PROCESSING AND PRODUCTS

Transcription

1 PROCESSING AND PRODUCTS Effects of a Buffered Propionic Acid in Diets on the Performance of Broiler Chickens and on Microflora of the Intestine and Carcass 1 A. L. IZAT, N. M. TIDWELL, R. A. THOMAS, M. A. REIBER, M. H. ADAMS, M. COLBERG, and P. W. WALDROUP Department of Animal and Poultry Sciences, University of Arkansas, Fayetteville, Arkansas (Received for publication August 23, 1988) ABSTRACT A buffered propionic acid (BPA) was added to broiler diets fed in floor pens with litter. The BPA was fed continuously at 0,.2,.4, and.8% in Trial 1 and at 0 and.4% in Trial 2. The BPA was also fed at.4% for the last 7 days in Trial 2. Natural salmonellae exposure versus periodic dosage with Salmonella typhimurium was compared in Trial 2. In Trial 1, the BPA supplement had no adverse effects on growth, feed utilization, or abdominal fat with a significant (P<.05) increase in the female dressing value at.8% of buffered propionic acid. The total number of coliforms and of Escherichia coli in the duodenum were significantly reduced by.4% BPA; in the jejunum, by all levels used in the trials; and in the ileum, by.4% and.8% of buffered propionic acid. The intestinal ph was not influenced by the BPA addition. In Trial 2, the BPA at.4% fed continuously had no adverse effect on growth, feed utilization, the abdominal fat of females, or the dressing percentage of males while significantly reducing the abdominal fat for males and increasing the dressing percentage for females. Feeding.4% BPA for the last 7 days had no effect on any of these parameters. Periodic dosage with S. typhimurium had no effect on body weight, feed utilization, or abdominal fat and significantly increased the dressing percentage. There was a significant interaction between the Salmonella dosage and the time of feeding BPA on dressing percentage. In Trial 2, the BPA at.4% fed continuously or during the last 7 days of the trial produced a significant reduction in the number of salmonellae on postchill carcasses when the birds were periodically dosed with Salmonella. There was no significant effect from BPA on the number of salmonellae or of coliforms in the small intestine, but there was a significant reduction in the total number of organisms when.4% BPA was fed either continuously or during the last 7 days of the trial. (Key words: salmonellae, processing, buffered propionic acid, growth, organic acid) INTRODUCTION Propionic acid and its salts have long been used in poultry diets as effective moldinhibitors (Stewart et al, 1977; Paster, 1979; Dixon and Hamilton, 1981). Some studies have indicated that the addition of propionic acid and other organic acids to the diet of broilers may influence the populations of microbes in the intestine and may promote growth or improve feed utilization (Vogt et al., 1981, 1982). Patten and Waldroup (1988) observed significant improvements in growth and feed utilization for broilers fed fumaric acid. Van der Wal (1980) found that including formic, acetic, propionic, or lactic acid resulted Published with the approval of the Director, Arkansas Agricultural Experiment Station Poultry Science 69: in significant reductions in the number of Enterobacteriaceae in the feed. When these acids were included in the diet of pigs, a reduction also occurred in the number of animals infected with Salmonella. In general, the primary source of carcass contamination is thought to derive from the intestinal tract of the broiler (Bryan et al., 1968; Morris and Wells, 1970; Rigby et al, 1980). Therefore, any means of reducing intestinal microbial levels should be reflected in reduced carcass levels following processing. Several field reports have suggested that the addition of a buffered propionic acid (BPA) mold-inhibitor at levels greater than the recommended one (.1 to.15% of the diet) resulted in performance enhancement for growing broilers with intestinal microbial alterations. The objectives of the present study were to determine the possible influence of a BPA

