A total of 9600 swab samples from 900 carcasses originating from ten different abattoirs were subjected to bacteriological examination. Two sampling sites, brisket and forearm, consistently showed the highest contamination rates. The following sites are recommended for sampling: on the lateral side of the carcass neck, forearm, shoulder, brisket and abdomen. The neck is recommended for the medial side. Compared to the large variance of contamination either on individual carcasses or between different carcasses, the differences in the variance of results between double swab and incision sampling techniques should be of minor importance. Considering this big variance of colony counts, it is suggested to take five to six swab samples from each of at least ten to 15 carcasses once a month. With a view to a more differentiated and evident evaluation the results should be recorded in a 'box plots' and not in the form of mean values and standard deviations. The data confirms bacteriological monitoring of beef carcasses as a useful tool for the verification of slaughter hygiene.