From 153 samples of fresh ground beef, aerobic plate counts (APC) and counts of viable gram-negative bacteria (GNB) were determined along with ng of endotoxin/g by use of the Limulus amoebocyte lysate (LAL) test. When the 153 meats were grouped by one-half log cycle ranges beginning with samples that contained < log 4.00 GNB/g, they were placed into eight groups. Mean ng endotoxin for these groups increased from 14 for the five samples with GNB < log 4.00 to 52,340 for the 16 samples with log GNB from 7.00 to 7.49. The APC to GNB ratios for the 13 meats with log GNB between 4.00 – 4.49 was 21, but decreased steadily to 2.2 for the 16 with log GNB of 7.00 – 7.49/g, reflecting the progressive decrease in the proportions of gram-positive bacteria and yeasts as total viable counts increased. The mean number of viable GNB/ng of total endotoxin for 148 of these samples was 1900. To calculate the number of GNB in a sample, the latter value was multiplied by the LAL-determined endotoxin in the sample. When GNB/g was multiplied by the respective APC to GNB ratio values, the approximate APC/g was derived. Although the variance and confidence limits of some of the APC and GNB ratio values were wide, use of LAL to estimate APC in 1 h was found to be workable.

This content is only available as a PDF.