within specific institutions, and it took a different form in different countries, but also for different parts of the body. Developments of institutionalisation and professionalisation of radiology in England and the Netherlands are presented to provide a small part of the background of this shaping of knowledge of shadows.