Journal of the Scientific Agricultural Society of Finland Vol. 49: 107-166, 1977 Maataloustieteellinen Aikakauskirja THE EFFECT OF SOME GRAM-NEGATIVE BACTERIA ON THE RIPENING AND QUALITY OF DRY SAUSAGE Selostus: Eräiden gram-negatiivisten bakteerien vaikutus kestomakkaran kyp- symiseen ja laatuun ESKO PETÄJÄ Institute of Meat Technology,University of Helsinki Viikki, 00710 Helsinki 71, Finland TO BE PRESENTED, WITH THE PERMISSION OF THE Faculty of Agriculture and Forestry of the University of Helsinki, for public criticism at Viikki in Auditorium E on September 23,1977 at 12 o’clock. SUOMEN MAATALOUSTIETEELLINEN SEURA HELSINKI https://www.c-info.fi/en/info/?token=nS8lGgD0g4Kihd5q.TA9a3ceJz_g8Ru_fp2yR7A.1zvYsBOazxYTbKUIzgqYbUELkH8gOsg2rb6jsVr6tXw4UnhWOc-jfv9eOZAHnX4RMIXEupgRnIyzEPIs816eG-dw-Kf0cGl1nBXaiWDrXytqgP7b4sjrA4ykJ9M-jdaJ0wT6DAxRvYapwyDiw-4qa0HHS1mXrGM7M01Ts8fowr_uKdg Acknowledgements I express my sincere thanks to Prof. F. P. Niinivaara, the head of the Institute of Meat Technology, University of Helsinki, for his encouragement and interest in this investigation. For technical assistance I express my thanks to Miss Irma Hukkala and Mrs Pirkko Kos- tamo. To Mrs Aira Kuparinen and Mrs Anna-Maija Ullstedt, who assisted with the typing, my cordial thanks. I also extend my thanks to Olavi Kaitila and Jorma Outinen for their help in manufacturing the sausages. To Prof. Esko Nurmi, the head of the State Veterinary Medical Institute, and Kyllikki Ala- Huikku, Lie. vet. mod. I extend my thanks for carrying out the mouse experiments to test the apathogenicity of the strains used in the investigation. Some of the strains used in the research were received from Rudolf Muller & Co. for which I am grateful. My colleagues at the Institute of Meat Technology have been interested in this work and have given their help on many occasions during the investigation. I would like to express my gratitude for this assistance. To Suomen Maataloustieteellinen Seura (the Scientific Agricultural Society of Finland) I extend my thanks for including this study in their series of publications. I am grateful to The English Centre for checking the language of this work. Finally I would like to dedicate this work to Kaarina, Pasi and Jussi for their help and contributions during my research. Helsinki, September 1977 Esko Petäjä CONTENTS page Abstract 113 1. INTRODUCTION 113 2. REVIEW OF THE LITERATURE 114 2.1. Natural bacterial flora of dry sausage 114 2.2. Use of gram-positive bacteria and moulds in dry sausage 115 2.2.1. Micrococci 115 2.2.2. Lactic acid bacteria 116 2.2.3. The mixture of micrococci and lactic acid bacteria 117 2.2.4. Bacilli 119 2.2.5. Moulds 119 2.3. Gram-negative bacteria and their use in fermented meat products 119 2.3.1. Taxonomy 119 2,3,2. Metabolic properties 120 2.3.3. Toxins and pathogenicity 120 2.3.4. Gram-negative bacteria in meat curing 121 2.3.5. Gram-negative bacteria in the preparation of dry sausage 122 3. EXPERIMENTAL 123 3.1. Purpose of the study 123 3.2. Material 123 3.2.1. Preparation of sausages 123 3.2.2. Bacterial cultures used 124 3.2.2.1. Micrococcus sp 124 3,2.2.2. Lactobacillus plantarum 124 3.2.2.3. Aeromonas x 124 3.2.2.4. Aeromonas 19 125 3.2.2.5. Vibrio costicolus 126 3.2.2.6. A chromobacter strains 126 3.2.2.7. Escherichia coli 126 3.2.2.8. Proteus vulgaris 126 3.3. Methods 127 3.3.1. Organoleptic evaluation 127 3.3.2. Physical and chemical examinations 128 3.3.2.1. pH value 128 3.3.2.2. Consistency 128 3.3.2.3. Weight losses 128 3.3.2.4. Nitrite 128 3.3.2.5. Nitrate 128 3.3.3. Microbiological examinations 129 3.3.3.1. Substrates used 129 3.3.3,2. Methods employed 129 3.3.4. Statistical methods 130 3.3.5. Grouping of experimental sausages 131 3.4. Results 131 3.4.1. Use of Aeromonas strains as a starter culture 131 3.4.1.1. Organoleptic evaluation 131 3.4.1.2. Physical and chemical examinations 138 3.4.1.2.1. pH value 138 3.4.1.2.2. Consistency 140 3.4.1.2.3. Weight losses 142 3.4.1.2.4. Nitrite 143 3.4.1.2.5. Nitrate 144 3.4.1.3. Microbiological examinations 146 3.4.1.3.1. Total bacteria 146 3.4.1.3.2. Micrococci 147 3.4.1.3.3. Total bacteria on tributyrine agar 148 3.4.1.3.4. Lipolytic bacteria 149 3.4.1.3.5. Lactobacilli 150 3.4.1.3.6. Streptococci 151 3.4.1.3,7. Coliform bacteria 153 3.4.2. Use of Vibrio costicolus as a starter culture 154 3.4.3. Use of Achromobacter strains as a starter culture 154 3.4.4. Use of Escherichia coli as a starter culture 155 3.4.4. Use of Proteus vulgaris as a starter culture 155 4. DISCUSSION 156 4.1. Effect of bacterial inoculations on the ripening of dry sausage 156 4.2. Effect of bacterial inoculations on the bacterial flora of dry sausage 158 5. SUMMARY AND CONCLUSION 159 REFERENCES 161 SELOSTUS 164 113 JOURNAL OF THE SCIENTIFIC AGRICULTURAL SOCIETY OF FINLAND Maataloustieteellinen Aikakauskirja Vol. 49: 107-166, 1977 Petäjä, E. 1977. The effect of some gram-negative bacteria on the ripening and quality of dry sausage. J. Sdent. Agric. Soc. Finl. 49: 107 166. Abstract. The possible use of gram-negative bacteria as starter cultures in dry sausage was studied by inoculating the following gram-negative bacteria into dry sausage both alone and with Lactobacillus plantarum: Aeromonas x, Aeromonas 19. Vibrio costicolus, Achromobacter 22, A chromobacter guttatus, Achromobacter X, Escherichia coli, or Proteus vulgaris. Sausage without any inoculation and sausage containing Micrococcus sp. + Lactobacillus plantarum were used as references. Aeromonas x and 19 strains had a very favourable effect on the quality of dry sausage when inoculated together with lactobacilli. Both strains reduced nitrate, and the colour formed during the first 3 days. The pH value of Aeromonas + Lactobacillus sausages decreased so quickly that the consistency of these sausages developed within one week and was at least as good as, and often firmer than in Micrococcus + Lactobacillus sausages. The consistency of Aeromonas 19 + Lactobacillus sausages was better at the 0.05 signifi- cance level than that of Micrococcus ■+■ Lactobacillus samples. The aroma and flavour of Aeromonas + Lactobacillus sausages were as good as or better than those of Micro- coccus 4- Lactobacillus sausages. The aroma and flavour of Aeromonas 19 + Lacto- bacillus sausages were significantly better (significance level 0.001) than Micrococcus -f Lactobacillus sausages. The flavour of Aeromonas x -(- Lactobacillus sausages was better at the 0.05 level than Micrococcus + Lactobacillus sausages. The number of bacteria, roughly the number of lactobacilli. was often significantly higher in Aeromonas + Lactobacillus sausages (range 2—4 x 108 /g between 3 and 21 days of ripening) than in Micrococcus + Lactobacillus sausages (range 8 X 107 1.5 X 108/g between 3 and 21 days of ripening). For this reason the former sausages ripened more quickly than the latter. Vibrio 21 and three A chromobacter strainsdid not thrive in dry sausage and disappeared during the first 7 days of ripening. When inoculated together with lactobacilli,Escherichia coli had almost as favourable an effect as micrococci on the quality of dry sausage. Ino- culations of E. coli alone and Proteus vulgaris, both alone and with lactobacilli, did not improve the quality of dry sausage. P. vulgaris actually had a detrimental effect. Conclusion The investigation has proved that gram-negative bacterial strains Aeromonas x and 19 inoculated with lactobacilli improve the quality of dry sausage and that, especially with the Aeromonas 19 + Lactobacillus inoculum, better dry sausage was obtained than with micrococci and lactobacilli. 1. Introduction The manufacture of fermented food products, e.g. various dairy products, is based on the action of micro-organisms. The use of microbes ensures ripening and avoids the development of faulty products. The use of bacterial cultures 114 in dairy products or different starting portions of ripe material in vegetable and bread products has been known for quite a long time. On the other hand, starter cultures in meat products have only been used during the last three decades. Bacteria used as starter culture for dry sausage have been solely gram- positive, generally lactic acid bacteria and micrococci. Mixing lactic acid bacte- ria with micrococci has proved beneficial for dry sausage. In cured meat pro- ducts gram-negative bacteria have also been important. The traditional method has been to add old brine to new brine to transfer bacterial inoculum to the new brine. Gram-negative bacteria often prevail in curing brines. They have also been inoculated artificially into the brines because they improve the colour formation and flavour of cured products. Some investigations into the effects of gram-negative bacteria on the quality of dry sausage have been made, but in practice gram-negative bacteria have not been used as starter culture for dry sausage. Most previous studies date from the 1950’s and are not quite applicable to modern processing techniques, because shorter ripening times are used nowadays. Because gram-negative bacteria can have a beneficial influence on meat curing and because their use in the manufacture of dry sausage has been little studied, the effect of some gram-negative bacteria on the ripening and quality of dry sausage will be clarified in this investigation. Attention will primarily be directed to the question of whether or not gram-negative bacteria can be used as starter culture for dry sausage. 2. Review of the literature 2. 1. Natural bacterial flora of dry sausage According to Coretti (1956 a, 1956 b) dry sausage contains mixed bacterial flora; the number of bacteria as ripening begins is about 4 x 108/g. Möller (1955) investigated the bacterial flora of the dry sausage masses from different factories and observed gram-negative bacteria dominating the flora. The number of bacteria increased according to Coretti (1956 a, 1956 b) from 4 X 106 /g to 3 X 108/g over 2 weeks. The number of enterococci varied from 1.6 x 104/g to 5 X 105/g, the number of yeasts being 4 X 103 3.1 X 104/g. According to the same investigation the number of lactobacilli was very small when ripen- ing began while in the ripened product lactobacilli predominated. According to Szczepaniak and Pezacki (1974) cocci formed the dominant microbial flora in dry sausage and smoking resulted in the destruction of gram-negative rods and yeasts. The total microbial counts increased from one million/g to 6—33 million/g during ripening. Reuter et ai. (1968) observed that the number of bacteria increased from 104 10 5/g to 108/g during the first 23 days of ripening. The increase in lacto- bacillus counts during that time was from I02 104/g to 108/g. The number of bacteria belonging to the genera Leuconostoc and Pediococcus was o—l02/g 115 during the first 4 days, increasing later to a level of 103—l0 4/g. The sausage mass before stuffing contained 104 —lO 5 enterobacter cells per gram, the count after 23 days being 102 or less per gram. The prevailing level of streptococcus and bacillus cells was 104 105/g throughout ripening. Lerche (1956 a, 1956 b) found that the lactobacillus count at the beginning of ripening was low but increased considerably over 45 days. During drying almost all gram-negative and a great part of the gram-positive bacteria disap- peared and the ready product contained almost exclusively lactobacilli, micro- cocci and yeasts. Fiszer (1970) stated that during the first 10 days of ripening the heterofermentative lactobacillus strains prevailed in dry sausage and there- after the homofermentative strains dominated. However Urbaniak and Pe- zacki (1975) proposed that the situation is just the opposite that the homo- fermentative strains predominate at the beginning with heterofermentative strains being found increasingly later on. They also stated that the number of gram-negative bacteria decreased during ripening. Niinivaara and Pohja (1956, 1957 a, 1957 b) found that during the be- ginning of the ripening of dry sausage the number of bacteiia increased abun- dantly, micrococci forming the majority. The number of lactobacilli was also considerable but the count of gram-negative bacteria decreased, especially after smoking. The final flora were chiefly composed of lactobacilli. Ten Cate (1960) observed that gram-negative bacteria disappeared from dry sausage as early as the first 2 or 3 days of ripening. Lactobacilli and Lancefield D group streptococci remained in the sausage. Also, according to Erikson (1961) gram-negative bacteria generally disappears completely and the micrococcus count decreases while Leuconosloc strains form the majo- rity of bacterial flora. Coretti (1958) found Leuconosloc species in faulty dry sausages. Kasbohm (1954) studied salami and cervelat sausages and found that the good products contained the most bacteria, especially lactobacilli. There was also an abundance of lactostreptococci, micrococci and yeasts. The sausages contained small but constant amounts of bacteria belonging to the mesentericus- subtilis group, and gram-negative bacteria. 2. 