Serum samples were collected from 3236 sows and 4712 market hogs in 1995. Sera were collected from sows on 226 farms, while market hog sera was obtained from 282 farms. Herds were randomly selected to participate in the 1995 National Animal Health Monitoring System (NAHMS) swine survey. Sera were assayed for antibodies against Toxoplasma gondii using the modified direct agglutination test. Herd data and serologic information were used to study the association between T. gondii infection in sows and specific herd characteristics and farm management practices. Overall, 15.1 percent of sows and 3.2 percent of finishers were positive for toxoplasmosis. Analysis of the data showed significant associations between toxoplasmosis in sows or herds and three factors: 1) method of rodent control, 2) type of production facility, and 3) access of cats to production facilities. In particular, seronegativity was associated with the use of "bait and/or traps only" for rodent control as compared to the use of cats for rodent control. Thus, use of cats as a method of rodent control should be avoided by producers. No regional differences in prevalence were detected and toxoplasmosis in sows was not associated with a reduced level of reproductive performance.