2 BUFFERED PROPIONIC ACID AND CARCASS SALMONELLA 819 TABLE 1. Composition of the diets Ingredients Yellow corn Soybean meal (48% CP) Propack (60% CP) 1 Blended fat Dicalcium phosphate Ground limestone Vitamin premix 2 Salt Trace-mineral mix DL-methionine (98%) Total 0 to 21 days Feeding period 21 to 42 days fm 42 to 49 days 'Blended animal protein (H. J. Baker and Bro., New York, NY). 2 Provides per kilogram of diet: vitamin A, 6,600 IU; vitamin D3, 2,200 IU; vitamin E, 6.6 IU; menadione sodium bisulfate, 3.3 mg; riboflavin, 5.5 mg; niacin, 33 mg; pantothenic acid, 8.8 mg; choline, 495 mg; thiamine, 1.1 mg; pyridoxine, 1.1 mg; vitamin Bj2,.01 mg; biotin,.11 mg; folacin,.66 mg; ethoxyquin, 125 mg; selenium,.1 mg. ^ovides per kilogram of diet: iron, 100 mg; zinc, 100 mg; manganese, 100 mg; copper, 10 mg; iodine, 1 mg. mold-inhibitor (Luprosil NC) 2 on the performance of broiler chickens, the microbial population of the intestinal tract, and selected carcass characteristics when fed at levels higher than those normally used for moldinhibition. Experiment 1 MATERIALS AND METHODS Day-old chickens of a commercial broiler strain (Cobb 500) were obtained from a local hatchery and were randomly assigned to pens in a broiler test house. The pens contained litter composed of wood shavings on which two previous groups of broilers had been grown. The birds were fed high-energy starter (0 to 21 days), grower (21 to 42 days), and finisher (42 to 49 days) diets for broilers based primarily on corn and soybean meal supplemented with a blended source of animal protein (Table 1). The diets were calculated to be nutritionally adequate (National Research Council, 1984). Salinomycin, 3 at 66 mg per kg of feed, was used as an anticoccidial in all diets until 42 days of age. 2 BASF Wyandotte Corp., Parsippany, NJ. 3 Bio-Cox, Agri-Bio Corp., Gainesville, GA. 4 BMD, A. L. Laboratories, Inc., Fort Lee, NJ. Roxarsone 10%, Rhone-Poulenc, Inc., Atlanta, GA. Aliquots of the basal diet within each age group were supplemented with the BPA at 0,.2,.4, and.8% by weight, replacing an equal amount of ground com. The BPA product is composed of 53.5% propionic acid, 9.5% ammonium hydroxide, 11.5% 1,2-propanediol, and 25.5% water, is soluble in water, and has a ph of 6.8 to 7.1. These diets were fed with and without the addition of 55 mg per kg of bacitracin methylene disalicylate 4 and 50 mg per kg of roxarsone. 5 These additives are commonly used in poultry diets to promote growth and to improve feed utilization (Miller Publishing Co., 1988a). This resulted in a total of eight diets. The roxarsone was removed from all diets at 42 days of age. Each of these diets was fed to six pens of chickens (5.2 m 2 ), each pen containing 25 male and 25 female chickens. The pens were randomly assigned to the dietary treatments. Pen body weights by sex were taken at 21,42, and 49 days of age. Feed consumption by pen was determined at the same intervals in order to determine feed utilization. Mortality was noted as it occurred, with the weight of birds that died used to adjust feed utilization. At 49 days of age, five males and five females were randomly selected from each pen for processing. Feed and water were not withdrawn from the birds prior to processing. The birds from each treatment were caught 1 h before processing, placed in coops, and transported to the University of Arkansas processing plant.

