Recent studies, using low-temperature perfusion of rat kidneys, have claimed the existence of renal charge selectivity simply on the basis of the differential excretion rates of uncharged Ficoll and charged proteins. To test for the existence of charge selectivity in vivo, we examined the clearance of negatively charged Ficoll compared with uncharged Ficoll. A short-term approach to steady state was used to study the fractional clearances. Relative clearances were also examined using an osmotic pump technique where the tracers reach a steady-state value in conscious rats after 7 days. Carboxymethyl Ficoll was stable during filtration and renal passage, was not taken up by the kidneys, and did not bind to plasma proteins. There was no significant difference in the fractional clearance of molecules with radius of 36 A for Ficoll (fractional clearance = 0.048 +/- 0.038, n = 5) and negatively charged carboxymethyl Ficoll (fractional clearance = 0.028 +/- 0.019, n = 5). For molecules with radii greater than 36 A, carboxymethyl Ficoll had facilitated clearance with respect to uncharged Ficoll [for example, at a radius of 60 A fractional clearance for Ficoll = 0.0012 +/- 0.0005 (n = 5), whereas that for carboxymethyl Ficoll = 0.015 +/- 0.005 (n = 5)]. Renal function was not compromised by carboxymethyl Ficoll as uncharged Ficoll in urine exhibited similar hydrodynamic size profiles when studied in the presence of excess unlabeled carboxymethyl Ficoll. The facilitated clearance of negatively charged Ficoll with respect to uncharged Ficoll reveals a property of the capillary wall, which has been previously observed with other nonproteinaceous polyanions. This study demonstrates that the glomerular capillary wall is not charge selective in the form of excluding negatively charged Ficoll. However, the charge properties of the capillary wall may influence the facilitated transport of charged Ficoll compared with uncharged Ficoll.