The primary function of rat, mouse and human descending colon is the dehydration of faeces and the recovery of fluid and electrolytes from the intestinal digesta. In humans the colon receives around 1.5 L of fluid a day with electrolytes approximately at the same concentration as plasma. Small intestinal contents arrive in the colon at a water : solid ratio of 20:1 and leave having been dehydrated to a ratio of 2: 1. As in other epithelia water transport is secondary to active solute transport via membrane ion channels and transporters. The majority of fluid absorption in the descending colon occurs through the surface epithelium against a low hydraulic resistance. However as the luminal contents become more dehydrated it requires a significant force to further dehydrate faeces. Faeces have a similar structure to clay and dehydration on this scale requires a large compressive force of around 4 atm. The generation of a sufficiently hypertonic absorbate could provide the osmotic force required to transport water against a large hydraulic resistance and evidence is accumulating that this process is localized in colonic crypts. Colonic crypts are surrounded by a layer of myofibroblast cells and extracellular matrix components termed the pericryptal sheath. This sheath provides the physical barrier required to create a ‘central’ hypertonic compartment allowing fluid absorption via crypt epithelial cells and generating a suction force within the crypt lumen. Initial evidence for the process of fluid absorption via crypts was provided by in vitro studies of colonic mucosa showing concentration polarization of fluorescently labeled dextrans within colonic crypt lumens¹. Additional in vivo experiments using agarose gel cylinders inserted into the colon showed that the descending colon, but not the caecum, is able absorb against large hydraulic resistances via colonic crypts². To characterize further the process of fluid convection into crypts, photobleaching studies of fluorescently labeled dextrans were performed in isolated rat descending colonic mucosa and showed that recovery of fluorescence within the crypt lumen was the same for widely different molecular weight dextrans and abolished by inhibition of epithelial Na transport. These studies provide evidence that fluid flow in colonic crypts is convective³. The generation of a sufficiently strong suction pressure at crypt openings requires a highly hypertonic compartment, postulated to be localized in the pericryptal space surrounding crypts. Confocal microscopy performed in vivo in mice using a low affinity ratiometric Na sensitive dye confirmed the presence of a pericryptal hypertonic compartment with a local Na concentration of 200-400 mM, which would be sufficient to provide the necessary osmotic pressure for convective flow into crypts. Inhibition of Na transport significantly reduced pericryptal Na concentration as did blockage of crypt opening using paraffin oil suggesting that pericryptal hypertonicity is dependent on salt transport from the crypt lumen. Colonic fluid transport can be altered by in a number of states resulting in changes in colonic crypt function. High doses of ionizing radiation result in a reduction in fluid absorption. Examination of colonic mucosa post-irradiation showed that this reduction in fluid absorption was associated with increased leakiness of colonic crypt cells and the loss of cell adhesion molecules. Additionally it was shown that the pericryptal sheath was significantly disrupted after radiation and these changes were preceded by the release of apoptotic enzymes and signaling molecules. The loss of the pericryptal sheath and crypt epithelial adhesion molecules coincided with increased permeability of FITC dextrans out of crypts. In contrast to irradiation, low dietary Na results in increased fluid absorption through increases in plasma aldosterone and angiotensin II. Investigation of the effects of low Na diet in rat descending colon revealed a significant trophic effect on myofibroblast cells and extracellular components of the pericryptal sheath in contrast to rat proximal colon. These changes suggest that increased pericryptal barrier function may be contribute to increased fluid absorption after low Na diet. Several in vivo and in vitro studies of colonic crypt function both in health and disease now provide evidence that crypts play an important role in colonic fluid absorption. Further studies are required to measure important determinants such as crypt luminal pressure gradients and the role of active secretion in states of disease on crypt and pericryptal function.