3 820 IZAT ET AL. Birds from the various treatments were caught at staggered intervals in order to equalize the time between catching and processing. The birds were stunned by an electrical knife, 6 killed, and allowed to bleed, then scalded at 54 C for 2 min in a dunking scalder 7 and picked in a rotarydrum picker. 8 The birds were manually eviscerated, taking care not to rupture the intestinal tracts. The amount of abdominal fat was determined along with the dressing percentage. "Abdominal fat" was defined as the fat surrounding the gizzard, extending within the ischium, and surrounding the bursa of Fabricius, cloaca, and the adjacent abdominal muscles. "Dressing percentage" was defined as the dressed carcass weight, without the neck or giblets, divided by the live carcass weight. All birds within a treatment were processed consecutively, the scald water was discarded, then all processing equipment was cleaned and sanitized before processing the next treatment group. Following evisceration, the intestinal tracts from three males taken from each replicate pen were divided into three segments: the duodenum, the jejunum (that part of the small intestine extending from the duodenum to Meckel's diverticulum), and the ileum. A sample of the contents from each segment was gently expelled into containers. Each intestinal sample was mixed with 10 ml of distilled water and the ph was determined. 9 To determine the total number of coliform and Escherichia coli organisms, six samples were collected from each of the three intestinal segments, each one being a composite sample collected from the three birds. The contents of each segment sample were weighed (for enumeration calculations), then added to 50 ml of the sterile, buffered-phosphate diluent. Immediately after sample collection, a threetube, most-probable-number (MPN) procedure (Oblinger and Koburger, 1984) was completed for the enumeration of coliforms and E. coli in each of the intestinal samples. The isolation and identification of the organisms was completed Tvlodel G, Cervin Automated Systems, Inc., Minneapolis, MN). 7>8 Pickwick Co., Cedar Rapids, IA. Accumet ph Meter 910, Fisher Scientific, Fair Lawn, NJ. 10 ATCC 14028, American Type Culture Collection, Rockville, MD. 'Plasson Broiler Drinker, Diversified Imports, Inc., Lakewood, NJ. by using the methods outlined by the Food and Drug Administration (1984). Experiment 2 Day-old chicks of a commercial broiler strain (Cobb 500) were randomly assigned to pens, as previously described, and were divided into three treatment groups. Each pen contained 25 males and 25 females. Twelve pens were assigned to each of the three treatment groups. The pens contained litter composed of wood shavings on which three previous flocks of birds had been reared without the use of BPA in the diet. One group received the diet previously described, fortified with salinomycin, BMD, and roxarsone at the levels used in the first experiment, but with no BPA. A second group was fed the same diet, supplemented with.4% BPA from Day 1 until the completion of the study at 49 days. The third group was fed the unsupplemented diet to 42 days of age, then receiving the diet containing.4% BPA from 42 to 49 days of age. The birds in six of the replicate pens were grown to 49 days of age with only natural exposure to salmonellae. The birds in other six replicate pens were inoculated with 10 5 to 10 6 cfu per ml of Salmonella typhimurium 10 in the water fountains 11 at 2,7,14, 21, 28, 35,42, and 47 days of age. Previous studies in the authors' laboratory have demonstrated that natural salmonellae contamination cannot be relied on to provide a significant intestinal population and that inoculation into the drinking water allows for positive individual bird contamination (Izat et al unpublished data). The drinking water was not chlorinated, and no attempt was made to monitor the survival of the organism after inoculation. Caretakers took care of the uninoculated birds first, then changed their footwear and outer clothing in order to care for the inoculated birds. The footwear was scrubbed with a disinfectant and outer clothing was washed before being used again. Immediately after weighing, at 49 days of age and without withdrawing feed or water, 5 males and 5 females from each pen were processed, as previously described, to determine the dressing percentage and content of abdominal fat. The first two males processed within each pen were selected for evaluation of the intestinal microflora. Their intestinal tracts were carefully removed, taking care not to contaminate the outside surface of the carcass. After eviscera-

