Reproducibility of Microbiological Counts on Frozen Cod: A Collaborative Study

Abstract

ABSTRACT Reproducibility of aerobic plate count (APC), coliform and fecal coliform counts in frozen cod was examined by 14 laboratories (3 government, 4 university and 7 industry) using primarily FDA recommended procedures. In order to assure homogeneity of samples, the fish was comminuted and thoroughly mixed. The 35°C and the “room temperature” (26°C) APC on Sample A (a good quality product) were 1.3 × 10 5 /g and 2.6 × 10 5 /g, respectively, and for Sample B (a questionable quality product) the counts were 2.9 × 10 5 /g and 6.4 × 10 5 /g, respectively. Standard deviations among counts (log 10 values) were 0.26 and 0.32 for the 35°C counts and 0.24 and 0.40 for the 26°C counts of Sample A and B, respectively. The std. dev. were reduced to 0.19, 0.22, 0.17 and 0.25, respectively, when counts that did not show significant overlap (based on Dur can's New Multiple Range Test) were excluded from the calculations. Coliform counts on Sample A (not inoculated) ranged from <30 to 4.6 × 10 5 /100g. Coliform and fecal coliform counts for Sample B (inoculated at estimated level of 1 × 10 4 /100g) were 5.8 × 10 3 /100g and 5.5 × 10 3 /100g, with a std. dev. of 0.43 and (.37, respectively. The std. dev. for the coliform counts was reduced to 0.30 when counts from one laboratory were excluded, based on Duncan's New Multiple Range Test.