Work on ion transport in plant cells and tissues is largely concerned with the properties of cells rather than of cell layers, and the evidence is on the whole against an important role for asymmetric ion transport across cell layers equivalent to animal epithelia. Cells structurally specialized for transport, having a large increase in surface area, seem to occur singly or in small groups, but not to be organized into closely packed layers; they are known as transfer cells and were described in a wide variety of situations by Gunning & Pate (1969). At the cell level, ion transport is best characterized in giant algal cells, but the situation may well be similar in higher plants. In Nitella translucens an ouabain-sensitive ATP-dependent sodium-potassium exchange pump at the plasmalemma maintains the high K/Na of the cell, and the high internal osmotic pressure is achieved by net salt uptake by a (chloride + cations) pump, also at the plasmalemma. The linkage between chloride and cations seems more likely to be chemical than electrogenic. This pump may be energized by a membrane redox system, but is not ATP-powered. The mechanism for initial entry of chloride to the cell seems also to control the distribution of tracer chloride between cytoplasm and vacuole, since the two processes of entry and transfer to the vacuole are very closely linked. The kinetics of vacuolar transfer are consistent with a pinocytotic entry of salt at the plasmalemma, fusion ofpinocytotic vesicles with the endoplasmic reticulum, from which new vacuole is formed. The process of discharge to the vacuole seems to be quantized, but the mechanism and significance of this observation are not understood.