[Bacterial regrowth in drinking water. IV. Bacterial flora in fresh and stagnant water in drinking water purification and in the drinking water distribution system]. / Bakterielle Wieder-Verkeimung im Trinkwasser. IV. Mitteilung: Bakterienflora in Frisch- und Stagnationswasser bei der Trinkwasseraufbereitung und im Trinkwasser-Verteilersystem.

Six dead end water pipes were installed inside a Zurich drinking water plant and five others over a distance of 12 km along the distribution system and the water was left stagnating in there for 2 weeks. A total of 1508 bacteria from fresh and stagnating water were isolated and identified. Of these, 241 bacterial isolates from the distribution system were examined using the nutrient-tolerance test, i.e. testing the ability to grow in tap water and in media with low and very high nutrient content. In the fresh water of the treatment plant specific bacterial populations were obtained, these occurring particularly after the filters. According to different chlorine dosage and chlorine demand, they were finally washed into the distribution system in varying amounts and compositions. It was shown that in the fresh water of the distribution system the genera of Pseudomonas, Azotobacter and Actinobacteria were each present at a level of approximately 30%. After two weeks stagnation non-fluorescing pseudomonads were dominating in the treatment plant as well as in the fresh water of the distribution system. All isolated Actinobacteria and Azotobacter and almost half of the Pseudomonads proved to be oligotrophic oligocarbotolerants or oligocarbophilic organisms in the nutrient-tolerance test. The other half of the Pseudomonads plus the Flexibacter species were mesotrophic oligocarbotolerants, since they could grow in tap water and in culture media with very high nutrient content. Attention is drawn to the unrecognized danger of recontamination of mesotrophic bacteria growing rapidly in stagnating drinking water, which is used as rinsing water for cleaning food processing equipment.

Temperature limits of growth, TNase and enterotoxin production of Staphylococcus aureus strains isolated from foods.

For 77 strains of Staphylococcus aureus freshly isolated from different foods, growth, enterotoxin and TNase production were determined in intervals of 1.5 degrees C +/- 0.5 degrees C by cultivating them in a temperature-gradient incubator between 5 and 50 degrees C for up to 7 days. All the strains were coagulase, DNase and lysostaphin positive but only 58% formed one or two enterotoxins type SEA, SEB or SEE. All strains grew within 7 days in brain heart infusion and had lower and upper temperature limits for growth and TNase production of between 6.5 and 12.5 degrees C, and 39.5 and 48.5 degrees C respectively. The lower and upper temperature limits for production of enterotoxins were between 14 and 38 degrees C, and between 35 and 44 degrees C respectively. Enterotoxin forming isolates either showed narrow (3 to 4 degrees C) or wide (10 to 20 degrees C) ranges of enterotoxin production, irrespective of their temperature range of growth and TNase production. None of the 12 specific physiological attributes used for differentiation could be correlated to toxin type or the temperature requirement of the toxin production. No correlation between the origin and the physiological characters could be detected.

The bacterial flora of non-carbonated, natural mineral water from the springs to reservoir and glass and plastic bottles.

Quantitative and qualitative determinations of the bacterial flora of non-carbonated natural mineral water at the most important steps during bottling at a large water source yielded the following results: (i) Colony counts (on 1:10 diluted plate count agar, incubated at 20 degrees C for 14 days) for water of the five springs and the mixed water were less than 1 to 4 cfu ml-1. The Gram-negative bacterial flora (n = 50 isolates) showed a very different but constant spring specific species distributions with predominance of either eutrophic fluorescent pseudomonads, oligotrophic non-fluorescent pseudomonads or oligotrophic yellow bacteria. (ii) In the reservoir and immediately after bottling the counts were in the range of 10 cfu ml-1. But nearly 30% of the species of the spring water were no longer detectable and there was a significant increase of Gram-positive bacteria. (iii) After 1 week of storage at 20 degrees C colony counts of more than 10(5) cfu ml-1 were found in plastic bottles, but only about 10(4) cfu ml-1 in glass bottles. Besides, a very distinct change of the composition of the microflora occurred. In glass bottles slow-growing oligotrophic non-fluorescent pseudomonads, yellow bacteria and Acinetobacter predominated. In plastic bottles fast-growing eutrophic and mesotrophic fluorescent pseudomonads, Flexibacter and Acinetobacter were dominating. In mineral water, bottled into thoroughly cleaned glass bottles, colony counts of more than 10(5) cfu ml-1 were found within 4 days. In bottles, cleaned mechanically as usual, the increase was significantly slower with a maximum of only 5 x 10(3) cfu ml-1 after 8 days. The results of inoculation experiments in sterile filtered mineral and distilled water led to the suggestion that the difference between the two types of bottles is caused firstly by an inhibition of growth due to residues of cleaning detergents in the glass bottles. Growth promotion by dissolved organic substances in the plastic bottles only played a minor role. After repairing of the pump at a depth of 300 m in a warm mineral water spring, the colony counts at 20, 37 and 42 degrees C on 1:10 diluted and normal plate count agar increased beyond the limits required by the EC directive for mineral water stored a month. Then colony counts decreased slowly and reached the initial level after 1 year, except for the colony counts 1:10 diluted agar at 20 degrees C which stabilized at a relatively high number and a significant alteration of the microflora.(ABSTRACT TRUNCATED AT 400 WORDS)

