The hygienic condition of the manufacturing beef collected at six carcass breaking plants for dispatch to hamburger patty manufacturing plants was examined. At each plant, 24 x 1 kg samples of meat were selected at random from the product being collected into bulk containers. Total aerobic, coliform and Escherichia cell counts per gram were enumerated for each sample. The log mean ((X) over bar) and standard deviation (s.d.) were calculated for the log(10) values of each set of 24 counts, on the assumption that the distribution of counts approximated the log normal. A value for the log(10) of the arithmetic mean (log A) was calculated for each set from the values for X and s.d. Log A values for total, coliform and E. coli counts ranged from 3.5 to 4.9, 0.7 to 3.0 and 0.2 to 2.6 log(10) cfu g(-1), respectively. For each set of samples, there was a weak or no correlation between log,, values for total counts and those for coliforms or E. coli, but correlations between log(10) values for coliform and E. coli counts ranged from non (R(2)=0.19) to close (R(2)=0.97). The results show that there are large differences between plants in the numbers of E. coli in the manufacturing beef which they produce. The differences in E. coli numbers are not reliably reflected by differences between total or coliform counts in product from different plants. In addition, the findings indicate that manufacturing beef obtained from culled cow carcasses may generally be less heavily contaminated with E. coli than the trimmings obtained from the carcasses of feedlotted steers. Clearly, some current processes for the production of manufacturing beef destined for hamburger patty manufacture are poorly controlled with respect to minimizing the contamination of product with E. coli and, presumably, other faecal bacteria. The methods used in this study offer a means of objectively identifying such hygienically inadequate processes. (C) 1996 Academic Press Limited