A number of aspects of the melano-macrophage centres of fish were investigated. In teleosts these centres consist of aggregates of pigment-laden macrophages and various leucocytes. They are usually embedded in haemopoietic tissue (mainly within the spleen and kidney) in association with the blood-supplying vessels and in a few species a distinct lymphoid cuff surrounds the entire structure. Salmonids are exceptional in that their melano-macrophages are scattered throughout the haemopoietic tissue and do not form distinct aggregates. At least four types of pigments have been described to occur in these phagocytes, viz. melanin, the lipogenic pigments ceroid and lipofuscin, and the haematogenous pigment haemosiderin. In light microscopy the centres appear in varying hues of yellow, brown and black. It has been suggested that these centres could well represent the primitive analogues of germinal centres of the lymph nodes of birds and mammals. As all previously available information came exclusively from a restricted number of teleost species it was considered of primary importance to carry out a study of the distribution and cytological organization of these centres in living representatives of Agnatha, Chondrichthyes and Osteichthyes. Seventy-two species of fish had their haemopoietic tissues examined by light microscopy for pigment-containing macrophages. Except for the lamprey Lampetra fluviatilis, all fish species were observed to possess these pigment cells. An evolutionary pattern was evident in both the distribution and the degree of organization of melano-macrophages; three major evolutionary trends were discernible, viz: i. a progressive increase in the abundance of pigment cells; ii. a structural evolution from a random distribution of individual pigmented macrophages observed in Agnatha and Chondrichthyes to organized centres characteristic of all Osteichthyes except the salmonids; iii. a change in organ location of these pigment cells from the liver in Agnatha, Chondrichthyes and the primitive bony fishes to the spleen and kidney in the advanced bony fishes. The increasing sophistication in cytological organization of the melano-macrophage centres is concomitant with the increasing levels of complexity of the cytoarchitecture of the lymphoid system. The increasing proclivity of the centres for the main lymphoid organs follows closely upon the evolution of the lymphoid system and represents a major advance in the evolution of lympho-reticular relationships. These analogies provide additional evidence that these centres are of a lymphatic nature and may well represent the primitive analogues of germinal centres of birds and mammals. Ontogenically pigment-bearing macrophages appear following upon first feeding. Immunological maturity appears to be attained at first feeding and the fact that it is shortly afterwards that melano-macrophages appear within the lymphoid tissues (eventually leading to melano-macrophage centre formation within them), seems to add weight to the evidence of a structural and functional relationship between melano-macrophage centres and lymphoid tissues. The very marked changes observed in the melano-macrophage centres during cachexia provided a convenient tool for studying these pigmented macrophages. In adult rainbow trout Salmo gairdneri and plaice Pleuronectes platessa kept at 12°C, the density of melano-macrophages and melano-macrophage centres respectively had increased considerably after 6 weeks of complete deprivation of food and by 10 weeks very high densities were observable. At higher temperatures (25°C), employing Tilapia zillii and swordtails Xiphophorus helleri, a very marked increase in the density of these centres was already evident after 3 weeks of complete starvation. With first feeding rainbow trout fry kept at 12°C, high densities of melano-macrophages were observed within the spleen and kidney after 3 weeks of complete starvation. No other treatment employed in this study was observed to induce any significant changes in the melano-macrophages of either fingerlings or adult fish. These results suggest that tissue atrophy is a major factor contributing to the formation of the pigments observed within the melano-macrophages. Electron microscopic observations employing normal and cachectic plaice indicated the following possible modes of origin for the pigments within melano-macrophages: i. melanin granules seem to be derived from their being simply phagocytosed from the classical melanin-containing cells that have been ruptured or otherwise damaged; ii. lipogenic pigments appear to derive from damaged cellular components such as effete mitochondria through the process of peroxidation of their unsaturated lipids; iii. haemosiderin is almost certainly derived from the breakdown of haemoglobin from effete erythrocytes. Since lipid peroxidation and recycling of iron compounds lead to the formation of free radicals and cations, these potentially toxic entities are bound to arise spontaneously within melano-macrophage centres. In view of this there is raised the possibility that the melanin within the centres could be playing a very important role through its well-recognized ability to absorb free radicals and its strong affinity for cations. While this would account for the presence of all these types of pigments within melano-macrophages, it could also explain why pigment cells are so often observed at sites of infection or tissue injury. All ingested cellular debris appears to be subjected to lysosomal enzyme activity and it is the indigestible residues (indigestible unsaturated lipids mainly) that give rise to the pigments which gradually accumulate. The absence of pigment in young larval fish, its steady accumulation with age in clinically normal fish and its presence without exception in older fish indicate that the pigments being studied seem to satisfy the criteria set forth for a basic biological aging process. The role of melano-macrophage centres in iron storage in normal and diseased fish was studied. The spleen, kidney and liver centres of fourteen species of clinically normal teleost fish were examined histochemically for haemosiderin. This was found to be present in varying amounts within the splenic centres of most specimens, but in contrast it was rarely found in the centres of the kidney and the liver. Linder conditions of starvation and in diseased fish, a markedly increased deposition of ferric iron occurred in the splenic centres of nearly all fish examined. By comparison, the iron content in the kidney and liver centres was generally still very low. These results suggest that although the centres in the various haemopoietic organs resemble each other morphologically and in their relation with the associated tissues, there could well be important functional differences between the centres of different organs. When rainbow trout that had been splenectomised were starved, accumulation of haemosiderin was diverted to the kidney melano- macrophages; the liver was still virtually devoid of iron. The possible implications of these findings are discussed. In conclusion, all the available evidence seem to indicate that the melano-macrophage centres of fish pnd the germinal centres of birds and mammals are similar in many ways. A major difference is the high levels of pigments in the centres of fish. This seems to be related to the inability of the latter to control their body temperature. Fatty acids of living organisms shift towards greater unsaturation under lower environmental temperatures as a means of maintaining protoplasmic viscosity within the range necessary for normal metabolic processes. Thus fish, because of their poikilothermic nature, have high levels of unsaturated fatty acids in their bodies and are thus more prone to lipofuscin formation. It has also been suggested that intracellular digestive processes of fish macrophages may not be well developed on the evolutionary scale. This could also lead to increased accumulation of indigestible materials within the melano-macrophage centres. Finally these results are discussed with special emphasis being placed on the following three points: i. that the melano-macrophage centres should be regarded as sites where a large variety of materials are aggregated, processed, sifted and disposed of in a variety of ways rather than regarding them as static areas passively accepting and storing any materials that come their way. Of special significance are those materials that are required for recycling such as iron-containing compounds; ii. that there are important functional differences between the melano-macrophage centres of different organs; iii. that as more information becomes available the melano-macrophage centres may well become useful as sensitive indicators of the state of health of the fish. A technique for bleaching the pigments within melano- macrophage centres in ultra-thin sections was developed. Treatment of the sections with permanganate for five minutes followed by fifteen minutes in metabisulphite resulted in complete bleaching of almost all the pigments.