Phosphoinositides have been in the limelight for over two decades following their identification as sources of the second messengers, diacylglycerol, inositol(1,4,5) trisphosphate (IP3) and phosphatidylinositol(3,4,5) trisphosphate (PtdIns(3,4,5)P3). The last 10 years have witnessed an explosion of experimental results that demonstrate that inositol lipids themselves should be considered as second messengers in their own right. The intact lipid molecule functions as a reversible recruiting device for proteins to transiently bind to membranes and as an allosteric regulator of many proteins. The inositol headgroup of phosphatidylinositol (PtdIns) can be phosphorylated at a single or a combination of positions (38, 48 or 58) to give rise to seven different phosphoinositides (Fig. 1). Phosphoinositides can control membrane trafficking via their structural role as membrane components and via their ability to engage specific protein domains (e.g., pleckstrin homology (PH) or FVYE) which bind with high affinity and specificity to the different phosphorylated species of phosphoinositides, and thus recruit molecular machinery driving specific membrane trafficking events and regulating actin assembly. The first hint that phosphoinositides were important in membrane traffic came from studies showing that a bacterial phospholipase C inhibited regulated exocytosis in PC12 cells [20] and from genetic studies in Saccharomyces cerevisiae where the secretory mutant, Sec14 showed defects in exit of secretory vesicles from the Golgi and where the gene product responsible for these defects was identified as the yeast phosphatidylinositol transfer protein (PITP) [5]. Subsequent studies led to the identification of PITP, the small GTPase ARF, phosphatidylinositol 4-kinases (PI4K), phosphatidylinositol 4-phosphate 5-kinase (PIP5K) and PIP phosphatases as regulators of membrane trafficking along the exocytic and endocytic pathways by virtue of their ability to control PtdIns(4,5)P2 levels [21, 30, 31, 34, 37, 44, 79]. As for 3-phosphorylated phosphoinositides, their role in membrane trafficking emerged with the identification of the product of the Vps34 gene (whose mutation was responsible for a defect in vacuolar protein sorting and membrane trafficking) as phosphoinositide 3-kinase (PI3K) which phosphorylates PtdIns to PtdIns(3)P [65, 66]. Since then many studies have provided evidence for a role of PI3Ks and their regulators, such as the small GTPase Rab5, in Golgi to endosomal/vacuole transport and along the endocytic pathway both in yeast and mammals and for the existence of a selective PI(3)P binding domain in some proteins, the FVYE domain [9, 25, 58]. Although the role of inositol lipids in membrane traffic appears to be central in yeast and in mammals, it is also clear that mammalian systems, due to their inherent complexity, have a richer and wider variety of functions where phosphoinositides are important. This is also apparent when one compares the inositol lipidmetabolizing enzymes and the inositol lipids identified in mammalian organisms and in S. cerevisiae (compare Fig. 1 and Fig. 2 and see also Table). PtdIns is the parent lipid from which all the phosphorylated species are derived (Fig. 1). In mammalian cells, PtdIns constitutes between 5–8% of total cellular lipids and the enzymatic Correspondence to: S. Cockcroft