The structure of macrophages at the site of subcutaneous injections of iron-dextran has been studied with the electron microscope. Iron-dextran has a macromolecular structure which can be distinguished from that of its metabolic product, the iron-storage protein ferritin. By a process of pinocytosis the macrophages ingest iron-dextran, then concentrate it within cytoplasmic vacuoles, apoferritin is synthesized and the iron is finally stored within this protein to form ferritin. At first individual molecules of ferritin are uniformly distributed throughout the cytoplasm, accompanied by a few molecules in the nucleoplasm but, as a rule, none in mitochondria. One week after injection, vacuoles containing a mixture of iron-dextran and ferritin are found, and these are gradually replaced by vacuoles containing only ferritin.COLLOIDAL iron preparations contain macromolecular particles which can be distinguished in electron micrographs from the images produced by the ferric hydroxide micelles of the iron-storage protein, ferritin [Richter, 1959;Bessis and Breton-Gorius, 1959;Muir, 1960]. Thus it is possible to observe the cellular transformation of administered iron preparations into ferritin and haemosiderin. After intraperitoneal injections of iron-dextran, saccharated iron oxide and hydrous ferric oxide in the mouse, Richter [1959] studied, with high resolution electron microscopy and electron diffraction, the incorporation of the injected material in hepatic and splenic macrophages. Some of these experiments have been repeated and the results confirmed by Bessis and Breton-Gorius [1959].In the inevitably small samples of tissue examined in the electron microscope, such studies of the synthesis of ferritin in the splenic macrophages could be confused by the random association of pre-existing ferritin with the injected colloidal iron compounds. It is preferable to work with tissues at a subcutaneous injection site, which normally does not contain ferritin; here it can be assumed that any ferritin synthesized stores the administered iron. Richter [1959] mentions briefly the subcutaneous changes 1-16 days after injection of hydrous ferric oxide, illustrating only a cytoplasmic iron depot at 5 days. The fate of colloidal iron and isolated haemoglobin injected subcutaneously in the mouse is reported bv Wessel and Gedigk [1959]. In the present study attention is given to the method of ingestion of irondextran by macrophages and the subsequent synthesis of ferritin in their cytoplasm.