Abstroct. The urine and plasma content of certain proteins was studied near term in healthy gravidas, in gravidas with essential hypertension, proteinuria, hypertension of pregnancy and gestosis (pre-eclampsia; toxaemia). besides this the ratio of urine-plasma concentrations for each protein was calculated. In gestosis only the urine-plasma ratio of ceruloplasmin differed from those of the other proteins, but in the other diseases studied this ratio showed no difference from that in healthy gravidas. The amounts of different proteins excreted in urine did not correlate with the molecular size of the respective protein.A n u m b e r of studies have shown that gestosis (preeclampsia; toxaemia) is always accompanied by thickening of the glomerular basement membrane a n d swelling of the endothelium (8,9), which is a result of disseminated intravascular coagulation (4, 11). T h e proteinuria in gestosis is the result of the increased glomerular permeability present in this disease (2). It has been s h o w n that the clearance rate of plasma proteins is dependent upon molecular weight a n d that it decreases in proportion with higher molecular weight (3, 5). This simple selectivity is also found in gestosis (10).Attempts have been made t o find, f o r s o m e proteins, a urine-plasma ratio characteristic of gestosis (6, 12) b u t no definitely specific protein has been detected. Since the discovery o f such a specific protein might be of great importance in differentialThe series studied consisted of 15 healthy pregnant women and 22 gravidas with essential hypertension, 3 with proteinuria of late pregnancy, 17 with hypertension of late pregnancy and 8 with gestosis. The criteria of gestosis were a pathological blood pressure and proteinuiia. The upper limit of normal blood pressure was set at 140/90 mmHg and the highest normal proteinuria at 0.3 g/24 h, measured by Esbach's method. Gestosis was considered as a complication of late pregnancy if it had begun in the 26th week of gestation at the earliest. At the time of study the pregnancies were all in the 36th4Oth week of gestation. All the fetuses were alive.Of the collected 24-hour urine, 400 ml was concentrated to 10 ml. Concentration was done under nitrogen pressure, using a magnetic mixer and ultrafiltration apparatus (Amicon N.V., Oosterhout, Holland). The same manufacturer's filter paper XM-50 was used, allowing particles of molecular weight 50 OOO to remain in the solution to be studied.Twenty ml of blood was obtained from the cubital vein and centrifuged immediately. The plasma was studied without concentration.From both groups of samples 8 proteins were determined in mg per to0 ml, using radial immunodiffusion containing specific antiserum (Partigen plates, Behringwerke AG, Marburg, BRD)' (7). The standard sera of the same manufacturer were used as standards. The standard deviation of the method was less than 2% of the mean.The proteins to be studied were selected so as to be representative of different molecular sizes, but nevertheless were within the limits...