4 BUFFERED PROPIONIC ACID AND CARCASS SALMONELLA 821 TABLE 2. Effects of buffered propionic acid (BPA)' fed in combination with bacitracin methylene disalicylate 2 and roxarsone on the body weight, feed utilization, and carcass characteristics of broiler chickens, Experiment 1 BPA (%) 0 x" SEM Feed additives 4 (-/) x" Males 2,137 2,100 2,119 2,134 2,118 2,131 2,255 2,126 2,191 2,165 2,205 2,185 2,175 2,137 2,156 Weight at 49 days Females Ctrl \b/ 1,7% 1,844 1,820 1,838 1,722 1,781 1,778 1,888 1,833 1,795 1,890 1,838 1,802 1,835 1, : 29.5 Gain:feed ratio, 0 to 49 days - (g/g) Males Abdominal fat-pad weight Females fr> a,b Within columns, means with no common superscripts differ significantly (P<.05). Luprosil NC, BASF Wyandotte Corporation, Parsippany, NJ. 2 BMD-50, A. L. Laboratories, Inc., Fort Lee, NJ. 3 Roxarsone 10%, Rhone-Poulenc, Inc., Atlanta, GA. 4 Addition of 55 mg per kg of BMD and 50 mg per kg of roxarsone, tion, the carcasses were placed in individual containers of nonagitated chill water. After being chilled for 1 h, the two male carcasses from each pen were subjected to a whole-carcass rinse, as described by Cox et al. (1983). Then, the contents of the small intestine and the wholecarcass rinse samples were evaluated for the total number of coliforms, bacteria, and MPN salmonellae using standard enumeration and identification methods (Oblinger and Koburger, 1984; FDA, 1984). The data obtained from these experiments were analyzed as a completely random design using the general linear model procedure of the Statistical Analysis System (SAS Institute, 1982). The model for all variables analyzed in Experiment 1 included the level of BPA, presence or absence of bacitracin MD and roxarsone, and interaction of the two main effects. In Experiment 2, the model included the time of feeding BPA, presence or absence of Salmonella inoculation, and the interaction of these two main effects. The residual mean square was used as the error term in the analyses of data from Experiment 1 and Experiment 2. In 1.8 Dressing percentage without fat pad Males ( M n\ Females b b ab l a each study, CONTRAST statements were used for a single-degree comparison on the levels of BPA versus the unsupplemented control birds. All statements of statistical significance are based on P<.05. Experiment 1 RESULTS AND DISCUSSION The addition of.2,.4, or.8% BPA to the diet of broiler chickens had no significant effect on the 49-day body weight of male and female broilers, feed utilization, the abdominal fat-pad weight for male and female broilers, or the dressing percentage of male broilers (Table 2). The dressing percentage of the female broilers fed.8% BPA was significantly higher than that of the females given the negative control diet or the diet with.2% buffered propionic acid. The addition of the combination of BMD and roxarsone had no significant effect on any of these factors, and there was no significant interaction between BPA and the combination of BMD and roxarsone..22

5 822 IZAT ET AL. TABLE 3. Effects of a buffered propionic acid (BPA) fed in combination with bacitracin methylene disalicylate 2 and roxarsonr on the total number of coliforms and Escherichia coli in the duodenum, jejunum, and ileum, Experiment 1, n = 6 BPA (%) Duodenum Jejunum * Ileum No Logio total coliforms Yes a 5.23 a 4.61 b 5.28 a 6.66" 5.52 b 5.86 b 5.69 b Bacitracin plus roxarsone " 7.03 ab 6.44 b 6.60 b No MPN^ ner a\ a_c Within columns, means with no common superscripts differ significantly (P<.05). 'Luprosil NC, BASF Wyandotte Corporation, Parsippany, NJ. 2 BMD-50, A. L. Laboratories, Inc., Fort Lee, NJ. 'Roxarsone 10%, Rhone-Poulenc, Inc., Atlanta, GA. 4 Addition of 55 mg per kg of BMD and 50 mg per kg of roxarsone (No, Yes). 5 MPN = most-probable number. ^EM =.31 for total coliforms and.37 for E. coli. 7 SEM =.41 for total coliforms and.41 for E. coli. 8 SEM =.46 for total coliforms and.52 for E. coli. The addition of BPA to the diet of broiler chickens resulted in significant reductions in the number of microorganisms in various segments of the small intestine, compared with untreated control birds (Table 3). In the duodenum, there was a significant reduction in both the total number of coliforms and of E. coli when.4% BPA was included in the diet. Adding.8% BPA also resulted in a significant reduction in number of E. coli, in comparison with the untreated control birds. In the jejunum, all levels of BPA in the diet resulted in significant reductions in both the total number of coliforms and off. coli, compared with the untreated control birds. In the ileum, adding.4 or.8% BPA produced significant reductions in the