12

Journal of Food Protection, Vol. 64, No. 1, 2001, Pages 1216

Copyright q, International Association for Food Protection

Salmonella in the Lairage of Pig Slaughterhouses

M. SWANENBURG, 1,2* H. A. P. URLINGS, 1,2 D. A. KEUZENKAMP, 1 AND J. M. A. SNIJDERS 1

1Department of the Science of Food of Animal Origin, Faculty of Veterinary Medicine, Utrecht University, P.O. Box 80.175, 3508 TD Utrecht,
The Netherlands; and 2Department of Food Science, Institute of Animal Science and Health, P.O. Box 65, 8200 AB Lelystad, The Netherlands

MS 00-76: Received 9 March 2000/Accepted 25 July 2000

ABSTRACT
The purpose of this study was to determine if lairages of pig slaughterhouses can act as a source of contamination of
slaughtered pigs with Salmonella. The prevalence and variety of serotypes of Salmonella in the lairages of two pig slaugh-
terhouses were determined, and the ef cacy of the usual cleaning and disinfection on the presence of Salmonella was estimated.
Lairages of two pig slaughterhouses were sampled three times when pigs were present. Furthermore, these lairages were
sampled after the usual cleaning and disinfection, whereas the lairage of one slaughterhouse was sampled an additional time
after improved cleaning and disinfection. Samples were collected by swabbing oor and wall surfaces and collecting the
residing uids on the oor throughout the lairage. Salmonella was isolated in 70 to 90% of the samples when pigs were
present. The usual cleaning and disinfection reduced the level of contamination with Salmonella to 25% positive samples,
whereas improved cleaning and disinfection reduced this level to 10% positive samples. It is concluded that the waiting period
in the lairage of at least 2 h contains a substantial risk for slaughter pigs to become infected with Salmonella, especially for
pigs originating from Salmonella-free herds. The usual cleaning and disinfection of the lairage were not suf cient to eliminate
this risk, whereas an improved procedure for cleaning and disinfection still was unsatisfactory.