2. Use of gram-positive bacteria and moulds in dry sausage 2.2.1. MICROCOCCI It has been shown in many investigations that micrococci play an essential part in the ripening of dry sausage. When studying the bacterial flora of dry sausage, Lerche (1956 b, 1957) found about 50 bacterial species, with micro- cocci forming an essential part of the microbial flora. He also stated that smoking reduced the number of micrococci in dry sausage and that micrococci improved the colour of dry sausage depending on their ability to reduce nitrate. Niinivaara (1955) carried out extensive investigations into the influence of Micrococcus strains on the ripening process of dry sausage. A great number of Micrococcus strains were isolated in his work. A strain denoted M53 proved to be the best as to its starter culture properties. It accelerated the drop in the 116 pH value so that in sausages inoculated with micrococci, a pH value of 5.4 was achieved in 2 days while in control sausages 5 days were needed for this. According to Niinivaara the development of colour depends on the re- duction of nitrate and nitrite and on the pH value. The development of normal colour took 5 days in the control sausages but only 2 days in the sausages inoculated with micrococci. Further, Niinivaara observed that the control sausages were sometimes spoiled because of the influence of other bacteria. The Micrococcus strain M53 was found to have an antagonistic effect on the growth of other bacteria. Inoculation ensured the successful ripening of dry sausage and it was also possible to reduce the ripening time from 14 days for control sausages to 9 days for sausages containing micrococci. The acceleration and assurance of ripening are based on the following properties of Micrococcus M 5 3: its ability to reduce nitrate and to reduce the pH value of sausage, and its antagonism towards other microbes. Niinivaara and Pohja (1956) further clarified the role of micrococci in the preparation of dry sausage. According to them the micrococcus count increased more than that of other bacteria during the predrying of dry sausage. They also found, like Lerche (1956 b), that smoking reduced the number of bacteria in dry sausage. In their later investigations (Niinivaara 1958, Niinivaara and Pohja 1957 a, 1957 b, 1957 c, Pohja and Niinivaara 1957) they stated that the Micrococcus strain M 53 had an antagonistic effect upon most of the bacteria isolated from dry sausage, but not, however, upcn the Lactobacillus strains identified as Lactobacillus plantarum and Lactobacillus leichmannii. In 1957 the Micrococcus strain M 53 began to be used in the meat industry in the form of a preparation called »Baktoferment» (manufactured by Rudolf Muller & Co., Giessen, ERG). This strain, isolated originally by Niinivaara, lost its activity fairly quickly and the strains used at present were isolated later. It is important, therefore, that new active strains be continually sought. Pohja (1960) has presented a method for the selection of new strains suitable for manu- facturing dry sausage. Kuchling (1963) found that nitrate-reducing micrococci promoted the deve- lopment of a red colour but only during the first day after preparation. He further showed that micrococci have no effect on the aroma of the sausage but that they may disturb acid formation and so bring about faulty products. Kucharkova et al. (1963) found a Micrococcus aquatis strain which improves the quality of dry sausage. In Bulgaria a Micrococcus strain called P 4 has been used in the manufacture of dry sausage (Djevizov 1973, Djedjeva 1973). It improves the quality of sausage and shortens the processing time. 2. 2. 2. LACTIC ACID BACTERIA Jensen and Paddock (1940) were the first to clarify the possible use of lactobacilli in the preparation of semi-dry sausage. The strains used were Lacto- bacillus plantarum, Lactobacillus brevis and Lactobacillus leichmannii. They used both pure and mixed cultures and stated that lactobacilli reduced the ripening time, prevented the development of faulty products and improved the aroma of sausages. 117 Niven et al. (1955, 1958) and Deibel et al. (1961 a, 1961 b) extensively studied the possible use of lactic acid bacteria in semi-dry sausage called summer sausage in the USA. They found a Pediococcus cerevisiae strain which accelerated the ripening of semi-dry sausage and produced a firmer pro- duct with a strong, clean aroma. This strain does not reduce nitrate but forms lactic acid from glucose. Its optimum temperature is -|-37° C and it is therefore applied in the manufacture of summer sausage, the internal temperature of which rises to 35—40° C. The preparation takes some days. This strain has not been adopted for use in the manufacture of dry sausage in Europe. The Pediococcus cerevisiae strain isolated by Niven and Deibel has been used commercially as a lyophilized preparation called »Accel» (manufactured by Merck & Co., Rahway, N. J., USA). Merck & Co. later developed a frozen pediococcus preparation (Lactacel) which is more active than the lyophilized one (Fwerson et al. 1970). More recently frozen cultures of Lactobacillus plan- tarum (Lactacel DS) and a mixture of L. plantarum and P. cerevisiae (Lacta- cel MC) have been made available (Deibel 1974). The starter culture preparations Saga and Saga II (Manufactured by Micro- life Technics, Sarasota, Fda, USA) contain lactic acid bacteria as a frozen suspension. Saga is for use in sausages fermented at smokehouse temperatures of 32—37° C and Saga II in sausages fermented at greenroom temperatures of 18—32° C (Shah 1977). The cultures of lactic acid bacteria are also manu- factured in France (Liepe 1975). According to Coretti (1977) the French pre- parate called »Fermentes lactiques» contains lyophilized streptococci (Strepto- coccus lactis) in dried fat-free milk powder. 2. 2. 3. THE MIXTURE OF MICROCOCCI AND LACTIC ACID BACTERIA Niven (1952 a, 1952 b) found that both bacteria producing lactic acid and nitrate-reducing bacteria are required in the ripening of dry sausage. In spite of this, mixed starter cultures containing only homofermentative and hetero- fermentative lactobacilli were used in many plants in the USA (Leistner 1963). Nurmi (1965, 1966) developed a method in which both micrococci and lactobacilli were used in the preparation of dry sausage. According to Nurmi, lactobacilli used alone to inoculate sausage produced an essentially better consistency than micrococci. The consistency of control sausages was also much poorer. However, when the usual nitrate quantity was used, lactobacilli brought about discolourations and an off-flavour. This was caused by hydrogen peroxide formed by lactobacilli and oxidation of fat caused by hydrogen per- oxide. The inoculation of just micrococci did not markedly accelerate the dec- rease in pH value and so the consistency developed no better than in the control sausages. The micrococci alone improved the colour of the cut surface of dry sausage a little. When lactobacilli were inoculated along with micrococci into dry sausage, the bacteria were observed to considerably improve the quality of dry sausage. The lactobacillus inoculation markedly accelerated the ripening process while, 118 on the other hand, micrococci prevented the discolourations caused by lacto- bacilli and reduced nitrate, thus promoting the formation of colour. Accor- ding to Nurmi it is probable that the catalase formed by micrococci destroys hydrogen peroxide before it causes any harm. Rozier et al. (1970) also observed the catalase activity of micrococci necessary to assure the colour formation and stabilization by decomposing peroxides formed by lactic acid bacteria. Nurmi’s experimental sausages inoculated with lactobacilli and micro- cocci ripened on average within 67 days with a weight loss of about 10 %. The ripening time of control sausages was 19 days and respective weight loss 20 %. The lactobacillus inoculation required was given as 1— 10 million cells per gram. Nurmi used a micrococcus inoculation of 5 lO million cells per gram. Nurmi concluded from his investigations that it is better to use lactobacilli and micrococci together as starter culture than either separately. He stated that lactobacilli considerably accelerate the decrease in pH and so promote the formation of a firm consistency. On the other hand micrococci promote colour formation and prevent discolouration and bad flavour which might be caused by inoculating only lactobacilli into sausages. From Nurmi’s investiga- tions a starter culture called »Duploferment 66» has been developed containing frozen or lyophilized micrococci and lactobacilli (manufactured by Rudolf Muller & Co., Giessen, FRG). The starter culture Saga 111 (manufactured by Microlife Technics, Sarasota, Fda, USA) also contains frozen micrococci and lactic acid bacteria and is used for dry sausages fermented at 10 32° C (Shah 1977). Some mixed culture preparations are also on the market in Spain. According to Coretti (1977), these preparations called »Lamirlac» are mixtures of lyophilized lactobacilli, pediococci and micrococci and are prepared for different dry sausages, and dried and cooked hams. Inal (1969) investigated the effect of Micrococcus aurantiacus and Pedio- coccus cerevisiae inoculated together on the ripening and quality of dry sausage. This starter culture combination ensured ripening and improved the colour, consistency and aroma of the sausages. Sutic and Joksimovic (1973) inoculated dry sausage separately with Micrococcus strain M-104 and Streptococcus strain Ak-60 and with both strains together. The best quality, especially taste and aroma, was achieved with a culture containing both streptococci and micrococci. There were only small differences between other experimental sausage types. The pH value of the sausages did not decrease during ripening. Reuter (1970, 1972) has studied the lactic acid bacteria of dry sausage. In addition to finding normal lactobacilli he found lactic acid bacteria which he called atypical streptobacilli. These bacteria were coccoid, grew at a lower temperature than normal lactobacilli and needed less oxygen than typical lactobacilli. Reuter (1972) also inoculated some atypical streptobacilli together with micrococci into dry sausage and observed that some strains had an un- favourable effect and some improved the quality, especially the flavour. Some strains even predominated in dry sausage over typical lactobacilli. 119 2. 2. 4. BACILLI In the 1950’s the role of bacilli was also studied. Schönberg (1953), Schön- berg and Walz (1954), Baldenius (1954) and Walz (1958) found that bacteria belonging to the genus Bacillus caused colour and flavour defects in dry sausage. Coretti (1958) inoculated Bacillus subtilis pure culture into dry sausage without any adverse effect on the products. In his opinion lactobacilli were the main reason for faulty products. Pohja and Niinivaara (1960) came to the conclusion that bacilli do not bring about any kind of change in dry sausage. 2. 2. 5. MOULDS Many countries manufacture dry sausage with mould flora on the surface. Most natural moulds are toxic and so not suitable for foodstuffs. Mintzlaff and Leistner (1972), however, have isolated a mould, Penicillium nalgiovensis, which is not toxic and has the right properties for use as a starter culture for dry sausage. The name of the commercial product developed from it is Edel- schimmel Kulmbach 72 (manufactured by Rudolf Muller & Co. Giessen, FRG). 2. 3. Gram-negative bacteria and their use in fermented meat products 2.3.1. TAXONOMY Bergey’s Manual (1974) classifies gram-negative bacteria according to cell shape, oxygen demands and metabolic character in the following way: aerobic rods and cocci facultatively anaerobic rods anaerobic bacteria cocci and coccobacilli anaerobic cocci chemolithotrophic bacteria. Gram-negative aerobic rods and cocci Pseudomonadaceae includes the following families: Azolobacteraceae Rhizobiaceae Methylomonadaceae Halobacteriaceae. This group also contains genera of uncertain affiliation: Alcaligenes, Aceto- bacter, Brucella, Bordetella, Francisella and Thermus. Gram-negative facultatively anaerobic rods includes two families: Entero- bacteriaceae and Vibrionaceae. Enterobacleriaceae contains 12 genera, among them being Escherichia, Salmonella, Shigella, Serratia and Proteus. Vibriona- ceae contains five genera; Vibrio, Aeromonas, Plesiomonas, Photobacterium and 120 Lucibacterium. There are also genera of uncertain affiliation like Chromobacte- rium, Flavobacterium, Haemophilus, Pasteurella etc. Gram-negative anaerobic bacteria includes the family Bacleroidaceae and genera of uncertain affiliation. The Neisseraceae family and two associated genera belong to the gram- negative cocci and coccobacilli group. Gram-negative anaerobic cocci consists of the family Veillonellaceae. Gram-negative chemolithotrophic bacteria contains the Nitrobacteriaceae family of organisms which oxidize ammonia or nitrite. This group also contains organisms that metabolize sulphur and organisms that deposit iron or manganese oxides. 2. 3. 