The authors estimated the rates of development of serum antibodies against Brucella abortus, between January 1987 and December 1990, in female Jersey cattle and buffalo weaned from seropositive and seronegative dams, and used logistic regression analysis to examine the epidemiological relationship of these rates with the serum antibody status of dams. The offspring from both seropositive and seronegative dams were reared together in calf pens, while separate pens were used for bovine and buffalo calves. Each of the bovine calves was manually fed two litres of bulked milk (pooled from seronegative and seropositive cows) twice a day, in the morning and evening. The buffalo calves, however, were allowed to suckle their respective dams before and after manual milking of each buffalo, in the morning and evening. Bovine calves and buffalo calves were weaned at approximately six months of age and moved to sheds for young livestock. At maturity, the female offspring were artificially inseminated, and pregnancy was subsequently diagnosed by rectal palpation. The rate of seroconversion in the progeny of seropositive Jersey cattle was 26.4% (14 of 53 animals) compared to 14.3% (11 of 77 animals) in the progeny of seronegative cattle; this difference was non-significant (P = 0.1342). In buffalo, however, the rate of seroconversion in the progeny of seropositive dams (43.3% = 13/30) was significantly higher (P = 0.0002) than in the progeny of seronegative buffalo (8.8% = 6/68). The logistic regression analysis revealed that the progeny of seropositive buffalo were approximately 6.2 times more likely to have developed serum antibodies by the time of first calving than the progeny of seronegative buffalo (adjusted odds ratio [OR]) = 6.2; 95% confidence interval [CI] = 1.5, 36.4). This relationship was non-significant, however, for the progeny of seropositive cattle (adjusted OR = 2.1; 95% CI = 0.6, 7.4). The implications of these results and potential sources of bias are discussed.