Reduction of Clostridium sporogenes spore outgrowth in natural sausage casings using nisin.

Preservation of natural sausage casings using dry salt or saturated brine is regarded as sufficient to inactivate vegetative pathogenic non-spore-forming bacteria present on the casings. Although the outgrowth of bacterial spores is prevented by salt or saturated brine preservation, these spores will remain present and develop into vegetative cells when conditions are more favourable. To prevent subsequent outgrowth additional preservation measures should be implemented. In the experiments described the use of nisin was evaluated to reduce outgrowth of spores in desalinated casings. The bacteriocin nisin was chosen because of its known efficacy against spore-forming bacteria and their spores in various foodstuffs. Clostridium spore suspensions (Clostridium sporogenes, ATCC 3584) were used in two concentrations to inoculate three nisin concentrations (10, 50, 100 µg/mL) in water containing gamma-irradiated casings. Additionally, the binding of nisin to casings, using (14)C-labeled nisin Z and subsequent availability of nisin were evaluated. Results demonstrate that nisin is partly reversibly bound to casings and can reduce the outgrowth of Clostridium spores in the model used by approximately 1 log(10) (90%). However, the biological relevance of these results needs to be determined further by conducting industrial trials before any recommendation can be made on the practical implementation of nisin in the preservation of natural sausage casings.

Evaluation of antioxidative properties of Geranium macrorrhizum and Potentilla fruticosa extracts in Dutch style fermented sausages.

Antioxidative properties of Geranium macrorrhizum, Potentilla fruticosa and Rosmarinus officinalis (as a reference) extracts were evaluated in Dutch style fermented sausages. Extracts were incorporated into sausages during preparation. The sausages were subsequently fermented, tested and compared to a standard spices mix, traditionally used for the production of such sausages. Formation of the primary oxidation products - peroxides, and secondary - TBARS and hexanal was monitored. The polar extracts from Potentilla showed some antioxidant activity, especially in combination with ascorbate, however the activity was low compared to the standard spices mix. Polar extracts from Geranium showed only negligible antioxidant activity.

Effect of trisodium phosphate on slip and textural properties of hog and sheep natural sausage casings.

The defective gliding of certain natural casings during the stuffing of sausages is an important problem in the meat processing industry. The gliding behaviour of (defective) hog and sheep casings was assessed with a newly developed instrument, and by a technologist during the stuffing of sausages. Casings were treated with 0.01 M trisodium phosphate; control casings were untreated. Cooked and smoked sausages were made in hog casings treated or untreated with phosphate and subjected to compression tests. In all cases the treatment with phosphate clearly facilitated the gliding of the casings over the test pipes, as compared to the control casings. The instrument to measure the casing gliding properties did not provide reliable information about the actual stuffing of sausages. The phosphate-treated casings had a lower shear force than the control casings after being used as skins for cooked and smoked sausages. If confirmed, the finding that mild phosphate treatment can diminish the force required to shear a casing will be of interest to the sausage industry because the toughness of certain hog casings is considered a problem.

Variations in product-related parameters during the standardised manufacture of a semi-dry fermented sausage.

In the industry there is a tendency to shorten production processes of semi-dry fermented sausages (SFDS). Emerging problems then are generally related to product flavour and texture. Five batches of the same product were produced under pilot-plant conditions, following strictly standardised procedures. At pre-determined time intervals in the early ripening stage weight loss, pH and a(w) values were recorded and Uniaxial compression tests were performed on cylindrical axial product samples. Uniaxial compression was introduced as a potential new methodology for determining the consistency of fermented sausages. The compression data were used to calculate stress and strain at fracture and Young's modulus. Despite careful standardisation of both the product formulation and the manufacturing process, the various parameters showed considerable variation. As anticipated, there was significant product weight loss, and decrease in pH and a(w) during the manufacturing process. Analyses of variance showed that neither stress, strain nor Young's modulus changed significantly over the period from 2-9 days. However, the averages of the Young's modulus tended to increase with time. These results suggest that the Young's modulus may become an important production parameter, in addition to the already applied product pH, a(w) and weight loss.

Growth inhibition of heat-injured Enterococcus faecium by oligophosphates in a cured meat model.