2. METABOLIC PROPERTIES Among gram-negative bacteria there are metabolically versatile organisms forming acid from carbohydrates, alcohols etc., and which are both lipolytic and proteolytic. Nearly all gram-negative bacterial genera contain the strains that form acids from sugars and other compounds (Bergey’s Manual 1974). According to Caselitz (1965, p. 31, 125) lipolytic strains are to be found among bacteria belonging to the genera Pseudomonas and Aeromonas. Of the family Enterobacteriaceae the genus Proteus contains the most lipolytic strains (Ed- wards and Ewing 1972, p. 324—328). The genera Pseudomonas and Aero- monas contain many proteolytic strains (Caselitz 1965, p. 22, 119, Bergey’s Manual 1974). There are also proteolytic bacteria which belong to the Entero- bacteriaceae family. According to Edwards and Ewing (1972, p. 324 328) the genus Proteus in particular contains proteolytic strains. The lipolytic, proteolytic and saccharolytic activities of gram-negative bac- teria are factors important in the aroma and flavour when these bacteria are used as starter cultures in dry sausage (Keller 1954, Keller and Meyer 1954). According to Buttiaux (1959) aeromonads play an important part in the biological processes on which the manufacture of raw sausage is based. Many gram-negative bacterial strains such as pseudomonads, aeromonads and enterobacteria reduce nitrate to nitrite (Bergey’s Manual 1974). As is known the nitrate reduction is an important factor in the manufacture of dry sausage. 2. 3. 3. TOXINS AND PATHOGENICITY Most gram-negative bacteria possess endotoxins. Endotoxins are complex lipopolysaccharides derived from the walls of bacterial cells and often liberated when bacteria lyse. The physiopathogenic effects of all endotoxins are similar, although they appear to be distinct antigenically. The sequence of monosaccha- ride units imparts antigenic specificity to the molecule. The specific antibody combines with the polysaccharide portion of the molecule (Jawetz et al. 1968, p. 294). The different types of toxins of gram-negative bacteria are classified accord- ing to their antigenic nature as follows: somatic (O), flagellar (H) and capsular (K) antigens. The common antigen (CA) and a thermostable antigen of salmo- 121 nellae are also known (Edwards and Ewing 1972, p. 48, 54). Mäkelä and Meyer (1976) suggest that all bacteria belonging to the family Enterobacteriaceae have the common antigen (ECA) and that other gram-negative bacteria do not. The presence of this antigen can be a virulence factor. It is known that there are many pathogenic organisms among gram-negative bacteria, especially bacteria belonging to facultatively anaerobic rods such as enterobacteria. Bacteria belonging to the genera Salmonella and Shigella of the Enterobacteriaceae family are the most pathogenic. However, some Esche- richia coli and Proteus strains are also pathogenic. Some E. coli strains may even be very pathogenic and cause diarrhoeal inflammations and other infec- tions (Bergey’s Manual 1974). Among pseudomonads Pseudomonas aeruginosa, and among vibrios Vibrio comma, are generally known to be pathogenic. Most Pseudomonas and Vibrio bacteria are, however, apathogenic and saprophytic. The genus Aeromonas includes bacteria which are pathogenic for fish, frogs and other water animals (Caselitz 1965, p. 57 —59). According to Caselitz (1965, p. 59—60) A. hydro- phila is also pathogenic for mice, guinea pigs, rats, pigeons, and rabbits. Caselitz also stated that pathogenic aeromonads are haemolytic. Dahle and Nordstoga (1968) also found aeromonas bacteria pathogenic for mice and possibly for blue fox but not for mink or seals. Caselitz (1965, p. 74 77) presents five cases in whichAeromonas hydrophila has been pathogenic for man. The diagnosis was septicaemia twice, diarrhoea twice and osteomyelitis once. 2. 3. 4. GRAM-NEGATIVE BACTERIA IN MEAT CURING Bacteria with a beneficial effect on meat curing may be both gram-positive and negative. In many countries a curing method in which old brine is mixed with new is used. The purpose of this method is to transfer bacteria to the brine to be made before the curing process. Leistner (1960) examined the effect of bacteria in brines made in the above way on the brine itself, and on the pro- ducts cured in those brines. In his opinion micro-organisms may improve the colour or flavour of the ready products, and stabilize brines. Leistner observed that micrococci and lactobacilli improve the stability of curing brines and prevent their spoilage. He is also of the opinion that bacteria belonging to the genera Vibrio, Spirillum, Achromobacter, Alcaligenes and Micrococcus improve colour formation by reducing nitrate or affecting the pH and redoxpotential values favourably. Leistner observed the influence of bacteria on the flavour of cured products. At least micrococci improve the flavour. Also, the optimal mixed cultures have a beneficial effect on the flavour of cured meat products. One mixed culture which Leistner found beneficial contained bacteria belonging to the following genera: Micrococcus, Vibrio, Spirillum, Achromobacter, Alcaligenes, Lacto- bacillus, Microbacterium and Corynebacterium. The effect is stronger if the cul- ture is inoculated both into the cover brine and the pumping brine. Buttiaux (1957) isolated a nitrate-reducing Vibrio costicolus strain which was used in meat curing in France. It was of great importance because the use 122 of nitrite was not allowed in France. In Germany V. costicolus was also found to have a beneficial effect on the flavour of cured products. In Great Britain Achromobacter strains have been used in brines to improve the quality of meat products (Hawthorn and Leitch 1962, p. 281). These strains have been found to have the same biochemical properties as the Vibrio costicolus of Buttiaux. Petäjä (1972, 1973) found Vibrio costicolus and Achromobacter strains which proved successful in meat curing, used both in cover brine and especially in the brine to be injected into meat. The strains were used as starter culture in dried hams. They had a favourable effect on the colour, consistency and flavour of hams. The respective control hams were raw after 7 days of ripening. Gorbatov et al. (1974) used Achromobacter guttatus with Streptococcus lactis, Lactobacillus plantarum, Micrococcus caseolyticus and Micrococcus varians in the curing of canned »Vetchina hams» and found that this kind of mixed culture improves the flavour of hams. 2. 3. 5. GRAM-NEGATIVE BACTERIA IN THE PREPARATION OF DRY SAUSAGE The influence of gram-negative bacteria on the quality of dry sausage, especially on the aroma and flavour, was investigated in the 1950’5. Kohnle (1953) stated that some Achromobacter and Escherichia strains produce the characteristic aroma of dry sausage and that micrococci and lactobacilli do not produce any kind of aroma. Alcaligenes, Pseudomonas and Aerobacter species are also able to create characteristic aromas. Keller (1954) investigated the effect of bacterial strains isolated by Kohnle and belonging to the genera Alca- ligenes, Pseudomonas, Achromobacter, Aerobacter and Escherichia on the aroma of dry sausage and found one Escherichia strain and two Achromobacter strains to have the best effect. Keller and Meyer (1954) inoculated dry sausages with 5 Escherichia, 2 Alcaligenes, 1 Pseudomonas and 2 Achromobacter strains and found one Escherichia strain to have an especially favourable effect on the flavour of dry sausage, this being caused by fat breakdown products. Meyer (1954) also investigated the effect of Kohnle’s strains on the aroma of dry sau- sage. One Escherichia strain proved best. Losem (1956) also stated that this strain improves the flavour of dry sausage. He also isolated it from dry sausage after a ripening period of 9 weeks. Eckert (1958) inoculated dry sausage with Micrococcus M 53 (Niinivaara 1955) together with an Escherichia strain and obtained sausages whose colour and aroma were especially good. Both strains disappeared from the sausage during the ripening period of 40 weeks. The ripening time of the sausages recorded by the school of Keller was long compared with processing times nowadays. Also, the pH value of the sausages was high over 6.0 making it possible for gram-negative bacteria to grow and thrive in dry sausage. Nurmi (1966) inoculated dry sausage with two nitrate-reducing strains, one being Escherichia coli and the other Aerobacter cloaceae according to Ber- gey’s Manual (1957). The strains were used together with Lactobacillus plan- 123 tarum. Aerobacler + Lactobacillus sausages had a peculiar mild flavour cha- racteristic of sausages inoculated with lactobacilli. When Escherichia coli was used with lactobacilli the peculiar flavour was not found. The colour of the sausages was essentially a lighter red than that of sausages inoculated with lactobacilli and micrococci. 3. Experimental 3. 1. Purpose of the study Gram-negative bacteria are known to be useful in curing brines and improve the aroma and flavour of the cured meat products. The purpose of this inves- tigation is to establish whether or not it is possible to use gram-negativebacteria as a starter culture in dry sausage and to find new, improved strains to be used as a starter culture in dry sausage. This is done by inoculating dry sausage with the following bacteria both alone and with lactobacilli: Aeromonas x, Aeromonas 19, Vibrio costicolus, three Achromohacter strains, Escherichia coli and Proteus vulgaris. The effect of inoculated bacteria on the ripening, quality and bacterial flora of dry sausage is examined. 3. 2. Material 3. 2. 1. PREPARATION OF SAUSAGES Twenty-two series of experiments were used in the investigation. Each series contained 6 different sausage groups and a group of 67 identical sausages. Sausages were prepared using the following formula: Beef 33.4 % Pork 33.4 % Pork fat 30.0 % Seasoning salt mixture 3.2 % 100.0 % Seasoning salt mixture: Sodium chloride 89.7 % Glucose 9.0 % Potassium nitrate 0.625 % (200 ppm in sausage) White pepper 0.7 % 100.0 % One batch contained 34.2 kg of sausage mass; the raw materials were mixed and ground in a cutter of 200 1 capacity (Seydelmann, manufactured by Maschi- nenfabrik Seydelmann, Stuttgart, FRG). The mixture was left quite rough and divided into 6 equal parts each containing 5.7 kg. Each portion was inoculated and grinding was completed in a 40 1 cutter (Seydelmann, Rasant 124 K-40, manufactured by Maschinenfabrik Seydelmann, Stuttgart, FRG). 100 ml of cell suspension was used for each 5.7 kg. The control portion was pre- pared first and the cutter was cleaned after the preparation of each portion. The sausage mass was stuffed into a casing 60 mm in diameter (Visko PK X, manufactured by Oy Visko Ab, Hanko, Finland), the weight of one sausage being about 0.5 kg. The sausages were ripened as follows: ripening time temperature humidity 0-4 days 22° C 95 % 4-7 days 21° C 90 % 7-21 days 15-16° C 80% The ripening of the sausages was followed for 21 days. The samples inoculated with lactobacilli ripened in 10 days while sausages with other inoculations or without any inoculation sometimes did not ripen even in 21 days. After 14 days the sausages had ripened so that they could best be compared with each other. 3. 2. 2. BACTERIAL CULTURES USED 3. 2.2. 1. Micrococcus sp. The strain employed was isolated from the commercial product »Bactoferment 61» (manufactured by Rudolf Muller & Co, Giessen, FRG). The cells were cultivated in the nutrient broth (see 3. 3. 3. 1. h) by aeration, the incubation temperature being 30° C. It took I—2 days to produce enough cells. The cells were separated centrifugally. The cell mass was covered with glucose, frozen and lyophilized (Lyophilizer Christ Delta 1/1 a, manufactured by Martin Christ, Osterode am Harz, FRG). The lyophilized cells were stored for 4 days before use. Lyophilized cells were suspended in 2 % sodium chloride solution just before inoculation into the sausage mass. The inoculations varied between 106 and 10 7 cells/g of sausage mass. 3. 2. 2. 2. Lactobacillus plantarum The Lactobacillus strain used was L. plantarum (Nurmi 1966). It was isolated from the commercial product »Duploferment 66» (manufactured by Rudolf Muller & Co, Giessen, FRG). The cells were cultivated in MRS broth (see 3. 3. 3. 1. i) by aeration, the incubation temperature being 30° C and the time one day. The cells were separated, covered with glucose, frozen, lyophilized and stored for 4 days before use, as with micrococci. Lyophilized cells were suspended in 2 % sodium chloride solution before inoculation. The inoculations varied between 106—lO 7 cells/g of sausage mass. 3. 2. 2. 