The condition of close contact between cells and emulsion is achieved by the technique of dip-coating autoradiography. In this process the glass slide on which the cells are carried is dipped into a melted photographic emulsion (see Figure 15a), a thin film of which clings to the slide. After it has been dried, the slide is placed in a lighttight box and kept in a refrigerator for the desired period of exposure, usually several days or weeks. During this period disintegrating radioactive atoms within the cells continue to emit beta particles, which, in turn, produce a latent image in the overlying emulsion. After the exposure is complete, the slide is developed and fixed like a photographic plate, and a stain is applied which penetrates the emulsion so that the outlines of the cells and their internal structures can be seen. The fixing process removes all silver bromide that has not been ionized so that the emulsion is reduced to a thin, transparent film of gelatin covering the stained cells and containing only the clusters of silver grains that were struck by the beta particles. [Illustration: Figure 16 _Radioautographs of tumor cells. Above, tumor cells and blood cells. Below, magnification of tumor cells._] When the finished autoradiograph is examined under the microscope, it will look like the radioautographs of tumor cells in Figure 16. In the upper micrograph the tumor cells are the larger ones and the smaller ones are blood cells. The dense structures in the center of the tumor cells are nuclei. The cells were exposed to tritium-labeled thymidine, and those synthesizing DNA at the time of exposure took up the thymidine and became radioactive. They can be identified by the black dots overlying the nuclei; the dots are the aggregates of silver grains struck by the beta particles. Notice that only the nuclei contain radioactivity; the reason for this is that radioactive thymidine is incorporated only into DNA localized in the nuclei of cells. This picture identifies not only the cells that were making DNA at the time the label was administered but also the cells that were destined to divide in the immediate future, since cells synthesize DNA in preparation for cell division. If we want to compare two populations of cells to find out which is proliferating (dividing) more actively, counting the fraction of cells labeled will give the number of cells synthesizing DNA in preparation for cell division. Of course, a rough approximation of the proliferating activity can be obtained by simply counting the number of cells actually dividing. But with tritiated thymidine we can obtain not only much more accurate measurements but also considerable information that cannot be obtained by simply counting the number of cells in mitosis. We shall discuss the cell cycle later on, but for the moment we should emphasize that much of our knowledge of the cell cycle stems from the use of high-resolution autoradiography.