The rate at which solutes may move in the cell wall apoplast of roots was assessed by measuring the diffusivity (D) of a dye, sulphorhodamine G (SR), in freshroots of field-grown maize. Pieces of axile and branch roots from regions with soil sheaths (with immature late metaxylem) and without sheaths (with mature late metaxylem) were placed in SR solution for 5, 20 and 80 min, removed, rinesed and freeze-substituted. The water-soluble SR was retained in position in the tissues during anhydrous embedding and sectioning. The distance of radial penetration of the dye was measured from micrographs taken in the fluorescence microscope. Fick's aecond law was used to calculate values of D. The values are expressed relative to the value of D in water (Daq). Mature axile roots uniformly gave values of D=Duq/1000. In immature axile roots D varied from 0 (no penetration through the hypodermal walls) through Duq/1000 to D aq/40 (the maximum value found). This variation probably arises from a developmental sequence. with D high in the region to 10 cm from the tip. zero in the region 10-25 cm from the tip, and low further back (merging with values for mature roots). In branches of both immature and mature roots also, D was highly variable, from zero to Daq/40. SR was found to penetrate through the endodermis of axes and branches, following sub-microscopic paths around the lining of pits in the secondary walls. It is concluded that soil solutes probably enter roots at the outermost layer of cells, and that the cell wall apoplast is most unlikely to be a path for flowing water.