A simple nutrient-tolerance (NT) test for the characterization of the different types of oligocarbotolerant and oligocarbophile water bacteria from non-carbonated mineral water.

Comparative determination of the specific growth kinetics in mineral water and low and higher concentrated broths at 20 degrees C of 25 selected Gram-negative bacteria isolated from natural non-carbonated mineral water yielded three groups: (1) facultative oligocarbotolerants--with faster growth in normal broth (In g l-1: yeast extract 2.5; casein peptone 5.0; glucose 1.0); (2) obligate oligocarbotolerants--with equal rates of growth in normal and 1:10 diluted broth; and (3) oligocarbophiles--with faster growth in 1:10 diluted broth and in mineral water. In addition, three nutrient types, 'eu-, meso- and oligotrophic' could be distinguished on the basis of full, weak and no growth in brain-heart infusion broth. Further characterization was made between slow and very slow growth types in 1:10 diluted broth. All 25 isolates were psychrotrophic with a minimum growth temperature below 0 degree C. The optimum and maximum temperatures of growth in 1:10 diluted broth, as determined in a temperature gradient incubator were between 20 and 32, and between 29 and 34 degrees C with an average of 26 and 31 degrees C, respectively. Based on these results a very simple nutrient-tolerance test was proposed. After inoculation of the three media, 1:10 diluted broth, normal broth and brain-heart infusion, it is only necessary to check whether or when visible turbidity occurs during 2 weeks incubation at 20 degrees C. This allows additional characterization of bacteria from natural mineral water, which are often difficult to identify, on the basis of growth characteristics in various types of nutrient media.

Disinfection with gaseous formaldehyde. Fifth Part: Influence of albumin, mucin and blood on the bactericidal and sporicidal effectiveness.

The influence of 0.1% peptone, 0.2% albumin, 1% mucin solutions and whole human blood on the inactivation of Staphylococcus aureus ATCC 6538 and spores of Bacillus subtilis var. niger DSM 675 was determined. The bactericidal and sporicidal effectiveness of 0.75, 1.5 and 3.2 mg HCHO l-1 air at temperatures of 35, 40, 45 degrees C and a relative humidity (RH) of 90% decreased in the following order of loading substances: peptone, albumin, mucin and blood. The calculated D-values of the microorganisms suspended and dried in 0.1% petone and in 0.2% albumin after exposure to 3.2 mg HCHO l-1 air at 40 degrees C and a relative humidity of approximately 90% were in both suspensions 1.9 min for S. aureus and 6.1 and 7.3 min respectively for B. subtilis spores. Under the same exposure conditions but with an addition of 1% mucin D-values of 2.5 min and 7.7 min for S. aureus and B. subtilis spores respectively were found. In the presence of blood, D-values of 12.3 and 18.3 min were obtained under the same conditions for S. aureus and B. subtilis spores respectively. Thus the suspension in blood caused a 2-fold increase in D-value compared to the other substances. The nature of the anticoagulant in the whole blood did not cause much difference in the inactivation of B.subtilis spores. Decreasing concentrations of blood caused an increase in sensitivity of B.subtilis spores to gaseous formaldehyde, whereby diluting blood 1:6 reduced the D-value from 23.9 min to only 7.1 min.

Disinfection with gaseous formaldehyde. Third Part: Bactericidal and sporicidal effectiveness of gaseous formaldehyde and level of residues in dependence on concentration, temperature and relative humidity.

