SummaryKunkel’s technic for precipitation of globulins was applied to hyperglobulinic serum of patients with multiple myeloma-with following results :Turbidity was in those cases too high to allow comparison with the McLaghan's SO4Ba scale.Dilution of the serum or precipitation with larger amounts of Kunkel’s reagent showed obvious errors in the dosage of globulins (Fig. 1).Further investigations made possible to understand the mechanism of these erroneous results.First, a linear correlation was noticed between photometric readings and concentration of pure globulins solutions (Fig. 2).Thereafter, several mixtures of pure albumin and globulins were prepared, definited by' their A/G ratio; they were diluted and precipitated by Kunkel’s reagent. Authors found a curvilinear relation of readings with globulin concentration (Fig. 3).Low levels of serumalbumins increased the zinc turbidity values. High levels inhibited the floculation of globulins. The lower the A/G ratio of the mixture, the highest are the amounts of precipitated proteins (Fig. 3).These results were also checked as follows : mixtures of globulins and pure albumin were precipitated with increasing amounts of Kunkel’s reagent, but pure albumin was added in order to avoid dilution of this protein when globulin- concentration decreased. Once more a protective effect on globulin floculation was observed with high concentrations of albumin, whereas small amounts of this seemed to increase the floculation of protein (Fig. 4).The authors call«critical concentration of albumin»this value which does not influence the floculation of globulins, giving a photometric reading equal to that of the pure material. This critical concentration of albumin seems to be only slightly influenced by the concentration of globulins in the mixtures.