Since their first description in 1988, glycopeptide-resistant enterococci (GRE) have emerged as a significant cause of nosocomial infections and colonisations, particularly in Europe and the USA. Two major genetically distinct forms of acquired resistance, designated VanA and VanB, are recognised, although intrinsic resistance occurs in some enterococcal species (VanC) and a third form of acquired resistance (VanD) has been reported recently. The biochemical basis of each resistance mechanism is similar; the resistant enterococci produce modified peptidoglycan precursors that show decreased binding affinity for glycopeptide antibiotics. Although VanA resistance is detected readily in the clinical laboratory, the variable levels of vancomycin resistance associated with the other phenotypes makes detection less reliable. Under-reporting of VanB resistance as a result of a lower detection rate may account, in part, for the difference in the numbers of enterococci displaying VanA and VanB resistance referred to the PHLS Laboratory of Hospital Infection. Since 1987, GRE have been referred from >1100 patients in almost 100 hospitals, but 88% of these isolates displayed the VanA phenotype. It is possible that, in addition to the problems of detection, there may be a real difference in the prevalence of VanA and VanB resistance reflecting different epidemiologies. Our present understanding of the genetic and biochemical basis of these acquired forms of glycopeptide resistance has been gained mainly in the last 5 years. However, these relatively new enterococcal resistances appear still to be evolving; there have now been reports of transferable VanB resistance associated with either large chromosomally borne transposons or plasmids, genetic linkage of glycopeptide resistance and genes conferring high-level resistance to aminoglycoside antibiotics, epidemic strains of glycopeptide-resistant Enterococcus faecium isolated from multiple patients in numerous hospitals, and of glycopeptide dependence (mutant enterococci that actually require these agents for growth). The gene clusters responsible for VanA and VanB resistance are located on transposable elements, and both transposition and plasmid transfer have resulted in the dissemination of these resistance genes into diverse strains of several species of enterococci. Despite extensive research, knowledge of the origins of these resistances remains poor. There is little homology between the resistance genes and DNA from either intrinsically resistant gram-positive genera or from the soil bacteria that produce glycopeptides, which argues against direct transfer to enterococci from these sources. However, recent data suggest a more distant, evolutionary relationship with genes found in glycopeptide-producing bacteria. In Europe, VanA resistance occurs in enterococci isolated in the community, from sewage, animal faeces and raw meat. This reservoir suggests that VanA may not have evolved in hospitals, and its existence has been attributed, controversially, to use of the glycopeptide avoparcin as a growth promoter, especially in pigs and poultry. However, as avoparcin has never been licensed for use in the USA and, to date, VanB resistance has not been confirmed in non-human enterococci, it is clear that the epidemiology of acquired glycopeptide resistance in enterococci is complex, with many factors contributing to its evolution and global dissemination.