3. Aeromonas x Aeromonas x was isolated from dry sausage (control sausage of experimental series IV in this study). The cell preparates used as an inoculum were made in 125 the same way as micrococci preparates. However the cultivation of one day was enough. The inoculations varied between 106 and 10 7 cells/g of sausage mass. Some properties of Aeromonas x were investigated. According to the results, Aeromonas x is a gram-negative rod with polar flagella. Its properties are as follows: fermentative oxidase positive catalase positive lipolytic grows well in salt concentrations under 5 % and at temperatures of 20-37° C reduces nitrate liquefies gelatin forms acids from the following substrates: glucose, fructose, mannose galactose, sucrose, maltose, xylose, raffinose, mannitol, sorbitol, glycerol and salicin. does not form acid from the following substrates; lactose, arabinose, rhamnose and esculin. not haemolytic Aeromonas x has been classified as belonging to Aeromonas genus on the basis of Bergey’s Manual (1974). In Bergey’s Manual three species (A. hydro- phila, A. punctata, A. salmonicida) and some subspecies are presented. The properties of Aeromonas x do not correspond to the properties of the three species and subspecies described. Pathogenicity The examination was carried out by inoculating intraperitoneally different numbers of bacterial cells into white mice. As a result of the apathogenicity examination of Aeromonas x, the LD 50 value lies between 5 X 107 and 10s . It is known that gram-negative bacteria produce endotoxins and are so able to kill mice when inoculated in sufficient quantities. According to Mäkelä (1975) gram-negative bacteria having an LD S 0 value of 10s or more are apathogenic. Valtonen (1970) considered Salmonella typhi- murium strains having a value of 10s avirulent. 3. 2. 2. 4. Aeromonas 19 Aeromonas 19 was isolated from dry sausage (control sausage of experimental series XVII in this research). The cell preparation was made and inoculated in the same way as that of Aeromonas x. The inoculations varied between 106 and 10 7 cells/g of sausage mass. Aeromonas 19 is a gram-negative rod with polar flagella. Its properties are as follows: fermentative oxidase positive catalase positive 126 lipolytic grows well in salt concentrations under 5 % and at the temperatures of 20-37° C reduces nitrate liquefies gelatin forms acids from the following substrates: glucose, fructose, mannose, galactose, sucrose, maltose, xylose, mannitol, sorbitol, glycerol and salicin. does not form acid from the following substrates: lactose, raffinose, rhamnose and esculin. Acid was formed only weakly from arabinose. not haemolytic Aeromonas 19 has also been identified according to Bergey’s Manual (1974) and does not correspond exactly with three species or subspecies pre- sented, either. Pathogenicity The LD 50 value of Aeromonas 19 was observed to be about 108 . So according to what has just been said (chapter 3. 2. 2. 3.) it can be regarded as apathogenic. 3. 2. 2. 5. Vibrio costicolus The Vibrio costicolus strain used had been isolated from curing brine (Pe- täjä et ai. 1973). A lyophilized V. costicolus preparation was made and inoculated in the same way as that of Aeromonas strains, the inoculum being 10 6 —l0 7 cells/g of sausage mass. 3. 2. 2. 6. Achromobacter strains Achromobacter 22 (Petäjä et ai. 1973), Achromobacter guttatus (received from Nora Winterhalter of Eidgenossische Technische Hochschule, Zurich) and Achromobacter X (isolated from control dry sausage, experimental series XIV) were used as starter cultures. The strains were classified according to Bergey’s Manual (1957). The lyophilized preparations were made and ino- culated in the same way as those of Aeromonas x and 19, the inoculum being 106—l0 7 cells/g of sausage mass. 3. 2. 2. 7. Escherichia coli The strain of Escherichia coli used was obtained from the Institute of Micro- biology of the University of Helsinki. E. coli cells were inoculated as a lyo- philized preparation made in the same way as Aeromonas preparations. The cells were suspended in 2 % sodium chloride solution just before inoculation, the inoculum being 10 6 107 cells/g of sausage mass. 3. 2. 2. 8. Proteus vulgaris The strain of Proteus lulga'is used originated from the Institute of Micro- biology of the University of Helsinki P. vulgaris was inoculated as a lyo- 127 philized preparation made in the same way as Aeromonas preparations. The cells were suspended in 2 % sodium chloride solution just before inoculation, the inoculum being 106—l0 7 cells/g of sausage mass. 3. 3. Methods 3.3.1. ORGANOLEPTIC EVALUATION The panel carrying out the evaluation consisted of five persons familiar with the organoleptic evaluation of dry sausage. Before evaluation, the quality desirable in experimental sausages was established. This was achieved by training evaluations. The experimental sausages were evaluated when they were 1,3, 7, 10, 14 and 21 days old. The following properties were evaluated: colour of sliced surface, consistency, aroma and flavour. The flavour of 1 and 3 day old sausages was not evaluated. A scoring system and a descriptive method were used side by side as the evaluation method. The choices offered to the panel members were as follows: Colour of sliced surface 3 bright red 2 red 1 brown, yellowish red 0 other bad discolourations Consistency 3 firm 2 quite firm 1 slightly soft 0 soft Aroma 3 excellent 2 good 1 odourless 0 unpalatable Flavour 3 excellent 2 good 1 moderate 0 bad Because the panel members could also award half points the points were multiplied by 2 before statistical processing. 128 3. 3. 2. PHYSICAL AND CHEMICAL EXAMINATIONS 3. 3.2. 1. pH value pH values were measured after 0,1, 3. 7, 10, 14 and 21 days of ripening using a Findip digital pH & mV meter 555 A (manufactured by Oy Findip Ab, Kauniainen, Finland). 3. 3. 2. 2. Consistency Consistency was measured using an Instron instrument MOB2l TM-M (manufactured by Instron Limited, High Wycombe, Bucks, England). A 7 cm long piece of sausage was used in the measurements. The sausage was com- pressed 1 cm with the measuring device which was 5.6 cm in diameter and the instrument registered the force needed. Measurements were made in three positions and their mean value was calculated. The results were expressed in kilograms. Consistency was measured after 1,3, 7, 10, 14 and 21 days of ripen- ing at room temperature. 3. 3. 2. 3. Weight losses Weight losses were determined by weighing the sample sausages to be stu- died before other determinations. This means that figures for weight losses were obtained after ripening times of 1,3, 7, 10, 14 and 21 days. 3. 3. 2. 4. Nitrite The quantity of nitrite was determined according to the method of Stoya (1969). The protein was precipitated using di-sodium tetraborate-10-hydrate and zinc sulphate-7-hydrate. In this study the precipitate was separated centrifugally instead of by filtration. Nitrite was determined from the liquid phase using sulfanilic acid and a-naphthole. The determination was carried out after 0, 1,3, 7, 10, 14 and 21 days of ripening. 3. 3. 2. 5. Nitrate The nitrate quantity was also determined by the method of Stoya (1969). The protein precipitation took place in the same way as in the nitrite determina- tion. In the liquid phase the nitrate was reduced to nitrite by cadmium released by reducing cadmium sulphate with zinc powder. The total nitrite content was determined using the sulfanilic acid and a-naphthole reagents. The nitrate content was calculated using nitrite and nitrate standards determined in con- junction with the analyses. The nitrate was determined after 1,3, 7, 10, 14 and 21 days of ripening. 129 3. 3. 3. MICROBIOLOGICAL EXAMINATIONS 3. 3.3. 1. Substrates used a) Plate count agar for counting total number of bacteria, Merck No 5463. Manufactured by Merck, Darmstadt, FRG. b) Mannitol salt agar for counting staphylococci and micrococci. Manu- factured by Orion Oy, Helsinki, Finland. c) Tributyrine agar for counting lipolytic bacteria. Manufactured by Orion Oy, Helsinki, Finland. d) Rogosa agar for counting lactobacilli (Rogosa et al. 1951). Manu- factured by Orion Oy, Helsinki, Finland. e) Slanetz agar for counting enterococci and streptococci of the Viridans group (Slanetz and Bartley 1957). Manufactured by Orion Oy, Helsinki, Finland. f) VRB (violet red bile) agar for counting coliform bacteria. Manufactured by Orion Oy, Helsinki, Finland. g) Hugh-Leifson semi-solid agar for fermentation and oxidation examinati- ons in identification tests (Hugh and Leifson 1953). h) Nutrient broth according to Petäjä (1977) for mass cultivation of diffe- rent bacteria (developed for Vibrio costicolus at the Institute of Meat Technology of the University of Helsinki). i) MRS broth for mass cultivation of lactobacilli (de Man et al. 1960). 3. 3. 3. 2. Methods employed Experimental sausages were studied microbiologically after 0, I, 3,7, 10, 14 and 21 days of ripening. Dilution 1 was made in the following way: 10 grams of sausage was weighed into 90 ml of 2 % sodium chloride solution with a sterile spoon. The mixture was homogenized using an Ultra Turrax homogenizer TP 18/2 (manufactured by Janke & Kunkel KG, Staufen i. Br., FRG). The other necessary dilutions were also made in 2 % sodium chloride solution. If possible the count of bacteria was made by using plates on which the number of colonies varied between 30 and 300. The following examinations were made: a) Cultivation on plate count agar, plating method. Incubation for 4 days at 30° C. The total number of bacteria was determined on plate count agar. All the colonies were recorded. b) Cultivation on mannitol salt agar, spreader method Incubation for 2 days at 37° C. Round and entire colonies were recorded, the count including both micro- cocci and staphylococci. c) Cultivation on tributyrine agar, spreader method Incubation for 4 days at 30° C. 130 The lipolytic bacteria, of which inoculated Aeromonas strains separately, were determined on tributyrine agar by counting the colonies having a distinct lipolysis halo. The total number of bacteria was also counted, d) Cultivation on Rogosa agar, plating method. Incubation for 4 days at 30° C. The number of lactobacilli was determined on Rogosa agar. All the colonies were recorded. e) Cultivation on Slanetz agar, spreader method. Incubation for two days at 37° C. On Slanetz agar all red and reddish entire colonies were counted, the results thus including enterococci and streptococci of the Viridans gioup. f) Cultivation on VRB agar, plating method. Incubation one day at 37° C. Coliform bacteria form deep red colonies on VRB agar I—2 mm in diameter. Only these kinds of colonies were recorded. g) Identification of Aeromonas strains. Some strains that grew well in good quality dry sausage were identified. The first studies consisted of the following examinations: gram staining and microscopic examination, growing experiments at temperatures of 20 37° C and in NaCl concentrations of 0— 5 %, reduction of nitrate, catalase and oxidase tests and further proteolytic, lipolytic and starch hydrolytic properties. Later the oxidation and fermentation tests and acid formation from the following substrates were investigated; glucose, fructose, mannose, galactose, sucrose, maltose, lactose, xylose, arabi- nose, raffinose, rhamnose, mannitol, sorbitol, glycerol, salicin and es- culin. h) Determination of pathogenicity of Aeromonas strains. The apathogenicity of Aeromonas strains was examined by inoculating intraperitoneally different numbers of bacterial cells into white mice. These studies were carried out in the State Veterinary Medical Institute, Helsinki. 3. 3. 4. STATISTICAL METHODS The results of the organoleptic evaluation were tested by analysis of variance both without separating the samples of different ages and also by handling sausages of different ages separately. The comparison of different sausage group pairs was carried out using the t-test. The different properties were treated separately. The following notations were used to express the degree of signi- ficance: +-f -f- samples different, significance level 0.05 The presence of any significant differences between evaluations of panel members was tested in order to criticize the organoleptic evaluation. The stan- dard deviation of the mean was used in the calculation of all the results. 131 3. 3. 5. GROUPING OF EXPERIMENTAL SAUSAGES There were altogether twenty-two experimental sausage series. The composition of these series varied so that in different sequences the effect of different bacteria was studied. The sequences of investigation can be recorded in the following way according to the experimental series: Experimental series Inoculated gram-negative bacteria I—6 Vibrio 21 and different Achromobacter strains 4 13 Aeromonas x, Escherichia coli and Proteus vulgaris 14—16 Achromobacter X 17 —22 Aeromonas 19 This work deals mainly with the effect of Aeromonas x and Aeromonas 19 on the quality of dry sausage. This is done by comparing their properties with those of control and Micrococcus + Lactobacillus sausages. The following schema will be used in describing the results: 1. Control group 2. Micrococcus -|- Lactobacillus group 3. Aeromonas x group 4. Aeromonas x -f- Lactobacillus group 5. Aeromonas 19 group 6. Aeromonas 19 + Lactobacillus group A short account will be given of the effect of other inoculated gram-negative bacteria on the ripening of dry sausage. 3. 4. Results 3. 4. 1. USE OF AEROMONAS STRAINS AS A STARTER CULTURE 3.4. 1. 