THE STRUCTURAL BASIS FOR GENETIC VARIATIONS OF NORMAL HUMAN -yGLOBULINS*

degree of clumping. In all studies, appropriate controls were used to insure that agglutination was not due to the test serum or the added fractions rather than the "rheumatoid factor." The test sera and red cell coating sera employed were similar to those used previously.25 Each fraction was tested for inhibitory activity with at least two separate sets of reagents for each Gm factor with the exception of In V(b), for which only one set of reagents exists to date. In some of the later experiments, tests for In V (b) activity were not possible since the reagent was no longer available. Inhibitory activity ( + ) or lack of inhibition (-) varies with the quality of the reagents used for the various test systems. For Gm 1 (a), the difference is striking and varies by a factor of at least 32 with both sets of reagents used. For In V (a), the difference is 8- and 16-fold with the two reagents used. For In V (b), the difference is 8-fold, and in the Gm 1 (b) system the differ- ence between a positive and a negative serum is 4- to 8-fold with two sets of test reagents. The results of the present studies on normal human 'y-globulins isolated from individuals of different genetic y-globulin groups (Gm groups) and similar studies on myeloma proteins by ourselves29 and by Harboe and collaborators30 present strong evidence that the Gm factors, controlled by two distinct Gm loci, reside in separate parts of the -y-globulin molecule and suggest that the two major genetic loci affect different parts of the protein.