The phenomenon of agglutination of plague bacilli by the serum of patients was first observed by Wissokowitsch and Zabolotny in 1897[8] and later confirmed by the German Plague Commission. Vagedes, Klein, and others[8] pointed out the defects of the reaction as a diagnostic means. Aside from the technical difficulties, the reaction was found inconstant, and its occurrence was not noticed until the second week of the disease and even then only in low dilutions of the serum. [8] Referred to in Kolle und Wassermann: Handbuch der pathogenen Mikroorganismen (1903), =2=, 524. Although the recent work of Strong[9] and of Strong and Teague[10] has reduced the technical difficulties, the fact remains that positive agglutination of plague bacilli by the patient's serum cannot be obtained in the first week of the disease, and, therefore, the isolation of plague bacilli from the body of the patient is still the only quick and safe method of plague diagnosis. [9] The Philippine Journal of Science, Sec. B. (1907), =2=, 155. [10] _Ibid._ (1912), =7=, 194-201. Having utilized the technic devised by Teague, I have had no difficulty in performing the agglutination test in plague. The emulsion of plague bacilli, to be used for the test, was prepared by suspending young cultures of virulent plague bacilli, grown at 30° C., in salt solution and filtering the suspension through filter paper. No antiseptic was added nor heat applied. Serial dilutions of unheated patient's serum were mixed with equal amounts of bacterial suspension in small test tubes. Incubation at 35° C. followed. Controls, consisting of serial dilutions of normal human serum as well as bacterial suspensions without serum, excluded any possible error which might have been caused by spontaneous sedimentation of the bacterial suspension; while a parallel test with highly agglutinant serum facilitated the reading of positive results. Altogether, 22 tests were performed on 15 patients, 11 of whom were fatal cases and 4 of whom recovered. In the negative reactions, the duration of the disease at the time of examination ranges from two to six days. The non-fatal cases showed slight agglutination from the sixth day on. From that day, the agglutination titer of the serum was found to rise, and the agglutinins persisted in the blood of convalescents up to the seventh week of the disease.[11] [11] It is hoped that it will be possible to examine some of the survivors for agglutination from time to time.