It was made clear that the difference in effect of the dose of serum administered intramuscularly which suppressed_ the appearance of SR and which produced negative SR was recognized between 20 AU/kg and 50 AU/kg; furthermore, the effect of suppression was distinct when serum was administered within 1 hour both before and after Schick tests; it was independent on the dose of serum and SR appeared usually positive. The observation on the diphtheria patients treated with the antitoxic serum by Rosling(41), Mazima(42), Hayashi(48), , Fukumi(44, 45) and with human serum by Thelander(46), Barenberg, Greenstein & Leighton(47), in which the-relation between the serum injection and SR was studied from the early stage of disease or in a long duration, might be explained, partially by this paper.As it is suggested that the fixation of toxin in skin to tissue cells becomes firmer and firmer as time passes and it can not be dissolved and neutralized by antitoxin any more, if once fixed: on the other hand, the serum injected intramuscularly appears fairly rapidly in blood, rises to the highest concentration in 2 days and then decreases gradually, but circulates for over 1 month. It is suggested that the antidiphtheric serum should be administered as early as possible.Moreover, concerning the dose of serum, Schick, Kassowitz u. Busacchi (2), Schick (48) obtained the result showing that the effective dose is 100 AU/kg to, the mild, middle and good promising case, 500 AU/kg to the severe. Madsen(16) also recommended a dose of 100-300 AU/kg, Weichsel(4) 100 AU/kg on guinea pigs. Nakamura et al(20, 21) mentioned that observing the clinical courses and results of 289 diphtheria children divided into two groups, of the large' injection dose (over 500 AU/kg and of the small (200 AU/kg), with the chief object of complete recovery from the disease, the following result was obtained: they can be cured by 200 AU/kg without grave faults except the cases in which long time elapsed before treatment was started or very malignant cases. But it would be permitted to presume that it is generally effective to some extent to treat with 50 AU/kg of serum except in the malignant cases in which a large dose of toxin is produced probably abruptly, although it is feared to be irrational to discuss the clinical treatment of diphtheria from these experimental results.And then, as mentioned above, the antitoxic titer for reversal is in a fairly wide range and this problem is confirmed by multiple Schick tests by Glenny & Waddington(49), Parish & Wright(37) and Kurokawa et. al.(50), hence the significance of SR in antitoxin immunity must be investigated more extensively.