Changes in Microbial Population and Growth of Bacillus cereus During Storage of Reconstituted Dry Milk

680 Journal of Food Protection, Vol. 49, No. 9, Pages 680-686 (September 1986) Copyright International Association of, Food and Environmental Sanitarians Changes in Microbial Population and Growth of Bacillus cereus During Storage of Reconstituted Dry MARCIA H. RODRIQUEZ and ERICKA L. BARRETT* Department of Food Science and Technology, University of California, Davis, California 95616 (Received for publication November 22, 1985) ABSTRACT Eight brands of retail nonfat and whole dry milk were reconstituted and analyzed for changes in the predominant bacterial population and for the proliferation of Bacillus cereus throughout storage at 30, 20, and 5 C. All brands yielded similar results. Bacillus and Micrococcus predominated in the freshly reconstituted milk. During storage at 30 C, the Bacillus population proliferated initially, but was gradually replaced by enterococci. At the time of spoilage, Bacillus counts had dropped by several orders of magnitude. The proportion of Micrococcus gradually declined. B. cereus counts reached hazardous levels as early as 10 h after reconstitution, which was before spoilage was evident. Similar changes occurred in reconstituted milk stored at 20 C, but the time course for the changes was longer, and the Bacillus counts did not decline as rapidly as they did at 30 C. Again, counts of B. cereus reached hazardous levels before the milk showed signs of spoilage. At 5 C, the milk showed no signs of spoilage for 4 to 5 weeks. Bacillus constituted more then 90% of the bacteria isolated after the first week. Bacillus counts continued to increase slowly, but the relative proportion decreased as the gram-negative rods, especially Enterobacter, proliferated. B. cereus never reached numbers great enough to cause disease. The results revealed that the microbial profile of reconstituted dry milk changed significantly over time and that the temperature of storage determined the eventual microbial composition. The results also showed that B. cereus is an omnipresent health hazard in reconstituted milk that is not properly refrigerated. The microbial composition, hence the safety, of reconstituted milk powder is a function of both the initial microbial load and the degree to which the various component organisms have survived and proliferated. Today the former variable can be assessed with some certainty. Microbial analysis of milk powders typically reveals thermoduric streptococci, micrococci, and aerobic sporeforming bacilli, including Bacillus cereus (3,9,17,22,25,30,31). Enterobacteriaceae, particularly Enterobacter species, have also been isolated from milk powders (9,75,25). Pathogens other than B. cereus are rarely isolated from the powders, but the fact that, on occasion, food poisoning outbreaks have been traced to dry milk (/,2,6,8,23) suggests that milk powders may sometimes contain them. Their numbers may be too small to permit detection amidst the background of the normal flora, yet great enough for eventual proliferation. Assessment of the second variable in the evaluation of the safety of milk powder, namely the extent of microbial survival and proliferation, is a more difficult proposition. As has been discussed by Mossel (26), there are many parameters to consider and few reported studies concerning their effect on the microbial ecology of specific foods. It has been shown that the microbial profile of the dry milk changes during storage: total counts tend to decrease, and the relative proportions are also altered because the cocci tend to die out more rapidly than do the sporeforming bacilli (9,18,35). Furthermore, it would be expected that the safety of the milk, once reconstituted, depends on its temperature of storage. The temperature not only directly affects growth of