Salmonella is an important cause of bacterial food-
borne infections. Each year approximately 450 per 100,000
persons in The Netherlands suffer from salmonellosis, and
it is estimated that about 15% of these salmonellosis cases
originate from pork (2).
Pigs can become infected with Salmonella either di-
rectly via uptake of infected feed or contact with another
infected pig or indirectly through a contaminated environ-
ment, such as surfaces, equipment, or persons. It is com-
monly thought that the lairage, where pigs are held from 2
to 8 h before they are slaughtered, can act as an important
source of bacterial infections. However, only few studies
have been carried out to determine the prevalence and the
variety of serotypes of Salmonella in pig lairages or to de-
termine the effect of cleaning and disinfection of the lairage
on the prevalence of Salmonella. Weber (12) found 16 to
50% Salmonella-positive samples in lairages of German
slaughterhouses after pigs had been in the lairage and 0 to
33% Salmonella-positive samples after cleaning and disin-
fection. Lazaro et al. (6) found 20% Salmonella-positive
samples in the lairage of a pig slaughterhouse in Brazil.
Since Fedorka-Cray et al. (5) and Blaha et al. (3) found
that Salmonella can spread through the body of the pig very
rapidly after uptake, it would be reasonable to assume that
pigs can easily become generally infected with Salmonella
in the lairage of a slaughterhouse before slaughter. This has
already been shown by Williams and Newell (13), who
found the same Salmonella serotypes in the lairage before
the pigs entered as in rectal and cecal swabs of the pigs
after their waiting period in the lairage.
* Author for correspondence. Tel: 131 320 238176; Fax: 131 320
238961; E-mail: M.Swanenburg@id.wag-ur.nl.
These experiments were carried out in the framework
of a research project on Salmonella in slaughter pigs in The
Netherlands. The purpose of the study was to determine if
lairages of pig slaughterhouses can act as a source for con-
tamination with Salmonella of slaughtered pigs. Swab sam-
ples and water samples were collected from the oors and
walls of the lairages of two slaughterhouses at two time
points, when pigs were present and after cleaning and dis-
infection. We determined the prevalence of Salmonella in
all samples, after which the isolated Salmonella strains were
genotyped by enterobacterial repetitive intergenic consen-
sus polymerase chain reaction. Furthermore, in every sam-
ple we estimated the total number of CFUs of aerobic bac-
teria and Enterobacteriaceae.
MATERIALS AND METHODS
Lairages of two Dutch pig slaughterhouses (slaughterhouse
1 and 2), with a capacity of slaughtering 800 and 500 pigs per h,
respectively, were each sampled three times during presence of
pigs (sampling days 1, 2, and 3) and one time after the usual
cleaning and disinfection (sampling day 4). The lairage of slaugh-
terhouse 1 was sampled an additional time after improved clean-
ing and disinfection (sampling day 5). Both lairages had basically
the same size and were constructed in the same way, which im-
plied that logistic procedures were comparable. Floors and walls
of the lairages were made of concrete.
Slaughtering occurred from Monday to Friday. Cleaning and
disinfection were carried out on Saturdays in the usual way for
both slaughterhouses. Sampling during presence of pigs was car-
ried out at three different days per slaughterhouse within a period
of 2 weeks, each day at the same time (in slaughterhouse 1: Tues-
day, rst week; Monday and Thursday, second week; in slaugh-
terhouse 2: Monday and Thursday, rst week; Tuesday, second
week). Sampling of the cleaned lairage was carried out on Sun-
J. Food Prot., Vol. 64, No. 1 SALMONELLA IN THE LAIRAGE OF PIG SLAUGHTERHOUSES
FIGURE 1. Schematic oor diagram of the sampled lairages. (A)
Entrance platforms, (B) corridor between entrance platforms and
pens, (C) waiting pens, (D) corridor to restrainer, and () walk-
ing direction of pigs.
days (in slaughterhouse 1: 2 months after rst samplings; in
slaughterhouse 2: 3 weeks after rst sampling). In slaughterhouse
1, cleaning started with water to ush away the dirt. Thereafter, a
cleaning solution (Kleencare CF6202, 5% active alkaline chloride
combination) was used (contact time at least 1 h), which was
washed away with water. Thereafter, the lairage was disinfected
with Kleencare DS6601 (5%, contains didecyldimethylammoni-
umchloride), which also was washed away with water. The clean-
ing and disinfection solutions were approximately 308C. In
slaughterhouse 2, cleaning was carried out with a lot of water. A
cleaning or disinfection solution was not used. Both slaughter-
houses used potable water (high pressure) for cleaning. The im-
proved cleaning in the lairage of slaughterhouse 1 was carried out
in the same way as the normal cleaning and disinfection, but
additional to the normal procedures the cleaning was visually
checked at the critical control points (drains, holes in the oor,
rough surfaces of oor and wall, corners) by an experienced vet-
erinary hygienist. Mistakes in cleaning were corrected, followed
by disinfection.
During the samplings in the presence of pigs, 25 swab sam-
