British Poultry Science (2000) 41: 575–583

Investigation of hygiene aspects during air chilling of poultry

carcases using a model rig
V.M. ALLEN, C.H. BURTON1 , J.E.L. CORRY, G.C. MEAD2 AND D.B. TINKER1
University of Bristol, Langford, 1 Silsoe Research Institute, Silsoe, and 2 Royal Veterinary College, University of London,
Potters Bar, England

Abstract 1. An experimental rig, designed and built to simulate conditions found in commercial poultry
chilling systems, was used to investigate the effects of varying air temperature and chilling duration, and the
effect of chlorinated water sprays, on the microbial load present on the skin and in the body cavity of
freshly eviscerated poultry carcases; deep muscle and skin temperatures were monitored during chilling at
three different temperatures.
2. During dry chilling for 2 h, total viable microbe counts (TVC) and counts of coliforms and pseu-
domonads from the body cavity fell by between half and one log unit; smaller reductions were observed in
samples from the breast skin.
3. The situation changed when chlorinated water sprays (50, 100 or 250 ppm available chlorine) were
applied for the first hour of chilling; spraying carcases enhanced the reduction in numbers on the skin; the
effect was most pronounced with 250 ppm chlorine; conversely, in the body cavity, the general effects of
sprays was to increase contamination by up to one log unit.
4. There was no evidence that sprays increased the rate of chilling.
5. When carcases were held overnight in the rig at 11°C after chilling, microbe counts on dry-chilled
carcases remained stable, but increased on carcases that had been sprayed with chlorinated water.

INTRODUCTION effect in relation to counts from the abdominal cavity

In the UK, air chilling has become the preferred
method of cooling poultry carcases because of the
increasing popularity of fresh rather than frozen
poultry meat. The system is often modified by the
incorporation of water sprays in the 1st stage for
the primary purpose of providing evaporative
chilling. This has the additional benefit of
minimising weight loss during cooling.
Some studies on air chilling have shown a
reduction in microbial contamination of carcases
by the end of the process (Berner et al., 1969)
although Knoop et al. (1971) found that the reduc-
tion was confined to psychrotrophic bacteria.
Following the work of Lahellec et al. (1973), it is
often assumed that air chilling has neither a qualita-
tive nor quantitative effect on the carcase micro-
flora. In an EEC study involving 8 Member States
(Anon 1978), air-chilling produced no change in
extent of carcase contamination and no effect on
shelf-life in comparison with water-chilling, although
only neck skins were examined. With regard to
evaporative chilling, Stephan and Fehlhaber (1994)
reported no change in the microbial counts from
the skin when samples were macerated prior to
examination. Graw et al. (1997), also using sample
maceration, found only a 0·5 log1 0 reduction in the
microbial load on the skin but observed no such
(rinse sampling).
In recent studies at commercial plants during
normal production, little change was found in
microbial numbers on skin when air or evaporative
chilling was used (Allen et al., 2000). The volume of
water sprayed on to each carcase was not closely
controlled and large numbers of pseudomonads
were usually present in the vicinity of the sprays. In
the absence of sprays, there was a progressive reduc-
tion in microbial numbers recovered from the body
cavity during the chilling process.
In some commercial plants the chillers are used
to store carcases overnight whilst the main plant is
closed down for cleaning. Whilst most of the stored
carcases will be held at an acceptable temperature
during this period, a few in peripheral regions (such
as those in the relatively warmer entry areas) may
cool relatively slowly. In order to investigate possible
hazards related to this practice, the effect of storing
carcases for prolonged periods at warmer
temperatures (3° to 12°C) needed to be investigated.
A thorough study of the effect on microbial
counts of chilling carcases under sub-optimal condi-
tions cannot be readily carried out in commercial
units because this could compromise food safety
and/or shelf-life. To enable a broader and more
objective study into the effect of higher cooling
temperatures or modifying the operation of water

