Some of the properties of myelin forms prepared by freezing, thawing, and homogenization of ghosts of human red cells are described, among these being the "apparent molecular weight" as found by light scattering, the effect of repeated freezing and thawing on the "apparent molecular weight," the quantity of fatty acids split off at various temperatures of freezing, the effect of the anticoagulant used, and the remarkable constancy in the "apparent molecular weight" met with when myelin forms are prepared from the ghosts of the red cells of normal people. Taken together, these observations lead to the description of a "standard technique" for the preparations of myelin forms; this technique can later be extended to the myelin forms obtained from animals other than man and also to those obtained from abnormal red cells.

The interaction between dextran and serum albumin, gamma globulin, and fibrinogen can be studied by an electrophoretic method, which depends on obtaining electrophoretic patterns, first of each protein, then of dextran, and finally of mixtures of each protein with dextran. The areas under the electrophoretic spikes for each protein, for dextran, and for the mixtures are measured. At pH's between 9.6 and 6.6, there is a transference of albumin to dextran when the two components are mixed, the amount of albumin lost being nearly equal to the gain in the new component, albumin plus dextran. This new species has a specific refractive index of about 0.00205 and seems to be composed of about one albumin molecule for every four dextran molecules. The method is unsuitable for studying interaction between dextran and gamma globulin because these substances are almost immobile and do not separate into two spikes. The method shows that a mixture of dextran and fibrinogen gives only one slowly moving spike (pH 6.6 to 8.6), the area under which is the sum of the areas under the dextran and the fibrinogen spikes taken separately. Either there is no interaction, or the new species has virtually the same refractive index increment as fibrinogen and dextran taken separately (0.0014 to 0.0015).

Ghosts prepared in CO 2 -saturated water from unwashed human red cells can be fragmented mechanically, but ghosts from thrice washed cells cannot. If the ghosts are prepared by freezing and thawing, this difference is not observed. The electrophoretic velocity varies also with the way in which the ghosts are prepared. The pH-mobility dependence of washed red cells flatten off to a plateau at pH 9, and the electrophoretic velocity is zero at about pH 2. Ghosts prepared by freezing and thawing have almost the same pH-mobility dependence, but if the ghosts are prepared in CO 2 -saturated hyptonic saline, the mobility at pH 9.4 is 0.75 times that of washed cells. Fragments of ghosts of unwashed red cells have a smaller mobility than that of the red cells. Trypsin reduces the mobility of washed red cells and of ghosts. Sols of lipid complexes (lecithin, cephalin, and lipositol), at varying pH's, have a mobility 1.2 times that of the washed red cell. The pH-mobility relation is otherwise similar. These complexes can be coated with dextran and trypsin.

The apparent protein content of a single spherical lymphocyte or a similar cell, as well as its diameter, can be measured by a modified and stabilized AO-Baker interference microscope which is either fitted with an AO half-shade eyepiece or connected to a photomultiplier. This paper gives results for the apparent protein content and the diameter of normal mouse lymphocytes and thymocytes; these do not differ significantly from each other. It also gives values for the apparent protein concentration and the diameter of the lymphocytes of mice of the same strain which have developed leukemia or lymphoma; these values are significantly different from those of normal mice. Related data are given for the apparent protein content and for the diameter of normal human lymphocytes and the white cells found in myeloblastic leukemia and in stem cell leukemia in man; here the values differ significantly from those of the normal human lymphocyte. The rule seems to be that the abnormal white cells become more watery and larger as the disease advances.

Sodium erucate reacts progressively ( i.e ., once the reaction is started in a time which is so short that the lysin is in contact with the red cells for 30 seconds, it cannot be stopped even by being diluted 10-fold) with human red cells at pH 7. At the same time, systems containing the lysin and human red cells show a zone phenomenon, lysis occurring most readily in a certain concentration of lysin but more slowly in larger or smaller concentrations. Sodium erucate-I 131 can be used to investigate both the zone phenomenon and the progressive character of the reaction. As regards the former, large concentrations of the lysin react relatively poorly with the red cell surfaces and the resistance of the red cells is relatively high. This may be due to the presence of an admixed inhibitor or to the development of an inhibitory state. The lysin is taken up and fixed by material in the red cell surface, so that the "internal phase" of lysin attached to the cell surfaces is so firmly fixed that a 10-fold dilution has no effect on it. It follows that lysis in these systems is progressive, as it is found to be.