ples and 20 water samples (residing uids on the oor: sprinkling
water, feces, urine) were collected per sampling day, except for
sampling day 1 at slaughterhouse 1 when 22 swab samples and
19 water samples were collected. The swab method was described
by Van den Elzen and Snijders (9). Per sample a surface of 50 by
50 cm was swabbed with both sides of the swab. Water samples
of 100 ml were collected with a sterile pipette. Samples were
collected throughout the whole lairage at the following locations:
the oor of the entrance platforms, the wall of the entrance plat-
forms, the oor of the corridor between entrance platforms and
pens, the oor of the pens, and the oor of the corridor leading
to the restrainer. Figure 1 shows a oor diagram of the lairages
to indicate sampling locations.
During the samplings after cleaning and disinfection, water
samples could not always be collected, because there was not
enough water present on the oors. Therefore, a different number
of samples was collected: 40 swabs and 10 water samples on
sampling day 4 in slaughterhouse 1 and 50 swabs on sampling
day 5 in slaughterhouse 1 and on sampling day 4 in slaughter-
house 2. After sampling, all samples were transported to the lab-
oratory at a temperature of 48C. Salmonella isolation procedures
started 1 to 2 h after sampling.
Pre-enrichment Salmonella isolation procedure. Fifty mil-
liliters of buffered peptone water (with 0.1% Tween 80) was added
13
to the swab samples. Then the swabs were put in a stomacher for
2 min, after which they were squeezed out and removed from the
stomacher bag. Ten milliliters of concentrated buffered peptone wa-
ter was added to the water samples. Then the samples were gently
stirred. All samples were incubated at 378C for 18 to 24 h.
Selective enrichment Salmonella isolation procedure. The
pre-enrichment broth was mixed, and 0.1 ml was transferred to
9.9 ml of prewarmed Rappaport-Vassiliadis enrichment broth (Ox-
oid CM 669). This was incubated in a water bath of 428C for 24
to 48 h.
Selective and elective growth. A loop of material from the
Rappaport-Vassiliadis broth was transferred and spread onto the
surface of a brilliant green agar plate (Oxoid CM 329) so that
isolated colonies could develop. The plates were incubated in in-
verted position at 378C for 18 to 24 h. After incubation the plates
were checked for growth of typical Salmonella colonies (lightly
transparent, reddish color). When no typical colonies were found
after 24 h of incubation, a loop of Rappaport-Vassiliadis (48 h
incubated) was plated out again on a brilliant green agar plate.
Con rmation. When typical colonies had grown on the bril-
liant green agar plate, up to 5 colonies per plate were transferred
and inoculated on triple sugar iron agar (Oxoid CM 277), which
was incubated at 378C for 18 to 24 h. When a positive reaction
occurred after incubation (black color of agar, gas), bacterial ma-
terial from the triple sugar iron agar was agglutinated with poly-
valent Salmonella antiserum (Pro-lab Diagnostics, Neston, Wirral,
UK). Isolated Salmonella strains were typed with monovalent spe-
ci c O antisera (Pro-lab Diagnostics) to identify O antigens.
The Salmonella strains were genotyped with a standardized
enterobacterial repetitive intergenic consensus polymerase chain
reaction, rst described by Versalovic et al. (11) and amended by
Swanenburg et al. (8). As described by Van Lith and Aarts (10)
and Swanenburg et al. (8), genotypes could be related to serotypes
(each serotype could be divided into one or more genotypes; each
genotype always had the same serotype). After genotyping, a rep-
resentative number of strains was serotyped, according to the
Kauffmann-White scheme, at the Salmonella reference center
RIVM (National Institute of Public Health and the Environment)
at Bilthoven, The Netherlands. Serotyping was carried out with
Salmonella antisera that were prepared at the Rijksinstituut voor
Volksgezondheid en Milieuhygiene (RIVM).
Furthermore, to measure the hygienic status of the lairage,
the total numbers of CFUs of aerobic bacteria and Enterobacte-
riaceae (per cm2 or ml) were determined by diluting the samples
and growing them on plate count agar, which was incubated 3
days at 308C, and violet red bile glucose agar (poured plates,
Oxoid CM 485), which was incubated 24 h at 378C. The statistical
detection level (the lowest number of Enterobacteriaceae that
could possibly be detected: at least 1 colony on plate) was 21.4
in log CFU/cm2 for the swab samples and 20.05 in log CFU/ml
for the water samples.
The results were evaluated statistically by the chi-square test
(comparison of numbers of Salmonella-positive samples) and by
an independent samples t test (two groups) or one-way analysis
of variance (more than two groups), followed by a least signi cant
difference test, when signi cance was established (comparison of
numbers of total CFUs of aerobic bacteria and Enterobacteri-
acea), using SPSS software, version 7.0 (1). Signi cance was de-
termined with P , 0.05.
RESULTS
Salmonella could be isolated from all locations of the
lairages in both slaughterhouses when pigs were present.
14 SWANENBURG ET AL. J. Food Prot., Vol. 64, No. 1