Cells of two heat-resistant strains of Enterococcus faecium were heated and incubated in meat suspensions containing curing ingredients. The concentrations of the curing ingredients were those frequently used for pasteurized ham-type products, except that the concentrations of the oligophosphates (triphosphate and diphosphate) varied. Heating tests at 69 degrees C were performed with inoculated meat suspensions in heat-sealed plastic pouches. Numbers of bacteria were counted immediately after heating and in parallel series of heated pouches incubated at 37 degrees C. Plating was performed in Tryptone Dextrose Yeast Meat Peptonised Milk Agar (TDYMP); in TDYMP Agar to which the curing ingredients were added; and in TDYMP Agar to which the curing ingredients except oligophosphates were added. The inclusion of oligophosphates in the heating medium increased the heat-injury sustained by the E. faecium cells, and in combination with rather severe heat treatment even completely blocked the growth of surviving organisms in the meat suspension incubated at 37 degrees C. The presence of oligophosphates in the culture medium TDYMP Agar severely reduced the counts of freshly heated cells; however, this effect disappeared after repair and growth of the surviving organisms in the meat suspension.

The potential of vancomycin-resistant enterococci to persist in fermented and pasteurised meat products.

Experiments with 148 isolates of vancomycin-resistant enterococci (VRE) were performed to assess their potential to persist and grow in fermented sausages and pasteurised meat products. All strains were meat isolates and Van-type A, except a single VanC1 strain. In total, 143 strains of Enterococcus faecium were involved. Eight selected strains were examined for their potential to grow at high salt and nitrite levels and at reduced pH. The same isolates were used in experiments with fermented sausages. All available strains were subjected to heating tests in meat suspensions with added curing ingredients. All but one of the eight tested isolates grew at pH 4.0 in tryptone soya broth (TSB). With the combination of 8% w/w NaCl, 400 ppm NaNO2 and 0.5% w/w glucose in the meat suspension, all isolates grew at 37 degrees C, whereas none grew at 7 degrees C even after 56 days. With the addition of 10% w/w NaCl, 200 ppm NaNO2 and 0.5% w/w glucose, still one E. faecium isolate grew at 37 degrees C, although very slowly. Overall, the strains tolerated high salt and nitrite concentrations and reduced pH very well, even beyond levels applied in the regular production of fermented and/or pasteurised meat products. The tested strains could be isolated after the fermentation and further ripening of "boerenmetworst" and "snijworst". Overall, their colony counts decreased on average about 1 log-unit over a period of 60 days after batter manufacture. All 148 isolates demonstrated a relatively weak thermal resistance compared to results for selected vancomycin-sensitive enterococci strains reported in the literature and to results collected under identical experimental conditions in this laboratory. None of the strains (log inoculation level about 5-6 ml(-1) for each isolate) could be cultured after heating at 70 degrees C for 10 min.

Nutritional and sensory evaluation of tempe-fortified maize-based weaning foods.

Maize-based weaning foods prepared from whole maize meal or dehydrated fermented maize (ogi) flour fortified with soybean, cowpea or groundbean tempe, with or without the addition of melon seed flour were fed to 40 albino weanling rats. The nutrient quality of the weaning foods was monitored by the measurement of protein efficiency ratio (PER), net protein ratio (NPR), true digestibility (TD), biological value (BV) and net protein utilization (NPU). Amino acid composition and amino acid scores were determined. The weaning foods were also evaluated for sensory properties and acceptability by nursing mothers and their children. Protein efficiency ratio, net protein retention and net protein utilization of the fortified weaning foods were not significantly different (P > 0.05) when compared to casein diet (control). True digestibility of tempe-fortified maize-based weaning foods averaged 85.4 +/- 3.4% compared to 94.4%, for casein diet. Results from sensory analysis showed that relative to tempe-fortified maize-based weaning foods, most mothers preferred Cerelac, a commercial weaning food, and its porridge because of its vanillic flavour and sweet taste. Despite the low scores recorded for aroma and taste of tempe-fortified maize-based weaning foods, most mothers and their children found porridges prepared from these foods acceptable, with a preference for ogi fortified with cowpea tempe. Tempe-fortified maize-based weaning foods had high protein quality and can support the growth of infants in developing countries especially during the critical weaning period (6-12 months). They were comparable to Cerelac, a commercial baby food in colour and texture.

Effects of dietary vitamin E supplementation and packaging on the colour stability of sliced pasteurized beef ham.

