1. Using confocal microscopy, the rate of fluid absorption into isolated perifused descending rat colonic crypt lumens is estimated from the concentration polarization and distribution of fluorescein sulphonate (FS) and fluorescein isothiocyanate-dextran (FITC-dextran; molecular weight, 10,000) within the crypt lumens and pericryptal fluid. 2. The probe dyes enter the crypt via the luminal opening, are concentrated in the lumen, then escape into the pericryptal space via the paracellular spaces spanning the crypt wall. 3. FITC-dextran is maximally accumulated at a luminal depth of 60 microns to 5 times the concentration at the crypt opening (p < 0.001) and penetrates 150-200 microns along the lumen. FS is maximally accumulated within crypt lumen close to the opening. At crypt luminal depths 10-60 microns from the opening FS is accumulated by a factor of 1.5-2.0 above that found in HgCl2-treated tissue (p < 0.001). 4. Dye enters the crypt lumen slowly from the basal side, but from this side does not accumulate above the bathing solution concentration. 5. HgCl2 (20 microM) or theophylline (10 mM) completely inhibit concentrative accumulation of FITC-dextran and FS within the crypts and pericryptal space (p < 0.001). 6. Computer simulation of the dye uptake indicates that the rate of water flow into the crypt luminal opening is 1 x 10(-3) cm s-1 which is equivalent to 15 microliters (cm mucosa)-2 h-1. Approximately 75% of the fluid entering the crypt is abosrbed across the proximal 50 microns of crypt wall as a consequence of the large osmotic pressure gradient between the pericryptal and crypt luminal solutions. A pericryptal diffusion barrier with lower permeability than that across the crypt wall is required to simulate dye accumulation in the pericryptal space. Differences between FITC-dextran and FS accumulation are explained by the lower diffusion coefficient within the crypt lumen, and lower crypt wall permeability of FITC-dextran.