CFUs of aerobic bacteria in the water samples differed sig-

FIGURE 2. Percentages of Salmonella-positive swab and water
samples in the lairages of two pig slaughterhouses. Days 1, 2,
and 3: during presence of pigs. Day 4: after usual cleaning (and
disinfection). Day 5: after improved cleaning and disinfection. Sl.
1, slaughterhouse 1; Sl. 2, slaughterhouse 2.
Figure 2 shows the percentages of Salmonella-positive
swab and water samples on each sampling day in both
slaughterhouses. Salmonella was isolated from 89% of the
swab samples and 95% of the water samples (mean per-
centage of 3 sampling days) when pigs were present in
slaughterhouse 1, whereas Salmonella was isolated from
79% of the swab samples and 62% of the water samples
when pigs were present in slaughterhouse 2. The number
of Salmonella-positive samples was not signi cantly in u-
enced by sampling day or slaughterhouse, except for sam-
pling day 1 at slaughterhouse 2, when the number of Sal-
monella-positive samples was signi cantly lower (P ,
0.05) than on the other sampling days. The number of Sal-
monella-positive samples was also not signi cantly in u-
enced by the kind of sample collected (swab or water sam-
ple).
In both slaughterhouses, the numbers of CFUs of aer-
obic bacteria were around 6 in log units per cm2 and num-
bers of CFUs of Enterobacteriaceae were around 3 to 4 in
log units per cm2 (Table 1) when pigs were present. Slaugh-
terhouse and sampling day had no signi cant effect on the
number of CFUs of Enterobacteriaceae in either swabs or
water samples, but the number of CFUs of aerobic bacteria
in both swab and water samples was signi cantly higher in
slaughterhouse 2 (P 5 0.009). The mean 6 SD number of
ni cantly between sampling day 2 (7.54 6 0.47) and 3
(7.08 6 0.48) in slaughterhouse 1 (P 5 0.002) and between
sampling day 1 (7.37 6 0.44) and 2 (7.72 6 0.47) in
slaughterhouse 2 (P 5 0.017).
After normal cleaning and disinfection (only cleaning
in slaughterhouse 2), the lairages still contained rests of
feces, especially around the water outlets. Also some rest
water of cleaning and disinfection was present on the oor
of both lairages. The results showed that neither cleaning
alone nor cleaning followed by disinfection was able to
eliminate all Salmonella, although the percentage of Sal-
monella-positive samples in the lairages of both slaughter-
houses was signi cantly decreased. Salmonella was isolated
from 25% of the swab samples and none of the water sam-
ples of slaughterhouse 1 and 24% of the swab samples of
slaughterhouse 2. This implies that the lairages still were
highly contaminated with Salmonella. After improved
cleaning and disinfection of the lairage at slaughterhouse 1,
Salmonella was isolated from 10% of the swab samples.
This result showed that it was possible to clean the lairage
better than is usually done but that it will be dif cult to
eliminate all salmonellae from the lairage.
After cleaning and disinfection, numbers of CFUs of
aerobic bacteria and Enterobacteriaceae were signi cantly
reduced (P , 0.05) in the lairages of both slaughterhouses
(Table 1). After improved cleaning and disinfection in
slaughterhouse 1, the number of Enterobacteriaceae was
lower than after normal cleaning and disinfection (P 5
0.051).
Genotyping of Salmonella isolates resulted in 13 dif-
ferent genotypes isolated from the lairage of slaughterhouse
1 and 10 different genotypes isolated from the lairage of
slaughterhouse 2. In total, 16 different genotypes were iso-
lated. Table 2 shows the different genotypes and corre-
sponding serotypes that were isolated in both slaughter-
houses. The most common serotype in the lairage of
slaughterhouse 1 during the presence of pigs was Salmo-
nella Panama, which was not isolated after cleaning any-
more. Another common serotype, isolated each sampling
day in this slaughterhouse, was Salmonella Typhimurium.
The most common serotype in the lairage of slaughterhouse

