At the Animal Health Service in the Netherlands initially an absorbed ELISA (IDEXX/CSL) based on the use of an anti-bovine conjugate was validated. For the European market, the manufacturer replaced this test by an absorbed ELISA with a protein G conjugate (HerdCheck M. pt., IDEXX, Scandinavia). We evaluated the specificity and relative sensitivity - as compared with faecal shedding - of this new commercially available absorbed ELISA (test B), in comparison with the previously validated IDEXX/CSL ELISA (test A). For estimation of the sensitivity, positive faecal culture was taken as a gold standard. Serum samples were obtained from 74 faecal culture positive cows, For test A and B, we found a sensitivity of 35 and 49 %, respectively. On a panel of 199 sera from once or twice faecal culture positive cows, belonging to low prevalence herds participating in an eradication program, we found a sensitivity of 27,3 % and 36,9 %, respectively. For test B, at a higher cut-off (S/P ratio > 0.3), for these serum panels the sensitivity was 40% and 27.8 %, respectively. The specificity of test A was evaluated using 821 sera from 36 herds, all herds being tested three times negative by pooled faecal culture. We found a specificity of 99.4 %. The specificity of test B was initially evaluated on a subset of 50 sera from three highly unsuspected Dutch herds and on a set of sera from France (n=260) and the USA (n=255) from herds that were found repeatedly negative after individual faecal culture. We found a specificity of 100 %, 99.4 % and 99.6 %, respectively. However, using a larger panel (n=339) from three highly unsuspected Dutch herds, a specificity was found of 98%. We decided to interpret results of test B at a higher cut-off (S/P ratio > 0.3) corresponding with a specificity of 99.4 % and a relative sensitivity of 40% for our herd certification program. In this voluntary program, all cows of three years and older are tested in the absorbed ELISA, and all serological reactors are confirmed by individual faecal culture. Preliminary results show, that 30 out of 46 ELISA positive (S/P ratio > 0.3) cows were confirmed by faecal culture, and a correlation was observed between S/P ratio and percentage of positive cultures. For a more extensive evaluation of the specificity of test B in Dutch herds, 1377 blood samples were taken from 28 herds, all herds being tested four times negative by pooled faecal culture. Surprisingly, at the relative high cut-off we use (S/P ratio > 0.3) we found a specificity of 96.5 % corresponding with 48 seropositive cows. However, 36/48 positives originated from five herds. Excluding these herds from the data set yielded a specificity of 99%. Further research is carried out to decide whether these seropositives are true-positives or false-positives.