To determine the immune status of persons receiving botulinum pentavalent (ABCDE) toxoid and to evaluate the effectiveness of the vaccine, we surveyed immunized individuals for neutralizing antibodies to type A and to type B botulinum toxins. After the primary series of three immunizations administered at 0, 2, and 12 weeks, 21 of 23 persons tested (91%) had a titer for type A that was greater than or equal to 0.08 international units (IU)/ml, and 18 (78%) had a titer for type B of greater than or equal to 0.02 IU/ml. (One international unit is defined as the amount of antibody neutralizing 10,000 mouse 50% lethal doses of type A or B botulinum toxin). Just before the first annual booster, 10 of 21 (48%) and 14 of 21 (67%) people lacked a detectable titer for type A and for type B, respectively. After the first booster, all individuals tested had a demonstrable titer to both types A and B. Of 77 persons who had previously received from one to eight boosts of the toxoid, 74 (96%) had an A titer of greater than or equal to 0.25 IU/ml and would not require an additional booster, according to the recommendations of the Centers for disease Control. However, only 44 of 77 (57%) had a B titer of greater than or equal to 0.25 IU/ml. In each group by booster number, even the group having had eight boosts, at least one person would require reimmunization on the basis of B titer. There was a wide range of antibody levels among individuals at the same point in the immunization scheme. Results from an enzyme linked immunosorbent assay, with purified type A or type B neurotoxin as the capture antigen, were compared with neutralization test results on 186 serum samples for type A and 168 samples for type B. Statistically, the correlation coefficients for results from the two assays were high (r = 0.69, P < 0.0001, for type A and r = 0.77, P < 0.0001, for type B). However, due to the wide dispersion of values obtained, using enzyme-linked immunosorbent assay results to predict neutralizing antibody levels is unwarranted.