A radiogram is not a photograph of the object exposed to the X-rays but merely a picture of its shadow, or rather of a series of shadows of the different structures, which vary in opacity. As the rays emanate from a single point in the vacuum tube, and as they are not, like the sun's rays, approximately parallel, the shadows they cast are necessarily distorted. Hence, in interpreting a radiogram, it is necessary to know the relative positions of the point from which the rays proceed, the object exposed, and the plate on which the shadow is registered. The least distortion takes place when the object is in contact with the plate, and the shadow of that part of the object which lies perpendicularly under the light is less distorted than that of the parts lying outside the perpendicular. The light and the plate remaining constant, the amount of distortion varies directly with the distance between the object and the plate. To ensure accuracy in the diagnosis of fracture by the X-rays, it is necessary to take two views of the limb--one in the sagittal and the other in the coronal plane. By the use of the fluorescent screen, the best positions from which to obtain a clear impression of the fracture may be determined before the radiograms are taken. Stereoscopic radiograms may be of special value in demonstrating the details of a fracture that is otherwise doubtful. Imperfect technique and faulty interpretation of the pictures obtained lead to certain fallacies. In young subjects, for example, epiphysial lines may be mistaken for fractures, or the ossifying centres of epiphyses for separated fragments of bone. The os trigonum tarsi has been mistaken for a fracture of the talus. In the vicinity of joints the bones may be crossed by pale bands, due to the rays traversing the cavity of the joint. In this way fracture of the olecranon or of the clavicle may be simulated. The neck of the femur may appear to be fractured if a foreshortened view is taken. It is possible, on the other hand, to overlook a fracture--for example, if there is no displacement, or if the line of fracture is crossed by the shadow of an adjacent bone. In deeply placed bones such as those about the hip, or in bones related to dense, solid viscera--for example, ribs, sternum, or dorsal vertebræ--it is sometimes difficult to obtain conclusive evidence of fracture in a radiogram.