This review summarizes biogenesis, composition, intracellular transport, and possible functions of trichocysts. Trichocyst release by Paramecium is the fastest dense core-secretory vesicle exocytosis known. This is enabled by the crystalline nature of the trichocyst "body" whose matrix proteins (tmp), upon contact with extracellular Ca , undergo explosive recrystallization that propagates cooperatively throughout the organelle. Membrane fusion during stimulated trichocyst exocytosis involves Ca mobilization from alveolar sacs and tightly coupled store-operated Ca -influx, initiated by activation of ryanodine receptor-like Ca -release channels. Particularly, aminoethyldextran perfectly mimics a physiological function of trichocysts, i.e. defense against predators, by vigorous, local trichocyst discharge. The tmp's contained in the main "body" of a trichocyst are arranged in a defined pattern, resulting in crossstriation, whose period expands upon expulsion. The second part of a trichocyst, the "tip", contains secretory lectins which diffuse upon discharge. Repulsion from predators may not be the only function of trichocysts. We consider ciliary reversal accompanying stimulated trichocyst exocytosis (also in mutants devoid of depolarization-activated Ca channels) a second, automatically superimposed defense mechanism. A third defensive mechanism may be effectuated by the secretory lectins of the trichocyst tip; they may inhibit toxicyst exocytosis in Dileptus by crosslinking surface proteins (an effect mimicked in Paramecium by antibodies against cell surface components). Some of the proteins, body and tip, are glycosylated as visualized by binding of exogenous lectins. This reflects the biogenetic pathway, from the endoplasmic reticulum via the Golgi apparatus, which is also supported by details from molecular biology. There are fragile links connecting the matrix of a trichocyst with its membrane; these may signal the filling state, full or empty, before and after tmp release upon exocytosis, respectively. This is supported by experimentally produced "frustrated exocytosis", i.e. membrane fusion without contents release, followed by membrane resealing and entry in a new cycle of reattachment for stimulated exocytosis. There are some more puzzles to be solved: Considering the absence of any detectable Ca and of acidity in the organelle, what causes the striking effects of silencing the genes of some specific Ca -release channels and of subunits of the H -ATPase? What determines the inherent polarity of a trichocyst? What precisely causes the inability of trichocyst mutants to dock at the cell membrane? Many details now call for further experimental work to unravel more secrets about these fascinating organelles.