THE serum of patients suffering from mb.ltiple myelomatosis has for a considerable time been known to show gross divergences in constitution in comparison with normal human serum. Among these abnormalities may be mentioned a marked hyperproteinemia, associated with a change in the albumin-globulin ratio. The condition is also frequently, but not invariably, associated with Bence-Jones proteinuria. An exhaustive review of the literature has recently been given by Bonsdorff et al. [1938]. The data on two. such sera [McFarlane, 1935] encouraged a further extensive study by the ultracentrifugal and recently improved electrophoretic techniques. This paper is a report of the results obtained in the examination of the sera from five cases by these methods.

1. Human alpha(2)-macroglobulin was prepared from a fraction obtained during the large-scale separation of normal human plasma proteins for clinical use. 2. Sedimentation-equilibrium measurements indicated a molecular weight of 725000. A value of 18.1S was obtained for s(0) (20,w). 3. The dissociation that occurs in the pH range 4.5-2.5 and in the region of neutrality in urea-containing solutions is consistent with a dimeric structure of the molecule. 4. The effects of the thiol reagents mercaptoethanol, mercaptoethylamine and N-acetylcysteine were investigated over a range of experimental conditions. Distinct components having sedimentation coefficients of 15, 12 and 8.5S were identified. 5. Conditions were found under which limited reduction with thiol liberated a subunit with a molecular weight approximately one-quarter of that of the intact molecule. This subunit retains the serological specificity of the whole molecule.

Tmis paper is a report upon a somewhat extensive investigation of the hydrogen electrode titration of crystalline egg albumin. The work was undertaken with the object of examining the limitations of the method as a means of establishing a quantitative definition of the amphoteric properties of a protein. Crystalline egg albumin was chosen because it is probably the best accredited example of a molecularly homogeneous protein. The extensive work of S0rensen and of his colleagues, of Svedberg and of others indicates that of the common protein preparations egg albumin exhibits the greatest degree of constancy in composition, molecular weight and solubility under varying conditions of preparation and treatment. This protein has the further advantage that it is soluble in water over the whole titratable range of PH.The data which will be considered derive from observations gathered over a period of five years upon five distinct preparations of the protein. These preparations had been submitted to considerable variations in treatment prior to titration. Each was crystallised four times-one by the original method of S0rensen [1917], the others by a modification of this method in which sodium sulphate replaced ammonium sulphate as the salting-out agent. Two of the latter preparations were employed in the electrometric work without having been reduced to the dry state. Two were converted into a dry powder and stored for some time before use. One of the dry preparations was dissolved in water and titrated without further purification, the other after one more recrystallisation. Some details of the method of preparing crystalline egg albumin in the dry state with the aid of sodium sulphate will be found in the experimental section.Each product was prepared for titration as a stock solution which had been dialysed in distilled water until the sulphate ion could not be detected in a dialysate after sixteen hours' contact with the solution. Varying amounts of dilute HCI or of NaOH were added to equal volumes of the stock solution and the mixtures were then diluted to a predetermined concentration of protein. The majority of the observations relate to systems containing 22-25 g. of protein in 1000 g. of water. In a few cases this value was reduced to 15 g. and in a few others raised to 45 g. The PE of each mixture was determined in a rocking hydrogen electrode [Clark, 1928] at 250, using a saturated calomel half-cell and a saturated potassium chloride liquid junction. The calomel half-cell was calibrated with the aid of 0.1 N HCI, for which we assumed a pH of 1-075 [Scatchard, 1925]. The interval between the preparation of a mixture and the determination of its PH did not exceed 1 hour.Commonwealth Fund Fellow. ( 227 )

For investigations concerning the mode of action of the clearing factor, estimations of free fatty acids in mixtures of normal rat plasma and lipaemic dog plasma were performed using a modification of van de Kamer's (1948) method as described for faeces. Because the method developed may be of value for similar or other uses, it is given here. In a 100 ml. flask 2 ml. of the plasma-mixture, containing either heparin or citrate as an anticoagulant, are run into 6 ml. of phosphoric acid solution (metaphosphoric acid, 50 g., NaCl, 250 g. water to 1 1.). After a few minutes 16 ml. of ethanol, containing 0-4 % amyl alcohol, are added, followed by 20 ml. of distilled light petroleum (b.p. 40-60°). The flask is stoppered and the mixture shaken by hand for 1 min. After separation, the upper layer is filtered through dry paper and the funnel covered in order to minimize evaporation. Care should be taken to avoid contamination with traces of the other phase. Ten ml. of the light petroleum extract is evaporated on a boiling-water bath and the residue dried in vacuo. Neutralized ethanol (5 ml.) containing 0-6 mg. of thymol blue per 100 ml. are added and the residue dissolved by boiling gently under reflux for 3 min. Finally, the fatty acids are titrated under a stream of N2 with O-O0N-NaOH.The results are corrected for a blank (usually about 0060 ml. of 001N-NaOH), obtained by running 2 ml. of water through the same procedure.The reproducibility of the method suffices for many purposes. In twenty-four estimations giving results ranging from 021 to 210 0,-equiv. of fatty acids per 2 ml. of plasma tested, the differences between duplicate determinations were found to have a standard deviation of + 0-14 ,u-equiv.

not be ascribed to a direct toxic effect in any of the cases where this was studied. 3. The compounds that proved to be most effective in reducing the amount of the incorporation of amino acids into the bacterial protein were also, in general, those most effective in inhibiting the growth of the Walker rat carcinoma.