SYNOPSIs The differential ferrioxamine test is a simple method for the measurement of chelation of body iron by desferrioxamine. A single six-hour specimen of urine is obtained after intravenous Desferal, accompanied by 59Fe-ferrioxamine. Two values are measured: Fd, the excretion of ferrioxamine derived from body iron by chelation, and Fex, the proportion of ferrioxamine excreted from a known intravenous dose. The data enables F,, chelation of iron in vivo, to be calculated by simple proportion. Desferrioxamine chelation proceeds for about half an hour after injection.The results in normal subjects, in cases with known high iron stores, and in cases of iron-deficiency The clinical assessment of the state of the body iron is made by indirect methods, which include haemoglobin concentration, mean corpuscular haemoglobin concentration, examination of the stained blood film, serum iron concentration, and the amount of iron-binding globulin, transferrin. There is a considerable literature dealing with the interrelationships of these parameters and their interpretation (see, for instance, Beutler, Fairbanks, and Fahey, 1963;Harris, 1963). In some situations the interpretation of these investigations taken together is unmistakable: in severe iron deficiency, for instance, a characteristic pattern of results is obtained. Nevertheless, these measurements are subject to many influences such as infection, neoplasm, renal failure, and metabolic disorders, which affect their values apart from iron status, and in such conditions interpretation may be speculative. Furthermore, they are of minor importance in the diagnosis of iron storage diseases.A more direct approach to assessing storage iron may be by needle biopsy of the marrow, staining the aspirated material for ferritin and haemosiderin by the prussian blue reaction (Rath and Finch, 1948).

Strains of Streptococcus salivarius were screened by negative staining for the presence of surface structures. Two structural subgroups were found, carrying either fibrils or fimbriae, projecting from the cell surface. Eight strains carried a very dense peritrichous array of fibrils of two distinct lengths. Long fibrils had an average length of 175 nm, and short fibrils had an average length of 95 nm. Two strains carried only long fibrils, one strain carried only short fibrils, and another strain carried a lateral tuft of very prominent fibrils of two lengths, with a fibrillar fuzz covering the remainder of the cell surface. In all the strains in which they were present, the long fibrils were unaffected by protease or trypsin treatment. In contrast, the short fibrils were completely digested by protease and partially removed by trypsin. Neither long nor short fibrils were affected structurally by mild pepsin digestion or by lipase. The Lancefield extraction procedure removed both long and short fibrils. These twelve fibrillar strains were therefore divisible into four structural subgroups. Extracts of all the fibrillar strains reacted with group K antiserum. The second main structural subgroup consisted of nine strains of S. salivarius, all of which carried morphologically identical, flexible fimbriae arranged peritrichously over the cell surface. The fimbriae were structurally distinct from fibrils and measured 0.5 to 1.0 ,um long and 3 to 4 nm wide, with an irregular outline and no obvious substructure. There was no obvious reduction in the number of fimbriae after protease or trypsin treatment. Extracts of the fimbriated strains did not react with the group K antiserum. The two serological and structural subgroups could also be distinguished by colony morphology.

The major iron-bearing cytosol components of human reticulocytes identified after incubation with 59 Fe-125I-transferrin have been studied further. Component C previously found to behave consistently as an intermediate in the iron transport pathway to haem is shown to consist entirely of ferritin. After a short pulse of labelled transferrin incubation, chase experiments showed a fall of ferritin label with time and a corresponding increase in haemoglobin-iron incorporation. There was no loss of ferritin to the culture medium. Restriction of iron uptake by reticulocytes using both p-hydroxymercuribenzoate inhibition of uptake and incubation with progressively lower saturations of iron-transferrin gave linearly related incorporation of 59Fe into ferritin and haemoglobin at all levels of iron uptake, thus negating the concept of ferritin as an 'overspill' form of reticulocyte iron. The results suggest that cytosol ferritin is an obligatory intermediate in reticulocyte iron transport.

SUMMARY The chelation of body iron by desferrioxamine methanesulphonate in patients with haemolytic anaemias of varying types, in megaloblastic and in sideroblastic anaemias was measured by the differential ferrioxamine test. Thirty‐five out of 44 patients studied had chelation values above the normal range, sometimes exceeding those found in untreated haemochromatosis. In the haemolytic anaemias, the amount of chelation increased with the severity of the anaemia. In the megaloblastic anaemias, specific therapy reduced the amount of iron chelated by desferrioxamine at the same time as normal erythropoiesis was re‐established. Concepts of labile iron and metabolically active iron are compared with desferrioxamine chelatable iron. These findings suggest that in addition to storage iron (ferritin‐haemosiderin) chelated with difficulty, there is a compartment of readily chelatable iron derived from haem catabolism, possibly sited in reticulo‐endothelium. A model of intracellular iron transport in reticulo‐endothelial cells is proposed which provides a primary pathway of metabolically active, highly chelatable iron derived from haemoglobin catabolism, and a secondary pathway to intracellular ferritin.

The low molecular weight of iron-sorbitol complex resulted in a rapid excretion in the urine. This was maximal in the,first twelve hours. The urinary excretion for the 48-hour period after injection ranged from 18% to 53%.The red-cell utilization at ten days ranged from 27% to 66 %. Utilization in the iron-deficient group was greater than in the control group. The utilization curve in the iron-deficient group showed that maximal utilization had not occurred by T+ 10 days.The surface counting data suggested that the retained fraction of iron not utilized for haemoglobin synthesis by T+ 10 days (16% to 40%) was stored predominantly in the liver.We are grateful to Dr. S. Wahlqvist, of Astra, Sweden, for providing the "9Fe sorbitol. The expenses of the investigation were partly borne by the Endowment Fund, St. Thomas's Hospital.