The aim of this study was to compare a domestic indirect enzyme-linked immunosorbent assay (DI-ELISA) and an in-house agar gel immunodiffusion (AGID) test with a commercial indirect ELISA (CI-ELISA) test for the detection of antibodies to bovine leukaemia virus (BLV). BLV proteins were harvested as described by OIE and used in both the DI-ELISA and AGID tests. Analysis of negative sera showed that consideration of a cutoff equivalent to three times the standard deviation value above the mean value of the negative control sera provided an acceptable specificity and reduced the risk of false positive results for the DI-ELISA test. From 460 serum samples, 425 (92%), 416 (90%) and 435 (94%) sera were found to be negative when using either the CI-ELISA, DI-ELISA or AGID test. Of the six serum samples which yielded suspicious results with the CI-ELISA, four were found to be positive by the DI-ELISA, but all of them were negative by the AGID test. DI-ELISA and AGID tests' relative (to CI-ELISA) sensitivities were 97% and 86%, respectively. DI-ELISA and AGID tests' relative (to CI-ELISA) specificities were 84% and 100%, respectively. Comparison of the results from a native breed, Sarabi, with Holstein showed that there is no significant (p<0.05) difference in the frequency of enzootic bovine leukosis between the two.