The effect of supplementation of vitamin E (2025 IU animal(-1) day(-1)) in the diet of beef bulls on the colour stability of pasteurized beef ham was studied. Control and enriched diets were provided for the last 136 days before slaughter. Pasteurized hams were manufactured from Mm. semitendinosus from eight animals per dietary group. Half of the samples of sliced ham from control (CON) and supplemented (SUP) bulls were packaged under vacuum (VAC) and half in low-oxygen modified atmosphere packs (FOG, gas mixture: CO(2)/N(2)=50/50). The packages were kept under constant illumination for 28 days at 8°C. During storage, the number of colony-forming units (CFU) reached a maximum of 5x10(7) g(-1). The microflora was dominated by lactic acid bacteria. The supplementation with vitamin E showed no effect on microbial growth. Lipid oxidation was stable during storage. A significant difference between both dietary groups was detected for the decrease in the redness values during storage. Redness values of CON vacuum-packaged samples decreased (P < 0.01) with time, whereas those for the SUP products only tended to decrease. The redness values of FOG-packed ham were higher than those of VAC-packed ham at the end of the display period, irrespective of the dietary group. Overall, colour appeared to be more stable in the FOG-packed products than in the VAC-packed products. It can be concluded that dietary supplementation of bulls with vitamin E appears to offer only a minor improvement in colour stability over current feeding regimens when the Mm. semitendinosus are used to make cured, pasteurized ham-type products.

Effect of dietary vitamin E supplementation, fat level and packaging on colour stability and lipid oxidation in minced beef.

The effect of addition of vitamin E (2025 IU animal(-1) day(-1)) to the diet of beef bulls on the colour stability and lipid oxidation of minced beef was studied. Control and enriched diets were provided for the last 136 days before slaughter. Batches of freshly minced meat were prepared containing approximately 1.3 and 22.2 wt% fat, respectively. Half of the samples of minced meat from control (CON) and supplemented (SUP) beef were packaged on trays with oxygen-permeable over wraps and half in modified atmosphere (MA) packs (initial gas mixture: O(2)/CO(2)/N(2)=65/25/10). The minced beef was stored for 10 days at 7°C in an illuminated environment. The SUP meat at both fat levels was consistently more resistant to lipid oxidation than was the CON meat. The additional vitamin E had a greater anti-oxidant effect for the lean meat product. MA packaging in comparison to the oxygen-permeable foil over-wrap did increase lipid oxidation, the effect being most pronounced for the CON meat. A sensory panel considered the colour of the lean SUP meat during display as more attractive than that of lean CON meat, irrespective of packaging. A similar effect was observed occasionally for the relatively fat minced meat. These subjective findings were confirmed by objective assessment of colour. The stability of the colour of the MA packed meat was better than that of the oxygen-permeable foil-wrapped meat. Microbial growth patterns of enriched and control meat were similar. MA packaging retards the multiplication of mesophilic aerobic spoilage micro-organisms and Enterobacteriaceae.

Effect of dietary vitamin E supplementation on beef colour stability.

The effect of dietary vitamin E upon colour, waterholding capacity, bacterial growth and lipid oxidation of beef longissimus thoracis (LT) and psoas major (PM) muscle were examined during aerobic display of fresh muscle and after aging in vacuum for 26 days. Forty crossbred beef bulls received a whole crop corn silage, supplemented with concentrate. Twenty bulls were each supplemented with 2025 mg vit E per day (added to the concentrate) for 136 day prior to slaughter and compared with non-supplemented control animals (n=20). In fresh LT muscle drip loss did not differ between treatment groups, while in PM muscle drip loss was significantly higher for the supplemented group. The treatment did not affect bacterial growth in fresh and aged muscles. Lipid oxidation during 12 day storage of fresh muscle was significantly lower for the supplemented group, as indicated by the lower TBA-values. No effect of the vitamin E treatment was observed on a (∗)-values of both fresh and aged LT muscle during display for 8 and 5 days, respectively. In PM muscle, supplemented beef had lower a (∗)-values in fresh (at day 1) and aged (at days 1 and 2) muscle, due to a lower oxygenation. The reason for this lower oxygenation is unclear. After aging, colour stability was decreased and more variable than in fresh muscle. Similar results were obtained when the difference in reflection values at 630 and 580 nm (R630-580), instead of the a (∗) value, was used as a parameter for colour stability. The absence of an effect of vit. E on the rate of discoloration, might possibly be explained from the observation that α-tocopherol levels in control muscle were relatively high (LT: 2.1 and PM: 3.2 µg/g muscle), compared with data from literature. Analysis of the feed for vit. E suggests that this was due to a relatively high natural vit. E uptake from the feed, which was calculated to be approx. 330 mg vit. E per animal per day for the control group.