1. Organoleptic evaluation Colour of sliced sausage Fig. 1 and Table 1 show the differences in the mean scores of the sausages after different ripening periods. The sausage groups containing Lactobacillus inoculations are the best, the differences between them being very small. The colour of Micrococcus -f- Lactobacillus sausages is slightly better than that of Aeromonas x + Lactobacillus aud Aeromonas 19 -\~ Lactobacillus samples after 14 days of ripening. The colour develops at about the same rate in all three groups. Table 1. Effect of bacterial inoculations on colour of sliced surface of experimental sausages (scores o—6). Sausage group 3 days 7 days 14 days 21 days 1. Control number of samples 21 21 20 19 mean 1.2 2.9 3.2 3.2 s.d.m 0.9 1.4 1.5 1.7 2. Micrococcus + Lactobacillus number of samples 15 15 15 13 mean 2.9 4.1 4.9 4.6 s.d.m 1.1 1.0 0.9 0.9 3. Aeromonas x number of samples 5 5 5 5 mean 1.3 2.6 3.1 4.0 s.d.m 1.3 1.8 1.3 1.4 4. Aeromonas x + Lactobacillus number of samples 6 6 6 6 mean 2.6 4.3 4.2 4.4 s.d.m 1.2 1.1 0.9 0.9 5. Aeromonas 19 number of samples 3 4 4 3 mean 2.7 3.0 3.9 3.6 s.d.m 0.1 1.7 1.6 1.6 6. Aeromonas 19 + Lactobacillus number of samples 6 6 6 5 mean 2.9 4.0 4.3 4.4 s.d.m 1.0 1.0 1.2 1.0 s.d.m. = standard deviation of mean In the control and Aeromonas x sausages the colour was significantly worse than in other samples (Table 2), being brownish and developing slowly. The colour of Micrococcus + Lactobacillus, Aeromonas x + Lactobacillus and Aeromonas 19 + Lactobacillus sausa- ges proved significantly better (sig- nificance level 0.05) than that of oth- er samples. There was, however, no significant difference between them. Aeromonas 19 sausages were better than the control and Aeromonas x samples (Table 2).Figure 1. Effect of different bacterial inoculations on colour of sliced surface of sausage. 132 133 Table 2. Comparison of colour of sausage pairs using the t-test. Scores for sausages of different ages are not separated. ++ + = samples different, significance level 0.05 Samples —>■ Control Micrococcus + Aeromonasx Aeromonas x + A eronion as 19 Lactobacillus Lactobacillus Micrococcus -f- ++ + Lactobacillus t Aeromonas x + + + Aeromonas x + + + + + + + Lactobacillus •<— t t Aeromonas 19 ++ + + + + + + «- Aeromonas 19+ + + + + + + + Lactobacillus 4 4 4 Consistency The mean consistencies of sausages of different ages are shown in Fig. 2 and Table 3. Aeromonas 19 + Lactobacillus sausages are generally the best with Micrococcus + Lactobacillus samples the next best followed by Aero- ntonas x + Lactobacillus sausages. This was also the order after 14 days of ripening. Taken throughout ripening, Aeromonas 19 sausages were better than the control and Aero- monas x sausages. The consistency of the control and Aeromonas x sausages was sig- nificantly worse than that of other samples (Table 4). The samples made using Lactobacillus inoculation were best. Sausages inoculated with mic- rococci and lactobacilli were not significantly better than Aeromonas x -f- Lactobacillus samples. Instead Aeromonas 19 -f- Lactobacillus and Micrococcus -f- Lactobacillus sausages differed at the 0.05 level, the former being the better. Aeromonas 19 + Lactobacillus samples were better than Aeromonas x + Lacto- bacillus ones at the 0.001 level. Aeromonas 19 sausages were better than samples with Aeromonas x. Figure 2. Effect of different bacterial ino- culations on consistency of sausage. Table 3. Effect of bacterial inoculations on consistency of experimental sausages (scores 0— 6) Sausage group 3 days 7 days 14 days 21 days 1. Control number of samples 21 21 20 19 mean 0.9 1.9 2.5 2.7 s.d.m 1.0 1.2 1.5 1.2 2. Micrococcus + Lactobacillus number of samples 15 15 15 13 mean 2.2 3.5 4.2 4.5 s.d.m 1.3 0.9 1.0 1.1 3. Aeromonasx number of samples 5 5 5 5 mean 0.6 1.5 2.5 3.0 s.d.m 0.8 1.4 0.9 1.3 4. Aeromonas x + Lactobacillus number of samples 6 6 6 6 mean 1.8 3.4 3.8 4.2 s.d.m 1.2 0.7 0.8 1.2 5. Aeromonas 19 number of samples 3 4 4 3 mean 1.3 2.5 3.2 3.5 s.d.m 1.2 1.3 1.1 1.5 6. Aeromonas 19 + Lactobacillus number of samples 6 6 6 5 mean 2.9 4,1 4.5 4.4 s.d.m 1.4 0.9 1.1 0.9 s.d.m. = standard deviation of mean Table 4. Comparison of consistency of sausage pairs using the t-test. Scores for sausages of different ages are not separated. ++ + = samples different, significance level 0.05 Samples Control Micrococcus + A eromonas x A eromonas x -)- Aeromonas 19 Lactobacillus Lactobacillus Micrococcus -f ++ + Lactobacillus t A eromonas x + + + Aeromonas x -f- ++ + ++ + Lactobacillus 4^— t t Aeromonas 19 + + + + + + ++ + + A eromonas 19+ + + + + + + + + + + + + + Lactobacillus •<— •<— •<— •<— 134 135 Aroma The Aeromonas 19 + Lactobacillus sausages were the best for most of the ripening time (Fig. 3 and Table 5). The order of preference of the Micrococcus + Lactobacillus, Aeromonas x + Lactobacillus and Aeromonas 19 sausages varied during ripening. The order of preference of sausages with Lactobacillus inoculation after 14 days of ripening was as follows; Aeromonas 19 -(- Lacto- bacillus sausages were the best, second came Aeromonas x + Lactobacillus samples and third Micrococcus + Lactobacillus samples. The aroma of the control and Aeromonas x sausages was significantly worse than that of other samples (Table 6). The Aeromonas 19 samples were clearly better than these, being statistically as good as Micrococcus -j- Lactobacillus and Aeromonas x + Lactobacillus samples, but worse than Aeromonas 19 + Table 5. Effect of bacterial inoculations on aroma of experimental sausages (scores 0 —6). Sausage group 3 days 7 days 14 days 21 days 1. Control number of samples 21 21 20 19 mean 1.4 2.4 3.0 2.7 s.d.m 1.0 1.3 1.2 1.3 2. Micrococcus + Lactobacillus number of samples 15 15 15 13 mean 1.9 3.0 3.6 3.6 s.d.m 0.9 1.0 1.0 0.8 3. Aeromonas x number of samples 5 5 5 5 mean 1.1 2.2 2.0 3.4 s.d.m 1.0 1.4 0,9 0.7 4. Aeromonas x + Lactobacillus number of samples 6 6 6 6 mean 1.6 3.2 4.0 4,3 s.d.m 1.2 1.0 1.0 0.7 5. Aeromonas 19 number of samples 3 4 4 3 mean 2.2 3.1 3.5 2.9 s.d.m 1.4 1.4 1.1 1.3 6. Aeromonas 19 + Lactobacillus number of samples 6 6 6 5 mean 2.5 3.9 4.6 4.2 s.d.m 0.8 0.9 0.8 1.0 s.d.m. = standard deviation of mean Lactobacillus samples. There was no statistical difference between Micro- coccus -+ Lactobacillus and Aeromonas x +Lactobacillus sausages, while Aero- monas 19 -f- Lactobacillus samples were significantly better (significance level 0.001) than the other ones. Table 6. Comparison of aroma of sausage pairs using the t-test. Scores for sausages of different ages are not separated. ++ + = samples different, significance level 0.05 Samples —>■ Control Micrococcus-]- Aeromonasx Aeromonas x + Aeromonas 19 I Lactobacillus Lactobacillus Micrococcus + ++ + Lactobacillus t Aeromonas x + + + Aeromonas x + + + + + + + Lactobacillus -i Aeromonas 19 + + +— ++ +- <- Aeromonas 19+ + + + ++ + + + + ++ + ++ + Lactobacillus -4— Flavour The flavour of Aeromonas 19 + Lactobacillus sausages was best after ripen- ing times of 7 and 14 days (Fig. 4 and Table 7). Sausages with Aeromonas x + Lactobacillus and Micrococcus + Lactobacillus were next in this order through- out ripening. Aeromonas 19 sausages were better than both the control and Aeromonas x sausages throughout the ripening period. The flavour of control and Aeromonas x sausages was the worst (Table 8). Aeromonas 19 samples were significantly better than these. However, the sau- sages inoculated with lactobacilli were best. Aeromonas x + Lactobacillus Figure 3. Effect of different bacterial inoculations on aroma of sausage. 136 Table 7. Effect of bacterial inoculations on flavour of experimental sausages (scores 0 —6). Sausage group 7 days 14 days 21 days 1. Control number of samples 17 18 18 mean 1.9 2.5 2.9 s.d.m 1.0 1.2 1.2 2. Micrococcus + Lactobacillus number of samples 14 14 13 mean 3.1 3.9 4.2 s.d.m 1.0 0.8 0.7 3. Aeromonas x number of samples 4 5 5 mean 1.9 2.6 2.7 s.d.m 1.6 1.1 0.9 4. Aeromonas x + Lactobacillus number of samples 6 6 6 mean 3.6 4.0 4.5 s.d.m 0.7 0.9 1.0 5. Aeromonas 19 number of samples 4 4 3 mean 2.5 3.7 3.3 s.d.m 1.1 1.1 1.0 6. Aeromonas 19 + Lactobacillus number of samples 6 6 5 mean 4.0 4.5 4.4 s.d.m 0.8 1.1 0.9 s.cl.m. = standard deviation of mean sausages were better at the 0.05 significance level and Aeromonas 19 + Lactobacillus sausages better at the 0,001 significance level than Micrococcus + Lactobacillus sausages. However, Aeromonas 19 + Lactoba- cillus samples were not significantly better than Aeromonas x + Lactoba- cillus samples. Figure 4. Effect of different bacterial inoculations on flavour of sausage. 137 138 Table 8. Comparison of flavour of sausage pairs using the t-test. Scores for sausages of different ages are not separated. ++ + = samples different, significance level 0.05 Samples —y Control Micrococcus 4- Acromonas x Aeronionas x + Aeromonas 19 Lactobacillus Lactobacillus Micrococcus + ++ + Lactobacillus t A cromonas x + + + A cromonas x + ++ + + ++ + Lactobacillus t t Aeromonas 19 + + + + + + ++ + + + <— +- Aeromonas 19+ + + + + + + + + + + + + Lactobacillus ■<— ■<— ■<— A— 3.4. 1. 2. Physical and chemical examinations 3. 4. 1. 2. 1. pH value a) Control group The pH value of control sausages showed the slowest drop. However, in these experiments the mean pH value of 7 day old sausages was 5.33 (Fig. 5 and Table 9). The pH value 5.30 or lower was attained in 54.4 % of 7 day old samples. During the next 2 weeks the pH decreased only slightly, the mean being 5.29 after 21 days of ripening. Fifty percent of the samples had a pH of 5.30 or lower. b) Micrococcus + Lactobacillus group The pH of Micrococcus + Lactobacillus sausages decreased so that its mean in 7 day old sausages was 5.20 (Fig. 5 and Table 9). 67.7 % of samples that were only 3 days old attained a pH of 5.30 or lower. The pH changes during the next 2 weeks were small. The mean pH of 3 week old sausages was 5.11 and the pH of 86.7 % of samples was 5.30 or lower. c) Aeromonas x group The pH of sausages inoculated with Aeromonas x strain was high throughout the ripening time, the mean always being over 5.30 (Table 9). Aeromonas x did not reduce the pH of dry sausage. d) Aeromonas x + Lactobacillus group The pH of the Aeromonas x + Lactobacillus sausage group decreased more quickly than that of the Micrococcus -+- Lactobacillus group, the mean being 5.22 when the sausages were 3 days old (Fig. 5 and Table 9). 67.7 % of the samples had a pH of 5.30 or lower. After 2 weeks the mean pH value was 4.97 and all samples had a pH value of 5.30 or lower e) Aeromonas 19 group The pH of Aeromonas 19 sausages was lower than that of sausages inoculated with Aeromonas x. After 10 days of ripening the mean pH value was less than 5.30, being 5.24 in 3 week old sausages (Table 9). Table 9. Effect of bacterial inoculation on pH value of experimental sausages. Sausage group 0 days 1 day 3 days 7 days 10 days 14 days 21 days 1. Control group number of samples 22 20 22 22 22 22 20 mean 5.70 5.70 5.58 5.33 5.30 5.25 5.29 s.d.m 0.11 0.12 0.18 0.17 0.15 0.16 0.14 % samples, pH == 5.30 0 0 9.1 54.5 46.5 72.2 50.0 2. Micrococcus + Lactobacillus number of samples 15 14 15 15 15 15 14 mean 5.69 5.66 5.28 5.20 5.10 5.06 5.11 s.d.m 0.69 0.33 0.23 0.27 0.16 0.15 0.14 % samples, pH == 5.30 0 0 67.7 73.3 86.7 86.7 86.7 3. Aeromonas x number of samples 4 4 4 4 4 4 4 mean 5.74 5.72 5.67 5.40 5.37 5.38 5.42 s.d.m 0.10 0.10 0.16 0.21 0.17 0.21 0.18 % samples, pH 5.30 0 0 0 50.0 50.0 50.0 25.0 4. Aeromonas x + Lactobacillus number of samples 6 6 6 6 6 6 6 mean 5.72 5.70 5.22 5.09 5.09 4.97 5.07 s.d.m 0.10 0.10 0.09 0.19 0.19 0.20 0.15 % samples, pH 5.30 0 0 67.7 83.3 83.3 100 100 5. Aeromonas 19 number of samples 4 4 4 4 4 4 3 mean 5.75 5.76 5.57 5.32 5.28 5.25 5.24 s.d.m 0.05 0.11 0.21 0.27 0.20 0.25 0.26 % samples, pH =5 5.30 0 0 0 50.0 50.0 75.0 75.0 6. Aeromonas 19 + Lactobacillus number of samples 6 6 6 6 6 6 5 mean 5.73 5.71 5.16 5.04 5.04 5.07 5.12 s.d.m 0.08 0.09 0.21 0.14 0.19 0.13 0.19 % samples, pH 5.30 0 0 83.3 100 100 100 100 s.d.m. = standard deviation of mean 139 f) Aeromonas 19 + Lactobacillus group The decrease in pH value of Aeromonas 19 -f- Lactobacillus saus- ages was quickest (Fig. 5 and Table 9). The mean pH value of 3 day old sausages was 5.16 and all the sausage samples attained pH 5.30 or lower after 7 days of ripening. The pH differences were small during the next 2 weeks. 3. 4. 1. 2. 2. Consistency a) Control group The consistency of control sausages developed slowly and was not firm enough. Some sausages reached a consistency value of 3 kg after 14 days and these comprised 53.8 % of the total sample (Fig. 6 and Table 10). The mean consistency value of 21 day old sausages was 3.83 kg. b) Micrococcus + Lactobacillus group The consistency of Micrococcus + Lactobacillus sausages was clearly firmer than that of control sausages. After 7 days 41.7 % of samples attained a con- sistency value of 3 kg or over (Fig. 6 and Table 10). After 3 weeks of ripening the mean was 5.08 kg. c) Aeromonas x group The consistency values of the control and Aeromonas x sausages were of the same order. However, after 7 days of ripening the consistency values of 33.3 %of Aeromonas x sausages were 3kg or over (Table 10). Later the respective percentages were higher than those of the control group. The mean of 21 day old sausages was 3.86 kg. d) Aeromonas x -f- Lactobacillus group After 3 days of ripening 50.0 % of samples had a consistency value of 3 kg or over (Fig. 6 and Table 10). So the consistency can be said to develop more quickly in the Aeromonas x + Lactobacillus group than in the Micrococcus -j- Lactobacillus group. The mean consistency values were also higher throughout ripening with the exception of 3 week old sausages, where the mean for Micro- coccms + Lactobacillus sausages was 0.24 kg higher, the mean for Aeromonas x + Lactobacillus sausages being 4.84. Figure 5. Effect of different bacterial inoculations on pH value of sausage. 