The highest rate of inactivation of Staphylococcus aureus (ATCC 6538), Streptococcus faecium (ATCC 6057) and spores of Bacillus subtilis var. niger (DSM 675) was observed at a relative humidity (RH) of 90-95%. At a RH of 60% D 0.75 mg HCHO l-1 air at 45 degrees C-values of 35 min and over 65 min were determined for S. aureus and for B. subtilis spores which decreased to D = 2.9 min and D = 20 min respectively at a RH of 90%. At 45 degrees C, an optimal formaldehyde (HCHO) concentration for the inactivation of S. aureus was within the range of 0.75 to 1.5 mg HCHO l-1 air, with D-values of 2.9 to 1.8 min respectively and for Bacillus subtilis spores from 1.5 to 3.2 mg HCHO l-1 air with D-values of 10.9 to 4.0 min respectively. The most effective bactericidal and sporicidal activity of HCHO was observed at temperatures ranging from 40 to 45 degrees C. The higher the exposure temperature the lower were the HCHO residues on 100 cm2 aluminium plates with a structured lacquer. Exposure to formaldehyde concentrations of 0.4 mg l-1 air for 60 minutes resulted in lower residual formaldehyde of 5.9, 5.6, 4.7 and 4.5 micrograms HCHO 100 cm-2 sample surface at 30, 35, 40 and 45 degrees C, respectively. In contrast, after exposures to 3.2 mg HCHO l-1 air residues of 48.5, 43.4, 35.6 and 30.6 micrograms HCHO 100 cm-2 were found at the same temperatures.

Disinfection with gaseous formaldehyde. Fourth Part: Influence of aeration on the level of formaldehyde residues.

After a formaldehyde exposition of 100 cm2 aluminium plates covered with a structured lacquer during 60 min at 40 degrees C and relative humidity of about 90% in a model chamber formaldehyde residues of 9.5, 21.9 and 57 micrograms HCHO 100 cm-2 were recorded for plates exposed respectively to 0.75, 1.5 and 3.2 mg HCHO l-1 air. In the model chamber tempered at 40 degrees C the reduction of the HCHO residues after an additional 60 min ventilation with an internal air circulation was 63%. After a passive aeration for 60 min in a room at ambient temperature (24 +/- 3 degrees C) without ventilation the formaldehyde residues decreased by 70-74%. In a heating cabinet tempered at 40 degrees C after an internal active aeration of 60 minutes the residual formaldehyde diminished by 76-82%.

Disinfection with gaseous formaldehyde. Second Part: Influence of test materials on formaldehyde residues and the bactericidal and sporicidal effectiveness.

The pararosaniline method for the determination of formaldehyde residues on test surfaces after exposure to gaseous formaldehyde was standardized as well as the methods of collection, preparation and desorption from the samples. The analysis of residues on plates of 5 different materials yielded the following amounts of residues after 60 minutes exposure to 3.2 mg HCHO1-1 air at 45 degrees C and a relative humidity (RH) of about 90%; silicon rubber 287.2 micrograms, lacquered polyurethane foam 109.6 micrograms, lacquered aluminium 30.3 micrograms, plexiglass 13 micrograms and stainless steel 4 micrograms HCHO 100 cm-2. The residues of formaldehyde on lacquered aluminum after an HCHO exposition with condensing layer at 20 and 30 degrees C were 10(3)-fold higher than after an exposition to gaseous formaldehyde without a condensing layer at a relative humidity of about 90% and a temperature of 40 degrees C. The inactivations of S. aureus and Bacillus subtilis spores on carriers of 5 different materials were determined under the same conditions (60 min, 3.2 mg HCHO 1-1 air, 45 degrees C and a RH of about 90%). The decimal reductions showed that Staphylococcus aureus ATCC 6538 was more readily inactivated on non porous plexiglass with a D-value of 0.7 min or stainless steel D = 1.1 min than on porous silicon rubber D = 3 min. For spores of Bacillus subtilis var. niger DSM 675, D-values of 1.6 min for plexiglass, 2.3 min for stainless steel, 2.7 min for lacquered aluminium, 3.2 min for lacquered polyurethane foam and 4.1 min for silicon rubber were registered.

Disinfection with gaseous formaldehyde. First Part: Bactericidal and sporicidal effectiveness of formaldehyde with and without formation of a condensing layer.