Correspondence to: Dr V.M. Allen, Division of Food Animal Science, Department of Clinical Veterinary Science, University of Bristol,
Langford, Bristol BS40 5DT, UK. Tel +44–01179 289 430. Fax +44 01934–853145. E-mail: viv.allen@bris.ac.uk
Accepted For publication 28th June 2000.
ISSN 0007-1668(print) /ISSN 1466-1799(online)/00/050575-0 9 © 2000, British Poultry Science Ltd
DOI: 10.1080/0007166002000910 8
576
sprays, an experimental rig was constructed to
simulate closely commercial chilling conditions while
allowing a wider range of environments to be
investigated under controlled conditions. This rig
has also been used to investigate cross-contamination
of pathogens (Mead et al. 2000).
MATERIALS AND METHODS
Design of the experimental rig
A rig was constructed based on the operation of
the continuous chillers used in commercial
processing plants. Warm, freshly eviscerated carcases
were required, so the equipment had to be located
next to a poultry processing plant but separate from
it to avoid potential hazards to food safety. The
basic design was thus an independent mobile unit.
The rig (Figure 1) consisted of a metal frame
fitted with 10 shackles of the type used in
commercial processing arranged in two rows of 5.
The layout of the shackles (150 mm apart and
300 mm between rows) followed the typical geometry
of a commercial chiller, where an air-duct passes
between alternate rows. The duct used in the rig
was made from 100 mm diameter plastic pipe with
two 10 mm longitudinal slots cut in it, angling an
air jet at 15 degrees off the vertical towards the
carcases, simulating the observed commercial opera-
tion. Likewise, the duct was positioned above the
Figure 1. Experimental rig used to cool batches of 10 carcases in a controlled environment. The air duct and water supply to the spray boom can be seen in the
middle of the rig. The apparatus was set up in a refrigerated van.
V.M. ALLEN ET AL.
carcases, allowing a 100 mm gap between the slot
and the entrance to the body cavity of the carcase. A
centrifugal blower (rated at 370 W and 509 m3 /h free
air delivery) was used to drive air through the duct
and slots. The rig was designed to deliver air at 10 m/s
at the slots based on the slot size and the specification
of the blower chosen. Furthermore, by strictly
following the geometry of the commercial shackle line
(which delivers 10 m/s at the slot), 1 m/s around the
carcase itself was expected and confirmed by measure-
ments made in commercial units using a sonic
anemometer. The performance of the rig was veri-
fied similarly by detailed measurements with a hot
wire anemometer during commissioning. The same
device was used to periodically check the airflow
throughout the experimental programme.
The provision of sprays simulated conditions
in the pre-chiller stage in a typical commercial air
chiller. A single boom was set up with 10 fine nozzles
fitted to direct conical sprays at each of the carcases.
The main water reservoir was a 100 litre vessel filled
from the factory supply (~ 10°C with 50 ppm
chlorine). A high pressure centrifugal pump
circulated the water through a loop, returning flow
to the reservoir via a back pressure valve which
enabled a 3 bar working pressure to be sustained. A
variable flow of water was taken from this loop to
supply the spray boom, typically set up to produce
a flow-rate of 500 ml/min per nozzle.
 HYGIENE IN AIR-CHILLING OF POULTRY 577