the component bacteria; it also affects the outcome of their competition. We still know very little regarding the differential growth of the milk powder bacteria after reconstitution. Higginbottom (17) studied bacterial growth at two temperatures in reconstituted milk powder obtained from the manufacturer. She found that the reconstituted milk stored at 22 C was characterized by a greater proportion of Bacillus than was milk stored at 15.5 C. Walthew and Luck (37) examined the growth rate of naturally occurring B. cereus in reconstituted milk obtained from manufacturers in South Africa. They found that the growth rate at 25 C was about nine times greater than the growth rate at 20 C, although they concluded that there was no danger of B. cereus poisoning even in milk stored at 25 C as long as it was stored less than 8 h. On the other hand, outbreaks of B. cereus food-poisoning have been attributed to contaminated dry milk (13,19). One possible contributing factor to such outbreaks is a possibly greater growth rate of Bacillus in milk obtained at the retail level, which, due to prolonged storage in the market channels, might contain fewer com-

MICROBIOLOGY OF RECONSTITUTED DRY MILK 681 peting microorganisms. The objective of the present study was to characterize the changing microbial profile of reconstituted retail milk powder during its incubation at different temperatures with particular attention to the differential proliferation of B. cereus. MATERIALS AND METHODS samples Eight different brands of dry milk were obtained at retail outlets. Three brands of dry nonfat milk (,, and ) and two of dry whole milk ( and ) were purchased from local supermarkets in California. Three brands of dry whole milk produced in Venezuela (VI,, and ) were purchased at retail outlets in Venezuela. Samples were prepared for analysis using the procedures outlined by Marth (24). Bacterial enumeration Standard plate count (SPC) procedure using plate count agar (Difco) as outlined by Marth (24) was used to obtain total counts. Psychrotrophs were enumerated on the same medium using the method of Baumann and Reinbold (5). Coliforms were determined by means of the MPN procedure of Marth (24). Staphylococcus aureus was enumerated on Baird-Parker agar (BBL) following non-selective enrichment in trypticase soy broth (Difco) (32). Salmonella detection followed the Food and Drug Administration procedures as specified for dry milk (12). Enterobacteriaceae were enumerated on modified MacConkey agar (28). Aerobic spore formers (i.e. genus Bacillus) were enumerated either by MPN estimation of organisms surviving brief exposure to 100 C in standard methods broth or by enumeration of Bacillus-like colonies on plate count agar containing polymixin B sulfate, which inhibits most gram-negative bacteria (27). Although gram-positive cocci can also grow on this medium, their colonies were easily distinguished from colonies of Bacillus. B. cereus was enumerated on KG agar (21). The reliability of colony recognition was checked periodically by subjecting typical B. cereus colonies to confirmatory tests as outlined by the Food and Drug Administration manual (12). Lactic acid bacteria as a group were enumerated on yeast-dextrose-calcium carbonate (YDC) agar which contains, per liter, 10 g of tryptone, 5 g of yeast extract, 1 g of sorbic acid, 10 g of glucose, 10 g of re-precipitated calcium carbonate, and 15 g of agar. On this medium, colonies of lactic acid bacteria are small and surrounded by clear zones in which the calcium carbonate has dissolved. Identification of bacteria taken from SPC plates Colonies to be identified were picked from the SPC plates, purified, and maintained on slants of the same medium. Standard methods were used for microscopic examination, catalase test, oxidase test, and determination of IMViC reactions. Triple sugar iron agar (Difco) and purple broth base (Difco) containing 0.1% glucose were used to distinguish bacteria capable of fermentation from strictly respiratory bacteria. Among Enterobacteriaceae, MacConkey agar (Difco) was used to determine lactose fermentation and triple sugar iron agar was used to test lactose/sucrose fermentation, hydrogen sulfide (H 2 S) production, and gas production. Salt tolerance was indicated by the ability to grow in brain heart infusion broth (Difco) containing 6.5% NaCl. Gas production by lactic acid bacteria was tested in phenol red broth (BBL) containing 1% glucose. The ability to form spores was indicated either by the presence of spores in stained preparations or by ability to survive 65 C for 30 min. The following criteria were used for genus or group assignment: Bacillus - gram-positive sporeforming rods, catalase-positive; Micrococcus - gram-positive cocci, catalase-positive, nonfermentative; Staphylococcus - gram-positive cocci, catalasepositive, fermentative; lactic acid bacteria - gram-positive, catalase-negative, small colonies surrounded by clear zones on YDC agar; Streptococcus - as for lactic acid bacteria, also coccoid and gas-negative; enterococci - as for Streptococcus, also salt tolerant (11); Pseudomonas or Alcaligenes - gram-negative motile rods, oxidase-positive, strictly respiratory; Acinetobacter - gram-negative immotile coccobacilli, oxidase-negative, strictly respiratory; Enterobacteriaceae - gram-negative rods, oxidasenegative, fermentative and respiratory; Enterobacter and Escherichia - as for Enterobacteriacea, also lactose-positive, H 2 S negative, and IMViC reactions typical of the respective genera; Proteus - as for Enterobacteriaceae, also lactose-negative, sucrose-positive, H 2 S-positive, and IMViC reactions typical of this genus. Storage life assessment The storage life of reconstituted milk held at 20 or 30 C was based on the time of coagulation. Keeping quality was defined as 12 h less than the time at which coagulation occurred (31). Coagulation did not occur in samples stored at 5 C. The keeping quality of these samples was defined as the time required for the development of off-odors. RESULTS AND DISCUSSION Microbial analysis of the dry milk powder None of the eight dry milk samples contained detectable coliforms, psychrotrophs, Salmonella, or S. aureus. Standard Plate Counts and counts of Enterobacteriaceae, aerobic sporeformers, and B. cereus are shown in Table 1. Enterobacteriaceae were found in one sample from Venezuela; in other respects the microbial composition of samples from Venezuela did not differ significantly from that of samples purchased in this country. Similarily, the nonfat dry milk was not found to differ from the whole milk. In general, the counts were low and were within safe microbiological limits. The Standard Plate Counts for all samples were well below the 30,000 per gram limit recommended by the U.S. Public Health Service TABLE 1. Bacterial composition of the dry milk powders (bacteria/g). a VI Standard Plate Count 4,700 2,600 2,100 1,300 2,500 3,000 510 5,000 Enterobacteriaceae 50-100 Aerobic Sporeformers 750 150 930 90 300 750 390 1500 Bacillus cereus 150 30 <10 <10 70 100 270 <10 "Counts presented are mean values obtained from three determinations for each sample using duplicate plates in each determination.

