Synaptic vesicles release neurotransmitter both actively (upon stimulation) and spontaneously (at rest). It has been long assumed that identical vesicles use both modes of release; however, recent evidence has challenged this view. Using several assays (FM dye imaging, pHluorin imaging and antibody-labeling of synaptotagmin), in neuromuscular preparations from Drosophila, frog and mouse as well as rat cultured neurons, we suggest that the same vesicles participate in active and spontaneous release.1 Synaptic vesicles fuse with the neuronal plasma membrane to release their neurotransmitter contents both upon the arrival of action potentials (active release) and at rest (spontaneous release). It has been assumed for almost six decades that identical vesicles exocytose in both cases, an assumption which has been used to generate the quantal (vesicular) release theory 1 . This assumption is in agreement with the fact that prolonged active 2 or spontaneous release 3 can both release essentially all vesicles (see also 4,5 ). The issue of whether the same vesicles can be released both spontaneously and actively has been tested recently by investigating the loading and unloading of styryl (FM) dyes during vesicle recycling 6 . In cultured hippocampal neurons, the vesicles loaded with FM dyes during active or spontaneous recycling were unloaded efficiently only by the same release paradigm. Also, inhibiting synaptic vesicle re-acidification (and therefore refilling with neurotransmitter) by use of folimycin during spontaneous release specifically depleted the spontaneous pool 6 . Surprisingly, both of these findings have been later contested in similar experiments -FM dyes taken up either at rest or during stimulation could be released with identical kinetics, and the inhibition of reacidification during depolarization led to a cessation of spontaneous release 7 (see also 5 ). However, later studies have again found different FM dye release kinetics when using different loading paradigms 8,9 , and labeling experiments using the expression of a biotinylated variant of the synaptic vesicle marker VAMP2 (synaptobrevin) also supported the hypothesis that spontaneously and actively recycling vesicles are different 10 .We tested here this controversial issue by combining several fluorescence imaging assays. We sought to reproduce the FM dye loading/unloading paradigms used in the past. A number of FM dyes have been used, including FM 1-43 and FM 2-10 6-9 ; both dyes have provided evidence for a difference between the spontaneously and actively recycling vesicles 8,9 . It has been recently suggested that FM 2-10 reports this difference better than FM 1-43 9 . However, as FM 2-10 is less bright when compared to FM 1-43, it is typically used at very high concentrations (400 µM; ~4-fold higher than its membrane dissociation constant 11 ). Since the FM dyes have a detergent-like structure, they inhibit vesicle recycling at high concentrations 11 . We therefore decided to employ FM 1-43 in our work (the concentration typically used,...