The complete apparatus was set up inside a
small refrigerated van with an operational range
between –10° and +15°C. Air temperature could
be controlled within limits of ~ 1·5°C
Experimental programme
For each trial, 12 warm carcases, of similar weight,
were removed from the production line immediately
prior to chilling and within 5 min were conveyed to
the test rig in a sterile, thermally-insulated box. Ten
were placed in the rig and the cooling trial started
immediately; the two remaining carcases were
retained as controls, representing the initial (zero-
time) samples.
Nine of the carcases in the rig were examined
microbiologically (details below). The 10th carcase
in each experiment was used to monitor
temperatures (see temperature measurement, below).
It was initially intended to run the rig at 3 nominal
temperatures, –3°, 3° and 11°C, representing
standard chilling, pre-chilling and poor chilling
respectively. For each of these temperatures, 4 spray
options were applied: no spray, and spraying with
chlorinated water at 50, 100 and 250 ppm making
a total of 12 trials (see Table). However, the trial
planned at –3°C with water sprays containing
50 ppm chlorine was aborted due to freezing of the
water sprays. This problem was avoided in
subsequent trials by raising the temperature to –1·1°
and –1·5°C but it was not possible to repeat the
trial at 50 ppm chlorine, hence only 11 trials were
carried out. Spray was applied intermittently for a
period of 60 s on 5 occasions during the first hour
(0, 15, 30, 45 and 60 min) applying a total of 2000
to 2500 ml per carcase.
Sets of 3 carcases were briefly removed from
the rig for microbiological examination after 30, 60
and 120 min and then replaced in the rig until the
end of the experiment. The carcases in each set
were, as far as possible, 3 shackles apart in order to
allow for any possible variations such as air move-
ment, spray dispersal and proximity of other carcases
within the rig. All trials except two were concluded
after 120 min, the typical duration of chilling in
continuous commercial units. The trials at a nominal
11°C were extended to 12 h, to represent the worst
conditions that might occur in a commercial unit
(that is, overnight storage near the entrance of a
chiller room). In these trials the sampling was
repeated on the carcases examined after 120 min
(see general methods).
At the end of each trial the carcases were
removed and the rig and vehicle interior were
thoroughly cleaned following a set procedure. The
chiller unit was switched off, the doors opened and
all equipment was hosed down with water, paying
especial attention to the rig itself and the shackles
that held the carcases, followed by spraying chlorin-
ated water (10,000 ppm) on to all surfaces. Water
drained easily because the chiller van was sited on a
gentle slope. The rig and vehicle interior were left
to dry and ventilate for at least 30 min before cooling
resumed in preparation for the next trial. The
ventilation allowed chlorine concentrations in the
air to fall below 10 ppm. Opportunity for recontami-
nation was minimal because all carcases had been
removed at least 20 m from the mobile chill room,
and as soon as ventilation was complete the doors
were closed.
General methods of microbiological
examination
Sample collection
Samples were collected (where appropriate) into
chilled diluent and all were transported in ice to the
laboratory where they were held at 1°C until
examination the next day. Two types of sample were
taken from the carcase:
1. Skin, representing the exterior surface that is most
exposed to the surrounding environment—
approximately 10 g were removed aseptically from
the breast.
2. Body cavity, representing internal surfaces which
are largely protected from the chiller environ-
ment and are most likely to be contaminated as a
result of the evisceration process – a large cotton
wool swab with wooden shaft (MW104J, Medical
Wire, Corsham, Wilts, UK) was wiped all round
the body cavity via the vent and the tip broken

Table 1. Temperatures (°C) recorded during trials.

No sprays 250 ppm chlorine 100 ppm chlorine 50 ppm chlorine

Trial 1 Trial 2 Trial 3 Trial 4 Trial 5 Trial 6 Trial 7 Trial 8 Trial 9 Trial 10 Trial 11

Nominal air temperature ± 3 3 11 ± 3 3 11 ± 3 3 11 3 11

Actual air temperature ± 2´8 3 12´3 ± 1´1 2´9 9´7 ± 1´4 2´8 9´7 4´8 12´2
One standard deviation 1´4 0´7 0´4 1´7 0´6 0´5 1´7 0´9 0´6 1´5 0´5
Cooling time deep tissue (min)
from 20° to 10°C 24 33 * 31 29 94 33 36 * 35 *
from 10° to 4°C 45 * * 35 58 * 25 68 * * *
Cooling time skin (min)
from 20° to 10°C 17 22 * 38 18 38 32 21 * 38 *
from 10° to 4°C 35 76 * 36 52 * 39 87 * * *

