Rates of infection with Borrelia burgdorferi were compared in adult Ixodes dammini ticks collected from deer at one coastal and two island sites with those collected from vegetation at the same sites. Ticks were examined using polyclonal direct fluorescent antibody. Spirochetes were observed in 47% of 288 ticks from vegetation as opposed to 13% of 276 ticks from deer (chi 2, P < .001). This disparity was increased when only male ticks were compared. Among female ticks from deer, the infection rate was higher in flat than in engorged ticks. These findings may reflect spirochete loss from ingestion of borreliacidal antibodies in deer blood or may result from factors related to the sensitivity of direct fluorescent antibody methods. They indicate that erroneously low estimates of regional risk for
Lyme disease may be obtained if ticks removed from deer are included in the determination of tick infection rates.