682 RODRIQUEZ AND BARRETT (36). The aerobic sporeformer count ranged from 30 to 1500 per gram, which agrees with typical values reported by previous investigators (4,9,17), and is below the target limit suggested by Mossel and Shennan (29). The B. cereus counts were generally higher than those reported by Walthew and Luck (37), who analyzed dry milk obtained directly from the manufacturer in South Africa, but slightly lower than those reported by Kim and Goepfert (22), who analyzed dry milk obtained at retail outlets in this country. Three of the counts equaled or exceeded the target limit of 100 per gram proposed by Mossel and Shennan (29), but all were below the maximum tolerance of 1000 per gram suggested by the same authors. Effect of storage temperature on keeping quality and microbial profile Portions of each dry milk were reconstituted with sterile water and stored at three different temperatures: 5 C (representing good refrigeration), 20 C (representing milk left standing in a cool room), and 30 C (representing milk left standing in a warm room). The keeping quality of each milk at each temperature was determined. Then samples of incubating milk were plated on nonselective media (SPC agar) midway during storage and at a time shortly before reaching the end of the storage life. Representative colonies from these plates were then purified and identified. Those picked for identification were, in each instance, a random sample consisting of about 10% of the colonies on an SPC plate containing 90-250 colonies. The results (Table 2a-2c) revealed that the predominant microbial population changed significantly during storage, and that the nature of changes were temperature-dependent. Enterococci and other Streptococcus were regularly isolated from reconstituted milk incubated at 20 and 30 C even though these organisms were only rarely detected in the freshly reconstituted samples. Similarily, gram-negative rods were among the predominant microorganisms in five of the eight milk samples stored at 5 C although they were completely overshadowed by Bacillus and Micrococcus in all but one of the fresh samples. The relative Bacillus population declined significantly during storage at 30 C and declined slightly at 20 C. At 5 C it increased during the first week of storage and then declined to about the levels found in the freshly reconstituted milk. The prevalence of organisms at the end of incubation which were not commonly found in the fresh sample indicates either that they were present in such small num- TABLE 2a. Changes in the microbial population of reconstituted dry milk during storage at 30 C. Genera or groups isolated (% of total)" Keeping quality Initial 1 d 2 d 21 h Enterococci (33) VI 25 h 26 h 31 h 18 h 22 h 27 h 19 h Micrococcus (33) Bacillus (86) Micrococcus (14) Bacillus (33) Micrococcus (56) Staphylococcus (11) Micrococcus (29) Micrococcus (17) Escherichia (17) Bacillus (45) Micrococcus (27) Staphylococcus (27) Micrococcus (33) Combined Bacillus (65) Micrococcus (25) Enterococci (4) Staphylococcus (4) Escherichia (2) Bacillus (46) Enterococci (39) Enterobacter (15) Bacillus (60) Micrococcus (40) Bacillus (46) Micrococcus (23) Enterococci (31) Enterococci (20) Other Strep. (30) Bacillus (57) Other Strep. (7) Escherichia (7) Enterobacter (7) Bacillus (12) Enterococci (62) Other Strep. (25) Enterococci (33) Bacillus (55) Micrococcus (11) Enterococci (31) Escherichia (2) Other Strep. (6) Bacillus (44) Enterococci (37) Micrococcus (19) Enterococci (64) Micrococcus (8) Other Strep. (28) Bacillus (31) Enterococci (62) Other Strep. (7) Enterococci (60) Other Strep. (40) Bacillus (15) Enterococci (54) Other Strep. (31) Other Strep. (27) Mirococcus (36) Bacillus (23) Enterococci (54) Other Strep. (23) Bacillus (40) Enterococci (60) Bacillus (18) Micrococcus (13) Enterococci (54) Other Strep. (16) Total count after 2 d.6x10" 7.9X10 1 1.9X10 9 5.0xl0 y 1.2X10 10 3.9X10" 3.6X10" 2.6x10 s ^jenus or group assignments were determined as described in Materials and Methods. ''Pseudomonas" refers to Pseuodomonas or Alcaligenes. "Other Strep." refers to Streptococcus other than enterococci. Bacteria present initially are listed in Table 2a only. JOURNAL OF FOOD PROTECTION. VOL. 49, SEPTEMBER 1986

MICROBIOLOGY OF RECONSTITUTED DRY MILK 683 bers in the powder that they escaped detection or that they were present in injured form. No matter which explanation is correct, the results show that enumeration of bacterial types in the powder is not the best predictive index for the microbial profile of the reconstituted milk which has been allowed to stand. Growth of Bacillus in the reconstituted milk Danger of B. cereus food poisoning would be expected to be associated with conditions favoring the proliferation of Bacillus. The significant changes in Bacillus population during storage were further examined by enumerating Bacillus at shorter time intervals during storage. Lactic acid bacteria (predominantly enterococci) were also enumerated in the samples incubated at 30 and 20 C. All milk samples were included in this study. In our initital experiments, Bacillus was enumerated by means of spore counts. However, we noted a significant discrepancy between the spore counts and the number of Bacillus obtained indirectly by identification of colonies appearing on the SPC plates. This was especially true of the reconstituted milk samples incubated at 5 C C. The lower spore counts indicated that not all the Bacillus in the dry milk had formed spores. Inefficient sporulation of Bacillus in foods has been noted by others (27), and has been shown to be an important problem in the microbial analysis of refrigerated foods (33). To obtain more accurate counts of Bacillus in our growth studies, we plated samples on SPC agar containing polymixin B. This method also yielded lower Bacillus counts than were obtained indirectly when colonies on SPC plates without antibiotic were analyzed for representative microbial types, but the difference was much less significant. There was very little variation among the patterns obtained for the eight reconstituted milk samples incubated at 20 and 30 C C. Typical plots for these temperatures, (those of sample ) are shown in Fig. 1 and 2, respectively. In each case, Bacillus proliferated initially, but the lactic acid bacteria eventually predominated. The Bacillus counts after prolonged incubation were several orders of magnitude lower than the peak counts for this bacterial group, whereas the counts of lactic acid bacteria declined only slightly after reaching their maxima. An antagonistic action of enterococci and other spoilage bacteria on the growth of aerobic sporeformers has been noted by others (20,34). The results for the reconstituted milk samples incubated at 5 C formed two types of patterns, as exhibited TABLE 2b. Changes in the microbial population of reconstituted dry milk during storage at 20 C. Genera or groups isolated (% of total) 3 Keeping quality 1 d 2d 3 d 3 d Bacillus (60) Bacillus (64) Enterococci (10) Other Strep. (9) Other Strep. (30) Micrococcus (30) 2.5 d Bacillus (47) Bacillus (60) Enterococci (27) Enterococci (30) Other Strep. (20) Micrococcus (10) Micrococcus (10) 3d Micrococcus (29) 3 d 3 d 4d 3d 3d Bacillus (78) Enterococci (22) Bacillus (80) Enterococci (20) Combined Bacillus (76) Enterococci (12) Micrococcus (4) Other Strep. (8) Bacillus (33) Enterococci (50) Staphylococcus (17) Enterococci (29) Bacillus (78) Enterococci (12) Micrococcus (6) Other Strep. (2) Staphylococcus (2) Enterococci (22) Micrococcus (10) Bacillus (29) Enterococci (57) Micrococcus (14) Other Strep. (14) Bacillus (90) Enterococci (10) Bacillus (30) Enterococci (20) Enterobacter (30) Proteus (20) Bacillus (30) Enterococci (50) Other Strep. (20) Enterobacter (33) (not done) Enterococci (30) Micrococcus (4) Other Strep. (5) Enterobacter (8) Proteus (3) Total count after 3 d 1.7 XlO 9 1.6X10 9 4.0 XlO 8 6.3 xlo 8 8.9 XlO 8 6.3 XlO 8 4.0 xlo 9 1.6 xlo 9 "Genus or group assignments were determined as described in Materials and Methods, "Pseudomonas" refers to Pseudomonas or Alcaligenes. "Other Strep." refers to Streptococcus other than enterococci. Bacteria present initially are listed in Table 2a only. JOURNAL OF FOOD PROTECTION. VOL. 49, SEPTEMBER 1986

684 RODRIQUEZ AND BARRETT TABLE 2c. Changes in the microbial population of reconstituted dry milk during storage at 5 C. VI 5 wks 4 wks 7 wks 6 wks 7 wks 7 wks 5 wks Combined Bacillus (94) Escherichia (4) Other 5lrep. (2) Genera or groups isolated (% of total)" Keeping quality 1 week 3 weeks 5 weeks 4 wks Escherichia (29) Bacillus (80) Other Strep. (20) Bacf7/i«(38) Enterobacter (38) Other 5/rep. (25) Other 5?rep. (50) Bacillus (75) Other 5?rep. (25) Sac(7/«j (100) fiac///«i (38) Enterobacter (50) Pseudomonas (12) Bad/to (100) fiac(7to (91) Micrococcus (9) Enterobacter (12) Other S/rep. (13) Micrococcus (2) Pseudomonas (2) Bacillus (73) Enterobacter 27) Bac///u.s (33) Enterobacter (11) Proteus (22) Pseudomonas (22) Acinetobacter (11) Sac;7/«i (83) Pseudomonas (17) Micrococcus (40) Other 5?rep. (10) Bacillus (18) Enterobacter (73) Proteus (9) Bari//i«(100) Bacillus (22) Micrococcus (11) Enterobacter (67) Sac///«i (100) Bacillus (54) Enterobacter (27) Other 5?rep. (1) Micrococcus (7) Pseudomonas (5) Proteus (5) Acinetobacter (1) Total count after 5 weeks 1.3 X10 5 7.9 XlO 7 5.1 xlo 5 7.1 X10 4 4.0 XlO 7 5.6X10 3 1.0 XlO 7 3.2 xlo 4 "Genus or group assignments were determined as described in Materials and Methods. " Pseudomonas" refers to Pseudomonas or Alcaligenes. "Other Strep." refers to Streptococcus other than enterococci. Bacteria present initially are listed in Table 2a only. by samples and VI in Fig. 3. In general, it was found that slower overall growth corresponded to a higher proportion of Bacillus. Samples which developed a significant population of gram-negative rods (see Table 2c) reached the highest total counts, including higher counts of Bacillus. Apparently the psychrotrophic spoilage bacteria extended the growth potential of Bacillus at the lower temperatures. The different spoilage patterns, however, could not have been predicted by the composition of the dry powder as psychrotrophs had not been detected in any of the milk powders. It has been reported that the level of contamination necessary for initiation of B. cereus gastroenteritis is 10 6-10 8 cells per gram or ml of food (10,14). To determine how many hours of incubation would be required for B. cereus to reach this level we followed the growth of this organism in the five milk powders in which it was detected (see Table 1). Growth curves were obtained for reconstituted milk incubated at 30 and 20 C C (Fig. 4). Each point on the curves represents the mean value for two trial experiments, each using duplicate plates. The total time was selected as the mean time for spoilage at the given temperature as determined in the previous experiments. In all instances, B. cereus reached the levels necessary to cause disease before spoilage was evident. Depending on the milk sample, 12-22 h were required for B. cereus counts to reach 10 6 /g at 30 C C, and 24 to 56 h at 20 C C. Although it is unlikely that reconstituted milk left standing so long would be served as a beverage, its use in the preparation of uncooked foods (such as instant puddings) and the evaluation of the freshness of such foods left standing may not always receive as much scrutiny. Similar experiments conducted at 5 C C (data not shown) revealed no apparent danger of B. cereus poisoning for up to 5 weeks of incubation. At that time the numbers of B. cereus were 10 3-10 4 per ml. Thus B. cereus behaved as a mesophilic organism. Psychrotrophic strains of B. cereus have been isolated from raw milk, but even these strains, when grown at refrigeration temperatures, exhibited a lag of 14-18 d and a relatively slow growth rate (7). It appears that the risk of B. cereus poisoning from reconstituted milk is associated primarily with its storage at or above room temperature. The results for the 30 C incubation illuminate the difficulties in predicting food safety as well as in tracking the cause of food poisoning when food samples are analyzed before incubation or long after the illness has been initiated. Neither the freshly reconstituted milk nor the milk incubated for 2 d contained enough Bacillus to

MICROBIOLOGY OF RECONSTITUTED DRY MILK 685 10 10 20 30 4-0 50 Hours of Incubation 96 1 O 20 30 AO 50 Hours of Incubation 60 96 Figure 1. Bacterial growth in reconsituted dry milk incubated at 30 C (milk sample ). O O, aerobic plate count; A A, lactic acid bacteria (primarily enterococci); D D, typical Bacillus. constitute danger of B. cereus poisoning, yet the numbers of B. cereus had reached high enough to levels cause disease during the storage period. ACKNOWLEDGMENTS The foregoing studies were supported in part by funds from the California Agricultural Experiment Station. We thank Richard Bradford for his technical artistry. REFERENCES 1. Anderson, P. H. R., and D. M. Stone. 1955. Staphylococcus food poisoning associated with spray-dried milk. J. Hyg. 53:387-397. 2. Armijo, R., D. A. Henderson, R. Timothee, and H. B. Robinson. 1957. Food poisoning outbreaks associated with spray-dried milk - an epidemiologic study. Amer. J. Public Health 47:1093-1100. 3. Arun, A. P. S., C. Prasad, B. K. Sinha, and B. N. Prasad. 1979. Mesophilic bacteria in spray dried skim milk powder and its public health importance. Indian J. Dairy Sci. 32:468-471. 4. Atwal, J. S., P. C. Reddy, R. Chann, and R. A. Srinivason. 1974. Occurrence of aerobic sporeforming bacteria in milk products. Indian J. Dairy Sci. 27:297-299. 5. Bauman, D. P., and G. W. Reinbold. 1963. Enumeration of psychrophilic microorganisms in dairy products. J. Food Technol. 26:351-356. 6. Bryan, F. L. 1983. Epidemiology of milk-borne disease. J. Food Prot. 46:637-649. Figure 2. Bacterial growth in reconstituted dry milk incubated at 20 C (milk sample ). O O, aerobic plate count; A A, lactic acid bacteria (primarily enterococci);, typical Bacillus. 7. Chung, B. H., and R. Y. Cannon. 1971. Psychrotrophic sporeforming bacteria in raw milk supplies. J. Dairy Sci. 54:448. 8. Collins, R. N., M. D. Treger, J. B. Goldsby, J. R. Boring III, D. B. Coohon, and R. N. Barr. 1968. Interstate outbreak of Salmonella newbrunswick infection traced to powdered milk. J. Am. Med. Assoc. 203:838-844. 9. Crossley, F. L., and W. A. Johnson. 1942. Bacteriological aspects of the manufacture of spray-dried milk and whey powders, including some observations concerning moisture content and solubility. J. Dairy Res. 13:5-44. 10. Davies, F. L., and G. Wilkinson. 1973. Bacillus cereus in milk and dairy products. Pages 57-67. In B. C. Hobbs and J. H. B. Christian (eds)., The microbiological safety of food. Academic Press, New York. 11. Deibel, R. H., and P. A. Hartman. 1976. The enterococci. Pages 370-373. In M. L. Speck (ed.), Compendium of methods for the microbiological examination of foods. American Public Health Association, 12. Food and Drug Administration. 1976. Bacteriological analytical manual for foods. Association of Official and Analytical Chemists, 13. Goepfert, J. M., W. M. Spira, and H. U. Kim. 1972. Bacillus cereus: food poisoning organism. A review. I. Food Technol. 35:213-227. 14. Goepfert, J. M., W. M. Spira, B. A. Glatz, and H. U. Kim. 1973. Pathogenicity of Bacillus cereus. Pages 69-75. In B. C. Hobbs and J. H. B. Christian (eds)., The microbiological safety of food. Academic Press, New York. 15. Hall, H. E., D. F. Brown, H. M. Robinson, C. B. Donnelly, and A. L. Reyes. 1967. Bacteriological examination of grade A dry milk powder. J. Food Technol. 30:219-221. 16. Higginbottom, C. 1944. Bacteriological studies of roller-dried milk

