Life events and supportive social networks are often treated as separate independent variables in their relations with supposed dependent variables such as depressive disorder. It is important therefore to establish that they are independent of one another. One hundred and twenty one men and women attending psychiatric hospitals with depressive disorders were interviewed at the time of their initial contact by a research psychiatrist and a week later by a non-clinical research worker who enquired about life events (LEDS) and about social networks (IMSR). One hundred and eleven patients were successfully followed up in the same manner, typically 4 months later. Social networks were very stable over time. Where they did change, this could not be explained in terms of losses of social contact due to events such as death of a primary group member. The rating of threatening events and difficulties was also unrelated to the quality of social support networks. Institutional rearing in childhood was associated with substantially smaller adult primary networks. It was concluded that over short periods of time, measures based on the LEDS and IMSR are independent, but that major social disruptions may have life long effects on social support networks.