Suspensions of Staphylococcus aureus ATCC 6538, Streptococcus faecium ATCC 6057, and of spores of Bacillus subtilis var. niger DSM 675 dried on polished stainless steel carriers were exposed in a model chamber to 3.2 mg HCHO l-1 air at temperatures of 20, 25, 30, 35 and 40 degrees C and the kinetics of their inactivation were determined by successive colony counting during the exposures. The HCHO treatment was carried out once with the formation of a condensing layer on the carriers and once without recondensation, at a RH of about 90%. In both procedures the cells or spores were suspended and dried in saline peptone water and additionally, in the case of the exposure without condensate layer, suspended in peptone water only. For S. aureus and S. faecium, significant differences between the two processes were only observed at 20 degrees C, whereby S. aureus showed for example a D-value of 14.8 minutes and after an exposure with condensate a D-value of 28.1 min. However at higher temperatures the effectiveness of HCHO in gaseous or condensate form was rather similar. At 35 degrees C the D-values after exposure to HCHO in condensate and gaseous form was for S. aureus 4.1 min and 5.9 min respectively, whereas after treatment at 40 degrees C D-values of 3.2 min and 3.8 min respectively were determined. At 35 and 40 degrees C D-values for S. aureus suspended in peptone water and exposed to gaseous HCHO were 50% lower than when suspended in saline peptone water. At lower exposure temperatures large differences were not registered. For the B. subtilis spores exposure to formaldehyde without condensation showed D-values of 34.8 and 5.6 min at 20 and 40 degrees C respectively. These are 8- and 4-fold lower than those of a corresponding exposure with a condensing layer. No D-value differences were observed for spores suspended and dried in saline peptone water or in peptone water after exposures to gaseous formaldehyde.

[Bacterial regrowth in drinking water. III. Reasons for regrowth with oligocarbotolerant bacteria]. / Bakterielle Wieder-Verkeimung im Trinkwasser III. Mitteilung: Ursachen der Verkeimung mit oligocarbotoleranten Bakterien.

Investigations have been undertaken into the bacterial regrowth in a Zurich drinking water plant and over a distance of 12 km along the drinking water distribution system. This required installation of eleven chromium steel dead-end water pipes. Counts of oligocarbotolerant bacteria were carried out in 7 or 8 repetitions in fresh water and water left to stagnate for 7 and 14 days respectively (7,8). Additionally ten different biological, chemical and physical parameters were determined in the fresh water samples. Multiple linear regression analysis was used to determine what influence the different parameters had on the bacterial regrowth during stagnation. Two regression models were evaluated, one for the data obtained during treatment, and the other for the data obtained along the distribution system. In both models the content of DOC (Dissolved Organic Carbon) and the content of phosphate was correlated with the growth of oligocarbotolerant bacteria. Further, a relationship between the contents of organic matter and bacterial regrowth was discovered during two measuring series conducted in different seasons. The model of data obtained during treatment generated two additional parameters correlated with aftergrowth: Firstly the initial colony content, which probably resulted because the stagnation period was only half that of the model. Secondly, the oxygen content which resulted because of ozonization. Under the given test conditions the following parameters did not appear in the regression model (Cp-statistics according to Daniel and Wood, 5): AOC (Assimilatible Organic Carbon) and UV-absorption used for measuring organic matter, nitrate content, content of chlorine/chlorine dioxide, electrical conductivity, pH-values and the room temperature during the 14 days stagnation in the distribution system model.

[Colony count and LPS content of gram-negative bacteria in cold storage foods and water]. / Koloniezahlen und LPS-Gehalte Gram-Negativer Bakterien in kühlgelagerten Lebensmitteln und Wasser.

In storage trials, the multiplication of gram negative bacteria was monitored by means of both colony counts and LPS-formation as determined by the three LAL tests methods (the Capillary test, the "Mini" Endotoxin test and the Coatest endotoxin method). The detection limits of the colony forming units which could be determined by the three LAL tests were initially set in model experiments in which beef (M. cleidooccipitalis) was variously inoculated with stationary-phase cells of Pseudomonas sp. previously grown at 30 degrees C. In all three methods, measurable amounts of LPS were possible only at colony counts above 10(3)/ml, g or cm2. The detection limits for colony counts of vacuum packed, aerobically stored beef muscle (Caput long. of M. triceps brachii) were found to lie between 1.0 x 10(3) and 1.0 x 10(4)/g for the Capillary test and the Coatest endotoxin method, and 5.0 x 10(4)/g for the "Mini" Endotoxin test. In the case of poultry carcasses the detection limits lay between 2.0 x 10(1) and 7.0 x 10(2)/cm2 and were thus considerably lower than for beef. With very low levels of bacterial loads, substrate interference in the LPS-LAL reaction must be taken into account. Linear regression analysis gave satisfactory correlation between the concentration of LPS and colony forming units for beef, poultry carcasses, ground beef and mixed salad. An acceptable proportionality was established only for beef and poultry carcasses. Potable water, stagnant for a fortnight in an experimental piping system and sampled at five different points, showed significant regrowth of oligocarbotolerant aquatic bacteria. However, only very low levels of LPS could be determined. All three LAL test methods can be recommended for rapid determination of the load of gram negative bacteria in meat and meat products.

