Serum samples were obtained from blood donors in eleven selected locations in Ireland and tested for antibodies to Borrelia burgdorferi s.l. by enzyme immunoassay (EIA) and immunoblot (IB). The highest seroprevalence (8.7%) was found in Portumna, an area rated as high risk because of the presence of public access woodland harbouring both ticks and spirochaete reservoir hosts. The lowest seroprevalence (0.0%) was found in Ballina, an area where there is no suitable woodland though ticks are common on pastures grazed by cattle and sheep. These results support the findings of an earlier less extensive study and suggest that infection with B. burgdorferi s.l. in Ireland is mainly associated with mixed woodland containing deer and a variety of spirochaete reservoir hosts. The lack of association of tick-infested farmland with seropositive samples suggests that exposure to ticks alone is not a reliable indicator of risk of Lyme borreliosis. This is supported by the fact that none of 38 park rangers, all of whom are regularly bitten by ticks, were seropositive for B. burgdorferi s.l. The low overall seroprevalence of 3.4% in Ireland correlates with the rarity of clinical cases compared with continental European countries, and is due in part to the scarcity of high risk Lyme borreliosis habitat.