total number of coliforms and off. coli, versus the untreated control birds. There was no effect by the dietary treatment on the ph of the various gut segments (data not Log 10 E. coli Yes a 5.14 ab 4.42 c 4.61 bc 6.50 a 5.42 b 5.68 b 5.32 b 7.46 a 6.94 ab 6.32 b 6.25 b shown). Since BPA is a buffered product with a neutral ph, a change in gut ph was not anticipated. Therefore, the reduction in intestinal microbial numbers in the present study apparently was not caused by a change in the ph within the intestine. Bacitracin is not considered to be a strong coliform-inhibitor, but has been shown to weaken the cell walls of these organisms, thus making them more susceptible to other antibiotics or to acids or alkalies present in the digestive tract (Walton and Bird, 1975; Walton, 1978, 1984; Walton and Wheeler, 1987). In the current study, however, adding the combination of BMD and roxarsone had no significant effect on the number of total coliforms and of E. coli in the various segments of the small intestine. There was no significant interaction between BPA and the combination of Bacitracin MD and

6 BUFFERED PROPIONIC ACID AND CARCASS SALMONELLA 823 roxarsone on the number of microorganisms. The authors are not aware of any studies concerning the possible effect of roxarsone on coliforms. Experiment 2 Table 4 shows the effects from the dietary supplementation of BPA and periodic dosage with S. typhimurium in the water fountains on growth rate, feed utilization, and the carcass characteristics of broiler chickens. The addition of.4% BPA to the diet of broiler chickens for 49 days or from 42 to 49 days had no significant effect on the 49-day weight of males or females or on feed utilization, compared with the nonsupplemented controls. Feeding of.4% BPA from 0 to 49 days significantly reduced the abdominal fat-pad weight for males but not for females. Feeding BPA, either continuously or during the last week, had no effect on the dressing percentage of males; but the addition of.4% BPA during the entire 49-day feeding period significantly increased the dressing percentage of females. This outcome agrees with the increased dressing percentage observed in females fed.8% BPA continuously during the first experiment. Periodic dosage with S. typhimurium in the water fountains had no significant effect on the 49-day body weight for male or female broilers or on the feed utilization or the abdominal fat content for male or female broilers (Table 4). The dressing percentage for both male and female broilers was significantly increased in birds subjected to periodic dosage in the water fountains with S. typhimurium. There was a significant interaction between the Salmonella dosage and the BPA treatment on the dressing percentage of both males and females. When the male birds were not dosed periodically with S. typhimurium in the water fountains and were fed.4% BPA in the diet during 42 to 49 days of age, there was a significant reduction in the dressing percentage as compared to those fed no BPA; this reduction was not observed in males dosed periodically with S. typhimurium and fed.4% BPA from 42 to 49 days of age. When female birds were not dosed periodically with S. typhimurium in the water fountains and were fed.4% BPA from 42 to 49 days of age, there was no 'Myco Curb, Kemin Industries, Des Moines, IA. reduction in the dressing percentage, compared with those fed no buffered propionic acid. However, when the female birds were dosed periodically with S. typhimurium in the water fountains, the addition of.4% BPA, either from 0 to 49 days of age or from 42 to 49 days of age, significantly increased the dressing percentage compared to the birds fed no buffered propionic acid. Table 5 shows the effects produced by the dietary supplementation of BPA and by periodic dosage with S. typhimurium in the water fountains on the number of MPN salmonellae, coliforms, and total organisms in the intestinal contents and on postchill carcasses. The addition of.4% BPA to the diet of broiler chickens for 49 days or from 42 to 49 days had no significant effect on the number of MPN salmonellae or coliforms in the intestine of chickens. However, the addition of.4% BPA for 0 to 49 days or from 42 to 49 days significantly reduced the number of total organisms in the intestine. Rouse et al. (1988) reported that adding a propionic acidbased chemical preservative 12 reduced the number of fecal and intestinal samples that were positive for Salmonella when fed at.5 and 1 % for the last 7 days. Periodic dosage of the birds with S. typhimurium in the water fountains had no significant effect on the level of MPN