An asterisk indicates that the cooling temperature was not reached at the end of the 2 h period.
578 V.M. ALLEN ET AL.
off into 10 ml Maximum Recovery Diluent
(MRD: Oxoid CM733, Basingstoke, UK) at 0°C,
containing glass beads. Results were expressed as
cfu (colony forming units) per body cavity. In the
2 trials which were extended overnight, only half
of the body cavity was sampled at 120 min and
the remaining half sampled at 12 h. These results
were doubled to give values per whole body cavity.
Sample examination
Organisms were released from the swabs by agitating
the bottle containing the swab on a vortex mixer
for ~ 30 s. Skin samples were weighed and then
homogenised for 1 min in 50 ml MRD, using a stom-
acher (Seward, London). Decimal dilutions were
prepared in MRD and appropriate dilutions surface-
plated on the media listed below. Each dilution was
plated in duplicate. The lower limit of detection
was approximately 10 cfu per g for skin and 10 cfu
per body cavity.
Media, microbial groups studied
Total viable counts were obtained using Plate Count
Agar (Oxoid CM325, Basingstoke, UK) incubated
aerobically at 30°C for 48 h. Presumptive coliforms
were detected using MacConkey Agar No. 3 (Oxoid
CM115) incubated aerobically at 37°C for 24 h.
Only pink/red colonies > 2 mm diameter were
counted. Presumptive Pseudomonas spp. were enumer-
ated on cephaloridine-fucidin-centrimide agar
(Oxoid CM559 with supplement SR103E) incubated
aerobically at 25°C for 48 h and counting only
oxidase positive colonies.
Temperature measurement
The temperatures of the circulating air and of the
carcase hung on the first shackle were measured at
frequent intervals using thermister probes, recorded
on a Grant 1200 series Squirrel Meter/Logger
(Grant Instruments, Cambridge, UK). To measure
air temperature the probe was located on the test
rig in the flow of air from the duct. The remaining
3 probes were inserted in the carcase as follows: (1)
under the skin; (2) into a deep cut in the upper leg
muscle; (3) into the body cavity.
Chlorine determination
At the start of each trial, 25 ml water were collected
from the rig spray reservoir in a plastic bottle in
order to estimate and adjust the chlorine content of
the sprays using 14% to 15% sodium hypochlorite
as supplied by the processing plant. The amounts of
total and free residual chlorine were estimated using
a Camlat Hach colorimeter (Camlab, Cambridge
UK) with DPD indicators (Mead and Thomas 1973).
Experimental design, presentation and
analysis of results
The carcases were of approximately uniform size
and within the factory target weight range (~ 2·18 kg).
Three carcases, each representing one replicate for
analysis purposes, were examined on each sampling
occasion, within each treatment. In order to allow
for variations in the initial microbial load attribut-
able to pre-chill processes or flock differences, the
results for carcases were reported as mean
proportional log1 0 change from counts obtained from
the unchilled carcases (zero-time controls) in that
particular treatment group. The results can therefore
be either positive if the microbial numbers increase
or negative if they decrease. Analysis of variance
(ANOVA) was undertaken using ‘Minitab’ software.
RESULTS
Effect of chilling environment on carcase
temperature
The mean cooling (rig air) temperatures achieved
during each trial and the variations observed (given
as one standard deviation) are presented in the Table
which also includes the time required for the skin
and deep tissue to cool from 20°C to 10°C and 10°
to 4°C. The body cavity temperature data were less
consistent than those from the skin and deep tissue,
because of variations in the size of the opening
which allowed some air to pass through some
carcases but not others. When air passed through,
the temperature in the cavity was lower than the
deep tissue temperature, although the skin
temperature was always lower than the cavity.
Figure 2 shows the temperature profile from
Trial 9, which is similar to that observed in the
other trials. A fairly typical Newtonian pattern was
observed throughout (the rate of temperature fall
was proportional to the body temperature). The skin
temperature fell faster than the deep tissue
temperature, with the cavity close to the latter. The
air temperature remained fairly steady, oscillating
between the control points of the chiller unit. No
