The important question of the formation of serum proteins has been the subject of a number of investigations. Among these, the extensive studies of Whipple and his collaborators (1-4) should be mentioned especially. The authors brought evidence to show that the liver plays an important part in the production of plasma proteins, in particular albumin. This opinion is based on the results of several sorts of experiments, for example the effect of various types of liver injury such as that obtained by the establishment of Eck fistulae, and feeding experiments with dogs having low plasma proteins. The liver also is held by most investigators to be the organ responsible for the production of fibrinogen. Other workers suggest the participation of the bone marrow in the production of serum proteins. For a more complete analysis of the literature on the subject, the reader is referred to recent reviews by Whipple (5), Juergens (6a) and Reimann (6b).Lately, an approach to the question by histological analysis has been made by Sabin (7), who based her thesis upon the relationship between antibodies and serum globulins. The production of antibodies, which are generally considered to be modified serum globulins, is usually attributed to the reticulo-endothelial system (cf. 8) and current hypotheses imply an intimate connection between the formation of antibodies and normal serum globulins (see 9, 10). It would follow, then, that normal globulins are likewise formed by this system. And by examination of tissues and exudates of rabbits during immunization with an artificial antigen having intense red color, Sabin obtained histological pictures which led her to conclude that antibodies are formed by the reticulo-endothelial cells.In the hope of obtaining further information concerning the origin of serum proteins, an attempt was made to grow tissue cells in the plasma of a foreign species and to detect the presence of serum proteins by means of precipitin reactions with immune sera. Interesting experiments to deter-231 on