140 1. Control 3. Aeromonas x 5. Aeromonas 19 Table 10. Effect of bacterial inoculations on consistency (kg) of experimental sausages. Sausage group 1 day 3 days 7 days 10 days 14 days 21 days number of samples 10 13 12 13 13 11 mean 0.87 1.40 2.37 2.53 3,32 3.83 s.d.m 0.14 0.65 0.88 0.65 3.21 0.93 % samples, 3 kg 0 0 0 0 53.8 90.9 2. Micrococcus + Lactobacillus number of samples 11 12 12 13 13 12 mean 1.07 2.09 3.33 3.58 4.51 5.08 s.d.m 0.39 1.11 1.36 1.16 1.34 1.41 % samples, 3 kg 0 0 41.7 61.7 92.3 100 number of samples 3 3 3 3 3 3 mean 1.45 1.68 2.57 2.55 3.82 3.86 s.d.m 0.45 0.72 0.42 0.73 0.43 1.36 % samples, 3 kg 0 0 33.3 66.7 66.7 100 4. Aeromonas x + Lactobacillus number of samples 3 4 4 4 4 4 mean 1.35 2.61 4.17 3.98 5.27 4.84 s.d.m 0.45 1.86 2.68 1.12 1.76 1.40 % samples, == 3 kg 0 50.0 75.0 75.0 100 100 number of samples 4 4 4 4 4 3 mean 0.88 1.06 1.62 2.18 2.93 3.83 s.d.m 0.11 0.11 0.51 0.76 0.60 0.74 % samples, 3 kg 0 0 0 0 75.0 100 6. Aeromonas 19 + Lactobacillus number of samples 6 6 6 6 6 5 mean 0.86 1.86 2.91 3.27 3.78 4.74 s.d.m 0.13 0.33 0.73 1.05 1.00 1.38 % samples, 3 kg 0 0 16.7 66.7 83.3 100 s.d.m. = standard deviation of mean e) Aeromonas 19 group The consistency values stayed under 3 kg during the first 10 days of ripening, until after 2 weeks 75.0 % of samples had values over 3 kg. The sausages of this group were softest throughout ripening up to 3 weeks when the mean consistency value was the same as that of the sausages of the control group - 3.83 kg (Table 10). f) Aeromonas 19 + Lactobacillus group In Aeromonas 19 -f- Lactobacillus sausages the consistency developed surp- risingly slowly especially at the beginning of ripening. After 7 days only 16.7 % 141 142 of sausages had attained the consist- ency value of 3 kg or over (Fig. 6 and Table 10). The corresponding percentage in Micrococcus + Lacto- bacillus sausages was 41.7 and in Aeromonas x + Lactobacillus saus- ages 75.0. After 21 days the mean consistency value of Aeromonas 19 + Lactobacillus sausages was 4.74 being 0.34 kg lower than that of Micro- coccus + Lactobacillus sausages. 3. 4. 1. 2. 3. Weight losses The mean weight losses for each sausage group are presented in table 11. The table shows that the weight losses for each group developed similarly. The ranges of the means for all sausage groups are as follows: Table 11. Percentile weight losses of experimental sansages. Sausage group 1 day 3 days 7 days 10 days 14 days 21 days 1. Control number of samples 12 15 13 16 17 14 mean 2.6 7.1 13.9 17.1 19.0 24.6 s.d.m 1.0 1.8 2.3 3.1 4.4 3.6 2. Micrococcus + Lactobacillus number of samples 10 16 13 16 15 12 mean 2.7 8.4 14.1 17.8 20.6 25.4 s.d.m 1.0 2.1 1.4 2.1 2.7 2.8 3. A eromonas x number of samples 4 5 5 5 4 5 mean 2.5 7.8 15.3 18.0 20.6 26.0 s.d.m 1.0 2.1 2.3 3.2 4.9 2.8 4. A eromonas x + Lactobacillus number of samples 5 6 6 6 5 5 mean 3.3 8.3 15.2 18.9 20.5 25.3 s.d.m 1.4 2.3 2.5 2.4 1.7 6.1 5. Aeromonas 19 number of samples 3 4 4 3 3 3 mean 3.0 7.8 13.5 17.2 20.5 24.9 s.d.m 0.6 1.9 1.5 1.0 2.6 1.0 6. Aeromonas 19 -+■ Lactobacillus number of samples 5 6 6 6 5 5 mean 3.3 8.8 14.5 18.1 22.4 26.7 s.d.m 1.5 2.2 1.6 2.0 1.6 0.8 s.d.m. = standard deviation of mean Figure 6. Elfect of different bacterial inoculations on consistency (determined by Instron apparatus) of sausage. 1 day 3 days 7 days 10 days 14 days 21 days 2.5-3.3% 7.1-8.8 % 13.5-15.3% 17.1-18.9% 19.0-22,4% 24.6-26.7 % When the control sausages ripen for 3 weeks and Micrococcus + Lacto- bacillus, Aeromonas x + Lactobacillus and Aeromonas 19 + Lactobacillus sau- sages ripen for 10 days, the weight losses during ripening can, in the last 3 cases, be considerably reduced. The following mean weight losses for ripened sausages can be taken from table 11 control group 24.6 % Micrococcus + Lactobacillus group 17.8 % Aeromonas x group 26.0 % Aeromonas x + Lactobacillus group 18.9 % Aeromonas 19 group 24.9 % Aeromonas 19 + Lactobacillus group 18.1 % It was sometimes found that in the control, Aeromonas x and Aeromonas 19 groups the sausages did not always ripen within 3 weeks. 3. 4. 1.2. 4. Nitrite The nitrite content of experimental sausages rose during the first 3 days of ripening so that the percentages of samples containing 30 ppm or more of sodium nitrite for the different sausage groups were as follows: control group 29.6 % Micrococcus + Lactobacillus group 66.7 % Aeromonas x group 80.0 % Aeromonas x -f Lactobacillus group 66.7 % Aeromonas 19 25.0 % Aeromonas 19 + Lactobacillus 0 % (Table 12) The highest mean nitrite values after 3 days of ripening were found in sausages inoculated with Aeromonas x (56.5 ppm), Aeromonas x + Lacto- bacillus (37.2 ppm) and Micrococcus + Lactobacillus (34.1 ppm) (Table 12). Even the control sausages contained 25.6 ppm sodium nitrite while Aeromonas 19 sausages contained only 12.0 ppm and Aeromonas 19 + Lactobacillus sausages only 9.4 ppm. After 3 days of ripening nitrite concentrations decreased so that 7 day old sausages contained about 10 ppm sodium nitrite. During the next 2 weeks there was a small decrease in nitrite concentrat- ion in all sausage groups (Fig. 7 and Table 12). The results can be inter- preted as showing that Aeromonas x and Micrococcus are effective nit- rate reducers. On the other hand the content of nitrate is also lower in Aeromonas 19 + Lactobacillus sausages (p. 145). Consequently Aeromonas 19 could be regarded as a stronger nitrate reducer. Figure 7. Effect of different bacterial inoculations on sodium nitrite content of sausage. 143 144 Table 12. Effect of bacterial inoculations on sodium nitrite content (ppm) of experimental sausages. Sausage group 0 days 1 day 3 days 7 days 10 days 14 days 21 days 1. Control number of samples 20 19 21 21 22 22 20 mean 1.0 1.4 25.6 13.3 8.0 6.4 6.4 s.d.m 0.7 1.3 25.0 9.4 3.5 3.7 3.0 % samples, => 30 ppm 0 0 29.6 0 0 0 0 2. Micrococcus + Lactobacillus number of samples 15 14 15 15 15 15 14 mean 1.4 12.5 34.1 10.4 8.6 7.0 6.9 s.d.m 1.2 20.8 18.0 5.4 3,7 2.9 3.6 % samples, 30 ppm 0 14.4 66.7 0 0 0 0 3. Aeromonas x number of samples 5 5 5 5 5 5 5 mean 0.8 2.3 56.5 13.3 12.8 9.6 8.7 s.d.m 0.6 2.8 28.3 9.3 7.5 4.7 3.3 % samples, 30 ppm 0 0 80.0 0 0 0 0 4. Aeromonas x + Lactobacillus number of samples 6 6 6 6 6 6 6 mean 0.8 1.8 37.2 10.1 7.5 8.2 8.1 s.d.m 0.8 1.7 24.6 3.6 4.0 4.3 2.4 % samples, => 30 ppm 0 0 66.7 0 0 0 0 5. Aeromonas 19 number of samples 4 4 4 4 4 4 4 mean 1.2 8.0 12.0 11.1 5.5 7.6 6.5 s.d.m 0.5 7.0 12.6 11.3 1.4 3.9 5,1 % samples, 30 ppm 0 0 25.0 25.0 0 0 0 6. Aeromonas 19 + Lactobacillus number of samples 6 5 6 6 6 6 5 mean 1.3 3.6 9.4 6.8 6.5 7.1 6.4 s.d.m 0.5 4.6 7.5 1.2 1.7 2.4 2.7 % samples, 30 ppm 0 20.0 0 0 0 0 0 s.d.m. = standard deviation of mean 3. 4. 1. 2. 5. Nitrate The potassium nitrate content decreased most markedly in sausages inoc- ulated with lactobacilli. After one day the means were as follows: Micrococcus 4- Lactobacillus group 119.0 ppm Aeromonas x + Lactobacillus group 100.3 ppm Aeromonas 19 + Lactobacillus group 122.4 ppm (Table 13). The mean potassium nitrate contents of other samples after one day of ripening were as follows: control group 176.9 ppm Aeromonas x group 176.6 ppm Aeromonas 19 group 159.0 ppm (Table 13). 1. Control 3. A eromonas x 5. Aeromonas 19 Table 13. Effect of bacterial inoculations on potassium nitrate content (ppm) of experimental sausages Sausage group 1 day 3 days 7 days 10 days 14 days 21 days number of samples 13 11 13 11 14 12 179.7 82.8 97.8 64.7 53.5mean 176.9 s.d.m 55.3 62.0 43.8 54.8 41.7 23.6 2. Micrococcus 4- Lactobacillus number of samples 11 12 12 13 12 9 106.0 86.6 54,9 42.8 39.1mean 119.0 s.d.m 39.0 32.4 43.3 37.7 22.9 21.5 number of samples 5 5 5 6 4 6 mean 176.6 107.8 94.0 90.8 87,0 81.6 65.5 73.0 54.4 29.1 51.9s.d.m 65.0 4. Aeromonas x 4- Lactobacillus number of samples 6 7 7 7 6 6 mean 100.3 115.1 73.7 71.0 36.7 46.5 35.8 18.9 26.5 13.5 30.0s.d.m 33.1 number of samples 3 3 3 3 3 3 99.7 55.3 57.7 44.7 55.3mean 159.0 s.d.m 18.7 70.5 38.3 33.7 5.5 45.2 6. Aeromonas 19 4- Lactobacillus number of samples 5 mean 122.4 s.d.m 27.1 s.d.m. = standard deviation of mean 5 6 6 5 5 72.4 68.2 49.0 50.6 33.2 26.850.5 37.033.0 24.9 The potassium nitrate concen- tration decreased throughout the time of ripening in sausages inocul- ated with lactobacilli and some other bacteria (Fig. 8). After 3 weeks of ripening the lowest mean values were in these sausages the values being as follows: Micrococcus -f Lactobacillus group 39.1 ppm Aeromonas x A- Lactobacillus group 46.5 ppm Aeromonas 19 + Lactobacillus group 33.2 ppm (Table 13). Figure 8. Effect of different bacterial inoculations on potassium nitrate content of sausage. 145 146 The mean potassium nitrate contents in other sausages after 3 weeks of ripening were as follows: control group 1 53.5 ppm Aeromonas x group 81.6 ppm Aeromonas 19 group 55.3 ppm (Table 13). 3. 4. 1.3. Microbiological examinations 3. 4. 1. 3. 1. Total bacteria The results of total bacterial counts are presented in Fig. 9 and Table 14. On the first day of ripening there was little or no increase in numbers of bacteria in any of the sausage groups. During the next 2 days the essential growth of bacterial counts took place in all groups. After that the total bacterial counts stayed at the same level or changed only slightly. The highest numbers of bacteria were in the sausages of Aeromonas x + Lactobacillus and Aeromonas 19 Table 14. Number of bacteria x 106/g on plate count agar in different sausage groups. Sausage group 0 days 1 day 3 days 7 days 10 days 14 days 21 days 1, Control number of samples 15 14 15 14 15 15 14 mean 3.2 2.7 82.6 86.4 70.5 59.9 66.9 s.d.m 4.6 3.8 84,0 69.9 41.2 46,2 40.3 2. Micrococcus + Lactobacillus number of samples 10 13 13 14 14 13 12 mean 16.4 23.1 135.1 143.6 92.5 77.4 77.7 s.d.m 8.3 15.4 107.0 145.2 71.6 52.3 70.0 3. Aeromonas x number of samples 4 4 4 4 4 4 4 mean 17.3 17.3 38.8 92.5 65.3 71.4 65.7 s.d.m 10.9 10.2 54.3 80.2 52.4 68.1 25.5 4. Aeromonas x + Lactobacillus number of samples 6 6 6 6 6 6 6 mean 16.7 12.7 356.7 208.0 195.3 171.5 211.8 s.d.m 8.9 8.7 215.1 81.3 99.0 77.6 86.6 5. Aeromonas 19 number of samples 4 4 3 4 3 4 3 mean 12.0 12.5 49.0 42.8 39.7 69.8 213.3 s.d.m 4.6 3.4 5.6 12.7 15.6 65.2 161.7 6. Aeromonas 19 + Lactobacillus number of samples 6 6 6 6 6 6 5 mean 15.0 36.0 350.0 360.0 391.0 264.8 698.0 s.d.m 7.639.8 284.7152.2 352.6170.8 436.0 s.d.m. =standard deviation of mean 1. Control 3. A eromonas x 5. Aeromonas 19 -f- Lactobacillus groups, with mean values well over 108/g. In the Micro- coccus + Lactobacillus group the mean bacterial count increased to a little over 108/g during the first 3 days of ripening and decreased to less than 108/g after 7 days of ripening. The control sausages contained slightly fewer bacteria than the sausages inoculated with micrococci and lactobacilli. The bacterial counts of Aeromonas x and Aeromonas 19 sausages remained between 107/g and 2 X 108/g throughout ripening. The reason is that they did not contain inoculated lactobacilli. 3. 4. 1. 3. 2. Micrococci Mannitol salt agar was used to determine, in particular, the number of inoculated micrococci. Staphylococci also grow on this medium, and this must be borne in mind when checking the results. Table 15. Number of micrococci X 104/g on mannitol salt agar in different sausage groups. Sausage group 0 days 1 day 3 days 7 days 10 days 14 days 21 days number of samples 20 19 17 14 19 15 17 mean 7.5 6.0 14.9 33.2 10.0 5.1 8.7 s.d.m 11.9 6.6 13.7 33.6 11.5 6.9 11.3 2. Micrococcus + Lactobacillus number of samples 15 13 14 15 15 14 14 mean 870 1267 1607 1240 1547 1593 1293 s.d.m 566 849 1440 771 1164 1408 975 number of samples 5 5 5 5 5 5 5 mean 1.5 2.9 1.9 3.4 4.3 4.2 4.5 s.d.m 1.2 2.1 1.1 3.3 4.7 2.3 4.5 4. Aeromonas x + Lactobacillus number of samples 6 6 6 6 6 6 6 mean 3.1 3.3 3.3 2.1 1.4 0.9 0.3 s.d.m 3.0 3.2 3.0 2.4 1.7 0.9 0.2 number of samples 4 4 4 4 4 4 4 mean 7,5 23.3 10.6 23.3 1.9 3.6 4.8 s.d.m 10.4 24.5 8.8 24.4 1.4 2.6 4.7 6. Aeromonas 19 + Lactobacillus number of samples 6 6 6 6 6 6 4 mean 7.3 4.4 4.1 0.8 2.1 2.7 0.6 s.d.m 9.0 3.4 4.9 0.7 2.2 2.5 0.8 s.d.m. = standard deviation of mean Figure 9. Number of bacteria/g on plate count agar in different sausage groups. 147 148 The number of micrococci in control sausages was about 10 6/g throughout ripening (Fig. 10). The sausages inoculated with micrococci and lactobacilli contained the most micrococci. Throughout ripening the mean count was a little over 10’/g. The number of micrococci in Aero- monas x + Lactobacillus and Aero- monas 19 + Lactobacillus sausages decreased from about 5 X 104/g to about 5 X 103/g during ripening (Fig. 10 and Table 15). The mean micrococcus count of Aeromonas x sausages stayed between 104 /g and 5 X 104/g while that of Aeromonas 19 sausages exceeded 105/g after one day and stayed there about one week before falling back to between and 106/g. 3. 