effect was observed on the temperature profiles when
the doors were briefly opened to remove and replace
carcases (see Figure 2). These trials produced no
evidence of enhanced cooling by the use of sprays.
Effect of chilling environment on microbial
numbers on carcases during chilling
The mean microbiological counts for the 22 control
carcases (zero-time before chilling) were 4·81 ± 0·32,
3·06 ± 0·36 and 2·78 ± 0·61 log1 0 cfu/g of breast
skin and 4·98 ± 0·52, 4·25 ± 0·48 and 2·82 ± 0·76
log1 0 cfu/body cavity for TVC, coliforms and pseu-
domonads respectively. Resulting proportional log1 0
changes are illustrated in Figures 3 to 5.
Overall there was no consistent effect of time
or temperature on TVC or counts of coliforms and
 HYGIENE IN AIR-CHILLING OF POULTRY
Figure 2. Temperature profile during the cooling of a chicken carcase in Trial 9, at 11°C with intermittent water sprays chlorinated at 100 ppm during the 1st
60 min.
pseudomonads recovered from the skin or body
cavity during chilling. There was, however, a
significant overall effect of the use of water sprays
on TVC and counts of coliforms from the breast
skin (P < 0·001) and for coliforms and pseudomonads
(P = 0·001 and P = 0·045 respectively) recovered from
the body cavity.
On the skin there was little change in microbial
contamination during dry chilling, whereas the use
of sprays reduced the counts, indicating a washing
effect. Overall there was a reduction after 1 h of
intermittent spraying of log1 0 cfu/g 0·28, 0·48 and
0·73 for TVC, coliforms and pseudomonads
respectively. Spraying ceased at 60 min, after which
there was no significant change in counts for the
remaining hour of processing time.
In the body cavity with dry chilling, however,
there were also reductions in TVC and counts of
coliforms and pseudomonads of log1 0 cfu/body
cavity of up to 0·67, 1·01 and 0·88 (at 3°C for
120 min) respectively.
Effect of water chlorination
Overall the concentration of chlorine had little effect
on microbial counts on the skin. However, when
the changes in counts of the different microbial
groups were considered, statistical analysis indicated
a significant influence of chlorine concentration for
TVC, coliforms and pseudomonads on the skin
(P < 0·001, 0·02 and < 0·001 respectively). There was
a marginally greater reduction in TVC and colif-
orms when using the highest concentration (decrease
of log 1 0 cfu/g 0·43 compared to 0·09 for TVC and
a decrease of 0·62 compared to 0·26 for coliforms
following processing with 250 and 50 ppm chlorine
respectively). This trend was reversed in the case of
pseudomonads (decrease of log1 0 cfu/g 0·29 and
1·29 for 250 and 50 ppm chlorine respectively).
There was no significant influence of sampling time
579
or significant interaction between sampling time and
chlorine concentration. There was little change in
TVC and coliform counts from the body cavity
during these trials, although use of the higher
concentration resulted in a decrease of 0·29
log 1 0 cfu/body cavity for pseudomonads compared
to no change in numbers when 50 ppm chlorine
was used (P = 0·01).
Effect of extended storage at 11°C on
microbial counts on the carcases (Figure 5)
After storage for approximately 12 h at 11°C, there
was little change in microbial numbers recovered
from the breast skin or body cavity following dry
chilling. However, after spraying with water
containing 250 ppm of chlorine during the 1st h,
there was an overnight increase in the numbers of
coliforms and pseudomonads (mean log1 0 cfu/g 1·34
and 0·79 respectively) recovered from the skin, and
a similar effect in the body cavity (increase of
log 1 0 cfu/body cavity of 0·49, 0·64 and 1·10 for
TVC, and coliforms and pseudomonad counts
respectively).
DISCUSSION
In more than half of the measurements, there was
no significant change in counts, and where significant
change occurred, this was mostly a fall. Significant
increases in counts were observed in relatively few
situations. In particular counts of all types but
especially pseudomonads increased in the body
cavity after prolonged storage at 11°C when sprays
(250 ppm chlorine) had been applied during chilling.
The effect of prolonged storage after use of sprays
with lower concentrations of chlorine was not
investigated. Prolonged storage after dry chilling
caused no significant increase in counts. These key
findings using an experimental rig in a controlled
580 V.M. ALLEN ET AL.