TABLE 1. Average numbers of colony-forming units of aerobic bacteria and Enterobacteriaceae in swabs and water samples collected
in the lairages of two pig slaughterhouses
Log CFU/cm2 Log CFU/cm2
or ml of or ml of
Slaughter- Sampling No. of aerobic bacteria Enterobacteriaceae
house Situation technique samples 6 SD 6 SD

1 Pigs present Swabs 72 5.7 6 0.9 3.5 6 1.2

Water 59 7.3 6 0.5 5.0 6 0.6
After cleaning Swabs 40 3.1 6 1.4 20.4 6 1.6
Water 10 2.1 6 1.1 0.0 6 0.2
After improved Swabs 50 Not done 20.7 6 1.2
cleaning
2 Pigs present Swabs 75 6.0 6 0.5 3.6 6 0.8
Water 60 7.5 6 0.5 5.1 6 0.6
After cleaning Swabs 50 4.5 6 0.9 20.1 6 1.5
J. Food Prot., Vol. 64, No. 1 SALMONELLA IN THE LAIRAGE OF PIG SLAUGHTERHOUSES 15

TABLE 2. Genotypes and corresponding serotypes of Salmonella in two pig slaughterhousesa

Distribution of Salmonella genotypes in contaminated samples (%)b

Slaughterhouse 1 Slaughterhouse 2

Genotype/serotype D1 D2 D3 D4 D5 D1 D2 D3 D4

B-A/Typhimurium 39 40 31 90 50 95 77 80 83
B-C/Derby 3 17 12 17
B-E/Typhimurium 3
B-G/Brandenburg 3 3 24 20
B-J/Bredeney 5
B-L/Brandenburg 10 20
B-S/Senftenberg 8
C-A/Infantis 18 10 20 10
C-B/Livingstone 3 15 2 11
C-C/Bovismorbi cans 8
C-G/Goldcoast 20
C-F/Mbandaka 7 5
D-A/Panama 68 58 60 9
E-A/Give 29
E-B/London 11 7 10 10 8
E-E/Elisabethville 29
No. of positive samples 38/41 40/45 42/45 10/50 5/50 20/45 35/45 41/45 12/50
a Days 1, 2, and 3: during presence of pigs. Day 4: after usual cleaning (and disinfection). Day 5: after improved cleaning and disinfection.
b The total of one day can be more than 100% due to the presence of several genotypes in a single sample.