Effect of initial mild curing, with additives, of hog and sheep sausage casings on their microbial quality and mechanical properties after storage at difference temperatures.

Sausage containers, derived from animal intestines, are usually preserved by salting and/or drying. Adequately salted final products are microbiologically fully acceptable. However casings, even those packed in dry salt, sometimes deteriorate in quality. Experiments were performed to improve salting procedures by adding food-grade additives to the salt to improve the microbiological and mechanical properties of the casings. Before storage, casings were cured by slush- or dry-salting with or without additives for 3 weeks, and after that the rinsed and re-salted (dry- or slush-salting) casings were stored for 6 months at different temperatures (10, 20, and 40°C). During storage, growth of halophylic bacteria was observed in control casings (salted, no additives) but not in casings cured with citric or lactic acid and their relative sodium salts. The casings cured with citric acid/Na(3)-citrate had good mechanical properties and filling characteristics when assessed after prolonged storage at 10°C.

Effect of the dietary supplementation with Vitamin E on colour stability of packaged, sliced pasteurized ham.

The effect of dietary supplementation with vitamin E (200IU kg(-1) feed) on the colour stability of pasteurized ham was studied. Pigs were fed on control and enriched diets for the last 12 weeks before slaughter. Pasteurized ham was manufactured from the hams from 6 barrows and 6 gilts per dietary group. Half of the samples of sliced ham from control and supplemented pigs were packaged under vacuum (VAC) and half in low-oxygen modified atmosphere packs (FOG, gas mixture: CO(2)/N(2)=60/40). Half the packages were kept under constant illumination and the other half in the dark, both for 22 days at 7°C. The redness component of the VAC-packaged ham prepared from vitamin E-supplemented pigs was slightly more stable than that of comparably packaged ham prepared from control pigs. The opposite was observed for the FOG-packaged products. Overall, colour changes were greater in the ham in FOG-packs than in the ham in VAC-packs. In addition, the colour of the FOG-packaged ham was clearly affected by illumination, whereas the colour of the VAC-packaged ham appeared more stable. It is concluded that dietary supplementation of pigs with vitamin E does not appear to offer significant advantages over currently used feeding regimens with regard to the quality of the ham produced.

Effect of the dietary supplementation with vitamin E on colour stability and lipid oxidation in packaged, minced pork.

The effect of supplementation of vitamin E (200 W kg(-1) feed) in the diet of pigs on colour stability and lipid oxidation in minced pork was studied. Control and enriched diets were provided for the last 12 weeks before slaughter. Half of the samples of minced shoulder meat from control and supplemented pigs were packaged on trays with oxygen-permeable overwraps and half in modified atmosphere packs (initial gas mixture: O(2)/CO(2)/N(2) = 66/ 27/7). Meats were stored for 10 days at 7 °C in an illuminated retail display cabinet. The meat from vitamin E-supplemented pigs was more resistant to lipid oxidation than was the control meat. Gas packaging appeared to increase lipid oxidation in control meat, whereas lipid oxidation was stable in meat from vitamin E-supplemented pigs. Colour stability for gaspacked meat was comparable for both dietary groups. However, oxygen-permeable overwraps had a negative effect on colour stability in vitamin E-enriched meat. The reason for this is not known. The shelf-life of enriched and control meat was similar. Thus supplementation of pig feeds with vitamin E is recommended if an improved stability against lipid oxidation of (minced) pork is required.

Effect of dietary vitamin E supplementation on pork quality.

