Diphtheria toxoid-antitoxin precipitates formed in antitoxin excess can prepare guinea pigs, rats, and rabbits for a secondary type of antitoxin response. Priming may occur without the development of detectable serum antibody. In rats, toxoid-antitoxin precipitates are more efficient than "free" toxoid in priming, whereas in guinea pigs, the magnitude of the anamnestic response varies with the precipitate employed. The possibility that priming is due to "free" antigen released from the specific precipitate rather than the precipitate itself is discussed. The anamnestic antitoxin response can be inhibited by passive antitoxin, but less efficiently than primary antitoxin formation. Partial suppression of the secondary antitoxin response was accomplished by injection of excess horse antitoxin as long as 4 days after reimmunization with toxoid. The importance of these findings for the understanding of passive-active immunization in the human is discussed.