salmonellae, total coliforms, or total organisms in the intestinal tract (Table 5). There was no interaction between the S. typhimurium dosage of the birds and the dietary addition of BPA on the level of intestinal microbial organisms. Overall, the addition of.4% BPA to the diet for 49 days or from 42 to 49 days produced a significant reduction in the MPN salmonellae on postchill carcasses, but had no effect on the number of coliforms or of total organisms on the postchill carcasses (Table 5). The reduction in MPN salmonellae is due primarily to the lower number in birds inoculated with Salmonella. The periodic dosage of the birds with S. typhimurium in the water fountains caused a significant increase in the MPN salmonellae on postchill carcasses, but had no significant effect on number of coliforms or on the total organisms on the postchill carcasses. The increased levels of salmonellae on the postchill carcasses but not in the intestinal material from inoculated birds suggests either that the organism had not colonized in the small intestine, but perhaps in the cecal tonsils, or that the outside environment of the bird was a more

7 824 IZAT ET AL. TABLE 4. Effects of buffered propionic acid (BPA) 1 and artificial dosage with Salmonella on the growth rate, feed utilization, and carcass characteristics of broiler chickens. Experiment 2 BPA feeding 3 Male weight at 49 days, g (SEM = 24.5) Female weight at 49 days, g (SEM = 26.7) Gain per gram of feed consumed, 0 to 49 days (SEM =.007) Abdominal fat content, males, g (SEM = 2.67) Abdominal fat content, females, g (SEM = 2.82) Dressing percentage, males, % (SEM =.37) Dressing percentage, females, % (SEM =.36) No ,272 2, ,931 1,921 1,937 1, b 633^ 62.3 C 63. iy (,2.9* 64.7 bc 62.9" ? Salmonella inoculation 2 Yes 2,230 2,284 2,208 2,240 1,917 1,938 1,867 1, ab 64.9" 65.0 a 64.7" 64.7 C 65.8 a 65.6 ab 65.4" 2,264 2,278 2,220 1,924 1,930 1, P m P P Row and column means with no common superscripts differ significantly (P<05). P'lColumn means with no common superscripts differ significantly (P<.05). x,y Row means with no common superscripts differ significantly (P<.05). 'Luprosil NC, BASF Wyandotte Corporation, Parsippany, NJ. 2 At 2, 7, 14, 21, 28, 35,42, and 47 days of age, the birds were inoculated with 10 5 to 10 6 cfu per ml of Salmonella typhimurium in the water fountains. 3 Dosage times with.4% BPA in the diet. important source of carcass contamination than the intestinal material. This view is in agreement with USDA findings (U.S. Department of Agriculture, 1988b). Cox etal. (1988) suggested that salmonellae colonization occurs in the ceca and that extremely high challenges are required in order to colonize chicks through the upper gastrointestinal tract. The report by Cox et al. (1988) was published after the present experiments had been conducted; otherwise, cecal material would also have been examined. Inoculating the water may temporarily infect the birds with salmonellae, but perhaps the organism did not readily colonize. The greater problem is that the organism undoubtedly is shed into the litter where it apparently contaminates the outside surface of the birds prior to processing.

8 BUFFERED PROPIONIC ACID AND CARCASS SALMONELLA 825 TABLE 5. Effects of dietary supplementation with a buffered propionic acid (BPA)' and artificial dosage with Salmonella on most probable number (MPN) salmonellae, coliforms, and total organisms in the small intestine and on broiler carcasses, Experiment 2, n = 12 BPA feeding 3 MPN salmonellae (SEM x Coliforms (SEM x Total organisms (SEM x No Intestinal contents Salmonella inoculation Postchill carcass data Yes x No Yes x 16.3 for intestine and 339 for carcass) * (MPN per 10 g) a a a 15.00".74 for intestine and.67 for carcass) x (Logio cfu per g) " 5.08 a 4.40 a 3.85 a.63 for intestine and.34 for carcass) x (Log 10 cfu per g) a b b 8.77" (MPN per 2,500 cm 2 ) 4.5 3,350.0 l,677.2 a l,313.6y 194.7" b x " (Logio cfu per cm 2 ) a a 2.87" (Log^ cfu per cm 2 ) * a a 5.00 x a,b Column means with no common superscripts differ significantly (P<.05). x,y Row means with no common superscripts differ significantly (P<.05). 'Luprosil NC, BASF Wyandotte Corporation, Parsippany, NJ. 2 At 2,7, 14, 21, 28, 35, 42, and 47 days of age, the birds were inoculated with 10 5 to 10 6 cfu per ml of Salmonella typhimurium in the water fountains. 