The purpose of this experiment was to examine the effect of dietary vitamin E supplementation on pork quality, and in particular on colour stability. Crossbred pigs (n = 72) at a mean weight of 44 kg were assigned to one of two treatments. One group received, during a period of 84 days prior to slaughter, a tapioca based diet, which contained 8 mg vitamin E per kg feed. The other group received during this period the same diet, except it was supplemented with 200 mg vitamin E per kg feed. Muscle samples of longissimus thoracis and lumborum (LL) and psoas (PM) were collected at 24 hr post mortem and meat quality was assessed: pH, drip and cooking loss, shear force and intramuscular fat content. Colour stability was evaluated in fresh muscle (LL and PM) and after freezing (LL only) by measuring redness (a(∗)-values) during 6 days of storage at 7 °C. TBA-values and microbiological counts were also determined during storage. Results showed that extra dietary vitamin E had no effect on pig performance (daily gain, feed efficiency, lean meat percentage) nor on meat quality traits. The vitamin E levels were five times higher in the muscles of the treated group than the control group. In comparison with fresh LL muscle, colour stability was lower in PM and after freezing. In both muscles, the vitamin E treatment reduced TBA-values, in particular after frozen storage. No effect was found on microbiological counts. Colour stability was improved in LL after 6 days of storage, but not in PM. The effect in LL is too late to be of practical significance, since pork is usually sold well before that time in The Netherlands. It is suggested that variation in feedstuff composition of the diet may possibly explain part of the variable results reported in literature for the effect of vitamin E supplementation on colour stability of pork.

The effect of dietary vitamin E supplementation on drip loss of bovine Longissimus lumborum, psoas major and Semitendinosus muscles.

The effect of dietary vitamin E supplementation (2150 IU/head/day) on drip loss and related quality traits of bovine M. longissimus lumborum M. psoas major and M. semitendinosus was examined. The effect of vitamin E supplementation on drip loss seemed to depend on muscle studied. Drip loss of longissimus muscles was not significantly influenced, whereas supplemented semitendinosus muscles lost significantly less (p < 0.05) and supplemented psoas major muscles significantly more (p < 0.05) drip than did control counterparts. In both supplemented and control samples, sarcolemma failure occurred. No ultimate pH differences were detected between control and supplemented samples regardless of the muscle considered. In supplemented semitendinosus muscles, the decrease in drip loss was accompanied by an increase of sarcoplasmic protein solubility and sarcomere length. It is discussed that both these variables may be related to the stability of mitochondria and sarcoplasmic reticulum, as affected by dietary supplementation of vitamin E. However, this view deserves further investigation and more evidence is needed to establish the mechanism by which vitamin E influences drip loss.

Mechanical and microstructural characteristics of meat doughs, either heated by a continuous process in a radio-frequency field or conventionally in a waterbath.

Meat doughs, all having the same chemical composition, were pasteurised to a comparable heat intensity (calculated as Cook values: target level of 5 min at 100°C): (i) while flowing through a glass tube (inner diameter 50 mm) mounted in a special radio-frequency (27 MHz) heating section; (ii) after flowing unheated through the glass tube at the same rate and heated in a waterbath; and (iii) after sampling immediately after the pump and heated in a waterbath. The cooked products were sampled in the core and at the rim of the product for rheological (oscillation and uniaxial compression tests at small strain), fracture measurements (uniaxial compression tests at high strain) and microstructural evaluation (light microscopy and video image analysis). Additional core samples were used for a sensory evaluation (triangle tests) of the texture of the differently processed doughs. The fast heating rate (25-30 K/min) at a mass flow of the dough of 100 kg/h (mean velocity 0.014 m/s) during dielectrical pasteurisation affected the mechanical character, the microstructure and the triangle test results of core samples from the sausages, compared to heating in a waterbath. Flow of the unheated dough through the tube of the continuous processing equipment, followed by heating in a waterbath, had little effect on the results of the mechanical tests, the microscopical evaluation and the triangle tests. The radio-frequency heated products had both higher storage and loss moduli (were more firm), fractured at higher stress values and were considered more firm in the sensory evaluation. The microstructure of dielectrically heated versus other samples displayed a more open structure of the protein matrix with larger irregularly shaped fat particles that were surrounded by relatively thin and compact protein bridges. The effects of flow and heating method on the behaviour of rim samples were very similar to their effects on the core of the products. A comparison of the mechanical behaviour of core and rim samples only was significant for radio-frequency heated doughs. The rim samples had lower storage and loss moduli and fractured at lower stress values than the core samples. Micrographs of the dielectrically heated rim versus core samples displayed more orientation of connective tissue particles in the direction of flow and of elongated, larger and irregularly shaped fat particles. Probably, shear at the wall of the tube affected the characteristics of the rim samples. All heated doughs displayed hardly and cooking losses. The radio-frequency heated products always displayed a thin layer of moisture on their surface and occasionally a little fat separation.

Microbiological changes in mawè during natural fermentation.

