We have evaluated the epidemiological evidence for an aetiological role of varicella zoster virus (VZV) infection in the development of multiple sclerosis (MS). A MEDLINE search of the English language literature for 1965-99 identified 40 studies. These studies were categorized as seroepidemiological (13), case-control (23), historical cohort (2) or ecological (2). One study used both case-control and historical cohort methodologies. Studies were then classified according to methodological rigour, using criteria derived from published guidelines for the epidemiological study of MS. There was a large variability in the quality of evidence. The five studies with the best methodology failed to show an increased risk of MS associated with varicella or zoster infections. At the present time there is insufficient evidence to support an important aetiological role of VZV infection in the development of MS.