Coliphages recovered from 38 samples of ten different food products were characterized with regard to temperature of infectivity. High temperature (HT) phages were capable of reproduction at or above 30°C, mid temperature (MT) phages over a range of 20 to 42°C and low temperature (LT) phages at or below 20°C. The percentage of HT coliphages isolated with Escherichia coli C-3000 host were consistently higher than corresponding percentages with E. coli C host. Coliphages recovered from all products were primarily HT or MT coliphages.

The distribution of coliphages in various foods and the relationship between the incidences of coliphages and bacterial indicators were investigated. A total of 120 food samples comprising twelve products and including fresh meats, shellfish, vegetables and processed meats, were analyzed for indigenous coliphages using Escherichia coli hosts C, C-3000 and B. Bacterial analyses included enumeration of E. coli , fecal coliforms and coliforms, as well as aerobic plate counts and Salmonella analyses. Coliphages were detected (≥10 PFU/100 g) in 56% of samples and eleven of twelve products. Coliphages, E. coli , fecal coliforms and coliforms were recovered at a level of at least 30 organisms per 100 g in 43, 43, 68 and 81% of samples, with overall mean recoveries of 13, 19, 93 and 4300 organisms/100 g, respectively. Highest and lowest recoveries of coliphages and E. coli were from fresh meats and vacuum-packaged processed meats, respectively. Significant nonparametric correlations between coliphages, E. coli , fecal coliforms and coliforms were found among all food samples.

The relationship between microbial ATP measurements and aerobic plate counts (APC's at 35, 20 and 7°C) was investigated for 75 ground beef samples. Samples (n = 27) were obtained from several local retail markets in one experiment, and ground beef samples obtained from a single, processing facility were sampled throughout 15 d of storage at 1°C for a total of 48 samples in another experiment. Bioluminescent assay time for a given sample was less than 1 h. The correlation coefficient (r) between log 10 microbial ATP and log 10 APC (20°C) per g was 0.86 and 0.99 for retail and single source samples, respectively. Differences between actual log 10 APC (20°C)/g and corresponding values predicted by linear regression equations were ≤log 10 0.5 for 25 of 27 retail samples and 48 of 48 single source samples. Variation was noted in values of ATP per bacterial cell and relative bioluminescent quenching (ATP per relative light unit, RLU) for most retail samples and for single source samples having low APC (20°C) levels (≤log 10 7.0).

Coliphages were recovered from 18 of 18 fresh chicken and pork sausage samples as well as from 2 of 6 processed delicatessen meat samples employing a rapid technique using EC medium as both an eluent and as a modified phage assay plating medium. Coliphage recoveries ranged from approximately 3.3 to 4.4 log 10 , 1.2 to 3.5 log 10 and zero to 2.7 log 10 plaque forming units per 100 g in fresh chicken, fresh pork sausage and roast turkey breast, respectively. High coliphage levels generally reflected high fecal coliform counts, particularly for fresh meat samples. These data indicate that coliphages can be readily enumerated in foods within 16 h.

The effect of subjective condition of beef quarters, as determined by United States Department of Agriculture (USDA) personnel, on the microbial and sensory quality of vacuum-packaged clods and ground beef during refrigerated or frozen storage was investigated. In addition, the effect of reconditioning or trimming beef quarters considered to be in off-condition before fabrication into clods or ground beef patties was studied. The microbial quality of beef quarters was directly related to the subjective condition classification in that aerobic plate counts (APC's) of beef quarters and derived products were greater with increasing off-condition on the basis of condition “1” being “excellent” and condition “5” being “unfit for consumption.” No significant differences (P<0.05) were observed in APC's between clods from quarters in various conditions initially or following 14 or 28 d of vacuum packaged storage at 1 to 2°C. Patties derived from quarters in condition “1” had significantly lower (P<0.05) APC's than those of patties from quarters in condition “4” or “5”. The microbial quality of ground beef patties prepared from quarters in conditions “3” and “4” was not significantly affected by trimming of the quarters before fabrication. Few consistent differences in the sensory quality of ground beef patties were noted as a result of condition classification or trimming of quarters from which they were fabricated. These data indicate that reconditioning or trimming of “off-condition” beef quarters cannot be relied upon to improve the microbial quality of derived products such as ground beef.