Lactic acid bacteria increased from 3.2 × 10(6) and 1.6 × 10(7) c.f.u./g (wet wt) to 2 × 10(9) and 1.6 × 10(9) c.f.u./g after 12 to 24 h of fermentation of home-produced mawè (a dough produced from dehulled maize) and commercial mawè, respectively. In commercial mawè, the yeast count increased from 1.3 × 10(5) to 2.5 × 10(7) c.f.u./g after 48 h of fermentation before decreasing, whereas in the home-produced mawè it increased from 2.5 × 10(4) to 3.2 × 10(7) c.f.u./g after 72 h of fermentation; the dominant yeasts were mainly Candida krusei, although C. kefyr, C. glabrata and Saccharomyces cerevisiae were also present. Enterobacteriaceae counts increased slightly during the initial stage ofthe fermentation, but decreased below the detection level after 24 to 48 h. Enterobacter cloacae was mostly found in commercial mawè and Escherichia coli mostly in homeproduced mawè.

Characterization and frequency distribution of species of lactic acid bacteria involved in the processing of mawè, a fermented maize dough from Benin.

Lactic acid bacteria involved in the natural fermentation of both home-produced and commercial mawè were investigated during a 72 h fermentation period. Lactobacillus spp. constitute the majority (94%) of the strains of the lactic acid bacteria isolated, among which 89% represent the Betabacterium group. They include L. fermentum (biotype cellobiosus) (41%), L. fermentum or L. reuteri (19%), L. brevis (26%), L. confusus (less than 2%), L. curvatus (less than 1%) and L. buchneri (less than 1%). Other isolated lactic acid bacteria were L. salivarius, Lactococcus lactis, Pediococcus pentosaceus, Pediococcus acidilactici and Leuconostoc mesenteroides. Several species were detected at the early stage of fermentation, but the final stage was dominated by L. fermentum (biotype cellobiosus) and L. fermentum or L. reuteri totalling 90% of the isolated strains. The trend was the same for both home-produced and commercial mawè. No strains of L. plantarum, generally reported as dominating lactic acid bacteria at the final stage of fermentation of most plant foods, were isolated.

Effect of citric acid and citrate on the shelf stability of canned liver paste processed at reduced F(0) values.

Citric acid and the combination of citric acid and citrate, in a few cases enriched with potassium sorbate, were tested for their stabilizing effect in a canned liver paste processed at reduced F(0) values. pH values of the rest products ranged from 5·24 to 5·76; a(w) values varied from 0·950 to 0·972. F(0) values of 0·05, 0·30 and 0·85 were applied. Before sterilization, the paste doughs were contaminated with mixtures of bacterial spores. Heated packages were challenged by incubation at 30 and 55°C for up to about 2 years. Incubated packages with the standard formulation at an F(0) of 0·05 showed the first blown cans at 30°C after 24 days. Starting from this point, at this F(0) value products with a variable, but increased, bacteriological stability were observed. At an F(0) of 0·30 a perfectly stable product at 30°C was observed after addition of 0·14% w/w citric acid only (product pH 5·69). Moreover, all tested combinations of additives at F(0) values of 0·30 and 0·85 yielded stable products at this incubation temperature. At the higher F(0) value a clear effect with respect to delaying thermophilic spoilage was observed. The effect of the application of potassium sorbate-in combination with both other additives-at F(0) values of 0·30 and 0·85 proved to be negligible.

Effect of citric acid, citrate and slight a(w) decreases on the bacteriological stability of Hague liver sausage.

Combinations of citric acid and citrate, with several carbohydrates and KCl, were tested in a canned Hague liver sausage model for their antimicrobial properties. pH values for the test products ranged from 5·38 to 6·48; a(w) values varied from 0·960 to 0·977. Before pasteurization, the sausage doughs were contaminated with mixtures of bacterial spores. Since the bacteriological stability sometimes appeared to be considerably improved, heated packages were challenged by incubation at 20° and 30°C; some of them for up to 2 years. Incubated packages from the standard formulation spoiled at 30°C within 3 days. Starting from this point, products with a variable, but increased, bacteriological stability were obtained. For the best stabilized batches, first symptoms of spoilage were observed after periods of incubation at 30°C, ranging from 49 to 150 days. Combinations of pH and a(w) values which can be relied upon to yield stable products consistently, cannot be derived from the accumulated data. Citric acid appeared a far more effective stabilizer than its salt. If we regard the undissociated citric acid concentration for these products (a(w) 0·960-0·972) as a significant parameter, values above approximately 13 µmkg(-1), yielded products with remarkable bacteriological stability.