Chart No. 1 shows that during this study the agglutination titre of only two cows changed to any extent. Of these two cows the titre changed from a slight reaction in low dilution to a high reaction in high dilutions. However, both gave birth to normal, healthy calves, one in February, the other in March. Cows No. 22, 28 and 309 were sold because they were sterile. Cows No. 220 and 251 were sold because on differential test they were classified as infected with Brucella abortus, and they were from the beef herd where no isolation is practiced. Cow No. 7 died of unknown cause, No. 30 died of peritonitis, and No. 42 died of tumors. Chart No. 2 shows that the reactions to the differential test remained practically the same in all the cows except numbers 68 and 76. Their differential agglutination titres rose correspondingly with their agglutination titres. Isolation of Brucella abortus was tried from each cow, provided proper samples could be had. Isolation from foetuses was possible in only three cases, as only three cows aborted during this study. Cow No. 21 aborted twin calves on August 23, 1929. Isolation attempts from the foetuses were negative as were the uterus and milk examinations. This abortion was not due to Brucella abortus. Heifer No. 315 aborted a six weeks old calf on November 24, 1929. Uterus and foetus examinations for Brucella abortus were negative. This heifer had never lactated so milk examinations were impossible. The cause of abortion was undetermined. Isolation of brucella abortus was attempted from each cow until the organism was isolated unless the cow stopped lactating, but no more than three attempts were made on any cow. By comparison of charts No. 1, 2, and 3, it can easily be seen that the isolation of Brucella abortus with one exception, cow No. 16, was from cows that gave high agglutination titre reactions, and in all cases from cows that gave infected reactions to the differential test. Records of isolation of Brucella abortus indicate that 14 out of 23 infected reactors to the differential test are known carriers of Brucella abortus. Repeated isolation tests were run on the other nine when proper material could be obtained. In no case was Brucella abortus isolated from the non-infected reactors to the differential test. / M.S.