Thin section electron microscopic studies of stimulated secretory cells in the 60’s and the 70’s suggested that secretory vesicles often contact the cell plasma membrane at sites where the intervening cytoplasm has been expressed. It was believed that this results in the formation of the so called ‘pentalaminar figures’, which gave rise to a state preceding membrane fusion [1–7]. In the late 70’s and the early 80’s ultrastructural freeze-fracture studies on mast cells and the frog neuromuscular junction suggested that, following stimulation of secretion, indentations appear at the plasma membrane. It was believed that indentations form at points closest to the secretory vesicle membrane and in some cases invaginate, giving rise to a tiny ‘dimple or pore’. Some of these indentations were found to have made contact or fused with the secretory vesicle membrane [8, 9]. Morphological studies, using freeze-fracture electron microscopy of stimulated mast cells, showed the presence of plasma membrane depressions ranging in size from 50–100 nm in diameter, which were thought to form following a secretory stimulus . At the neck of these depressions or pores tiny protuberances called intramembrane particles were identified. In both stimulated and unstimulated mast cells small fibrils appeared to reach out to secretory vesicle membrane from the cytoplasm underneath the plasma membrane. The fibers appeared to be 50–70 nm in length and contact secretory vesicles found in close proximity to the plasma membrane (up to 100 nm), although the two membranes were separated by cytosol. Although firmly believed at the time, the total incorporation of the secretory vesicle membrane at the cell plasma membrane could not be demonstrated . Even after fusion of the vesicle membrane with the so-called ‘plasma membrane pore’ and full pore distension, the remainder of the granule membrane was always found well separated from the plasma membrane . Freeze-fracture studies [8, 11, 12] supported the view that fusion pores originated from specific sites at the plasma
membrane, perhaps at the fibrillar region. Analogous to the presumed role of clathrin in vesiculation, it was hypothesized that the fibers may regulate pore structure and dynamics .
Then during the late 80’s and early 90’s, electrophysiological measurements in mast cells [13–18] and in adrenal chromaffin cells  suggested the presence of fusion pores at the cell plasma membrane as a ‘dynamic entity’ . The results from cell capacitance and conductance measurements suggested that following stimulation of secretion, the ‘exocytotic fusion pore’ abruptly appears as a 1–2 nm in diameter pore at the plasma membrane, with conductance similar to a large ion channel . The fusion pore either irreversibly expands or closes [10, 14, 20, 21]. The latter process, where the fusion pore opens allowing secretory vesicles to fuse momentarily and subsequently closes, was referred to as transient fusion [14, 15, 17, 20, 21]. Patch-clamp measurements of rat mast cells were the first to suggest the existence of such transient fusions [18, 21]. Experimental data from these and other studies, and from theoretical considerations, gave rise to a working model of the fusion pore [10, 22). According to this model, the fusion pore is formed by the regulatory function of a group of proteins responsible for bringing the plasma membrane into a highly curved dimple, close to a tense secretory vesicle membrane. The resultant structure causes the plasma membrane and secretory vesicle membranes to form a ‘hemifusion’ intermediate and completion of fusion by formation of an aqueous pore following rupture of the shared bilayer .
One needs to be critically aware that these earlier studies, performed primarily using electron microscopy and electrophysiological approaches, were based on the assumption that following stimulation of secretion, the secretory vesicle membrane fully merged with the cell plasma membrane, consequently releasing the intravesicular contents to the outside by diffusion. The incorporated vesicle membrane at the cell plasma membrane is later retrieved by endocytosis. It is precisely due to this hypothetical concept of the secretory process in cells that electron micrographs of various stages of plasma membrane invaginations were often chosen and arranged in sequential order, in support of and giving the impression that the fusion pore is formed as a consequence of dimpling of the cell plasma membrane, which eventually makes contact and fuses with the secretory vesicle membrane. The vesicle membrane then completely merges (flattens out, for lack of a better term, becoming one with the plasma membrane) with the cell plasma membrane and the excess membrane is retrieved by compensatory endocytosis
at a later time. Since at the time there was no way of determining membrane structure and dynamics at nm resolution and in real time in live cells, these EM
studies and their supporting electrophysiological measurements enforced the dogma and became widely accepted. However, with this model of cell secretion, numerous observations such as the appearance of empty and partially empty vesicles following cell secretion, or no loss in vesicle number following cell secretion, could not be explained. The complete merger of synaptic vesicle membrane at the presynaptic membrane, and the compensatory retrieval of excess membrane by endocytosis, would be too tardy of a process at the nerve terminal. Moreover, the presence of neurotransmitter transporters at the synaptic vesicle membrane for refilling of spent vesicles would be of little use if vesicles completely fused at the membrane [23, 24]. In the past decade, the discovery of the new cellular structure – the ‘Porosome’, the universal molecular machinery for cell secretion, has resolved these conundrums.