Figure 3. The proportional change (log1 0 cfu per g (skin) or per body cavity) in numbers of microorganisms recovered from carcases processed in the rig at –1°C,
with or without intermittent water sprays during the 1st 60 min. (Overall significant effect of sprays on coliforms on skin, P = 0·02, and pseudomonads on skin
and in the body cavity, P = 0·01)

environment reflected parallel findings in
commercial trials (Allen et al., 2000) in that the use
of water sprays during air chilling had a significant,
but not always detrimental effect on the microbial
numbers recovered from the carcases. In the absence
of sprays, there was some reduction in microbial
contamination of the body cavity during the chilling
process as also found in separate trials in a
commercial chiller. This would seem to be due to
some internal drying of the body cavity in the
absence of sprays, leading to a fall in microbe
numbers. Equally, the presence of sprays might be
expected to increase microbial load by washing
contaminants into the carcases cavity. The degree
of drying and the extent of accumulation of
microbes in the body cavity would depend on the
sizes of opening at the vent and neck ends of the
carcase, which were observed to vary. A clear airway
through carcases appears important for efficient air
cooling.
Conversely, there was little change on the skin
of the carcases which was in agreement with both
the commercial findings and information in some
of the literature (Lahellec et al., 1973; Stephan and
Fehlhaber, 1994). When water sprays were used
during the 1st h of chilling in the model rig, there
was, as expected, a reduction in microbial
contamination of the skin because of a washing
effect, but numbers increased after spraying had
ceased, so that after a further h there was little overall
 HYGIENE IN AIR-CHILLING OF POULTRY 581

Figure 4. The proportional change (log1 0 cfu per g (skin) or per body cavity) in numbers of microorganisms recovered from carcases processed in the rig at 3°C, with
or without intermittent water sprays during the 1st 60 min. (Overall significant effect of sprays on coliforms, P = 0·04 and TVC, P = 0·02 on skin)