4. 1. 3. 3. Total bacteria on tributyrine agar The total number of bacteria in control sausages remained lower on tri- butyrine agar than on plate count agar, the mean lying between 106/g and 107/g throughout ripening (Figs. 9, 11 and Tables 14, 16). The total bacterial count Table 16. Number of bacteria X 106 /g on tributyrine agar in different sausage groups. Sausage group 0 days 1 day 3 days 7 days 10 days 14 days 21 days 1. Control number of samples 19 16 21 15 22 21 21 mean 1.4 1.3 7.1 10.0 7.1 4.9 6.2 s.d.m 1.2 1.5 8.9 9.1 4.7 6.8 7.2 2. Micrococcus + Lactobacillus number of samples 15 14 15 13 14 13 14 mean 8.1 11.2 14.5 11.1 6.6 7.0 6.0 s.d.m 5.2 9.1 13.3 9.0 3.5 4.3 4.7 3. Aeromonas x number of samples 5 5 5 5 5 5 5 mean 12.5 43.8 45.4 38.2 59.2 39.8 35.6 s.d.m 8.7 36.3 27.9 27.9 55.2 27.7 24.3 4. Aeromonas x -j- Lactobacillus number of samples 6 6 6 6 6 5 4 mean 20.6 33.8 57.0 10.6 7.1 4.0 2.1 s.d.m 17.1 13.0 50.0 3.7 4.8 3.0 3.3 5. Aeromonas 19 number of samples 4 4 3 4 4 4 3 mean 22.1 28.0 10.8 35.5 45.8 49.1 25.0 s.d.m 13.8 6.5 8.0 28.8 46.0 52.2 15.0 6. Aeromonas 19 + Lactobacillus number of samples 6 5 6 6 6 6 5 mean 32.0 23.0 49.2 13.3 132.1 78.6 20.2 s.d.m 32.011.8 36.912.7 151.787.2 19.3 s.d.m. = standard deviation of mean Figure 10. Number of micrococci/g on mannitol salt agar in different sausage groups. of Micrococcus -)- Lactobacillus sausages on tributyrine agar reflects most closely the number of micrococci. The mean count first rose to a little over 107/g and then fell to less than 107 /g after one week. On mannitol salt agar the mean count remained in excess of 107/g. The mean bacterial count of Aeromonas x + Lactobacillus sausages increased from about 2 X 10 7/g to about 5 X 10 7/g during the first 3 days of ripening (Fig. 11 and Table 16). Thereafter the count fell to less than 107 /g and after 3 weeks of ripening was around 10 6/g. The total number of bacteria in Aeromonas 19 + Lactobacillus sausages ranged be- tween 10 7/g and 108/g during ripen- ing. The numbers of bacteria shown greatly resemble the lipolytic bac- terial count (= inoculated Aero- monas bacteria) because lactobacilli do not grow on tributyrine agar (Fig. 11 and Table 16). The mean quantity of bacteria in the Aeromonas x and Aeromonas 19 sausage groups varied between 10 7/g and 6 X 10 7/g during ripening. 3. 4. 1. 3. 4. Lipolytic bacteria The sausage was cultivated on tributyrine agar to determine the number of inoculated Aeromonas strains. In the sausages inoculated with Aeromonas x -\- Lactobacillus and Aeromonas 19 -f- Lactobacillus the mean lipolytic bacterial counts decreased from about 2 x 107/g to about 2.5 X 10 6/g during ripening (Fig. 12 and Table 17). The numbers of strain x were a little smaller than those of strain 19. The mean quantity of lipolytic bacteria in sausages inoculated with Aeromonas x and Aeromonas 19 � � Control group □ a Micmoccus + Laciobacmus groap varied between 107/g and 5 X 107/g log N O -O Aeromonas x + Lactobacillus group . • # Aeromonas 19 + Lactobacillus group ( A 3.D10 17j . v The mean lipolytic bacterial 7 counts in the control sausages rose °--i5 • from about 105/g to about 10 6/g 6 - y* —* —* -S during the first 3 days of ripening y* and then remained at a level of 106/g (Fig- 12 and Table 17). The D — —□ number of lipolytic bacteria varied □ 4 T—i 1 1 1 1 r between 106 /g and 104/g in Micro- -1,1 coccus + Lactobacillus sausages Figure 12. Number of lipolytic bacteria/g on during ripening, tributyrine agar in different sausage groups. Figure 11. Number of bacteria/g on tri- butyrine agar in different sausage groups. 149 Table 17. Number of lipolytic bacteria x 106 /g on tributyrine agar in different sausage groups. Sausage group 0 days 1 day 3 days 7 days 10 days 14 days 21 days 1. Control number of samples 19 17 19 19 19 19 19 mean 0.2 0.4 1.3 1.0 0.8 1.5 1.0 s.d.m 0.3 1.2 2.0 2.4 1.4 2.5 2.4 2. Micrococcus + Lactobacillus number of samples 12 13 14 11 13 11 14 mean 0.07 0.09 0.1 0.04 0.03 0.03 0.02 s.d.m 0.1 0.2 0.2 0.06 0.07 0.04 0.03 3. Aeromonas x number of samples 5 5 5 5 5 5 4 mean 15.0 26.2 42.2 32.6 17.6 16.2 4.7 s.d.m 14.3 19.9 27.0 26.7 11.4 14.7 7.0 4. Aeromonas x + Lactobacillus number of samples 6 6 6 6 6 6 6 mean 14.8 25.8 30.2 11.2 3.8 2.4 1.2 s.d.m 11.2 17.8 27.9 5.0 2.5 2.2 2.2 5. Aeromonas 19 number of samples 4 4 3 4 4 4 3 mean 22.0 28.0 41.0 35.5 35.3 9.1 25.5 s.d.m 13.9 6.5 15.0 28.8 35.5 6.8 30.4 6. Aeromonas 19 + Lactobacillus number of samples 6 5 6 6 6 6 5 mean 32.0 23.0 49.2 18.5 9.4 3.0 3.3 s.d.m 31.4 11.8 36.9 16.8 7.1 3.7 4.0 s.d.m. = standard deviation of mean 3. 4. 1. 3. 5. Lactobacilli The mean lactobacillus counts of the control sausages rose steeply during the first 3 days of ripening from 5 X 103/g to 4X 107/g and ranged between 107/gandl0 8/gfortherest of the ripen- ing time (Fig. 13 and Table 18). The mean quantity of lactohacilli in sau- sages inoculated with lactohacilli rose from B—lo X 106/g to over 108/g in 3 days. Thereafter the mean lactoba- cillus counts remained at more than 8x 107 /g for the rest of the ripening time so that the mean lactobacillus level in Micrococcus + Lactobacillus sausages was less than that in Aero- monas x + Lactobacillus and Aeromo- nas 19 + Lactobacillus sausages, the mean lactobacillus counts after 14 days of ripening being as follows: Figure 13. Number of lactobacilli/g on Ro- gosa agar in different sausage groups. 150 1. Control 3. Aeromonas x 5. Aeromonas 19 Table 18. Number of lactobacilli x 106/g on Rogosa agar in different sausage groups. Sausage group 0 days 1 day 3 days 7 days 10 days 14 days 21 days number ot samples 18 18 21 20 20 17 20 mean 0.005 0.02 43.8 94.6 34.1 56.1 40.1 s.d.m 0.006 0.02 52.5 90.5 24.1 55.6 44.8 2. Micrococcus + Lactobacillus number of samples 13 12 15 15 15 12 12 mean 9.2 17.8 147.3 112.2 103.9 102.0 77.2 s.d.m 7.2 14.4 143.2 116.8 109.4 101.1 78.3 number of samples 5 5 5 5 5 5 5 mean 0.0031 0.014 10.0 44.6 41.4 41.8 41.6 s.d.m 0.0053 0.022 8.1 30.8 22.5 34.3 24.5 4. Aeromonas x + Lactobacillus number of samples 6 6 6 6 6 6 6 mean 8.0 14.7 300.0 162.7 247.3 320.0 187.5 s.d.m 8.2 9.7 248,9 75.7 162.5 234.8 95.0 number of samples 4 4 3 4 4 4 4 mean 0.0073 0.037 9.7 41.0 78.0 65.5 102.7 s.d.m 0.0083 0.043 7.2 36.4 50.9 69.9 64.7 6. Aeromonas 19 + Lactobacillus number of samples 6 6 6 6 6 6 5 mean 10.0 31.2 311.7 345.0 423.0 240.7 456.0 s.d.m 3.9 27.6 190.8 190.0 300.0 144.9 218.0 s.d.m. = standard deviation of mean Micrococcus + Lactobacillus group 1.0 X 108/g Aeromonas x + Lactobacillus group 3.2 X 108/g Aeromonas 19 -f- Lactobacillus group 2.4 X 108/g The number of lactobacilli in Aeromonas x and Aerotnonas 19 sausages rose from 103 —lO4/g to a mean value of 107/g during the first 3 days (Table 18). During the rest of the ripening the mean lactobacillus counts varied between 4 X 107/g and 108/g. 3. 4. 1. 3. 6. Streptococci The mean streptococcus counts of control sausages rose from about 103/g to 5 X 105/g during the first 3 days of ripening (Fig. 14 and Table 19). There- after the number of streptococci rose further attaining a mean value ot 2 X 106/g after 10 days of ripening. After that, the value remained in excess of 10*/g. The mean quantity of streptococci in sausages inoculated with lactobacilli and some other bacteria rose from 3 X 102 —2 X 103/g to values between 10!/g 151 Table 19. Number of streptococci x 104/g on Slanetz agar in different sausage groups. Sausage group 0 days 1 day 3 days 7 days 10 days 14 days 21 days 1. Control number of samples 21 18 19 19 20 21 20 mean 0.091 0.21 45.0 84.7 201.5 166.0 107.0 s.d.m 0.100 0.18 35.0 93.0 311.0 235.0 140.0 2. Micrococcus + Lactobacillus number of samples 12 13 13 13 13 15 14 mean 0.05 0.13 2.3 3.1 1.9 3.4 2.9 s.d.m 0.04 0.15 3.1 5.8 3.0 5.7 6.7 3. Aeromonas x number of samples 5 5 5 5 5 4 5 mean 0.30 0.21 20.2 133.0 84.6 53.5 103.3 s.d.m 0.46 0.23 14.7 161.8 79.9 71.3 90.6 4. Aeromonas x + Lactobacillus number of samples 6 6 6 6 6 6 4 mean 0.20 0.11 4.6 4.4 5.7 6.1 2.6 s.d.m 0.35 0.052 5.8 4.7 7.7 8.6 3.4 5. Aeromonas 19 number of samples 3 4 3 4 3 4 3 mean 0.038 0.70 58.8 90.0 333.3 132.7 217.3 s.d.m 0.025 0.89 37.1 132.1 305.5 171.3 300.3 6. Aeromonas 19 + Lactobacillus number of samples 6 6 5 6 6 6 5 mean 0.027 0.42 8.2 14.4 13.4 7.4 23.4 s.d.m 0.023 0.37 7.4 20.9 21.6 8.0 33.7 s.d.m. = standard deviation of mean and 105/g over 3 days (Fig. 14 and Table 19). Thereafter the strep- tococcus counts remained at a level of about 104/g to 105/g so that Aero- monas 19 + Lactobacillus samples contained most and Micrococcus + Lactobacillus sausages least strepto- cocci. The mean streptococcus counts in Aeromonas x and Aeromonas 19 sausages resembled the values for control sausages, exceeding 106/g after 7 and 10 days of ripening and the latter remaining in excess of this figure for the rest of the ripening time (Table 19). Figure 14. Number of streptococci/g on Slanetz agar in different sausage groups. 152 1. Control 3. Aeromonas x 5. Aeromonas 19 Table 20. Number of coliform bacteria x 102/g on VRB agar in different sausage groups. Sausage group 0 days 1 day 3 days 7 days 10 days 14 days 21 days number of samples 17 18 15 13 16 13 18 mean 7.1 12.6 14.7 15.7 14.4 15.5 21.7 s.d.m 10.8 12.9 25.2 18.5 23.5 20.3 29.3 2. Micrococcus -f- Lactobacillus number of samples 12 13 12 12 13 13 12 mean 16.3 21.5 12.9 11.3 4.6 4.0 3.2 s.d.m 15.8 24.9 15.2 21.7 5.9 8.8 6.8 number of samples 6 6 6 6 6 6 6 mean 2.1 3.6 1.0 1.3 0.1 0.8 1.8 s.d.m 3.1 4.3 1.4 1.1 0.2 0.8 1.8 4. Aeromonas x + Lactobacillus number of samples 6 6 6 6 6 6 6 mean 1.6 1.2 2.0 0.9 1.0 0.6 0.7 s.d.m 1.3 1.5 2.6 1.2 0.6 0.7 0.6 number of samples 4 4 4 4 4 4 4 mean 0.6 9.3 15.6 5.3 3.8 4.4 4.7 s.d.m 0.5 15.3 18.7 5.1 5.5 4.0 6.4 6. Aeromonas 19 + Lactobacillus number of samples 6 6 6 6 6 5 4 mean 1.0 3.6 6.3 5.3 3.8 4.1 4.2 s.d.m 1.2 3.4 6.8 4.9 5.1 3.9 6.2 s.d.ra. = standard deviation of mean 3. 4. 1. 3. 7. Coliform bacteria The mean quantity of coliform bacteria in control sausages rose from 7 X 102/g to a little over 103/g and stayed at this level for the rest of the ripening time (Fig. 15 and Table 20). The mean level of coliform bacteria in Micrococcus + Lactobacillus sausages, on the other hand, decreased from 2.0 X 103 /g to less than 103/g after 10 days of ripening and remained at less than 103/g for the rest of the ripening time. The mean quantity of coliform bacteria in Aeromonas x + Lactobacillus sausages decreased from just over 102/g to less than 102/g after 10 days of ripening (Fig. 15 and Table 20). The corresponding number of bacteria in Aeromonas 19 + Lactobacillus sau- sages ranged between 102/g and 103/g. The mean quantity of coliform bac- teria in Aeromonas x and Aeromonas 19sausages ranged between 102/g and 103/g throughout ripening (Table 20). The highest values appeared after one day or 3 days of ripening. Figure 15. Number of coliform bacteria/g on VRB agar in different sausage groups. 153 154 3.4.2. USE OF VIBRIO COSTICOLUS AS A STARTER CULTURE The effect of Vibrio costicolus on the quality of dry sausage was studied. The strain was isolated from curing brine and improves the quality of dried hams (Petäjä 1972, Petäjä et ai. 1973). V. costicolus did not thrive in dry sausage. The amount of V. costicolus fell noticeably during the first 3 days of ripening from 10 7/g to about 103/g. In 7 day old sausages no V. costicolus was found. As ripening began V. costicolus reduced nitrate, producing 20 —3O ppm nitrite in the first 3 days of ripening thereby securing colour formation. This nitrate reduction did not happen regularly in every experimental series. When lactobacilli were inoculated along with V. costicolus the colour formation was missing because vibrios dis- appeared and the lactobacilli therefore produced discolourations. The consis- tency of these sausages was firm. On the basis of these observations, especially that of the disappearance of V. costicolus from dry sausage, it can be concluded that V. costicolus cannot be used as a starter culture for dry sausage. 3. 4. 