The relationship between foot process fusion (retraction) and proteinuria in glomerular diseases has been known for many years, but only a few quantitative studies have been performed. The quantitative morphometrical method used here makes it possible to estimate the distribution of true foot process width from the observed apparent width on electron micrographs. The true foot process width was analyzed in a consecutive material of patients with glomerulonephritis and related to the diurnal urinary protein excretion. Only a weak correlation was found between these two parameters. Albeit an abnormal foot process width was invariably associated with proteinuria, a rather large fraction of patients with gross proteinuria had normal foot process width compared to control subjects. Thus, if foot process fusion causes proteinuria, this can not be the only mechanism leading to this permeability alteration. On the other hand, our results show that foot process fusion is not a necessary consequence of proteinuria. In addition, a methodologic study was performed, analyzing the relationship between the quantitative and a semiquantitative method for estimating foot process fusion. It is concluded that the latter method is fast and sufficiently precise for most evaluations of foot process width in human kidney biopsies.