A total of 150 honey, 43 syrup and 40 dry cereal samples were analyzed for Clostridium botulinum spores, each in triplicate quantities of 25 g. The foods were sampled randomly, except for two lots of honey which were potentially associated with illness. Botulinal spores were detected in a sample of honey associated with infant botulism and in a single sample of rice cereal.

Shelf-stable canned cured meats (SSCCM) are preserved by thermodestruction of the vegetative microflora, partial destruction of the microbial spores, and subsequent inhibition of the surviving spores. Inhibition depends primarily on the concentration of salt, nitrite input and the severity of the thermoprocess. On the basis of published experimental work, the botulism risk of SSCCM may be estimated, either as the equivalent of the decimal heat destruction of spores, taking into account the combined destruction and inhibition in SSCCM (log 1/P), or as the decimal number of cans needed for a single can to become toxic, i.e., log 1/(P × i), where P is the probability of individual spores developing and producing toxin in the processed can, and i is the incidence of Clostridium botulinum spores in the raw product. The experimental data demonstrate that the 12 D concept (or its equivalent) is not applicable to SSCCM and cast doubt on the adequacy of the F 0 value as a measure of the effective heat treatment for this type of product. Research data on the safety of SSCCM are limited and do not allow us at present to propose minimal requirements of the essential safety factors. Instead, our risk assessment relies heavily on commercial practice and experience. In analogy to log 1/(P × i), the safety of commercial cans is expressed as the decimal number of cans produced per can causing severe illness, arbitrarily designated SU (safety units). Estimated SU values for defined commercial luncheon meats, canned cured ham and sausages, range from > 7 to > 9. Provided that the microbial spore levels of both the raw meat and non-meat ingredients of SSCCM are rigidly controlled, the SU estimates allow us to specify minimum safety requirements for the major groups of shelf-stable canned cured meats.

Vacuum-packaged beef, inoculated with Clostridium botulinum spores of types A and B and incubated at 25°C, was monitored for the production of toxin and for changes in appearance and odor. Toxin was detected first after 6 d of incubation and was always accompanied by significant organoleptic changes when compared to meat stored at 4°C. Uninoculated meat samples stored at 25°C remained non-toxic, whereas their sensory scores, in particular with respect to appearance, were similar to those of the inoculated samples.

The salt crystal method of Northolt and Heuvelman (J. Food Prot. 45:537–540, 1982) for testing the water activity of foods was modified to facilitate the recognition of crystal liquefaction. The proposed device is assembled from basic laboratory ware, i.e., an Erlenmeyer flask, a rubber stopper and a test tube.

Samples of 75 g of commercial liver sausage were cultured, with and without prior heating, for the presence of viable Clostridium botulinum . Three of 276 heated cultures and 2 of 276 unheated cultures produced botulinal toxin, all of type A. The most probable number of botulinal spores was estimated at 0.15/kg. The estimate for “total” C. botulinum , based on 5 toxic cultures in 276 heated and unheated pairs, was 0.24/kg. The 99% confidence limits were 0.02 to 0.53 per kg and 0.05 to 0.69 per kg, respectively .

Liver sausage was formulated with different brine and nitrite concentrations, challenged with a mixture of spores of five strains each of Clostridium botulinum , types A and B at 10-fold increasing concentrations, temperature-abused at 27°C, and assayed for botulinal toxin after various periods. From the number of toxic sausages and initial spore concentrations, the probability (P) of a single spore to give rise to toxin within a given period of abuse was estimated. At moderate brine concentrations (3.8–4.2% salt), 50 or 100 ppm of nitrite had little or no effect on toxigenesis; the estimated P values for one week at 27°C were from 10 −2 to 10 −4 with 0 and 50 ppm of nitrite, and from 10 −2 to 10 −5 with 100 ppm. With 150 ppm, however, P was consistently <2 × 10 −6 At a higher brine concentration, an appreciable delay in toxigenesis was also obtained with 50 or 100 ppm of nitrite. Randomly typed extracts from 44 toxic sausages all contained C. botulinum type A toxin only. At toxin levels ≥100 mouse MLD/g the sausages had a putrid odor, but sausages with less toxin often appeared organoleptically acceptable. Storage of sausages at 8°C for 6 weeks before incubation at 27°C resulted in nearly complete disappearance of detectable nitrite, but did not diminish the inhibitory effect of the initial nitrite. Increased processing temperature and/or prolonged time of processing reduced the inhibitory effect of nitrite.

Of 416 75-g samples of commercial bacon cultured with or without prior heating. only one gave rise to formation of botulinal toxin. The most probable number of Clostridium botulinum was estimated at 0.064 per kg, with a 99% confidence limit of 0.0004 to 0.478 per kg.

Reports of outbreaks of food- and waterborne disease in Canada and the United States were searched for data on the number of cases of illness initially reported to health and food protection agencies. These were compared to the number of cases identified by thorough epidemiological investigations or to the number estimated. For a total of 51 outbreaks of bacterial, viral, and parasitic disease, the median ratio of the estimated cases to initially reported cases was 25:1. Based on this ratio and the systems of data transmission from the first level of reporting to national reporting agencies, the estimates of annual food and waterborne disease cases for 1974 to 1975 were from 150,000 to 300,000 in Canada and 1,400,000 to 3,400,000 in the United States. For a total of 26 outbreaks of salmonellosis, the median ratio of estimated cases to initial buman isolations of Salmonella was 29.5. Based on this ratio and reported isolations of Salmonella from 1969 to 1978, the estimates of all cases of human salmonellosis were 150,000 in Canada and 740,000 in the United States, annually.

Bottled lumpfish caviar was prepared with different salt (NaCl) concentrations and pH, and injected with spores of Clostridium botulinum . Under abusive storage conditions (30 C), outgrowth and toxigenesis occurred at combinations of ≤ 3.95% salt in the water phase and pH ≥ 5.2, and of ≤ 4.67% salt and pH ≥ 5.6. No toxin was formed at salt concentrations of ≥ 5.56% or at pH ≤ 5.0. A survey of commercial caviar products showed that most of these had salt-pH combinations which would effectively inhibit C. botulinum at abusive temperatures during storage.

Clostridium perfringens type A food poisoning is caused by ingestion of food contaminated with large numbers of C. perfringens cells. The cells multiply and sporulate in the intestine and produce an enterotoxin in situ . The enterotoxin is released upon cell lysis and causes increased capillary permeability, vasodilation, and excess fluid movement into the intestinal lumen resulting in diarrhea. Purified C. perfringens enterotoxin has a UV absorption spectrum characteristic for proteins, and is essentially free of nucleic acids, lipids, and reducing sugars. Its apparent molecular weight is 36,000 ± 4,000, its Stokes radius 2.6 nm, and its isoelectric point pH 4.3. Its specific toxicity is about 2,000 mouse MLD/mg N.