The present study evaluated a commercial heterologous enzymeimmunoassay (EIA) method for the determination of prolactin (PRL) in canine blood plasma and compared the results found with this heterologous method to the PRL concentrations measured by use of a commercial homologous PRL EIA. The heterologous PRL EIA was characterized by intra-assay coefficients of variation (CV %) of 3-5 % for replicate determinations, and inter-assay CV %'s of 5-7 % between means of duplicate determinations. The least detectable concentration was 0.1 microgram/L. A satisfying accuracy was settled by a recovery of added PRL not different from 100 % at varying PRL levels. The cross-reactivity (CR) to canine growth hormone (GH) was 5 %. Both methods outlined equally well the changes in PRL concentrations induced by administration of metoclopramid and bromocriptin; the specificity of both methods was further confirmed by a lack of response to clonidin GH stimulation test. The results achieved with the two methods were not identical despite of identical dose response relations. This difference in test results was, however, considered to be of minor practical importance if locally established reference ranges are used for evaluation of test results.