This article attempts to review some of the advances made during the past few years in our understanding of the nature of the barrier presented by the endothelial cell wall and how it may contribute to the regulation of exchange between blood and tissues. It has concentrated on a small number of experimental techniques which have yielded information on the correlation between structure and function of the endothelial cell wall and which have emphasized the potentially dynamic characteristics of the barrier. Whilst there now seems to be little dispute as to the location of the fluid conducting channels across the endothelial cell wall, within the clefts, fenestrae and in inflammation the open cell junctions, it has proved difficult to identify the molecular filter which limits macromolecular exchange across these pathways. In fenestrated endothelium it has been suggested that the filter resides at the fenestral diaphragms or in the underlying basement membrane, while in continuous endothelium there is strong support in the literature that the filter is located within the intercellular cleft, at regions of closely apposed cell membranes, or in the case of a vesicular pathway, at the necks or diaphragms of the vesicle openings. Alternatively, there is a considerable and increasing body of experimental evidence that macromolecular movement is retarded by the endothelial cell coat which lines the whole of the endothelial cell surface and covers the openings of interendothelial cell clefts, fenestral diaphragms and vesicle openings. It is believed to comprise glycoproteins secreted and regulated by the endothelial cells themselves and to have associated with it plasma proteins, particularly serum albumin. Expression of this glycocalyx and its modification have been demonstrated in vivo and in cultures of isolated endothelial cells, in vitro. Experiments using single microvessels in which a correlation between structure and function can be most readily made, offer further evidence that the clefts between endothelial cells are quantitively more than sufficient in extent to accommodate the fluid fluxes measured in even the most highly permeable vessels. They further demonstrate that the dramatic increases in fluid flux seen in inflammation result from a modulation of endothelial cell shape to form interendothelial cell gaps by activation of intracellular contractile mechanisms, mediated by changes in intracellular calcium. Increases in macromolecular leakage may only be seen when gap formation is accompanied by extensive modulation of the intercellular cement substance, or glycocalyx filling those gaps.(ABSTRACT TRUNCATED AT 400 WORDS)