The effective control and eradication of brucellosis in developing countries can be achieved by quick and accurate diagnosis and effective vaccination. Protective efficacy of existing vaccines as well as diagnostic accuracies of various serological tests have not been thoroughly studied and verified in buffaloes, the prime dairy animal of South Asia. In the Present studies Brucella abortus strain 19 (S19) and strain RB51 (SRB51) vaccines were tested for inducing humoral and cell mediated immune responses in riverine buffaloes of various age groups. The protective efficacy of these vaccines against abortions was also evaluated by challenge protection studies. Protein antigen of Brucella abortus vaccinal strains involved in protection was identified by western blot analysis. Moreover an indirect enzyme linked immunosorbent assay (I-ELISA) for detection of anti-brucella antibodies in buffaloes was developed. The diagnostic sensitivity and specificity of developed test was compared with commercial ELISA test. Agreement between the two tests for negative sera was 100% and for positive sera it was 78%. In the vaccination study using S19 the peak antibody titres (P<0.05) were observed 30 days post vaccination using I-ELISA and Rose Bengal plate test (RBPT). The mean ELISA titre persisted for 3 to 4 months in the vaccinated buffaloes but waned quickly in calves than adults and heifers. The agglutinating antibodies detected by RBPT declined below the analytical threshold earlier than ELISA. Vaccinated and non vaccinated control animals were challenged subcutaneously with 5× 109 cfu of virulent Brucella abortus during 1st gestation after vaccination. Significantly lower (P<0.1) abortions were observed in vaccinated animals. In SRB51 vaccinates significant increase (P<0.05) in the humoral immune responses were detected in all the three age groups by an I-ELISA using acetone killed SRB51 antigen 30 days post vaccination. However, all SRB51 vaccinates were detected negative by RBPT. The vaccinated and control animals, when challenged with locally isolated virulent biovar 1 between 6-7 months of gestation, revealed non-significant (P>0.1) differences between the probabilities of abortion occurrence. However, the percentage of abortion occurrence in vaccinated animals (44%) was lower than the non-vaccinated buffaloes (78%). Lymphocyte proliferation assays detected moderately high and consistent proliferative responses in all of the S19 vaccinated adults, heifers and calves after 6 weeks of immunization. The lymphocyte proliferative responses of SRB51 vaccinate were low and inconsistent and only 77% of the immunized animals exhibited positive proliferative responses. Western immunoblot analysis using antisera raised against all three B. abortus strains in buffaloes indicated a near similar pattern of immune-reactive OMPs in S19 and field strain. Two OMPs of molecular weight 37-38 and 19 kDa were immuno-reactive in all strains in buffaloes. Two distinct proteins of molecular weights of 190.5 and 151.3 kDa were identified in field strain but not in both of the vaccinal strains of B. abortus.