3. USE OFACHROMOBACTER STRAINS AS A STARTER CULTURE The effect of Achromobacter 22, Achromobacter guttatus and Achromobacter X strains on the quality of dry sausage was studied. The first strain was isolated from curing brine (Petäjä et ai. 1973), strain X from a control dry sausage in this research and A. guttatus had been received from Nora Winterhalter of Eidgenossische Technische Hochschule, Zurich, Switzerland. Achromobacter 22 failed to thrive in dry sausage. It disappeared completely after one week. Clearly then, it did not have any marked effect on the quality of dry sausage. The sausages inoculated with it alone were in the same class as control sausages and those inoculated with Achromobacter 22 and lacto- bacilli had better consistency than the control. Two experimental series were carried out. The Achromobacter guttatus strain also failed to grow in dry sausage and disappeared even more quickly than Achromobacter 22. It had no effect on the quality of dry sausage. Only one series of experiments was carried out. Dry sausages were inoculated with Achromobacter X alone and with lacto- bacilli in three experimental series. In one series the inoculum was successful. Achromobacter X thrived in dry sausage, the number of cells being about 2 X 10 7/g after 7 days of ripening after which it fell. The effect of the Achromo- bacter strain used when inoculated with lactobacilli was the same as that of micrococci: the colour was red, consistency firm and aroma and flavour good. When inoculated with just Achromobacter X the colour of the sausages was good but the consistency, aroma and flavour were in the same class as control sausages. In two other experimental series Achromobacter X not only failed to thrive but disappeared during the first 3 days of ripening. Neither did this strain survive very well after lyophilization. It can be concluded then, that none of the Achromobacter strains used is suitable for use as a starter culture in dry sausage. 155 3. 4. 4. USE OF ESCHERICHIA COLI AS A STARTER CULTURE Escherichia coli was selected as a standard enterobacter for an experimental starter culture. Three experimental series contained sausages inoculated with E. coli alone and with E. coli and lactobacilli. The sausages inoculated with E. coli alone were in the same quality class as control sausages. However, the quality of E. coli -f- Lactobacillus sausages was clearly better though not as good as that of either Aeromonas x or 19 -f Lactobacillus or Micrococcus + Lactobacillus sausages. The colour was not as clear, consistency not as firm and aroma and flavour not as palatable as in these latter samples. The pH value of sausages inoculated with E. coli alone was quite high (5.40). The pH values of E. coli + Lactobacillus sausages were between 5.0 and 5.10. E. coli reduced nitrate quite strongly, producing a mean nitrite content after 3 days of ripening of 48 ppm in E. coli sausages and 44 ppm in E. coli -f- Lacto- bacillus sausages. The number of E. coli in sausages inoculated with E. coli alone and E. coli + Lactobacillus decreased during ripening from 106 —lO7/g to 105 -10*/g. 3.4.5. USE OF PROTEUS VULGARIS AS A STARTER CULTURE Proteus vulgaris was inoculated alone and with lactobacilli into dry sausage in three experimental series to investigate how this kind of strongly proteolytic and spoiling bacteria grows in dry sausage and affects its properties. When dry sausages were inoculated with P. vulgaris alone, the bacteria had an ad- verse effect on the quality of the sausages. The consistency did not develop. Instead, colour was produced, which is natural since P. vulgaris is a strong nitrate reducer. As a rule, however, the colour faded in the later stages cf ripening. The aroma and flavour of P. vulgaris sausages were unpalatable. When P. vulgaris was inoculated together with lactobacilli into dry sausage the consistency of the sausage was only slightly better than without lactobacilli. The colour did not develop in P. vulgaris + Lactobacillus sausages and the aroma and flavour were unsatisfactory. The pH value of P. vulgaris sausages decreased, as in the control sausages. The pH value of P. vulgaris + Lactobacillus sausages decreased, as in the sau- sages inoculated with micrococci and labtobacilli as a rule. However, in an experimental series of sausages containing more than 10 7 Proteus cells/g the pH value remained above 5,35 throughout ripening. The number of Proteus cells changed considerably. In one experimental series there were more than 107cells/g throughout ripening while in two other series the number decreased from a level of 3 X 10 s/g and P. vulgaris bacteria practically disappeared after 3—7 days of ripening. 156 4. Discussion 4. 1. Effect of bacterial inoculations on the ripening of dry sausage It is accepted that the use of bacterial cultures improves the quality of dry sausage. Since the Pediococcus cerevisiae starter culture developed by Niven et al. (1955) and the Micrococcus starter culture developed by Niinivaara (1955), many new starter cultures have been developed: different Lactacel and Saga products in the USA, French and Spanish starter cultures, and starters in Bulgaria, the Soviet Union, Yugoslavia, the GDR etc. Perhaps the most important of these is the mixed culture of micrococci and lactobacilli developed by Nurmi (1966). Also, according to Liepe (1971, 1975, 1976) the use of lacto- bacilli and micrococci is necessary to assure the controlled ripening of dry sausage. In Nurmi’s method lactobacilli produce the lactic acid necessary for the development of the consistency and aroma of dry sausage. Micrococci reduce nitrate to nitrite and avoid the discolourations which lactobacilli would produce. This method also considerably reduces the processing time. This mixed starter culture of micrococci and lactobacilli is used in Europe in particular, but in other countries, too. In addition to »Duploferment 66» developed on the basis of the studies of Nurmi, there are some other commercial products which contain micrococci and lactobacilli (Saga III). The Micrococcus -f- Lactobacillus starter culture is also the reference with which other experimental sausages are compared in this investigation. Like the bacteria of this starter culture, all cultures used in dry sausage have in practice been gram-positive. The effect of gram-negative bacteria on the aroma and flavour of dry sausage has been studied most notably by Kohnle (1953), Keller (1954), Keller and Meyer (1954), Meyer (1954), Losem (1956) and Eckert (1958). The purpose of this examination was to investigate the possible use of gram-negative bacteria in dry sausage. The effect of the following gram-negative bacteria on the quality of dry sausage was studied by inoculating the bacteria alone and with Lactobacillus planlarum (Nurmi 1965, 1966): two Aeromonas strains, Vibrio costicolus, three Achromobacter strains, Escherichia coli and Proteus vulgaris. Aeromonas strains x and 19 According to Buttiaux (1959) Aeromonas bacteria play an important part in the biological process on which the manufacture of raw sausage is based. When Aeromonas strains x and 19 were inoculated alone their favourable effect on the quality of dry sausage was restricted to improving colour due to the nitrate-reducing capacity of the Aeromonas strains. When inoculated along with lactobacilli, both Aeromonas x and Aeromonas 19 improved the ripening of dry sausage. The colour developed over 3 days and was at least as good as that of Micrococcus -f- Lactobacillus sausages throughout the ripening period. The pH value of sausages with Aeromonas x or 19 + Lactobacillus inocula- tions decreased as quickly as in Micrococcus + Lactobacillus sausages and often 157 even more quickly. The pH values, too, are as low or a little lower than in Micro- coccus -f- Lactobacillus sausages. According to Nurmi (1965) the inoculation of lactobacilli caused a rapid decrease of the pH value of dry sausage. Concurrently the desired consistency also developed rapidly. Also, the consistency scores of sausages inoculated with Micrococcus -f- Lactobacillus, Aerontonas x + Lacto- bacillus or Aerontonas 19 + Lactobacillus were best. The last of these, however, was better (significance level 0.05) than that of Micrococcus -f- Lactobacillus samples and better (significance level 0.001) than that of Aerontonas x + Lacto- bacillus samples. Results from samples of different ages were taken into account for this comparison. The consistency, measured mechanically, was more often firmer in Aerontonas -)- Lactobacillus sausages than in Micrococcus + Lacto- bacillus sausages. However, the values obtained did not differ very much from each other. The aroma of sausages inoculated with the Aerontonas 19 strain and lacto- bacilli was better (significance level 0.001) than the aroma of sausages inocu- lated with micrococci and lactobacilli. There were no statistical differences between Micrococcus -f- Lactobacillus and Aerontonas x + Lactobacillus samples. The flavour of Aerontonas x -f- Lactobacillus (significance level 0.05) and Aero- monas 19 -j- Lactobacillus (significance level 0.001) sausages was significantly better than the flavour of Micrococcus -f- Lactobacillus sausages. The flavour of all the sausages mentioned was certainly better than in the other sausage groups. Nurmi (1965) observed that micrococci and lactobacilli improved the flavour of dry sausage. The Aerontonas + Lactobacillus inoculations used have proved to have an even more beneficial effect on the flavour. The scores of the evaluations of the sausages of different ages were processed together using the t-test when the aroma and flavour of different sausage groups were compared. There were no significant differences between the evaluations of different panel members with respect to each property or with respect to the products. Products of different ages were tested separately. Nitrate was strongly reduced by Micrococcus and Aerontonas strains, the nitrite level ranging from 10 to 100 ppm after 3 days of ripening. After 3 days the nitrite content was lower in sausages inoculated with lacto- bacilli because of the low pH value. Many countries have discussed banning the use of nitrate. Puolanne (1977) concluded that it is safe from the point of view of technology to manufacture dry sausage using 150 ppm KN0 3 or 100 ppm NaN02 or 50 ppm NaN02 and 75 ppm KN0 3 when starter culture is used. Leistner et al. (1973) state that gram-negative bacteria do not resist sodium nitrite. The count of enterobacteria in 12 day old brine (11 % NaCl) is 105/ml but only 102/ml when 0.06 % nitrite is used. In some experimental series of this investigation the effect of nitrite on Aerontonas strains in dry sausage has also been examined. Sausages were prepared using nitrite and Aerontonas x + Lactobacillus and Aerontonas 19 + Lactobacillus inoculations. According to the results nitrite does not disturb the growth or prevent the Aerontonas strains used from thriving. Based on the weight losses, the following calculations can be made. If control sausages ripen in 3 weeks with a mean weight loss of 24.6 % and the Micro- 158 coccus Lactobacillus and Aeromonas xorl9 + Lactobacillus sausages ripen in 10 days with a mean weight loss of about 18.3% this represents a saving in weight loss of 6 % units. Nurmi (1966) found that sausages with Micro- coccus Lactobacillus inoculation ripened in just 67 days with a weight loss of 10 %. In this study too, the sausages often ripened in one week but were not fit for sale before they had ripened for 10 days. Vibrio costicolus and Achromobacter strains Neither the Vibrio costicolus strain nor the Achromobacter strains thrived in dry sausage and both disappeared during the first few days of ripening. This means that their effect on the ripening of dry sausage is very small, both alone and when inoculated together with lactobacilli. Escherichia coli Escherichia coli did not have a very beneficial effect on the ripening of dry sausage when inoculated alone. When inoculated with lactobacilli its effect on the quality of the sausage very much resembled that of micrococci inoculated with lactobacilli. However, the colour, consistency and flavour were not as good. Proteus vulgaris Inoculation with Proteus vulgaris was used to establish how this kind of strongly proteolytic organism affects dry sausage. The results show that it is harmful even when inoculated with lactobacilli. The consistency, colour, aroma and flavour were not acceptable. 4. 2. Effect of bacterial inoculations on bacterial flora of dry sausage According to Nurmi (1966) Lactobacillus inoculation is necessary to reduce the pH value and so to develop the consistency of dry sausage. The lactobacilli alone produced discolourations because they form hydrogen peroxide and little or no catalase to decompose hydrogen peroxide. Therefore bacteria which are catalase positive and which reduce nitrate are needed besides lactobacilli to prevent discolourations and to produce a good colour. Niinivaara (1955) had developed a method in which micrococci were used as starter culture in dry sausage. So Nurmi used micrococci to prevent discoloura- tions. In this research two gram-negative Aeromonas strains were used instead of micrococci and good quality sausages resulted. Nurmi (1966) stated that micrococci have a strongly depressing effect on the number of lactobacilli in dry sausage the number of lactobacilli is smaller when inoculated together with micrococci. In this investigation the number of lactobacilli was often significantly higher when inoculated with Aeromonas x or 19 than when inocu- lated with micrococci (Table 21). This property of Aeromonas strains proved favourable. It was stated earlier that most Aeromonas + Lactobacillus sau- sages were better than Micrococcus -)- Lactobacillus samples in colour, con- 159 Table 21. Comparison of lactobacillus counts of Micrococcus -f- Lactobacillus sausages with those of Aeromonas x Lactobacillus and Aeromonas 19 4- Lactobacillus sausages. 4-4- = more lactobacilli than in Micrococcus 4- Lactobacillus sausages, significance level