2, as well as in the presence of pigs after cleaning, was
Salmonella Typhimurium. Seven genotypes were isolated
from lairages of both slaughterhouses, whereas six geno-
types were exclusively isolated from the lairage of slaugh-
terhouse 1 and three genotypes were exclusively isolated
from the lairage of slaughterhouse 2.
DISCUSSION
These experiments show that almost the whole lairage
was contaminated with several strains of Salmonella when
pigs were present. Salmonella could be isolated from al-
most every sample collected in the lairages of the two
slaughterhouses, and up to eight different serotypes could
be isolated per sampling day. Numbers of CFUs of aerobic
bacteria were around 6 log units per cm2, and numbers of
CFUs of Enterobacteriaceae were around 3 to 4 log units
per cm2 . On sampling day 1 at slaughterhouse 2, Salmo-
nella was isolated from signi cantly fewer samples than on
the other sampling days in both slaughterhouses. We do not
have an explanation for this observation.
Only a few studies have been conducted on the Sal-
monella contamination of a lairage. Weber (12) found 16
to 50% Salmonella-positive swab samples in lairages of
German slaughterhouses after pigs had been in and 0 to
33% Salmonella-positive swab samples after cleaning and
disinfection. Lazaro et al. (6) found 20% Salmonella-posi-
tive swab samples in the lairage of a pig slaughterhouse in
Brazil. Furthermore, experiments have been done in which
the effect of the waiting time in the lairage on the number
of pigs infected with Salmonella was tested (4, 7, 12), but
the conclusions of these experiments were not consistent.
Morgan et al. (7) concluded that the number of pigs car-
rying Salmonella in the ceca increased signi cantly in pigs
held for extended periods (2 or 3 days) in lairage. Weber
(12) concluded that more Salmonella could be isolated from
pigs that were kept in lairage for a waiting period of 18 to
20 h compared with pigs that were slaughtered directly after
arrival at the slaughterhouse. In contrast, Craven and Hurst
(4) concluded that fewer pigs had Salmonella in the ceca
after 2 or 3 days in lairage.
The results of our experiments strongly suggest that the
lairage, as a result of the high level of contamination with
Salmonella, can have a signi cant effect on the number of
Salmonella-infected pigs at slaughter. Exploring the envi-
ronment is part of the normal behavior pattern of pigs. This
exploring behavior consists of snif ng, rooting, and drink-
ing, which offers a great possibility for the pigs to infect
themselves with Salmonella, after which the infection can
spread rapidly through the whole pig. This suggestion is
supported by the results of the study by Fedorka-Cray et
al. (5), who concluded from their experiments that Salmo-
nella Typhimurium could be detected from tonsil, cecum,
colon, and thoracic tissues 3 h after intranasally inoculation
of pigs, and by the results of the study by Blaha et al. (3),
who concluded that orally given Salmonella Typhimurium
could be detected in ileocecal lymph nodes 4 h after Sal-
monella was administered to the pigs and in tonsils already
30 min after administration.
Furthermore, this study shows that the normal
cleaning and disinfection procedures of the lairages are un-
able to eliminate Salmonella. Twenty- ve percent of the
swab samples still contained Salmonella after cleaning and
disinfection, indicating a relatively high level of contami-
nation. This is supported by the fact that still three (slaugh-
terhouse 1) or four (slaughterhouse 2) different serotypes
were isolated after cleaning and disinfection. Moreover,
even improved cleaning and disinfection were not suf cient
to eliminate all Salmonella, although the number of positive
16 SWANENBURG ET AL.
samples decreased. The main factor that causes the dif -
culty in cleaning and disinfection of the lairage is the way
it is constructed. The oor and wall surfaces are very un-
even and contain many small and bigger holes. Cleaning
and disinfection solutions penetrate badly in these holes.
Furthermore, water and rests of feces can easily stay behind
in the holes after the cleaning and disinfection solutions are
washed away, which is one of the reasons why even im-
proved cleaning and disinfection were not able to eliminate
all Salmonella from the lairage of slaughterhouse 1. The
observation that there was no difference between the num-
bers of Salmonella-positive samples after cleaning and dis-
infection in slaughterhouse 1, compared with only cleaning
in slaughterhouse 2, shows that disinfection will have no