3 Dosage times with.4% BPA in the diet. When the birds were inoculated with S. typhimurium in the water, carcass contamination by salmonellae was decreased significantly by the inclusion of BPA, either continuously or during the last week. The fact that the intestinal levels were not significantly increased while the carcass contamination did increase significantly with inoculation suggests that methods of decreasing contamination during the production of broilers need to be investigated; also, that the microbiological contamination of postchill carcasses may be directly related to the growing environment in addition to processing procedures. The reduction in the total number of bacteria within the intestinal tract due to the feeding of BPA are in agreement with the reports by Vogt et al. (1981, 1982), who observed reductions in the intestinal microflora when organic acids were fed to growing chickens. The results of the present study suggest that feeding the BPA for the entire broiler growing period may not be necessary in order to reduce carcass contamination. The selection of proper usage levels, the effectiveness of other organic acids, or the use of a nonbuffered propionic acid in reducing carcass contamination remain to be determined. ACKNOWLEDGMENTS The present research was supported in part by grants from the BASF Wyandotte Corp., Parsippany, NJ, and by the Southeastern Poultry and Egg Association. REFERENCES Bryan, F. L., J. C. Ayres, and A. A. Kraft, Contributory sources of salmonellae on turkey products. Am. J. Epidemiol. 87: Cox, N. A., J. S. Bailey, L. C. Blankenship, and F. McHan, Colonization dose-50% of baby chicks following oral or intracloacal inoculation of Salmonella typhimurium. Poultry Sci. 67(Suppl. 1):71. (Abstr.) Cox, N. A., J. E. Thomson, and J. S. Bailey, Procedure

9 826 IZAT ET AL. for isolation and identification of Salmonella from poultry carcasses. Agriculture Handbook No United States Department of Agriculture, Agricultural Research Service, Washington, DC. Dixon, R. C, and P. B. Hamilton, Effect of feed ingredients on the antifungal activity of propionic acid. Poultry Sci. 60: Food and Drug Administration, Bacteriological Analytical Manual. 6th ed. Food and Drug Administration, Association of Official Analytical Chemists, Arlington, VA. Miller Publishing Co., 1988a. Feed Additive Compendium. Miller Publishing Co., Minnetonka, MN. Morris, G. K., and J. G. Wells, Salmonella contamination in a poultry-processing plant. Appl. Microbiol. 19: National Research Council, Nutrient Requirements of Domestic Animals. 1. Nutrient Requirements of Poultry. 8th ed. National Academy Press, Washington, DC. Oblinger, J. L., and J. A. Koburger, The most probable number technique. Pages in: Compendium of Methods for the Microbiological Evaluation of Foods. 2nd ed. M. L. Speck, ed. American Public Health Assoc., Inc., Washington, DC. Paster, N., A commercial scale study of the efficacy of propionic acid and calcium propionate as fungistats in poultry feed. Poultry Sci. 58: Patten, J. D., and P. W. Waldroup, The use of organic acids in broiler diets. Poultry Sci. 67: Rigby, C. E., J. R. Pettit, M. F. Baker, A. H. Bentley, M. O. Salomons, and H. Lior, Sources of salmonellae in an uninfected commercially-processed broiler flock. Can. J. Comp. Med. 44: Rouse, J., A. Rolow, and C. E. Nelson, Research Note: Effect of chemical treatment of poultry feed on survival of Salmonella. Poultry Sci. 67: SAS Institute, SAS User's Guide. SAS Inst. Inc., Cary, NC. Stewart, R. G., R. D. Wyatt, and J. Kiker, Effect of commercial antifungal compounds on the performance of broiler chickens. Poultry Sci. 56: U.S. Department of Agriculture, Chicken contamination study leads USDA back to farms. Food Chem. News, pp Van der Wal, P., Salmonella control of feedstuffs by pelleting or acid treatment. Zootechnia, Nov: Vogt, H., S. Matthes, and S. Harnisch, The effect of organic acids in the rations on die performances of broilers and laying hens. Arch. Geflugelkd. 45: Vogt, H., S. Matthes, and S. Harnisch, The effect of organic acids on the performances of broilers-2nd report. Arch. Geflugelkd. 46: Walton, J. R., The effect of zinc bacitracin on the susceptibility of selected gram negative and gram positive bacteria to therapeutic antibiotics. Zemralbl. Veterinaermed. Reihe B, 25: Walton, J. R., The effect of dietary zinc bacitracin on the resistance status of intestinal Escherichia coli and enterococci from broiler chickens. Zentralbl. Veterinaermed. Reihe B, 31:1-8. Walton, J. R., and R. G. Bird, A possible mechanism to explain the growth promotion effect of feed antibiotics in farm animals: Zinc bacitracin induced cell wall damage in Escherichia coli in vitro. Zentralbl. Veterinaermed. Reihe B, 22: Walton, J. R., and J. E. Wheeler, Loss of resistance to the tetracyclines for porcine Escherichia coli in contact with dietary bacitracin methylene disalicylate. J. Vet. Med. B, 34: