Thirty-two water buffalo (Bubalus bubalis) calves aged 6–10 months were used to evaluate serological responses to Brucella abortus strain RB51 (RB51) vaccination in a dose-response study and to compare the use of two selective media for the isolation of RB51. The animals were randomly divided into three treatment groups. Groups I-III received the recommended vaccine dose (RD) twice 4 weeks apart, RD twice 18 weeks apart and saline once, respectively. Lymph nodes were excised from the three groups and subjected to bacteriological examination to determine the frequency of detection of RB51. Pre- and post-vaccination blood samples were collected and tested for B. abortus antibodies using the buffered plate agglutination test (BPAT), complement fixation test (CFT), and dot-blot assay. Sera taken at all post-inoculation weeks (PIW) were negative for field strain B. abortus using the BPAT. Antibody responses to RB51 were demonstrated in all vaccinates but not in controls by CFT and dot-blot assay from 1 PIW up to 16 weeks following booster vaccination. The agreement for both assays was 80.7% and there was a linear interdependence with a Pearson's correlation coefficient value of 0.578. The frequency of isolation of RB51 from the two selective media used was not significantly different (P > 0.05).