Sodium oleate reacts progressively with human red cells at pH 7. By progressive is meant a reaction which is not adequately described as reversible or irreversible; such reactions cannot be stopped once they are under way, and are probably associated with a more or less stable "internal" lysin phase at the cell surfaces. The uptake of the lysin and the effect of dilution on the uptake can be studied by converting sodium oleate into the radioactive form, sodium oleate-I 131 . The uptake is a parabolic function of the lysin initially present in the system, and the effect of a tenfold dilution of systems in which red cells have remained in contact with the lysin for 2 minutes is to reduce the lysin taken up at the cell surfaces twofold. The lysin rapidly forms a relatively stable layer at the cell interfaces, and this layer is little affected by the dilution of the system as a whole.

Inhibition of hemolysis by plasma has been studied in systems containing saponin, digitonin, and sodium lauryl sulfate, using the methods developed for the study of the kinetics of progressive reactions. The results are that the progressive nature of the hemolytic reaction in saponin systems becomes less when the inhibitor is added, that the addition of inhibitor to digitonin systems has no effect on the final result although the velocity of the progressive reaction is reduced, and that the effect of plasma in lauryl sulfate systems is intermediate between the effects in saponin systems and digitonin systems. A simple explanation is that the lysin is very strongly fixed, to form an internal phase, to the cell surfaces in digitonin systems, less strongly in laurate systems, and still less strongly in saponin systems. To answer the question as to whether, in a system in which some of the lysin forms as internal phase, the addition of an inhibitor results in a redistribution of the lysin between the internal phase and the bulk phase, sodium lauryl sulfate-S 35 and sodium cetyl sulfate-S 35 were prepared, and their distribution between the internal phase and the bulk phase was measured before and after the addition of plasma, the lysins being added to the cells either before or after the addition of the inhibitor. The results show that there is a large uptake of these lysins at the red cell surfaces when they are added first, and that the subsequent addition of plasma greatly reduces the quantity of lysin held in the internal phase. Further, if the inhibitor is added first and the lysin subsequently, the internal lysin phase is very incompletely formed. Serum albumin, used in place of plasma, gives essentially similar results.

The quantity of a radioactive hemolysin, sodium dodecyl sulfonate-S 35 , taken up by red cells from concentrations too small to produce hemolysis varies with the lysin concentration, and does so in a way which can be described by an adsorption isotherm. Attempts to use color reactions or surface tension measurements to determine the quantity of digitonin, saponin, and the bile salts taken up by red cells from hypolytic concentrations have failed, principally because chromogenic, and also surface-active, substances are liberated from the cells when the lysin is added. Color reactions with the anthrone reagent show that digitonin and saponin are both taken up by or fixed to red cell ghosts; the extent of the uptake, however, is uncertain, again because of the liberation of chromogenic substances. Comparison of the results of the various methods which measure the apparent amount of lysin fixed, or utilized in reactions between lysins and red cells or ghosts show discrepancies between results given by direct methods (measurement of radioactivity or of color) and indirect methods (addition of a second population after lysis of a first, and dependence of the position of the asymptote of the time-dilution curve on the number of red cells). The discrepancies are traceable to the inhibitory effects of substances liberated from the red cells or ghosts. The ease with which a lysin, once taken up by red cells, can be detached by diluting the system determines the extent to which the hemolytic reaction is "progressive," but has no observed connection with the quantity taken up in the first place. There is now ample evidence that lysis in systems containing simple hemolysins is a process involving two stages in time and two phases, and that it is usually complicated by reactions between the hemolysin and liberated inhibitory material.