Counts and taxonomic distribution of typical colonies on violet red bile agar (VRB) and VRB with 1% glucose (VRBG) incubated at 45, 35, 20, 7 and 1 °C from 23 retail food samples were compared. Aerobic plate counts were also obtained at each incubation temperature. Samples included fresh meats, processed meats, frozen processed products and fresh vegetables. Overall mean VRBG counts were slightly higher than VRB counts at each incubation temperature although there was some variation according to sample type. No significant differences (p>0.05) between the two media were noted overall or for any food type. Highest counts for VRB and VRBG generally occurred at 20°C incubation followed by 35, 7, 45 and 1 °C; counts at 20 and 35°C were not significantly different (p>0.05). The taxonomic distribution of typical colonies varied according to incubation temperature and sample type but there were few differences between VRB and VRBG for a given sample and incubation temperature. At 45°C, Escherichia coli was the most frequently recovered organism from both media overall but Klebsiella pneumoniae , Erwinia herbicola or Enterobacter cloacae predominated in many samples. Serratia marcescens and Erwinia herbicola comprised the majority of isolates from both media at 35, 20, 7 and 1 °C with taxonomic diversity being greatest at 35 and 20°C.

The microflora associated with fresh and temperature abused beef livers, kidneys, hearts, tongues and pork livers was identified. Variety meats were obtained from a packing plant and allocated to three packaging treatments, i.e., vacuum packaging, polyvinyl chloride film wrapping and no wrapping (unwrapped). Isolates were characterized from fresh variety meats following frozen storage for three weeks at −29 ± 2 C; and following simulated temperature abuse. Classification of 1555 isolates obtained from aerobic plate counts at 35, 20 and 7 C is provided. Fresh variety meats were found to be contaminated with a variety of bacteria commonly associated with fresh red meats immediately post-mortem, with Micrococcus sp. being the most frequently isolated gram-positive bacterium and Escherichia coli the predominating gram-negative isolate. Frozen variety meats before simulated temperature abuse reflected a higher proportion of gram-positive organisms and fewer Enterobacteriaceae than fresh variety meats. Abused variety meats yielded predominately Pseudomonas strains, except where vacuum packaging was used, in which case isolates were predominately Lactobacillus sp. and Micrococcus .

The development of microflora on hot-boned and conventionally processed beef plate cuts was investigated from time of slaughter and/or fabrication throughout vacuum-packaged storage for 6 weeks at 0–1 C. Cuts from each processing treatment were analyzed immediately post-mortem and after 0.5, 1, 1.5, 2, 4, 7, 14, 21, 28 and 42 days of storage. Fabrication, packaging and chilling of beef plates were carefully controlled to minimize differences in chilling rates and contamination of hot and conventionally processed cuts. Microbial analyses included enumeration of mesophilic, psychrotrophic and total Enterobacteriaceae populations as well as taxonomic characterization of corresponding microbial isolates. Microbial counts of hot-boned cuts were generally higher than corresponding counts of conventionally processed cuts with significant differences (p<0.05) detected between mesophilic and psychrotrophic counts at most storage intervals between 14 and 42 days. Earlier predominance of organisms such as Lactobacillus spp. and Brochothrix thermosphacta on hot-boned vs. conventionally processed cuts was indicated by taxonomic determinations. Psychrotrophic Enterobacteriaceae , including Hafnia alvei and Yersinia enterocolitica -like organisms, were recovered in high numbers from a few samples after 28 and 42 days of storage regardless of processing technique. Differences in the development of microbial flora on hot and conventionally processed beef cuts could not be explained on the basis of differences in initial chill rates between treatments.

Effects of various handling, packaging, temperature-abuse and storage conditions were determined on the microbiological characteristics of beef livers and tongues. These organs were evaluated: ( a ) initially following slaughter, ( b ) immediately following the frozen storage period of 2–4 weeks at −29°C and ( c ) following a simulated shipping-temperature abuse of 24 h at 22–28°C followed by 13 days of storage at −1 ± 0.5°C. Initial counts (log/cm 2 ) of coliforms, coagulase-positive Staphylococcus aureus and Clostridium perfringens ranged from 0.19–1.37. Generally, neither freezing nor temperature-abuse had a significant effect on these microorganisms. Vacuum-packaged beef tongues and livers, generally, had lower bacterial counts than did either naked or polyvinyl chloride film-wrapped products. Generally, it was observed that abusive storage temperatures, in conjunction with the naked and film-wrapped packaging systems, appear to present potential microbial spoilage problems when compared with vacuum packaging.

Various fermentation schedules and finishing temperatures used in the summer sausage industry were investigated for their ability to eliminate Salmonella newport or Salmonella typhimurium inoculated into raw ingredients during the formulation phase of production. Summer sausage was produced with and without the aid of a starter culture. Use of the starter culture, a Lactobacillus plantarum strain, provided for a more rapid fermentation and subsequent decrease in pH. Elimination of Salmonella was dependent upon the initial contamination level, serotype involved, rate of fermentation and processing temperature.

The two tail muscles, ilio caudalis and ischio caudalis, along with hide surface areas from four alligators, Alligator mississippiensis , were evaluated for microbial numbers and types. Microbial analyses of hide surfaces yielded a mean aerobic plate count (APC, 35 C) of approximately 4.45 logs/cm 2 at 35 C, and total coliform counts were low. Salmonella was not recovered from hide or tissue samples. APCs of fresh meat samples were low (2.88 – 3.02 logs/g) regardless of muscle type, with no recovery of fecal coliforms or Escherichia coli . Microbial development in meat stored at 1.7 C for up to 15 days was comparable to that reported for fresh beef. Typical microorganisms recovered included Corynebacterium , Staphylococcus , Micrococcus , Flavobacterium , Pseudomonas , Acinetobacter , Arthrobacter and yeasts.

Approximately 100 cells of Yersinia enterocolitica were inoculated onto surfaces of beef hearts and livers among 1.6 – 6.6 × 10 4 indigenous bacteria. Two strains known to be sensitive to selenite were tested for recovery in a selenite broth: T AMU 54, which was isolated in an enteritis outbreak, was recovered in 16 of 20 trials, while a clinical strain, IP 107, was recovered in 9 of 20 trials. Variables tested in the enrichment procedure included selenite levels, beef extract levels and incubation periods. Selective agents encouraging recovery of the organism included both MacConkey agar with Tween 80 and refrigerated Bismuth Sulfite agar plates, with and without a potassium hydroxide pretreatment of the enrichment broth before streaking onto selective agars. Use of refrigerated Bismuth Sulfite agar and a 3-day incubation period of the enrichment broth provided the most consistent recovery of Y. enterocolitica .

The surface plate method, using freshly pre-poured agar plates and/or stored pre-poured plates and the pour plate method were compared for enumeration of microorganisms in fresh bologna, fresh ground beef, frozen turkey pot pie and bacterial suspensions of Pseudomonas fluorescens and Streptococcus faecalis . Stored pre-poured Plate Count Agar (PCA) plates were packaged in plastic bags and held at 5 C for up to 6 weeks before use. Aerobic plate counts were derived from plates incubated at 35 C for 48 h, 20 C for 5 days and 7 C for 10 days. Differences in counts between methods for a given sample, incubation and pre-poured plate storage period were less than 0.5 log cycle in 97% of the comparisons. Regression and correlation coefficients between methods were highly significant; correlation coefficients varied from 0.987 to 0.999, and regression coefficients from 0.977 to 1.068 between any pair of methods. Storage of pre-poured plates for up to 6 weeks appeared to have no significant effect on recovery of microorganisms, using the surface plate technique.

Roast beef, turkey breast, pastrami, corned beef and pickle and pimento loaf were obtained from delicatessens in eight Gainesville (FL) supermarkets. A total of 80 samples were analyzed for total aerobic counts (APC at 35 and 20 C), yeasts and molds, fecal streptococci, Staphylococcus aureus , Clostridium peifringens , total coliforms, fecal coliforms, Escherichia coli and Salmonella . No Salmonella was recovered from any samples and only one sample contained C. peifringens . S. aureus was recovered from 12.5% of the samples at levels less than 1 log/g and E. coli was found in 11.3% of the samples. All samples contained fecal streptococci and 73.8% contained coliforms. APCs (35 C) were highest in corned beef and pastrami samples with mean counts of 6.75 and 6.91 logs/g, respectively; roast beef samples had lowest APCs with a mean count of 4.84 logs/g. High APCs and coliform counts in many samples indicate a need for improved sanitation procedures at the processor and retail levels.

Forty samples of frozen imported lean beef pieces from six countries were obtained from two centralized meat processing operations. The samples were analyzed for total aerobic counts (35, 20, 7 C), yeasts and molds, fecal streptococci, Staphylococcus aureus , Clostridium perfringens , total coliforms, fecal coliforms, Escherichia coli , and Salmonella . Characterization of the microbial flora from 20 of the samples was also done. Microbial counts were consistently low in all analyses; no Clostridium perfringens or Salmonella was recovered from any samples. The microbial flora was predominantly Pseudomonas , Flavobacterium , Moraxella , Acinetobacter , Corynebacterium , Micrococcus , Staphylococcus and Lactobacillus ; the remaining isolates included Alcaligenes , Erwinia , Citrobacter , Klebsiella , Streptococcus , Bacillus and Arthrobacter .

The top surfaces of 20 canned drinks obtained from vending machines were analyzed for their microbial flora. Members of the genus Flavobacterium predominated and were found on 85% of the cans; the remaining isolates included Pseudomonas , Bacillus , Staphylococcus , Corynebacterium , Erwinia , Klebsiella , Micrococcus , Arthrobacter , Lactobacillus , Nocardia , yeasts, and molds. Pseudomonus aeruginosa , Klebsiella pneumoniae , and Staphylococcus aureus were among these isolates. Total bacterial counts ranged from < 10 organisms per in 2 . to 3.2 × 10 4 organisms per in 2 . on can-top surface, with a mean count of 2.2 × 10 3 organisms per in 2 . Highest counts often occurred on samples with flora dominated by genera in the family Enterobacteriaceae .

Twenty samples of cold ready-to-eat cereal were analyzed for total organisms using the pour plate and Most Probable Number technique (MPN). Nine samples enhibited growth in MPN tubes but not on pour plates. Of the 28 isolates obtained from the MPN tubes, 16 were members of the genus Bacillus and the remainder were predominately other gram-positive organisms. The data support the working hypothesis that aside from the mathematical bias associated with the MPN procedure, microbial recovery is more favorable in a liquid environment than in an agar medium.

Shrimp dipped in sodium bisulfite solutions of 1.25 and 5.0% were analyzed for residual bisulfite during iced storage for up to 15 days. Most of the bisulfite was located on the shell and was readily leached away by the melting ice. A further reduction in bisulfite occurred when shrimp were rinsed in either water or hypochlorite solutions after storage, with the greatest reductions occurring during the hypochlorite rinses.

An evaluation of the microbiological quality of retail ground beef prepared in a centralized operation (which has nearly statewide distribution) and in four local stores was undertaken. Forty retail samples were analyzed for total aerobes (22 C and 35 C), yeasts and molds, Clostridium perfringens , Staphylococcus aureus , total coliforms, Escherichia coli , enterococci, Salmonella , and fat content. Products prepared in the centralized operation exhibited trends toward better microbiological quality than that of traditionally prepared products. Salmonella screening of the samples resulted in the identification of S. infantis in one traditionally prepared sample.

A comparison was made between Butterfield's buffered phosphate solution. 0.1% peptone in distilled water. 0.5% peptone in distilled water, 0.85% NaCl in distilled water, and pure distilled water as diluents for total bacterial counts of 28 different food samples. Plates were incubated at 35 and 22 C for 2 and 5 days, respectively. Butterfield's diluent afforded the highest overall mean count for most samples regardless of the incubation temperature. Highest counts occurred with all diluents at 22 C. Although Butterfield's diluent. 0.1% peptone, and 0.5% peptone appeared to yield higher counts for most samples than did distilled water or 0.85% NaCl, statistically the mean counts obtained for the live diluents used were not significantly different from each other. It appears that no single diluent is suited for every possible sample type.