[The effect of temperature on the growth and lipopolysaccharide production of gram-negative bacteria]. / Einfluss der Temperatur auf Vermehrung und Lipopolysaccharidbildung von Gram-negativen Bakterien.

5 strains of Pseudomonas and 7 of Enterobacteriaceae were cultivated in a Temperature Gradient Incubator (TGI) in intervals of 5 degrees C over a linear temperature gradient of 4 to 48 degrees C. After attaining the stationary growth phase the amount of lipopolysaccharides (LPS) formed was determined by means of the three LAL tests, i.e. the Capillary test, the "Mini" Endotoxin test and the Coatest endotoxin method. Simultaneously, the colony count was carried out. The same was done for one strain of E. coli in the exponential growth phase after continuous cultivation in 5 degrees C intervals over the range 10 to 35 degrees C. With the exception of the E. coli mutant P400, the lowest amounts of LPS produced in the stationary phase per 10(9) cfu was determined between 20 and 30 degrees C. Increasing or decreasing temperatures caused a more or less sharp rise in the quantity of LPS formed in the stationary phase. This was also demonstrated for E. coli in the exponential phase. For the E. coli mutant P400, however, the LPS content was highest at lower growth temperatures but decreased with increasing temperatures. Changes in the composition of fatty acids and sugars of the LPS and of membrane protein, which are dependent on the temperature of growth apparently alter the steric structure of the LPS which react with the LAL system, and could thus be responsible for the increased LPS found at the lower and higher growth temperatures. In the use of the LAL-tests methods for the assessment of the bacterial load of foods with gram negative bacteria, it is necessary, especially for cold stored products, to test for and determine the LPS-cfu relationship beforehand.

[The growth and lipopolysaccharide production of Escherichia coli and Pseudomonas aeruginosa]. / Vermehrung und Lipopolysaccharidbildung von Escherichia coli und Pseudomonas aeruginosa.

The course of growth and LPS production of two strains of type cultures of Escherichia coli (ATCC 11229, ATCC 25922), one E. coli mutant strain P400 and one type strain of Pseudomonas aeruginosa (ATCC 15442), grown partly by repeated cultures in BHI and partly also in minimal medium or in 1:10 diluted PC broth in a gyratory shaker (60 rpm) at 30 degrees C, was monitored respectively by counting the cfu and by simultaneous determination of LPS by means of the three miniaturized LAL-tests, i.e. the capillary test, the "Mini" endotoxin test and the Coatest endotoxin method. All three tests yielded generally comparable and reproducible results. The LPS content for a defined number of cfu was virtually in the same order of magnitude in all cases, regardless of the nutrient content of the culture medium. The quantity of LPS was relatively high in the initial phases of growth but then decreased significantly to constant levels in the stationary phase. There was a remarkable increase in the yield of LPS in the mid- and late stages of the exponential phase in the three strains, in contrast to the mutant in which the LPS content declined continuously. A possible explanation for this variation could be due to the fact that specific cell membrane proteins, which are lacking in the mutant, react differently with the LPS and thus with the Limulus Amoebocyte lysate. When the LAL tests are used for the rapid determination of the gram negative bacterial load of in particular perishable fresh foods, in which generally bacterial cells are at different stages of the exponential growth phase, then it is necessary to standardize each method specifically both for the product and for the storage conditions.

[Bacterial regrowth in drinking water. II. Drinking water distribution systems]. / Bakterielle Wieder-Verkeimung im Trinkwasser II. Mitteilung: Trinkwasser-Verteilersystem.

Five chromium steel dead-end water pipes were installed over a distance of 12 km along the Zurich city drinking water distribution system. Cell counts were determined in two series of four samplings in fresh water and stagnating water using three different methods. The colony counts of oligocarbon tolerant bacteria (1:10 diluted plate count agar, 20 degrees C, 14 d) in the fresh water was increasing along the distribution line. Initially there were counts around 1 CFU ml-1 and after 12 km between 120 and 1100 CFU ml-1. Water taken from house tabs showed higher colony counts than water taken after reservoirs. After a stagnating time of 14 d all 40 water samples showed aftergrowth from 10(3) up to 10(4) CFU ml-1. Water from the two sampling locations with the longest distance from the treatment plant showed less regrowth tendency. Epifluorescence microscopy and the INT-method for determining the electron transport system positive bacteria (ETS+) were less useful for monitoring bacterial regrowth. However, in the stagnating water there occurred a significantly higher percentage of ETS+ units as compared to the colony forming units (CFU) with growing distance from the treatment plant.