686 RODRIQUEZ AND BARRETT 2 3 4 5 Weeks of Incubation Figure 3. Bacterial growth in reconstituted dry milk incubated at 5 C. Closed symbols, milk sample ; open symbols, milk sample VI. and O O, aerobic plate count; and D D, typical Bacillus. Figure 4. Growth of Bacillus cereus in naturally contaminated reconstituted dry milk incubated at 30 C (A) and at 20 C (B). Hexagons, milk sample ; circles, milk sample ; square, milk sample ; closed triangles, milk sample VI; open triangles, milk sample. powders, roller-dried buttermilk, and of roller and spray-dried whey. J. Dairy Res. 13:308-323. 17. Higginbottom, C. 1948. Bacterial growth in reconstituted spray dried milk. J. Dairy Res. 15:285-291. 18. Higginbottom, C. 1953. Effect of storage at different relative humidities on the survival of microorganisms in milk powder and in pure cultures dried in milk. J. Dairy Res. 20:65-75. 19. Johnson, K. M. 1984. Bacillus cereus foodborne illness: an update. J. FoodProt. 47:145-153. 20. Kafel, S., and J. C. Ayres. 1969. The antagonism of enterococci on other bacteria in canned hams. J. Appl. Bacteriol. 32:217-232. 21. Kim, H. U., and J. M. Goepfert. 1971. Enumeration and identification of Bacillus cereus in food. I. 24 hour presumptive test medium. Appl. Microbiol. 22:581-587. 22. Kim, H. U., and J. M. Goepfert. 1971. Occurrence of Bacillus cereus in selected dry food products. J. Food Technol. 34:12-15. 23. Marth, E. H. 1969. Salmonella and salmonellosis associated with milk and milk products - a review. J. Dairy Sci. 52:283-316. 24. Marth, E. H. (ed.). 1978. Standard methods for the examination of dairy products. 14th ed. American Public Health Association, 25. Mattick, A. T. R., E. R. Hiscox, E. L. Crossley, C. H. Lea, J. D. Findlay, J. A. B. Smith, S. Y. Thompson, and S. K. Kon. 1945. Effects of temperature of pre-heating, of clarification, and of bacteriological quality of the raw milk on the keeping properties of whole-milk powders dried by Kestner spray process. J. Dairy Res. 14:116-114. 26. Mossel, D. A. A. 1983. Essentials and perspectives of the microbial ecology of foods, pages 1-45. In T. A. Roberts and F. A. Skinner (eds.), Food microbiology: Advances and prospects. Academic Press, Inc. New York. 27. Mossel, D. A. A., M. J. Koopman, and E. Jongerius. 1967. Enumeraton of Bacillus cereus in foods. Appl. Microbiol. 15:650-653. 28. Mossel, D. A. A., W. H. J. Mengerink, and H. H. Scholts. 1962. Use of a modified MacConkey agar medium for the selective growth and enumeration of Enterobacteriaceae. J. Bacteriol. 84:381. 29. Mossel, D. A. A., and J. L. Shennan. 1976. Microorgansims in dried foods: their significance, limitations, and enumeration. J. Food Technol. 11:205-220. 30. Mull, L. E., and K. L. Smith. 1965. Some observations on the bacterial quality of instant nonfat dry milk in consumer packages. J. Food Technol. 28:188-191. 31. Nichols, A. A. 1939. Bacteriological studies of spray-dried milk powder. J. Dairy Res. 10:202-230. 32. Ordal, Z. J., J. J. Iandola, B. Ray, and A. G. Sinskey. 1976. Detection and enumeration of injured microorganisms. Pages 163-169. In M. L. Speck (ed.), Compendium of methods for the microbiological examination of foods. American Public Health Association, 33. Shehata, T. B., A. Duran, and E. B. Collins. 1971. Influence of temperature on the growth of psychrophilic strains of Bacillus. J. Dairy Sci. 54:1579-1582. 34. Steele, J. E., and M. E. Stiles. 1981. Food poisoning potential of artificially contaminated vacuum packaged sliced ham in sandwiches. J. Food Prot. 44:430-434, 444. 35. Thompson, S. S., L. G. Harmon, and C. M. Stine. 1978. Survival of selected organisms during spray drying of skim milk and storage of nonfat dry milk. J. Food Prot. 41:16-19. 36. U. S. Public Health Service. 1969. Grade A dry milk products. Supplement I to the Ordinance and Code - 1965. Recommendations of the Public Health Service. U. S. Public Health Service, 37. Walthew, J., and H. Luck. 1978. Incidence of Bacillus cereus in milk powder. S. Afr. J. Dairy Technol. 10:47-50 (Food Sci. Technol. Abstr. 11:1P7, 1979).