An improved technique for passive hemagglutination (HA) for titration of tetanus antitoxin was described. The use of highly purified tetanus toxoid and of improved diluent increased the specificity and reproducibility of the test. Several hundreds of specimens of guinea-pig serum taken at various stages of immunization were titrated by HA and toxin neutralization (NT) in mice. The ratio of HA to NT titers varied significantly depending on the immunization stage; higher at early stages and lower at later stages. The high HA/NT ratio was not due to the IgM antitoxin, which is very rare in guinea pigs. The variation in discrepancy between HA and NT titers decreased considerably by grouping the serum specimens with respect to the stage of immunization. Thus, it is possible to predict the in vivo titer of a tetanus antitoxin accurately enough for clinical study. The HA test may be useful as an alternative method for titrating tetanus antitoxin in the field trials. Moreover, it can be used for the study of characteristics of antitoxins.