change either on the skin or in the body cavity. This
temporary reduction on the skin during spraying
was less evident under commercial conditions,
although Graw et al. (1997) reported a 0·5 log1 0
reduction in the microbial load on the skin.
The discrepancy may be due to the water
sprays in the rig. This aspect of modelling a
commercial system was the most difficult because
carcases in the commercial system pass directly
through the sprays twice but are in the vicinity of
the spray and the related mist several times.
However, in the rig there were 5 discrete applica-
tions of spray, which is likely to have led to a washing
rather than a damping action. Furthermore, samples
of breast skin, which received the sprays directly,
were examined in these rig trials. Neck skin samples
were used in the trials in commercial chillers (Allen
et al., 2000) and the neck skins would have
accumulated more contamination as the carcases
drained. In addition, there is likely to have been less
microbial contamination in the experimental rig
environment, and hence less possibility of microbial
transfer via the water sprays.
The addition of high concentrations of chlorine
to the water sprays had little lasting effect on the
microbial load. Following initial spraying, despite
the fact that the water was strongly chlorinated
(250 ppm), there was an increase in microbial counts
from both skin and body cavity sampling sites after
overnight storage at 11°C by up to 1 log unit. The
chlorine in the spray was probably quickly
neutralised by organic matter on the carcase surface
(Patterson 1968) so that chlorine would be unlikely
to be present after the sprays had stopped. This did
not occur following dry chilling where counts on
the skin remained unchanged throughout, dry
582
Figure 5. The proportional change (log1 0 cfu per g (skin) or per body cavity) in numbers of microorganisms recovered from carcases processed in the rig at 11°C,
with or without intermittent water sprays during the 1st 60 min. (Overall significant effect of sprays on coliforms, TVC and pseudomonads in the body cavity,
P = 0·07, P = 0·03 and P < 0·001 respectively).
conditions offering the better long-term storage
prospects. These results are applicable to the situa-
tion in the hanging-on area in the commercial chiller,
for carcases there would be wet from passing through
the final internal/external carcase washing stage.
From a microbiological perspective, the findings in
this study indicate that, unless sprays are used spar-
ingly and under very hygienic conditions with well
chlorinated water, dry chilling is preferable.
Although the exterior surface of the carcase
cooled relatively quickly, the deep tissue and body
cavity remained warmer for longer. Of particular
note is the time for the deep tissue of the carcase to
cool to the safe value of 4°C (Table 1). This took at
best, just under 1 h and at worst, when the cooling
temperature was inadequate, was not achieved by
V.M. ALLEN ET AL.
2 h. Possibly of greater concern was the length of
time that the carcases remained above 10°C, even
with air at –3°C, this could still be up to 1 h. Any
delay in the chilling operation, causing for example,
carcases to remain in the entry or pre-chill area
would add to this time.
In summary, use of a model chilling rig
confir med results obtained previously in
commercial plants that air chilling, with or without
chlorinated water sprays, usually caused small
reductions or no overall change in numbers of
microbes on the external surface of carcases.
Numbers of microbes in the body cavity only
increased when (chlorinated) water sprays were
used during chilling. Dry storage, of previously
sprayed carcases, for extended periods (for example
 HYGIENE IN AIR-CHILLING OF POULTRY
11 to 12 h) at the temperature often encountered
at the entrance to commercial air chillers, resulted
in significant increases in numbers of microbes.
Addition of chlorine to the spray did not appear
to reduce this effect.
ACKNOWLEDGEMENTS
This work was funded by the Ministry of Agriculture,
Fisheries and Food. We are grateful for the willing
cooperation and technical advice of a UK poultry
company, including provision of facilities and carcases.
Thanks are also due to Bryce Henderson and Tony
Kendall of Pennine Environmental Services Ltd for
advice on design of poultry chilling systems.
REFERENCES
ALLEN, V.M., CORRY, J.E.L., BURTON , C.H. WHYTE, R.T. & MEAD,
G.C. (2000) Hygiene aspects of modern poultry chilling.
International Journal of Food Microbiology, 58: 39–48
ANON (1978) Microbiology and shelf-life of chilled poultry carcasses.
Commission of the European Communities, Information on
Agriculture No. 61.
BERNER, H., KLEEBERGER, A. & BUSSE, M. (1969) Untersuchung
583
über ein neues Kühlverfahren für Schlachtgeflügel. III. Hygi-
enische uberprufung des Verfahrens. Fleischwirtschaft, 49: 1617,
1620, 1623.
GRAW, C., KOBE, A. & FRIES, R. (1997) Luft- und Luft- Sprühküh-
lung in der Geflügelfleischgewinnung—Ein mikrobiologischer
Vergleich 1. Gesamtkeimzahl. Fleischwirtschaft, 77: 78–80.
KN OOP, G.N., PAR MELEE, C.E. & STAD ELMAN, W.J. (1971)
Microbiological characteristics of wet- and dry-chilled poultry.
Poultry Science 50: 530–536.
LAHELLEC, C., MEURIER, C. & BENEJEAN, G. (1973) Influence de la
technique de refroidissement sur l’importance des contamina-
tions superficielle des carcasses de vollailles. Bulletin
d’Information, Station Experimentale d’Aviculture de Ploufragan, 13:
131–138.
MEAD , G.C. & THOMAS, N.L. (1973) Factors affecting the use of
chlorine in the spin-chilling of eviscerated poultry. British Poultry
Science, 14: 99–117.
MEAD, G.C. (1989 ) Hygiene problems and control of process
contamination, in: MEAD G.C. (Ed) Processing of Poultry pp.
183–220 (London, Elsevier applied science).
MEAD , G.C., ALLEN, V.M., BURTON , C.H. & CORRY, J.E.L. (2000)
Microbial cross-contamination during air chilling of poultry.
British Poultry Science, 41: 158–162.
PATTERSON, J.T. (1968) Chlorination of water used for poultry
processing. British Poultry Science, 9: 129–169.
STEPHAN, F. & FEHLHABER, K. (1994) Geflügelfleischgewinnung –
Untersuchungen zur Hygiene des Luft-Sprüh-Kühlverfahrens.
Fleischwirtschaft, 74: 870–873.