obvious effect if cleaning cannot be carried out completely.
The presence of feces after the regular cleaning processes
also strongly suggests that cleaning was not carried out
properly.
The Salmonella ora of the lairages seemed to consist
of a residential ora of Salmonella genotypes that were
present consistently and a transient ora of Salmonella ge-
notypes that were only present on 1 or 2 sampling days.
Genotype B-A and B-G constituted the largest part of the
residential ora in the lairage of slaughterhouse 1, whereas
genotype C-B and especially D-A also constituted an im-
portant part of the residential ora but were not isolated
after cleaning anymore (Table 2). Genotype B-A and B-C
constituted the main part of the residential ora in the lair-
age of slaughterhouse 2. Each lairage seemed to have its
own speci c residential ora, which will probably be con-
tinuously refreshed by Salmonella-infected pigs passing the
lairage. Furthermore, it might be possible that some Sal-
monella serotypes are more resistant against the applied
cleaning and disinfection compared with other serotypes,
which can be illustrated by genotype B-A in the lairage of
both slaughterhouses, as an example for a resistant Sal-
monella type, and by genotype D-A in the lairage of slaugh-
terhouse 1, as an example for a nonresistant Salmonella
type (Table 2).
From the results of the experiments, we conclude that
the waiting period in the lairage of a slaughterhouse con-
tains a substantial risk for pigs to become infected with
Salmonella, especially for pigs originating from Salmonel-
la-free herds. Regular cleaning and disinfection are insuf-
cient to eliminate all Salmonella in these lairages. Con-
J. Food Prot., Vol. 64, No. 1
icting interests concerning animal welfare (rough oors)
and hygiene (smooth oors) are a basic dilemma to improve
this situation.
ACKNOWLEDGMENTS
We thank the Dutch Product Boards for Livestock, Meat and Eggs
for nancing the project. We also thank the staff and personnel of the two
slaughterhouses for their willingness to participate in the study and their
cooperation during the experiments.
REFERENCES
1. Anonymous. 1996. SPSS Base 7.0 for Windows users guide. SPSS
Inc., Chicago.
2. Berends, B. R., F. van Knapen, D. A. A. Mossel, S. A. Burt, and J.
M. A. Snijders. 1998. Impact on human health of Salmonella on
pork in The Netherlands and the anticipated effects of some currently
proposed control strategies. Int. J. Food Microbiol. 44:219229.
3. Blaha, Th., G. Solano-Aguilar, and C. Pijoan. 1997. The early col-
onization pattern of S. typhimurium in pigs after oral intake. Pro-
ceedings of the Second International Symposium of Epidemiology
and Control of Salmonella in Pork. Danish Agricultural and Veteri-
nary Research Council, European Commission, Copenhagen.
4. Craven, J. A., and D. B. Hurst. 1982. The effect of time in lairage
on the frequency of Salmonella infection in slaughtered pigs. J. Hyg.
88:107111.
5. Fedorka-Cray, P. J., L. Collins-Kelley, T. J. Stabel, J. T. Gray, and
J. A. Laufer. 1995. Alternate routes of invasion may affect patho-
genesis of Salmonella typhimurium in swine. Infect. Immun. 63:
26582664.
6. Lazaro, N. S., A. Tibana, and E. Hofer. 1997. Salmonella spp. in
healthy swine and in abattoir environments in Brazil. J. Food Prot.
60:10291033.
7. Morgan, I. R., F. L. Krautil, and J. A. Craven. 1987. Effect of time
in lairage on caecal and carcass Salmonella contamination of slaugh-
ter pigs. Epidemiol. Infect. 98:323330.
8. Swanenburg, M., H. A. P. Urlings, D. A. Keuzenkamp, and J. M. A.
Snijders. 1998. Validation of ERIC PCR as a tool in epidemiologic
research of Salmonella in slaughter pigs. J. Ind. Microbiol. Biotech-
nol. 21:141144.
9. Van den Elzen, A. M. G., and J. M. A. Snijders. 1993. Critical points
in meat production lines regarding the introduction of Listeria mon-
ocytogenes. Vet. Q. 15:143145.
10. Van Lith, L. A. J. T., and H. J. M. Aarts. 1994. Polymerase chain
reaction identi cation of Salmonella serotypes. Lett. Appl. Micro-
biol. 19:273276.
11. Versalovic, J., T. Koeuth, and J. R. Lupski. 1991. Distribution of
repetitive DNA sequences in eubacteria and application to nger-
printing of bacterial genomes. Nucl. Acids Res. 19:68236831.
12. Weber, E. 1996. Epidemiologische Untersuchungen zum Eintrag von
Salmonellen aus Schweinemastbestanden in die Lebensmittelkette.
Inaugural-Dissertation. Tierarztliche Hochschule Hannover, Hanno-
ver, Germany.
13. Williams, L. P., and K. W. Newell. 1968. Sources of salmonellas in
market swine. J. Hyg. Camb. 66:281293.