P L du Noüy’s scientific contributions

Experiments are reported which indicate that a shift toward higher concentrations is observed in the minimum value of the static surface tension when plasma instead of serum solutions is used. The amount of the shift, expressed as a function of the concentration, shows that the figures are in satisfactory agreement with the determined amount of fibrinogen in the plasma. Some evidence is given that "plasma molecules" capable of organizing themselves on adsorbing surfaces exist in plasma, and that their length would be approximately 4.3 mmicroin round figures,instead of 4.0 mmicro for the serum. The area occupied in the plane of adsorption by one individual molecule is, however, smaller than that occupied by the "serum molecule," thus indicating a marked structural difference between the two, the "plasma molecule" being narrower but longer than the "serum molecule." This difference may be due either to a different orientation accompanied by an increase in one of the dimensions, or else to an actual difference in structure with respect to the main axis, resulting in a decrease in the mean diameter of the "serum molecule" with an increase in the length of its main axis. The mass of the "plasma molecule" is about 6.3 per cent larger than that of the "serum molecule, " in the case of rabbit serum.

The nature of the relation between the time-drop and antibody formation cannot be stated as yet. It is certain that the physicochemical change in the serum, detected by means of surface tension measurements, is not due directly to the antibody itself. Neither is it due to the direct action of the antigen on the serum, or to a change in the albumin-globulin ratio in this latter. Rather does it appear like antibody formation itself to be due to a tissue activity. Further experiments are being carried out on this subject.

1. The injection of antigen into an animal determines a gradual change in the blood fluid which finds expression in two physicochemical manifestations that can readily be followed, namely a decrease in the static value of the surface tension of serum solutions, and a special form of crystallization when serum diluted with isotonic sodium chloride solution is allowed to evaporate under certain conditions. 2. The change in the blood is at a maximum around the 13th day after the first antigen injection, and decreases progressively thereafter until it can no longer be observed, which is usually around the 30th day. 3. The change follows the same course, whether a single large injection of antigen is made, or many smaller ones. It begins at the same time in either case, it comes to a maximum after the same period, and in its subsequent course it is not affected by the reinjection of antigen. The manifestations of the change would appear to be independent of the presence of antigen in the circulation. 4. The mean length of the protein molecules of the immune serum obtained after the injection of the antigen dealt with is little if at all different from that of the protein molecules of normal serum. 5. It is possible that this reaction is independent of the antibody formation.

A technique is described for the study of the adsorption in function of the time of molecules of colloidal substances in the surface layer of water. This technique eliminates previous causes of error, in particular the individual coefficient. The curves obtained by plotting the data for serum at various dilutions give a clear picture of the phenomenon, and show that the low values of the initial surface tension of pure serum and serum at low dilutions are due to the extreme rapidity of adsorption of the molecules in the surface layer, owing to their high concentration. Evidence is given that colloidal substances in solution substantially lower the surface tension of the solvent, even when they are in too small a number to form an homogeneous monolayer.

An attempt was made to apply the assumption that a monolayer of serum exists at a certain dilution, in order to calculate the thickness of this layer or, that is to say, the mean value of one of the dimensions of the molecules of the serum proteins. The criterion taken for the existence of such a monolayer was the existence at a given concentration (1/11,000 for rabbit serum) of a maximum drop in the surface tension of serum solutions kept in watch-glasses. A series of preliminary experiments showed: 1. That the maximum drop in 2 hours took place, for the material used, at a concentration of 1/11,000, and that it always corresponded to an absolute minimum value of the surface tension of the solution, this minimum being quite sharp and well defined. 2. That adsorption took place on the glass as well as on the free surface of the liquid, and that apparently the same part of the molecule, in both cases, was drawn toward the water. 3. That the specific gravity of the anhydrous proteins of the rabbit serum studied was 1.275, whence it followed, on the basis of 6.51 per cent protein content, that the mean thickness of the protein molecules was 35.4 x 10(-8) cm. The same method applied to crystalline egg albumin, pH 6.8, in water, gave 52.8 x 10(-8) cm. for the probable molecular length.

The assumption has been made in preceding papers that, at a given concentration of 1:10,000, in the vessels used in our experiments, a monomolecular oriented layer was formed. Such a layer might be supposed to prevent the free escape of water molecules from the surface into the air, at least to a certain extent. In order to check this assumption, the rate of evaporation of solutions of serum at different concentrations was measured. It was found that, under the conditions of the experiments, in a progression of dilutions from 10(-1) to 10(-6), the slowest evaporation took place at a concentration of 1:10,000. In a few cases (less than 20 per cent), evaporation was slower at a different concentration, but always within the same range (between 10(-3) and 10(-5)), not far from 10(-4).