The Addis count, or number of cells excreted in the urine in a fixed time, has been assumed to give the best estimate in quantitative determinations. The method is based on the assumption that an inverse relationship exists between the number of cells per unit volume and the diuresis.
Some of the factors involved in the procedure were studied in a material comprising 75 males without urinary tract diseases. There was no correlation between the number of erythrocytes per unit volume and the volume in which they were excreted, a slight negative correlation being found between the number of white cells per unit volume and the diuresis. This observation lends some support to the hypothesis that under normal conditions erythrocytes enter the urine through the glomeruli, white cells being sloughed off the lining of the urinary tract.
In the determination of excretion rates the random variability of haemocytometer counts may cause great variations, while the diuresis, under standardized conditions, shows small variations. At high diuresis the concentration of the urine, as reflected in the specific gravities, becomes so low that hypotonic haemolysis may occur. The precision of results expressed as excretion rates is unknown, because the arithmetic applied makes it impossible to evaluate the precision of the main variable: the haemocytometer count. In the author's opinion the method ought therefore to be abandoned. The cells should be counted in 1 mm3 of urine and the result expressed as number of cells per mm3, the magnitude of the theoretical error due to the random distribution of cells in the haemocytometer being approximately known.