[Bacterial regrowth in drinking water. I. The upgrading of drinking water]. / Bakterielle Wieder-Verkeimung im Trinkwasser. I. Mitteilung: Trinkwasseraufbereitung.

Seven dead-end water pipes were installed after each treatment step in a drinking water plant. During a period of 7 weeks the bacterial load of freshwater and stagnating water was investigated with different methods. A modified surface spread plate count (Plate Count Agar, 10-fold diluted, 14 days incubation at 20 degrees C) proved to be more effective than the traditional pour plate method, because it gave consistently higher colony counts and had a lower level of detection (0.001 CFU ml-1). The enumerating of electron-transport-system positive bacteria yielded higher numbers than the colony count methods, but is not recommended when recently oxidized water samples are to be investigated. Highest cell counts were attained when using epifluorescence microscopic counting, yet bacterial regrowth could not be monitored thus. The tendency of bacterial regrowth was highest in freshly ozonized water. In stagnating lake water no regrowth occurred after 1 and 3 weeks because of the balance of bacteria and their predators.

[Comparison of three miniaturized Limulus-amoebocyte-lysate methods]. / Vergleich von drei miniaturisierten Limulus-Amoebocyten-Lysat Methoden.

Comparative examinations with three different micro-models for the LAL-test methods,--namely the capillary test, Endotoxin-test "Mini" and Coatest-Endotoxin-method-, tested with 11 Enterobacteriaceae- and 6 Pseudomonas-type-strains and isolates from foods gave only slight differences for the endotoxin content. Out of 250 single measurements per test method the capillary test as well as the Coatest-Endotoxin-method gave count limits between 2.8 X 10(1) and 7.4 X 10(3) CFU/ml. For the endotoxin-test method "Mini" the limits were one log unit higher-namely between 8.6 X 10(2) and 5.0 X 10(4) CFU/ml. The maximum values for count limits of all three methods correspond to approximately 1 ng Endotoxin. For the two gel test methods a relative error of +/- 15% has to be taken into account; for the chromogen method however, the relative error lies within +/- 25%, if the concentration is 0.0125 ng LPS/ml or lower. Between 0.025 and 0.1 ng LPS/ml or 0.075 and 0.1 ng LPS/ml the relative error is reduced to +/- 16% and +/- 5% respectively. The study has shown that the three micro tests in question as tested with pure cultures give comparable and reproducible results; inspite of the high relative error.

Microbiological characteristics of natural mineral water.

Natural, non-carbonated mineral water is, like every other natural water from a spring, never sterile. However, the microbial level is always very low. But after its bottling, the level rises rapidly and numbers of more than 10,000 to 100,000/ml can be reached. In principle 2 groups of bacteria of very different origin and properties can be found in the microbial flora of the bottled, non-carbonated mineral water. Allochthonous bacteria will get into the water by contamination from the containers, closures, air or the bottling machines. They are mostly transitory as they cannot grow in a substrate with an extremely low nutritive level and die off more or less rapidly. From the hygienic point of view the permanently contaminating flora with Pseudomonas aeruginosa as main representative is more serious. These special gram-negative bacteria are oligocarbotolerant and can therefore multiply in the mineral water of extremely low nutrient level after a certain adaptation. Their effective bacteriological control is possible by colony counting with incubation at +37 degrees C but only just after bottling. The autochthonous microbial flora consists of psychrotrophic and of distinctly oligocarbophilic, mainly gram-negative bacteria such as Achromobacter, Flavobacteria, Pseudomonas as well as gram-positive Arthrobacter-species. According to indirect experiences, this autochthonous microbial flora must be growing in the open system of the underground source and renew itself constantly. The bottling of the natural spring water implies a drastic environmental change from this open system into a closed one. Then the bacteria start multiplying more or less rapidly like in a batch culture. Main reason for this is the extension of the inners surface of the system. The multiplication of bacteria after bottling of a mineral water of extremely low nutrient level therefore is an entirely normal biological process. For this reason, limits of the aerobic colony count at +20 degrees C incubation for natural mineral water seem not to be justified.